Reliable direct identification method for characteristic parameters of single-tuned filter

Abstract

The invention discloses a reliable direct-type identification method for characteristic parameters of a single-tuned filter, which comprises the following steps of: initializing the single-tuned filter to be detected to be in a zero state, applying a direct-current excitation source with constant amplitude to the single-tuned filter to be detected, sampling, storing a current sequence of the single-tuned filter, carrying out sliding mean filtering processing on a sample, and carrying out direct-current identification on the characteristic parameters of the single-tuned filter. Calculating an identification element, calculating an identification basic parameter, calculating an identification characteristic quantity, and obtaining the characteristic parameters of the single-tuned filter, such as resistance R, inductance L and capacitance C; according to the method, the single-tuned filter is used as a detection object, the characteristic parameters of the single-tuned filter are monitored in real time, a simulation model of the single-tuned filter is constructed, the identification method of the characteristic parameters in the filter is verified through experiments, and the identification method is confirmed to be accurate and reliable on the premise that the filter device is not disassembled and not damaged through analysis of experimental results. And the characteristic parameters of the single-tuned filter in the interference environment can be accurately and reliably estimated.

Description

Technical Field

[0002] This invention relates to the field of power system filter parameter detection technology, and in particular to a reliable direct identification method for characteristic parameters of a single-tuned filter. Background Technology

[0004] With the modernization of various sectors of the national economy, a large number of nonlinear loads, mainly composed of various power electronic devices, are being used more and more widely, and harmonic mitigation is receiving increasing attention. A single-tuned filter, consisting of a filter reactor and a filter capacitor connected in series, is the basic unit of a power system filter. As the service life of a single-tuned filter increases and the operating environment temperature changes, its internal components may experience temperature increases and malfunctions. Under these circumstances, the characteristic parameter values ​​of various types of internal components may change, leading to aging and weakening the filter's filtering performance. Therefore, monitoring the parameters of the internal components of a single-tuned filter is of great practical significance for understanding its filtering performance and ensuring its rational and effective use.

[0005] However, single-tuned filter products are often pre-packaged, and disassembling them to test these parameters can easily damage the filter. Testing these parameters without disassembly is currently extremely difficult. Therefore, research in this area has begun. The filtering performance of a single-tuned filter actually depends on the values ​​of its three internal lumped components—resistance, inductance, and capacitance—that are the characteristic parameters R, L, and C. These parameters collectively determine the frequency characteristics of the single-tuned filter, characterizing its filtering performance. The equivalent circuit model of a traditional single-tuned filter is as follows: Figure 1 As shown, it mainly consists of a resistor R, an inductor L, and a capacitor C connected in series. Let its input voltage be u. i (t), capacitor voltage is u c (t), current is i s For capacitor C, (t), we have:

[0006] (1.1)

[0007] As can be seen from circuit knowledge:

[0008] (1.2)

[0009] Input voltage u i (t) represents a step signal with amplitude U0, i.e.:

[0010] (1.3)

[0011] t<0 means that before operation, there is no voltage input, i.e., ui When t = 0, the voltage across the capacitor is 0V, and the current flowing through the capacitor is 0A, meaning the single-tuned filter is in a zero-state. When t ≥ 0, the input voltage u of the single-tuned filter is... i (t) is a constant value U0. Solving equations (1.1) to (1.3) simultaneously, the voltage u in the time domain can be obtained. c (t), current i s (t). For commercially available single-tuned filters, their u is measured. c (t) is relatively difficult.

[0012] Existing methods for identifying the characteristic parameters of single-tuned filters generally only become effective after an actual fault has occurred, failing to predict or prevent fault occurrence in advance, and cannot directly obtain the magnitude of the characteristic parameters within the single-tuned filter. Some identification methods only focus on the aging of capacitors, ignoring the aging of other components such as inductors and resistors. Existing nonlinear identification methods are highly dependent on the initial point, potentially yielding local optima rather than global optima, resulting in significant errors. Currently proposed direct identification methods have relatively weak anti-interference capabilities; at a noise / signal ratio of 15.5%, the relative error in characteristic parameter identification reaches approximately 2%. Many applications experience significant interference, with noise / signal ratios often exceeding 20%. Therefore, this invention proposes a reliable direct identification method for the characteristic parameters of single-tuned filters to address the problems existing in the prior art. Summary of the Invention

[0014] To address the aforementioned problems, the present invention aims to propose a reliable direct identification method for the characteristic parameters of a single-tuned filter. This method solves the problems of existing identification methods for the characteristic parameters of single-tuned filters, which often fail to predict or prevent faults in advance, cannot directly obtain the magnitude of the characteristic parameters within the single-tuned filter, ignore the aging of other components such as inductors and / or resistors, have weak anti-interference capabilities, and have large errors.

[0015] To achieve the objectives of this invention, the invention is implemented through the following technical solution: a reliable direct identification method for characteristic parameters of a single-tuned filter, comprising the following steps:

[0016] Step 1: Set the input of the single-tuned filter under test to zero to eliminate the energy stored in the filter and bring the filter to a zero state, that is, the current output and internal state of the filter are both equal to zero. Then prepare a DC voltage source with a constant amplitude of U0 and a current acquisition system.

[0017] Step 2: Apply the DC voltage source prepared in Step 1 to the input terminal of the single-tuned filter under test, and simultaneously use the prepared current acquisition system to acquire the filter current i.s (t), and save it, denoted as i(k) = i s (k·T s Let M be the current value collected in the k-th cycle, where k = 1, 2, 3, ..., T. s The sampling period is M, and the number of samples is M.

[0018] Step 3: Perform moving mean filtering on sample i(k), and denote the processed sequence as h(k), i.e. ω is the width of the filter window, k = 1 + 0.5ω, 2 + 0.5ω, 3 + 0.5ω, ..., M ‒ 0.5ω;

[0019] Step 4: Calculate the identification primitives σ(bp,ep,0, 0), σ(bp,ep,0, 1), σ(bp,ep,0, 3), σ(bp,ep,1, 1), and σ(bp,ep,1, 3) using h(k), where bp and ep are the starting and ending points of h(k) used to obtain the identification primitives, respectively.

[0020] Step 5: Calculate and identify the basic parameters u and v;

[0021] Step 6: Calculate the identification features α and β;

[0022] Step 7: Obtain the characteristic parameters of the single-tuned filter: resistor R, inductor L, and capacitor C.

[0023] A further improvement is made in step four, where the calculation formulas for the identification primitives σ(bp,ep,0, 0), σ(bp,ep,0, 1), σ(bp,ep,0, 3), σ(bp,ep,1, 1), and σ(bp,ep,1, 3) are as follows:

[0024]

[0025]

[0026]

[0027]

[0028] .

[0029] A further improvement is made in step five, where the calculation formulas for the identification of the basic parameters u and v are as follows:

[0030]

[0031] .

[0032] A further improvement is made in step six, where the formulas for calculating the identification features α and β are as follows:

[0033]

[0034]

[0035]

[0036] Where φ is the difference coefficient.

[0037] A further improvement is made in step seven, where the calculation formulas for the characteristic parameters of the single-tuned filter—resistance R, inductance L, and capacitance C—are as follows:

[0038]

[0039]

[0040]

[0041] ,

[0042]

[0043]

[0044]

[0045] Where p1 and p2 are the first and second characterization factors, respectively, A is the cumulative ratio parameter, and γ1 and γ2 are the first and second characterization parameters, respectively.

[0046] The beneficial effects of this invention are as follows: This invention uses a single-tuned filter as the detection object and monitors the characteristic parameters of its internal components in real time. A simulation model of the single-tuned filter is constructed based on the Matlab / Simulink platform, and the calculation method of the characteristic parameters of the filter's internal components is verified through experiments. Analysis of the experimental results confirms that the identification method of this invention, without disassembling or damaging the filter device, can accurately estimate the magnitude of the characteristic parameters of the single-tuned filter in an interference environment by real-time monitoring of the filter's current. It has the advantages of strong anti-interference capability, timely data acquisition, high accuracy, and no damage to the filter. Furthermore, it has high identification accuracy, reliable and stable identification, and is easy to operate. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the equivalent circuit model of a traditional single-tuned filter in the background art of this invention;

[0050] Figure 2 This is a flowchart illustrating the reliable direct identification method for the characteristic parameters of a single-tuned filter according to the present invention.

[0051] Figure 3 This is a schematic diagram of the experimental system architecture for a single-tuned filter circuit in an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0054] See Figure 1 , Figure 2 , Figure 3 In this embodiment, a circuit model of a single-tuned filter circuit is established in Simulink under the Matlab environment. Experiments are then conducted on this circuit model to verify the method of the present invention. The specific steps are as follows:

[0055] Step 1: First, establish a single-tuned filter circuit model, such as... Figure 3 As shown, the capacitor C = 0.001F, the inductor L = 0.0022H, the resistor R = 20Ω, the white noise gain is 0 (i.e., no noise interference input), and a step signal with an amplitude of 100000 controls the controlled voltage source to generate a step voltage source u0 = 100000V. i (t) is used for a single-tuned filter; the current sensor collects the current i of the single-tuned filter. s (t), the voltage across the capacitor is initialized to 0V, the inductor current is initialized to 0A, that is, the single-tuned filter is initialized to a zero state;

[0056] Step 2: Set the sampling period T s Set T=7.0×10 -7 Seconds, start simulation, ui (t) Connect to the input of a single-tuned filter, and simultaneously, the current sensor collects i s (t) and store it, denoted as i(j) = i s (jT s ), representing the j-th sampled value, j=1, 2, 3, ..., M, when M=2.1×10 5 When i(j) changes very little, the experiment ends;

[0057] Step 3: Perform moving mean filtering on sample i(k), and denote the processed sequence as h(k), i.e. ω is the width of the filter window, taken as ω=51, k =27, 28, 29, …, 209974;

[0058] Step 4: Calculate the identification primitives σ(bp,ep,0,0), σ(bp,ep,0,1), σ(bp,ep,0,3), σ(bp,ep,1,1), and σ(bp,ep,1,3) using h(k), where bp and ep are the starting and ending points of h(k) used to obtain the identification primitives, respectively. Take bp = 50 and ep = 500. The calculation formula for the identification primitives is as follows:

[0059]

[0060]

[0061]

[0062]

[0063] ;

[0064] Step 5: Calculate the basic identification parameters u and v. The calculation formula is as follows:

[0065]

[0066] ;

[0067] Step 6: Calculate the identification features α and β, using the following formula:

[0068]

[0069]

[0070] ;

[0071] Step 7: Obtain the characteristic parameters of the single-tuned filter: resistance R, inductance L, and capacitance C. The calculation formula is as follows:

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] .

[0080] Therefore, the single-tuned filter characteristic parameter identification method proposed in this invention has relative identification errors of 0.0004%, 0.0001%, and 0.0004% for R, L, and C in the specific experimental case, respectively. This demonstrates that the single-tuned filter characteristic parameter identification method proposed in this invention can reliably identify parameters in interference-free systems with high accuracy.

[0081] In practical applications, interference (typically manifested as white noise) is always present in both the input and acquired signals. To examine the impact of interference on the identification method proposed in this invention, in... Figure 3 The model shown incorporates a white noise signal module to simulate interference signals experienced by the system. The white noise module is set to a power spectral density of 0.01 and a seed number of 23341. By varying the noise amplitude, it simulates the interference intensity in a real-world working environment. Additional parameters are also set. Figure 3 The gain values ​​of the gainers in the experiment are 1000, 5000, 10000, 30000, 50000, 70000, and 100000, corresponding to noise / signal ratios of 1%, 5%, 10%, 30%, 50%, 70%, and 100% for the input signal, respectively. Similarly, each time the interference amplitude or noise / signal ratio is changed, steps one through seven are repeated to perform an identification experiment. Table 1 records the basic identification parameters, identification characteristic quantities, and variables p1, p2, and A from step seven, calculated based on the current responses in the experiment. Table 2 records the values ​​of the identified filter characteristic parameters: capacitance C, inductance L, and resistance R.

[0082] Table 1. Details of the parameters calculated in the experiment.

[0083]

[0084] Table 2 shows the specific details of the characteristic parameters of the single-tuned filter identified in the experiment.

[0085]

[0086] As shown in Table 2, the identification method proposed in this invention can stably and reliably identify the characteristic parameters of a single-tuned filter with comparable accuracy. While the identification error gradually increases with the increase of white noise amplitude or noise / signal ratio, the identification accuracy remains high, demonstrating stability, reliability, and robustness. The method exhibits the best identification performance in the absence of noise, demonstrating high accuracy and precision, with relative identification errors for R, L, and C all significantly less than 0.001%. When the interference amplitude gradually increases to 100,000V (i.e., a noise / signal ratio of 100%), the identification errors for each characteristic parameter gradually stabilize and remain below 3.15%. Overall, the identification method proposed in this invention demonstrates good robustness, strong anti-interference capability, high identification accuracy, and stable and reliable identification.

[0087] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A reliable direct identification method for characteristic parameters of a single-tuned filter, characterized in that, Includes the following steps: Step 1: Set the input of the single-tuned filter under test to zero to eliminate the energy stored in the filter and bring the filter to a zero state, that is, the current output and internal state of the filter are both equal to zero. Then prepare a DC voltage source with a constant amplitude of U0 and a current acquisition system. Step 2: Apply the DC voltage source prepared in Step 1 to the input terminal of the single-tuned filter under test, and simultaneously use the prepared current acquisition system to acquire the filter current i. s (t), and save it, denoted as i(k) = i s (k·T s () represents the current value collected in the k-th cycle, k=1,2,3,……, M,T s The sampling period is M, and the number of samples is M. Step 3: Perform moving mean filtering on sample i(k), and denote the processed sequence as h(k), i.e. ω is the width of the filter window, k = 1 + 0.5ω, 2 + 0.5ω, 3 + 0.5ω, ..., M ‒ 0.5ω; Step 4: Calculate the identification primitives σ(bp,ep,0,0), σ(bp,ep,0,1), σ(bp,ep,0,3), σ(bp,ep,1,1), and σ(bp,ep,1,3) using h(k), where bp and ep are the starting and ending point numbers of h(k) used to obtain the identification primitives, respectively. Step 5: Calculate and identify the basic parameters u and v; Step 6: Calculate the identification features α and β; Step 7: Obtain the characteristic parameters of the single-tuned filter: resistor R, inductor L, and capacitor C.

2. The reliable direct identification method for characteristic parameters of a single-tuned filter according to claim 1, characterized in that: In step four, the calculation formulas for the identification primitives σ(bp,ep,0,0), σ(bp,ep,0,1), σ(bp,ep,0,3), σ(bp,ep,1,1), and σ(bp,ep,1,3) are as follows: 。 3. The reliable direct identification method for characteristic parameters of a single-tuned filter according to claim 1, characterized in that: In step five, the formulas for calculating the basic identification parameters u and v are as follows: 。 4. The reliable direct identification method for characteristic parameters of a single-tuned filter according to claim 1, characterized in that: In step six, the formulas for calculating the identification features α and β are as follows: Where φ is the difference coefficient.

5. The reliable direct identification method for characteristic parameters of a single-tuned filter according to claim 1, characterized in that: In step seven, the formulas for calculating the characteristic parameters of the single-tuned filter—resistance R, inductance L, and capacitance C—are as follows: ， Where p1 and p2 are the first and second characterization factors, respectively, A is the cumulative ratio parameter, and γ1 and γ2 are the first and second characterization parameters, respectively.

Images