A method of harmonic suppression based on frequency adaptive resonance control

Abstract

The application discloses a harmonic suppression method based on frequency adaptive resonance control and belongs to the field of power system harmonic suppression. In view of the problem that SVG under a traditional control strategy is difficult to effectively suppress harmonics, a resonance control strategy is designed based on an internal model theorem and is applied to a voltage control link of SVG. Meanwhile, in order to adapt to the uncertainty of harmonic frequency, a recursive least square method is introduced to identify the harmonic frequency, the resonance control parameter is dynamically adjusted, and the harmonic compensation bandwidth of the SVG device is widened. The simulation result shows that the suppression method provided by the application is simple in operation, can effectively eliminate the (intermediate) harmonic of the offshore wind power system without additionally adding electrical primary equipment, and has strong robustness and adaptability.

Description

Technical Field

[0001] This invention belongs to the field of power system harmonic suppression, specifically relating to a harmonic suppression method based on frequency adaptive resonance control. Background Technology

[0002] Offshore wind power has experienced rapid development due to its abundant resources and proximity to load centers. However, most offshore wind farms are typically connected to the onshore power grid via long-distance submarine AC cables. Considering that the distributed capacitance of AC cables at the same transmission distance and voltage level is usually 20 to 40 times that of overhead lines, while the line impedance is 50% to 100% of that of overhead lines, offshore wind power systems may experience harmonic resonance and harmonic amplification phenomena uncommon in onshore wind farms, seriously jeopardizing the safe and stable operation of the power system.

[0003] Currently, research on harmonic suppression in offshore wind power systems mainly focuses on passive LC filtering and active power filters (APF). Passive LC filtering typically requires additional electrical equipment, making it economically inefficient. APF has a small capacity and limited harmonic compensation capability. However, optimizing SVG control strategies to achieve harmonic suppression in offshore wind power systems has attracted significant attention due to its lack of additional hardware costs and larger device capacity. However, existing SVG harmonic suppression strategies primarily rely on the instantaneous reactive power theorem, calculating the DQ-axis components of the grid current to obtain the harmonic current to be compensated, neglecting the impact of power flow on SVG harmonic suppression performance. Furthermore, existing suppression strategies mainly target standard harmonics, rarely considering phenomena such as fundamental frequency shift and simple harmonics that may exist in the system. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a harmonic suppression method based on frequency adaptive resonant control. The method designs a resonant control strategy using the internal mode theorem for use in the voltage control stage of the SVG, and identifies harmonic frequencies using the least squares method to dynamically adjust the resonant control parameters, thereby effectively suppressing harmonic and interharmonic phenomena in offshore wind power systems.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A harmonic suppression method based on frequency adaptive resonance control, characterized by comprising the following steps:

[0007] S1: Select an electrical signal that characterizes the harmonic phenomenon of the offshore wind power system as the input signal for harmonic suppression control;

[0008] S2: Acquire the selected input signal and perform preprocessing;

[0009] S3: Based on the preprocessed signal, the harmonic parameters are identified using the recursive least squares method;

[0010] S4: The control parameters of the harmonic suppressor are optimized using the particle swarm optimization algorithm;

[0011] S5: Dynamically adjust the resonant frequency of the harmonic suppressor based on the harmonic parameters identified in step S3;

[0012] S6: Input the signal from step S2 to the harmonic suppressor to generate a harmonic suppression command;

[0013] S7: Send the harmonic suppression command generated in step S6 to the SVG device.

[0014] To optimize the above technical solution, the specific measures also include:

[0015] Furthermore, in step S1, the selected electrical signal is not affected by the operating mode.

[0016] Furthermore, in step S2, the input signal collected is the three-phase voltage data of the SVG node.

[0017] Furthermore, in step S2, the preprocessing of the input signal includes DQ transformation and signal DC blocking. After the preprocessing operation, the fundamental component in the original signal is removed, leaving the harmonic components.

[0018] Furthermore, in step S3, a fourth-order RLS signal model is selected to identify the harmonic parameters and obtain the frequency and amplitude change rate of the harmonics.

[0019] Furthermore, when the amplitude change rate is greater than or equal to zero, the SVG device is locked, the RLS parameters are re-initialized, and the pre-processed signal is re-input into the initialized fourth-order RLS signal model to obtain the frequency and amplitude change rate of the harmonics; when the amplitude change rate is less than zero, the resonant frequency of the harmonic suppressor is kept unchanged, and the SVG harmonic suppression strategy is kept in operation.

[0020] Furthermore, the step of selecting a fourth-order RLS signal model to identify harmonic parameters and obtain the frequency and amplitude change rate of the harmonics specifically includes the following steps:

[0021] S3.1: Determine initial values, including estimates of state variables. Error covariance matrix P(k) and noise variance R of measurement data; For time k The estimated values ​​are x1, x2, x3, and x4, which are the state variables of the fourth-order RLS.

[0022] S3.2: Calculate the observation matrix H(k) at time k;

[0023] S3.3: Calculate the RLS gain matrix K(k) and error covariance matrix P(k) at time k using the observation matrix H(k) and the error covariance matrix P(k-1) at time k-1:

[0024]

[0025] P(k)=(IK(k)H(k))P(k-1)(IK(k)H(k)) T +K(k)R[K(k)] T

[0026] In the formula, I represents the fourth-order identity matrix;

[0027] S3.4: Calculate the estimated value of the state variable at time k using the RLS gain matrix K(k):

[0028]

[0029] Combining the error covariance matrix P(k) obtained in step S3.3 with the estimated value of the current state variable Return to step S3.2 and continue iterative calculations until RLS converges;

[0030] S3.5: After RLS convergence, the harmonic parameters are calculated using the following formula:

[0031] a = x1

[0032] f k =x3

[0033] In the formula, a and f k These represent the rate of change of the harmonic amplitude and the frequency, respectively.

[0034] Furthermore, in step S4, the method for optimizing the control parameters of the harmonic suppressor is as follows:

[0035] S4.1: Establish the objective function based on the product integral index J of time and absolute error:

[0036]

[0037] In the formula, t f The rise time of the harmonic suppressor is represented by e(t), and the static error is e(t) = u. ref -u(t), u ref The reference value of the voltage is represented by u(t), and the actual value of the voltage is represented by u(t).

[0038] S4.2: Set control parameters [k] p ,k r The initial value of k. p Represents the proportional gain, kr Indicates the resonant gain;

[0039] S4.3: The control parameters are optimized using the particle swarm optimization algorithm;

[0040] S4.4: Calculate the objective function value;

[0041] S4.5: Determine if the maximum number of iterations has been reached. If so, output the solution [k] that minimizes the objective function value. p k r [The parameters are determined as the optimal parameters for the harmonic suppressor; otherwise, proceed to step S4.4 to continue the next optimization iteration.]

[0042] Furthermore, the present invention proposes a computer-readable storage medium storing a computer program, characterized in that the computer program causes a computer to execute the harmonic suppression method based on frequency adaptive resonance control as described above.

[0043] Furthermore, the present invention proposes an electronic device, characterized in that it includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the harmonic suppression method based on frequency adaptive resonance control as described above.

[0044] The beneficial effects of this invention are:

[0045] 1. The frequency adaptive resonance control harmonic suppressor of the present invention dynamically adjusts the resonant frequency of the suppressor by detecting changes in harmonic frequency online, and has strong adaptability;

[0046] 2. Compared with methods such as LC filtering, the harmonic suppressor of the present invention with frequency adaptive resonance control does not require the addition of primary electrical equipment, has lower economic cost, and higher engineering applicability.

[0047] 3. This invention proposes a harmonic suppression method based on frequency adaptive resonant control, which can effectively suppress harmonics and interharmonics in offshore wind power systems. This method is applicable to harmonic suppression with uncertain frequencies and only requires modification of the SVG control stage. It does not require any changes to the operation mode of the offshore wind farm and has low operating costs. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the method in a specific embodiment of the present invention.

[0049] Figure 2 This is a schematic diagram of the harmonic suppressor structure of the present invention.

[0050] Figure 3 This is a flowchart of the parameter optimization design for the present invention.

[0051] Figure 4 This is a diagram showing the installation location of the harmonic suppressor of the present invention in an offshore wind power system.

[0052] Figure 5a and Figure 5b The figures show a comparison of the harmonic suppression effects of the harmonic suppressor of this invention and the LC filter.

[0053] Figure 6 This is a time-domain waveform diagram of the Q-axis component of the SVG grid connection point voltage according to the present invention.

[0054] Figure 7a and Figure 7b The figures show a comparison of the harmonic suppression effects of the harmonic suppressor of this invention and the simple harmonic suppression effect of the LC filter. Detailed Implementation

[0055] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0056] This invention discloses a harmonic suppression method based on frequency adaptive resonance control, the specific process of which is as follows: Figure 1 As shown, it includes the following steps.

[0057] S1: Select an electrical signal that can characterize the harmonic phenomenon of the offshore wind power system and is basically unaffected by the system operation mode as the input signal for harmonic suppression control.

[0058] S2: Acquire the selected input signal and perform preprocessing.

[0059] In order to characterize the harmonic phenomena of offshore wind power systems and be largely unaffected by the system's operating mode, the collected measurement data mainly consisted of the three-phase voltage data of the SVG nodes.

[0060] The preprocessing of the measurement data includes two parts: DQ transformation and signal DC blocking. After preprocessing, the fundamental component in the original signal is removed, leaving the harmonic components, thus avoiding the influence of the fundamental component on the harmonic suppression strategy.

[0061] S3: Identify harmonic parameters using the recursive least squares (RLS) method.

[0062] This invention selects a fourth-order RLS model to identify harmonic voltage signals. The harmonic components of the Q-axis voltage of the SVG grid-connected power supply can be expressed as:

[0063]

[0064] Among them, A k (t), f k and These represent the amplitude, frequency, and phase angle of the harmonic, respectively, and t represents time.

[0065] To calculate the frequency and amplitude change rate of the harmonic signal, a fourth-order RLS signal model is introduced, as shown in equation (2):

[0066] u k (t)=(x1t+x2)sin(2πx3t+x4)(2)

[0067] The fourth-order RLS signal model is discretized, where k∈N is a discrete-time variable. A measurement equation is introduced, as shown in equation (3):

[0068] z(k)=h(X(k))(3)

[0069] in, x1, x2, x3, and x4 are state variables of a fourth-order RLS.

[0070] In the above equation, X(k) is the state vector; h(X(k)) is the nonlinear function of the measurement equation, and the overall process h(X) is expressed as in equation (4):

[0071] h(X)=(x1t+x2)sin(2πx3t+x4)(4)

[0072] Performing a Taylor expansion on equation (4) and retaining the first-order terms yields the observation matrix H(k), as shown in equation (5):

[0073]

[0074] X represents the state variable, and its meaning is: The meaning is the estimated value of X at the k-th time.

[0075] In step S3, the specific steps for identifying harmonic parameters using the four-state least squares method (taking the fifth harmonic as an example) are as follows:

[0076] S3.1: Determine the initial values:

[0077]

[0078] Where P(k) is the error covariance matrix; R is the noise variance of the measurement data.

[0079] S3.2: Calculate the observation matrix H(k) at time k using equation (5).

[0080] S3.3: Calculate the RLS gain matrix K(k) and covariance matrix P(k) at time k using the observation matrix H(k) and the covariance matrix P(k-1) at time k-1.

[0081]

[0082] P(k)=(IK(k)H(k))P(k-1)(IK(k)H(k)) T +K(k)R[K(k)] T (8)

[0083] Where I is a fourth-order identity matrix.

[0084] S3.4: Calculate the state variables at time k using Equation (9) with the RLS gain matrix K(k). When the gain matrix K(k) is not zero, i.e., RLS has not converged, combine the covariance matrix P(k) obtained in step S3.3 at time k with the estimated value of the current state variables. Return to step S3.2 and repeat the above process to continue the iteration.

[0085]

[0086] S3.5: After RLS converges, the harmonic parameters can be obtained from equations (10) to (13):

[0087]

[0088] a = x1(11)

[0089] f k =x3(12)

[0090]

[0091] Among them, A k (t), a, f k and These represent the amplitude, rate of change of amplitude, frequency, and phase angle of the harmonic, respectively.

[0092] S4: The particle swarm optimization algorithm is used to optimize the control parameters of the harmonic suppressor.

[0093] The structure of a harmonic suppressor is as follows: Figure 2 As shown, the steps of the parameter optimization method for the adaptive harmonic suppressor are as follows:

[0094] S4.1: Establish the objective function. The function established based on the product integral index J of time and absolute error is as follows:

[0095]

[0096] Among them, t f The rise time of the harmonic suppressor is represented by e(t); the static error e(t) = u ref -u(t). u ref The reference value of the voltage is represented by u(t); the actual value of the voltage is represented by u(t).

[0097] S4.2: Set [k] p ,k r The initial values ​​of these parameters will affect the harmonic suppression capability and dynamic process performance of the SVG. They need to be configured appropriately based on the relationship between the parameters and system performance.

[0098] The design principles are as follows:

[0099] 1) Based on the harmonic parameters identified in step S3, set the resonant frequency for harmonic suppression;

[0100] 2) Proportional gain k p The larger the value, the greater the static error of the system, but the smaller the phase margin, making it more prone to oscillation;

[0101] 3) Resonant gain k r The larger the value, the faster the system response, but the more pronounced the overshoot.

[0102] S4.3: The control parameters are optimized using the Particle Swarm Optimization (PSO) algorithm, with a particle swarm size of 20 and a maximum number of iterations of 200. p The range is set to [5, 20], k r The range is set to [100, 500].

[0103] S4.4: Calculate the objective function value.

[0104] S4.5: Determine if the maximum number of iterations has been reached. If so, output the solution [k] that minimizes the objective function value. p k r [This is] used as the optimal parameter for the harmonic suppressor; otherwise, proceed to step S4.4 to continue the next optimization iteration.

[0105] S5: Dynamically adjust the resonant frequency of the harmonic suppressor based on the harmonic parameters identified in step S3.

[0106] S6: Input the signal from step S2 to the harmonic suppressor to generate the harmonic suppression command required by the system.

[0107] S7: Send the harmonic suppression command generated in step S6 to the SVG controller.

[0108] Two embodiments of the present invention are described below.

[0109] Example 1

[0110] The verification example used in this invention is an AC submarine cable transmission system from a offshore wind farm in Jiangsu Province. The system structure diagram of the harmonic suppression system with frequency adaptive resonance control of this invention is shown below. Figure 4 As shown. The offshore wind farm is connected to the onshore 220kV system via a three-stage voltage boosting method. Specifically, the offshore wind farm is connected to the same 35kV busbar via a submarine collection cable. The voltage is then boosted to 110kV via an offshore substation, and subsequently connected to an onshore substation via a double-circuit 110kV AC submarine cable. This substation includes a 35 / 110 / 220kV three-winding transformer. The 220kV node is connected to the onshore power grid via an overhead line, and an SVG device is installed on the 35kV side. According to... Figure 3 The flowchart optimizes the harmonic suppression parameters of this invention, with a set value of: k. p =10.7, k r =120.5.

[0111] Set up in MATLAB / SIMULINK Figure 4 The transient simulation model shown has a sampling frequency of 20kHz. The operating conditions are set as follows: a 5th harmonic voltage of 0.88% is added to the onshore 220kV equivalent grid, and a 6.7A 5th harmonic current is injected into the system from the wind farm. The harmonic suppression strategy is activated at t=1.2s. The voltage waveform at the SVG grid connection point under harmonic influence is shown below. Figure 5a , 5b As shown.

[0112] It can be observed that when the system does not take harmonic suppression measures, the voltage distortion rate (THD) at the SVG grid connection point reaches 3.98%, exceeding the 2% specified in the national standard GB / T14549-1993. After t=1.2s, after implementing the harmonic suppression strategy of this invention and LC passive filtering, it can be observed that the system voltage is significantly improved and the voltage distortion rate is greatly reduced.

[0113] To further compare the differences between the strategy proposed in this invention and LC filtering, Figure 6 The Q-axis voltage u at the SVG grid connection point is given. tq It can be seen that the transition process of the harmonic suppression strategy of the present invention when suppressing harmonics is about 0.2s, which is much faster than the 0.7s of the LC filtering strategy. On the other hand, after the suppression measures are implemented, the harmonic suppression strategy of the present invention brings an impulse voltage of 0.05pu to the system, which is only one-eighth of that of the LC filtering strategy, which is beneficial to the stable operation of the system.

[0114] Example 2

[0115] Abundant simple harmonics still exist in the actual operation of offshore wind power systems. Therefore, it is necessary to verify the suppression effect of the harmonic suppressor of this invention on simple harmonics. The operating conditions are set as follows: 0.01% of the 4.5th interharmonic is added to the onshore 220kV equivalent grid, and the harmonic suppression strategy is activated at t=1.2s. The voltage waveform at the SVG grid connection point under the action of interharmonics is as follows. Figure 7a and Figure 7b As shown.

[0116] It can be observed that without the implementation of suppression strategies, the distortion rate of the SVG grid-connected voltage is 1.09%, and the 4.5th harmonic content is 0.82%, exceeding the 0.4% specified in the national standard GB / T 24337-2009, which is detrimental to the stable operation of the system. However, when the harmonic suppression method based on frequency adaptive resonant control of this invention is implemented, the THD of the SVG grid-connected voltage is improved to 0.58%, and the 4.5th harmonic content is reduced to 0.35%, meeting the national standard. However, after the LC filter is implemented, the THD of the grid-connected voltage is 1.03%, the 4.5th harmonic content is 0.78%, and voltage amplification occurs, which is also detrimental to the stable operation of the system. Therefore, this demonstrates that the harmonic suppression method based on frequency adaptive resonant control for SVG of this invention still has a significant suppression effect on simple harmonics.

[0117] Furthermore, the present invention also proposes a computer-readable storage medium storing a computer program that causes a computer to execute the harmonic suppression method based on frequency adaptive resonance control as described above.

[0118] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0119] Furthermore, the present invention also proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the harmonic suppression method based on frequency adaptive resonance control as described above.

[0120] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A harmonic suppression method based on frequency adaptive resonance control, characterized in that, Includes the following steps: S1: Select an electrical signal that characterizes the harmonic phenomenon of the offshore wind power system as the input signal for harmonic suppression control; S2: Acquire the selected input signal and perform preprocessing; S3: Based on the preprocessed signal, the harmonic parameters are identified using the recursive least squares method; S4: The control parameters of the harmonic suppressor are optimized using the particle swarm optimization algorithm; S5: Dynamically adjust the resonant frequency of the harmonic suppressor based on the harmonic parameters identified in step S3; S6: Input the signal from step S2 to the harmonic suppressor to generate a harmonic suppression command; S7: Send the harmonic suppression command generated in step S6 to the SVG device.

2. The harmonic suppression method based on frequency adaptive resonance control as described in claim 1, characterized in that: In step S1, the selected electrical signal is not affected by the operating mode.

3. The harmonic suppression method based on frequency adaptive resonance control as described in claim 1, characterized in that: In step S2, the input signal collected is the three-phase voltage data of the SVG node.

4. The harmonic suppression method based on frequency adaptive resonance control as described in claim 1, characterized in that: In step S2, the preprocessing of the input signal includes DQ transformation and signal blocking. After the preprocessing operation, the fundamental component in the original signal is removed, leaving the harmonic components.

5. The harmonic suppression method based on frequency adaptive resonance control as described in claim 1, characterized in that: In step S3, a fourth-order RLS signal model is selected to identify the harmonic parameters and obtain the frequency and amplitude change rate of the harmonics.

6. The harmonic suppression method based on frequency adaptive resonance control as described in claim 5, characterized in that: When the amplitude change rate is greater than or equal to zero, the SVG device is locked, the RLS parameters are re-initialized, and the pre-processed signal is re-input into the initialized fourth-order RLS signal model to obtain the frequency and amplitude change rate of the harmonics; when the amplitude change rate is less than zero, the resonant frequency of the harmonic suppressor remains unchanged.

7. The harmonic suppression method based on frequency adaptive resonance control as described in claim 5, characterized in that: The step of selecting a fourth-order RLS signal model to identify harmonic parameters and obtain the frequency and amplitude change rate of harmonics specifically includes the following steps: S3.1: Determine initial values, including estimates of state variables. Error covariance matrix P(k) and noise variance R of measurement data; For time k The estimated values ​​are x1, x2, x3, and x4, which are the state variables of the fourth-order RLS. S3.2: Calculate the observation matrix H(k) at time k; S3.3: Calculate the RLS gain matrix K(k) and error covariance matrix P(k) at time k using the observation matrix H(k) and the error covariance matrix P(k-1) at time k-1: P(k)=(I-K(k)H(k))P(k-1)(I-K(k)H(k)) T +K(k)R[K(k)] T In the formula, I represents the fourth-order identity matrix; S3.4: Calculate the estimated value of the state variable at time k using the RLS gain matrix K(k): Combining the error covariance matrix P(k) obtained in step S3.3 with the estimated value of the current state variable Return to step S3.2 and continue iterative calculations until RLS converges; S3.5: After RLS convergence, the harmonic parameters are calculated using the following formula: a = x1 f k =x3 In the formula, a and f k These represent the rate of change of the harmonic amplitude and the frequency, respectively.

8. The harmonic suppression method based on frequency adaptive resonance control as described in claim 1, characterized in that: In step S4, the method for optimizing the control parameters of the harmonic suppressor is as follows: S4.1: Establish the objective function based on the product integral index J of time and absolute error: In the formula, t f The rise time of the harmonic suppressor is represented by e(t), and the static error is e(t) = u. ref -u(t), u ref The reference value of the voltage is represented by u(t), and the actual value of the voltage is represented by u(t). S4.2: Set control parameters [k] p ,k r The initial value of k. p Represents the proportional gain, k r Indicates the resonant gain; S4.3: The control parameters are optimized using the particle swarm optimization algorithm; S4.4: Calculate the objective function value; S4.5: Determine if the maximum number of iterations has been reached. If so, output the solution [k] that minimizes the objective function value. p k r [The parameters are determined as the optimal parameters for the harmonic suppressor; otherwise, proceed to step S4.4 to continue the next optimization iteration.] 9. A computer-readable storage medium storing a computer program, characterized in that, The computer program causes the computer to execute the harmonic suppression method based on frequency adaptive resonance control as described in any one of claims 1-8.

10. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the harmonic suppression method based on frequency adaptive resonance control as described in any one of claims 1-8.

Images