A Fault Diagnosis System and Method for a Heater Used in Wind Tunnel Tests

Abstract

The present invention provides a fault diagnosis system and method for a heater used in a wind tunnel test, which relates to the technical field of heater fault diagnosis and includes: an infrared imaging component and a voiceprint positioning component for comprehensively monitoring the heater; a heater control module for controlling the start of the heater and obtaining the environmental parameters of the heater; the air inside the heater is heated to obtain an air flow for the wind tunnel test; when the heater is started, the infrared imaging component and the voiceprint positioning component are respectively used to collect the infrared image and the voiceprint image outside the heater in real time; a switch is used to send the collected data to the fault diagnosis module; the fault diagnosis module determines the current test working condition according to the environmental parameters, and compares the environmental parameters, the infrared image and the voiceprint image with the historical data under the current test working condition to obtain a fault diagnosis result. This solution realizes non-contact comprehensive fault monitoring and diagnosis, and improves the fault diagnosis accuracy of the heater used in the wind tunnel test.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis for wind tunnel tests, and particularly to a fault diagnosis system and method for a heater used in wind tunnel tests. Background Art

[0002] Wind tunnel tests are used to simulate the performance of a research object in an air flow. Among them, a high-temperature wind tunnel is an experimental device that can simulate the air flow environment and aerodynamic effects when an aircraft and a spacecraft are flying at high speeds. By passing high-temperature and high-pressure air through the pipeline in the main body of the wind tunnel, a high-speed air flow is formed to simulate the air flow environment when the aircraft is flying at high speeds. The air flow environment in the high-temperature wind tunnel is generated by a heater. As a special equipment with a high usage proportion in the high-temperature wind tunnel, the annual usage frequency exceeds 6,000 times, and the working temperature can reach 1,000K to 7,000K, running at full load all year round. Therefore, once the heater fails, it may cause major losses. For example, damage to heating elements, overheating, damage to the housing, failure of the sealing ring, etc. will cause leakage of high-temperature gas, seriously affecting the safe operation of the wind tunnel.

[0003] Currently, the heaters used in wind tunnel tests mainly adopt two methods: regular maintenance every year and after-fact maintenance. Regular maintenance is to carry out equipment maintenance for one month at the beginning of each year and stop the test. At this time, the heater is in the best state, and there is even a phenomenon of over-maintenance. After-fact maintenance is carried out after the heater has failed. Since the heater is a high-temperature and dangerous equipment, once failures such as breakage, leakage, and overheating occur, the consequences will be unimaginable. After a failure occurs, the wind tunnel test will also be forced to stop, and potential hazards are difficult to be discovered in the first time, thus affecting the efficiency of the wind tunnel test. Therefore, there is an urgent need to provide a fault diagnosis system and method for a heater used in wind tunnel tests. Summary of the Invention

[0004] The present invention provides a fault diagnosis system and method for a heater used in wind tunnel tests. The system realizes non-contact comprehensive fault monitoring and diagnosis, improves the fault diagnosis accuracy of the heater used in wind tunnel tests, and reduces the missed detection rate.

[0005] In a first aspect, the present invention provides a fault diagnosis system for a heater used in wind tunnel tests, including: an infrared imaging component, a sound pattern positioning component, a switch, a heater, a heater control module, and a fault diagnosis module;

[0006] The heater control module controls the start of the heater and obtains the environmental parameters of the heater; the air inside the heater is heated to obtain an air flow for wind tunnel tests; the environmental parameters include total temperature and total air source pressure;

[0007] The infrared imaging component and the voiceprint positioning component are installed around the heater in a non-contact manner to monitor the heater in all directions; when the heater is started, the infrared imaging component and the voiceprint positioning component are respectively used to collect infrared images and voiceprint images outside the heater in real time;

[0008] The switch is used to send the environmental parameters, the infrared images and the voiceprint images to the fault diagnosis module;

[0009] The fault diagnosis module is used to determine the current test condition according to the environmental parameters, and compare the environmental parameters, the infrared images and the voiceprint images with the historical data under the current test condition to obtain a fault diagnosis result; and when it is determined that the fault diagnosis result is a fault, the moment is determined as the fault moment, and a fault solution is determined according to the infrared images collected within a preset time period before and after the fault moment.

[0010] In a second aspect, the present invention also provides a fault diagnosis method for a heater used in a wind tunnel test, including:

[0011] When the heater is started, obtain the environmental parameters of the heater, and collect infrared images and voiceprint images outside the heater in real time;

[0012] Determine the current test condition according to the environmental parameters, and compare the environmental parameters, the infrared images and the voiceprint images with the historical data under the current test condition to obtain a fault diagnosis result.

[0013] The present invention provides a fault diagnosis system and method for a heater used in a wind tunnel test. By combining an infrared imaging component and a voiceprint positioning component, the system realizes all-round monitoring of the heater used in the wind tunnel test. When the heater is started, it can perform comprehensive fault diagnosis in real time based on the obtained environmental parameters and the collected external infrared images and voiceprint images, thereby providing a non-contact fault diagnosis method for the heater used in the wind tunnel test, realizing automatic association of multiple state parameters of the heater, improving the accuracy of typical fault diagnosis of the heater, reducing the missed detection rate, and improving the intelligent level of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1It is a schematic diagram of a fault diagnosis system for a heater used in a wind tunnel test provided by an embodiment of the present invention;

[0016] Figure 2 It is a schematic diagram of another fault diagnosis system for a heater used in a wind tunnel test provided by an embodiment of the present invention;

[0017] Figure 3 It is a flowchart of a fault diagnosis method for a heater used in a wind tunnel test provided by an embodiment of the present invention;

[0018] Among them, 10 - infrared imaging component, 20 - acoustic fingerprint positioning component, 30 - switch, 40 - heater control module, 50 - fault diagnosis module. Detailed implementation manners

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] The following describes the specific implementation manners of the above concepts.

[0021] Please refer to Figure 1 , an embodiment of the present invention provides a fault diagnosis system for a heater used in a wind tunnel test, and the system includes: an infrared imaging component 10, an acoustic fingerprint positioning component 20, a switch 30, a heater control module 40, and a fault diagnosis module 50;

[0022] The heater control module 40 controls the heater to start and obtains the environmental parameters of the heater; the air is heated inside the heater to obtain an air flow for the wind tunnel test; the environmental parameters include total temperature and total air source pressure;

[0023] The infrared imaging component 10 and the acoustic fingerprint positioning component 20 are installed around the heater in a non-contact manner to monitor the heater in all directions; when the heater starts, the infrared imaging component 10 and the acoustic fingerprint positioning component 20 are respectively used to collect the infrared image and the acoustic fingerprint image outside the heater in real time;

[0024] The switch 30 is used to send the environmental parameters, the infrared image, and the acoustic fingerprint image to the fault diagnosis module 50;

[0025] The fault diagnosis module 50 is used to determine the current test condition according to the environmental parameters, compare the environmental parameters, infrared images and acoustic fingerprint images with the historical data under the current test condition to obtain the fault diagnosis result; and when the fault diagnosis result indicates a fault, confirm this moment as the fault moment, and determine the fault solution according to the infrared images collected within a preset time period before and after the fault moment.

[0026] In the present invention, by combining the infrared imaging component and the acoustic fingerprint positioning component, the omnidirectional monitoring of the heater for wind tunnel tests is realized. When the heater is started, the comprehensive fault diagnosis can be carried out in real time based on the obtained environmental parameters and the collected external infrared images and acoustic fingerprint images. Thus, a non-contact fault diagnosis method is provided for the heater for wind tunnel tests, realizing the automatic correlation of multiple state parameters of the heater, improving the accuracy of diagnosing typical faults of the heater, reducing the missed detection rate, and enhancing the intelligent level of fault diagnosis.

[0027] In a preferred embodiment, the heater is obtained by connecting at least two-stage heating components in series; each stage of the heating component employs at least two infrared imaging components.

[0028] In a preferred embodiment, the heater includes a heating element, electrode rods, a heat-bearing layer, a heat-insulating layer, a pressure-bearing layer and a cooling structure; inside the heater, the air is heated by the heating element, and the cooling structure is used for the circulating cooling water to cool the electrode rods.

[0029] In a preferred embodiment, the fault diagnosis result includes having a fault and no fault; having a fault includes: having an abnormal temperature, having a gas or liquid leakage.

[0030] Specifically, for example, in a wind tunnel test, a heater obtained by connecting five-stage heating components in series is used to heat the air, and the temperature can reach 1100K to obtain the air flow for high-temperature wind tunnel tests. Each stage of the heating component is composed of a heating element, electrode rods, a heat-bearing layer, a heat-insulating layer, a pressure-bearing layer, a cooling structure and a flange, etc. Each stage of the heating component has 18 electrode rods. One end of the electrode rod is connected to 380V high-voltage electricity, and the other end is connected to the internal heating element. It penetrates into the heater through the flange, and requires insulation ceramics for insulation and sealing rings for sealing. At the same time, the cooling structure is used to guide the cooling water to the electrode rods to achieve effective cooling through heat exchange, which can not only ensure the cooling efficiency, but also avoid the risk of fluid leakage through the rigid structure of the pressure-bearing layer. The heating element is located inside the heat-bearing layer, there is a heat-insulating layer outside the heat-bearing layer, and a pressure-bearing layer outside the heat-insulating layer. Therefore, for this type of heater, during its operation, there may be situations of liquid leakage, gas leakage and abnormal temperature.

[0031] In a more preferred embodiment, the resolution of the infrared imaging component is not less than 256*192, and the frame rate is not less than 25fps; the frequency band of the voiceprint localization component is 5kHz to 72kHz.

[0032] As Figure 1 shown, there can be several infrared imaging devices and voiceprint localization devices, which are specifically determined by the heaters used in the wind tunnel test. Specifically, as described in the previous example, as Figure 2 shown, 10 infrared imaging devices and 1 voiceprint localization device are used in this fault diagnosis system. The resolution of the infrared imaging device is not less than 256*192, the frame rate is not less than 25fps, the temperature measurement distance is 3m, and the measurement range covers 0 to 300°C; every two infrared imaging devices monitor one-level heating components, so as to achieve full-range monitoring of the electrode rods of the heater. The voiceprint localization device has the functions of sound source localization and visual imaging, covering the frequency band of 5kHz to 72kHz, the field of view angle: horizontal ≥ 60°, vertical ≥ 45°, and the voiceprint localization device is installed at a position 10m horizontally and 10m high from the heater, with a 45° downward view, so as to achieve full-range sound source monitoring of the five-level heater.

[0033] It should be noted that the heater of the present invention can be a high-temperature pressure-bearing heater or a direct-connected heater, and can also realize fault diagnosis for the leakage of normal-temperature high-pressure containers.

[0034] In a preferred embodiment, the infrared imaging component, the voiceprint localization component, and the heater control module are deployed in different network segments, as Figure 2 shown, that is, through different network switches, which can avoid network conflicts between device systems and improve the network stability of the system; among them, the fault monitoring and diagnosis server is the fault diagnosis module.

[0035] In a preferred embodiment, the fault diagnosis module includes a working condition determination unit and a temperature diagnosis unit;

[0036] The working condition determination unit is used to compare the environmental parameters with the historical environmental parameters in the historical data to determine the current test working condition; among them, the historical data records the historical environmental parameters corresponding to each test working condition and the historical temperature ranges of each electrode rod during normal operation;

[0037] The temperature diagnosis unit is used to determine the temperature values of each electrode rod according to the infrared images under the current test working condition, compare the temperature values of each electrode rod with the historical temperature ranges under the current test working condition, and when the temperature value is not within the historical temperature range, determine that the fault diagnosis result is that there is a temperature anomaly.

[0038] It should be noted that the historical environmental parameters corresponding to each test condition are the environmental parameters when there is no fault. Continuing with the previous example, when the gas temperature inside the heater reaches 1100K under the current test condition, the historical temperature range of the electrode rods during normal operation of the heater is around 100°C. That is, the historical temperature range is the normal temperature threshold range of the electrode rods. If it exceeds this range, it is considered that there is a temperature anomaly and the heater is operating abnormally. Since different test conditions have different manifestations, the temperature values outside the heater are different under different test conditions.

[0039] In a preferred embodiment, the condition determination unit is used to perform the following operations:

[0040] When the environmental parameters are the same as the historical environmental parameters, determine that the current test condition of the environmental parameters is the test condition corresponding to the historical environmental parameters; where the environmental parameters include total temperature and total gas source pressure.

[0041] When the historical data does not contain the environmental parameters, determine the first historical environmental parameter closest to the environmental parameters from the historical data, and judge whether the difference between the first historical environmental parameter and the environmental parameters does not exceed a preset threshold. If so, determine that the current test condition of the environmental parameters is the test condition corresponding to the first historical environmental parameter; otherwise, generate the current test condition and store it in the historical data.

[0042] It should be noted that the infrared imaging component and the voiceprint positioning component collect data in real time, but only start data storage or stop data storage after receiving a specified OPC signal (such as a heater start signal, a heater shutdown signal), and at the same time record all environmental parameters during the operation of the heating equipment: total temperature, total gas source pressure, heater power, etc. In this way, irrelevant data when the heater is not working is avoided, the quality and effectiveness of the data are improved, and false alarms are reduced.

[0043] In the present invention, the total temperature and the total gas source pressure are important environmental parameters affecting the heating condition of the heater. Therefore, the system adaptively selects the corresponding threshold from the database of historical environmental parameters by automatically obtaining the OPC signals (total temperature, total gas source pressure) of the environmental parameters of the heater as the current test condition of the current heater. In this way, using the total temperature and the total gas source pressure of the environmental parameters as index values, if there are no corresponding total temperature and total gas source pressure in the historical environmental parameters, find the test condition where the threshold corresponding to the relatively close total temperature and total gas source pressure is located. For example, the threshold is that the total temperature fluctuation range does not exceed 50K and the total gas source pressure fluctuation range does not exceed 0.5 Mpa. If there is still no suitable threshold within this range, the system will remind the administrator of the new condition, directly generate the current test condition, and set the threshold of the current test condition.

[0044] In a preferred embodiment, the temperature diagnosis unit is configured to perform the following operations:

[0045] Under the current test condition, determine whether the temperature value is measured during the first wind tunnel test of the heater on the current day;

[0046] If not, use the historical temperature range without faults corresponding to the current test condition as the normal threshold. When the temperature value is within the normal threshold, determine that the fault diagnosis result is normal operation; when the temperature value is not within the normal threshold, determine that the fault diagnosis result is a temperature anomaly.

[0047] It should be noted that the historical temperature range of the electrode rod is the temperature value outside the heater. As described in the previous example, when operating normally without faults, this temperature value is about 100 °C.

[0048] In the present invention, since the temperature during the first wind tunnel test of the heater every day is different from the temperature during subsequent stable operation, the temperature value at the electrode rod is also different. When the temperature value is within the normal threshold under the current test condition, it is considered that the heater is operating normally; otherwise, the heater is operating abnormally. At the same time, during this process, the normal threshold can also be optimized and updated by accumulating the measured temperature values.

[0049] In a preferred embodiment, the fault diagnosis module includes a working condition determination unit and a leakage diagnosis unit;

[0050] The working condition determination unit is configured to compare the environmental parameters with the historical environmental parameters in the historical data to determine the current test condition; wherein, the historical data records the historical environmental parameters corresponding to each test condition, as well as the historical infrared images and historical sound pattern images marked with fault diagnosis results;

[0051] The leakage diagnosis unit is configured to compare the infrared image under the current test condition with the historical infrared images, and compare the sound pattern image under the current test condition with the historical sound pattern images to obtain the fault diagnosis result.

[0052] In the present invention, the final fault diagnosis result is determined by analyzing the infrared image and sound pattern image under the current test condition through the leakage diagnosis unit.

[0053] In a preferred embodiment, the fault diagnosis module is configured to perform the following operations:

[0054] Preprocess the infrared image to obtain the infrared image to be recognized;

[0055] Input the infrared image to be recognized into a pre-trained first recognition model to output the fault diagnosis result;

[0056] Among them, the first recognition model is trained through at least two groups of first sample sets. Each group of first sample sets includes a historical infrared image as the input and a fault diagnosis result of the historical infrared image as the output; the fault diagnosis result is the presence of a fault and no fault.

[0057] In a preferred embodiment, the fault diagnosis module is used to perform the following operations:

[0058] Divide the voiceprint image according to the structure of the heater to obtain a voiceprint image composed of several sub-images;

[0059] After preprocessing the voiceprint image composed of several sub-images, input it into a pre-trained second recognition model, and output a fault diagnosis result; the fault diagnosis result is the presence of a fault and no fault;

[0060] Among them, the second recognition model is trained through at least two groups of second sample sets. Each group of second sample sets includes a historical voiceprint image divided into several sub-images as the input and a fault diagnosis result of each sub-image as the output.

[0061] Specifically, the preprocessing involved in the fault diagnosis module includes operations such as image denoising, image quality evaluation, and image enhancement to ensure the provision of high-quality images with obvious features. In order to complete the training of the first recognition model and the second recognition model, an expanded sample algorithm can also be adopted for the historical infrared images and historical voiceprint images to achieve sample expansion and increase the number of samples.

[0062] In the present invention, when the fault diagnosis result of any recognition model is the presence of a fault, it is determined that the heater has a fault. For the second recognition model, the voiceprint image is further refined and divided. As described in the previous example, the voiceprint image is divided into five regions, and each region corresponds to a primary heating component, obtaining a voiceprint image composed of five sub-images, so as to further determine the specific heating component where the fault occurs based on the second recognition model.

[0063] Specifically, since there are obvious differences between the sound frequencies measured when there is a leak and the ambient noise frequencies measured when there is no leak, the voiceprint positioning component can initially judge whether a leak has occurred through the collected leak flag signal. If the leak flag signal is collected, it is considered that a leak has occurred. In order to determine whether it is the heater that has a leak, a further judgment can be made based on the second recognition model; if the leak flag signal is not collected, it is considered that no leak has occurred, but in order to confirm whether there is a false alarm, a further judgment can also be made based on the second recognition model.

[0064] In a preferred embodiment, the fault diagnosis module is further used to perform the following operations:

[0065] For the current test condition, when the fault diagnosis result is first obtained as a fault, this moment is confirmed as the fault moment;

[0066] Taking the fault moment as the central moment, obtain the infrared images collected within a preset time duration before and after the fault moment, and sort them in chronological order to obtain an infrared image sequence;

[0067] Extract the features of each infrared image in the infrared image sequence to obtain an infrared feature sequence composed of the infrared feature information of each infrared image in chronological order; among them, the infrared feature sequence includes a temperature gradient feature sequence, a shape feature sequence, and a texture feature sequence;

[0068] Determine the highest temperature value and its target position in the temperature gradient feature sequence, and calculate the temperature change rate at the target position per unit time;

[0069] Determine the shape of the high-temperature region where the target position is located in the shape feature sequence, and calculate the area diffusion rate per unit time; among them, the temperature range of the high-temperature region is the highest temperature value ± a preset temperature threshold;

[0070] Determine the leakage direction according to the shape of the high-temperature region that changes in chronological order in the shape feature sequence;

[0071] Determine the maximum gradient value and the minimum gradient value along the leakage direction in the texture feature sequence;

[0072] Determine the fault score according to the historical temperature range, the highest temperature value, the temperature change rate, the shape of the high-temperature region where the target position is located, the area diffusion rate, the maximum gradient value, and the minimum gradient value outside the heater for the current test condition;

[0073] Determine the fault level according to the fault score, and determine the fault case based on the fault level, so as to determine the fault solution through the fault case.

[0074] It should be noted that the infrared feature information includes temperature gradient features, shape features, and texture features.

[0075] In a preferred embodiment, the fault score is determined by the following formula:

[0076]

[0077] Among them, M is the fault score; T max is the highest temperature value; T a 、 T b are the starting temperature value and the ending temperature value of the historical temperature range respectively; VT is the temperature change rate; α is the material thermal inertia coefficient; A is the area of the high-temperature region, P is the perimeter of the high-temperature region; V S is the area diffusion rate; G max is the maximum gradient value; G min is the minimum gradient value; λ 1, λ 2, λ 3, λ 4 are all weight coefficients, and λ 1 + λ 2 + λ 3 + λ 4 = 1.

[0078] Specifically, taking the fault moment as the central moment t , infrared images collected within a preset time duration Δ t before and after the fault moment are obtained and sorted in chronological order to obtain an infrared image sequence; that is, the infrared image sequence is from t 1 = t -Δ t to t 2 = t +Δ t duration of infrared images. It should be noted that the highest temperature value in the temperature gradient feature sequence is the highest temperature value determined from all infrared images within the duration from t 1 to t 2. T max ; By statistically analyzing the temperature gradient features of each infrared image, the temperature change rate at the target position from t 2 to t 1 is obtained. According to the areas of the shapes of all infrared images in the high-temperature region in the shape feature sequence, and calculating the temperature change rate from t 2 to t 1, the area diffusion rate per unit time is obtained. The temperature range of the high-temperature region is the highest temperature value T max ± the preset temperature threshold ε . The leakage direction with the maximum diffusion rate in a certain direction; then the maximum gradient value and the minimum gradient value in the leakage direction are determined from the texture feature sequence within the duration from t 1 to t 2.

[0079] It should be noted that αIt is related to the specific heat capacity and thermal conductivity of the material used in the heater, and is used to adjust the sensitivity of the temperature change rate; among them, α The smaller it is, the more sensitive to rapid temperature rise, and it is suitable for metal materials; α The larger it is, it will suppress the temperature rise noise, and it is suitable for heat insulation materials. It is the shape complexity factor. The larger this value is, the more irregular the shape is, and the more likely it is to appear jet flow, so the degree of failure is higher.

[0080] In the present invention, the degree of temperature anomaly is quantified by the highest temperature value and the historical temperature range, excluding environmental interference; and the hyperbolic tangent function is used to constrain the influence of rapid temperature change, the influence of the shape is determined by the shape complexity factor, and by directly linearly coupling the temperature change rate and the shape complexity, the synergistic amplification effect of the two is considered; at the same time, the logarithmic function is used to suppress the excessive influence of the area diffusion rate, and the non-uniformity of the temperature field is characterized by the maximum gradient value and the minimum gradient value, so as to obtain the failure score. The higher the failure score, the more serious the failure and the higher the failure level.

[0081] In a preferred embodiment, based on the failure level, failure cases are determined to determine the failure solution through the failure cases, including:

[0082] Screen the failure cases included in this failure level under this test condition from the failure case library; the failure case library records the failure cases under each failure level under any test condition, and the failure cases include failure phenomena, failure diagnosis results and failure solutions;

[0083] For each failure case, the following is performed: compare the failure infrared image included in the failure phenomenon in this failure case with the infrared image under the current test condition to obtain the similarity;

[0084] Determine the failure solution included in the failure case corresponding to the minimum similarity as the current failure solution.

[0085] In a preferred embodiment, the failure diagnosis module further includes an output unit and an alarm unit; the output unit is used to display the failure diagnosis result; the alarm unit is used to give an alarm when the failure diagnosis result is that there is a failure. Specifically, the failure diagnosis module can be a device such as a server that implements the above functions.

[0086] Please refer to Figure 3 , the embodiment of the present invention provides a failure diagnosis method for a heater used in a wind tunnel test. This method is implemented based on any system embodiment in the specification. This method includes:

[0087] Step 300, when the heater is started, obtain the environmental parameters of the heater, and collect the infrared image and sound pattern image outside the heater in real time;

[0088] Step 302: Determine the current test condition according to the environmental parameters, and compare the environmental parameters, infrared images, and voiceprint images with the historical data under the current test condition to obtain the fault diagnosis result.

[0089] Regarding Step 302, determining the current test condition according to the environmental parameters, and comparing the environmental parameters, infrared images, and voiceprint images with the historical data under the current test condition to obtain the fault diagnosis result, including:

[0090] S1: When the environmental parameters are the same as the historical environmental parameters, determine that the current test condition of the environmental parameters is the test condition corresponding to the historical environmental parameters; where the environmental parameters include total temperature and total gas source pressure.

[0091] When there are no environmental parameters in the historical data, determine the first historical environmental parameter closest to the environmental parameters from the historical data, and judge whether the difference between the first historical environmental parameter and the environmental parameters does not exceed the preset threshold. If so, determine that the current test condition of the environmental parameters is the test condition corresponding to the first historical environmental parameter; otherwise, generate the current test condition and store it in the historical data.

[0092] S2: Under the current test condition, judge whether the temperature value is measured during the first wind tunnel test of the heater on the current day.

[0093] If not, use the historical temperature range without faults corresponding to the current test condition as the normal threshold. When the temperature value is within the normal threshold, determine that the fault diagnosis result is normal operation; when the temperature value is not within the normal threshold, determine that the fault diagnosis result is a temperature anomaly.

[0094] S3: After preprocessing the infrared image, obtain the infrared image to be recognized.

[0095] Input the infrared image to be recognized into the pre-trained first recognition model to output the fault diagnosis result.

[0096] Among them, the first recognition model is trained by at least two groups of first sample sets. Each group of first sample sets includes the historical infrared image as the input and the fault diagnosis result of the historical infrared image as the output; the fault diagnosis result is the presence of a fault and no fault.

[0097] S4: Divide the voiceprint image according to the structure of the heater to obtain a voiceprint image composed of several sub-images.

[0098] After preprocessing the voiceprint image composed of several sub-images, input it into the pre-trained second recognition model to output the fault diagnosis result; the fault diagnosis result is the presence of a fault and no fault.

[0099] Among them, the second recognition model is trained by at least two groups of second sample sets. Each group of second sample sets includes historical voiceprint images divided into several sub-images as inputs and fault diagnosis results for each sub-image as outputs.

[0100] S5. When the fault diagnosis result indicates the existence of a fault, determine the fault level based on the infrared image of the current test condition, and determine a fault case based on the fault level, so as to determine a fault solution through the fault case.

[0101] Regarding the content of the above method, since it is based on the same concept as the system embodiment of the present invention, the specific content can be referred to the description in the system embodiment of the present invention and will not be elaborated here.

[0102] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0103] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the aforementioned storage medium includes various media such as ROM, RAM, magnetic disk or optical disc that can store program codes.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fault diagnosis system for a heater for a wind tunnel test, characterized in that: include: Infrared imaging components, voiceprint positioning components, switches, heater control modules and fault diagnosis modules; The heater control module controls the heater to start and obtains the environmental parameters of the heater; the interior of the heater heats the air to obtain the airflow for the wind tunnel test, and the heater includes an electrode rod; the environmental parameters include total temperature, air source pressure and total pressure; The infrared imaging component and the voiceprint positioning component are installed around the heater in a non-contact manner to monitor the heater in all directions; when the heater is started, the infrared imaging component and the voiceprint positioning component are used to collect an infrared image and a voiceprint image of the outside of the heater in real time, respectively; The switch is used to send the environmental parameters, the infrared image and the voiceprint image to the fault diagnosis module; The fault diagnosis module is used to determine the current test condition according to the environmental parameters, and compare the environmental parameters, the infrared image and the voiceprint image with the historical data under the current test condition to obtain a fault diagnosis result; and when the fault diagnosis result is that a fault exists, the moment is confirmed as the fault moment, and a fault solution is determined according to the infrared images collected within a preset time period before and after the fault moment; The fault diagnosis module includes an operating condition determination unit and a temperature diagnosis unit; The working condition determination unit is used to compare the environmental parameters with the historical environmental parameters in the historical data to determine the current test working condition; wherein the historical data records the historical environmental parameters corresponding to each test working condition and the historical temperature range of each electrode rod during normal operation; The temperature diagnosis unit is used to determine the temperature value of each electrode rod according to the infrared image under the current test condition, compare the temperature value of each electrode rod with the historical temperature range under the current test condition, and determine that the fault diagnosis result is that there is a temperature abnormality when the temperature value is not within the historical temperature range; The operating condition determination unit is used to perform the following operations: When the environmental parameter is the same as the historical environmental parameter, determining the current test condition of the environmental parameter as the test condition corresponding to the historical environmental parameter; When the environmental parameter does not exist in the historical data, determine the first historical environmental parameter closest to the environmental parameter from the historical data, and judge whether the difference between the first historical environmental parameter and the environmental parameter does not exceed a preset threshold value; if so, determine that the current test condition of the environmental parameter is the test condition corresponding to the first historical environmental parameter; otherwise, generate the current test condition and store it in the historical data; The temperature diagnosis unit is used to perform the following operations: Under the current test condition, determining whether the temperature value is measured during the first wind tunnel test of the heater on that day; If not, the fault-free historical temperature interval corresponding to the current test condition is used as the normal threshold. When the temperature value is within the normal threshold, the fault diagnosis result is determined to be normal operation; when the temperature value is not within the normal threshold, the fault diagnosis result is determined to be a temperature anomaly.

2. The system according to claim 1, characterized in that The heater is obtained by connecting at least two stages of heating components in series; each stage of heating components uses at least two of the infrared imaging components.

3. The system according to claim 1, characterized in that The heater comprises a heating element, a heat-bearing layer, a heat-insulating layer, a pressure-bearing layer and a cooling structure; inside the heater, the heating element is used to heat the air, and the cooling structure is used to circulate cooling water to cool the electrode rod.

4. The system according to claim 1, characterized in that The fault diagnosis module includes an operating condition determination unit and a leakage diagnosis unit; The working condition determination unit is used to compare the environmental parameters with the historical environmental parameters in the historical data to determine the current test working condition; wherein the historical data records the historical environmental parameters corresponding to each test working condition and the historical infrared images and historical voiceprint images marked with the fault diagnosis results; The leakage diagnosis unit is used to compare the infrared image under the current test condition with the historical infrared image, and to compare the voiceprint image under the current test condition with the historical voiceprint image, to obtain the fault diagnosis result.

5. The system according to claim 1, characterized in that The fault diagnosis module is used to perform the following operations: After preprocessing the infrared image, an infrared image to be identified is obtained; Inputting the infrared image to be identified into a pre-trained first identification model, and outputting the fault diagnosis result; The first recognition model is obtained by training at least two groups of first sample sets, each of which includes historical infrared images as input and fault diagnosis results of the historical infrared images as output; the fault diagnosis results are the presence of fault and the absence of fault.

6. The system according to claim 1, characterized in that The fault diagnosis module is used to perform the following operations: Dividing the voiceprint image according to the structure of the heater to obtain a voiceprint image composed of a plurality of sub-images; After preprocessing the voiceprint image composed of several sub-images, a pre-trained second recognition model is input to output the fault diagnosis result; the fault diagnosis result is the presence of a fault or no fault; The second recognition model is obtained by training at least two groups of second sample sets, each of which includes a historical voiceprint image divided into several sub-images as input and a fault diagnosis result for each sub-image as output.

7. The system according to claim 1, characterized in that The fault diagnosis module is also used to perform the following operations: For the current test condition, when the fault diagnosis result is obtained for the first time that a fault exists, confirming the moment as the fault moment; Taking the fault moment as the central moment, acquiring infrared images collected within a preset time period before and after the fault moment, and sorting them in chronological order to obtain an infrared image sequence; Extracting features from each infrared image in the infrared image sequence to obtain an infrared feature sequence composed of infrared feature information of each infrared image in chronological order; wherein the infrared feature sequence includes a temperature gradient feature sequence, a shape feature sequence, and a texture feature sequence; Determine the highest temperature value and its target position from the temperature gradient characteristic sequence, and calculate the temperature change rate at the target position per unit time; Determine the shape of the high temperature area where the target position is located from the shape feature sequence, and calculate the area diffusion rate per unit time; wherein the temperature range of the high temperature area is the maximum temperature value ± a preset temperature threshold; determining the leakage direction according to the shape of the high temperature area that changes with time in the shape feature sequence; Determine the maximum gradient value and the minimum gradient value along the leakage direction from the texture feature sequence; Determine a fault score according to the historical temperature range of the heater exterior, the highest temperature value, the temperature change rate, the shape of the high temperature area where the target position is located, the area diffusion rate, the maximum gradient value, and the minimum gradient value of the current test condition; A fault level is determined according to the fault score, and a fault case is determined based on the fault level, so as to determine a fault solution through the fault case.

8. The system according to any one of claims 1 to 7, characterized in that: The resolution of the infrared imaging component is not less than 256*192, and the frame rate is not less than 25fps; the frequency band of the voiceprint positioning component is 5kHz~72kHz.

9. The system according to any one of claims 1 to 7, characterized in that: The fault diagnosis result includes the presence of a fault and the presence of no fault; the presence of a fault includes: the presence of abnormal temperature, and the presence of gas or liquid leakage.

10. The system according to any one of claims 1 to 7, characterized in that: The fault diagnosis module also includes an output unit and an alarm unit; the output unit is used to display the fault diagnosis result; the alarm unit is used to issue an alarm when the fault diagnosis result indicates that a fault exists.

11. A fault diagnosis method for a heater for wind tunnel testing based on the system according to any one of claims 1 to 10, characterized in that: include: When the heater is started, the environmental parameters of the heater are obtained, and an infrared image and a voiceprint image of the outside of the heater are collected in real time; The current test condition is determined according to the environmental parameters, and the environmental parameters, the infrared image and the voiceprint image are compared with the historical data under the current test condition to obtain a fault diagnosis result.

Images