A method for online estimation of grid impedance without measuring remote weak grid voltage

Abstract

The present invention discloses an online estimation method for grid impedance without measuring the voltage of a remote weak grid. The three-phase grid-connected inverter system is taken as the research object. First, U PCC Point voltage and grid current i 2x ; Secondly, transform the coordinates to the αβ coordinate axis and calculate the ideal grid voltage e through the phase-locked loop α 、e β , and then calculate ΔU under the αβ axis PCCα , ΔU PCCβ and i 2α 、i 2β Component; then perform Kalman filtering to obtain the real-time changing ΔU PCC and i 2x , the difference between the common coupling point voltage and the ideal grid voltage ΔU PCC It can be regarded as the voltage fluctuation at PCC point caused by the change of grid impedance. PCC and i 2x Iterate N times to estimate the grid inductance and grid resistance parameters in real time; the present invention can reduce the number of sampling channels by constructing a state equation and a Kalman filter design method, and is beneficial to improving the grid-connected current control stability under a weak grid.

Description

Technical Field

[0001] The invention belongs to the technical field of power electronics, and in particular relates to an online estimation method for grid impedance without measuring a remote weak grid voltage. Background Art

[0002] With the expansion of the scale of distributed generation systems, renewable energy generation systems are continuously integrated into the large power grid, which not only brings flexibility to the large power grid in terms of power dispatching, but also reduces the exploitation of traditional fossil energy and reduces pollution to the environment. However, affected by the distribution of renewable energy sources such as wind and solar energy, the integration of distributed generation systems into the large power grid usually requires longer transmission lines and transformers, which leads to non-negligible line impedance. At the same time, due to the increase in the penetration rate of distributed generation, as well as the line impedance and the large number of linear and nonlinear loads in the actual power grid, the local power grid presents weak grid characteristics. In general, the short-circuit ratio is used to analyze the relative strength of the AC large power grid and the stability of the system when the distributed generation system is integrated into the large power grid. Under the weak power grid, there is voltage fluctuation at the common coupling point of the grid voltage, and the grid impedance cannot be ignored.

[0003] Under weak power grid, the system damping and stability margin are reduced, which will undoubtedly greatly increase the control difficulty of grid-connected inverter. Grid-connected inverter can convert the direct current of renewable energy into alternating current, and then match it with a suitable filter to filter out the harmonic components in the alternating current, so as to merge it into the large power grid. Line impedance under weak power grid cannot be ignored, and the voltage at the common coupling point fluctuates, resulting in a significant reduction in the quality of grid-connected current and power. In order to cope with the fluctuations of grid impedance and grid voltage under weak power grid and improve control stability, online estimation of grid impedance is indispensable. When using the traditional Kalman filtering method to perform online estimation of grid impedance, it is necessary to collect the grid current and the voltage at both ends of the grid impedance, and the remote voltage of the weak power grid is difficult to measure, which brings inevitable difficulties to the above-mentioned online estimation method of grid impedance, and it is difficult to accurately estimate the grid impedance. Based on the Kalman filtering parameter identification method, the present invention proposes a more convenient Kalman filtering design method, which only needs to collect the grid current and the voltage at the common coupling point, and can obtain accurate grid impedance information using fewer sampling channels. Summary of the invention

[0004] The purpose of the present invention is to provide a method for online estimation of grid impedance without measuring the voltage of a remote weak grid. By using a Kalman filter parameter identification algorithm, only the voltage at the common coupling point and the grid current need to be sampled to accurately estimate the grid impedance online.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A method for estimating grid impedance online without measuring the voltage of a remote weak grid is proposed. The method takes a three-phase grid-connected inverter system as the research object and specifically includes the following steps:

[0007] Step 1: Establish the state space equation at the three-phase common coupling point;

[0008] Step 2: Simplify the state space equation, eliminating the need to sample the remote weak grid voltage to reduce sampling channels;

[0009] Step 3: Design a Kalman filter method to estimate the grid impedance online.

[0010] Furthermore, the specific method of step 1 is as follows:

[0011] Sampling three-phase common coupling point voltage U PCCx And the grid current i x , using Clarke transformation, the voltage and current in the three-phase DEF coordinate system are converted to the αβ coordinate system. The coordinate transformation formula T abc-＞αβ As shown in formula (1):

[0012]

[0013] The state space equation at the common coupling point is shown in equation (2):

[0014]

[0015] Where x = a, b, c, i 2α , U PCCα and e α I x2 , U PCCx and grid voltage e x The component on the α axis; i 2β , U PCCβ and e β I x2 , U PCCx and grid voltage e x The component on the β axis; R g and L g are the resistance and inductance of the grid impedance to be estimated; ω is the grid angular frequency.

[0016] Furthermore, the specific method of step 2 is as follows:

[0017] In the grid-connected current control system of the grid-connected inverter, the voltage at the common coupling point needs to be phase-locked to obtain the phase information of the grid voltage. Therefore, in the actual control process, the grid phase information obtained by the phase-locked loop can be used to construct an ideal three-phase grid voltage in the controller, so there is no need to measure the remote weak grid voltage, which can reduce the number of sampling channels of the traditional Kalman filter method, reduce the amount of calculation and hardware cost;

[0018] Grid voltage x It is an ideal three-phase sinusoidal power supply. Therefore, the difference between the common coupling point voltage in formula (2) and the ideal grid voltage can be regarded as the common coupling point voltage fluctuation caused by the grid impedance change. Therefore, formula (2) can be converted into formula (3):

[0019]

[0020] In the formula, ΔU PCCα and ΔU PCCβ are the components of the voltage fluctuation at the common coupling point on the α-axis and β-axis respectively.

[0021] Furthermore, the specific method of step 3 is as follows:

[0022] By discretizing formula (3), the state space equation in the discrete domain can be obtained as follows:

[0023]

[0024] Where, T s is the inverter sampling frequency, k and k+1 represent the sampling values ​​at the kth moment and the k+1th moment respectively;

[0025] Formula (4) can be simplified as

[0026]

[0027] In the formula,

[0028]

[0029] Therefore, the Jacobian matrix of the Kalman filter method can be expressed as:

[0030]

[0031] Step 3.1: Using the formula Predicting initial state variables

[0032] Step 3.2: Calculate the error covariance Among them G k is the Jacobian matrix, Q k-1 is the process noise covariance matrix;

[0033] Step 3.3: Calculate Kalman gain Where R k is the measurement noise covariance matrix;

[0034] Step 3.4: Calculate the updated state variables and the updated error covariance

[0035] Step 3.5: Enter the calculated difference between the ideal grid voltage and the actual grid voltage ΔU in the Kalman filter module PCCα , ΔU PCCβ , and grid-connected current i 2α 、i 2β The initial value of the ideal grid voltage and the actual grid voltage difference ΔU PCCα , ΔU PCCβ , and grid-connected current i α 、i β ;

[0036] Step 3.6: Calculate the updated value according to steps 3.1, 3.2, 3.3, and 3.4 D.U. PCCα , ΔU PCCβ , and grid-connected current i 2α , i 2β ;

[0037] Step 3.7: Repeat step 3.6 and iterate to get real-time updates. D.U. PCCα , D.U. PCCβ , and grid-connected current i 2α , i 2β And determine whether the number of iterations K is greater than the required number of iterations N. If K < N, resample U PCC The point voltage and grid-connected current are converted into 3 / 2 to calculate ΔU PCCα , ΔU PCCβ 、i 2α 、i 2β Repeat step 3.7 again. If K>N, proceed to the next step;

[0038] Step 3.8: Output real-time updates D.U. PCCα , D.U. PCCβ , i 2α , i 2β, the difference between the real-time updated common coupling point voltage and the ideal grid voltage D.U. PCCα , D.U. PCCβ , and grid current i 2α , i 2β It can be regarded as the grid inductance L g and grid resistance R g The PCC voltage and grid current fluctuations caused by the change of D.U. PCCα , D.U. PCCβ , i 2α , i 2β It can be equivalent to obtaining the updated grid inductance L g and grid resistance R g .

[0039] The beneficial effects of the present invention are:

[0040] The present invention provides a method for estimating grid impedance online without measuring the voltage of a remote weak grid. By sampling the voltage of a three-phase common coupling point and the grid-connected current, an ideal three-phase grid voltage is constructed in an actual controller, which can reduce the number of sampling channels. At the same time, the present invention can simultaneously estimate the resistance and inductance values ​​of the grid impedance with the help of a Kalman filter identification parameter algorithm, which is beneficial to improving the stability of grid-connected current control under a weak grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a topological structure diagram of the three-phase common coupling point of the present invention;

[0042] Figure 2 This is a flow chart based on the implementation of the Kalman filter method;

[0043] Figure 3 The present invention is a flow chart of a method for online estimating grid impedance without measuring the voltage of a remote weak grid. DETAILED DESCRIPTION

[0044] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] An online estimation method for grid impedance without measuring the voltage of a remote weak grid is proposed. The grid connection of a three-phase grid-connected inverter system is taken as the research object. When the grid impedance changes, the corresponding common coupling point voltage will fluctuate. That is to say, on the basis of an ideal grid, the grid impedance and grid fluctuation have a corresponding relationship. In order to improve the stability of inverter control, the three-phase common coupling point voltage and the three-phase grid-connected current are sampled. By constructing a state equation and a Kalman filter design method, the grid impedance is estimated online, which is beneficial to improving the stability of grid-connected current control.

[0046] An online estimation method of grid impedance without measuring the voltage of a remote weak grid is proposed, taking the grid-connected three-phase inverter system as the research object. Figure 3 As shown, the specific steps include:

[0047] Step 1: Establish the state space equation at the three-phase common coupling point, as follows;

[0048] like Figure 1 As shown, the three-phase common coupling point voltage U is sampled PCCx And the grid current i 2x , using Clarke transformation, the voltage and current in the three-phase DEF coordinate system are converted to the αβ coordinate system. The coordinate transformation formula T abc-＞αβ As shown in formula (1).

[0049]

[0050] The state space equation at the common coupling point is shown in equation (2).

[0051]

[0052] Where x = a, b, c, i 2α , U PCCα and e α I x2 , U PCCx and grid voltage e x The component on the α axis. 2β , U PCCβ and e β I x2 , U PCCx and grid voltage e x The component on the β axis. R g and L g are the resistance and inductance of the grid impedance respectively, and ω is the grid angular frequency.

[0053] Step 2: Simplify the state space equation, eliminating the need to sample the remote weak grid voltage, so as to reduce sampling channels, reduce costs, and improve estimation accuracy. The specific steps are as follows:

[0054] Grid voltage x It is an ideal three-phase sinusoidal power supply. Therefore, the difference between the common coupling point voltage in formula (2) and the ideal grid voltage can be regarded as the common coupling point voltage fluctuation caused by the grid impedance change. Therefore, formula (2) can be converted into formula (3).

[0055]

[0056] In the formula, ΔU PCCα and ΔU PCCβ are the components of the voltage fluctuation at the common coupling point on the α-axis and β-axis respectively.

[0057] Generally speaking, in the grid-connected current control system of the grid-connected inverter, the voltage at the common coupling point needs to be phase-locked to obtain the phase information of the grid voltage. Therefore, in the actual control process, the grid phase information obtained by the phase-locked loop can be used to construct an ideal three-phase grid voltage in the controller, so that there is no need to measure the remote weak grid, which can reduce the number of sampling channels of the traditional Kalman filter method, reduce the amount of calculation and hardware cost.

[0058] Step 3: Design a Kalman filter method to estimate the grid impedance online.

[0059] By discretizing formula (3), the state space equation in the discrete domain can be obtained as follows:

[0060]

[0061] Where, T S is the inverter sampling frequency, k and k+1 represent the sampling values ​​at the kth moment and the k+1th moment respectively.

[0062] Formula (4) can be simplified as

[0063]

[0064] In the formula,

[0065]

[0066] Therefore, the Jacobian matrix of the Kalman filter method can be expressed as:

[0067]

[0068] Figure 2 This is a specific implementation flow chart of the Kalman filter method. First, use the formula Predicting initial state variables Then calculate the error covariance Among them G kis the Jacobian matrix, Q k-1 is the process noise covariance matrix; then calculate the Kalman gain Where R k is the measurement noise covariance matrix; finally, the updated state variable is calculated and the updated error covariance

[0069] The specific steps are:

[0070] Step 3.1: Using the formula Predicting initial state variables

[0071] Step 3.2: Calculate the error covariance Among them G k is the Jacobian matrix, Q k-1 is the process noise covariance matrix;

[0072] Step 3.3: Calculate Kalman gain Where R k is the measurement noise covariance matrix;

[0073] Step 3.4: Calculate the updated state variables and the updated error covariance

[0074] Step 3.5: Enter the calculated difference between the ideal grid voltage and the actual grid voltage ΔU in the Kalman filter module PCCα , ΔU PCCβ , and grid-connected current i 2α 、i 2β The initial value of the ideal grid voltage and the actual grid voltage difference ΔU PCCα , ΔU PCCβ , and grid-connected current i 2α 、i 2β ;

[0075] Step 3.6: Calculate the updated value according to steps 3.1, 3.2, 3.3, and 3.4 D.U. PCCα , D.U. PCCβ , and grid-connected current i 2α , i 2 β ;

[0076] Step 3.7: Repeat step 3.6 and iterate to get real-time updates. D.U. PCCa , D.U. PCCβ , and grid-connected currenti 2α , i 2β And determine whether the number of iterations K is greater than the required number of iterations N. If K < N, resample U PCC The point voltage and grid-connected current are converted into 3 / 2 to calculate ΔU PCCα , ΔU PCCβ 、i 2α 、i 2β Repeat step 3.7 again. If K>N, proceed to the next step;

[0077] Step 3.8: Output real-time updates D.U. PCCα , D.U. PCCβ , i 2α , i 2β , the difference between the real-time updated common coupling point voltage and the ideal grid voltage D.U. PCCα , D.U. PCCβ , and grid current i 2α , i 2β It can be regarded as the grid inductance L g and grid resistance R g The PCC voltage and grid current fluctuations caused by the change of D.U. PCCα , D.U. PCCβ , i 2α , i 2β It can be equivalent to obtaining the updated grid inductance L g and grid resistance R g .

[0078] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for online estimation of grid impedance without measuring remote weak grid voltage, taking a three-phase grid-connected inverter system as the research object, characterized in that: The specific steps include: Step 1: Establish the state space equation at the three-phase common coupling point; The specific steps are: Sampling three-phase common coupling point voltage U PCCx And the grid current i 2x , using Clarke transformation, the voltage and current in the three-phase abc coordinate system are converted to the αβ coordinate system. The coordinate transformation formula T abc-＞αβ As shown in formula (1): The state space equation at the common coupling point is shown in equation (2): In the formula, x = a, b, c, i 2α , U PCCα and e α I x2 , U PCCx and grid voltage e x The component on the α axis; i 2β , U PCCβ and e β I x2 , U PCCx and grid voltage e x The component on the β axis; R g and L g are the resistance and inductance values ​​of the grid impedance to be estimated respectively; Step 2: Simplify the state space equation, eliminating the need to sample the remote weak grid voltage to reduce sampling channels; The specific steps are: In the grid-connected current control system of the grid-connected inverter, the voltage at the common coupling point needs to be phase-locked to obtain the phase information of the grid voltage. Therefore, in the actual control process, the grid phase information obtained by the phase-locked loop can be used to construct an ideal three-phase grid voltage in the controller, so there is no need to measure the remote weak grid voltage, which can reduce the number of sampling channels of the traditional Kalman filter method, reduce the amount of calculation and hardware cost; Grid voltage x It is an ideal three-phase sinusoidal power supply. Therefore, the difference between the common coupling point voltage in formula (2) and the ideal grid voltage can be regarded as the common coupling point voltage fluctuation caused by the grid impedance change. Therefore, formula (2) can be converted into formula (3): In the formula, ΔU PCCα and ΔU PCCβ are the components of the voltage fluctuation at the common coupling point on the α-axis and the β-axis respectively; Step 3: Design a Kalman filter method to estimate the grid impedance online; The specific steps are: By discretizing formula (3), the state space equation in the discrete domain can be obtained as follows: Where, T s is the inverter sampling frequency, k and k+1 represent the sampling values ​​at the kth moment and the k+1th moment respectively, and ω is the grid angular frequency; Formula (4) can be simplified as In the formula, Therefore, the Jacobian matrix of the Kalman filter method can be expressed as: Step 3.1: Using the formula Predicting initial state variables Step 3.2: Calculate the error covariance Among them G k is the Jacobian matrix, Q k-1 is the process noise covariance matrix; Step 3.3: Calculate Kalman gain Where R k is the measurement noise covariance matrix; Step 3.4: Calculate the updated state variables and the updated error covariance Step 3.5: Enter the calculated difference between the ideal grid voltage and the actual grid voltage ΔU in the Kalman filter module PCCα , ΔU PCCβ , and grid-connected current i 2α 、i 2β The initial value of the ideal grid voltage and the actual grid voltage difference ΔU PCCα , ΔU PCCβ , and grid-connected current i 2α 、i 2β ; Step 3.6: Calculate the updated value according to steps 3.1, 3.2, 3.3, and 3.4 ΔU PCCα , ΔU PCCβ , and grid-connected current i 2α , i 2β ; Step 3.7: Continue from Step 3.6 and continuously iterate to obtain real-time updated ΔU PCCα 、 ΔU PCCβ 、 and grid-connected current i 2α 、 i 2β and determine whether the iteration count K is greater than the required iteration count N. If K < N, resample the U PCC point voltage and grid-connected current for 3 / 2 transformation and calculate ΔU PCCα 、ΔU PCCβ 、i 2α 、i 2β Then repeat Step 3.

7. If K > N, execute the next step; Step 3.8: Output real-time updates ΔU PCCα , ΔU PCCβ 、i 2α 、i 2β , the difference between the real-time updated common coupling point voltage and the ideal grid voltage ΔU PCCα , ΔU PCCβ , and the grid current i 2α 、i 2β It can be regarded as the grid inductance L g and grid resistance R g The PCC voltage and grid current fluctuations caused by the change of ΔU PCCα , ΔU PCCβ 、i 2α 、i 2β It can be equivalent to obtaining the updated grid inductance L g and grid resistance R g .

Images