A Design Method of Bandstop Filter

Abstract

The present invention relates to a design method of a band-stop filter. The method obtains the elastic modal data of a rocket through environmental tests, analyzes the obtained elastic modal data of the rocket, determines the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate, establishes a model of the filter to be designed, establishes a corresponding full-coverage filter parameter value matrix table for each network parameter of the model of the filter to be designed, performs two-dimensional interpolation on each obtained full-coverage filter parameter value matrix table to generate a filter; performs a modal simulation experiment on the generated filter through the rocket attitude control system. This method reduces the number of ground tests for obtaining the natural frequencies of elastic motion before the rocket takes off, reduces manpower and material resources, improves the efficiency of the band-stop filter design process, has a high tolerance for the deviation of the natural frequencies of the elastic motion of the rocket, can achieve an optimal match with the actual frequencies of the launch vehicle, and improves the performance of the filter.

Description

Technical Field

[0001] The present invention relates to the field of attitude control of rockets, and in particular to a design method of a band-stop filter. Background Art

[0002] With the rise of commercial spaceflight, the demand for mass production and rapid launch of low-cost rockets has become increasingly prominent, and filters are essential devices on rockets.

[0003] Traditionally, the method for designing a band-stop filter is to obtain the natural frequency of the elastic motion of the launch vehicle through a large number of ground tests before the rocket takes off, and then use the one-dimensional interpolation method to obtain the parameters of the filter according to the real-time time t; the traditional filter design method has the following defects: 1. The number of ground tests for obtaining the natural frequency of the elastic motion of the launch vehicle before the rocket takes off is extremely large. Due to the complexity and high difficulty of the test process, it takes a lot of time, manpower and material resources; 2. Due to the differences between different individuals of the rocket body products, accurate data of a specific rocket cannot be obtained. Coupled with the differences between the ground and the flight state of the rocket, there are large deviations in the natural frequency of the elastic motion of the launch vehicle obtained by the traditional test method. The deviation is usually about 5%. In recent years, due to the extensive use of composite materials represented by carbon fiber in the rocket structure, this deviation often expands to 10% - 30%, bringing great difficulties to attitude control; 3. The tolerance range for the deviation of the natural frequency of the elastic motion is at most 10%. Beyond this range, the stability of the filter cannot be guaranteed; 4. The center frequency of the designed band-stop filter is a fixed value, and it is difficult to adapt to a larger deviation range exceeding 10%; 5. The designed filter and the actual frequency of the launch vehicle are not necessarily the optimal match, there is a risk. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a design method of a band-stop filter that can reduce the number of ground tests, has a high tolerance for the deviation of the natural frequency of the elastic motion of the rocket, and can achieve the optimal match with the actual frequency of the launch vehicle.

[0005] A design method of a band-stop filter, the method includes the following steps:

[0006] S1. Obtain the natural frequency of the elastic motion of the rocket at different time points through environmental tests, and the natural frequency of the elastic motion obtained from the environmental tests and the time points form an elastic modal data table in one-to-one correspondence;

[0007] S2. Analyze the elastic modal data table of the rocket obtained in step S1 to determine the range of the natural frequency of the elastic motion of the rocket that the filter to be designed needs to accommodate;

[0008] S3. Establish a model of the filter to be designed, and set the network parameters of the model of the filter to be designed as A0, A1, A2, B1, and B2. According to the elastic mode data table obtained in step S1 and the range of the natural frequencies of the rocket's elastic motion that the filter to be designed needs to accommodate obtained in step S2, establish a corresponding full-coverage filtering parameter value matrix table for each network parameter of the model of the filter to be designed;

[0009] S4. Real-time collect the natural frequency f_i of the rocket's elastic motion and the corresponding current flight point time t of the natural frequency f_i of the elastic motion. According to the natural frequency f_i of the elastic motion and the corresponding current flight point time t, perform two-dimensional interpolation on each full-coverage filtering parameter value matrix table obtained in step S3, and obtain the values of each network parameter of the model of the filter to be designed through two-dimensional interpolation, so as to generate a filter;

[0010] S5. Build a rocket attitude control system according to the filter generated in step S4, perform a modal simulation experiment on the filter generated in step S4 through the rocket attitude control system, and judge whether the filter generated in step S4 meets the design requirements according to the results of the modal simulation experiment. If the results of the modal simulation experiment are within the index range, then the filter generated in step S4 meets the design requirements. If the results of the modal simulation experiment are not within the index range, then the filter generated in step S4 does not meet the design requirements, and return to step S4 to regenerate the filter.

[0011] The beneficial effects of the present invention are as follows: By adopting the above-mentioned band-stop filter design method, the number of ground tests for obtaining the natural frequency of elastic motion before the rocket takes off is reduced, the manpower and material resources are reduced, and the efficiency of the band-stop filter design process is improved; The filter designed by adopting the above-mentioned band-stop filter design method has a high tolerance for the deviation of the natural frequency of the rocket's elastic motion, ensures the stability of the filter, and can achieve the optimal matching with the actual frequency of the launch vehicle, thereby improving the performance of the filter.

[0012] Preferably, in step S3, the expression of the model of the filter to be designed is:

[0013] Among them, represents the pitch attitude angle deviation without passing through the filter, Δψ represents the yaw attitude angle deviation before entering the filter, and the represents the pitch attitude angle deviation output after passing through the filter, Δψ lb represents the yaw attitude angle deviation after entering the filter, and the A0, A1, A2, B1, and B2 represent the network parameters of the model of the filter to be designed.

[0014] Preferably, in step S1, the elastic modal data table is an n-row and two-column matrix, where n represents the number of time points; in the two columns of data in the matrix, one column of data is the time point, and the other column of data is the natural frequency of elastic motion corresponding to the time point; in step S3, according to the elastic modal data table obtained in step S1 and the range of the natural frequency of rocket elastic motion that the filter to be designed needs to accommodate obtained in step S2, for each network parameter of the model of the filter to be designed, the specific process of establishing a corresponding full-coverage filter parameter value matrix table includes the following steps:

[0015] S3.1. According to the range of the natural frequency of rocket elastic motion that the filter to be designed needs to accommodate obtained in step S2, expand the elastic modal data table into an n-row and m-column matrix [T, W], where T represents a time point of n rows and 1 column, and W represents a frequency matrix M of n rows and m - 1 columns. w ij represents the natural frequency of elastic motion at the i-th row and j-th column in the matrix M.

[0016] S3.2. Design the continuous transfer function of the filter according to the n-row and m-column matrix obtained in step S3.1. The continuous transfer function of the filter is expressed as: where ξ1 and ξ2 are the damping ratios of the filter, and S represents the transfer function symbol; thus, i * j continuous transfer functions of the filter are obtained.

[0017] S3.3. Discretize the i * j continuous transfer functions of the filter obtained in step S3.2 to obtain the values of the discrete form of the network parameter A0, the values of the discrete form of the network parameter A1, the values of the discrete form of the network parameter A2, the values of the discrete form of the network parameter B1, and the values of the discrete form of the network parameter B2.

[0018] S3.4. The values of the discrete form of the network parameter A0 form the full-coverage filter parameter value matrix table corresponding to the network parameter A0, the values of the discrete form of the network parameter A1 form the full-coverage filter parameter value matrix table corresponding to the network parameter A1, the values of the discrete form of the network parameter A2 form the full-coverage filter parameter value matrix table corresponding to the network parameter A2, the values of the discrete form of the network parameter B1 form the full-coverage filter parameter value matrix table corresponding to the network parameter B1, and the values of the discrete form of the network parameter B2 form the full-coverage filter parameter value matrix table corresponding to the network parameter B2.

[0019] Preferably, in step S5, the rocket attitude control system includes a filter generated in step S4, a correction network connected to the filter generated in step S4, an actuator connected to the correction network, a rocket connected to the actuator, and a sensing device connected to the rocket. The rated state of the rocket attitude control system needs to satisfy that the elastic amplitude range is greater than 10 dB, and the offset state needs to satisfy that the elastic amplitude range is greater than 6 dB.

[0020] Preferably, in step S5, a modal simulation experiment is performed on the filter generated in step S4 through the rocket attitude control system. The specific process of judging whether the filter generated in step S4 meets the design requirements according to the results of the modal simulation experiment includes the following steps:

[0021] S5.1. Sense the rocket attitude information of the actual movement of the rocket during flight through the sensing device;

[0022] S5.2. Compare the rocket attitude information obtained in step S5.1 with the standard program angle to obtain the deviation between the two;

[0023] S5.3. Input the deviation obtained in step S5.2 into the filter generated in step S4, and the filter filters the high-frequency elastic signal in the deviation to obtain the filtered deviation;

[0024] S5.4. Input the filtered deviation obtained in step S5.3 into the correction network for amplitude and phase compensation to obtain a control command;

[0025] S5.5. Judge whether the control command obtained in step S5.4 contains the high-frequency elastic signal. If it contains, it means that the generated filter does not meet the design requirements, and return to step S4 to regenerate the filter; if it does not contain, it means that the generated filter meets the design requirements, and the actuator tracks the control command to control the rocket to perform corresponding deflection actions. When the rocket performs corresponding deflection actions, return to step S5.1 to complete the closed-loop control of the rocket attitude control system. Description of the Drawings

[0026] Figure 1 It is a flowchart of a design method of a band-stop filter according to the present invention;

[0027] Figure 2 It is a system diagram of a rocket attitude control system according to the present invention. Detailed Embodiments

[0028] The following further describes the invention with reference to the drawings and in combination with specific embodiments, so that those skilled in the art can implement it according to the description in the specification. The protection scope of the present invention is not limited to this specific embodiment.

[0029] An embodiment of the present invention provides a method for designing a band-stop filter, as Figure 1 shown. The method includes the following steps:

[0030] S1. Obtain the natural frequencies of the elastic motion of the rocket at different time points through environmental tests. The natural frequencies of the elastic motion obtained from the environmental tests and the time points correspond one by one to form an elastic mode data table;

[0031] S2. Analyze the elastic mode data table of the rocket obtained in step S1 to determine the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate;

[0032] S3. Establish a model of the filter to be designed. The network parameters of the model of the filter to be designed are set as A0, A1, A2, B1, and B2. According to the elastic mode data table obtained in step S1 and the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate obtained in step S2, establish a corresponding full-coverage filter parameter value matrix table for each network parameter of the model of the filter to be designed;

[0033] S4. Real-time collect the natural frequency f_i of the elastic motion of the rocket and the current flight point time t corresponding to the natural frequency f_i of the elastic motion. Perform two-dimensional interpolation on each full-coverage filter parameter value matrix table obtained in step S3 according to the natural frequency f_i of the elastic motion and the corresponding current flight point time t. Obtain the values of each network parameter of the model of the filter to be designed through two-dimensional interpolation, thereby generating a filter;

[0034] S5. Build a rocket attitude control system according to the filter generated in step S4. Perform a modal simulation experiment on the filter generated in step S4 through the rocket attitude control system. Judge whether the filter generated in step S4 meets the design requirements according to the results of the modal simulation experiment. If the results of the modal simulation experiment are within the index range, then the filter generated in step S4 meets the design requirements. If the results of the modal simulation experiment are not within the index range, then the filter generated in step S4 does not meet the design requirements, and return to step S4 to regenerate the filter.

[0035] A method for designing a band-stop filter provided in this embodiment. The band-stop filter designed by this method reduces the number of ground tests for obtaining the natural frequency of elastic motion before the rocket takes off compared to the original design method, reduces manpower and material resources, and improves the efficiency of the band-stop filter design process. For the filter designed by the above method for designing a band-stop filter, a range that the natural frequency of rocket elastic motion needs to accommodate is determined. The rocket has a high tolerance for the deviation of the natural frequency of elastic motion, ensuring the stability of the filter. By designing a full-coverage filter parameter value matrix table and performing two-dimensional interpolation on the full-coverage filter parameter value matrix table according to the real-time natural frequency of elastic motion \(f_i\) of the rocket and the corresponding current flight point time \(t\), the designed filter can achieve an optimal match with the actual frequency of the launch vehicle, improving the performance of the filter.

[0036] In step S3, the expression of the model of the filter to be designed is:

[0037] Wherein, represents the pitch attitude angle deviation without passing through the filter, \(\Delta\psi\) represents the yaw attitude angle deviation without passing through the filter, and the represents the pitch attitude angle deviation output after passing through the filter, \(\Delta\psi\) lb represents the yaw attitude angle deviation output after passing through the filter, and \(A0\), \(A1\), \(A2\), \(B1\), \(B2\) represent the network parameters of the model of the filter to be designed.

[0038] In step S1, the elastic mode data table is a matrix with \(n\) rows and two columns. Among them, \(n\) represents the number of time points; in the two columns of data in the matrix, one column of data is the time point, and the other column of data is the natural frequency of elastic motion corresponding to the time point. In step S3, according to the elastic mode data table obtained in step S1 and the range of the natural frequency of rocket elastic motion that the filter to be designed needs to accommodate obtained in step S2, the specific process of establishing a corresponding full-coverage filter parameter value matrix table for each network parameter of the model of the filter to be designed includes the following steps:

[0039] S3.1. According to the range of the natural frequency of rocket elastic motion that the filter to be designed needs to accommodate obtained in step S2, expand the elastic mode data table into a matrix \([T, W]\) with \(n\) rows and \(m\) columns, where \(T\) represents a time point with \(n\) rows and 1 column, and \(W\) represents a frequency matrix \(M\) with \(n\) rows and \(m - 1\) columns. w ijIt represents the elastic motion natural frequency at the i-th row and j-th column within the matrix M; for example: in the elastic mode data table obtained in step S1, the first row and first column is time T = 0, and the first row and second column is the natural frequency w0. Then this row of the array can be extended to the array [0, w0 - 0.1*k, …, w0 - 0.1*2, w0 - 0.1, w0, w0 + 0.1, w0 + 0.1*2, …, w0 + k*0.1], where k is a constant, and the number of elements in the extended array is m.

[0040] S3.2. Design the continuous transfer function of the filter according to the n×m matrix obtained in step S3.1. The continuous transfer function of the filter is expressed as: where ξ1 and ξ2 are the damping ratios of the filter, which are determined by the specific filtering depth and width; S represents the transfer function symbol; thus, i×j continuous transfer functions of the filter are obtained. In this embodiment, the center frequency of the filter to be designed is 9 Hz, the frequency range is 6 Hz to 12 Hz, the interval is 0.1 Hz, and the number of selected natural frequencies is 60. The flight time range is set to 0 to 100 s.

[0041] S3.3. Discretize the i×j continuous transfer functions of the filter obtained in step S3.2 to obtain the values of the discrete form of the network parameter A0, the values of the discrete form of the network parameter A1, the values of the discrete form of the network parameter A2, the values of the discrete form of the network parameter B1, and the values of the discrete form of the network parameter B2. Each network parameter has 60×100 values.

[0042] S3.4. The values of the discrete form of the network parameter A0 form the full-coverage filtering parameter value matrix table corresponding to the network parameter A0, as shown in Table 1. The values of the discrete form of the network parameter A1 form the full-coverage filtering parameter value matrix table corresponding to the network parameter A1, as shown in Table 2. The values of the discrete form of the network parameter A2 form the full-coverage filtering parameter value matrix table corresponding to the network parameter A2, as shown in Table 3. The values of the discrete form of the network parameter B1 form the full-coverage filtering parameter value matrix table corresponding to the network parameter B1, as shown in Table 4. The values of the discrete form of the network parameter B2 form the full-coverage filtering parameter value matrix table corresponding to the network parameter B2, as shown in Table 5.

[0043] Table 1 Full-coverage filtering parameter value matrix table corresponding to A0

[0044]

[0045]

[0046] Table 2 Full-coverage filtering parameter value matrix table corresponding to A1

[0047]

[0048] Table 3 Full-coverage filtering parameter value matrix table corresponding to A2

[0049]

[0050] Table 4 Full-coverage filtering parameter value matrix table corresponding to B1

[0051]

[0052] Table 5 Full-coverage filtering parameter value matrix table corresponding to B2

[0053]

[0054]

[0055] In the above five full-coverage filtering parameter value matrix tables, the frequency values on the abscissa are from the frequency values within the inherent frequency range of the rocket elastic motion that the filter to be designed needs to accommodate determined in step S2, the vertical on the ordinate is the corresponding time point, and the set flight time range is 0 to 100 s.

[0056] In step S5, as Figure 2 shown, the rocket attitude control system includes the filter generated in step S4, a correction network connected to the filter generated in step S4, an actuator connected to the correction network, a rocket connected to the actuator, and a sensing device connected to the rocket. The rated state of the rocket attitude control system needs to satisfy that the elastic amplitude range is greater than 10 dB, and the pulled-off state needs to satisfy that the elastic amplitude range is greater than 6 dB.

[0057] In step S5, the specific process of performing a modal simulation experiment on the filter generated in step S4 through the rocket attitude control system and judging whether the filter generated in step S4 meets the design requirements according to the results of the modal simulation experiment includes the following steps:

[0058] S5.1. Sense the rocket attitude information of the actual motion of the rocket during flight through the sensing device;

[0059] S5.2. Compare the rocket attitude information obtained in step S5.1 with the standard program angle to obtain the deviation amount between the two;

[0060] S5.3. Input the deviation amount obtained in step S5.2 into the filter generated in step S4, and the filter filters the high-frequency elastic signals in the deviation amount to obtain the filtered deviation amount;

[0061] S5.4. Input the filtered deviation obtained in step S5.3 into the correction network for amplitude and phase compensation to obtain a control command;

[0062] S5.5. Determine whether the control command obtained in step S5.4 contains the high-frequency elastic signal. If it does, it indicates that the generated filter does not meet the design requirements, and return to step S4 to regenerate the filter. If it does not, it indicates that the generated filter meets the design requirements. The actuator tracks the control command to control the rocket to perform corresponding deflection actions. When the rocket performs corresponding deflection actions, return to step S5.1 to complete the closed-loop control of the rocket attitude control system.

Claims

1. A method for designing a band-stop filter, characterized in that: The method includes the following steps: S1. Obtain the natural frequencies of the elastic motion of the rocket at different time points through environmental tests. The natural frequencies of the elastic motion obtained from the environmental tests and the time points correspond one by one to form an elastic mode data table. S2. Analyze the elastic mode data table of the rocket obtained in step S1 to determine the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate. S3. Establish a model of the filter to be designed. The network parameters of the model of the filter to be designed are set as A0, A1, A2, B1, and B2. According to the elastic mode data table obtained in step S1 and the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate obtained in step S2, a corresponding full-coverage filtering parameter value matrix table is established for each network parameter of the model of the filter to be designed. S4. Real-time collect the natural frequency fi of the elastic motion of the rocket and the current flight point time t corresponding to the natural frequency fi of the elastic motion. According to the natural frequency fi of the elastic motion and the corresponding current flight point time t, perform two-dimensional interpolation on each full-coverage filtering parameter value matrix table obtained in step S3. The values of each network parameter of the model of the filter to be designed are obtained through two-dimensional interpolation, thereby generating a filter. S5. Build a rocket attitude control system according to the filter generated in step S4. Perform a modal simulation experiment on the filter generated in step S4 through the rocket attitude control system. According to the results of the modal simulation experiment, determine whether the filter generated in step S4 meets the design requirements. If the results of the modal simulation experiment are within the index range, then the filter generated in step S4 meets the design requirements. If the results of the modal simulation experiment are not within the index range, then the filter generated in step S4 does not meet the design requirements, and return to step S4 to regenerate the filter.

2. The design method of a band-stop filter according to claim 1, wherein: In step S3, the expression of the model of the filter to be designed is as follows: Wherein, represents the pitch attitude angle deviation without passing through the filter, Δψ represents the yaw attitude angle deviation before entering the filter, and the represents the pitch attitude angle deviation output after passing through the filter, and Δψ lb represents the yaw attitude angle deviation after entering the filter. A0, A1, A2, B1, and B2 represent the network parameters of the model of the filter to be designed.

3. A method for designing a band-stop filter according to claim 2, characterized in that: In step S1, the elastic mode data table is a matrix with n rows and two columns, where n represents the number of time points. In the two columns of data of the matrix, one column of data is the time point, and the other column of data is the natural frequency of the elastic motion corresponding to the time point. In step S3, according to the elastic mode data table obtained in step S1 and the range of the natural frequencies of the elastic motion of the rocket that the filter to be designed needs to accommodate obtained in step S2, the specific process of establishing a corresponding full-coverage filtering parameter value matrix table for each network parameter of the model of the filter to be designed includes the following steps: S3.

1. According to the range of the natural frequencies of the rocket elastic motion that the filter to be designed needs to accommodate obtained in step S2, expand the elastic mode data table into an n-row and m-column matrix [T, W], where T represents the time points in a 1-column matrix of n rows, and W represents the frequency matrix M in an (m-1)-column matrix of n rows. ω ij represents the natural frequency of the i-th row and j-th column in the matrix M; S3.

2. Design a continuous transfer function of the filter according to the n×m matrix obtained in step S3.

1. The continuous transfer function of the filter is expressed as: where ξ1 and ξ2 are the damping ratios of the filter, and S represents the transfer function symbol. Thus, i×j continuous transfer functions of the filter are obtained. S3.

3. Discretize the i*j filter continuous transfer functions obtained in step S3.2 to obtain the values of the discrete form of the network parameter A0, the values of the discrete form of the network parameter A1, the values of the discrete form of the network parameter A2, the values of the discrete form of the network parameter B1, and the values of the discrete form of the network parameter B2. S3.

4. The discrete values of the network parameter A0 form the full-coverage filter parameter value matrix table corresponding to the network parameter A0, the discrete values of the network parameter A1 form the full-coverage filter parameter value matrix table corresponding to the network parameter A1, the discrete values of the network parameter A2 form the full-coverage filter parameter value matrix table corresponding to the network parameter A2, the discrete values of the network parameter B1 form the full-coverage filter parameter value matrix table corresponding to the network parameter B1, and the discrete values of the network parameter B2 form the full-coverage filter parameter value matrix table corresponding to the network parameter B2.

4. A method for designing a band-stop filter according to claim 3, characterized in that: In step S5, the rocket attitude control system includes the filter generated in step S4, a correction network connected to the filter generated in step S4, an actuator connected to the correction network, a rocket connected to the actuator, and a sensing device connected to the rocket. The rated state of the rocket attitude control system needs to satisfy that the elastic amplitude range is greater than 10 dB, and the offset state needs to satisfy that the elastic amplitude range is greater than 6 dB.

5. A method for designing a band-stop filter according to claim 4, characterized in that: In step S5, a modal simulation experiment is carried out on the filter generated in step S4 through the rocket attitude control system. The specific process of judging whether the filter generated in step S4 meets the design requirements according to the results of the modal simulation experiment includes the following steps: S5.

1. The sensing device senses the rocket attitude information of the actual movement of the rocket during flight. S5.

2. The rocket attitude information obtained in step S5.1 is compared with the standard program angle to obtain the deviation between the two. S5.

3. The deviation obtained in step S5.2 is input into the filter generated in step S4, and the filter filters the high-frequency elastic signal in the deviation to obtain the filtered deviation. S5.

4. The filtered deviation obtained in step S5.3 is input into the correction network for amplitude and phase compensation to obtain a control command. S5.

5. Judge whether the control command obtained in step S5.4 contains the high-frequency elastic signal. If it contains, it means that the generated filter does not meet the design requirements, and return to step S4 to regenerate the filter. If it does not contain, it means that the generated filter meets the design requirements. The actuator tracks the control command to control the rocket to perform corresponding deflection actions. When the rocket performs corresponding deflection actions, return to step S5.1 to complete the closed-loop control of the rocket attitude control system.

Images