Electromechanical transient modeling simulation method for public charging and battery swap station of electric vehicle

Abstract

The invention discloses an electromechanical transient modeling simulation method for an electric vehicle public charging and battery swap station, which belongs to the field of electric vehicle charging and battery swap, and comprises the following steps: S1, fusing a battery electrochemical-electromechanical transmission-power grid interaction multi-physical field model to obtain a full-dimensional coupling model; s2, describing an EV arrival rule by improving a Gamma-fractional order Poisson process, predicting battery charge state distribution in combination with deep learning, and obtaining a charge state sample set when the battery reaches a charging station; s3, constructing a charging and swapping station-power grid wide area cooperative control framework, and realizing cooperative optimization of battery recombination and power grid voltage support through fractional order sliding mode control; and S4, verifying the precision of the full-dimensional coupling model through a multi-scene pressure test. By adopting the electromechanical transient modeling simulation method for the public charging and battery swap station of the electric vehicle, high-precision modeling, load prediction and stability optimization of the electromechanical transient process of the charging and battery swap station are realized.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of electric vehicle charging and battery swapping, and in particular to an electromechanical transient modeling and simulation method for a public charging and battery swapping station of an electric vehicle. BACKGROUND

[0002] With the continuous increase of global electric vehicle (EV) penetration rate, the operation characteristics of public charging and battery swapping stations as key infrastructure have an increasingly significant impact on power grid stability. Since the charging and battery swapping station involves multi-physical field coupling processes such as battery electrochemistry, electromechanical transmission and power grid interaction, and the EV load has strong spatiotemporal randomness, the traditional modeling and control method is difficult to accurately depict the nonlinear dynamic characteristics in the transient process. In addition, the coupling effect of EV charging and battery swapping behavior and traffic flow aggravates the complexity of load prediction, and the existing transient control strategy has limitations in multi-time scale coordination, resulting in insufficient system stability. SUMMARY

[0003] The purpose of the application is to provide an electromechanical transient modeling and simulation method for a public charging and battery swapping station of an electric vehicle, which solves the above technical problems.

[0004] To achieve the above purpose, the application provides an electromechanical transient modeling and simulation method for a public charging and battery swapping station of an electric vehicle, comprising the following steps: S1, a multi-physical field model of battery electrochemistry-electromechanical transmission-power grid interaction is fused to obtain a full-dimensional coupled model; S2, a regional traffic flow model and an EV trip chain are integrated, the EV arrival law is described by improving the Gamma-fractional order Poisson process, and the battery state of charge distribution is predicted by combining deep learning to obtain a state of charge sample set when arriving at the charging and battery swapping station; S3, a charging and battery swapping station-power grid wide area collaborative control framework is constructed, and the collaborative optimization of battery reorganization and power grid voltage support is realized through fractional order sliding mode control; S4, a hardware-in-the-loop system including a real-time digital simulator and a mechanical arm physical prototype is constructed, and the full-dimensional coupled model accuracy is verified through multi-scenario stress testing.

[0005] Therefore, the application adopts the above-mentioned electromechanical transient modeling and simulation method for a public charging and battery swapping station of an electric vehicle, which has the beneficial effects of: Multi-physical field coupling improves modeling accuracy 1. The full-dimensional coupled model fuses the fractional order polarization characteristics of the battery, the rigid-flexible coupled vibration of the mechanical arm and the frequency-time domain interaction of the power grid, which improves the transient voltage prediction accuracy by 42% compared with the traditional integer order model, and reduces the battery swapping impact prediction error from 25% to 8%; 2. Fractional-order Deep Learning Enhances Load Prediction Robustness: Improved Gamma-fractional Poisson process captures the long memory characteristics of EV arrival, combined with LSTM to predict state of charge distribution, the root mean square error of spatiotemporal load prediction is reduced by 37% compared with the traditional Poisson process, and the load fluctuation prediction deviation is ≤5.2% under extreme conditions; 3. Wide-area collaborative control enhances grid stability: Fractional sliding mode control achieves coordinated battery reconfiguration and grid voltage support, improving the voltage transient index (VSI) by 28% during grid faults, and reducing low-frequency oscillation suppression by 40% compared to traditional PID control; 4. Hardware-in-the-loop verification ensures model reliability: The uncertainty of the model is quantified by the HIL system. After Bayesian update, the error of the fully coupled model parameters is ≤3%, the 95% confidence interval covers the response to extreme conditions, and the robustness is improved by 55% compared with the uncalibrated model.

[0006] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0007] Figure 1 This is a flowchart of a method for electromechanical transient modeling and simulation of a public charging and swapping station for electric vehicles, as described in this invention. Detailed Implementation

[0008] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0009] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0010] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0011] like Figure 1 As shown, a method for electromechanical transient modeling and simulation of public charging and battery swapping stations for electric vehicles includes the following steps: S1. A multi-physics model integrating battery electrochemistry, electromechanical transmission, and power grid interaction is obtained to obtain a full-dimensional coupled model; Step S1 specifically includes the following steps: S11. Based on the SPDT switch reorganization strategy, the fractional Thevenin model is used to describe the battery polarization characteristics, and the series-parallel adaptive reorganization is achieved through 2N-2 switch configuration to suppress battery imbalance during transient processes. Step S11 specifically includes the following steps: S111, Fractional Polarization Characteristic Modeling: ; In the formula, express The terminal voltage of a single battery cell at any given time; Indicates open-circuit voltage; Indicates the state of charge of a single battery cell; Indicates the internal resistance of the ohm; express Constant charging and discharging current; Indicates the polarization internal resistance of a single battery cell; express Order integral operator, Denotes the order of integration, and ; S112, SPDT switch reconfiguration modeling: ; In the formula, express Constantly reorganize and compensate voltage; Indicates the battery recombination coefficient; Indicates the number of switches, and , Indicates the number of individual battery cells; express Time of the first Each switch state; express Time of the first The state of charge of each individual battery cell; express The average state of charge of the battery pack at any given time; Indicates the number of individual battery cells; S113, Temperature-Charge State Coupling Modeling: ; in, ; In the formula, express State of charge at any given moment; express; Indicates the battery's rated capacity; Indicates the temperature influence coefficient; Indicates battery temperature; Indicates the reference temperature; Indicates the temperature sensitivity coefficient; S12. Add an elastic deformation term to the rigid body dynamics equation to describe the vibration characteristics of the robotic arm during rapid battery swapping; at the same time, use digital twin technology to map the real-time relationship between the joint displacement, torque and motor current of the robotic arm. Step S12 specifically includes the following steps: S121. Rigid Body Dynamics Modeling: ; In the formula, This represents the equivalent moment of inertia of the robotic arm; and These represent the joint angles of the robotic arm. The first and second derivatives; This represents the driving torque of the robotic arm joints; This represents the mechanical frictional resistance torque; S122, Elastic Modeling of Flexible Joints: ; In the formula, Indicates the joint damping coefficient; Indicates joint stiffness; Indicates the angle of rotation at the equilibrium position; Indicates the vibration disturbance torque; S123. Mechatronic transmission rigid-flexible coupling modeling: Couple motor current with mechanical torque to establish electromechanical dynamic equations, reflecting the conversion characteristics of electrical energy to mechanical energy during transient processes. ; In the formula, Indicates the motor inductance; Indicates the motor current; Indicates the internal resistance of the motor; Indicates the motor terminal voltage; Represents the torque constant; S13. Based on the PQ node model, the transient impedance frequency characteristics of the power grid are introduced to establish a frequency-time domain coupled interactive model; at the same time, the impact of power grid voltage drop and frequency fluctuation on the charging and swapping station is mapped through digital twin; Step S13 specifically includes the following steps: S131, Norton equivalent modeling of power grid: ; In the formula, express The system injects current at all times; Indicates short-circuit current; express Constantly monitor the bus voltage of the charging and battery swapping station; express Equivalent impedance of the system at any given time; S132, Frequency-Time Domain Coupled Modeling: ; In the formula, express Constant-time grid voltage deviation; Indicates the inverse Laplace transform; Represents the system impedance transfer function. Represents complex frequency; Laplace transform of the current change at a charging / swapping station; S133. Transient Stability Modeling: Introducing transient stability criteria, a correlation model is established between voltage drop depth, duration, and the withstand capability of charging and swapping station equipment. ; In the formula, express Voltage transient index at time moment; Indicates the lowest voltage during the transient process; Indicates the rated voltage; Indicates the voltage recovery time constant; S14. Digital Twin Mapping and Model Update: By fusing measured data and model predictions through Kalman filtering, dynamic matching between the virtual model and the physical system is achieved. In step S14, the following state-space equation and Kalman filter update expression are established: ; ; In the formula, and They represent Time and The state vector at time t, and , , , and These represent the grid voltage phase angle, grid angular frequency, inductor current, and capacitor voltage, respectively. , and Both represent system matrices; express The control input vector at each time step; and These represent process noise and measurement noise, respectively. Represents the measurement vector; This represents the estimated state vector value; Represents the Kalman gain matrix; Indicates the predicted state; Represents the error covariance matrix; S15. By combining electrochemical-electromechanical coupling, electromechanical-grid coupling, and grid-electrochemical coupling through cross-coupling terms, a full-dimensional coupling model is formed.

[0012] In step S15, the expression for the full-dimensional coupling model is as follows: Electrochemical-electromechanical coupling model: ; Electromechanical-grid coupling model: ; Power grid-electrochemical coupling model: ; In the formula, Indicates electromagnetic torque; Represents the torque constant; This indicates the real-time rotational speed of the robotic arm joints; Indicates the mass of a single battery cell; Indicates the specific heat capacity of a single battery cell; express The first derivative; This represents the convective heat transfer coefficient between the battery and the environment. This indicates the effective area for heat exchange between the battery and the environment. Indicates ambient temperature; This represents the active power injected into the power grid by the electromechanical system; Indicates the energy conversion efficiency of the motor; This indicates the real-time driving torque of the robotic arm joints; Indicates the equivalent resistance of the power grid line; Indicates the equivalent reactance of the power grid line; This represents the reactive power injected into the power grid by the electromechanical system; Indicates the real-time voltage of the power grid bus; This represents the equivalent impedance of the inverter.

[0013] S2. Integrate regional traffic flow models with EV travel chains, describe EV arrival patterns by improving the Gamma-fractional Poisson process, and combine deep learning to predict battery state of charge distribution to obtain a sample set of state of charge when arriving at the charging and swapping station. Step S2 specifically includes the following steps: S21. Traffic flow-load mapping modeling based on LSTM: A nonlinear mapping model of traffic flow and charging / swapping load is constructed by using a spatiotemporal mapping matrix and combining it with a long short-term memory network to capture the lag correlation between traffic congestion and load growth. Step S21 specifically includes the following steps: S211. Traffic Flow Feature Extraction: Noise in traffic flow data is removed using Gaussian filtering to extract traffic flow features, including congestion index, average vehicle speed, and vehicle density. ; In the formula, Indicates the vehicle speed after filtering; Indicates the Gaussian kernel standard deviation; Indicates a long time window; This represents the original vehicle speed data; Represents the integral variable; S212. Constructing a spatiotemporal mapping matrix: Based on the road network zoning method, construct a weight matrix of regional traffic flow and charging / switching station load to reflect the impact of spatial location on load distribution. ; In the formula, Indicates the first The charging and battery swapping station and the first Mapping weights between network nodes; Indicates the first The charging and battery swapping station and the first Spatial distance between network nodes; Indicates the total number of nodes in the road network; S213, LSTM prediction model training: A three-layer LSTM network is used, with the input being traffic flow characteristics and load data from six historical time steps, and the output being the load prediction value for the next step. The network parameters are optimized through backpropagation. ; ; In the formula, and They represent Time and The state of the hidden layer is always hidden; This represents the Sigmoid activation function; and Both represent weight matrices; express The time-feature input vector, and , Indicates traffic density. Historical samples representing the spatiotemporal distribution of load; express Forecast load values ​​in real time; and Both represent bias vectors; S22. EV Arrival Modeling Using Fractional Poisson Processes: In load forecasting, a fractional Poisson process is introduced to describe the long memory of EV arrival, and the model is modeled using the integral order. Adjusting the non-exponential nature of arrival intervals; Step S22 specifically includes the following steps: S221. Improved Gamma Distribution Modeling: Based on NHTS data, a fractional-order Gamma distribution model with correction terms is established to describe the initial arrival time of EVs. ; In the formula, Indicates EV in The probability density of arriving at the charging / swapping station at any given time; Indicates the scale parameter; Indicates shape parameters; Represents the gamma function; Indicates the time exponentiation term; Indicates the time when the EV arrives; Represents a fractional correction term; and All represent correction factors; S222. Constructing a fractional Poisson process: The arrival probability of a fractional Poisson process is defined using the Mittag-Leffler function to reflect the time correlation of EV arrivals. ; In the formula, express Arrive within the time limit The probability of a vehicle being an EV; This represents the Mittag-Leffler function; S223. Arrival Quantity Prediction: By utilizing the mean characteristics of a fractional Poisson process and adjusting the arrival rate according to the type of regional activity, a time-specific arrival quantity prediction is generated. ; In the formula, express Arrival rate at any given time; Indicates the baseline arrival rate; Indicates the time period correction function; Indicates the weight of the region type; S23. Prediction of battery state of charge degradation trajectory by integrating battery health status: Combining battery health status estimation methods, the memory characteristics of state of charge decay are described by fractional differential equations, and the influence factor of battery health status on energy consumption is introduced.

[0014] Step S23 specifically includes the following steps: S231, Fractional-order state of charge decay modeling: The nonlocal characteristics of state of charge decay are described using Caputo fractional derivatives, reflecting the cumulative impact of different driving habits on the battery. ; In the formula, express Caputo fractional derivative; express Constant battery state of charge; Indicates the energy consumption coefficient; Indicates real-time vehicle speed; Indicates the battery's rated capacity; Indicates the influence coefficient of battery health status; Indicates the battery's health status; S232. Dynamic Update of Battery Health Status: Based on the capacity decay curve, the battery health status is updated using the Arrhenius equation, reflecting the impact of temperature and depth of charge / discharge on battery aging. ; In the formula, express Monitor battery health status at all times; This indicates the amount of content decay during the period; Indicates activation energy; Represents the gas constant; S233. Generating EV driving trajectories through Monte Carlo simulation, and combining a fractional-order state-of-charge decay model with the battery health state update equation, generating a state-of-charge sample set upon arrival at the charging / swapping station: ; In the formula, This indicates the state of charge when the device reaches a charging / swapping station. Indicates the initial state of charge; Indicates travel time; This represents the battery health degradation coefficient.

[0015] S3. Construct a wide-area collaborative control framework between charging and battery swapping stations and the power grid, and achieve collaborative optimization of battery reconfiguration and grid voltage support through fractional sliding mode control; Step S3 specifically includes the following steps: S31, Adaptive Fractional-Order Battery Reconfiguration Control: Integrates SPDT switching reconfiguration strategy and fractional-order PID control, and dynamically optimizes the series and parallel states of the battery pack by adaptively adjusting the fractional order through fuzzy logic. Step S31 specifically includes the following steps: S311, Fractional PID Control Modeling: ; In the formula, Indicates control signals; , and These represent the proportional, integral, and differential coefficients, respectively. This indicates the deviation in the state of charge, and , and These represent the maximum and minimum states of charge, respectively. express fractional differential operators, and ; express Fractional integral operator of order, and ; S312. Reorganization Cost Function Optimization: A dynamic cost function is constructed with state-of-charge equilibrium and energy loss as optimization objectives. The optimal reorganization state is then solved using SPDT switching combinations. ; in, ; In the formula, Represents the optimal switch state vector; Represents the recombination cost function; Indicates the weighting coefficient; express Fractional differential operators; Indicates the first The state of charge of each individual battery cell; Indicates the penalty coefficient for switching actions; and They represent Time and The first moment Each switch state; Indicates the number of switches; Represents the switch state vector; S313. Fuzzy Adaptive Order Adjustment: Based on the state-of-charge deviation rate and the grid voltage fluctuation amplitude, the fractional order is adjusted in real time using fuzzy logic. and Optimize transient response characteristics: ; ; In the formula, and Indicates the adjusted fractional order; and They represent and The baseline score order; and They represent and The adjustment range; Represents a fuzzy mapping function; The derivative representing the deviation of the state of charge; S32. Wide-area fractional-order damping control: Based on the model-free control concept, a wide-area damping controller is designed. The frequency signals of adjacent substations are used to construct regional coordinated control to suppress low-frequency oscillations caused by charging and swapping. Step S32 specifically includes the following steps: S321. Wide-area frequency signal fusion: Multi-node frequency deviations are fused using weighted averaging to construct regional oscillation characteristics. ; in, ; In the formula, Indicates wide-area frequency deviation; Indicates the first The weight of each road network node; Indicates the first Frequency deviation of individual road network nodes; Indicates the first The equivalent impedance from each road network node to the charging and battery swapping station; Indicates the number of wide-area measurement points; S322, Fractional Damping Control Law Design: Utilizing the memory characteristics of fractional integrals, a damping control law related to historical frequency deviation is constructed: ; In the formula, Indicates damping power; Indicates control gain; express Fractional integral operator; express Wide-area frequency deviation at any given moment; S323, Transient Voltage Support Coordination: When the grid voltage drops below a threshold, the damping power and battery reconfiguration strategy are dynamically adjusted. ; In the formula, Indicates the output power of the charging and swapping station; Indicates the voltage-supported power; Indicates real-time voltage; S33. Electromechanical Transient Cooperative Optimization Control: Establish a cooperative optimization model for electromechanical transient processes, and use model prediction control to continuously optimize charging and swapping power and robotic arm motion sequence.

[0016] Step S33 specifically includes the following steps: S331. Electromechanical Coupled State-Space Modeling: Couples electrical dynamics with mechanical dynamics to construct state-space equations containing fractional terms. ; In the formula, express The input vector at time t, and , and These represent reconfiguration control and inverter control, respectively. Represents the interference vector; Indicates the output vector; and Both represent system matrices; S332. Multi-objective optimization cost function: A multi-objective cost function is constructed with electrical loss, mechanical vibration, and voltage deviation as optimization objectives, and the priority is controlled by adjusting the weighting coefficients. ; in, ; ; ; In the formula, Indicates the prediction time domain; , and All represent weights; , and These represent electrical costs, mechanical costs, and voltage costs, respectively. and All represent penalty coefficients; Indicates the inverter current; Indicates the first The driving torque of each joint; Indicates the first Reference values ​​for the driving torque of each joint; Indicates the first Angular acceleration of each joint; This indicates the bus voltage of the charging and battery swapping station.

[0017] S4. Construct a hardware-in-the-loop system that includes a real-time digital simulator and a physical prototype of a robotic arm, and verify the accuracy of the full-dimensional coupled model through multi-scenario stress testing.

[0018] Step S4 specifically includes the following steps: S41. Hybrid Scenario Generation: Based on the multi-scenario optimization method and combined with the boundary conditions set by the power grid specifications, test sequences containing normal operation and extreme conditions are generated through Monte Carlo sampling and Latin hypercube algorithm. Step S41 specifically includes the following steps: S411. Define typical scenario parameters: ; In the formula, Represents a set of test scenarios; , and These represent load scenarios, fault scenarios, and control scenarios, respectively. Represents the probability distribution function of the scene; S412, Mixed Scene Sequence Generation: A uniformly distributed test sequence is generated in a multidimensional parameter space using the Latin hypercube sampling algorithm. ; In the formula, This represents a scene sequence matrix, where each row corresponds to a parameter combination for a scene; Indicates the number of scenes; This represents the Latin hypercube sampling operator; S413. Setting Boundary Conditions: Based on the power grid transient characteristics specification and the withstand capability of charging and swapping station equipment, define extreme operating boundaries such as voltage deviation and frequency fluctuation. Generate boundary scenarios through Monte Carlo sampling to verify system robustness. ; In the formula, Indicates the percentage of grid voltage deviation; Indicates the power grid frequency deviation; Indicates the boundary of the charged state; Indicates the duration of the fault; S42 and HIL system synchronization control: Real-time simulation technology is used to synchronize the electrical-mechanical subsystem through the IEEE 1588 clock protocol to achieve collaborative verification of microsecond-level electrical dynamics and millisecond-level mechanical response; Step S42 specifically includes the following steps: S421. Multi-subsystem time synchronization modeling: Based on a clock synchronization protocol, establish a timestamp alignment mechanism between the electrical simulation subsystem and the mechanical simulation subsystem to eliminate sampling delay errors. ; In the formula, Indicates the system time after synchronization; Indicates the electrical simulation time; Indicates the mechanical simulation time; Indicates the amount of delay compensation, and , and These represent signal transmission delay and signal processing delay, respectively. S422. Data Interaction Interface Design: Based on real-time simulation technology, design a data interaction protocol for the electrical-mechanical subsystem, and adopt the OPC UA standard to achieve cross-platform data transmission. ; ; In the formula, Indicates the transmission of data frames; , and These represent the electrical state vector, mechanical state vector, and timestamp, respectively. Indicates data transmission delay; and These represent the data transmission time and reception time, respectively. S423. Transient Process Co-simulation: Combining a hybrid control strategy, electromechanical response data are synchronously collected during transient disturbances in the HIL system to verify the transient matching degree of the multiphysics model. ; In the formula, Indicates transient synchronization error; Indicates the number of sampling points; and They represent the first Simulated and measured values ​​for each sampling point; S43. Model Uncertainty Quantification: Introducing the Bayesian update method, the model parameters are calibrated using HIL test data to quantify the electromechanical transient uncertainty of the fully coupled model.

[0019] Step S43 specifically includes the following steps: S431, Bayesian parameter update modeling: ; in, ; ; In the formula, Represents given observation data Post-parameter The posterior distribution of; Represents the likelihood function; Indicates the prior distribution of the parameters. This represents the vector of parameters to be verified. Indicates the prior distribution of the parameters. This represents the mean vector of the prior distribution of the parameters. The covariance matrix represents the prior distribution of the parameters; Represents the likelihood distribution. The predicted values ​​of the fully coupled model for the observed data The covariance matrix representing the measurement noise; S432, Markov Chain Monte Carlo Sampling: Adaptive MCMC algorithm is used to sample from the posterior distribution and estimate the confidence intervals of the parameters. ; In the formula, and They represent The next iteration and The parameter values ​​for the next iteration; Indicates the step size factor; Indicates the suggested distribution. Indicates the first The proposed distribution covariance matrix for the next iteration; Indicates the probability of acceptance; express Probability of the next iteration; express Probability of the next iteration; S433. Uncertainty Propagation Analysis: This involves propagating parameter uncertainties to the system response through Monte Carlo simulation, quantifying the confidence intervals of model predictions, and evaluating the system robustness under extreme conditions. ; ; In the formula, Represents the system response variable; Represents the system response function; These represent the input parameter vectors respectively; This represents a 95% confidence interval; and All represent the quantiles of the confidence interval.

[0020] Step S4 is followed by S5, which combines a multi-objective optimization framework to construct an evaluation index system from three dimensions: transient stability, control accuracy, and energy efficiency, and quantitatively verify the effectiveness of electromechanical transient modeling and control strategies. The transient stability evaluation expression is as follows: ; ; In the formula, Indicates overshoot; This represents the peak value of the response during the transient process; Indicates the steady-state value; Indicates the adjustment time; Represents the time variable of a transient process; Represents transient response variables; The expression for evaluating control accuracy is as follows: ; ; In the formula, Indicates the root mean square error; and They represent the first Reference and actual values ​​for each sampling point; Indicates the maximum tracking error; The energy efficiency evaluation expression is as follows: ; ; In the formula, Indicates system efficiency; and These represent the output energy and the input energy, respectively. This indicates energy loss.

[0021] Simulation Experiment Simulation conditions Model platform: MATLAB / Simulink + OPAL-RT real-time simulator, building a fully coupled model and comparing it with the traditional integer-order model.

[0022] Operating conditions: Normal operation: EV arrival rate follows an improved Gamma-fractional Poisson process (where, =10 vehicles / hour =1.8, =0.5), the initial state of charge follows Extreme load: 25 EVs arriving within 30 minutes (125% over rated capacity), accompanied by a 20% voltage drop. Grid fault: A three-phase short circuit occurred, lasting 0.15 seconds, with a system inertia of [value missing]. .

[0023] Simulation process Normal operation: Run for 24 hours and compare the charge balance and grid frequency fluctuation of this application with the traditional model. Extreme load: Trigger a concentrated arrival event of EVs and record the voltage deviation and damping power response of the charging and swapping station bus. Grid fault: After simulating a fault, compare the frequency recovery time and voltage overshoot of this application (fractional-order damping control) with the traditional PID control.

[0024] Table 1 Simulation results under different working conditions

[0025] As shown in Table 1, this application, through the integration of multiphysics modeling, fractional-order theory and deep learning, significantly outperforms traditional methods in terms of transient response speed, parameter matching accuracy and adaptability to extreme conditions, thus verifying the advanced nature of the technical solution described in this application in electromechanical transient analysis and control of charging and swapping stations.

[0026] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for electromechanical transient modeling and simulation of public charging and battery swapping stations for electric vehicles, characterized in that: Includes the following steps: S1. A multi-physics model integrating battery electrochemistry, electromechanical transmission, and power grid interaction is obtained to obtain a full-dimensional coupled model; S2. Integrate regional traffic flow models with EV travel chains, describe EV arrival patterns by improving the Gamma-fractional Poisson process, and combine deep learning to predict battery state of charge distribution to obtain a sample set of state of charge when arriving at the charging and swapping station. S3. Construct a wide-area collaborative control framework between charging and battery swapping stations and the power grid, and achieve collaborative optimization of battery reconfiguration and grid voltage support through fractional sliding mode control; S4. Construct a hardware-in-the-loop system that includes a real-time digital simulator and a physical prototype of a robotic arm, and verify the accuracy of the full-dimensional coupled model through multi-scenario stress testing.

2. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11. Based on the SPDT switch reorganization strategy, the fractional Thevenin model is used to describe the battery polarization characteristics, and the series-parallel adaptive reorganization is achieved through 2N-2 switch configuration to suppress battery imbalance during transient processes. S12. Add an elastic deformation term to the rigid body dynamics equation to describe the vibration characteristics of the robotic arm during rapid battery swapping; at the same time, use digital twin technology to map the real-time relationship between the joint displacement, torque and motor current of the robotic arm. S13. Based on the PQ node model, the transient impedance frequency characteristics of the power grid are introduced to establish a frequency-time domain coupled interactive model; at the same time, the impact of power grid voltage drop and frequency fluctuation on the charging and swapping station is mapped through digital twin; S14. Digital Twin Mapping and Model Update: By fusing measured data and model predictions through Kalman filtering, dynamic matching between the virtual model and the physical system is achieved. S15. By combining electrochemical-electromechanical coupling, electromechanical-grid coupling, and grid-electrochemical coupling through cross-coupling terms, a full-dimensional coupling model is formed.

3. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 2, characterized in that: Step S11 specifically includes the following steps: S111, Fractional Polarization Characteristic Modeling: ； In the formula, express The terminal voltage of a single battery cell at any given time; Indicates open-circuit voltage; Indicates the state of charge of a single battery cell; Indicates the internal resistance of the ohm; express Constant charging and discharging current; Indicates the polarization internal resistance of a single battery cell; express Order integral operator, Denotes the order of integration, and ; S112, SPDT switch reconfiguration modeling: ； In the formula, express Constantly reorganize and compensate voltage; Indicates the battery recombination coefficient; Indicates the number of switches, and , Indicates the number of individual battery cells; express Time of the first Each switch state; express Time of the first The state of charge of each individual battery cell; express The average state of charge of the battery pack at any given time; Indicates the number of individual battery cells; S113, Temperature-Charge State Coupling Modeling: ； in, ； In the formula, express State of charge at any given moment; express; Indicates the battery's rated capacity; Indicates the temperature influence coefficient; Indicates battery temperature; Indicates the reference temperature; Indicates the temperature sensitivity coefficient; Step S12 specifically includes the following steps: S121. Rigid Body Dynamics Modeling: ； In the formula, This represents the equivalent moment of inertia of the robotic arm; and These represent the joint angles of the robotic arm. The first and second derivatives; This represents the driving torque of the robotic arm joints; This represents the mechanical frictional resistance torque; S122, Elastic Modeling of Flexible Joints: ； In the formula, Indicates the joint damping coefficient; Indicates joint stiffness; Indicates the angle of rotation at the equilibrium position; Indicates the vibration disturbance torque; S123. Mechatronic transmission rigid-flexible coupling modeling: Couple motor current with mechanical torque to establish electromechanical dynamic equations, reflecting the conversion characteristics of electrical energy to mechanical energy during transient processes. ； In the formula, Indicates the motor inductance; Indicates the motor current; Indicates the internal resistance of the motor; Indicates the motor terminal voltage; Represents the torque constant; Step S13 specifically includes the following steps: S131, Norton equivalent modeling of power grid: ； In the formula, express The system injects current at all times; Indicates short-circuit current; express Constantly monitor the bus voltage of the charging and battery swapping station; express Equivalent impedance of the system at any given time; S132, Frequency-Time Domain Coupled Modeling: ； In the formula, express Constant-time grid voltage deviation; Indicates the inverse Laplace transform; Represents the system impedance transfer function. Represents complex frequency; Laplace transform of the current change at a charging / swapping station; S133. Transient Stability Modeling: Introducing transient stability criteria, a correlation model is established between voltage drop depth, duration, and the withstand capability of charging and swapping station equipment. ； In the formula, express Voltage transient index at time moment; Indicates the lowest voltage during the transient process; Indicates the rated voltage; Indicates the voltage recovery time constant; In step S14, the following state-space equation and Kalman filter update expression are established: ； ； In the formula, and They represent Time and The state vector at time t, and , , , and These represent the grid voltage phase angle, grid angular frequency, inductor current, and capacitor voltage, respectively. , and Both represent system matrices; express The control input vector at each time step; and These represent process noise and measurement noise, respectively. Represents the measurement vector; This represents the estimated state vector value; Represents the Kalman gain matrix; Indicates the predicted state; Represents the error covariance matrix; In step S15, the expression for the full-dimensional coupling model is as follows: Electrochemical-electromechanical coupling model: ； Electromechanical-grid coupling model: ； Power grid-electrochemical coupling model: ； In the formula, Indicates electromagnetic torque; Represents the torque constant; This indicates the real-time rotational speed of the robotic arm joints; Indicates the mass of a single battery cell; Indicates the specific heat capacity of a single battery cell; express The first derivative; This represents the convective heat transfer coefficient between the battery and the environment. This indicates the effective area for heat exchange between the battery and the environment. Indicates ambient temperature; This represents the active power injected into the power grid by the electromechanical system; Indicates the energy conversion efficiency of the motor; This indicates the real-time driving torque of the robotic arm joints; Indicates the equivalent resistance of the power grid line; Indicates the equivalent reactance of the power grid line; This represents the reactive power injected into the power grid by the electromechanical system; Indicates the real-time voltage of the power grid bus; This represents the equivalent impedance of the inverter.

4. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 3, characterized in that: Step S2 specifically includes the following steps: S21. Traffic flow-load mapping modeling based on LSTM: A nonlinear mapping model of traffic flow and charging / swapping load is constructed by using a spatiotemporal mapping matrix and combining it with a long short-term memory network to capture the lag correlation between traffic congestion and load growth. S22. EV Arrival Modeling Using Fractional Poisson Processes: In load forecasting, a fractional Poisson process is introduced to describe the long memory of EV arrival, and the model is modeled using the integral order. Adjusting the non-exponential nature of arrival intervals; S23. Prediction of battery state of charge degradation trajectory by integrating battery health status: Combining battery health status estimation methods, the memory characteristics of state of charge decay are described by fractional differential equations, and the influence factor of battery health status on energy consumption is introduced.

5. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 4, characterized in that: Step S21 specifically includes the following steps: S211. Traffic Flow Feature Extraction: Noise in traffic flow data is removed using Gaussian filtering to extract traffic flow features, including congestion index, average vehicle speed, and vehicle density. ； In the formula, Indicates the vehicle speed after filtering; Indicates the Gaussian kernel standard deviation; Indicates a long time window; This represents the original vehicle speed data; Represents the integral variable; S212. Constructing a spatiotemporal mapping matrix: Based on the road network zoning method, construct a weight matrix of regional traffic flow and charging / switching station load to reflect the impact of spatial location on load distribution. ； In the formula, Indicates the first The charging and battery swapping station and the first Mapping weights between network nodes; Indicates the first The charging and battery swapping station and the first Spatial distance between network nodes; Indicates the total number of nodes in the road network; S213, LSTM prediction model training: A three-layer LSTM network is used, with the input being traffic flow characteristics and load data from six historical time steps, and the output being the load prediction value for the next step. The network parameters are optimized through backpropagation. ； ； In the formula, and They represent Time and The state of the hidden layer is always hidden; This represents the Sigmoid activation function; and Both represent weight matrices; express The time-feature input vector, and , Indicates traffic density. Historical samples representing the spatiotemporal distribution of load; express Forecast load values ​​in real time; and Both represent bias vectors; Step S22 specifically includes the following steps: S221. Improved Gamma Distribution Modeling: Based on NHTS data, a fractional-order Gamma distribution model with correction terms is established to describe the initial arrival time of EVs. ； In the formula, Indicates EV in The probability density of arriving at the charging / swapping station at any given time; Indicates the scale parameter; Indicates shape parameters; Represents the gamma function; Indicates the time exponentiation term; Indicates the time when the EV arrives; Represents a fractional correction term; and All represent correction factors; S222. Constructing a fractional Poisson process: The arrival probability of a fractional Poisson process is defined using the Mittag-Leffler function to reflect the time correlation of EV arrivals. ； In the formula, express Arrive within the time limit The probability of a vehicle being an EV; This represents the Mittag-Leffler function; S223. Arrival Quantity Prediction: By utilizing the mean characteristics of a fractional Poisson process and adjusting the arrival rate according to the type of regional activity, a time-specific arrival quantity prediction is generated. ； In the formula, express Arrival rate at any given time; Indicates the baseline arrival rate; Indicates the time period correction function; Indicates the weight of the region type; Step S23 specifically includes the following steps: S231, Fractional-order state of charge decay modeling: The nonlocal characteristics of state of charge decay are described using Caputo fractional derivatives, reflecting the cumulative impact of different driving habits on the battery. ； In the formula, express Caputo fractional derivative; express Constant battery state of charge; Indicates the energy consumption coefficient; Indicates real-time vehicle speed; Indicates the battery's rated capacity; Indicates the influence coefficient of battery health status; Indicates the battery's health status; S232. Dynamic Update of Battery Health Status: Based on the capacity decay curve, the battery health status is updated using the Arrhenius equation, reflecting the impact of temperature and depth of charge / discharge on battery aging. ； In the formula, express Monitor battery health status at all times; This indicates the amount of content decay during the period; Indicates activation energy; Represents the gas constant; S233. Generating EV driving trajectories through Monte Carlo simulation, and combining a fractional-order state-of-charge decay model with the battery health state update equation, generating a state-of-charge sample set upon arrival at the charging / swapping station: ； In the formula, This indicates the state of charge when the device reaches a charging / swapping station. Indicates the initial state of charge; Indicates travel time; This represents the battery health degradation coefficient.

6. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 5, characterized in that: Step S3 specifically includes the following steps: S31, Adaptive Fractional-Order Battery Reconfiguration Control: Integrates SPDT switching reconfiguration strategy and fractional-order PID control, and dynamically optimizes the series and parallel states of the battery pack by adaptively adjusting the fractional order through fuzzy logic. S32. Wide-area fractional-order damping control: Based on the model-free control concept, a wide-area damping controller is designed. The frequency signals of adjacent substations are used to construct regional coordinated control to suppress low-frequency oscillations caused by charging and swapping. S33. Electromechanical Transient Cooperative Optimization Control: Establish a cooperative optimization model for electromechanical transient processes, and use model prediction control to continuously optimize charging and swapping power and robotic arm motion sequence.

7. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 6, characterized in that: Step S31 specifically includes the following steps: S311, Fractional PID Control Modeling: ； In the formula, Indicates control signals; , and These represent the proportional, integral, and differential coefficients, respectively. This indicates the deviation in the state of charge, and , and These represent the maximum and minimum states of charge, respectively. express fractional differential operators, and ; express Fractional integral operator of order, and ; S312. Reorganization Cost Function Optimization: A dynamic cost function is constructed with state-of-charge equilibrium and energy loss as optimization objectives. The optimal reorganization state is then solved using SPDT switching combinations. ； in, ； In the formula, Represents the optimal switch state vector; Represents the recombination cost function; Indicates the weighting coefficient; express Fractional differential operators; Indicates the first The state of charge of each individual battery cell; Indicates the penalty coefficient for switching actions; and They represent Time and The first moment Each switch state; Indicates the number of switches; Represents the switch state vector; S313. Fuzzy Adaptive Order Adjustment: Based on the state-of-charge deviation rate and the grid voltage fluctuation amplitude, the fractional order is adjusted in real time using fuzzy logic. and Optimize transient response characteristics: ； ； In the formula, and Indicates the adjusted fractional order; and They represent and The baseline score order; and They represent and The adjustment range; Represents a fuzzy mapping function; The derivative representing the deviation of the state of charge; Step S32 specifically includes the following steps: S321. Wide-area frequency signal fusion: Multi-node frequency deviations are fused using weighted averaging to construct regional oscillation characteristics. ； in, ； In the formula, Indicates wide-area frequency deviation; Indicates the first The weight of each road network node; Indicates the first Frequency deviation of individual road network nodes; Indicates the first The equivalent impedance from each road network node to the charging and battery swapping station; Indicates the number of wide-area measurement points; S322, Fractional Damping Control Law Design: Utilizing the memory characteristics of fractional integrals, a damping control law related to historical frequency deviation is constructed: ； In the formula, Indicates damping power; Indicates control gain; express Fractional integral operator; express Wide-area frequency deviation at any given moment; S323, Transient Voltage Support Coordination: When the grid voltage drops below a threshold, the damping power and battery reconfiguration strategy are dynamically adjusted. ； In the formula, Indicates the output power of the charging and swapping station; Indicates the voltage-supported power; Indicates real-time voltage; Step S33 specifically includes the following steps: S331. Electromechanical Coupled State-Space Modeling: Couples electrical dynamics with mechanical dynamics to construct state-space equations containing fractional terms. ； In the formula, express The input vector at time t, and , and These represent reconfiguration control and inverter control, respectively. Represents the interference vector; Indicates the output vector; and Both represent system matrices; S332. Multi-objective optimization cost function: A multi-objective cost function is constructed with electrical loss, mechanical vibration, and voltage deviation as optimization objectives, and the priority is controlled by adjusting the weighting coefficients. ； in, ； ； ； In the formula, Indicates the prediction time domain; , and All represent weights; , and These represent electrical costs, mechanical costs, and voltage costs, respectively. and All represent penalty coefficients; Indicates the inverter current; Indicates the first The driving torque of each joint; Indicates the first Reference values ​​for the driving torque of each joint; Indicates the first Angular acceleration of each joint; This indicates the bus voltage of the charging and battery swapping station.

8. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 7, characterized in that: Step S4 specifically includes the following steps: S41. Hybrid Scenario Generation: Based on the multi-scenario optimization method and combined with the boundary conditions set by the power grid specifications, test sequences containing normal operation and extreme conditions are generated through Monte Carlo sampling and Latin hypercube algorithm. S42 and HIL system synchronization control: Real-time simulation technology is used to synchronize the electrical-mechanical subsystem through the IEEE 1588 clock protocol to achieve collaborative verification of microsecond-level electrical dynamics and millisecond-level mechanical response; S43. Model Uncertainty Quantification: Introducing the Bayesian update method, the model parameters are calibrated using HIL test data to quantify the electromechanical transient uncertainty of the fully coupled model.

9. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 8, characterized in that: Step S41 specifically includes the following steps: S411. Define typical scenario parameters: ； In the formula, Represents a set of test scenarios; , and These represent load scenarios, fault scenarios, and control scenarios, respectively. Represents the probability distribution function of the scene; S412, Mixed Scene Sequence Generation: A uniformly distributed test sequence is generated in a multidimensional parameter space using the Latin hypercube sampling algorithm. ； In the formula, This represents a scene sequence matrix, where each row corresponds to a parameter combination for a scene; Indicates the number of scenes; This represents the Latin hypercube sampling operator; S413. Setting Boundary Conditions: Based on the power grid transient characteristics specification and the withstand capability of charging and swapping station equipment, define extreme operating boundaries such as voltage deviation and frequency fluctuation. Generate boundary scenarios through Monte Carlo sampling to verify system robustness. ； In the formula, Indicates the percentage of grid voltage deviation; Indicates the power grid frequency deviation; Indicates the boundary of the charged state; Indicates the duration of the fault; Step S42 specifically includes the following steps: S421. Multi-subsystem time synchronization modeling: Based on a clock synchronization protocol, establish a timestamp alignment mechanism between the electrical simulation subsystem and the mechanical simulation subsystem to eliminate sampling delay errors. ； In the formula, Indicates the system time after synchronization; Indicates the electrical simulation time; Indicates the mechanical simulation time; Indicates the amount of delay compensation, and , and These represent signal transmission delay and signal processing delay, respectively. S422. Data Interaction Interface Design: Based on real-time simulation technology, design a data interaction protocol for the electrical-mechanical subsystem, and adopt the OPC UA standard to achieve cross-platform data transmission. ； ； In the formula, Indicates the transmission of data frames; , and These represent the electrical state vector, mechanical state vector, and timestamp, respectively. Indicates data transmission delay; and These represent the data transmission time and reception time, respectively. S423. Transient Process Co-simulation: Combining a hybrid control strategy, electromechanical response data are synchronously collected during transient disturbances in the HIL system to verify the transient matching degree of the multiphysics model. ； In the formula, Indicates transient synchronization error; Indicates the number of sampling points; and They represent the first Simulated and measured values ​​for each sampling point; Step S43 specifically includes the following steps: S431, Bayesian parameter update modeling: ； in, ； ； In the formula, Represents given observation data Post-parameter The posterior distribution of; Represents the likelihood function; Indicates the prior distribution of the parameters. This represents the vector of parameters to be verified. Indicates the prior distribution of the parameters. This represents the mean vector of the prior distribution of the parameters. The covariance matrix represents the prior distribution of the parameters; Represents the likelihood distribution. The predicted values ​​of the fully coupled model for the observed data The covariance matrix representing the measurement noise; S432, Markov Chain Monte Carlo Sampling: Adaptive MCMC algorithm is used to sample from the posterior distribution and estimate the confidence intervals of the parameters. ； In the formula, and They represent The next iteration and The parameter values ​​for the next iteration; Indicates the step size factor; Indicates the suggested distribution. Indicates the first The proposed distribution covariance matrix for the next iteration; Indicates the probability of acceptance; express Probability of the next iteration; express Probability of the next iteration; S433. Uncertainty Propagation Analysis: This involves propagating parameter uncertainties to the system response through Monte Carlo simulation, quantifying the confidence intervals of model predictions, and evaluating the system robustness under extreme conditions. ； ； In the formula, Represents the system response variable; Represents the system response function; These represent the input parameter vectors respectively; This represents a 95% confidence interval; and All represent the quantiles of the confidence interval.

10. The electromechanical transient modeling and simulation method for a public charging and swapping station for electric vehicles according to claim 9, characterized in that: Step S4 is followed by S5, which combines a multi-objective optimization framework to construct an evaluation index system from three dimensions: transient stability, control accuracy, and energy efficiency, and quantitatively verify the effectiveness of electromechanical transient modeling and control strategies. The transient stability evaluation expression is as follows: ； ； In the formula, Indicates overshoot; This represents the peak value of the response during the transient process; Indicates the steady-state value; Indicates the adjustment time; Represents the time variable of a transient process; Represents transient response variables; The expression for evaluating control accuracy is as follows: ； ； In the formula, Indicates the root mean square error; and They represent the first Reference and actual values ​​for each sampling point; Indicates the maximum tracking error; The energy efficiency evaluation expression is as follows: ； ； In the formula, Indicates system efficiency; and These represent the output energy and the input energy, respectively. This indicates energy loss.

Images