Solid rocket engine shell damage monitoring sensor network arrangement method

Abstract

The invention discloses a solid rocket engine shell damage monitoring sensor network arrangement method, and aims at monitoring damage of a solid rocket engine shell cylinder section. In a monitoring experiment, a piezoelectric patch (PZT) is used as excitation, an optical fiber (FBG) sensor is used as receiving, and a probability imaging method is adopted in post-processing to position damage. According to the invention, the non-blind area monitoring of the shell cylinder section of the solid rocket engine is realized, and the layering damage with the diameter greater than 40mm can be positioned by utilizing a probability imaging method based on guided waves.

Description

Technical Field

[0001] The invention belongs to the technical field of health detection, and in particular relates to a method for laying out a sensor network for monitoring damage of a solid rocket engine casing. Background Art

[0002] The solid rocket engine casing works in a complex environment and has a unique failure mode, which places high demands on its damage monitoring technology: the monitoring system should be able to identify damage in the casing in a timely manner, have strong anti-electromagnetic interference capabilities and measurement stability and reliability, and the number of sensors and demodulators should be as small as possible and the size should be as small as possible to avoid affecting the normal operation of the casing.

[0003] Ultrasonic guided wave technology has the advantage of detecting tiny damages and can generate ultrasonic waves with low-cost piezoelectric sensors and very little energy. In addition, guided waves can propagate over long distances with little attenuation, and can cover a large area with relatively few sensors. Ultrasonic guided wave technology has attracted widespread attention from researchers due to its high sensitivity, wide monitoring range, and relatively high positioning accuracy.

[0004] Domestic and foreign scholars have conducted extensive research on the structural health monitoring method of composite materials based on ultrasonic guided waves. Castaings et al. in France investigated the influence of the size of impact damage in composite plates on the S0 mode guided wave signal and found that ultrasonic guided waves can quantitatively detect the size of delamination damage in composite materials. Luca et al. in Italy studied the interaction between the characteristics of low-speed impact damage of composite materials and guided waves. They believed that the damage of composite materials caused by impact is complex. When guided waves are used to locate damage, there will be inaccurate results or low precision. The mechanism of action between the two needs further study. Based on the above research, Fenza et al. in Italy determined the location and degree of damage in composite laminates and established the corresponding relationship between guided wave characteristics and laminate damage index. Harb et al. used a non-contact empty probe and a laser ultrasonic generator to excite Lamb waves dominated by the A0 mode in composite laminates, and used this mode guided wave to detect damage in laminates. The results show that these two excitation forms have high sensitivity and efficiency. Based on this study, Ramadas et al. and Schaal et al. investigated the wave conduction of Lamb waves in composite T-joints and variable thickness composite plates, respectively, and discussed in detail the influence mechanism of interface delamination damage and thickness change in T-joints on ultrasonic guided waves. Yang et al. used three-dimensional finite element simulation and experimental technology to investigate the influence mechanism of parameters such as composite bolt connection joint form, number of connection holes and load size on guided wave conduction behavior, and established the correlation between guided wave characteristic signals and joint failure modes.

[0005] It can be seen that a lot of research has been carried out on the damage monitoring of composite structures using ultrasonic guided wave methods. However, most of the above research is aimed at simple structures such as flat plates or stiffened plates, while there are few studies on the structural damage of solid rocket motor casings, and the corresponding sensor network layout method also needs to be studied. Summary of the invention

[0006] In order to overcome the shortcomings of the prior art, the present invention provides a method for laying a network of sensors for monitoring damage in a solid rocket engine casing, and carries out damage monitoring for the barrel section of the solid rocket engine casing. The monitoring experiment uses a piezoelectric film (PZT) as an excitation and an optical fiber (FBG) sensor as a receiver, and a probabilistic imaging method is used in post-processing to locate the damage. The present invention realizes blind-spot monitoring of the barrel section of the solid rocket engine casing, and can locate layered damage with a diameter greater than 40 mm using a probabilistic imaging method based on guided waves.

[0007] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0008] Step 1: Select the size of the piezoelectric film and the length of the fiber grating;

[0009] Step 2: Paste a piezoelectric sheet on the solid rocket engine casing, and paste optical fiber sensors at different distances from the piezoelectric sheet along the circumference of the casing; use the same excitation signal frequency and amplitude to conduct a set of experiments, monitor the amplitude and signal quality of each received signal, and thus determine the range of the circumferential length of the sensing path;

[0010] Step 3: Make the axial distance between sensors equal to the distance between sensor groups to determine the number of sensors in each group;

[0011] Step 4: Measure the circumference and axial length of the shell monitoring area, and determine that the distance between sensor groups is 1 / 2 of the circumferential length of the sensing path; divide the circumference of the shell monitoring area equally under the condition that the circumferential length range measured in step 2 is met, and determine the number of sensors in each group and arrange them evenly according to the method in step 3;

[0012] Step 5: Take the shell specimen to be tested and lay out the sensor network on the specimen according to the designed network distance; evenly arrange PZT sensors as excitation and FBG sensors as reception around the circumference of the shell, and alternately arrange two groups of PZT and two groups of FBG in the circumferential direction, that is, there are four groups of sensors in one cycle. In the experiment, PZT and FBG separated by one group of sensors are used for damage monitoring experiments.

[0013] Step 6: Paste all sensors and cure them at room temperature for about 8 hours to ensure that the PZT sensor, FBG sensor grid area and shell are firmly pasted; connect the experimental system to check whether the signal can be excited and received normally;

[0014] Step 7: Select the monitoring area and debug the wavelength of each FBG sensor in the monitoring area; use the NI signal generator to output the Hanning window modulated sine signal to excite the PZT, and use the FBG sensor to receive and collect the signal of each excitation-sensing path in the monitoring area as the reference signal, and save the signal data file;

[0015] Step 8: Apply coupling agent at any position in the monitoring area of ​​the specimen and place a simulated damage, record the position of the simulated damage, use the same experimental conditions as step 7, collect the signal of each excitation-receiving path under the damage state, and save the signal data file;

[0016] Step 9: For each stimulus-receive path, the measured reference signal and the impairment signal are used to calculate the impairment index SDC as the reference impairment index, recorded as DI. The SDC index is determined by the following formula:

[0017]

[0018] Among them, b i represents the reference signal of path i, c i represents the monitoring signal of path i, and N represents the number of waveguide signal data points; represents the average value of the reference signal of path i, represents the average value of the monitoring signal of path i;

[0019] Step 10: Assuming that the sensor network has n excitation-receiving sensing paths, divide the rectangular monitoring area into several discrete pixels, and use the following formula to calculate the distance R from any pixel position (x, y) to the nth sensing path: n (x,y):

[0020]

[0021] Where: D n is the distance between the actuator and the receiver in the nth sensing path, that is, the distance between the PZT and the FBG; D an (x,y) and D sn (x, y) are the distances from the pixel point (x, y) to the PZT and FBG respectively;

[0022] Step 11: According to the distance R from the pixel point (x, y) to the nth sensing path n (x, y), calculate the weight factor W of this point relative to the nth sensing path n , as shown below:

[0023]

[0024] Among them: β is a parameter that controls the size of the area affected by the damage factor of a single path. It is an elliptical area with the sensor FBG and the actuator PZT as the focus, indicating the maximum range of damage influence. The probability of damage beyond this range is zero;

[0025] Step 12: Calculate the probability of damage at each pixel (x, y), expressed as follows:

[0026]

[0027] Where DI is the damage factor on the nth sensor path, W n [R n (x, y)] is the damage probability weighted distribution function that maps the damage factor value to the pixel point (x, y); p n (x,y) represents the damage probability value brought to the point (x,y) by the nth sensing path;

[0028] Step 13: Display the damage probability of each pixel in the monitoring area in the form of an image. The location with the maximum probability is the monitored simulated damage location. Adjust the threshold to display the maximum damage location.

[0029] Step 14: Perform the monitoring process from step 2 to step 8 on each monitoring area of ​​the shell to conduct comprehensive damage monitoring on the shell.

[0030] Preferably, the diameter of the piezoelectric sheet is 10 mm.

[0031] Preferably, the fiber grating length is 10 mm.

[0032] Preferably, in step 6, epoxy structural adhesive is used to adhere all sensors.

[0033] The beneficial effects of the present invention are as follows:

[0034] The present invention uses a sensor networking method with two groups of excitations and two groups of receptions interlaced. The blind spots in the sensor axis area are covered by interlacing the monitoring areas, thereby realizing blind-spot monitoring of the barrel section of the solid rocket engine casing. The layered damage with a diameter greater than 40 mm can be located by a probabilistic imaging method based on guided waves. Most of the monitoring areas can be covered twice by a sensor network composed of different sensors, so that the networking scheme has a certain fault tolerance to sensor damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a physical diagram of the experimental system used in the present invention;

[0036] Figure 2 This is a schematic diagram of the sensor network layout scheme of the present invention;

[0037] Figure 3 A physical map of the monitoring area in the embodiment of the present invention;

[0038] Figure 4 This is the imaging result of the damage localization according to the embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0040] The purpose of the present invention is to propose a sensor networking method for monitoring damage to solid rocket engine casings. The method solves the current problem of lack of a sensor networking method for monitoring damage to solid rocket engine casings. The proposed networking method has comprehensiveness in the monitoring range and fault tolerance to sensor damage, thereby achieving accurate positioning of simulated damage.

[0041] The purpose of the present invention is solved by the following technical solutions:

[0042] The barrel section of the solid rocket motor wrapped around the casing is usually a weak link prone to damage, so the present invention carries out damage monitoring for the barrel section of the solid rocket motor casing. The monitoring experiment uses a piezoelectric film (PZT) as an excitation and an optical fiber (FBG) sensor as a receiver, and a probabilistic imaging method is used in post-processing to locate the damage.

[0043] Step 1: Select the size of the piezoelectric piece and the fiber grating length. Under the condition that the piezoelectric piece fits well with the shell, select a piezoelectric piece with as large a diameter as possible; select a fiber sensor with a larger grating length (about 10mm). When sensing the same guided wave strain field, the amplitude of the received guided wave signal is larger for a fiber with a larger grating length.

[0044] Step 2: Paste a piezoelectric sheet on the shell, and paste optical fiber sensors at different distances from the piezoelectric sheet along the circumferential direction. Use the same excitation signal frequency and amplitude to conduct a set of experiments, observe the amplitude and signal quality of each received signal, and thus determine the range of the circumferential length of the sensing path.

[0045] Step 3: Determine the axial distance between sensors, that is, determine how many sensors a group contains. The axial distance of sensors should be roughly equal to the distance between sensor groups.

[0046] Step 4: Measure the circumference and axial length of the shell monitoring area, and specifically determine the distance between sensor groups (1 / 2 of the circumferential length of the sensing path in step 2). Divide the circumference equally while meeting the circumferential length range measured in step 2. Use the same method to determine the number of sensors in each group and arrange them evenly.

[0047] Step 5: Take the shell specimen to be tested and lay out the sensor network on the specimen according to the designed network distance. PZT sensors (as excitation) and FBG sensors (as reception) are evenly arranged around the circumference of the shell, and two groups of PZT and two groups of FBG are arranged alternately in the circumferential direction, that is, there are four groups of sensors in one cycle. In the experiment, PZT and FBG separated by one group of sensors are used for damage monitoring experiments.

[0048] Step 6: Use epoxy structural adhesive to stick all sensors and cure them at room temperature for about 8 hours to ensure that the PZT sensor and FBG sensor grid area are firmly attached to the shell; connect the experimental system to check whether the signal can be stimulated and received normally.

[0049] Step 7: Select the monitoring area and debug the wavelength of each FBG sensor in the area. Use the NI signal generator to output a Hanning window modulated sine signal to excite the PZT, and use the FBG sensor to receive and collect the signal of each excitation-sensing path in the monitoring area as the reference signal, and save the signal data file;

[0050] Step 8: Apply coupling agent at any position in the monitoring area of ​​the specimen and place a simulated damage, record the position of the simulated damage, use the same experimental conditions as step 7, collect the signal of each excitation-receiving path under the damage state, and save the signal data file;

[0051] Step 9: For each stimulus-receive path, the measured reference signal and the damage signal are used to calculate the damage index SDC as the reference damage index, recorded as DI. The SDC index is determined by the following formula:

[0052]

[0053] Step 10: Assume that the sensor network has n excitation-receiving sensing paths, divide the rectangular monitoring area into several discrete pixels, and use the following formula to calculate the distance R from any pixel position (x, y) to the nth sensing path: n (x,y):

[0054]

[0055] Step 11: According to the distance R from the pixel point (x, y) to the nth sensing path n (x, y), calculate the weight factor Wn of this point relative to the nth sensing path, as shown in the following formula:

[0056]

[0057] Among them: β is a parameter that controls the size of the area affected by the damage factor of a single path. It is an elliptical area with the sensor FBG and the actuator PZT as the focus, indicating the maximum range of damage influence. The probability of damage beyond this range is zero, and its value has a certain degree of experience.

[0058] Step 12: Calculate the probability of damage at each pixel (x, y), expressed as follows:

[0059]

[0060] Step 13: Display the damage probability of each pixel in the monitoring area in the form of an image. The point with the maximum probability is the monitored simulated damage position. Adjust the threshold to display the maximum damage position.

[0061] Step 14: Perform the monitoring process from step 2 to step 8 on each monitoring area of ​​the shell to conduct comprehensive damage monitoring on the shell.

[0062] Example:

[0063] At a reduced ratio Taking the wound composite shell test piece as an example, the monitoring method of the present invention is further described in detail:

[0064] Step 1: Select the size of the piezoelectric sheet and the length of the optical fiber grating. Under the condition of ensuring that the piezoelectric sheet fits well with the shell, a piezoelectric sheet with a diameter of 10 mm was selected; and an optical fiber sensor with a grating length of 10 mm was selected.

[0065] Step 2: Paste a piezoelectric sheet on the shell, and paste optical fiber sensors at 100mm, 150mm, and 200mm away from the piezoelectric sheet along the circumferential direction. Use a 70kHz Hanning window modulated sinusoidal signal as the excitation signal, observe the amplitude and signal quality of each received signal, and determine that the circumferential length of the sensing path is in the range of 100-150mm.

[0066] Step 3: Determine the axial distance between sensors, that is, determine how many sensors a group contains. The axial distance of sensors should be roughly equivalent to the distance between sensor groups (1 / 2 of the circumferential length of the path), about 50-75mm, so as to minimize the number of sensors and make the sensor network as dense as possible.

[0067] Step 4: The circumference of the shell monitoring area is measured to be 680mm, and the axial length is 200mm. Specifically, the distance between the sensor groups is determined to be 56mm. At this time, the circumferential length of the sensing path is 112mm, which meets the requirements of the circumferential length range measured in step 2. In the same way, the number of sensors in each group is determined to be 3 and evenly arranged.

[0068] Step 5: Take the shell specimen to be tested and lay out the sensor network on the specimen. Arrange PZT sensors and FBG sensors around the circumference of the shell, and alternately arrange two groups of PZT and two groups of FBG in the circumferential direction, that is, one cycle includes four groups of sensors. In the experiment, PZT and FBG with one group of sensors are used to conduct damage monitoring experiments. For example, in monitoring Figure 2 When the damage occurs in the overlapping area of ​​the blue and red dashed boxes, PZT group 1 and FBG group 1 can be used for monitoring, or PZT group 2 and FBG group 2 can be used for monitoring. For damage occurring at the sensor axis, the adjacent PZT-FBG sensor groups on both sides can also be used for monitoring experiments. The axial distance between each group of sensors is 65 mm; the circumferential spacing between each group of sensors is 56 mm.

[0069] Step 6: Use epoxy structural adhesive to paste all sensors and cure them at room temperature for about 8 hours to ensure that the PZT sensor, FBG sensor grid area and shell are firmly attached; connect the experimental system to check whether the signal can be excited and received normally;

[0070] Step 7: Select the monitoring area and debug the wavelength of each FBG sensor in the area. Use the NI signal generator to output a 3.5-cycle Hanning window modulated sinusoidal signal with a gain of 1.8 to excite the PZT. At the same time, use the FBG sensor to receive and collect the signal of each excitation-sensing path in the monitoring area as the reference signal. When collecting each signal, perform 128 averages and save the signal data file.

[0071] Step 8: Apply coupling agent in the axis area of ​​the FBG group in the middle of the monitoring area, 30mm away from the edge FBG, and place a 20mm diameter weight block as a simulated damage. The simulated damage coordinates are (70,40). Use the same experimental conditions as step 7 to collect the signal of each excitation-receiving path under the damaged state, and also average each signal 128 times, and save the signal data file;

[0072] Step 9: For the 9 stimulus-receive paths, the measured reference signal and the damage signal are used to calculate the damage index SDC as the reference damage index, recorded as DI. The SDC index is determined by the following formula:

[0073]

[0074] Step 10: The sensor network in the experiment has 9 excitation-receiving sensing paths. The rectangular monitoring area is divided into several discrete pixels. The distance R from any pixel position (x, y) to the nth sensing path is calculated using the following formula: n (x,y):

[0075]

[0076] Step 11: According to the distance R from the pixel point (x, y) to the nth sensing path n (x, y), calculate the weight factor Wn of this point relative to the nth sensing path, as shown in the following formula:

[0077]

[0078] Among them: β is a parameter that controls the size of the area affected by the damage factor of a single path. It is an elliptical area with the sensor FBG and the actuator PZT as the focus, indicating the maximum range of damage influence. The probability of damage beyond this range is zero. Its value is empirical and is taken as 0.097 here.

[0079] Step 12: Calculate the probability of damage at each pixel (x, y), expressed as follows:

[0080]

[0081] Step 13: Display the damage probability of each pixel in the monitoring area in the form of an image. The point with the maximum probability is the monitored simulated damage position. Adjust the threshold to display the maximum damage position.

[0082] Step 14: Perform the monitoring process from step 2 to step 8 on each monitoring area of ​​the shell to conduct comprehensive damage monitoring on the shell.

Claims

1. A method for laying out a sensor network for monitoring damage of a solid rocket engine casing, characterized in that: The steps include: Step 1: Select the size of the piezoelectric film and the length of the fiber grating; Step 2: Paste a piezoelectric sheet on the solid rocket engine casing, and paste optical fiber sensors at different distances from the piezoelectric sheet along the circumference of the casing; use the same excitation signal frequency and amplitude to conduct a set of experiments, monitor the amplitude and signal quality of each received signal, and thus determine the range of the circumferential length of the sensing path; Step 3: Make the axial distance between sensors equal to the distance between sensor groups to determine the number of sensors in each group; Step 4: Measure the circumference and axial length of the shell monitoring area, and determine that the distance between sensor groups is 1 / 2 of the circumferential length of the sensing path; divide the circumference of the shell monitoring area equally under the condition that the circumferential length range measured in step 2 is met, and determine the number of sensors in each group and arrange them evenly according to the method in step 3; Step 5: Take the shell specimen to be tested, and arrange the sensor network on the specimen according to the designed network distance; evenly arrange PZT sensors as excitation and FBG sensors as reception around the circumference of the shell, and alternately arrange two groups of PZT and two groups of FBG in the circumferential direction, that is, there are four groups of sensors in one cycle. In the experiment, PZT and FBG separated by one group of sensors are used for damage monitoring experiments; Step 6: Paste all sensors and cure them at room temperature for about 8 hours to ensure that the PZT sensor, FBG sensor grid area and shell are firmly pasted; connect the experimental system to check whether the signal can be excited and received normally; Step 7: Select the monitoring area and debug the wavelength of each FBG sensor in the monitoring area; Use the NI signal generator to output a Hanning window modulated sinusoidal signal to excite the PZT, and use the FBG sensor to receive it. Collect the signal of each excitation-sensing path in the monitoring area as the reference signal and save the signal data file; Step 8: Apply coupling agent at any position in the monitoring area of ​​the specimen and place a simulated damage, record the position of the simulated damage, use the same experimental conditions as step 7, collect the signal of each excitation-receiving path under the damage state, and save the signal data file; Step 9: For each stimulus-receive path, the measured reference signal and the impairment signal are used to calculate the impairment index SDC as the reference impairment index, recorded as DI. The SDC index is determined by the following formula: Among them, b i represents the reference signal of path i, c i represents the monitoring signal of path i, and N represents the number of waveguide signal data points; represents the average value of the reference signal of path i, represents the average value of the monitoring signal of path i; Step 10: Assuming that the sensor network has n excitation-receiving sensing paths, divide the rectangular monitoring area into several discrete pixels, and use the following formula to calculate the distance R from any pixel position (x, y) to the nth sensing path: n (x,y): Where: D n is the distance between the actuator and the receiver in the nth sensing path, that is, the distance between the PZT and the FBG; D an (x,y) and D sn (x, y) are the distances from the pixel point (x, y) to the PZT and FBG respectively; Step 11: According to the distance R from the pixel point (x, y) to the nth sensing path n (x, y), calculate the weight factor W of this point relative to the nth sensing path n , as shown below: Among them: β is a parameter that controls the size of the area affected by the damage factor of a single path. It is an elliptical area with the sensor FBG and the actuator PZT as the focus, indicating the maximum range of damage influence. The probability of damage beyond this range is zero; Step 12: Calculate the probability of damage at each pixel point (x, y), expressed as follows: Where DI is the damage factor on the nth sensor path, W n [R n (x, y)] is the damage probability weighted distribution function that maps the damage factor value to the pixel point (x, y); p n (x,y) represents the damage probability value brought to the point (x,y) by the nth sensing path; Step 13: Display the damage probability of each pixel in the monitoring area in the form of an image. The location with the maximum probability is the monitored simulated damage location. Adjust the threshold to display the maximum damage location. Step 14: Perform the monitoring process from step 2 to step 8 on each monitoring area of ​​the shell to conduct comprehensive damage monitoring on the shell.

2. A solid rocket engine casing damage monitoring sensor network deployment method according to claim 1, characterized in that: The diameter of the piezoelectric sheet is 10 mm.

3. A solid rocket engine casing damage monitoring sensor network deployment method according to claim 1, characterized in that: The length of the fiber grating is 10 mm.

4. A solid rocket engine casing damage monitoring sensor network deployment method according to claim 1, characterized in that: In step 6, all sensors are glued using epoxy structural adhesive.

Images