Discrete sliding mode control method and device for voltage outer ring of energy storage converter and medium

Abstract

The invention relates to a discrete sliding mode control method and device for a voltage outer ring of an energy storage converter and a medium, and the method comprises the following steps: obtaining related operation data of a super-capacitor energy storage system, and building a sampling data model of the super-capacitor energy storage system; designing a sliding mode surface function for a converter voltage outer loop control subsystem of the super-capacitor energy storage system; based on the sampling data model of the super-capacitor energy storage system, considering lumped disturbance of the system, and establishing a tight-format dynamic linearization data model of a converter voltage outer loop control subsystem; and based on the sliding mode surface function and the tight-format dynamic linearization data model, constructing a discrete sliding mode controller based on data driving, estimating lumped disturbance by adopting a delay estimation technology and compensating the lumped disturbance into the discrete sliding mode controller, and outputting a converter control voltage signal by the discrete sliding mode controller to realize a control process. Compared with the prior art, the method has the advantages of stabilizing the DC bus voltage, improving the transient characteristics of the system and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage converter control, and in particular to a discrete sliding mode control method, device and medium for a voltage outer loop of an energy storage converter. Background Art

[0002] Sustainable production of renewable energy is an effective means for countries to cope with resource scarcity and environmental pollution. However, the randomness and intermittency of renewable energy production will affect the stable operation of DC power grids. Usually, additional high-power fast energy storage devices are required to alleviate the impact of renewable energy on grid voltage. Energy storage converters are the hub for energy flow between high-voltage grids and low-voltage energy storage devices. Therefore, it is extremely necessary to ensure the rapidity of energy exchange between the grid and the energy storage device and maintain the stability of the grid voltage, and further improve the control performance of DC converters.

[0003] Patent application CN117977964A discloses an improved high-order sliding mode control method for a three-phase staggered parallel bidirectional DC-DC converter. This method combines a two-stage finite time extended state observer with fast convergence to improve the estimation accuracy of the system's lumped disturbance. Secondly, a dual closed-loop control system with inner loop differential flat control and outer loop improved super-helical sliding mode control is adopted. This can not only improve the system's dynamic response process, but also use a high-order sliding mode control algorithm to suppress jitter, while solving the non-minimum phase problem in the converter's boost mode; it can resist disturbances well and improve the system's transient performance. However, this method does not consider the adverse effects of unmodeled modeling due to modeling errors and unmodeled dynamics.

[0004] When the supercapacitor energy storage system is subject to external disturbances and circuit parameter perturbations, the PI control based on linear error feedback cannot effectively guarantee the stability of the DC bus voltage. In the supercapacitor energy storage system, the presence of nonlinear elements (such as switching tubes) and the complex and variable external operating conditions make it very difficult to accurately model the system. Designing a controller that relies on an accurate mathematical model usually requires simplifying the system structure and working process, which may lead to differences between the model and the actual system behavior. Summary of the invention

[0005] The purpose of the present invention is to provide a discrete sliding mode control method, device and medium for the voltage outer loop of an energy storage converter that can effectively suppress DC bus voltage fluctuations.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A discrete sliding mode control method for a voltage outer loop of an energy storage converter, characterized in that it comprises the following steps:

[0008] Obtain relevant operating data of the supercapacitor energy storage system and establish a sampling data model of the supercapacitor energy storage system;

[0009] Design sliding surface function for converter voltage outer loop control subsystem of supercapacitor energy storage system;

[0010] Based on the supercapacitor energy storage system sampling data model, considering the lumped disturbance of the system, a compact dynamic linearization data model of the converter voltage outer loop control subsystem is established;

[0011] Based on the sliding surface function and the compact dynamic linearization data model, a data-driven discrete sliding mode controller is constructed, and the delay estimation technology is used to estimate the lumped disturbance and compensate it in the discrete sliding mode controller. The discrete sliding mode controller outputs the converter control voltage signal to realize the control process.

[0012] Furthermore, the step of establishing a supercapacitor energy storage system sampling data model includes:

[0013] Select the inductor current i L With DC bus voltage u dc is the state variable, the voltage at the supercapacitor terminal As input variables, the state equation is established according to the state space averaging method:

[0014]

[0015] In the formula, is the first-order derivative of the inductor current, is the first-order derivative of the DC bus voltage, is the voltage across the equivalent internal resistance of the supercapacitor, u dc 、u sc , L, C0 are DC bus voltage, supercapacitor terminal voltage, energy storage inductor, and support capacitor respectively; R eq , R es , C sc are the equivalent load resistance, the equivalent internal resistance of the supercapacitor, and the supercapacitor value respectively; d is the duty cycle of the switch tube;

[0016] Define the sampling time T s , considering the first-order Euler method, the state equation is discretized to obtain the sampling data model of the supercapacitor energy storage system working in Boost mode:

[0017]

[0018] In the formula, i L (k), i L (k+1) is the inductor current at time k and k+1, d(k) is the switch duty cycle at time k, u dc (k) and u dc(k+1) is the DC bus voltage at time k and k+1, u c (k) is the voltage at the supercapacitor terminal at time k;

[0019] Define the sampling period T s , in each kT s The DC bus voltage and inductor current are sampled at the sampling time, that is: u dc (k) = u dc (kT s ),i L (k) = i L (kT s ).

[0020] Furthermore, the expression of the sliding surface function is:

[0021] s(k)=c1e(k)+c2e(k-1)

[0022] Where s(k) is the sliding surface function, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) is the DC bus voltage reference at time k, u dc (k) is the DC bus voltage sampling value at time k.

[0023] Furthermore, when the supercapacitor energy storage system enters the ideal sliding mode, the following equation holds true:

[0024] s(k)=s(k+1)

[0025] =c1e(k+1)+c2e(k)

[0026] =c1(u ref (k+1)-u dc (k+1))+c2e(k)

[0027] Where s(k) is the sliding surface function, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) and u ref (k+1) is the DC bus voltage reference at time k and k+1, u dc (k) and u dc (k+1) is the DC bus voltage sampling value at time k and k+1.

[0028] Furthermore, the steps of establishing the compact format dynamic linearized data model include:

[0029] According to the sampling data model of the supercapacitor energy storage system, combined with the model-free adaptive control theory, the voltage outer loop control is described as L (k) and control output u dc SISO form of (k):

[0030] u dc (k+1)=f((u dc (k),…,u dc (kn y ),i L (k),…i L (kn u ))+f NL (k)

[0031] In the formula, u dc (k) and u dc (k+1) is the DC bus voltage sampling value at time k and k+1, i L (k) is the sampling value of the inductor current at sampling time k, n y 、n u are the order of the system input and output, f(·) is an unknown nonlinear function, f NL (k) is the aggregate disturbance at sampling time k;

[0032] According to the condition that the converter voltage outer loop control subsystem satisfies the controllable and observable input and output conditions and meets the generalized Lipschitz condition, a compact dynamic linearized data model of the converter voltage outer loop control subsystem is established, and the expression is:

[0033] Δu dc (k+1)=φ(k)Δi L (k)+Δf NL (k)

[0034] In the formula, Δu dc (k+1)=u dc (k+1)-u dc (k), Δu dc (k+1) is the change of DC bus voltage between two adjacent sampling moments, φ(k) is the pseudo partial derivative, and |φ(k)|<b, b>0, Δi L (k) = i L (k)-i L (k-1), Δi L (k) is the change of inductor current between two adjacent sampling moments, i L (k-1) is the inductor current sampling value at the k-1 sampling time, ΔfNL (k) = f NL (k)-f NL (k-1), Δd(k) is the change of the lumped disturbance between two adjacent sampling moments, f NL (k-1) is the lumped disturbance at the k-1 sampling time.

[0035] Furthermore, the construction process of the data-driven discrete sliding mode controller includes:

[0036] Based on the sliding surface function and the compact dynamic linearization data model, an equivalent control law is constructed, which is expressed as follows:

[0037]

[0038] In the formula, Δi L eq (k) is the equivalent control law, φ(k) is the pseudo partial derivative, and |φ(k)|<b, b>0, δ is the additional correction term, and δ>0, u dc (k) is the DC bus voltage sampling value at time k, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) and u ref (k+1) is the DC bus voltage reference at time k and k+1, u dc (k) is the DC bus voltage sampling value at time k, s(k) is the sliding surface function, Δf NL (k) = f NL (k)-f NL (k-1), Δf NL (k) is the difference of disturbances at adjacent sampling times, f NL (k-1) is the aggregate disturbance at sampling time k-1;

[0039] Construct the switching control law, the expression is:

[0040]

[0041] In the formula, Δi L sw(k) is the switching control law, λ is the switching gain, and sgn is the sign function;

[0042] Combining the equivalent control law and the switching control law, a data-driven discrete sliding mode controller is constructed, and the expression is:

[0043]

[0044] In the formula, Δi L(k) is a discrete sliding mode controller.

[0045] Furthermore, the pseudo partial derivative φ(k) is an unknown time-varying parameter variable, which is estimated based on the input / output sampling data. The specific estimation process includes:

[0046] Define the pseudo partial derivative estimation criterion function:

[0047] J(φ(k))=|u dc (k)-u dc (k-1)-φ(k)Δi L (k-1)-Δf NL (k)| 2 +|μ(φ(k)-φ(k-1))| 2

[0048] Where J(φ(k)) is the pseudo partial derivative estimation criterion function, u dc (k-1) is the DC bus voltage sampling value at time k-1, Δi L (k-1) is the change of the inductor current between two adjacent sampling moments, Δf NL (k) is the change of the lumped disturbance between two adjacent sampling moments, μ is the weight factor, and μ>0.

[0049] Furthermore, the step of estimating the lumped disturbance comprises:

[0050] make:

[0051]

[0052] In the formula, is Δf NL The estimated value of (k), Δf NL (k-1) is Δf NL (k) The delay of one sampling moment;

[0053] The compact format dynamic linearized data model is delayed by one step to obtain:

[0054] Δf NL (k-1) = Δu dc (k)-φ(k-1)Δi L (k-1)

[0055] In the formula, Δu dc (k) is the change of DC bus voltage between two adjacent sampling moments, φ(k-1) is the pseudo partial derivative at moment k-1, Δi L (k-1) is the change of inductor current between two adjacent sampling moments;

[0056] Combining the above equations, we can get the estimated value of the lumped disturbance:

[0057]

[0058] In the formula, is Δf NL The estimated value of (k), Δf NL (k) is the change of the lumped disturbance between two adjacent sampling moments, u dc (k) and u dc (k-1) are the DC bus voltage sampling values ​​at time k and k-1 respectively.

[0059] The present invention also provides an electronic device, comprising: one or more processors; a memory; and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the discrete sliding mode control method of the voltage outer loop of the energy storage converter as described above.

[0060] The present invention also provides a computer-readable storage medium, comprising one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the discrete sliding mode control method of the voltage outer loop of the energy storage converter as described above.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] (1) The present invention first establishes a compact dynamic linearization data model of the voltage outer loop, and then designs a discrete sliding mode controller based on the data model. The delay estimation technology is used to estimate and compensate for the lumped disturbance in the data model, thereby further improving the accuracy of the control algorithm update. The problem of DC bus voltage instability caused by the deterioration of control performance when the energy storage converter is affected by external disturbances and unmodeled dynamics is effectively solved, thereby improving the transient characteristics of the system.

[0063] (2) The data-driven control method of the present invention only uses the input and output measurement data of the system to design the controller, which can get rid of the dependence on the model of the controlled system and effectively avoid the adverse effects caused by modeling errors and unmodeled dynamics. In addition, sliding mode control is essentially a switching type of control. The combination of the two is very suitable for solving the control problems of supercapacitor energy storage systems (which have strong nonlinear characteristics, are difficult to accurately model, and have structures that change with the switching state).

[0064] (3) The present invention can not only effectively reduce the DC bus voltage fluctuation, but also suppress the influence of external disturbance and system parameter mismatch through disturbance compensation, showing stronger robustness.

[0065] (4) The present invention avoids the need to establish a complex model for the supercapacitor energy storage system, has strong portability, and is applicable to different types of converter control systems.

[0066] (5) The switching characteristics of the sliding mode control method of the present invention are more suitable for supercapacitor energy storage systems whose structures change with the switching state. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0068] Figure 2 This is the structure diagram of the supercapacitor energy storage system;

[0069] Figure 3 This is the equivalent Buck mode diagram;

[0070] Figure 4 is the equivalent Boost pattern diagram;

[0071] Figure 5 A comparison diagram of the DC bus voltage under output voltage disturbance between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0072] Figure 6 A comparison diagram of the DC bus voltage under load disturbances between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0073] Figure 7 A comparison diagram of the DC bus voltage under parameter mismatch disturbance between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0074] Figure 8 A comparison diagram of the total current of the energy storage inductor between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0075] Fig. 9 This is a comparison diagram of the voltage at the supercapacitor terminal between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0076] Fig.10 A comparison diagram of the charge state of supercapacitors between the method of the present invention and the traditional PI control method and the linear active disturbance rejection control method;

[0077] Fig.11 A comparison diagram of the DC bus voltage under output voltage disturbance between the method of the present invention and the traditional discrete sliding mode control method;

[0078] Fig.12 A comparison diagram of the DC bus voltage under load disturbance between the method of the present invention and the traditional discrete sliding mode control method;

[0079] Fig.13 This is a comparison diagram of the DC bus voltage under parameter mismatch disturbance between the method of the present invention and the traditional discrete sliding mode control method. DETAILED DESCRIPTION

[0080] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0081] The structure of supercapacitor energy storage system is as follows: Figure 2 As shown, the DC bus side is the power input end of the system, the load is the power output end of the system, the three interleaved parallel DC / DC converters are responsible for adjusting the flow direction and size of the power, and the supercapacitor is used for the storage and release of power.

[0082] The three-phase interleaved parallel DC / DC converter is essentially composed of three single-channel bidirectional DC / DC converters interleaved in parallel. Its operating principle is basically the same as that of the bidirectional DC / DC converter. Therefore, when analyzing its mathematical model, it can be equivalent to a single bidirectional DC / DC converter. Figure 3 and Figure 4 As shown, S1 and S2 are power switch tubes, D1 and D2 are anti-parallel diodes; u dc 、u sc , L, C0 are DC bus voltage, supercapacitor terminal voltage, energy storage inductor, and support capacitor respectively; R eq , R es , C sc are the equivalent load resistance, the equivalent internal resistance of the supercapacitor, and the supercapacitor value, respectively. The on-duty cycle of the switch tube S1 is defined as d∈[0,1], and the on-duty cycle of the switch tube S2 is complementary to that of S1.

[0083] For Figure 2 The circuit topology shown applies data-driven discrete sliding mode control, which requires the establishment of a supercapacitor energy storage system sampling data model and the sampling of the DC bus voltage and inductor current at each sampling moment. Therefore, the embodiment of the present invention provides a discrete sliding mode control method for the voltage outer loop of the energy storage converter, which aims to solve the problem of DC bus voltage instability caused by the deterioration of control performance when the energy storage converter is affected by external disturbances and unmodeled dynamics. Figure 1 As shown, the method comprises the following steps:

[0084] Step 1: Establish a sampling data model for the supercapacitor energy storage system and sample the DC bus voltage and inductor current at each sampling moment.

[0085] The switch tubes S1 and S2 are turned on complementary, and the energy storage converter is taken as an example to work in Boost mode. When S1 is turned off, S2 is turned on, and D1 and D2 are turned off, the differential equation is established according to the circuit theorem:

[0086]

[0087] When S1 and S2 are turned off, D2 is cut off, and D1 is turned on, the differential equation is established according to the circuit theorem:

[0088]

[0089] Select the inductor current i L With DC bus voltage u dc is the state variable, the supercapacitor terminal voltage u c =u sc +u Res is the input variable, and the state equation is established according to the state space averaging method by combining equations (1) and (2):

[0090]

[0091] In the formula, u dc 、u sc , L, C0 are DC bus voltage, supercapacitor terminal voltage, energy storage inductor, and support capacitor respectively; R eq , R es , C sc are the equivalent load resistance, the equivalent internal resistance of the supercapacitor, and the capacitance of the supercapacitor respectively; d is the duty cycle of the switch tube.

[0092] Define the sampling time T s , considering the first-order Euler method, discretizing equation (3), we can get the sampling data model of the converter working in Boost mode:

[0093]

[0094] Through the discretization analysis and processing of the converter mathematical model, the system sampling data model can be decomposed into a series connection of a voltage outer loop first-order system and a current inner loop first-order system.

[0095] Define the sampling period T s , in each kT s The DC bus voltage and inductor current are sampled at the sampling time, that is: u dc (k) = u dc (kT s ),i L (k) = i L (kT s ).

[0096] Step 2: Design a sliding surface function for the transformer voltage outer loop control subsystem of the supercapacitor energy storage system.

[0097] Specifically: Design sliding surface function:

[0098] s(k)=c1e(k)+c2e(k-1) (6)

[0099] Where c1 and c2 are adjustable parameters, and c1>0, c2>0; e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) is the DC bus voltage reference at time k; u dc (k) is the DC bus voltage sampling value at time k.

[0100] When the system enters the ideal sliding mode, we have:

[0101]

[0102] Step 3: Considering the lumped disturbance of the system, a compact dynamic linearization data model of the transformer voltage outer loop control subsystem is established.

[0103] Specifically: Based on the model-free adaptive control theory, the voltage outer loop control system can be described as the control input i L (k) and control output u dc SISO form of (k):

[0104] u dc (k+1)=f((u dc (k),…,u dc (kn y ),i L (k),…i L (kn u ))+f NL (k) (8)

[0105] In the formula, i L (k) is the sampling value of the inductor current at sampling time k; u dc (k+1) is the DC bus voltage sampling value at time k+1; n y 、n u are the order of the system input and output respectively; f(·) is an unknown nonlinear function; f NL (k) is the lumped disturbance.

[0106] The voltage outer loop control system meets the conditions that the input and output are controllable and observable, and meets the generalized Lipschitz condition. Therefore, the compact dynamic linearization data model of the voltage outer loop control system is established as follows:

[0107] Δu dc (k+1)=φ(k)Δi L (k)+Δf NL (k) (9)

[0108] In the formula, Δudc (k+1)=u dc (k+1)-u dc (k); φ(k) is a pseudo partial derivative, and φ(k)|<b, b>0; Δi L (k) = i L (k)-i L (k-1), |Δi L (k)|≠0,|i L (k)|<c,c>0;Δf NL (k) = f NL (k)-f NL (k-1).

[0109] Step 4: Design a data-driven discrete sliding mode controller.

[0110] Specifically, by combining equation (7) and equation (9), the equivalent control law is as follows:

[0111]

[0112] In the formula, Δi L eq (k) = i L eq (k)-i L eq (k-1); δ is an additional correction term, and δ>0.

[0113] In order to realize the robust control of the system to the lumped disturbance, the switching control law is designed as follows:

[0114]

[0115] Where λ is the switching gain and sgn is the sign function.

[0116] Combining equations (10) and (11), the complete form of the data-driven discrete sliding mode controller is as follows:

[0117]

[0118] In formula (12), φ(k) is an unknown time-varying parameter variable, which needs to be estimated based on the input / output sampling data.

[0119] Define the pseudo partial derivative estimation criterion function:

[0120] J(φ(k))=|u dc (k)-u dc (k-1)-φ(k)Δi L (k-1)-Δf NL (k)| 2+|μ(φ(k)-φ(k-1))| 2 (13)

[0121] From equations (9) and (13), we can get the estimation algorithm of pseudo partial derivative:

[0122]

[0123] Where, μ is the weight factor, and μ>0; η∈(0,1] is the step coefficient; is the estimated value of φ(k); yes The initial value of Update condition: If or |Δi L (k-1)|<σ, then

[0124] Step 5: Use the delay estimation technology to estimate the lumped disturbance and compensate it in the controller.

[0125] Specifically, for the discrete nonlinear system in equation (9), if the sampling time is short, Δf in the equation NL (k) can be estimated using the following formula:

[0126]

[0127] In the formula, is Δf NL (k) is the estimated value.

[0128] Delaying equation (9) by one sampling time, we can get:

[0129] Δf NL (k-1) = Δu dc (k)-φ(k-1)Δi L (k-1) (16)

[0130] Combining equation (15) and equation (16), we can get Δf NL Estimated value of (k) as follows:

[0131]

[0132] The estimated value of Δd(k) in the above formula Substituting into equation (12), we can get the unknown term Δf in the controller: NL (k) Estimated compensation.

[0133] A discrete sliding mode control method for a voltage outer loop of an energy storage converter disclosed in an embodiment of the present invention can effectively suppress DC bus voltage fluctuations, improve the dynamic characteristics of the system, and enhance the robustness of the system.

[0134] The discrete sliding mode control (DD-DSMC) method of supercapacitor energy storage converter provided by the present invention is simulated and verified in MATLAB / SIMULINK, and compared with the traditional PI control method, linear active disturbance rejection control (LADRC) method and traditional discrete sliding mode control (DSMC) method. The system model parameters are as follows: the bus reference voltage is 1500V, the load equivalent resistance is 20Ω, the supercapacitor internal resistance is 1.8×10 -4 Ω, the supercapacitor capacitance is 46F, and the support capacitance is 5.8×10 -3 F, the sampling period is 1×10 -5 s, the energy storage inductance of each phase is 2.5×10 -3 L.

[0135] Experiment 1: Input voltage periodic disturbance simulation experiment. At t=1, the input voltage suddenly increases by 80V, and at t=2, the input voltage suddenly drops by 90V. The DC bus voltage change waveform is as follows: Figure 5 As shown in the figure, in the initial stage, PI control presents a large voltage pulse, while DD-DSMC has basically no voltage pulse; when the input voltage suddenly changes, DD-DSMC exhibits smaller voltage fluctuation and faster adjustment time.

[0136] Experiment 2: Load resistance step disturbance simulation experiment. At t=3, the load resistance suddenly increases by 20Ω, and at t=4, the load resistance suddenly decreases by 10Ω. Figure 6 It can be seen that the comprehensive performance of DD-DSMC is superior to that of PI and LADRC.

[0137] Experiment 3: The impact of circuit parameter mismatch disturbance on the system is simulated by a sudden change in the inductance value of each circuit of the three-phase staggered parallel DC / DC converter. The initial inductance value is 1mH, which suddenly drops to 0.5mH at t=5.5 and suddenly increases to 1.5mH at t=6.5. Figure 7 As shown in Figure 2, when the inductance value changes suddenly, the DC bus voltage fluctuates significantly. However, when DD-DSMC is used, the DC bus voltage fluctuates less and recovers faster.

[0138] Figure 8 and Fig. 9 It reflects the energy exchange process of the energy storage system. The flow direction of the inductor current of the converter in Boost mode is defined as the positive direction. As can be seen from the figure, at t=1, the input voltage suddenly increases, the inductor current flows negatively, the supercapacitor stores energy, and its terminal voltage increases. At t=2, the input voltage suddenly drops, the inductor current flows positively, the supercapacitor releases energy, and its terminal voltage decreases. Similarly, when the load resistance changes suddenly, the flow direction of the inductor current corresponds to the change of the supercapacitor terminal voltage.

[0139] Fig.10 It is the supercapacitor state of charge (SOC) change curve. It can be seen that under DD-DSMC, the supercapacitor SOC changes the least and is more energy-efficient.

[0140] The simulation results of the control effect comparison between DD-DSMC and DSMC are shown in the figure. Fig.11 , Fig.12 and Fig.13 As shown in the figure, it can be seen that in the initial stage, DSMC has a small voltage pulse; when the input voltage, load resistance and inductance value suddenly change, the voltage fluctuation of DD-DSMC is smaller, showing better robustness, but the adjustment time does not change much.

[0141] In a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0142] Example 2

[0143] This embodiment provides an electronic device, including: one or more processors; a memory; and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the discrete sliding mode control method of the voltage outer loop of the energy storage converter as described above.

[0144] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0145] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A discrete sliding mode control method for a voltage outer loop of an energy storage converter, characterized in that: The following steps are involved: Obtain relevant operating data of the supercapacitor energy storage system and establish a sampling data model of the supercapacitor energy storage system; Design sliding surface function for converter voltage outer loop control subsystem of supercapacitor energy storage system; Based on the supercapacitor energy storage system sampling data model, considering the lumped disturbance of the system, a compact dynamic linearization data model of the converter voltage outer loop control subsystem is established; Based on the sliding surface function and the compact dynamic linearization data model, a data-driven discrete sliding mode controller is constructed, and the delay estimation technology is used to estimate the lumped disturbance and compensate it in the discrete sliding mode controller. The discrete sliding mode controller outputs the converter control voltage signal to realize the control process.

2. A discrete sliding mode control method for a voltage outer loop of an energy storage converter according to claim 1, characterized in that: The step of establishing a supercapacitor energy storage system sampling data model comprises: Select the inductor current i L With DC bus voltage u dc is the state variable, the voltage at the supercapacitor terminal As input variables, the state equation is established according to the state space averaging method: In the formula, is the first-order derivative of the inductor current, is the first-order derivative of the DC bus voltage, is the voltage across the equivalent internal resistance of the supercapacitor, u dc 、u sc , L, C0 are DC bus voltage, supercapacitor terminal voltage, energy storage inductor, and support capacitor respectively; R eq , R es , C sc are the equivalent load resistance, the equivalent internal resistance of the supercapacitor, and the supercapacitor value respectively; d is the duty cycle of the switch tube; Define the sampling time T s , considering the first-order Euler method, the state equation is discretized to obtain the sampling data model of the supercapacitor energy storage system working in Boost mode: In the formula, i L (k), i L (k+1) is the inductor current at time k and k+1, d(k) is the switch duty cycle at time k, u dc (k),u dc (k+1) is the DC bus voltage at time k and k+1, u c (k) is the voltage at the supercapacitor terminal at time k; Define the sampling period T s , in each kT s The DC bus voltage and inductor current are sampled at the sampling time, that is: u dc (k) = u dc (kT s ),i L (k) = i L (kT s ).

3. The discrete sliding mode control method of the voltage outer loop of the energy storage converter according to claim 1 is characterized in that: The expression of the sliding surface function is: s(k)=c1e(k)+c2e(k-1) Where s(k) is the sliding surface function, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) is the DC bus voltage reference at time k, u dc (k) is the DC bus voltage sampling value at time k.

4. A discrete sliding mode control method for a voltage outer loop of an energy storage converter according to claim 3, characterized in that: When the supercapacitor energy storage system enters the ideal sliding mode, the following equation holds true: s(k)=s(k+1) =c1e(k+1)+c2e(k) =c1(u ref (k+1)-u dc (k+1))+c2e(k) Where s(k) is the sliding surface function, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k),u ref (k+1) is the DC bus voltage reference at time k and k+1, u dc (k),u dc (k+1) is the DC bus voltage sampling value at time k and k+1.

5. The discrete sliding mode control method of the voltage outer loop of the energy storage converter according to claim 1, characterized in that: The steps of establishing the compact format dynamic linearized data model include: According to the sampling data model of the supercapacitor energy storage system, combined with the model-free adaptive control theory, the voltage outer loop control is described as L (k) and the SISO form of the control output udc(k): u dc (k+1)=f((u dc (k),…,u dc (k-n y ),i L (k),…i L (k-n u ))+f NL (k) In the formula, u dc (k) and u dc (k+1) is the DC bus voltage sampling value at time k and k+1, i L (k) is the inductor current sampling value at sampling time k, n y 、n u are the order of the system input and output, f(·) is an unknown nonlinear function, f NL (k) is the aggregate disturbance at sampling time k; According to the condition that the converter voltage outer loop control subsystem satisfies the controllable and observable input and output conditions and meets the generalized Lipschitz condition, a compact dynamic linearized data model of the converter voltage outer loop control subsystem is established, and the expression is: Δu dc (k+1)=φ(k)Δi L (k)+Δf NL (k) In the formula, Δu dc (k+1)=u dc (k+1)-u dc (k), Δu dc (k+1) is the change of DC bus voltage between two adjacent sampling moments, φ(k) is the pseudo partial derivative, and φ(k)<b, b>0, Δi L (k) = i L (k)-i L (k-1), Δi L (k) is the change of inductor current between two adjacent sampling moments, i L (k-1) is the inductor current sampling value at the k-1 sampling time, Δf NL (k) = f NL (k)-f NL (k-1), Δd(k) is the change of the lumped disturbance between two adjacent sampling moments, f NL (k-1) is the lumped disturbance at the k-1 sampling time.

6. The discrete sliding mode control method of the voltage outer loop of the energy storage converter according to claim 1, characterized in that: The construction process of the data-driven discrete sliding mode controller includes: Based on the sliding surface function and the compact dynamic linearization data model, an equivalent control law is constructed, which is expressed as follows: In the formula, Δi L eq (k) is the equivalent control law, φ(k) is the pseudo partial derivative, and φ(k)|<b, b>0, δ is the additional correction term, and δ>0, u dc (k) is the DC bus voltage sampling value at time k, c1 and c2 are adjustable parameters, and c1>0, c2>0, e(k) is the DC bus voltage tracking error at time k, and e(k)=u ref (k)-u dc (k),u ref (k) and u ref (k+1) is the DC bus voltage reference at time k and k+1, u dc (k) is the DC bus voltage sampling value at time k, s(k) is the sliding surface function, Δf NL (k) = f NL (k)-f NL (k-1), Δf NL (k) is the difference of disturbances at adjacent sampling times, f NL (k-1) is the aggregate disturbance at sampling time k-1; Construct the switching control law, the expression is: In the formula, Δi L sw(k) is the switching control law, λ is the switching gain, and sgn is the sign function; Combining the equivalent control law and the switching control law, a data-driven discrete sliding mode controller is constructed, and the expression is: In the formula, Δi L (k) is a discrete sliding mode controller.

7. A discrete sliding mode control method for a voltage outer loop of an energy storage converter according to claim 6, characterized in that: The pseudo partial derivative φ(k) is an unknown time-varying parameter variable, which is estimated based on the input / output sampling data. The specific estimation process includes: Define the pseudo partial derivative estimation criterion function: J(φ(k))=|u dc (k)-u dc (k-1)-φ(k)Δi L (k-1)-Δf NL (k) 2 +μ(φ(k)-φ(k-1)) 2 Where J(φ(k)) is the pseudo partial derivative estimation criterion function, u dc (k-1) is the DC bus voltage sampling value at time k-1, Δi L (k-1) is the change of the inductor current between two adjacent sampling moments, Δf NL (k) is the change of the lumped disturbance between two adjacent sampling moments, μ is the weight factor, and μ>0.

8. The discrete sliding mode control method of the voltage outer loop of an energy storage converter according to claim 1, characterized in that: The step of estimating the lumped disturbance comprises: make: In the formula, is Δf NL Estimated value of (k), Δf NL (k-1) is Δf NL (k) The delay of one sampling moment; The compact format dynamic linearized data model is delayed by one step to obtain: Δf NL (k-1)=Δu dc (k)-φ(k-1)Δi L (k-1) In the formula, Δu dc (k) is the change of DC bus voltage between two adjacent sampling moments, φ(k-1) is the pseudo partial derivative at moment k-1, Δi L (k-1) is the change of inductor current between two adjacent sampling moments; Combining the above equations, we can get the estimated value of the lumped disturbance: In the formula, is Δf NL The estimated value of (k), Δf NL (k) is the change of the lumped disturbance between two adjacent sampling moments, u dc (k) and u dc (k-1) are the DC bus voltage sampling values ​​at time k and k-1 respectively.

9. An electronic device, characterized in that: include: one or more processors; Memory; and One or more programs stored in a memory, the one or more programs comprising instructions for executing the discrete sliding mode control method of the voltage outer loop of the energy storage converter as claimed in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: It comprises one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs comprise instructions for executing a discrete sliding mode control method for a voltage outer loop of an energy storage converter as claimed in any one of claims 1 to 8.

Images