A heterogeneous aggregation control method for a group of electric vehicle loads

Abstract

The application discloses a heterogeneous aggregation control method for group charging electric vehicle loads, comprising the following steps: S1, constructing an equivalent mapping relationship between electric vehicles and occupied charging modules, and obtaining aggregated power; S2, based on the aggregated power, performing group charging load heterogeneous aggregation to obtain an aggregated model in the form of a state space equation; S3, determining a high-order sliding mode control law suitable for the aggregated model in the form of the state space equation, and determining a constraint condition of a control coefficient to obtain a final control law. The application constructs a heterogeneous aggregated model with a two-dimensional state space and tensor coefficients, describes a dynamic evolution law of total power of loads in a charging process, and realizes unified aggregation of a large number of group charging loads.

Description

Technical Field

[0001] This invention belongs to the field of power regulation technology, specifically relating to a heterogeneous aggregation control method for the load of a group-charged electric vehicle. Background Technology

[0002] Electric vehicles (EVs) are an important flexible regulation resource for promoting the balance of power supply and demand, possessing considerable regulation potential in short-term grid responses such as peak shaving, frequency regulation, and clean energy consumption. With the further increase in the number of EVs, charging demand has surged. Traditional single-pile charging infrastructure, due to its slow charging speed and low efficiency, is gradually failing to meet the growing charging demand. Therefore, group charging systems (hereinafter referred to as group charging) have gradually been adopted and are continuously developing in the charging industry as a new type of charging facility. Group charging systems consist of multiple charging terminals and shared charging modules, offering greater charging capacity and flexible charging services. In the future, more and more EV users will choose group charging for charging and participating in grid regulation; therefore, researching its aggregation and control methods is of great significance.

[0003] Existing aggregation control models for electric vehicles (EVs) focus on single-pile charging loads and do not address the regulation of group-charging EVs. The core characteristic of group-charging systems is the intensive utilization of charging modules, with multiple modules forming a shared power pool, dynamically allocating module resources to charge connected vehicles. Compared to traditional aggregation control models, group-charging load aggregation modeling presents several challenges. First, in group-charging systems, charging modules are allocated to multiple different vehicles. Depending on user needs, these modules have different states, such as the number allocated and the time. Therefore, different models are needed to describe the charging process for each module. Second, EVs charged by group-charging systems have heterogeneous battery and power parameters, making it difficult to describe the load state using a homogeneous aggregation model. Constructing a heterogeneous aggregation model for group-charging loads is crucial. Furthermore, group-charging systems utilize fast charging technology, where charging power decreases as the EV's state of charge increases, resulting in a non-uniform charging rate coefficient in the aggregation model. This characteristic needs to be modeled to ensure the accuracy and stability of the aggregation control model. Finally, after aggregating the load into a stable and controllable system, the control algorithm is critical for adjusting the charging power. However, the regulation tasks such as clean energy consumption are dynamic and fluctuating, requiring precise reduction of tracking errors and rapid adaptation to the fluctuation speed. Therefore, higher requirements are placed on the control accuracy and response speed of the control algorithm.

[0004] In summary, group-charged electric vehicle loads exhibit significant parameter heterogeneity and power attenuation, making it difficult to directly form a unified and controllable aggregated resource. During tracking control, they face challenges such as strong target fluctuations and response differences, impacting control accuracy and responsiveness. Therefore, it is necessary to propose heterogeneous aggregation and improved control methods for group-charged electric vehicle loads. Summary of the Invention

[0005] To address the above problems, this invention proposes a heterogeneous aggregation control method for the load of group-charged electric vehicles.

[0006] The technical solution of the present invention is: a method for heterogeneous aggregation control of load in a group-charged electric vehicle, comprising the following steps:

[0007] S1. Construct the equivalent mapping relationship between the electric vehicle and the charging module it occupies, and obtain the aggregated power;

[0008] S2. Based on the aggregation power, perform heterogeneous aggregation of group-loaded materials to obtain an aggregation model in the form of state-space equations;

[0009] S3. Determine the higher-order sliding mode control law suitable for the aggregation model in the form of state-space equations, and determine the constraint conditions of the control coefficients to obtain the final control law.

[0010] Furthermore, S1 includes the following sub-steps:

[0011] S11. Construct a virtual battery model for the charging module;

[0012] S12. Based on the virtual battery model of the charging module, construct a piecewise linear power-SOC function;

[0013] S13. Based on the piecewise linear power-SOC function, construct the equivalent mapping relationship between the electric vehicle and the occupied charging module, and obtain the aggregated power.

[0014] Furthermore, in S11, the virtual battery model is as follows:

[0015] ;

[0016] In the formula, x(t) represents the battery state of charge (SOC) at time t, u(t) represents the control signal, C represents the rated battery capacity, and P[x(t)] represents the function relating charging power and state of charge.

[0017] In S12, the piecewise linear power-SOC function is:

[0018] ;

[0019] In the formula, x represents the state of charge (SOC), x tsP represents the breakpoint. max Indicates the maximum charging power. Let φ represent the charging power function, φ represent the slope of the first stage, and σ represent the ratio between the charging power at full charge and the maximum power.

[0020] In S13, the equivalent mapping relationship between the electric vehicle and the occupied charging module is as follows:

[0021] ;

[0022] In the formula, P EV C represents the charging power of an electric vehicle. EV Indicates the rated battery capacity of an electric vehicle, x 0,EV P represents the initial state-of-charge parameter of an electric vehicle. CM C represents the charging power of the charging module. CM Indicates the rated battery capacity of the charging module, x 0,CM This indicates the initial state of charge parameters of the charging module;

[0023] In S13, the polymerization power P agg (t) is:

[0024] .

[0025] Furthermore, S2 includes the following sub-steps:

[0026] S21. Based on the generalized convection equation, perform heterogeneous aggregation analysis of group-filled loads;

[0027] S22. Based on the analysis results of heterogeneous polymerization under group loading, the differential equation form of the heterogeneous polymerization model is obtained;

[0028] S23. Based on the aggregation power, the differential equation form of the heterogeneous aggregation model is discretized to obtain the aggregation model in state-space equation form.

[0029] Furthermore, in S21, the generalized convection equation is:

[0030] ;

[0031] In the formula, Q(x,y,t) represents the state of charge x, battery capacity y, and number of loads at time t, ω(x,y,t) represents the random number of loads entering or leaving the charging station, ▽ represents the gradient operator, J(x,y,t) represents the load flow rate through section xy at time t, and v(x,y,t) represents the two-dimensional velocity vector.

[0032] In S22, the differential equation form of the heteropolymerization model is:

[0033] ;

[0034] In the formula, This represents the flow velocity in the x-direction. Let u(t) represent the flow velocity in the y direction, and u(t) represent the control signal.

[0035] In S23, the aggregation model in state-space equation form is:

[0036] ;

[0037] In the formula, u(t) represents the charging power coefficient, A a Let A represent the state transition tensor. b P represents the bias tensor. agg (t) represents the polymerization power. Indicates the rate of change of load quantity. Indicates the load quantity. This indicates a random perturbation. This indicates left multiplication of the output tensor. This indicates right multiplication of the output tensor, where K represents the number of intervals divided by the battery state and R represents the number of intervals divided by the rated battery capacity.

[0038] Furthermore, S3 includes the following sub-steps:

[0039] S31. The difference between the clean energy consumption target and the aggregated power is used as the tracking error, and a sliding surface is constructed based on the tracking error;

[0040] S32. Based on the sliding surface, the higher-order sliding control law suitable for the aggregation model in the form of state-space equations is determined using the sign function, and the constraint conditions of the control coefficients are determined to obtain the final control law.

[0041] Furthermore, in S31, the sliding surface s(t) is:

[0042] ;

[0043] In the formula, e(t) represents the tracking error. γ represents the rate of change of error, and γ represents a constant.

[0044] Furthermore, in S32, the constraint conditions for the control coefficients are:

[0045] ;

[0046] In the formula, η represents the gain coefficient. Let represent the rate of change of error, γ represent a constant, and u(t) represent the control signal. The second derivative of the target being tracked. Indicates control signal, This represents the total load power in an uncontrolled state. and Let represent its first derivative and second derivative, respectively;

[0047] In S32, the final control law is:

[0048] ;

[0049] In the formula, represents a higher-order sliding mode control law, and sat() represents an adjustable saturation function.

[0050] The beneficial effects of this invention are:

[0051] (1) This invention uses a virtual battery model to describe the charging module resources of the group charging system as basic aggregation units. Considering that the charging module of the group charging system is the most basic physical unit and reflects the charging state of the group charging electric vehicle load, it is described as a basic aggregation unit. A virtual battery model is used to represent their charging power and duration, and a battery model parameter mapping relationship is established between the charging module and the electric vehicle it powers.

[0052] (2) This invention addresses the problem of heterogeneous coefficients in the group charging load aggregation model. Based on the generalized convection equation with source terms and velocity gradients, it constructs a heterogeneous aggregation model with two-dimensional state space and tensor coefficients to describe the dynamic evolution of the total load power during the charging process and realize the unified aggregation of a large number of group charging loads.

[0053] (3) This invention regulates the load power of the control group, tracks and absorbs clean energy, proposes an improved sliding mode control algorithm based on a higher-order control law, designs a linear combination of the tracking error and its derivative as the sliding surface, and derives a control law suitable for the proposed aggregation model based on the Lyapunov stability theorem, so as to achieve fast response and smooth adjustment in the regulation process. Attached Figure Description

[0054] Figure 1 A flowchart of a heterogeneous aggregation control method for the load of a group-charged electric vehicle;

[0055] Figure 2 This is a diagram showing the internal structure of the group charging system according to an embodiment of the present invention and its relationship to electric vehicle charging.

[0056] Figure 3 This is a heterogeneous aggregation model diagram of the load of a group-charged electric vehicle according to an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the dimensions of the coefficients in the heterogeneous polymerization model of this invention.

[0058] Figure 5This is a diagram illustrating the improved sliding mode control structure of an embodiment of the present invention.

[0059] Figure 6 This is a schematic diagram of the load power data used in the experimental verification of this invention embodiment;

[0060] Figure 7 This is a schematic diagram of the heteropolymerization results verified by experiments in an embodiment of the present invention;

[0061] Figure 8 This is a schematic diagram illustrating the improved sliding mode control results verified by experiments in an embodiment of the present invention. Detailed Implementation

[0062] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0063] like Figure 1 As shown, the present invention provides a heterogeneous aggregation control method for the load of a group-charged electric vehicle, comprising the following steps:

[0064] S1. Construct the equivalent mapping relationship between the electric vehicle and the charging module it occupies, and obtain the aggregated power;

[0065] S2. Based on the aggregation power, perform heterogeneous aggregation of group-loaded materials to obtain an aggregation model in the form of state-space equations;

[0066] S3. Determine the higher-order sliding mode control law suitable for the aggregation model in the form of state-space equations, and determine the constraint conditions of the control coefficients to obtain the final control law.

[0067] This invention uses a virtual battery model to describe the charging module resources of a group charging system as basic aggregation units. Addressing the heterogeneity of coefficients in the group charging load aggregation model, a heterogeneous aggregation model with a two-dimensional state space and tensor coefficients is constructed based on the generalized convection equation with source terms and velocity gradients. This model describes the dynamic evolution of the total load power during charging, enabling the unified integration of a large number of group charging loads. To provide clean energy consumption services by adjusting charging power, and addressing the control performance degradation caused by fluctuations in consumption targets, an improved sliding mode control algorithm based on a higher-order control law is derived, achieving fast response and smooth adjustment.

[0068] In this embodiment of the invention, S1 includes the following sub-steps:

[0069] S11. Construct a virtual battery model for the charging module;

[0070] S12. Based on the virtual battery model of the charging module, construct a piecewise linear power-SOC function;

[0071] S13. Based on the piecewise linear power-SOC function, construct the equivalent mapping relationship between the electric vehicle and the occupied charging module, and obtain the aggregated power.

[0072] As the most basic physical unit, the charging module of the group charging system reflects the charging state of the group charging load, and is therefore modeled first. Considering that the charging modules in the group charging system have different states, a unified battery model is used to represent their charging power and duration. A function describing the decrease in charging power as the state of charge (SOC) increases is adopted to establish the battery model parameter mapping relationship between the charging module and the electric vehicle it powers.

[0073] Figure 2 The diagram illustrates the internal structure of the group charging system in this invention and its relationship to electric vehicle charging. It shows that each group charging system has multiple charging terminals, capable of charging different electric vehicles. Charging module resources are aggregated to form a power pool, and the output power of the charging terminals is flexibly adjusted via a distribution unit to meet the charging needs of different electric vehicle users.

[0074] In this embodiment of the invention, it is noted that the charging module resource is the smallest physical execution unit for realizing power output and energy conversion. Therefore, when constructing the aggregation model of the charging load of the electric vehicle fleet, it is represented as the basic aggregation modeling object. Due to the differences in the charging needs of connected electric vehicles, the charging modules have different usage quantities and times. To facilitate the description of these characteristics, the charging module is equivalent to a virtual battery model.

[0075] In S11, the virtual battery model is as follows:

[0076] ;

[0077] In the formula, x(t) represents the battery state of charge (SOC) at time t, u(t) represents the control signal, C represents the rated battery capacity, and P[x(t)] represents the function relating charging power and state of charge.

[0078] A piecewise linear power-SOC function is used to accurately model the fast charging power decay. Considering the possibility of power fluctuations in the first stage in reality, coefficients φ and σ are used to represent the slope of the first stage and the ratio between the fully charged charging power and the maximum power, respectively.

[0079] In S12, the piecewise linear power-SOC function is:

[0080] ;

[0081] In the formula, x represents the state of charge (SOC), x ts P represents the breakpoint. max Indicates the maximum charging power. Let φ represent the charging power function, φ represent the slope of the first stage, and σ represent the ratio between the charging power at full charge and the maximum power.

[0082] Assuming that a certain electric vehicle connected to the group charging system occupies n charging modules, the charging power and rated battery capacity parameters between the electric vehicle and the charging modules can be expressed as an n-fold proportional relationship, and the initial state of charge and charging rate parameters can be expressed as an equal relationship, thereby ensuring the consistency of the mapping process.

[0083] In S13, the equivalent mapping relationship between the electric vehicle and the charging module it occupies is as follows:

[0084] ;

[0085] ;

[0086] In the formula, P EV C represents the charging power of an electric vehicle. EV Indicates the rated battery capacity of an electric vehicle, x 0,EV P represents the initial state-of-charge parameter of an electric vehicle. CM C represents the charging power of the charging module. CM Indicates the rated battery capacity of the charging module, x 0,CM This indicates the initial state of charge parameters of the charging module;

[0087] In S13, the polymerization power P agg (t) is:

[0088] .

[0089] In this embodiment of the invention, S2 includes the following sub-steps:

[0090] S21. Based on the generalized convection equation, perform heterogeneous aggregation analysis of group-filled loads;

[0091] S22. Based on the analysis results of heterogeneous polymerization under group loading, the differential equation form of the heterogeneous polymerization model is obtained;

[0092] S23. Based on the aggregation power, the differential equation form of the heterogeneous aggregation model is discretized to obtain the aggregation model in state-space equation form.

[0093] To address the heterogeneity of load parameters and the power reduction issue in the later stages of SOC, a heterogeneous aggregation method for the loads of group-charged electric vehicles is constructed based on the generalized convection equation with source terms and velocity gradients. This method characterizes the heterogeneous load concentration migration characteristics during charging by generalizing the convection mechanism and describes the non-uniformity of charging rate by using the gradient transfer mechanism, thereby uniformly representing the charging dynamics of heterogeneous loads.

[0094] In this embodiment of the invention, the load aggregation cluster changes with the state of charge x and battery capacity y during the charging process. This dynamic evolution process can be described using fluid dynamics equations. Addressing the aggregation challenges caused by heterogeneous aggregation parameters and uneven charging rates, this paper describes the principle of heterogeneous aggregation of group-charged loads in a two-dimensional state space with respect to the state of charge and battery capacity, based on a generalized convection equation containing source terms and velocity gradients in fluid dynamics theory. This characterizes the load differentiation phenomenon caused by heterogeneous battery capacities and the gradient effect caused by uneven charging rates.

[0095] In S21, the generalized convection equation is:

[0096] ;

[0097] In the formula, Q(x,y,t) represents the state of charge x, battery capacity y, and number of loads at time t, ω(x,y,t) represents the random number of loads entering or leaving the charging station, ▽ represents the gradient operator, J(x,y,t) represents the load flow rate through section xy at time t, and v(x,y,t) represents the two-dimensional velocity vector.

[0098] ω(x,y,t) is the source term, representing the random number of loads entering or leaving the charging station; ▽ is the gradient operator, describing the rate of change of the load quantity in the xy direction and the non-uniformity of the charging rate; J(x,y,t) is the load flow rate passing through the cross-section xy at time t, which is the product of the load quantity and the two-dimensional velocity vector v(x,y,t); the velocity v in the x-direction... x This refers to the load charging rate, and the flow velocity v in the y-direction. y A value of 0 indicates that the rated battery capacity will not change during the charging process.

[0099] .

[0100] Polymerization power P agg (t) is the double integral of the number of heterogeneous loads multiplied by the charging power over the xy-plane domain, specifically: Heterogeneous load clusters move dynamically during charging, forming fluid on this plane, such as... Figure 3 As shown.

[0101] Figure 3 This is a schematic diagram of the heterogeneous aggregation of loads in a group-charged electric vehicle according to an embodiment of the present invention. The heterogeneous load cluster moves dynamically during the charging process, forming a fluid on this plane.

[0102] By combining the gradient multiplication rule and the power decay function, and substituting the velocity components and their partial derivatives, the differential equation form of the heterogeneous polymerization model is derived. A bias term related to dP(x) / dx is introduced, the value of which is related to the derivative of the power decay function, to model the power decay characteristics.

[0103] In S22, the differential equation form of the heteropolymerization model is:

[0104] ;

[0105] In the formula, This represents the flow velocity in the x-direction. Let u(t) represent the flow velocity in the y direction, and u(t) represent the control signal.

[0106] In S23, to facilitate simulation and model control, the above aggregate model is discretized with respect to variables x and y to obtain the state-space equation form.

[0107] The results of discretizing variables x and y are as follows:

[0108] ;

[0109] The battery status (x) and rated battery capacity (y) are divided into K and R intervals within their respective ranges, with lengths Δx and Δy, and indices k and r, respectively. k,r (t), ω k,r (t), P(x) k ,y r ) represents the number of loads, random numbers, and charging power for the k-th power state interval and the r-th battery capacity interval.

[0110] The above equation is rearranged into a state-space equation form of an aggregated model, which consists of two parts: state equations and output equations.

[0111] The aggregation model in state-space equation form is:

[0112] ;

[0113] In the formula, u(t) represents the charging power coefficient, A a Let A represent the state transition tensor. b P represents the bias tensor. agg (t) represents the polymerization power. Indicates the rate of change of load quantity. Indicates the load quantity. This indicates a random perturbation. This indicates left multiplication of the output tensor. This indicates right multiplication of the output tensor, where K represents the number of intervals divided by the battery state and R represents the number of intervals divided by the rated battery capacity.

[0114] The state transition tensor is given by A a and A b It consists of two parts, i.e., A=A a +A bThe tensor A in two-dimensional matrix form when the battery capacity y takes the r-th value. a , denoted as A a,r Charging power is a function P(x,y) related to SOC and battery capacity. b This is called the bias tensor, and it relates to the derivative of the power decay function. The tensor A in two-dimensional matrix form is the tensor when the battery capacity y takes the r-th value. b , denoted as A b,r The value of B represents the charging power corresponding to the state of charge k and the battery capacity r, while the value of C is always 1. Specifically:

[0115] ;

[0116] ;

[0117] ;

[0118] .

[0119] In this embodiment of the invention, S3 includes the following sub-steps:

[0120] S31. The difference between the clean energy consumption target and the aggregated power is used as the tracking error, and a sliding surface is constructed based on the tracking error;

[0121] S32. Based on the sliding surface, the higher-order sliding control law suitable for the aggregation model in the form of state-space equations is determined using the sign function, and the constraint conditions of the control coefficients are determined to obtain the final control law.

[0122] The aggregated power of the group-charged electric vehicle load is used to track and absorb clean energy. An improved high-order sliding mode control algorithm is designed, with the linear combination of the tracking error e(t) and its derivative as the sliding surface. Based on the group-charged load aggregation model, a second-order sliding mode control law is derived.

[0123] The tracking error e(t) is defined as follows: the clean energy consumption target is P. des (t) and the aggregate power P of the group charging load agg The difference between s(t) and the sliding surface s(t) is constructed as a linear combination of the tracking error and its derivative. The Lyapunov function V(t) = 1 / 2s is selected. 2 (t);

[0124] To facilitate the derivation of the control law, the aggregated power P of the group-charged load is... agg (t) can be simplified to the following formula:

[0125] .

[0126] In this embodiment of the invention, in S31, the sliding surface s(t) is:

[0127] ;

[0128] In the formula, e(t) represents the tracking error. γ represents the rate of change of error, and γ represents a constant.

[0129] The parameter γ is a constant and is related to the convergence speed. Once the system enters the sliding surface s(t)=0, then de(t) / dt =-γe(t), and the error and its derivative decrease proportionally, ensuring synchronous convergence. Furthermore, the error converges exponentially.

[0130] Based on Lyapunov's stability theorem and the sign function sgn(), a higher-order sliding mode control law d suitable for heterogeneous aggregation models is derived. 2 u(t) / dt 2 =ηsgn[de(t) / dt+γe(t)], and the constraint conditions of the control coefficient. Finally, in order to reduce chattering, the adjustable saturation function sat() is used to replace the sign function sgn().

[0131] According to the stability criterion, the derivative of the Lyapunov function along the system trajectory must satisfy dV(t) / dt < 0 to ensure the asymptotic reachability of the sliding surface. The final derivative result of the Lyapunov function is:

[0132] ;

[0133] By introducing the sign function of s(t) for control law design, an improved sliding mode control algorithm based on a higher-order control law is obtained. Furthermore, the constraint conditions for the gain coefficient η are obtained.

[0134] In this embodiment of the invention, in S32, the constraint condition for the control coefficient is:

[0135] ;

[0136] In the formula, η represents the gain coefficient. Let represent the rate of change of error, γ represent a constant, and u(t) represent the control signal. The second derivative of the target being tracked. Indicates control signal, This represents the total load power in an uncontrolled state. and Let represent its first derivative and second derivative, respectively;

[0137] Substituting the control law and the constraints of η into the derivative of the Lyapunov function, the result is less than 0, thus proving the convergence of the improved sliding mode control algorithm.

[0138] To reduce chattering, an adjustable saturation function sat() is usually used to replace the sign function, with ε as the error boundary layer, to obtain the final control law.

[0139] In S32, the final control law is:

[0140] ;

[0141] In the formula, represents a higher-order sliding mode control law, and sat() represents an adjustable saturation function.

[0142] Figure 5 To improve the sliding mode control structure in this embodiment of the invention, the controller receives the error signal e(t), and calculates the charging power control signal u(t) after passing through a series of control modules such as gain, differentiator, summer, saturation function, and integrator.

[0143] Figure 6 To verify the load power data used in the experimental design of this invention, 120 group-charged electric vehicles of 6 different types were selected. The battery capacity and charging power parameters of each type of vehicle are different, and the charging power data are from a publicly available dataset.

[0144] Figure 7 The heteropolymerization results verified by the experiments in the embodiments of the present invention show an average relative polymerization error of 0.57%, which verifies the polymerization accuracy of the present invention.

[0145] Figure 8 The improved sliding mode control results, verified experimentally in this embodiment of the invention, are compared with those of the traditional sliding mode control algorithm. In comparison, the range of tracking error is reduced by approximately 85%, demonstrating the advantages of this invention.

[0146] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A method for heterogeneous aggregation control of load in a group-charged electric vehicle, characterized in that, Includes the following steps: S1. Construct the equivalent mapping relationship between the electric vehicle and the charging module it occupies, and obtain the aggregated power; S2. Based on the aggregation power, perform heterogeneous aggregation of group-loaded materials to obtain an aggregation model in the form of state-space equations; S3. Determine the higher-order sliding mode control law suitable for the aggregation model in the form of state-space equations, and determine the constraint conditions of the control coefficients to obtain the final control law.

2. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 1, characterized in that, S1 includes the following sub-steps: S11. Construct a virtual battery model for the charging module; S12. Based on the virtual battery model of the charging module, construct a piecewise linear power-SOC function; S13. Based on the piecewise linear power-SOC function, construct the equivalent mapping relationship between the electric vehicle and the occupied charging module, and obtain the aggregated power.

3. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 2, characterized in that, In S11, the virtual battery model is as follows: ； In the formula, x(t) represents the battery state of charge (SOC) at time t, u(t) represents the control signal, C represents the rated battery capacity, and P[x(t)] represents the function relating charging power and state of charge. In S12, the piecewise linear power-SOC function is: ； In the formula, x represents the state of charge (SOC), x ts P represents the breakpoint. max Indicates the maximum charging power. Let φ represent the charging power function, φ represent the slope of the first stage, and σ represent the ratio between the charging power at full charge and the maximum power. In step S13, the equivalent mapping relationship between the electric vehicle and the occupied charging module is as follows: ； In the formula, P EV C represents the charging power of an electric vehicle. EV Indicates the rated battery capacity of an electric vehicle, x 0,EV P represents the initial state-of-charge parameter of an electric vehicle. CM C represents the charging power of the charging module. CM Indicates the rated battery capacity of the charging module, x 0,CM This indicates the initial state of charge parameters of the charging module; In S13, the polymerization power P agg (t) is: 。 4. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 1, characterized in that, S2 includes the following sub-steps: S21. Based on the generalized convection equation, perform heterogeneous aggregation analysis of group-filled loads; S22. Based on the analysis results of heterogeneous polymerization under group loading, the differential equation form of the heterogeneous polymerization model is obtained; S23. Based on the aggregation power, the differential equation form of the heterogeneous aggregation model is discretized to obtain the aggregation model in state-space equation form.

5. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 4, characterized in that, In S21, the generalized convection equation is: ； In the formula, Q(x,y,t) represents the state of charge x, battery capacity y, and number of loads at time t, ω(x,y,t) represents the random number of loads entering or leaving the charging station, ▽ represents the gradient operator, J(x,y,t) represents the load flow rate through section xy at time t, and v(x,y,t) represents the two-dimensional velocity vector. In S22, the differential equation form of the heteropolymerization model is: ； In the formula, This represents the flow velocity in the x-direction. Let u(t) represent the flow velocity in the y direction, and u(t) represent the control signal. In S23, the aggregation model in the form of state-space equations is: ； In the formula, u(t) represents the charging power coefficient, A a Let A represent the state transition tensor. b P represents the bias tensor. agg (t) represents the polymerization power. Indicates the rate of change of load quantity. Indicates the load quantity. This indicates a random perturbation. This indicates left multiplication of the output tensor. This indicates right multiplication of the output tensor, where K represents the number of intervals divided by the battery state and R represents the number of intervals divided by the rated battery capacity.

6. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 1, characterized in that, S3 includes the following sub-steps: S31. The difference between the clean energy consumption target and the aggregated power is used as the tracking error, and a sliding surface is constructed based on the tracking error; S32. Based on the sliding surface, the higher-order sliding control law suitable for the aggregation model in the form of state-space equations is determined using the sign function, and the constraint conditions of the control coefficients are determined to obtain the final control law.

7. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 6, characterized in that, In S31, the sliding surface s(t) is: ； In the formula, e(t) represents the tracking error. γ represents the rate of change of error, and γ represents a constant.

8. The heterogeneous aggregation control method for the load of a group-charged electric vehicle according to claim 6, characterized in that, In S32, the constraint condition for the control coefficient is: ； In the formula, η represents the gain coefficient. Let represent the rate of change of error, γ represent a constant, and u(t) represent the control signal. The second derivative of the target being tracked. Indicates control signal, This represents the total load power in an uncontrolled state. and Let represent its first derivative and second derivative, respectively; In S32, the final control law is: ； In the formula, represents a higher-order sliding mode control law, and sat() represents an adjustable saturation function.

Images