Detection, maintenance method and system of carbon fiber composite material defects in hydrogen storage bottles

Abstract

The present invention discloses a method and system for detecting and maintaining defects of carbon fiber composite materials in hydrogen storage bottles, including: obtaining prediction results of defects of carbon fiber composite materials in hydrogen storage bottles according to new ultrasonic detection signals of hydrogen storage bottles and trained MRNN models; when the defect prediction result is that the performance of fiber or epoxy resin is degraded, the Gamma random process is introduced into the trained MRNN model through maximum likelihood estimation to improve the health index of carbon fiber composite materials; when the defect prediction result is fiber breakage or delamination defect, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to obtain the health index of carbon fiber composite materials; according to the health index of carbon fiber composite materials, a maintenance strategy for hydrogen storage bottles is given. The defect type and health index of carbon fiber composite materials are determined, and a specific maintenance strategy is given.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrogen storage bottle detection and maintenance, and in particular to a method and system for detecting and maintaining defects of carbon fiber composite materials in hydrogen storage bottles. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] As an indispensable new engineering material in modern industry, carbon fiber reinforced polymer (CFRP) is widely used in aerospace, new energy equipment, automobiles, sporting goods, transportation, engineering equipment, medical equipment and construction. Carbon fiber reinforced polymer has the characteristics of high strength, light weight, high toughness, corrosion resistance, fatigue resistance and high temperature resistance.

[0004] Carbon fiber reinforced polymers are wrapped around the outer periphery of the hydrogen storage bottle liner (such as aluminum alloy or polymer) in a spiral or hoop method to increase the structural strength of the liner, which can meet extremely high safety and lightweight requirements. The weight of hydrogen storage bottles using carbon fiber reinforced polymers is about 70% less than that of steel cylinders. However, during the use of hydrogen storage bottles, CFRP may have defects such as fiber or epoxy resin performance degradation, or even fiber breakage and delamination, which will seriously affect the performance of hydrogen storage bottles. Timely detection of the health status of carbon fiber reinforced polymers in hydrogen storage bottles and providing maintenance suggestions for hydrogen storage bottles are of great significance to the safety of hydrogen storage bottles. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a method and system for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles. Based on the detection of defects in carbon fiber composite materials in hydrogen storage bottles, the health index of the carbon fiber composite materials is further determined, and a specific maintenance strategy is given according to the health index of the carbon fiber composite materials.

[0006] To achieve the above object, the present invention adopts the following technical solution:

[0007] First, a method for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles is proposed, including:

[0008] Obtain new ultrasonic detection signals of hydrogen storage bottles, and obtain prediction results of carbon fiber composite material defects in hydrogen storage bottles based on the trained MRNN model;

[0009] When the defect prediction result is the degradation of fiber or epoxy resin performance, the Gamma random process is introduced into the trained MRNN model through maximum likelihood estimation to improve it and obtain the health index of carbon fiber composite material.

[0010] When the defect prediction result is fiber breakage or delamination defect, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to obtain the health index of carbon fiber composite materials.

[0011] According to the health index of carbon fiber composite materials, the maintenance strategy of hydrogen storage bottles is given.

[0012] Secondly, a detection and maintenance system for defects in carbon fiber composite materials in hydrogen storage bottles is proposed, including:

[0013] The prediction result acquisition module is used to collect new ultrasonic detection signals of the hydrogen storage bottle and obtain the prediction results of carbon fiber composite material defects in the hydrogen storage bottle according to the trained MRNN model of the hydrogen storage bottle;

[0014] The health index acquisition module is used to introduce the Gamma random process into the trained MRNN model through maximum likelihood estimation to improve the health index of the carbon fiber composite material when the defect prediction result is the degradation of fiber or epoxy resin performance; when the defect prediction result is fiber breakage or delamination, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to obtain the health index of the carbon fiber composite material;

[0015] The maintenance strategy acquisition module gives the corresponding maintenance strategy based on the defect prediction results and health index of carbon fiber composite materials.

[0016] In a third aspect, an electronic device is proposed, comprising a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein when the computer instructions are run by the processor, the steps described in the method for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles are completed.

[0017] In a fourth aspect, a computer-readable storage medium is proposed for storing computer instructions. When the computer instructions are executed by a processor, the steps described in the method for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles are completed.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] 1. Based on the detection of defects in carbon fiber composite materials in hydrogen storage bottles, the present invention adopts the Gamma random process or proportional risk regression model to improve the MRNN model according to the predicted defect type, predicts the health index of carbon fiber composite materials, and gives specific maintenance strategies according to different health indexes, thereby realizing the monitoring and maintenance of long-term performance attenuation of carbon fiber composite materials.

[0020] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0022] Figure 1 This is a flow chart of the method disclosed in Example 1. DETAILED DESCRIPTION

[0023] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0024] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0025] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0026] Example 1

[0027] In this embodiment, a method for detecting and maintaining defects in carbon fiber composite materials in a hydrogen storage bottle is disclosed, comprising:

[0028] Obtaining ultrasonic detection signals of hydrogen storage bottles;

[0029] The MRNN model is trained according to the ultrasonic detection signal of the hydrogen storage bottle to obtain a training set of the defect detection results of the carbon fiber composite material in the hydrogen storage bottle;

[0030] Obtain new ultrasonic detection signals of hydrogen storage bottles, and obtain prediction results of carbon fiber composite material defects in hydrogen storage bottles based on the trained MRNN model;

[0031] When the defect prediction result is the degradation of fiber or epoxy resin performance, the Gamma random process is introduced into the trained MRNN model through maximum likelihood estimation (MLE) to improve it and obtain the health indicator (HI) of carbon fiber composite materials.

[0032] When the defect prediction result is fiber breakage or delamination, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to improve it and obtain the health index of carbon fiber composite materials.

[0033] According to the health index of carbon fiber composite materials, the maintenance strategy of hydrogen storage bottles is given.

[0034] The method for detecting and maintaining defects of carbon fiber composite materials in the hydrogen storage bottle disclosed in this embodiment is described in detail.

[0035] like Figure 1 As shown, the detection and maintenance method of carbon fiber composite material defects in hydrogen storage bottles includes:

[0036] S1: Obtain the ultrasonic detection signal of the hydrogen storage bottle.

[0037] The ultrasonic detection signal of the hydrogen storage bottle and the defect type of the carbon fiber composite material in the hydrogen storage bottle corresponding to the ultrasonic detection signal are obtained to form a training sample. The mapping relationship between the ultrasonic detection signal and the corresponding defect type is established using the meta-recurrent neural network (MRNN) model. After the training is completed, the trained MRNN is obtained to obtain a training set of the defect detection results of the carbon fiber composite material in the hydrogen storage bottle.

[0038] The MRNN model takes the ultrasonic detection signal of the hydrogen storage bottle as input and the defect detection result of the carbon fiber composite material of the hydrogen storage bottle as output.

[0039] Defect detection results obtained include: early defects such as fiber or epoxy resin performance degradation, fiber breakage or delamination and other defects.

[0040] MRNN is a time-dependent neural network method that can monitor the health of carbon fiber composites over a long period of time. However, existing time-independent neural network methods find it difficult to monitor the long-term performance degradation of CFRP.

[0041] S2: Obtain new ultrasonic detection signals from the hydrogen storage bottle, and obtain the prediction results of carbon fiber composite material defects in the hydrogen storage bottle based on the trained MRNN model; when the defect prediction result is the degradation of fiber or epoxy resin performance, the Gamma random process is introduced into the trained MRNN model through maximum likelihood estimation (MLE) to improve the health index (HI) of the carbon fiber composite material; when the defect prediction result is fiber breakage or delamination defects, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to improve the health index of the carbon fiber composite material.

[0042] In order to accurately predict the health status of carbon fiber composites in hydrogen storage bottles, when identifying the new ultrasonic detection signals of hydrogen storage bottles, the trained MRNN model is first used to identify them to obtain the defect prediction results of carbon fiber composites in hydrogen storage bottles; based on the defect prediction results, the Gamma random process or proportional risk regression model is used to improve the MRNN model to more accurately predict the health index of carbon fiber composites.

[0043] For early defects such as degradation of fiber or epoxy resin performance, the Gamma random process is introduced into the trained MRNN through maximum likelihood estimation. The critical threshold of fiber or epoxy resin performance degradation is defined in the Gamma random process, and the health index of the carbon fiber composite material is calculated based on the critical threshold. Introducing the Gamma random process into the trained MRNN through maximum likelihood estimation can better predict model degradation and more accurately predict the health index of the carbon fiber composite material in the hydrogen storage bottle. Specifically: the Gamma random process is introduced into the trained MRNN model through maximum likelihood estimation to obtain the first health index prediction model, and the new ultrasonic detection signal of the hydrogen storage bottle is input into the first health index prediction model to obtain the health index of the carbon fiber composite material.

[0044] For defects such as fiber breakage or delamination, considering the effects of aging and covariates, the proportional hazard regression (PH) model is introduced into the trained MRNN through maximum likelihood estimation to predict model degradation and predict the health index of carbon fiber composite materials in hydrogen storage bottles. The proportional hazard regression model includes a baseline risk function that describes the effects of aging and a link function that considers the influencing factors of covariates, where covariates include temperature and workload. Generally, temperature and workload have a greater impact on the detection of defects such as fiber breakage and delamination, so the proportional hazard regression model is introduced to improve the accuracy of health index prediction. Specifically, the proportional hazard regression model is introduced into the trained MRNN model through maximum likelihood estimation to obtain a second health index prediction model, and the new ultrasonic detection signal of the hydrogen storage bottle is input into the second health index prediction model to obtain the health index of the carbon fiber composite material.

[0045] Maximum likelihood estimation is to use known sample result information to infer the model parameter values ​​that are most likely (with the highest probability) to cause these sample results.

[0046] In specific implementation, the range of the health index can be set to [0,1]. A health index of 0 represents that the carbon fiber composite material is in a healthy state. When the health index is between 0 and 1, it can indicate that the hydrogen storage bottle is in a state of fiber or epoxy resin performance degradation, and the higher the health index, the more serious the defects of the carbon fiber composite material, such as delamination. When the health index is 1, it means that the carbon fiber composite material contains the most serious defects.

[0047] S3: Based on the health index of carbon fiber composite materials, a maintenance strategy for hydrogen storage bottles is given. The maintenance strategy includes no maintenance required, preventive general repairs, minor repairs, major repairs and replacement.

[0048] The health status of the hydrogen storage bottle is represented by (k, HI, S), where k represents the service life of the hydrogen storage bottle, HI represents the health index of the carbon fiber composite material of the hydrogen storage bottle, and S∈{1, 2, 3, 4, 5} represents the repair strategy. S=1 means no maintenance is required, S=2 means preventive general repair is required, S=3 means minor repair is required, S=4 means major repair is required, and S=5 means replacement is required. When the health index is 0 and the hydrogen storage bottle is in a healthy state, the hydrogen storage bottle does not require maintenance; when the health index is between 0 and 1, and the hydrogen storage bottle is in the state of fiber or epoxy resin degradation, the hydrogen storage bottle requires preventive general repair; when the health index is between 0 and 1, and the hydrogen storage bottle is in the early stage of fiber breakage or delamination, minor repair is required, and when the hydrogen storage bottle is in a period of more serious fiber breakage or delamination, major repair is required; when the health index is 1, the hydrogen storage bottle has the most serious defects and needs to be replaced.

[0049] The state set of the constructed hydrogen storage bottle is expressed as S=S_1∪S_2∪S_3∪S_4∪S_5, S_1={(k,HI,1)|k=0,...,k-1}, S_2={(k,HI,2)|k=0,...,k-1}, S_3={(k,HI,3)|k=0,...,k-1}, S_4={(k,HI,4)|k=0,...,k-1} and S_5={(k,HI,5)|k=0,...,k-1}.

[0050] The detection and maintenance method disclosed in this embodiment, based on the detection of carbon fiber composite materials in hydrogen storage bottles through the MRNN model, uses the Gamma random process or proportional risk regression model to improve the MRNN model according to the predicted defect type, so as to obtain a more accurate health index of the carbon fiber composite materials, and give a specific maintenance strategy based on the predicted health index, thereby realizing the monitoring and maintenance of the long-term performance degradation of the carbon fiber composite materials.

[0051] Example 2

[0052] In this embodiment, a detection and maintenance system for defects in carbon fiber composite materials in hydrogen storage bottles is disclosed, including:

[0053] The prediction result acquisition module is used to collect new ultrasonic detection signals of the hydrogen storage bottle and obtain the prediction results of carbon fiber composite material defects in the hydrogen storage bottle according to the trained MRNN model of the hydrogen storage bottle;

[0054] The health index acquisition module is used to introduce the Gamma random process into the trained MRNN model through maximum likelihood estimation to improve the health index of the carbon fiber composite material when the defect prediction result is the degradation of fiber or epoxy resin performance; when the defect prediction result is fiber breakage or delamination, the proportional risk regression model is introduced into the trained MRNN model through maximum likelihood estimation to obtain the health index of the carbon fiber composite material;

[0055] The maintenance strategy acquisition module gives the corresponding maintenance strategy according to the defect detection results and health index of carbon fiber composite materials.

[0056] Example 3

[0057] In this embodiment, an electronic device is disclosed, including a memory and a processor, and computer instructions stored in the memory and running on the processor. When the computer instructions are run by the processor, the steps described in the method for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles disclosed in Example 1 are completed.

[0058] Example 4

[0059] In this embodiment, a computer-readable storage medium is disclosed for storing computer instructions. When the computer instructions are executed by a processor, the steps described in the method for detecting and maintaining defects in carbon fiber composite materials in hydrogen storage bottles disclosed in Example 1 are completed.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. Detection and maintenance methods of carbon fiber composite material defects in hydrogen storage bottles, It is characterized in that include: Obtain new ultrasonic detection signals of hydrogen storage bottles, and obtain prediction results of carbon fiber composite material defects in hydrogen storage bottles based on the trained Meta-recurrent neural network model; When the defect prediction result is the degradation of fiber or epoxy resin performance, the Gamma random process is introduced into the trained Meta-recurrent neural network model through maximum likelihood estimation to improve the health index of carbon fiber composite materials. When the defect prediction result is fiber breakage or delamination defect, the proportional risk regression model is introduced into the trained Meta-recurrent neural network model through maximum likelihood estimation to obtain the health index of carbon fiber composite materials. According to the health index of carbon fiber composite materials, the maintenance strategy of hydrogen storage bottles is given.

2. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 1, It is characterized in that The critical threshold of fiber or epoxy resin performance degradation is defined in the Gamma random process, and the health index of carbon fiber composites is calculated based on this critical threshold.

3. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 1, It is characterized in that The proportional hazards regression model included a baseline hazard function to describe the effects of aging and a link function to account for the effects of covariates, including temperature and workload.

4. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 1, It is characterized in that Obtain ultrasonic detection signals of the hydrogen storage bottle; and train a Meta-recurrent neural network model based on the ultrasonic detection signals of the hydrogen storage bottle.

5. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 1, It is characterized in that The range of health index is [0,1].

6. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 5, It is characterized in that When the health index is 0, the hydrogen storage bottle is in a healthy state. The higher the health index, the more serious the defects of the carbon fiber composite material.

7. The method for detecting and maintaining defects of carbon fiber composite materials in a hydrogen storage bottle according to claim 1, It is characterized in that Maintenance strategies include: no maintenance required, preventive general repair required, minor repair required, major repair required, and replacement required.

8. Detection and maintenance system for defects in carbon fiber composite materials in hydrogen storage bottles, It is characterized in that include: The prediction result acquisition module is used to collect new ultrasonic detection signals of the hydrogen storage bottle and obtain the prediction results of carbon fiber composite material defects in the hydrogen storage bottle according to the trained Meta-recurrent neural network model of the hydrogen storage bottle; The health index acquisition module is used to introduce the Gamma random process into the trained Meta-recurrent neural network model through maximum likelihood estimation to improve the health index of the carbon fiber composite material when the defect prediction result is the degradation of fiber or epoxy resin performance; when the defect prediction result is fiber breakage or delamination, the proportional risk regression model is introduced into the trained Meta-recurrent neural network model through maximum likelihood estimation to obtain the health index of the carbon fiber composite material; The maintenance strategy acquisition module gives the corresponding maintenance strategy according to the defect detection results and health index of carbon fiber composite materials.

9. An electronic device, It is characterized in that The invention comprises a memory and a processor and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of the method for detecting and maintaining defects of carbon fiber composite materials in hydrogen storage bottles as described in any one of claims 1 to 7 are completed.

10. A computer-readable storage medium, It is characterized in that Used to store computer instructions, which, when executed by a processor, complete the steps of the method for detecting and maintaining defects in carbon fiber composite materials in a hydrogen storage bottle as described in any one of claims 1 to 7.

Images