Single-loop nonlinear control method, three-phase inverter and equipment

Abstract

The invention discloses a single-loop nonlinear control method, which comprises the following steps that a nonlinear three-phase inverter model is constructed, the nonlinear three-phase inverter model comprises a direct-current power supply, a three-phase inverter circuit, an LC filter circuit and a three-phase load, and initial conditions are set; time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation; based on the energy characteristics of the filter capacitor, defining a system state variable, and representing the output voltage of the filter capacitor as flat output in an electrostatic energy form; deducing a control variable of the three-phase inverter by combining a state variable of a nonlinear three-phase inverter system through the flatly output time derivative; according to a feedback control rule, the flat output error is converged to zero; and generating a PWM modulation signal based on the control variable, and adjusting the AC output voltage of the three-phase inverter. According to the invention, the AC output voltage of the three-phase inverter is effectively adjusted, and power is supplied to the load when the load needs.

Description

Technical Field

[0001] The present invention relates to the technical field of converter control systems, and in particular to a single-loop nonlinear control method, a three-phase inverter, and equipment. Background Art

[0002] In renewable energy power generation systems, three-phase inverters are the core devices that convert direct current into alternating current. Their control technology directly affects the power conversion efficiency and power supply stability. Currently, the most mainstream method in the field of traditional three-phase inverter control is the decoupled dq vector control technology based on the proportional integral (PI) compensator. This method achieves inverter output regulation by coordinating PI control of the DC voltage with PI control of the actual current. However, its core relies on the linear approximation of the system operating point, which can only guarantee control accuracy within a small local range. When the system operating point deviates from the initial set range, the error of the linear approximation increases significantly, resulting in reduced control performance and even system instability.

[0003] The limitations of traditional linear control methods are further reflected in their dynamic response and scope of application: due to the inherent characteristics of linear approximation, they cannot effectively handle the wide range of dynamic behaviors of nonlinear systems, such as large load changes and grid voltage imbalances. Complex parameter adjustments are required to adapt to different operating conditions, increasing the difficulty of engineering implementation. Furthermore, PI control relies on an integrator to eliminate steady-state errors, which can easily introduce overshoot or delay, making it difficult to meet the high-precision, fast-response demands of modern power supply systems. Therefore, there is an urgent need for a nonlinear control method that does not require linear approximation and can maintain high robustness over a large operating range. Summary of the Invention

[0004] One of the objectives of the present invention is to provide a single-loop nonlinear control method, device, and medium to effectively regulate the AC output voltage of a three-phase inverter and provide power to a load when the load needs it.

[0005] The object of the present invention is to provide a single-loop nonlinear control method, comprising the following steps:

[0006] Constructing a nonlinear three-phase inverter model, wherein the nonlinear three-phase inverter model includes a DC power supply, a three-phase inverter circuit, an LC filter circuit, and a three-phase load, and setting initial conditions;

[0007] The time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation;

[0008] Based on the energy characteristics of the filter capacitor, the system state variables are defined, and the output voltage of the filter capacitor is expressed as a flat output in the form of electrostatic energy;

[0009] The control variables of the three-phase inverter are derived by combining the time derivative of the flat output with the state variables of the nonlinear three-phase inverter system.

[0010] According to the feedback control rule, the flat output error is converged to zero;

[0011] A PWM modulation signal is generated based on the control variable to adjust the AC output voltage of the three-phase inverter.

[0012] In the above technical solution, the total current at the output end of the three-phase inverter is zero, and the relationship between voltage and current is as follows:

[0013] i a +i b +i c =0

[0014] v a +v b +v c =0

[0015] where i a ,i b ,i c They are the three branch currents at the output end of the three-phase inverter, v a , v b , v c They are the three branch voltages at the output end of the capacitor filter.

[0016] In the above technical solution, the time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation by Park transformation, including:

[0017]

[0018]

[0019] Among them, i d is the DC component of the three-phase inverter output current in the dq coordinate system, i q is the AC component of the three-phase inverter output current in the dq coordinate system;

[0020] v d is the DC component of the three-phase inverter output voltage in the dq coordinate system,

[0021] v q is the AC component of the three-phase inverter output voltage in the dq coordinate system,

[0022] e d is the control variable of the three-phase inverter, that is, the DC component of the grid voltage in the dq coordinate system, e qis the control variable of the three-phase inverter, that is, the AC component of the grid voltage in the dq coordinate system, m a ,δ are the modulation index and phase shift of sinusoidal PWM respectively, and the parasitic resistance of the line group and output filter is R s .

[0023] In the above technical solution, the control variables of the three-phase inverter are derived by combining the time derivative of the flat output with the state variables of the nonlinear three-phase inverter system, including:

[0024] The output voltage of the three-phase inverter in the filter capacitor is defined as the electrostatic energy y d and y q , y=[y d ,y q ] t =[y1,y2] t ;

[0025] The state variable of the three-phase inverter is: x = [x1, x2, x3, x4] t =[i d ,i q ,v d ,v q ] t

[0026] The control variables of the three-phase inverter are estimated by taking the time derivative of the flat output:

[0027]

[0028] where v d and v q is the control variable of the three-phase inverter.

[0029] In the above technical solution, the generation parameters of the sinusoidal PWM modulation signal are determined by the control variable v d and v q Sure.

[0030] In the above technical solution, the feedback control rule is:

[0031]

[0032]

[0033] Among them, K 11 ,K 12 ,K 13 ,K 21 ,K 22 ,K 23 are controller parameters.

[0034] In the above technical solution, the controller parameters are:

[0035]

[0036] where ζ is the desired main damping ratio, ω n is the natural frequency.

[0037] In the above technical solution, the Parker transform is:

[0038]

[0039] Where P represents the PARK transformation matrix, and θ is the angle at the corresponding moment of any angular frequency ω.

[0040] The present invention further provides a three-phase inverter, which executes any one of the single-loop nonlinear control methods described above.

[0041] The present invention also provides a device having a computer program stored thereon, wherein the computer program, when executed, implements a single-loop nonlinear control method as described in any one of the above.

[0042] To achieve the above object, the present invention provides the following technical solutions:

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention avoids the limitations of traditional linear approximate control by constructing a nonlinear model, can maintain high precision in a wider range of operating conditions, simplifies the control design complexity of the time-varying system, makes trajectory planning more natural, converges errors faster and has better dynamic response, and significantly improves the control accuracy, robustness and output voltage stability of the three-phase inverter. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of a flow chart of an embodiment of the present invention.

[0046] Figure 2 Schematic diagram of the structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] This embodiment provides a single-loop nonlinear control method to effectively regulate the AC output voltage of a three-phase inverter and provide power to a load when the load needs it.

[0049] Specifically, such as Figure 1 As shown, the following steps are included:

[0050] Step S1: construct a nonlinear three-phase inverter model, wherein the nonlinear three-phase inverter model includes a DC power supply, a three-phase inverter circuit, an LC filter circuit and a three-phase load, and set initial conditions.

[0051] The nonlinear three-phase inverter model uses a DC power supply as its energy input source, which is directly connected to the input port of the three-phase inverter circuit. The three-phase inverter circuit is responsible for converting DC power into three-phase AC power, and its output port is connected to the LC filter circuit. The LC filter circuit is used to filter out the high-frequency noise generated during the inverter process to ensure a smooth output voltage waveform. The filtered AC power is ultimately delivered to the three-phase load to meet the load's power supply needs.

[0052] The initial condition is set as the total current at the output of the three-phase inverter is zero, and the output voltage at the LC filter capacitor is zero at the initial time of the system. That is, the voltage-current relationship of the inverter is as follows:

[0053] i a +i b +i c =0(1)

[0054] v a +v b +v c =0(2)

[0055] where i a ,i b ,i c They are the three branch currents at the inverter output end, v a , v b , v c These are the three branch voltages at the output of the capacitor filter. The initial condition setting conforms to the physical characteristics of current balance in a three-phase system and simplifies the initial state of the model, avoiding complex initial bias interference in the subsequent control algorithm design.

[0056] The DC power supply provides a stable DC input. The three-phase inverter circuit converts the DC power into three-phase AC power in the form of pulse-width modulation (PWM) through the high-frequency on-off of the internal power switch tube. The LC filter circuit filters out the high-frequency components in the PWM wave through a combination of inductors and capacitors, outputting three-phase AC power that is close to a sine wave. Finally, the filtered AC power drives the three-phase load, realizing the conversion and transmission of electrical energy from DC to AC.

[0057] The construction of a nonlinear three-phase inverter model avoids the limitations of traditional control methods that rely on linear approximations of system operating points. Traditional methods rely on proportional-integral control, assuming system linearization within the small signal range, and are only applicable under specific operating conditions. This model, however, directly establishes nonlinear relationships based on actual physical characteristics, maintaining high accuracy over a wider operating range. This provides an accurate model foundation for subsequent nonlinear control based on flat characteristics, ensuring that the system trajectory is directly defined by the flat output without relying on integrators or differential equation compensation, thereby improving control robustness and dynamic response.

[0058] Step S2: converting the time-varying voltage and current signals output by the three-phase inverter into two-phase time-invariant rotating coordinate system signals through coordinate transformation;

[0059] Perform PARK transformation on the three-phase voltage and three-phase current output from the AC port, as shown in the following formula:

[0060]

[0061] Where P represents the PARK transformation matrix, and θ is the angle at the corresponding moment of any angular frequency ω.

[0062] The three-phase time-varying system is transformed into a two-phase time-invariant rotating dq coordinate system, and the time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation, including:

[0063]

[0064] Among them, i d is the DC component of the three-phase inverter output current in the dq coordinate system, i q is the AC component of the three-phase inverter output current in the dq coordinate system;

[0065] v d is the DC component of the three-phase inverter output voltage in the dq coordinate system,

[0066] v q is the AC component of the three-phase inverter output voltage in the dq coordinate system,

[0067] e d is the control variable of the three-phase inverter, that is, the DC component of the grid voltage in the dq coordinate system, e q is the control variable of the three-phase inverter, that is, the AC component of the grid voltage in the dq coordinate system, m a ,δ are the modulation index and phase shift of sinusoidal PWM respectively, and the parasitic resistance of the line group and output filter is R s .

[0068] The original three-phase time-varying system is converted into a two-phase time-invariant system. In the original abc coordinate system, the amplitude, frequency and phase of the three-phase signal all change with time, resulting in the model containing time-varying coefficients, making it difficult to directly apply linear control theory; after Park transformation, the i in the dq coordinate system d ,i q ,v d and v q In steady state, it manifests as a DC component, or an AC component containing only low-order harmonics, greatly simplifying the complexity of the system model. This time-invariant property eliminates the need to deal with time-varying parameters in subsequent nonlinear control designs based on flat characteristics, allowing the direct use of linear feedback control rules, significantly improving the feasibility and robustness of the control algorithm.

[0069] Step S3: Based on the energy characteristics of the filter capacitor, define the system state variable and express the output voltage of the filter capacitor as a flat output in the form of electrostatic energy;

[0070] The output voltage of the three-phase inverter in the filter capacitor is defined as the electrostatic energy y d and y q , y=[y d ,y q ] t =[y1,y2] t (5);

[0071] Define the state variables of the three-phase inverter as: x = [x1, x2, x3, x4] t =[i d ,i q ,v d ,v q ] t (6)

[0072] At the same time, define the system input as shown below:

[0073] u=[u1,u2] t =[e d ,e q ] t (7)

[0074] According to formula (4), the following formulas (8)(9)(10)(11) can be obtained:

[0075]

[0076] in,

[0077]

[0078] Based on these two voltage components and their time derivatives, the electrostatic energy of the filter capacitor is calculated to obtain flat outputs x3 and x4; at the same time, combined with the current component i in the dq coordinate system d ,i q Constructing a quaternary state variable consisting of current and voltage. This process combines the physical energy properties of capacitance with the mathematical flatness theory to ensure that the flat output fully characterizes the dynamic behavior of the system. In other words, all states and inputs of the system can be represented by the flat output and its finite-order derivatives.

[0079] The flat output definition based on energy characteristics eliminates zero dynamics and fully linearizes the system, avoiding the control failure problem that may be caused by zero dynamic instability in traditional state feedback linearization. At the same time, the state variables include current and voltage components, fully covering the electrical state of the inverter output. This provides complete state information for the subsequent derivation of control variables through flat output derivatives, ensuring the accuracy and robustness of the feedback control rule design.

[0080] Step S4: deriving the control variables of the three-phase inverter by combining the time derivative of the flat output with the state variables of the nonlinear three-phase inverter system;

[0081] The control variables of the three-phase inverter are estimated by taking the time derivative of the flat output:

[0082]

[0083] where v d and v q is the control variable of the three-phase inverter. The design of the reference signal required for the dq current of the three-phase inverter can be obtained by Calculated.

[0084] The control variable is calculated directly from the system's measurable state and its derivative, eliminating the need for complex observers or model predictions, thus reducing the complexity of engineering implementation. The derivation process is based on the precise dynamic relationships of the nonlinear model, avoiding the errors caused by linear approximations in traditional PI control. This ensures high-precision voltage regulation under a wide range of operating conditions, including sudden load changes and grid voltage imbalances. The control variable is directly linked to the derivative of the flat output, allowing subsequent feedback control rules to directly affect the inverter input, achieving rapid tracking and stabilization of the output voltage.

[0085] Step S5: According to the feedback control rule, the flat output error is made to converge to zero.

[0086] The dynamic error of the flat output is constrained to zero through linear feedback rules, and the stability and rapidity of the system response are ensured through parameter design.

[0087] The control of the flat output is ensured by using linear feedback control, and the calculation of this virtual control is derived through the feedback control law (15)(16), which allows the flat output error to converge to zero.

[0088]

[0089] where K 11 ,K 12 ,K 13 ,K 21 ,K 22 ,K 23 are controller parameters. v1 and v2 directly determine the voltage reference values ​​of the d-axis and q-axis through the inverse transformation of the system flattening. These are fed into the PWM to generate the duty cycle signal that drives the three-phase bridge arm to determine the damping and natural frequency of the error dynamics. In order to ensure that the closed-loop poles fall strictly on the desired second-order underdamped position, they are set to the desired characteristic polynomial. The optimal choice of the designed controller parameters is defined as:

[0090]

[0091] Where ζ is the desired main damping ratio, which is usually set to 0.7 to balance response speed and overshoot; ω n is the closed-loop natural frequency, which can be selected according to the grid fundamental wave or system bandwidth requirements; p1 is an additional real pole, which is used to increase the system phase margin and suppress the amplification of high-frequency noise on the integral link. Then, according to equations (15) and (16), v1 and v2 are solved and reversely calculated into v through equations (12) and (13). d and v q Finally, SPWM generates a pulse train that acts on the power devices. Because the error differential, proportional, and integral loops are connected in parallel to form a third-order closed loop with two conjugate poles plus one real pole, the system not only accurately locks y1 and y2 in steady state but also achieves zero overshoot and rapid convergence during load disturbances or reference steps.

[0092] The single-loop architecture eliminates the switching and linearization compensation required for the traditional PI voltage outer loop and PI current inner loop, simplifying hardware while increasing bandwidth. Furthermore, the error dynamics incorporate an integral term, maintaining zero steady-state error despite device parameter drift and grid-side imbalance, while significantly reducing the output voltage total harmonic distortion (THD). Furthermore, the additional real pole p1 provides the controller with adjustable phase margin, ensuring robust system stability at high frequencies. According to tests, transient recovery time is nearly halved compared to the classic PI-dq strategy, while ripple is reduced by approximately ten percentage points.

[0093] Step S6: Generate a PWM modulation signal based on the control variable to adjust the AC output voltage of the three-phase inverter.

[0094] First, the control variable vd and v q An inverse Parker transform generates a sinusoidal modulation wave in a three-phase stationary coordinate system. These three modulation waves are then compared with a high-frequency triangular carrier to generate switching signals for the three-phase bridge arms. When the modulation wave exceeds the carrier, the upper transistors turn on; otherwise, they turn off. Finally, the three-phase inverter outputs a PWM voltage through the high-frequency switching of the switching transistors, which is then filtered by LC to produce a smooth sinusoidal AC voltage. The modulation index and phase shift directly control the amplitude and phase of the output voltage, enabling dynamic regulation of the three-phase inverter's AC output voltage. This process converts control commands into hardware operations, ensuring a stable output voltage that adapts to load requirements.

[0095] This method avoids the limitations of traditional linear approximate control by constructing a nonlinear model, and can maintain high accuracy over a wider range of operating conditions; coordinate transformation is used to convert the time-varying system into a two-phase time-invariant model, simplifying the complexity of the control design; the flat output defined based on the energy characteristics of the filter capacitor has no zero dynamics, so that the system trajectory is directly determined by the flat output, without relying on integral or differential compensation, and the trajectory planning is more natural; the control variable instructions derived from the state variables are accurate, and the feedback control rules ensure that the flat output error converges quickly and the dynamic response is better; real-time adjustment of the output voltage is achieved through PWM modulation, which overall improves the control accuracy, robustness and adaptability to load fluctuations of the three-phase inverter, providing a more stable and reliable AC voltage output for scenarios such as renewable energy generation.

[0096] Based on the same inventive idea, Figure 2 As shown, the present application also provides a three-phase inverter that executes the above-mentioned single-loop nonlinear control method.

[0097] The present application also provides a device for executing the above-mentioned single-loop nonlinear control method.

[0098] The specific control process is the same as the above method and will not be described in detail.

[0099] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A single-loop nonlinear control method, characterized in that: The following steps are involved: Constructing a nonlinear three-phase inverter model, wherein the nonlinear three-phase inverter model includes a DC power supply, a three-phase inverter circuit, an LC filter circuit, and a three-phase load, and setting initial conditions; The time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation; Based on the energy characteristics of the filter capacitor, the system state variables are defined, and the output voltage of the filter capacitor is expressed as a flat output in the form of electrostatic energy; The control variables of the three-phase inverter are derived by combining the time derivative of the flat output with the state variables of the nonlinear three-phase inverter system. According to the feedback control rule, the flat output error is converged to zero; A PWM modulation signal is generated based on the control variable to adjust the AC output voltage of the three-phase inverter.

2. A single-loop nonlinear control method according to claim 1, characterized in that: The total current at the output end of the three-phase inverter is zero, and the relationship between voltage and current is as follows: i a +i b +i c =0 v a +v b +v c =0 where i a ,i b ,i c They are the three branch currents at the output end of the three-phase inverter, v a , v b , v c They are the three branch voltages at the output end of the capacitor filter.

3. The single-loop nonlinear control method according to claim 1, characterized in that: The time-varying voltage and current signals output by the three-phase inverter are converted into two-phase time-invariant rotating coordinate system signals through coordinate transformation, including: Among them, i d is the DC component of the three-phase inverter output current in the dq coordinate system, i q is the AC component of the three-phase inverter output current in the dq coordinate system; v d is the DC component of the three-phase inverter output voltage in the dq coordinate system, v q is the AC component of the three-phase inverter output voltage in the dq coordinate system, e d is the control variable of the three-phase inverter, that is, the DC component of the grid voltage in the dq coordinate system, e q is the control variable of the three-phase inverter, that is, the AC component of the grid voltage in the dq coordinate system, m a ,δ are the modulation index and phase shift of sinusoidal PWM respectively, and the parasitic resistance of the line group and output filter is R s .

4. The single-loop nonlinear control method according to claim 1, characterized in that: By combining the time derivative of the flat output with the state variables of the nonlinear three-phase inverter system, the control variables of the three-phase inverter are derived, including: The output voltage of the three-phase inverter in the filter capacitor is defined as the electrostatic energy y d and y q , y=[y d ,y q ] t =[y1,y2] t ; The state variable of the three-phase inverter is: x = [x1, x2, x3, x4] t =[i d ,i q ,v d ,v q ] t The control variables of the three-phase inverter are estimated by taking the time derivative of the flat output: where v d and v q is the control variable of the three-phase inverter.

5. A single-loop nonlinear control method according to claim 4, characterized in that: The generation parameters of the sinusoidal PWM modulation signal are determined by the control variable v d and v q Sure.

6. A single-loop nonlinear control method according to claim 5, characterized in that: The feedback control rule is: Among them, K 11 ,K 12 ,K 13 ,K 21 ,K 22 ,K 23 are controller parameters.

7. The single-loop nonlinear control method according to claim 6, characterized in that: The controller parameters are: where ζ is the desired main damping ratio, ω n is the natural frequency.

8. The single-loop nonlinear control method according to claim 3, characterized in that: The Parker transform is: Where P represents the PARK transformation matrix, and θ is the angle at the corresponding moment of any angular frequency ω.

9. A three-phase inverter, characterized in that: The three-phase inverter executes a single-loop nonlinear control method as claimed in any one of claims 1 to 8.

10. A device, characterized in that A computer program is stored thereon, and when the computer program is executed, the single-loop nonlinear control method according to any one of claims 1 to 8 is implemented.

Images