A parameter optimization method for oscillation suppression of photovoltaic grid-connected system with nonlinear load

Abstract

The present application relates to the technical field of power system control, and discloses a parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing a nonlinear load, comprising: establishing a sequence impedance model of the photovoltaic grid-connected system containing a controllable nonlinear load by using a harmonic linearization method; obtaining key parameters affecting system stability according to the obtained sequence impedance model of the photovoltaic grid-connected system containing a controllable nonlinear load, and dividing the initial value range of the key parameters into a stable region and an unstable region; on the basis of the stable region, performing parameter optimization through a simulation model of the photovoltaic grid-connected system containing a controllable nonlinear load, and using a particle swarm algorithm to optimize the key parameters affecting system stability to obtain optimized parameters; and according to the obtained optimized parameters, performing oscillation suppression on the photovoltaic grid-connected system containing a controllable nonlinear load. The present application can effectively improve the problem of slow system response speed caused by improving system stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system control, and particularly relates to a parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing a nonlinear load. BACKGROUND

[0002] The power energy industry is the main source of CO2 emissions, accounting for 87% of the total carbon emissions of the whole society.

[0003] On the power supply side, new energy power generation represented by photovoltaic occupies an important position. Generally, photovoltaic energy needs to be converted by power electronic converters to realize grid connection due to its intermittent and fluctuating characteristics. In addition, in order to reduce carbon emissions on the load side and improve energy utilization efficiency, new load elements based on power electronic interfaces are also widely used in industrial production and daily life, such as electrolytic hydrogen production, electric arc furnaces, variable frequency transmission / speed regulating motors and electric vehicle chargers. These new load elements are mainly based on power conversion systems consisting of front-end rectification links and back-end power conversion (DC / DC or DC / AC) links. Among them, three-phase thyristor controlled rectifiers are widely used in real life due to their cost-effectiveness and large capacity.

[0004] With the large-scale grid connection of new energy power generation represented by photovoltaic and the extensive use of nonlinear loads based on power electronic device interfaces, the new power system presents the characteristics of "high proportion of renewable energy" and "high proportion of power electronic devices". Due to the nonlinear and strong coupling characteristics of power electronic devices and the multi-time scale control links, the interaction between photovoltaic, nonlinear load and alternating current grid may cause system oscillation, threatening the stable operation of the grid and its connected equipment and causing risks. Therefore, it is of great significance to analyze the oscillation characteristics of the photovoltaic grid-connected system containing nonlinear loads and to study the oscillation suppression method. SUMMARY

[0005] The present application aims to overcome the shortcomings of the prior art and provide a parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing a nonlinear load, comprising the following steps:

[0006] Step one, a harmonic linearization method is used to establish a sequence impedance model of a photovoltaic grid-connected system containing a controllable nonlinear load, and a corresponding simulation model of the photovoltaic grid-connected system containing the controllable nonlinear load is built;

[0007] Step two, according to the obtained photovoltaic grid-connected system with controllable nonlinear load sequence impedance model, the relative sensitivity of the real part of impedance and the relative sensitivity of the imaginary part of impedance are calculated, and the relative sensitivity of the real part of impedance and the relative sensitivity of the imaginary part of impedance are sorted according to the absolute value, the parameter with larger absolute value of relative sensitivity of real part or imaginary part of impedance is identified as the key parameter affecting system stability, and the initial value range of the key parameter is divided into stable region and unstable region;

[0008] Step three, on the basis of the stable region, the parameter optimization is carried out through the simulation model of photovoltaic grid-connected system with controllable nonlinear load, and the particle swarm algorithm is used to optimize the key parameters affecting system stability, so as to obtain the optimized parameters;

[0009] Step four, according to the obtained optimized parameters, the oscillation suppression of photovoltaic grid-connected system with controllable nonlinear load is carried out.

[0010] Further, the harmonic linearization method is used to establish the sequence impedance model of photovoltaic grid-connected system with controllable nonlinear load, which includes:

[0011] The photovoltaic grid-connected system with controllable nonlinear load includes photovoltaic power generation unit, nonlinear load and equivalent alternating current grid; wherein the simplified equivalent model is used on the direct current side of the photovoltaic power generation unit, the outer ring adopts constant power control, and the inner ring adopts current control; the three-phase controllable nonlinear load based on thyristor rectification is used for the nonlinear load, which includes measurement link, phase-locked loop, direct current control, and constant direct current voltage source is used for the equivalent of direct current side; the sequence impedance modeling is carried out on the photovoltaic power generation unit and controllable nonlinear load respectively through the harmonic linearization method, and then the sequence impedance model of photovoltaic grid-connected system with controllable nonlinear load is obtained according to the equivalent of alternating current grid impedance.

[0012] Further, according to the obtained photovoltaic grid-connected system with controllable nonlinear load sequence impedance model, the relative sensitivity of the real part of impedance and the relative sensitivity of the imaginary part of impedance are calculated, and the relative sensitivity of the real part of impedance and the relative sensitivity of the imaginary part of impedance are sorted according to the absolute value, the parameter with larger absolute value of relative sensitivity of real part or imaginary part of impedance is identified as the key parameter affecting system stability, including:

[0013] The parameters include circuit parameters and control parameters: the circuit parameters are two filter inductors, one filter capacitor and one damping resistor for LCL type filter; the control parameters are PI control parameters corresponding to phase-locked loop and current inner ring of photovoltaic generator set, and PI control parameters corresponding to phase-locked loop and direct current control of controllable nonlinear load;

[0014] The relative sensitivity of the real part of impedance is:

[0015]

[0016] The relative sensitivity of the impedance imaginary part is:

[0017]

[0018] Further, the initial value range of the key parameter is divided into a stable region and an unstable region, including:

[0019] The value range of the key influence parameter corresponding to the system equivalent resistance value greater than 0 is the stable region, which is the initial optimization interval of the parameter; the value range of the parameter corresponding to the system equivalent resistance value less than 0 is the unstable region.

[0020] Further, the initial value range of the key parameter is divided into a stable region and an unstable region, including:

[0021] The optimization objective function includes two parts of the system stability index and the response performance index, and the calculation formula is as follows:

[0022]

[0023] J = min [k1 (-R min ) + k2ψ]

[0024] In the formula, ψ is the system response performance index; i d (t) and i dref are the actual value and the reference value of the d-axis current of the photovoltaic power generation unit respectively; i dc (t) and i dc_ref are the actual value and the reference value of the DC side current of the controllable nonlinear load respectively; t s is the calculation starting time, T ψ is the calculation duration; R min is the system stability index, which means the minimum value of the equivalent resistance of the system in the easy oscillation frequency band; k1 and k2 are the weight coefficients of the stability index R min and the response performance index ψ respectively.

[0025] The beneficial effects of the present application are: (1) the parameter stability domain analysis proposed in the present application narrows the initial optimization range of the key influence parameter, and improves the effectiveness and efficiency of parameter optimization.

[0026] (2) The parameter optimization method considering system stability and response performance, the optimization objective function of which contains system stability index Rmin and system response performance index ψ, can effectively improve the damping level of the system in the oscillation frequency band, and also considers the influence of the change of the optimization parameter on the system response performance, thereby effectively solving the problem of slow system response caused by improving system stability. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 It is a parameter optimization method flowchart for oscillation suppression of a photovoltaic grid-connected system containing a nonlinear load.

[0028] Figure 2 It is a structure diagram of a photovoltaic grid-connected system containing a controllable nonlinear load.

[0029] Figure 3 It is a circuit topology and control system structure diagram of a photovoltaic power generation unit.

[0030] Figure 4 It is a schematic diagram of a three-phase controllable nonlinear load based on thyristor rectification.

[0031] Figure 5 It is a schematic diagram of an equivalent small signal model of a photovoltaic grid-connected system containing a controllable nonlinear load.

[0032] Figure 6 It is a schematic diagram of impedance characteristic curves of a photovoltaic grid-connected system containing a controllable nonlinear load and an alternating current power grid.

[0033] Figure 7 It is a schematic diagram of relative sensitivity of real parts of impedances of various parameters of the system.

[0034] Figure 8 It is a schematic diagram of relative sensitivity of imaginary parts of impedances of various parameters of the system.

[0035] Figure 9 It is a schematic diagram of changes of equivalent resistance of the system when k p2 , k p4 changes.

[0036] Figure 10 It is a parameter optimization method flowchart considering system stability and response performance. DETAILED DESCRIPTION

[0037] The technical solutions of the present application will be further described in detail below in combination with the drawings, but the protection scope of the present application is not limited to the following description.

[0038] The features and performance of the present application will be further described in detail below in combination with the embodiments.

[0039] As Figure 1As shown, a parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing a nonlinear load, comprising the following steps:

[0040] Step one, a harmonic linearization method is used to establish a sequence impedance model of a photovoltaic grid-connected system containing a controllable nonlinear load, and a corresponding simulation model of a photovoltaic grid-connected system containing a controllable nonlinear load is built.

[0041] Step two, according to the obtained sequence impedance model of the photovoltaic grid-connected system containing a controllable nonlinear load, the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance of each parameter are calculated, and the absolute values of the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance are sorted to identify the parameters with larger absolute values as the key parameters affecting the stability of the system, and the initial value range of the key parameters is divided into stable and unstable regions.

[0042] Step three, on the basis of the stable region, parameter optimization is carried out through the simulation model of the photovoltaic grid-connected system containing a controllable nonlinear load, and the particle swarm algorithm is used to optimize the key parameters affecting the stability of the system to obtain the optimized parameters.

[0043] Step four, according to the obtained optimized parameters, the oscillation suppression of the photovoltaic grid-connected system containing a controllable nonlinear load is carried out.

[0044] A harmonic linearization method is used to establish a sequence impedance model of a photovoltaic grid-connected system containing a controllable nonlinear load, comprising:

[0045] The photovoltaic grid-connected system containing a controllable nonlinear load includes a photovoltaic power generation unit, a nonlinear load and an equivalent alternating current grid; wherein the photovoltaic power generation unit adopts a simplified equivalent model on the DC side, the outer ring adopts a constant power control, and the inner ring adopts a current control; the nonlinear load adopts a three-phase controllable nonlinear load based on a thyristor rectifier, which includes a measurement link, a phase-locked loop, a DC current control, and a constant DC voltage source equivalent on the DC side; the sequence impedance modeling of the photovoltaic grid-connected system containing a controllable nonlinear load is carried out by the harmonic linearization method on the photovoltaic power generation unit and the controllable nonlinear load, and the sequence impedance model of the photovoltaic grid-connected system containing a controllable nonlinear load is obtained according to the equivalent impedance of the alternating current grid.

[0046] According to the obtained sequence impedance model of the photovoltaic grid-connected system containing a controllable nonlinear load, the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance of each parameter are calculated, and the absolute values of the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance are sorted to identify the parameters with larger absolute values as the key parameters affecting the stability of the system, comprising:

[0047] The parameters include circuit parameters and control parameters; the circuit parameters are two filter inductances, one filter capacitance and one damping resistance for an LCL type filter; the control parameters are PI control parameters corresponding to a phase-locked loop and a current inner loop of the photovoltaic generator set, and PI control parameters corresponding to a phase-locked loop and a direct current control of the controllable nonlinear load;

[0048] The relative sensitivity of the real part of the impedance is:

[0049]

[0050] The relative sensitivity of the imaginary part of the impedance is:

[0051]

[0052] The initial value range of the key parameters is divided into a stable region and an unstable region, and the division includes:

[0053] The value range of the key influence parameter corresponding to the system equivalent resistance value greater than 0 is the stable region, and is used as the initial optimization interval of the parameter; the value range of the parameter corresponding to the system equivalent resistance value less than 0 is the unstable region.

[0054] The parameter optimization is performed on the basis of the stable region through the photovoltaic grid-connected system simulation model containing the controllable nonlinear load, and the particle swarm algorithm is used to optimize the key parameters affecting the system stability to obtain the optimized parameters, and the optimization includes:

[0055] The optimization objective function includes two parts of the system stability index and the response performance index, and the calculation formula is as follows:

[0056]

[0057] J = min [k1 (-R min ) + k2ψ]

[0058] In the formula, ψ is the system response performance index; i d (t) and i dref are respectively the actual value and the reference value of the d-axis current of the photovoltaic power generation unit; i dc (t) and i dc_ref are respectively the actual value and the reference value of the direct current of the controllable nonlinear load; t s is the calculation starting time, T ψ is the calculation time length; R min is the system stability index, which means the minimum value of the equivalent resistance of the system in the easy oscillation frequency band; k1 and k2 are respectively the weight coefficients of the stability index R min and the response performance index ψ.

[0059] Specifically, the embodiment takes Figure 2The effectiveness of the parameter optimization method for the photovoltaic grid-connected system with nonlinear load is verified by taking the photovoltaic grid-connected system with controllable nonlinear load as an example.

[0060] The impedance model of the photovoltaic grid-connected system with controllable nonlinear load is established, and the circuit topology and control system of the photovoltaic power generation unit are simulated as shown in the figure. Figure 3 In the figure, L1, L2, C f , and R d are the inverter-side inductance, grid-side inductance, filter capacitance, and damping resistance of the LCL filter, respectively; S1-S6 are switching elements; u ci , u f , and u j are the inverter output voltage, filter voltage, and PCC voltage; i ci , i f , and i j are the inverter-side inductance current, filter capacitance current, and grid-side inductance current, respectively. The PCC point voltage is input to a phase-locked loop (PLL) to obtain the PCC point phase θ PLL ; the outer loop adopts a constant power control mode, and the inner loop is a current control, H i (s) is a current inner loop PI controller; k s is a coupling coefficient.

[0061] The topology structure and control system of a three-phase controllable nonlinear load based on thyristor rectification are shown in the figure. Figure 4 In the figure, i dc2 and u dc2 are the DC side current and voltage of the nonlinear load; R4, L4, and C4 are the DC side resistance, inductance, and capacitance; it is assumed that the back-end power conversion link can be equivalent to a constant DC source V dc0 ; the PLL link has the same phase-locked loop structure as in Figure 3 , and the proportional and integral coefficients are k p3 and k i3 , respectively; G im (s) is a gain and first-order filter link for DC current measurement; G i (s) is a DC current control PI link, and the proportional and integral coefficients are k p4 and k i4 , respectively; i dc2_ref is a DC current reference value; θ l (0) is the phase lag of the lth switching tube; θ l is the equivalent grid voltage phase angle; α is the trigger angle; and mod is the modulo operation.

[0062] According to the circuit structure and control system principle of the photovoltaic power generation unit and the controllable nonlinear load, the impedance model Z pv (s) of the photovoltaic power generation unit is established by using the harmonic linearization method load (s) of the controllable nonlinear load. On the basis of Z pv (s) and Z load (s), according to the AC power grid impedance equivalent method as shown in Figure 5 , the photovoltaic power generation unit and the controllable nonlinear are regarded as a whole, and this whole is referred to as a parallel subsystem for the convenience of description in the following, and the impedance of the parallel subsystem is represented by Z s (s).

[0063] According to the circuit structure and control system principle of the system, the corresponding photovoltaic grid-connected system simulation model with controllable nonlinear load is built in the simulation platform.

[0064] System stability analysis and key influence parameter analysis

[0065] If the photovoltaic power generation unit and the controllable nonlinear load respectively meet the stability condition under the power supply of an ideal voltage source, the stability of the photovoltaic grid-connected system with controllable nonlinear load depends on whether Z g (s) / Z s (s) meets the Nyquist stability criterion, wherein Z g (s) represents the AC power grid impedance. The impedance characteristic curve of the system is as shown in Figure 6 when the logarithmic frequency stability criterion is used for stability judgment.

[0066] As can be seen from Figure 6 , the impedance amplitude-frequency characteristics of the parallel subsystem and the AC power grid intersect at 119.04 Hz, at which time the impedance phase difference between the parallel subsystem and the AC power grid is 184.02°, which is greater than the stability boundary 180°. Therefore, the photovoltaic grid-connected system with controllable nonlinear elements has an oscillation risk near 119 Hz.

[0067] The stability of the photovoltaic grid-connected system with controllable nonlinear load is closely related to system parameters. In order to analyze the key parameters affecting the stability of the system, the impedance relative sensitivity is used to quantitatively analyze the influence degree of the circuit parameters and the control parameters on the total impedance characteristic, and by calculating and sorting the impedance real part relative sensitivity and the impedance imaginary part relative sensitivity of each parameter, the key parameters affecting the stability of the system can be obtained. The calculation formulas of the impedance real part relative sensitivity and the impedance imaginary part relative sensitivity are as shown in formulas (5) and (6).

[0068]

[0069]

[0070] where Z total s is the total impedance of the system, K i represents each system parameter, H Re is the real part of the relative sensitivity, H Im is the imaginary part of the relative sensitivity.

[0071] The calculation results are shown in Figure 7 , Figure 8 From the figure, it can be seen that in the frequency band of 80-150 Hz, the changes of the controllable nonlinear load DC current control proportional coefficient k p4 and the photovoltaic power generation unit current inner loop proportional coefficient k p2 have a greater impact on the impedance of the photovoltaic grid-connected system containing controllable nonlinear elements, and are the key factors affecting system oscillation.

[0072] 3. The parameter optimization method considering system stability and response performance and the particle swarm algorithm for parameter setting are used to determine the initial optimization range of k p2 and k p4 . Based on the negative resistance theory, k p2 and k p4 are taken as independent variables to analyze the changes of the equivalent resistance of the photovoltaic grid-connected system containing controllable nonlinear loads at 119 Hz, and the results are shown in Figure 9 From Figure 9 it can be seen that as k p2 and k p4 increase, the corresponding equivalent resistance value of the system gradually decreases and even becomes negative resistance, and there is a part of the region in the parameter plane where the equivalent resistance value of the system is less than 0. The part where the equivalent resistance value of the system is less than 0 is divided into an unstable region, and the part where the equivalent resistance value of the system is greater than 0 is taken as a stable region, and the range corresponding to the stable region is the initial optimization range of the parameters, i.e. the following conditions need to be met:

[0073] (k p2 , k p4 ) ∈ parameter stable domain (7)

[0074] To quantify the response performance and oscillation suppression effect of the system, the photovoltaic power generation unit d-axis current i d (t) and the controllable nonlinear load DC side current i dc2 (t) are taken as references to define an oscillation energy function ψ, and the calculation formula is shown in formula (8).

[0075]

[0076] where t s is the starting time of calculation, T ψ is the calculation time length of the oscillation energy function, i dref (t) is the reference value of the photovoltaic power generation unit d-axis current, and i dc2_refReference value of direct current side current of the controllable nonlinear element.

[0077] In order to improve system stability and response performance, an optimization objective function is defined:

[0078] J = min [k1(-R min )+k2ψ] (9)

[0079] In the formula, R min is the minimum value of the equivalent resistance of the system in the easy oscillation frequency band, and k1 and k2 are weight coefficients of the stability index R min and the performance index ψ. The optimization process is shown in Figure 10

[0080] The above only describes the preferred embodiments of the present application, and it should be understood that the present application is not limited to the forms disclosed herein, and should not be considered as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concepts described herein by the above teachings or related art or knowledge. Any modification and change made by those skilled in the art without departing from the spirit and scope of the present application shall be within the protection scope of the claims of the present application.​

Claims

1. A parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing nonlinear loads, characterized in that: The steps include: Step 1: Use the harmonic linearization method to establish a sequence impedance model of the photovoltaic grid-connected system with controllable nonlinear loads, and build a corresponding simulation model of the photovoltaic grid-connected system with controllable nonlinear loads; Step 2: Based on the obtained sequence impedance model of the photovoltaic grid-connected system with controllable nonlinear loads, the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance of each parameter are calculated, and the parameters are ranked according to their absolute values. The parameters with larger absolute values ​​of the relative sensitivity of the real part or imaginary part of the impedance are identified as key parameters affecting the stability of the system, and the initial value range of the key parameters is divided into a stable region and an unstable region. Step 3: Based on the stable region, parameter optimization is performed using a simulation model of a photovoltaic grid-connected system with controllable nonlinear loads, and a particle swarm algorithm is used to optimize the key parameters that affect system stability to obtain the optimized parameters. Step 4: Suppress oscillation of the photovoltaic grid-connected system containing controllable nonlinear loads according to the obtained optimization parameters; Based on the stable region, the parameters of the photovoltaic grid-connected system simulation model containing controllable nonlinear loads are optimized, and the particle swarm algorithm is used to optimize the key parameters affecting the system stability to obtain the optimized parameters, including: The optimization objective function includes two parts: system stability index and response performance index. The calculation formula is as follows: J=min[k1(-R min )+k2ψ] Where, ψ is the system response performance index; i d (t), i dref are the actual value and reference value of the d-axis current of the photovoltaic power generation unit respectively; i dc (t), i dc_ref are the actual value and reference value of the DC side current of the controllable nonlinear load respectively; t s To calculate the starting time, T ψ is the calculation time; R min is the system stability index, which means the minimum value of the equivalent resistance of the system in the easy oscillation frequency band; k1 and k2 are the stability index R min The weight coefficient of the response performance index ψ.

2. The parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing nonlinear loads according to claim 1, characterized in that: The harmonic linearization method is used to establish a sequence impedance model of a photovoltaic grid-connected system containing a controllable nonlinear load, including: The photovoltaic grid-connected system containing a controllable nonlinear load comprises a photovoltaic power generation unit, a nonlinear load and an equivalent AC power grid; wherein a simplified equivalent model is adopted on the DC side of the photovoltaic power generation unit, the outer loop adopts constant power control, and the inner loop adopts current control; the nonlinear load adopts a three-phase controllable nonlinear load based on thyristor rectification, which includes a measurement link, a phase-locked loop, and DC current control, and a constant DC voltage source is adopted on the DC side for equivalent; sequence impedance modeling is performed on the photovoltaic power generation unit and the controllable nonlinear load respectively through the harmonic linearization method, and then the sequence impedance model of the photovoltaic grid-connected system containing the controllable nonlinear load is obtained based on the AC power grid impedance equivalent.

3. The parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing nonlinear loads according to claim 2, characterized in that: The sequence impedance model of the photovoltaic grid-connected system with a controllable nonlinear load is obtained, and the parameters are ranked according to their absolute values ​​by calculating the relative sensitivity of the real part of the impedance and the relative sensitivity of the imaginary part of the impedance. The parameters with larger absolute values ​​of the relative sensitivity of the real part of the impedance or the imaginary part of the impedance are identified as key parameters affecting the stability of the system, including: The parameters include circuit parameters and control parameters. The circuit parameters are specifically two filter inductors, a filter capacitor, and a damping resistor for the LCL filter. The control parameters are the PI control parameters corresponding to the phase-locked loop and current inner loop of the photovoltaic generator set, as well as the PI control parameters corresponding to the phase-locked loop and DC current control of the controllable nonlinear load. The relative sensitivity of the real part of impedance is: The relative sensitivity of the imaginary part of impedance is: Where Ztotal(s) is the total impedance of the system, K i Represents various system parameters, H Re is the real part relative sensitivity, H Im is the imaginary relative sensitivity.

4. The parameter optimization method for oscillation suppression of a photovoltaic grid-connected system containing nonlinear loads according to claim 3, characterized in that: The aforementioned division of the initial value range of the key parameters into a stable region and an unstable region includes: The range of key influencing parameters corresponding to the system equivalent resistance value greater than 0 is the stable area, which serves as the initial optimization interval of the parameters; the range of parameter values ​​corresponding to the system equivalent resistance value less than 0 is the unstable area.

Images