Wind speed stabilizing control method for open type high-speed wind tunnel

Abstract

The invention discloses a steady wind speed control method for an open-type high-speed wind tunnel, and the method comprises the steps: opening a pneumatic control valve to a certain opening degree through the calculation of a pre-opening degree, then instantaneously opening a rapid valve, releasing the pressure of a gas storage tank, generating high-speed airflow, and rapidly building a flow field; collecting data during wind tunnel operation in real time, filtering the data and inputting the data into a control system; the Mach number of a test section is calculated in real time according to data collected in real time during operation of a wind tunnel, the optimal control quantity of a pneumatic control valve is calculated through predictive control, and the optimal control quantity is weighted to participate in final output of fuzzy control and PI control; an error and an error change rate between the actually measured Mach number and an expected Mach number are calculated, when the error is larger than or equal to 0.05, fuzzy control is used for adjustment, the error and the error change rate serve as input of fuzzy control, and control output is generated through fuzzy rule reasoning to preliminarily adjust a wind tunnel flow field; and when the error is less than 0.05, further adjusting according to the feedback error of the control system by using a position type PI control algorithm.

Description

Technical Field

[0001] The invention relates to a method for controlling steady wind speed of an open high-speed wind tunnel. Background Art

[0002] High-speed wind tunnels are key experimental equipment for studying the aerodynamic characteristics of aircraft in high-speed airflow environments. Traditional high-speed wind tunnels are mostly used to test large aircraft or complex engineering projects. Their control systems are complex and costly, making it difficult to meet the needs of small scientific research laboratories or teaching. However, small open-type high-speed wind tunnels are gradually becoming an important tool for universities and research institutions to conduct aerodynamic experiments due to their small footprint and low cost. The current open-type high-speed wind tunnel control system usually adopts a simple open-loop control or semi-automatic control method, which has the following problems:

[0003] 1) Poor flow field control accuracy: The existing system cannot accurately maintain the required Mach number when adjusting the gas source pressure and the opening of the pneumatic control valve, which easily causes fluctuations in experimental conditions and affects the accuracy of the data.

[0004] 2) Slow response speed: Traditional control methods have a slow response speed during the flow field establishment and stabilization process, making it difficult to quickly reach the target Mach number, reducing the experimental efficiency.

[0005] Therefore, it is necessary to develop a control method that integrates high-precision control, fast response and efficient data processing functions to make up for the shortcomings of existing technologies and meet the special needs of open high-speed wind tunnel control. Summary of the invention

[0006] The present invention provides a method for controlling the steady wind speed of an open high-speed wind tunnel in order to solve the problems existing in the prior art.

[0007] The technical solutions adopted in the present invention are:

[0008] A method for controlling the steady wind speed of an open high-speed wind tunnel, wherein the wind tunnel comprises a stable section and a test section, wherein the test section is located downstream of the stable section, and a fast valve and a pneumatic regulating valve are arranged in the flow direction upstream of the stable section, wherein the method comprises the following steps:

[0009] S1: Use an air compressor and an air tank to provide the high-pressure air source required for wind tunnel operation. The pneumatic regulating valve is opened to a certain opening through pre-opening calculation, and then the fast valve is opened instantaneously to release the pressure of the air tank to generate high-speed airflow and quickly establish the flow field.

[0010] S2: Collect wind tunnel operation data in real time, filter the collected data in real time using the set filtering algorithm, and input the filtered data into the control system;

[0011] S3: The control system calculates the Mach number of the test section in real time based on the real-time data collected during wind tunnel operation, calculates the optimal control amount of the pneumatic control valve through predictive control, and weights the optimal control amount to participate in the final output of fuzzy control and PI control;

[0012] S4: Calculate the error and error change rate between the actual measured Mach number and the expected Mach number set before the experiment. When the error is ≥0.05, use fuzzy control to make adjustments. The error and error change rate are used as the input of fuzzy control. The control output is generated through fuzzy rule reasoning to make preliminary adjustments to the wind tunnel flow field. When the error is <0.05, use the position PI control algorithm to further fine-tune according to the feedback error of the control system.

[0013] Furthermore, the formula for calculating the pre-opening degree is:

[0014]

[0015] Among them, Ma is the expected Mach number of the experiment; P is the gas source pressure measured before the experiment; K is the value of the pre-opening obtained before the experiment.

[0016] Furthermore, the data during wind tunnel operation include the total pressure of the stable section and the static pressure data of the test section. The Mach number is calculated by the total pressure of the stable section and the static pressure data of the test section. The calculation formula is:

[0017]

[0018] Among them, P0 is the total pressure in the stable section, P S is the static pressure of the test section.

[0019] Furthermore, in S2, the filtering algorithm is set as follows: the collected data is processed by combining low-pass filtering with median filtering. First, the high-frequency noise in the data is removed by low-pass filtering, and then the remaining noise is further removed by median filtering. The formula is as follows:

[0020]

[0021] Where: x(ti) represents the total pressure and static pressure data during wind tunnel operation, N represents the order of the low-pass filter, h(i) is the coefficient of the low-pass filter, K represents the size of the median filter window, (tK),...,(t+K) represents a sliding window, and y[t] is the filtered total pressure and static pressure data.

[0022] Furthermore, the predictive control in S3 includes:

[0023] First, in predictive control, a prediction model is established, and its formula is as follows:

[0024] x(k+1)=A·x(k)+B·u(k)+B·v(k)

[0025] Among them, x(k) represents the current state, that is, the current Mach number of the test section; u(k) represents the control input, that is, the opening of the pneumatic control valve; v(k) represents the disturbance; A represents the state transfer matrix, that is, the state change process from time k to time k+1; B represents the control input matrix, B the disturbance transfer matrix;

[0026] Then predict the state of the next N steps, the formula is as follows:

[0027]

[0028] in, represents the control system state at the prediction time (k+i); x(k) represents the current state, and u(k+j) represents the control input sequence;

[0029] Then define the cost function, the formula is as follows:

[0030]

[0031] Among them, x ref (k+i) represents the Mach number that should be reached at time k+i; Q represents the weight matrix; u(k+i) represents the control action at time k+i; Ru represents the weight matrix; γ represents the trade-off factor;

[0032] Finally, the cost function is minimized under the condition of satisfying the constraints through quadratic programming to obtain the optimal control input sequence:

[0033]

[0034] The first control quantity u*(k) of the optimization result is extracted as the optimal control quantity of the pneumatic control valve calculated by predictive control, that is, the final output of predictive control, u mpc (k) = u * (k).

[0035] Furthermore, in S4, when fuzzy control is used, the error and error change rate are calculated from the expected Mach number and the data collected during wind tunnel operation. The error calculation formula is:

[0036] e(t)=Ma ref -M(t)

[0037] Where e(t) is the error in the Mach number. ref is the expected Mach value, M(t) is the actual Mach value collected and filtered during wind tunnel operation;

[0038] The error rate is as follows:

[0039]

[0040] Among them, e(t-1) is the Mach number error at the previous moment, e(t) is the Mach number error at the current moment, Δt is the sampling period, and ed(t) is the rate of change of the Mach number error. When the sampling period is constant, Δt can be omitted and simplified to:

[0041] ed(t)=e(t)-e(t-1);

[0042] The Mach number error e(t) and the error change rate ed(t) are normalized, and the formula is as follows:

[0043] Normalized error:

[0044]

[0045] Normalized error rate of change:

[0046]

[0047] in, is the normalized value of the Mach number error, and e(t) is the Mach number error e max With e min are the maximum and minimum values ​​of the error, is the normalized value of the error change rate, ed(t) is the error change rate, ed max With ed min is the maximum and minimum value of the error change rate, r max With r min are the maximum and minimum values ​​of the target normalized range;

[0048] According to the normalized values ​​of the error and the error change rate, the fuzzy subsets are divided into negative large NB, negative small NS, zero ZE, positive small PS, and positive large PB. The fuzzy subsets calculate the fuzzy membership according to the normalized input error and error change rate, and calculate the fuzzy output according to the conditions of the fuzzy rules. The fuzzy rule formula is:

[0049] U output =min(u inpute ,u outputed )

[0050] Among them, u inpute and u outputed are the membership degrees of the fuzzy subsets of the Mach number error and the Mach number error change rate, respectively. output Then it is the membership degree of fuzzy output;

[0051] Finally, the fuzzy output is defuzzified by weighted average to obtain ufc , and its calculation formula is as follows:

[0052]

[0053] Among them, y i is the output of the ith rule, u i is the membership degree corresponding to the i-th rule, and n is the total number of rules;

[0054] The predictive control output is added as a weighted term to the final output of the fuzzy control, and the formula is as follows:

[0055] u fc =α(k)·u fc +γ·u mpc +k

[0056] Among them, u fc is the fuzzy output control quantity, α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

[0057] Furthermore, in S4, when the position PI control algorithm is used for control, the formula is as follows:

[0058] u pi =k p ×e(t)+k i ×∫e(t)dt

[0059] Among them, u pi is the control quantity, e(t) is the error, ∫e(t)dt is the integral term of the error, k p is the proportional gain, k i is the integral gain;

[0060] The predictive control output is added as a weighted term to the final output of the position PI control, and the formula is as follows:

[0061] u pi =α(k)·u pi +γ·u mpc +k

[0062] Among them, u pi is the final output control quantity of the position PI, α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

[0063] The present invention has the following beneficial effects:

[0064] The present invention adopts a combination of predictive control plus fuzzy control and PI control. The fuzzy control effectively shortens the response time of the system and gradually stabilizes the flow field. The PI control further stabilizes the flow field, thereby improving the control accuracy. In addition, the control quantity of the predictive control is used as a reference input, and the weighted strategy is combined with the output of the fuzzy control and PI control, so that the system can still operate stably under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a structural diagram of the wind tunnel of the present invention.

[0066] Figure 2 This is a control flow chart of the present invention.

[0067] Figure 3 This is the control principle diagram of the present invention.

[0068] Figure 4 This is a diagram showing the results of implementing the present invention.

[0069] Figure 5 This is a diagram showing the results of implementing the present invention. DETAILED DESCRIPTION

[0070] The present invention will be further described below in conjunction with the accompanying drawings.

[0071] like Figure 1 As shown, the pneumatic structure of the wind tunnel consists of multiple components. The wind tunnel includes a stable section 1 and a test section 2. The test section 2 is located downstream of the stable section 1. A quick valve 3 and a pneumatic regulating valve 4 are set in the flow direction upstream of the stable section. The air compressor 5 cooperates with the air storage tank 6 to provide the high-pressure air source required for the operation of the wind tunnel.

[0072] like Figure 2 As shown, the operation steps of the entire system are:

[0073] S1: The calculation of Mach number is based on the total pressure of the stable section and the static pressure data of the test section. The calculation formula is as follows:

[0074]

[0075] Among them, P0 is the total pressure in the stable section, P S is the static pressure of the test section.

[0076] S2: The collected data is processed by combining low-pass filtering and median filtering. First, low-pass filtering is used to remove high-frequency noise in the signal, and then the remaining noise is further removed by median filtering. The formula is as follows:

[0077]

[0078] Where x(ti) represents the total pressure and static pressure data during wind tunnel operation, N represents the order of the low-pass filter, h(i) is the coefficient of the low-pass filter, K represents the size of the median filter window, (tK),...,(t+K) is a sliding window, and y[t] is the filtered total pressure and static pressure data.

[0079] S3: Control principle block diagram as shown Figure 3 As shown, predictive control is first used to calculate the optimal output based on state feedback and control objectives. In predictive control, a prediction model is first established, and its formula is as follows:

[0080] x(k+1)=A·x(k)+B·u(k)+B·v(k)

[0081] Where x(k) represents the current state, u(k) represents the control input (opening of the pneumatic control valve), v(k) represents the disturbance, A represents the state transfer matrix (the state change process of the system from time k to time k+1), B represents the control input matrix, and B represents the disturbance transfer matrix.

[0082] Then predict the state of the next N steps, the formula is as follows:

[0083]

[0084] in represents the system state at the prediction time (k+i). x(k) represents the current state, and u(k+j) represents the control input sequence.

[0085] Then define the cost function, the formula is as follows:

[0086]

[0087] where x ref (k+i) represents the Mach number that should be reached at time k+i, Q represents the weight matrix, u(k+i) represents the control action at time k+i, Ru represents the weight matrix, and γ represents the trade-off factor.

[0088] Finally, the cost function is minimized under the condition of satisfying the constraints through quadratic programming to obtain the optimal control input sequence:

[0089]

[0090] Extract the first control quantity u of the optimization result * (k) as the final output, i.e. u mpc (k) = u * (k)

[0091] S4: Fuzzy control is used when the error is large. The error and error change rate of the fuzzy control system are calculated from the expected Mach number and the data collected during wind tunnel operation. The error calculation formula is:

[0092] e(t)=Ma ref -M(t)

[0093] e(t) is the error in the Mach number, Ma ref is the expected Mach value, and M(t) is the actual Mach value collected and filtered during wind tunnel operation.

[0094] The error rate is as follows:

[0095]

[0096] Where e(t-1) is the Mach number error at the previous moment, e(t) is the Mach number error at the current moment, Δt is the sampling period, and ed(t) is the rate of change of the Mach number error. When the sampling period is constant, Δt can be omitted, and it can be simplified to:

[0097] ed(t)=e(t)-e(t-1)

[0098] The Mach number error e(t) and the error change rate ed(t) are normalized, and the formula is as follows:

[0099] Normalized error:

[0100]

[0101] Normalized error rate of change:

[0102]

[0103] in is the normalized value of the Mach number error, and e(t) is the Mach number error e max With e min are the maximum and minimum values ​​of the error, is the normalized value of the error change rate, ed(t) is the error change rate, ed max With ed min is the maximum and minimum value of the error change rate, r max With r min are the maximum and minimum values ​​of the target normalized range.

[0104] According to the normalized values ​​of the error and the error change rate, the fuzzy subsets are divided into NB (negative large), NS (negative small), ZE (zero), PS (positive small), and PB (positive large). The fuzzy subsets can calculate the fuzzy membership according to the normalized input error and error change rate, and calculate the fuzzy output according to the conditions of the fuzzy rules. The fuzzy rule formula is:

[0105] U output =min(u inpute ,u outputed )

[0106] where u inpute and u outputed are the membership degrees of the fuzzy subsets of the Mach number error and the Mach number error change rate, respectively. output is the membership degree of the fuzzy output.

[0107] Defuzzify the fuzzy output by weighted average to get u fc , and its calculation formula is as follows:

[0108]

[0109] where y i is the output of the ith rule, u i is the membership degree corresponding to the i-th rule, and n is the total number of rules.

[0110] The predictive control output is added as a weighted term to the final output of the fuzzy control, and the formula is as follows:

[0111] u fc =α(k)·u dc +γ·u mpc +k

[0112] Among them, u fc is the fuzzy output control quantity. α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

[0113] After the wind speed enters the error band, the system switches to position PI, the formula is as follows:

[0114] u pi =k p ×e(t)+k i ×∫e(t)dt

[0115] where u pi is the control quantity, e(t) is the error, ∫e(t)dt is the integral term of the error, k p is the proportional gain, k i is the integral gain.

[0116] S5: The predictive control output is added as a weighted term to the final output of the position PI control, and the formula is as follows:

[0117] u pi =α(k)·u pi +γ·u mpc +k

[0118] Among them, u pi is the final output control quantity of the position PI. α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

[0119] like Figure 4 and Figure 5 , experimental results at two different Mach numbers are given, namely 0.6 Mach and 0.85 Mach. It can be seen that when the set value is 0.6 Mach, the wind speed has entered the error band in the initial stage, that is, the error is ≤0.05, and the system begins to enter PI control and outputs a stable wind speed in combination with predictive control. When the set value is 0.85 Mach, the wind speed error is large in the initial stage of wind tunnel operation, that is, the error is ≥0.05. Fuzzy control is started and combined with predictive control to continuously reduce the wind speed error. When the error is ≤0.05, the control system switches to PI control combined with predictive control to further stabilize the wind speed.

[0120] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A method for controlling the steady wind speed of an open high-speed wind tunnel, the wind tunnel comprising a stable section and a test section, the test section being located downstream of the stable section, and a fast valve and a pneumatic regulating valve being arranged in the flow direction upstream of the stable section, characterized in that: The method comprises the following steps: S1: Use an air compressor and an air tank to provide the high-pressure air source required for wind tunnel operation. The pneumatic regulating valve is opened to a certain opening through pre-opening calculation, and then the fast valve is opened instantaneously to release the pressure of the air tank to generate high-speed airflow and quickly establish the flow field. S2: Collect wind tunnel operation data in real time, filter the collected data in real time using the set filtering algorithm, and input the filtered data into the control system; S3: The control system calculates the Mach number of the test section in real time based on the real-time data collected during wind tunnel operation, calculates the optimal control amount of the pneumatic control valve through predictive control, and weights the optimal control amount to participate in the final output of fuzzy control and PI control; S4: Calculate the error and error change rate between the actual measured Mach number and the expected Mach number set before the experiment. When the error is ≥ 0.05, use fuzzy control to make adjustments. The error and error change rate are used as the input of fuzzy control. The control output is generated through fuzzy rule reasoning to make preliminary adjustments to the wind tunnel flow field. When the error is < 0.05, use the position PI control algorithm to make further adjustments based on the feedback error of the control system.

2. The method for controlling the steady wind speed of an open high-speed wind tunnel according to claim 1, characterized in that: The data during wind tunnel operation include the total pressure of the stable section and the static pressure data of the test section. The Mach number is calculated by the total pressure of the stable section and the static pressure data of the test section. The calculation formula is: Among them, P0 is the total pressure in the stable section, P S is the static pressure of the test section.

3. The method for controlling the steady wind speed of an open high-speed wind tunnel according to claim 1, characterized in that: In S2, the filtering algorithm is set as follows: the collected data is processed by combining low-pass filtering with median filtering. First, low-pass filtering is used to remove high-frequency noise in the data, and then median filtering is used to further remove the remaining noise. The formula is as follows: Where: x(ti) represents the total pressure and static pressure data during wind tunnel operation, N represents the order of the low-pass filter, h(i) is the coefficient of the low-pass filter, K represents the size of the median filter window, (tK),…,(t+K) represents a sliding window, and y[t] is the filtered total pressure and static pressure data.

4. The method for controlling the steady wind speed of an open high-speed wind tunnel according to claim 1, characterized in that: Predictive control in S3 includes: First, in predictive control, a prediction model is established, and its formula is as follows: x(k+1)=A·x(k)+B·u(k)+B·v(k) Among them, x(k) represents the current state, that is, the current Mach number of the test section; u(k) represents the control input, that is, the opening of the pneumatic control valve; v(k) represents the disturbance; A represents the state transfer matrix, that is, the state change process from time k to time k+1; B represents the control input matrix, B the disturbance transfer matrix; Then predict the state of the next N steps, the formula is as follows: in, represents the control system state at the prediction time (k+i); x(k) represents the current state, and u(k+j) represents the control input sequence; Then define the cost function, the formula is as follows: Among them, x ref (k+i) represents the Mach number that should be reached at time k+i; Q represents the weight matrix; u(k+i) represents the control action at time k+i; Ru represents the weight matrix; γ represents the trade-off factor; Finally, the cost function is minimized under the condition of satisfying the constraints through quadratic programming to obtain the optimal control input sequence: Extract the first control quantity u of the optimization result * (k) is used as the optimal control quantity of the pneumatic control valve calculated by predictive control, that is, the final output of predictive control, u mpc (k) = u * (k).

5. The method for controlling the steady wind speed of an open high-speed wind tunnel according to claim 1, characterized in that: In S4, when fuzzy control is used, the error and error change rate are calculated from the expected Mach number and the data collected during wind tunnel operation. The error calculation formula is: e(t)=Ma ref -M(t) Where e(t) is the error in the Mach number. ref is the expected Mach value, M(t) is the actual Mach value collected and filtered during wind tunnel operation; The error rate is as follows: Among them, e(t-1) is the Mach number error at the previous moment, e(t) is the Mach number error at the current moment, Δt is the sampling period, and ed(t) is the rate of change of the Mach number error. When the sampling period is constant, Δt can be omitted and simplified to: ed(t)=e(t)-e(t-1); The Mach number error e(t) and the error change rate ed(t) are normalized, and the formula is as follows: Normalized error: Normalized error rate of change: in, is the normalized value of the Mach number error, and e(t) is the Mach number error e max With e min are the maximum and minimum values ​​of the error, is the normalized value of the error change rate, ed(t) is the error change rate, ed max With ed min is the maximum and minimum value of the error change rate, r max With r min are the maximum and minimum values ​​of the target normalized range; According to the normalized values ​​of the error and the error change rate, the fuzzy subsets are divided into negative large NB, negative small NS, zero ZE, positive small PS, and positive large PB. The fuzzy subsets calculate the fuzzy membership according to the normalized input error and error change rate, and calculate the fuzzy output according to the conditions of the fuzzy rules. The fuzzy rule formula is: IN output =min(in inpute ,in outputed ) Among them, u inpute and u outputed are the membership degrees of the fuzzy subsets of the Mach number error and the Mach number error change rate, respectively. output Then it is the membership degree of fuzzy output; Finally, the fuzzy output is defuzzified by weighted average to obtain u fc , and its calculation formula is as follows: Among them, y i is the output of the ith rule, u i is the membership degree corresponding to the i-th rule, and n is the total number of rules; The predictive control output is added as a weighted term to the final output of the fuzzy control, and the formula is as follows: you fc =α(k)·u fc +γ·u mpc +k Among them, u fc is the fuzzy output control quantity, α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

6. The method for controlling the steady wind speed of an open high-speed wind tunnel according to claim 1, characterized in that: In S4, when using the position PI control algorithm, the formula is as follows: u pi =k p ×e(t)+k i ×∫e(t)dt Among them, u pi is the control quantity, e(t) is the error, ∫e(t)dt is the integral term of the error, k p is the proportional gain, k i is the integral gain; The predictive control output is added as a weighted term to the final output of the position PI control, and the formula is as follows: you pi =α(k)·u pi +γ·u mpc +k Among them, u pi is the final output control quantity of the position PI, α(k) is the weighting coefficient based on the current error, γ is the predictive control weight coefficient, and k is the actual valve position of the current pneumatic control valve.

Images