A method for real-time adjustment of a helicopter rotor balance

Abstract

The application discloses a helicopter rotor balance real-time adjustment method, and calculates the adjustment amount ADJ of each intelligent variable-pitch pull rod in the vibration balance adjustment of a helicopter rotor by using a control model Vib , calculates the adjustment amount ADJ of each intelligent variable-pitch pull rod in the cone balance adjustment of the helicopter rotor by using a PID feedback control algorithm C one_ b , calculates the adjustment amount ADJ of each intelligent variable-pitch pull rod in the cone balance adjustment of the helicopter rotor by using a feedforward control algorithm Cone_f ; the ADJ Vib , ADJ Cone_b , ADJ Cone_f of each intelligent variable-pitch pull rod is weighted and adjusted, and the final weighted adjustment amount is ADJ fi na l The application realizes the simultaneous control of the vibration and the cone of the helicopter rotor by analyzing the coupling relationship between the vibration and the cone of the helicopter and adopting the relatively engineering weight value configuration to control the amplitude between the vibration and the cone.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of computers and relates to a real-time adjustment method for balance of a helicopter rotor. BACKGROUND

[0002] The vibration and excessive cone caused by the mass imbalance and aerodynamic imbalance of a helicopter rotor not only easily causes fatigue of a pilot, difficulty in operation, damage of airborne equipment and shortening of the service life of the helicopter, but also causes high-cycle fatigue of a large number of moving parts of the helicopter, easily causes mechanical failure and thus causes a flight accident. Therefore, the most important thing in the use and maintenance of the helicopter is the balance measurement and adjustment of the rotor. With the increasingly stringent requirements for safety, economy and comfort of a medium and large-sized civil helicopter, the periodic maintenance with obvious disadvantages has not adapted to the increasingly extensive application requirements of the helicopter. The method for adjusting the cone and dynamic balance of the rotor blade is an effective way to reduce the vibration level of the helicopter, improve the vibration environment of the helicopter and improve the flight quality of the helicopter. However, the cone and dynamic balance values of the helicopter rotor need to be accurately measured first. Only when the accurate cone and dynamic balance data are obtained, can the vibration reduction method and measure be researched and implemented.

[0003] The maintenance regulation of the dynamic balance of the helicopter first performs the rotor cone balance measurement and adjustment to eliminate the vibration of the helicopter body caused by the cone. Then, the body vibration measurement is performed to eliminate the vibration caused by the mass imbalance through the blade weight adjustment. Under the conventional condition, after the above two kinds of imbalance adjustment, the body vibration will be limited within the limited value required by the maintenance manual. The traditional dynamic balance adjustment of the rotor is an intermittent maintenance method on the ground. The length of the variable pitch pull rod, the angle of the trailing edge adjustment piece and the size of the hub weight can be changed to balance the aerodynamic force and centrifugal force of each blade of the rotor, so as to reduce the excitation load of the rotor. A large amount of fuel and the service life of key components such as the engine are consumed each time the maintenance is performed. The adjustment cannot be performed in the air. The adjustment opportunity and the position for balance compensation on the rotor are limited, which causes a large economic loss.

[0004] The current domestic and foreign research reduces the number of dynamic balance adjustment. After the adjustment, the verification is still needed. The real-time adjustment in the flight process cannot be realized. The use cycle and all flight states of the helicopter cannot be covered. Therefore, the safety, economy, comfort and service support of the helicopter need to be improved through advanced technical means. The use efficiency and maintenance level of the medium and large-sized civil helicopter are improved as a whole. SUMMARY

[0005] The application aims to provide a helicopter rotor balance real-time adjustment method.

[0006] The application aims to provide a helicopter rotor balance real-time adjustment method.

[0007] A helicopter rotor balance real-time adjustment method, the method steps are as follows:

[0008] Step 100, the adjustment amount ADJ of each intelligent variable pitch rod in the vibration balance adjustment of the helicopter rotor is calculated by using a control model Vib , the adjustment amount ADJ of each intelligent variable pitch rod in the balance adjustment of the helicopter rotor cone is calculated by using a PID feedback control algorithm Cone_b , and the adjustment amount ADJ of each intelligent variable pitch rod in the balance adjustment of the helicopter rotor cone is calculated by using a feedforward control algorithm Cone_f .

[0009] Step 200, the adjustment amount ADJ of the vibration adjustment of each intelligent variable pitch rod, the adjustment amount ADJ of the cone feedforward adjustment, and the adjustment amount ADJ of the cone feedback adjustment are weighted to obtain the final weighted adjustment amount ADJ. Vib Cone_b Cone_f final .

[0010] The application has the following advantages:

[0011] The application realizes the coupling control of vibration and cone. The automatic intelligent variable pitch rod is installed under each blade, the coupling relationship between the vibration and the cone of the helicopter is analyzed, the output limiting between the vibration and the cone is performed by using the relatively engineering weight configuration method such as the experience method or the trial and error method, the displacement amount of the intelligent variable pitch rod is calculated according to the helicopter rotor balance real-time adjustment method, the length of the intelligent variable pitch rod is controlled by issuing a digital instruction, if the returned vibration value and the cone value are higher than the set expected value, the control is continuously performed until the vibration and the cone value are reduced to the target value, the control of the intelligent variable pitch rod is stopped, the helicopter realizes the dynamic balance adjustment in a shorter time during the flight, the mutual difference of the rotor blade tip trajectory is not greater than 4mm, and the vibration is reduced to be less than 0.05IPS. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 It is a vibration vector screening flowchart of the application.

[0013] Figure 2 It is a planar view of three vectors of the application. ​​​

[0014] Figure 3 Vector plan for the first class of the present invention;

[0015] Figure 4 Vector plan for the second class of the present invention;

[0016] Figure 5 Vector plan A for the second class of the present invention;

[0017] Figure 6 Vector plan B for the second class of the present invention;

[0018] Figure 7 Vector plan for the third class of the present invention;

[0019] Figure 8 Vector plan A for the third class of the present invention;

[0020] Figure 9 Vector plan B for the third class of the present invention;

[0021] Figure 10 Vector plan A for the fourth class of the present invention;

[0022] Figure 11 Vector plan B for the fourth class of the present invention;

[0023] Figure 12 Vector plan C for the fourth class of the present invention;

[0024] Figure 13 Vector plan A1 for the fourth class of the present invention;

[0025] Figure 14 Vector plan A2 for the fourth class of the present invention;

[0026] Figure 15 Vector plan A3 for the fourth class of the present invention;

[0027] Figure 16 Vector plan A4 for the fourth class of the present invention;

[0028] Figure 17 Vector plan C1 for the fourth class of the present invention;

[0029] Figure 18 Vector plan C2 for the fourth class of the present invention;

[0030] Figure 19 Vector plan C3 for the fourth class of the present invention;

[0031] Figure 20 Vector plan C4 for the fourth class of the present invention;

[0032] Figure 21 It is a flow chart of the PID feedback control of the present invention;

[0033] Figure 22 This is a structural diagram of the feedforward control algorithm of the present invention;

[0034] Figure 23 This is a flow chart of the feedforward control process of the present invention;

[0035] Figure 24 It is a workflow diagram of the present invention;

[0036] Figure 25 A control flow chart for real-time adjustment of the present invention;

[0037] Figure 26 Figure (1) is the control result of the present invention;

[0038] Figure 27 FIG. 2 is a control result diagram of the present invention;

[0039] Figure 28 FIG. 1 is a diagram showing the vibration adjustment result of the present invention;

[0040] Figure 29 FIG. 2 is a diagram showing the vibration adjustment result of the present invention;

[0041] Figure 30 FIG. 3 is a diagram showing the vibration adjustment result of the present invention;

[0042] Figure 31 Figure (1) is the cone adjustment result of the present invention;

[0043] Figure 32 Figure (2) is the cone adjustment result of the present invention;

[0044] Figure 33 Figure (3) is the cone adjustment result of the present invention; DETAILED DESCRIPTION

[0045] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0046] like Figure 24 As shown, the method for real-time adjustment of helicopter rotor balance shown in this embodiment comprises the following steps:

[0047] Step 100: Calculate the adjustment amount ADJ of each intelligent variable pitch lever in the helicopter rotor vibration balance adjustment using the control model Vib , using PID feedback control algorithm to calculate the adjustment amount ADJ of each intelligent pitch lever in the helicopter rotor cone balance adjustment Cone_b, the adjustment amount ADJ of each intelligent pitch rod in the balance adjustment of the helicopter rotor cone is calculated by using a feedforward control algorithm Cone_f .

[0048] In the calculation process of the adjustment amount ADJ of the vibration balance adjustment Vib , the following steps are included:

[0049] Step 111, pre-processing of modeling data: screening the collected vibration data of the helicopter rotor, removing abnormal vibration vectors and retaining normal vibration vectors.

[0050] The vibration data is the amplitude and phase data of the X, Y and Z axes of the four intelligent pitch rods under three equally spaced positive force displacements (such as 0.2, 0.4, 0.6) and three equally spaced negative force displacements (such as -0.2, -0.4, -0.6). The vibration vector ρ is a matrix composed of the intelligent pitch rod number α, the rod displacement β, the amplitude δ and the phase θ, ρ = [α, β, δ, θ].

[0051] As shown in Figure 1 , in the process of screening the vibration vector, the positive force vector obtained under the positive force displacement and the negative force vector obtained under the negative force displacement of each intelligent pitch rod are respectively sorted by phase from small to large, respectively obtaining θ min , θ mid , θ max , and then obtaining two phase differences θ Δ1 , θ Δ2 , the formulas are as follows:

[0052] θ Δ1 = θ mid - θ min (1)

[0053] θ Δ2 = θ max - θ mid (2)

[0054] There are a total of 5 blades, and the angle interval between the blades is 72, the normal phase difference interval is (0, 72), and the abnormal phase difference interval is (288, +∞). When θ Δ1 and θ Δ2 are in the interval (0, 72), the vibration vectors corresponding to θ min , θ mid , θ max are retained; when θ Δ1 is in the interval (0, 72) and θ Δ2 is in the interval (288, +∞), the vibration vector corresponding to θ max is removed, and the vibration vectors corresponding to θ min and θ mid are retained; when θΔ2 In the interval (0,72), θ Δ1 In the interval (288, +∞), remove θ min The corresponding vibration vector, retaining θ mid and θ max The corresponding vibration vector; when θ Δ1 In the interval (288, +∞), θ mid ,θ max Subtract 360 from each and reorder and calculate θ Δ1 ,θ Δ2 , and then perform vibration vector screening; when θ Δ2 In the interval (288, +∞), θ max Reorder and calculate θ after subtracting 360 Δ1 ,θ Δ2 , and then perform vibration vector screening; when θ Δ1 and θ Δ2 When both are in the interval [72,288], remove θ min ,θ mid ,θ max The corresponding vibration vector.

[0055] Step 112, establishing a control model: establishing a control model according to the mapping relationship between the displacement, amplitude δ, phase θ of the intelligent variable pitch pull rod and the vibration vector of the helicopter.

[0056] The amplitude of the resultant force vector obtained by adding the positive force vectors screened out must be greater than the amplitude of the positive force vector. Experiments show that the rod displacement is linearly proportional to the vibration amplitude, that is, the larger the rod displacement, the greater the vibration amplitude of the helicopter. Although the amplitude of the resultant force vector is large, the maximum rod displacement cannot exceed the rod displacement of the positive force vector screened out. Therefore, the rod displacement of the resultant force vector takes the maximum rod displacement in the positive force vector, and the final positive vibration vector is expressed as ρ + =[α+,β+max,δ + ,θ + ], where α + is the number of the intelligent pitch-changing rod, such as 2, 3, 4, and 5; among them, β +max is the maximum displacement among the positive force vectors screened out. Similarly, the negative force vector is expressed as ρ-=[α - ,β -max ,δ - ,θ - ], where β -max is the maximum displacement among the negative force vectors screened out. Through the above steps, we finally obtain the positive and negative force vectors of the intelligent variable pitch tie rods 2, 3, 4, and 5, a total of 8 vectors, with a total of three axes: X-axis, Y-axis, and Z-axis. Therefore, the matrix composed of 24 vectors is the control model.

[0057] Step 113, vibration adjustment balance: real-time acquisition of vibration data of the helicopter rotor, rotation of vibration unbalance points on X-axis, Y-axis and Z-axis by 180 degrees respectively as compensation unbalance required compensation adjustment vector, then find the closest vector on both sides of the compensation adjustment vector on X-axis, Y-axis and Z-axis from the control model, decompose the vector to obtain the adjustment amount of each intelligent variable pitch pull rod on X-axis, Y-axis and Z-axis.

[0058] The rotor will generate unbalanced force during operation. In order to adjust the balance, the unbalanced force needs to be offset by the adjustment of the intelligent variable pitch pull rod. The X-axis, Y-axis and Z-axis have corresponding unbalanced forces respectively. The unbalanced force is composed of amplitude and phase, represented as U=[δ U ,θ U ]. The symmetric force of the unbalanced force is the offset force, represented as F=[δ F ,θ F ]. The calculation formula of the offset force is as follows:

[0059] F=U+[0,180] (3)

[0060] The adjustment balance algorithm needs to use the control model and the offset force F. The vibration adjustment balance algorithms of X-axis, Y-axis and Z-axis are the same. The following takes X-axis as an example for description.

[0061] Sort the 8 vibration vectors of X-axis in the control model by phase from small to large, find the vibration vectors corresponding to the adjacent left and right two phases of the phase θ F of the offset force, for example, in [θ L2 ,θ L1 ,θ F ,θ R1 ,θ R2 ], the vibration vector corresponding to θ F on the left side of θ L1 is LW1; the vibration vector corresponding to θ L2 is LW2; the vibration vector corresponding to θ F on the right side of θ R1 is UP1; the vibration vector corresponding to θ R2 is UP2.

[0062] When the left component of the offset force can be synthesized by using part of the components of LW1, the pull rod displacement amount of the vector corresponding to LW1 is calculated as the adjustment amount according to the pull rod adjustment ratio; when the left component of the offset force cannot be synthesized after using all the components of LW1, the output adjustment amount contains the pull rod displacement amount of the vector corresponding to LW2 calculated according to the pull rod adjustment ratio in addition to the pull rod displacement amount of the vector corresponding to LW1.

[0063] When UP1 partially uses its components to synthesize the right-hand component of the offsetting force, the rod displacement of the corresponding vector of UP1 is calculated according to the rod adjustment ratio and used as the adjustment variable. When UP1 uses all its components and still cannot synthesize the left-hand component of the offsetting force, the output adjustment variable includes not only the full rod displacement of UP1 but also the rod displacement of the corresponding vector of UP2 calculated according to the rod adjustment ratio. The following is a detailed description of the output of the adjustment variable.

[0064] The phase angles between the four vibration vectors and the offset force vector are calculated respectively. The formulas are as follows:

[0065] LW1θ=θ F -θ L1 (4)

[0066] LW2θ=θ F -θ L2 (5)

[0067] UP1θ=θ R1 -θ F (6)

[0068] UP2θ=θ R2 -θ F (7)

[0069] like Figure 2 As shown, using LW 1θ UP 1θ , LW1, UP1, F make a parallelogram, assuming that these three vectors form a plane graph, the two angles in ΔOFA can get the third angle TH 1θ The sine theorem formulas for triangles ΔOFA and ΔOFB are as follows:

[0070]

[0071] Given the amplitude of OF, i.e., the offsetting force vector F, TH1θ, LW1θ, and UP1θ, OA and OB are obtained according to formulas (8) and (9). By comparing the amplitudes of OA and LW1, and the amplitudes of OB and UP1, the adjustment balance algorithm is divided into four categories. When OA is smaller than the amplitude of LW1 and OB is smaller than the amplitude of UP1, it is the first category. Only LW1 and UP1 are needed to synthesize the offsetting force, that is, these two vectors can offset the unbalanced force, and LW2 and UP2 are not needed. When OA is larger than the amplitude of LW1 and OB is smaller than the amplitude of UP1, it is the second category. Only UP1 can be used to synthesize the right component of the offsetting force, while only LW1 cannot be used to synthesize the left component of the offsetting force. LW1 and LW2 need to be used to synthesize the left component of the offsetting force. When OA is smaller than the amplitude of LW1 and OB is smaller than the amplitude of UP1, it is the second category. Only UP1 can be used to synthesize the right component of the offsetting force, while only LW1 cannot be used to synthesize the left component of the offsetting force. When the amplitude of LW1 and OB are greater than the amplitude of UP1, it is the third category. Only LW1 can synthesize the left component of the offsetting force, but only UP1 cannot synthesize the right component of the offsetting force. It is necessary to use UP1 and UP2 to synthesize the right component of the offsetting force. When OA is greater than the amplitude of LW1 and OB is greater than the amplitude of UP1, it is the fourth category. LW1 cannot synthesize the left component of the offsetting force, and UP1 cannot synthesize the right component of the offsetting force. It is necessary to use LW1, LW2, UP1, and UP2 to synthesize the offsetting force. The specific methods of the four types of vibration adjustment algorithms are as follows:

[0072] ① Category 1

[0073] like Figure 3 In the vector plane diagram of the first type shown, LW1 and UP1 can synthesize a counteracting force using only partial components. Calculating the ratio of the amplitudes of OA and LW1, and the ratio of the amplitudes of OB and UP1, we can obtain the rod adjustment ratio of LW1 and UP1, as shown in the following formula:

[0074]

[0075] Calculate the rod adjustment of LW1 and UP1, which is the product of the rod adjustment ratio and the rod displacement of the corresponding vector. The formula is as follows:

[0076] ADJ LW1 =PER LW1 β LW1 (12)

[0077] ADJ UP1 =PER UP1 β UP1 (13)

[0078] The output regulation vector is [α LW1 ,ADJ LW1 ]、[α UP1 ,ADJ UP1 ].

[0079] ② Category 2

[0080] like Figure 4 In the second type of vector plane diagram shown in , UP1 can synthesize the right component of the offsetting force using some of its components, but LW1 can only synthesize the left component of the offsetting force. According to formulas (11) and (13), the pull rod adjustment amount of UP1 is obtained. When LW1 is fully adjusted, the entire pull rod displacement of LW1 is output, and the adjustment output is [α LW1 ,β LW1 ]、[α UP1 ,ADJ UP1 ]. The left component of the offsetting force that can be synthesized by LW1 at full adjustment is OO1 in the figure, and the remaining O1F needs to be synthesized by LW2.

[0081] like Figure 5 、 Figure 6 As shown, using LW 2θ UP 1θ , LW2, UP1, O1F make a parallelogram to form the second type of vector plane A and B. In the second type of vector plane A, O1C is greater than the amplitude of LW2, so the full adjustment output is the full displacement of the pull rod of LW2, and the adjustment output is [α LW2 ,β LW2 ]、[α LW1 ,β LW1 ]、[α UP1 ,ADJ UP1 ]、[α UP2 ,0]; In the second type of vector plane B, O1C is smaller than the amplitude of LW2, so the adjustment ratio of the lever of LW2 is calculated to obtain the adjustment amount of the lever of LW2. The final adjustment amount output is [α LW2 ,ADJ LW2 ]、[α LW1 ,β LW1 ]、[α UP1 ,ADJ UP1 ]、[α UP2 ,0].

[0082] ③The third category

[0083] like Figure 7 In the vector plane diagram of the third type shown, LW1 can synthesize the left component of the offsetting force using part of its components, but UP1 can only synthesize the right component of the offsetting force.

[0084] According to formulas (10) and (12), the adjustment amount of the pull rod of LW1 is obtained. When UP1 is fully adjusted, the entire pull rod displacement of UP1 is output. The adjustment amount output is [α LW1 ,ADJ LW1], [α UP1 , β UP1 ]. As can be seen from the figure, the right side component of the UP1 full adjustment energy synthesis counter force is OO1 in the figure, and the remaining O1F needs UP2 to synthesize.

[0085] As shown in Figure 8 , Figure 9 , the third type of vector plane A diagram and B diagram are formed by using LW1θ, UP2θ, LW1, UP1, O1F to form a parallelogram. In the third type of vector plane A diagram, O1B is greater than the amplitude of UP2, so the full rod displacement amount of the adjustment output UP2 is obtained, and the adjustment output amount is [α LW2 , 0], [α LW1 , ADJ LW1 ], [α UP1 , β UP1 ], [α UP2 , β UP2 ]; in the third type of vector plane B diagram, O1B is less than the amplitude of UP2, so the rod adjustment proportion of UP2 is calculated, the rod adjustment amount of UP2 is obtained, and finally the adjustment output amount is [α LW2 , 0], [α LW1 , ADJ LW1 ], [α UP1 , β UP1 ], [α UP2 , ADJ UP2 ].

[0086] ④ The fourth type

[0087] As shown in Figure 10 , Figure 11 , Figure 12 , in the fourth type of vector plane A diagram, B diagram and C diagram, LW1 and UP1 use all vectors and still cannot synthesize the counter force. The fourth type of vector plane A diagram is the first small class, and a parallelogram is formed by using LW1θ, UP1θ, LW1, UP1, F, O U1 is less than O L1 . Therefore, the partial vector OC of LW1 and UP1 can synthesize the partial counter force OO U1 of F, and the adjustment amount of LW1 is obtained, and the adjustment output amount is [α LW1 , ADJ LW1 ], [α UP1 , β UP1 ]. The remaining counter force O U1 F is synthesized by LW2 and UP2.

[0088] As shown in Figure 13 , the fourth type of vector plane A1 diagram is the first case in the first small class, and the algorithm principle is not repeated, and the adjustment output amount is [αLW2 ,ADJ LW2 ]、[α LW1 ,ADJ LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,ADJ UP2 ];like Figure 14 The vector plane A2 of the fourth category shown is the second case in the first subcategory, using LW 2θ UP 2θ , LW2, UP2, F make a parallelogram, use the partial vector O of LW2 and UP2 U1 D Synthetic Offsetting Force O U1 O U2 , LW2 is fully regulated, UP2 is proportionally regulated, and the regulated output is [α LW2 ,β LW2 ]、[α LW1 ,ADJ LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,ADJ UP2 ];like Figure 15 The fourth type of vector plane A3 shown is the third case in the first sub-category. LW2θ, UP2θ, LW2, UP2, and F are used to make a parallelogram, and the partial vector O of LW2 is used to make a parallelogram. U1 D and UP2 combine to counteract the force O U1 O U2 , LW2 is adjusted proportionally, UP2 is fully adjusted, and the adjustment output is [α LW2 ,ADJ LW2 ]、[α LW1 ,ADJ LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,β UP2 ];like Figure 16 The vector plane A4 diagram of the fourth category shown is the fourth case in the first subcategory, using LW 2θ UP 2θ , LW2, UP2, F form a parallelogram, LW2 and UP2 cannot synthesize the entire offset force O U1 F, so LW2 and UP2 are fully regulated, and the regulated output is [α LW2 ,β LW2 ]、[α LW1 ,ADJ LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,β UP2 ].

[0089] The fourth type of vector plane B is the second subtype, using LW 1θ UP 1θ , LW1, UP1, F make a parallelogram, O U1 Greater than O L1 Therefore, the partial vector OC of LW1 and UP1 can synthesize the partial offset force OO of F. U1 , the algorithm principle for calculating the adjustment amount is consistent with the first subcategory, and will not be repeated here.

[0090] The fourth type of vector plane C diagram is the third subtype, using LW 1θ UP 1θ , LW1, UP1, F make a parallelogram, O U1 Equal to O L1 Therefore, LW1 and UP1 can synthesize the partial offset force of F OO U1 , LW1 and UP1 are fully regulated, and the regulated output is [α LW1 ,β LW1 ]、[α UP1 ,β UP1 Using LW 2θ UP 2θ , LW2, UP2, F make parallelograms, and compare O U1 D and LW2, O U1 The size of E and UP2.

[0091] When O U1 D is less than LW2, O U1 When E is less than UP2, Figure 17 In the fourth type of vector plane C1 diagram shown, LW2 and UP2 can synthesize the remaining counteracting force O U1 F, calculate the adjustment ratio of LW2 and UP2 respectively, and get the adjustment amount. The adjustment amount output is [α LW2 ,ADJ LW1 ]、[α LW1 ,β LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,ADJ UP2 ]. When O U1 D is greater than LW2, O U1 When E is less than UP2, Figure 18 In the fourth type of vector plane C2 shown in the figure, UP2 is adjusted proportionally, LW2 is fully adjusted, and the adjustment output is [α LW2 ,β LW2 ]、[α LW1 ,β LW1 ]、[α UP1,β UP1 ]、[α UP2 ,ADJ UP2 ]. When O U1 D is less than LW2, O U1 When E is greater than UP2, such as Figure 19 In the fourth type of vector plane C3 shown in FIG, LW2 is adjusted proportionally, UP2 is fully adjusted, and the adjustment output is [α LW2 ,ADJ LW2 ]、[α LW1 ,β LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,β UP2 ]. When O U1 D is greater than LW2, O U1 When E is greater than UP2, such as Figure 20 In the fourth type of vector plane C4 shown in the figure, LW2 and UP2 are fully regulated, and the regulated output is [α LW2 ,β LW2 ]、[α LW1 ,β LW1 ]、[α UP1 ,β UP1 ]、[α UP2 ,ADJ UP2 ].

[0092] Feedback control algorithms are based on deviation control, meaning the regulator generates control to offset the interference caused by the deviation. In practice, cone measurements are collected continuously and in real time, so control adjustments fluctuate behind the increasing interference, resulting in inevitable steady-state position tracking errors. Feedforward control algorithms perform compensation control based on the disturbance, directly generating a corrective effect based on the magnitude of the disturbance, and theoretically can completely eliminate the deviation caused by the disturbance. In the present invention, both feedback and feedforward control algorithms are employed, resulting in a control system that offers both timely feedforward control and precise feedback control.

[0093] like Figure 21 As shown in Figure 1, the PID feedback control algorithm consists of a proportional unit P, an integral unit I, and a differential unit D. Assuming that the actual measured value of the cone is e(t), the output of the PID feedback control algorithm is u(t), and the relationship between them is shown in Equation (14):

[0094]

[0095] The upper and lower limits of the integral are 0 and t respectively, and the initial value of P is set, so the transfer function is shown in formula (15):

[0096]

[0097] Where KP is the proportional coefficient, T1 is the integral time constant, and TD is the differential time constant. Set the initial value of KP to 0.01, the initial value of T1 to 0.03, and the initial value of TD to 0.

[0098] The adjustment amount ADJ of each intelligent pitch lever in the calculation of helicopter rotor cone balance adjustment using PID feedback control algorithm Cone_b The process includes the following steps:

[0099] Step 121: Input the expected value of each cone adjustment; input the actual measured value of each cone; input the proportional coefficient KP of P, the proportional coefficient KI of I, and the proportional coefficient KD of D; input the initial values ​​of P_ and I_.

[0100] Step 122: Calculate the value of the output U of the proportional unit P, the integral unit I, the differential unit D, and the PID feedback control algorithm using the following formula:

[0101] P=EC (16)

[0102] I=I_+P (17)

[0103] D=P-P_ (18)

[0104] U=KP·P+KI·I+KD·D (19)

[0105] Step 123: Update the values ​​of P_ and I_. The formula is as follows:

[0106] P_=P (20)

[0107] I_=I (21)

[0108] Step 124: The calculation formula for the adjustment amount of each intelligent pitch-changing rod corresponding to the cone is as follows:

[0109] ADJ Cone_b =0 . 1(-U) (22)

[0110] like Figure 22 In the feedforward control algorithm structure diagram shown in the figure, G n (s) is the transfer function of the disturbance channel of the controlled object, D n (s) is the transfer function of the feedforward controller, G(s) is the control channel transfer function of the controlled object, n, u, and y are the disturbance variable, control variable, and output variable, respectively.

[0111] Assuming that the input u1=0, then:

[0112] Y(s)=Y1(s)+Y2(s)=[D n (s)G(s)+Gn (s)]N(s) (23)

[0113] When Y(s) = 0, the feedforward control can fully compensate for the disturbance, that is:

[0114] D n (s)G(s)+G n (s)=0 (24)

[0115] Therefore, the transfer function of the feedforward controller is:

[0116] D n (s)=-G n (s) / G(s) (25)

[0117] In the calculation of the adjustment amount ADJ of each intelligent pitch lever in the balance adjustment of the helicopter rotor cone using the feedforward control algorithm Cone_f The process consists of the following steps:

[0118] Step 131: Input the expected value E of cone adjustment f ; Input the actual measurement value A of the cone f ; Input the proportional parameter Per of the feedforward control. This parameter is the empirical value obtained during the experiment and is set to 0.35.

[0119] Step 132: Calculate the adjustment amount ADJ of each smart lever Cone_b =U f , if the measured value is positive, the calculation formula is as follows:

[0120]

[0121] If the measured value is negative, the calculation formula is as follows:

[0122]

[0123] Step 200: Adjust the vibration adjustment amount ADJ of each intelligent variable pitch rod. Vib , cone feedforward adjustment adjustment amount ADJ Cone_f , cone feedback adjustment amount ADJ Cone_b Perform weighted adjustment and obtain the final weighted adjustment amount as ADJ final .

[0124] There are respective weights for vibration regulation, cone feedforward regulation, and cone feedback regulation, which are W Vib 、W Cone_f 、W Cone_b Take the No. 2 pull rod as an example to illustrate the weighted adjustment of vibration and cone. For example, the adjustment amount of the No. 2 pull rod for vibration adjustment is ADJ Vib, the adjustment amount of the second pull rod of the conical feedforward adjustment is ADJ Cone_f , the adjustment amount of the second pull rod of the conical feedback adjustment is ADJ Cone_b , the final weighted adjustment amount of the second pull rod is ADJ final , the calculation formula is as follows:

[0125] ADJ final = W Vib ·ADJ Vib + W Cone_f ·ADJ Cone_f + W Cone_b ·ADJ Cone_b (28)

[0126] The weighted adjustment amount ADJ final is judged, because the maximum displacement amount of the intelligent variable pitch pull rod is ±0.66, if the absolute value thereof is less than 0.66, the weighted adjustment amount is outputted, if the absolute value thereof is greater than 0.66, the positive number is outputted as 0.66, and the negative number is outputted as -0.66.

[0127] The effect of the present application is described below by using a group of experimental data.

[0128] 1. Test state

[0129] The helicopter rotor test system has a total of 5 blades, which are marked as No. 1-5 in order, wherein No. 1 is the reference blade, and no intelligent variable pitch pull rod is installed, and the other 4 blades are all installed with intelligent variable pitch pull rods, all the intelligent variable pitch pull rods are placed at zero position, and the initial imbalance of the rotor is generated by adding weight blocks. The present application is verified under different total pitch and cyclic pitch states of the helicopter rotor, and Table 1 is the test state description.

[0130] Table 1 Test state description

[0131]

[0132]

[0133] 2. Real-time adjustment work flow

[0134] In the work flow chart of the present application, the corresponding control model is established by using the control algorithm, after the model is established, the control parameters need to be set before adjustment, the expected value of vibration control needs to be set, and the default value is 0; the expected value of conical control is 4; the adjustment control mainly includes vibration control and conical control, as shown in the control flow chart of real-time adjustment.

[0135] Vibration control: The compensation adjustment vector required to compensate the vibration unbalance point by rotating 180 degrees is obtained. The two closest vectors to the left and right of the compensation adjustment vector are found, and these two vectors are called the two interval. The compensation adjustment vector is decomposed into two vectors in the two interval, and if it can be completely decomposed, the pull rod control amount is output (first type of adjustment algorithm in step 113); if it can only be partially decomposed, the remaining compensation adjustment vector is decomposed into one vector in the third interval and one vector not used in the two interval (second and third types of adjustment algorithm in step 113); if it cannot be completely decomposed, the compensation adjustment vector is decomposed into two vectors in the four interval (fourth type of adjustment algorithm in step 113), and then the pull rod control amount is output.

[0136] Cone control: The cone height difference value of the current state of the system is first received, and the cone height error is obtained by using the error solution of the height difference value and the cone control expectation. The output proportion of the control amount is allocated by using the feedforward control based on empirical data and the feedback control based on error, and the output proportion of the control amount is allocated by using the feedforward control based on empirical data and the feedback control based on error. The adjustment parameter setting of the feedforward control based on empirical data is 0.35, the feedback control based on error uses a PID controller, the control input is the pull rod displacement, and the control output is the cone mutual difference value. The initial parameters used in the test are 0.01, 0.03, and 0, and the output control amount of the pull rod is calculated by the error of the cone error value and the error of the cone control expectation.

[0137] After the vibration control pull rod adjustment amount and the cone control pull rod adjustment amount are calculated, the control output proportion is allocated by the weight value. The coupling control of the vibration and the cone is completed. The adjustment amount is output to the pull rod, so as to realize the real-time adjustment control of the vibration and the cone of the helicopter rotor.

[0138] The method in the application is integrated into software, initial state data is received, and the adjustment amount of the pull rod is calculated. The calculated pull rod adjustment amount is issued as a digital instruction to the intelligent variable pitch pull rod and is actuated to the specified pull rod position. The vibration amplitude, phase, and cone data at the current position are received again, closed-loop control is realized, and the vibration and cone values reach the expected values, i.e., the control is stopped. If the adjustment effect is not good, the vibration weight value, the cone weight value, the coupling parameter, and the cone parameter can be re-set for continuous adjustment, or the model can be re-built. For example, as shown in the control results (1) and (2) in the test, the cone and dynamic balance amplitude of the rotor in the test are respectively within 4 mm and 0.05 IPS. Figure 26 、 Figure 27

[0139] 3. Test summary

[0140] After the test verification, Figure 28 、 Figure 29 、 Figure 30 ​The vibration adjustment results (1) (2) (3) charts can show that the X, Y, Z direction rotor rotation frequency amplitude is reduced by more than 50%, which shows that the application can effectively reduce the rotation frequency amplitude under the rotor working state and control it within 0.05IPS. From the Figure 31 、 Figure 32 、 Figure 33 The cone adjustment results (1) (2) (3) charts can show that the application can control the rotor cone value within 4mm.

[0141] It can be understood that the ordinary skilled in the art can make equivalent replacements or changes according to the technical solutions and inventive concepts of the application, and all these changes or replacements shall belong to the protection scope of the appended claims of the application.

Claims

1. A method for real-time adjustment of helicopter rotor balance, characterized by: The steps are as follows: Step 100: Calculate the adjustment amount ADJ of each intelligent variable pitch lever in the helicopter rotor vibration balance adjustment using the control model Vib , using PID feedback control algorithm to calculate the adjustment amount ADJ of each intelligent pitch lever in the helicopter rotor cone balance adjustment Cone_b , using the feedforward control algorithm to calculate the adjustment amount ADJ of each intelligent pitch lever in the helicopter rotor cone balance adjustment Cone_f ; Among them, the adjustment amount ADJ in vibration balance adjustment Vib The calculation process includes the following steps: Step 111: Screen the collected vibration data of the helicopter rotor to remove abnormal vibration vectors and retain normal vibration vectors. The vibration data screening method is: When a helicopter rotor has five blades, the amplitude and phase data of the four intelligent variable pitch levers on the X, Y, and Z axes are collected under three equally spaced positive force displacements and three equally spaced negative force displacements. The vibration vector ρ is a matrix consisting of the intelligent variable pitch lever number α, the lever displacement β, the amplitude δ, and the phase θ, where ρ = [α, β, δ, θ]. The positive force vector obtained under the positive force displacement and the negative force vector obtained under the negative force displacement of each intelligent variable pitch rod are sorted in order from small to large phases to obtain θ min ,θ mid ,θ max , and then we get two phase differences θ Δ1 ,θ Δ2 , the formula is as follows: i Δ1 =θ mid -θ min (1) i Δ2 =θ max -θ mid (2) When θ Δ1 and θ Δ2 When both are in the interval (0,72), keep θ min ,θ mid ,θ max The corresponding vibration vector; when θ Δ1 In the interval (0,72), θ Δ2 In the interval (288, +∞), remove θ max The corresponding vibration vector, retaining θ min and θ mid The corresponding vibration vector; when θ Δ2 In the interval (0,72), θ Δ1 In the interval (288, +∞), remove θ min The corresponding vibration vector, retaining θ mid and θ max The corresponding vibration vector; when θ Δ1 In the interval (288, +∞), θ mid ,θ max Subtract 360 from each and reorder and calculate θ Δ1 ,θ Δ2 , and then perform vibration vector screening; when θ Δ2 In the interval (288, +∞), θ max Reorder and calculate θ after subtracting 360 Δ1 ,θ Δ2 , and then perform vibration vector screening; when θ Δ1 and θ Δ2 When both are in the interval [72,288], remove θ min ,θ mid ,θ max The corresponding vibration vector; Step 112: Establish a control model based on the mapping relationship between the displacement, amplitude δ, phase θ of the intelligent variable pitch pull rod and the helicopter vibration vector; Step 113: Real-time vibration data of the helicopter rotor is collected. The vibration imbalance points on the X-axis, Y-axis, and Z-axis are rotated 180 degrees to obtain the adjustment vectors required to compensate for the imbalance. The closest vectors on either side of the compensation adjustment vectors on the X-axis, Y-axis, and Z-axis are then found from the control model. The vectors are then decomposed to obtain the adjustment values ​​of each intelligent variable pitch lever on the X-axis, Y-axis, and Z-axis. Step 200: Adjust the vibration adjustment amount ADJ of each intelligent variable pitch rod. Vib , cone feedforward adjustment adjustment amount ADJ Cone_b , cone feedback adjustment amount ADJ Cone_f Perform weighted adjustment and obtain the final weighted adjustment amount as ADJ final ; 2. A method for real-time adjustment of helicopter rotor balance according to claim 1, characterized in that The control model establishment process in step 112 is as follows: The positive force vectors that are screened are added together, and the rod displacement of the resultant force vector is the maximum rod displacement in the positive force vector, and the final positive vibration vector is expressed as ρ + =[α + ,β +max ,δ + ,θ + ], where α + is the intelligent pitch change lever number, β +max is the maximum displacement among the positive force vectors screened out; similarly, the negative force vector is expressed as ρ - =[α - ,β -max ,δ - ,θ - ], where β -max is the maximum displacement among the negative force vectors screened out; through the above steps, the positive force vectors and negative force vectors of each intelligent variable pitch rod on the X-axis, Y-axis, and Z-axis are finally obtained, and the matrix formed is the control model.

3. A method for real-time adjustment of helicopter rotor balance according to claim 1, characterized in that The calculation process of the adjustment amount of the X-axis, Y-axis, and Z-axis of each intelligent variable pitch lever in step 113 is the same, and includes the following process: Sort the vibration vector of a certain axis in the control model from small to large phase, and find the phase θ of the offset force on the axis. F The adjacent left and right phases [θ L2 ,θ L1 ,θ F ,θ R1 ,θ R2 ] corresponding vibration vector, θ F θ on the left L1 The corresponding vibration vector is LW1; θ L2 The corresponding vibration vector is LW2; θ F θ on the right R1 The corresponding vibration vector is UP1; θ R2 The corresponding vibration vector is UP2; When LW1 can synthesize the left component of the offsetting force by using only some of its components, the rod displacement of the corresponding vector of LW1 is calculated according to the rod adjustment ratio as the adjustment amount. When LW1 still cannot synthesize the left component of the offsetting force after using all its components, the output adjustment amount includes not only the full rod displacement of LW1 but also the rod displacement of the corresponding vector of LW2 calculated according to the rod adjustment ratio. When UP1 can synthesize the right component of the offsetting force by using part of its components, the rod displacement of the corresponding vector of UP1 is calculated as the adjustment amount according to the rod adjustment ratio; when UP1 still cannot synthesize the left component of the offsetting force after using all its components, the output adjustment amount includes not only the full rod displacement of UP1, but also the rod displacement of the corresponding vector of UP2 calculated according to the rod adjustment ratio.

4. A method for real-time adjustment of helicopter rotor balance according to claim 1, characterized in that Step 1: Use PID feedback control algorithm to calculate the adjustment amount ADJ of each intelligent pitch lever in the helicopter rotor cone balance adjustment. Cone_b The process includes the following steps: Step 121: Input the expected value E of each cone adjustment; input the actual measured value C of each cone; input the proportional coefficient KP of P, the proportional coefficient KI of I, and the proportional coefficient KD of D; input the initial values ​​of P_ and I_; Step 122: Calculate the value of the output U of the proportional unit P, the integral unit I, the differential unit D, and the PID feedback control algorithm using the following formula: P=EC (16) I=I_+P (17) D=P-P_ (18) U=KP·P+KI·I+KD·D (19) Step 123: Update the values ​​of P_ and I_. The formula is as follows: P_=P (20) I_=I (21) Step 124: The calculation formula for the adjustment amount of each intelligent pitch-changing rod corresponding to the cone is as follows: ADJ Cone_b = 0.1(-U) (22).

5. A method for real-time adjustment of helicopter rotor balance according to claim 1, characterized in that Step 1: Use the feedforward control algorithm to calculate the adjustment amount ADJ of each intelligent pitch lever in the helicopter rotor cone balance adjustment. Cone_f The process consists of the following steps: Step 131: Input the expected value E of cone adjustment f ; Input the actual measurement value A of the cone f ; Input the proportional parameter Per of feedforward control; Step 132: Calculate the adjustment amount ADJ of each smart lever Cone_b =U f , if the measured value is positive, the calculation formula is as follows: If the measured value is negative, the calculation formula is as follows:

6. A method for real-time adjustment of helicopter rotor balance according to claim 1, characterized in that Step 200 also includes the weighted ADJ final Make a judgment, if ADJ final If the absolute value of the intelligent variable pitch rod is less than the maximum displacement, the ADJ final ; if ADJ final If the absolute value of is greater than the maximum displacement of the intelligent variable pitch rod, the maximum displacement of the intelligent variable pitch rod is output.

Images