Eccentric butterfly valve with self-locking structure and performance detection method

Abstract

This invention belongs to the field of eccentric butterfly valve technology and discloses an eccentric butterfly valve with a self-locking structure and a performance testing method, including a valve body, an eccentric valve plate, a valve shaft, an aluminum seat, an actuator, and a linkage mechanism; the eccentric valve plate is fixedly installed on the valve shaft, and the eccentric valve plate and the valve shaft are installed eccentrically; the aluminum seat is fixed on the outside of the valve body, and the actuator is fixedly installed on the aluminum seat. The output shaft of the actuator is connected to the valve shaft through a linkage mechanism; the linkage mechanism includes a driving crank, a linkage body, and a driven crank. The driving crank is fixed to the output shaft of the actuator, and the driven crank is fixed to the extended end of the valve shaft. The two ends of the linkage body are respectively hinged to the driving crank and the driven crank through pins; when the eccentric butterfly valve is in the fully closed state, the hinge center of the linkage body and the driven crank, the rotation center of the driven crank and the valve shaft, and the hinge center of the linkage body and the driving crank are all on the same straight line. The linkage mechanism moves to the mechanical dead point position, forming a reverse self-locking structure.

Description

Technical Field

[0001] This manual relates to the field of eccentric butterfly valve technology, and in particular to an eccentric butterfly valve with a self-locking structure and a method for testing its performance. Background Technology

[0002] In industrial pipeline control systems such as petrochemicals, municipal pipelines, and power energy, butterfly valves are core components for controlling the flow and regulating the flow of media. Traditional centrally symmetrical butterfly valves have a structure in which the valve shaft passes through the center of the valve plate and the valve plate is centrally located. This structure has defects such as large opening and closing torque, easy fatigue failure of the valve shaft, and lack of mechanical self-locking when closed. Under media pressure fluctuations or emergency conditions, the valve plate is prone to accidental opening, leading to leakage and safety accidents.

[0003] Although existing eccentric butterfly valves reduce some torque through their eccentric structure, they generally suffer from the following problems:

[0004] 1. There is no dedicated linkage self-locking mechanism, or the self-locking structure has inaccurate dead point positioning and poor reliability, and cannot resist the reverse pressure of the medium;

[0005] 2. The sealing pair relies on static compression, resulting in high friction and easy wear of the sealing surface during the opening and closing process.

[0006] In terms of performance testing, existing technologies (such as CN120927225B) only perform optical testing on the sealing pairs of ordinary eccentric butterfly valves, which has obvious shortcomings:

[0007] 1. The contact pressure is calculated using only linear mapping, without considering the elastic-plastic deformation of the metal hard seal, resulting in low accuracy;

[0008] 2. It completely neglects the testing of self-locking structures, and fails to conduct a coupled correlation analysis between sealing performance and self-locking performance.

[0009] In summary, existing butterfly valve structures and testing methods cannot meet the demands of high safety, high precision, and intelligent industrial applications. There is an urgent need for a technical solution that features reliable self-locking, low torque operation, and complete testing and life prediction capabilities. Summary of the Invention

[0010] This invention addresses the problems existing in the prior art by proposing an eccentric butterfly valve with a self-locking structure and its performance testing method. The eccentric arrangement reduces the opening and closing torque, and the linkage mechanism achieves reverse self-locking when the valve reaches its mechanical dead point upon closing. The actuator and valve shaft are misaligned to reserve space for sensor installation. Based on an elastoplastic contact mechanics model, multimodal spatiotemporal feature fusion, a sealing-self-locking coupling correlation model, and a timing prediction model, the invention achieves dynamic contact pressure calculation of the sealing pair, micro-defect identification, self-locking-sealing coupling analysis, and life prediction. It solves the problems of traditional butterfly valves having high torque, lacking reliable self-locking, and existing testing methods failing to cover self-locking performance and sealing-self-locking correlation analysis, significantly improving valve operation safety, testing accuracy, and intelligence.

[0011] To achieve the above objectives, this application provides the following technical solution:

[0012] Firstly, an eccentric butterfly valve with a self-locking structure includes a valve body, an eccentric valve plate, a valve shaft, an aluminum seat, an actuator, and a linkage mechanism. The valve shaft is rotatably mounted on the valve body, and the eccentric valve plate is fixedly mounted on the valve shaft and located within the flow channel of the valve body, with the eccentric valve plate and valve shaft being eccentrically mounted. The aluminum seat is fixed to the outside of the valve body, and the actuator is fixedly mounted on the aluminum seat. The output shaft of the actuator is connected to the valve shaft via the linkage mechanism. The linkage mechanism includes a drive crank and a connecting rod. The connecting rod body and the driven crank are respectively connected to the driven crank and the valve shaft extension by pins at both ends. When the eccentric butterfly valve is in the fully closed state, the hinge center of the connecting rod body and the driven crank, the rotation center of the driven crank and the valve shaft, and the hinge center of the connecting rod body and the driven crank are on the same straight line. The connecting rod mechanism moves to the mechanical dead point position, forming a reverse self-locking structure.

[0013] Optionally, a valve seat is fixedly installed in the flow channel of the valve body. When the eccentric butterfly valve is in the fully closed state, the sealing surface of the eccentric valve plate is pressed tightly against the sealing surface of the valve seat to form a metal hard seal.

[0014] Optionally, the rotation center of the eccentric valve plate is eccentrically set with respect to its own geometric center, and the sealing surface of the eccentric valve plate and the sealing surface of the valve seat move along an eccentric arc trajectory during the opening and closing process.

[0015] Secondly, a method for testing the performance of an eccentric butterfly valve, characterized in that, for testing the eccentric butterfly valve with a self-locking structure described in the first aspect, the method includes the following steps:

[0016] S1. Simultaneously collect the test data of the sealing pair during the entire opening and closing cycle of the eccentric butterfly valve under test, as well as the self-locking performance parameters of the self-locking structure performance test.

[0017] S2. Based on the elastoplastic contact mechanics model of the metal hard seal pair, calculate the dynamic contact pressure distribution of the seal pair according to the detection data of the seal pair;

[0018] S3. Multimodal spatiotemporal feature fusion is performed on the static structural characteristics and dynamic response characteristics of the sealing pair during the opening and closing process to identify sealing pair defects and generate a leakage early warning area map.

[0019] S4. Construct a performance coupling correlation model between the sealing pair and the connecting rod self-locking structure, determine the influence relationship between the performance changes of the self-locking structure and the sealing performance, and generate coupling anomaly correlation results.

[0020] S5. Combine the leakage early warning area map with the coupling anomaly correlation results to generate a comprehensive test result of the butterfly valve sealing performance.

[0021] Optional, performance testing of the self-locking structure includes:

[0022] T1. Collect motion timing data and torque timing data of the linkage mechanism during the entire opening and closing cycle of the eccentric butterfly valve to be tested;

[0023] T2. Based on the motion timing data and torque timing data, generate the transmission performance parameters and dead point position parameters of the linkage mechanism;

[0024] T3. Construct a self-locking mechanical model based on the transmission performance parameters, and calculate the self-locking critical threshold and the actual self-locking holding torque;

[0025] T4. Generate the performance test results of the self-locking structure based on the self-locking critical threshold, the actual self-locking holding torque, and the dead point position parameters.

[0026] Optionally, step T1 includes:

[0027] T11. The high-speed vision acquisition unit continuously acquires motion trajectory time-series images of the active crank, connecting rod body, driven crank and each hinge point of the linkage mechanism within a preset opening and closing time period, as the motion time-series data.

[0028] T12. The torque timing data of the actuator output shaft and the load torque timing data of the valve shaft are collected by the torque sensing unit, and the valve opening and closing angle timing data are collected simultaneously, which together serve as the torque timing data.

[0029] Optionally, step T2 includes:

[0030] T21. Perform subpixel registration on the motion trajectory time image, extract the spatial displacement vector sequence of each hinge point in the entire opening and closing cycle, determine the actual spatial position of the hinge point between the driven crank and the connecting rod body in the fully closed state of the valve, and calculate the actual offset of the dead point position as the dead point position parameter.

[0031] T22. Based on the output torque timing data, load torque timing data, and opening / closing angle timing data, calculate the transmission ratio, torque amplification factor, and transmission efficiency of the linkage mechanism under different opening / closing angles, and use them as the transmission performance parameters.

[0032] Optionally, step T3 includes:

[0033] T31. Based on the transmission performance parameters, and combined with the reverse thrust load of the medium on the eccentric valve plate in the fully closed state of the valve, a self-locking mechanical model of the linkage mechanism is constructed.

[0034] T32. Calculate the minimum self-locking torque to prevent the linkage mechanism from rotating in the opposite direction using the self-locking mechanical model, and use it as the self-locking critical threshold.

[0035] T33. Based on the torque timing data and transmission performance parameters of the valve in the fully closed state, calculate the maximum reverse resistance torque of the linkage mechanism at the dead point position, which is used as the actual self-locking holding torque.

[0036] Optionally, step S2 includes:

[0037] S21. Based on the light transmission intensity distribution image and the structured light reflection interference image sequence, obtain the spatial distribution of light transmission intensity of the sealing pair and the evolution sequence of the micromorphology of the sealing surface, and establish the spatial reference coordinate system of the sealing pair.

[0038] S22. Construct an elastoplastic contact mechanics model for a metal hard seal pair and clarify the mapping relationship between the mechanical properties of the sealing surface material and contact deformation and contact pressure.

[0039] S23. Based on the elastoplastic contact mechanics model, the spatial distribution of light transmission intensity and the evolution sequence of micromorphology are calculated through a pre-trained contact pressure inversion convolutional neural network, and the dynamic contact pressure distribution matrix sequence of the sealing pair during the entire opening and closing cycle is output.

[0040] Optionally, step S4 includes:

[0041] S41. Using each sealing sub-region of the sealing pair as the first type of node, each transmission component of the linkage mechanism as the second type of node, and the physical relationship between sealing performance and self-locking performance as the edge, construct a heterogeneous graph network of the sealing pair-linkage self-locking structure.

[0042] S42. By coupling the correlation model with a pre-trained graph attention network, the dynamic correlation weights of each node in the heterogeneous graph network are learned, and the contribution of the performance change of the self-locking structure to the sealing pair defect is quantified.

[0043] S43. Based on the correlation weight and contribution, determine the correlation mapping relationship between sealing defects and self-locking structure anomalies, and generate a coupling anomaly correlation map as the coupling anomaly correlation result.

[0044] The beneficial effects of this invention are as follows:

[0045] 1. This application significantly reduces opening and closing torque through eccentric design and dead-point self-locking structure, achieving highly reliable reverse self-locking. At the same time, it reserves space for sensor installation, adapts to the intelligent transformation needs of industrial valves, and forms a complete technical solution with the detection method, comprehensively improving the operational safety and intelligent maintenance level of valves.

[0046] 2. This application replaces the linear mapping model of the existing technology with a metal hard seal elastoplastic contact constitutive model and a CNN inversion network, which solves the calculation error problem caused by the elastoplastic deformation of the metal sealing surface. The contact pressure calculation error is reduced by more than 60%, and the dynamic contact pressure evolution process of the entire opening and closing cycle can be accurately restored.

[0047] 3. This application uses a spatiotemporal fusion Transformer model to fuse multimodal data, and simultaneously extracts static structural defect features and dynamic response anomaly features, which solves the problem of high false negative rate in static feature identification in the prior art. The identification rate of early wear and micro-deformation defects larger than 5μm is increased to more than 95%, which can realize ultra-early leakage risk warning.

[0048] 4. This application targets eccentric butterfly valves with self-locking structures. It constructs a coupled correlation model of the sealing pair and the self-locking structure of the connecting rod through graph attention network. This model can accurately locate the root cause of the self-locking structure abnormality that leads to the deterioration of sealing performance. It solves the core pain point of existing technologies that cannot correlate the transmission structure and sealing performance, and realizes a comprehensive evaluation of the valve's core performance. This is fundamentally different from existing detection technologies.

[0049] 5. This application uses a bidirectional LSTM time-series prediction model to predict the sealing performance degradation trend and remaining service life based on multi-cycle dynamic data, filling the gap in the existing technology that lacks predictive evaluation capabilities. It can provide core data support for the preventive operation and maintenance of industrial pipeline valves and significantly reduce the risk of sudden leakage under operating conditions. Attached Figure Description

[0050] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The same numbers in the drawings denote the same structures or steps.

[0051] Figure 1 This is a schematic diagram of an eccentric butterfly valve with a self-locking structure according to Embodiment 1 of this application.

[0052] Figure 2 This is a schematic diagram of the linkage mechanism in the eccentric butterfly valve with a self-locking structure according to Embodiment 1 of this application.

[0053] Figure 3 This is a schematic diagram of the eccentric butterfly valve performance testing method according to Embodiment 2 of this application.

[0054] Figure 4 This is a schematic diagram of the performance testing of the self-locking structure in Embodiment 2 of this application.

[0055] Figure label:

[0056] 1. Valve body; 2. Eccentric valve plate; 3. Valve shaft; 4. Aluminum seat; 5. Actuator; 6. Valve seat; 7. Driving crank; 8. Connecting rod body; 9. Driven crank. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection of this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] Example 1:

[0059] like Figure 1-2 As shown, an eccentric butterfly valve with a self-locking structure includes a valve body 1, an eccentric valve plate 2, a valve shaft 3, an aluminum seat 4, an actuator 5, and a linkage mechanism. The valve body 1 is a standard valve body 1 for industrial pipelines, with a through-flow medium channel inside. An annular metal valve seat 6 is fixedly installed in the middle of the channel, and the sealing surface of the valve seat 6 is a conical sealing surface. The valve shaft 3 is rotatably mounted on the side wall of the valve body 1 via a bearing assembly. One end of the valve shaft 3 extends into the flow channel of the valve body 1, and the other end extends outward from the valve body 1 as the transmission end.

[0060] The eccentric valve plate 2 is a circular butterfly plate, fixedly installed at the extension end of the valve shaft 3 and located within the flow channel of the valve body 1, corresponding to and cooperating with the valve seat 6. The eccentric valve plate 2 and the valve shaft 3 are eccentrically mounted, with the rotation center of the eccentric valve plate 2 (i.e., the axis of the valve shaft 3) eccentrically positioned with respect to its own geometric center. Simultaneously, the sealing surface of the eccentric valve plate 2 is eccentrically positioned with respect to the axis of the valve shaft 3, forming a double eccentric structure. This structure allows the sealing surface of the eccentric valve plate 2 to move along an eccentric arc trajectory during opening and closing, completely disengaging from the valve seat 6 within 10° of valve opening, significantly reducing frictional torque during the opening and closing process.

[0061] The aluminum seat 4 is fixed to the outer wall of the valve body 1 by bolts, located on one side of the extended end of the valve shaft 3. The actuator 5 is fixedly installed on the aluminum seat 4. In this embodiment, the actuator 5 is existing technology. The actuator 5 includes a reducer and a drive motor. The drive motor is a servo motor, and its output end is connected to the input end of the reducer. The output shaft of the reducer is the output shaft of the actuator 5. The reducer is fixed to the outside of the valve body 1 by the aluminum seat 4. The output shaft of the actuator 5 and the valve shaft 3 are arranged in a staggered parallel configuration. Their axes are parallel to each other and not collinear. A reserved sensor installation space is formed between the output shaft of the actuator 5 and the valve shaft 3, which can be used to install monitoring elements such as valve opening sensors, torque sensors, and vibration sensors, adapting to the needs of intelligent valve transformation.

[0062] The output shaft of actuator 5 is connected to valve shaft 3 via a linkage mechanism. The linkage mechanism includes a driving crank 7, a connecting rod body 8, and a driven crank 9. One end of the driving crank 7 is fixed to the end of the output shaft of actuator 5 by a flat key, and one end of the driven crank 9 is fixed to the extended end of valve shaft 3 by a flat key. The two ends of the connecting rod body 8 are respectively hinged to the other ends of the driving crank 7 and the driven crank 9 by hinge pins, forming a crank-connecting rod transmission structure.

[0063] When the eccentric butterfly valve is fully closed (valve closing angle 90°), the hinge center of the connecting rod body 8 and the driven crank 9, the rotation center of the driven crank 9 and the valve shaft 3, and the hinge center of the connecting rod body 8 and the driving crank 7 are all on the same straight line. At this time, the connecting rod mechanism moves to the mechanical dead point position. When the medium pressure acts on the eccentric valve plate 2, generating a reverse opening torque on the valve shaft 3, this torque is transmitted to the connecting rod body 8 through the driven crank 9. Since the connecting rod mechanism is in the dead point position, the reverse torque cannot drive the driving crank 7 and the output shaft of the actuator 5 to rotate, thus forming a reverse self-locking structure, effectively preventing the valve plate from opening in reverse under the impact of medium pressure. Through the eccentric design and dead point self-locking structure, the opening and closing torque is significantly reduced, achieving highly reliable reverse self-locking. At the same time, space is reserved for sensor installation, adapting to the intelligent transformation needs of industrial valves, forming a complete technical solution with the detection method, and comprehensively improving the operational safety and intelligent maintenance level of the valve.

[0064] The action process is as follows:

[0065] Actuator 5 startup phase: After the drive motor receives the valve opening or closing signal, it drives the output shaft of the reducer to rotate. The output shaft of actuator 5 drives the active crank 7 to rotate. The active crank 7 pulls or pushes the connecting rod body 8 through the hinge pin. The connecting rod body 8 then transmits the motion to the driven crank 9, which drives the valve shaft 3 to rotate, thereby realizing the opening and closing drive of the eccentric valve plate 2.

[0066] In the initial and middle stages of opening and closing (valve opening degree 10°-80°): the valve shaft 3 drives the eccentric valve plate 2 to rotate. Due to the eccentric design, the sealing surface of the eccentric valve plate 2 quickly separates from the valve seat 6 and moves along the eccentric arc trajectory. During this process, the frictional resistance is extremely small, and the required operating torque is much lower than that of the traditional center butterfly valve, resulting in smooth and fluid operation.

[0067] In the final closing and self-locking stages (valve opening 80°-90°): When the eccentric valve plate 2 rotates to a near fully closed position, it begins to contact the sealing surface of the valve seat 6; at the same time, the linkage mechanism moves to a near dead center position, and the force output by the actuator 5 is greatly amplified through the lever effect, generating a strong sealing torque, which is transmitted to the eccentric valve plate 2 through the valve shaft 3, forcing the sealing surface of the valve plate to elastically press against the valve seat 6, forming a uniform and reliable metal hard seal; when the valve reaches the 90° fully closed position, the linkage mechanism reaches the mechanical dead center position, realizing mechanical self-locking and preventing the eccentric valve plate 2 from opening in the reverse direction due to medium pressure.

[0068] Example 2:

[0069] like Figure 3-4 As shown, a method for testing the performance of an eccentric butterfly valve, used to test an eccentric butterfly valve with a self-locking structure as described in Example 1, includes the following steps:

[0070] S1. Simultaneously collect the test data of the sealing pair during the entire opening and closing cycle of the eccentric butterfly valve under test, as well as the self-locking performance parameters of the self-locking structure performance test.

[0071] S2. Based on the elastoplastic contact mechanics model of the metal hard seal pair, calculate the dynamic contact pressure distribution of the seal pair according to the detection data of the seal pair;

[0072] S3. Multimodal spatiotemporal feature fusion is performed on the static structural characteristics and dynamic response characteristics of the sealing pair during the opening and closing process to identify sealing pair defects and generate a leakage early warning area map.

[0073] S4. Construct a performance coupling correlation model between the sealing pair and the connecting rod self-locking structure, determine the influence relationship between the performance changes of the self-locking structure and the sealing performance, and generate coupling anomaly correlation results.

[0074] S5. Combine the leakage early warning area map with the coupling anomaly correlation results to generate a comprehensive test result of the butterfly valve sealing performance.

[0075] In this embodiment, the performance testing of the self-locking structure includes:

[0076] T1. Collect motion timing data and torque timing data of the linkage mechanism during the entire opening and closing cycle of the eccentric butterfly valve to be tested;

[0077] T2. Based on the motion timing data and torque timing data, generate the transmission performance parameters and dead point position parameters of the linkage mechanism;

[0078] T3. Construct a self-locking mechanical model based on the transmission performance parameters, and calculate the self-locking critical threshold and the actual self-locking holding torque;

[0079] T4. Generate the performance test results of the self-locking structure based on the self-locking critical threshold, the actual self-locking holding torque, and the dead point position parameters.

[0080] For eccentric butterfly valves with self-locking structures, a coupled correlation model of the sealing pair and the self-locking structure of the connecting rod is constructed by graph attention network. This model can accurately locate the root cause of the self-locking structure abnormality that leads to deterioration of sealing performance. It solves the core pain point that existing technologies cannot correlate the transmission structure and sealing performance, and realizes a comprehensive evaluation of the valve's core performance. This is fundamentally different from existing testing technologies.

[0081] Step T1 includes:

[0082] T11. The high-speed vision acquisition unit continuously acquires motion trajectory time-series images of the active crank, connecting rod body, driven crank and each hinge point of the linkage mechanism within a preset opening and closing time period, as the motion time-series data.

[0083] T12. The torque timing data of the actuator output shaft and the load torque timing data of the valve shaft are collected by the torque sensing unit, and the valve opening and closing angle timing data are collected simultaneously, which together serve as the torque timing data.

[0084] Specifically, step T2 includes:

[0085] T21. Perform subpixel registration on the motion trajectory time image, extract the spatial displacement vector sequence of each hinge point in the entire opening and closing cycle, determine the actual spatial position of the hinge point between the driven crank and the connecting rod body in the fully closed state of the valve, and calculate the actual offset of the dead point position as the dead point position parameter.

[0086] T22. Based on the output torque timing data, load torque timing data, and opening / closing angle timing data, calculate the transmission ratio, torque amplification factor, and transmission efficiency of the linkage mechanism under different opening / closing angles, and use them as the transmission performance parameters.

[0087] Specifically, step T3 includes:

[0088] T31. Based on the aforementioned transmission performance parameters, and considering the reverse thrust load of the medium on the eccentric valve plate when the valve is fully closed, a self-locking mechanical model of the linkage mechanism is constructed. The calculation formula for the reverse thrust generated by the medium pressure on the eccentric valve plate when the valve is fully closed is shown below:

[0089]

[0090] in, P is the reverse thrust of the medium on the valve plate; S is the nominal pressure of the medium in the pipeline; and S is the effective pressure-bearing area of ​​the valve plate.

[0091] Formula for calculating the reverse opening torque generated by the medium thrust on the valve shaft:

[0092]

[0093] in, denoted as , where is the reverse opening torque acting on the valve bearing; e is the eccentricity from the point of application of the resultant force on the valve plate to the valve shaft axis.

[0094] T32. Calculate the minimum self-locking torque to prevent the linkage mechanism from rotating in the opposite direction using the self-locking mechanical model, and use it as the self-locking critical threshold; the calculation formula for the minimum self-locking torque (self-locking critical threshold) to prevent the valve from opening in the opposite direction when the linkage mechanism is in the dead position is as follows:

[0095]

[0096] in, is the self-locking critical threshold; i is the transmission ratio of the linkage mechanism at the dead point position; The mechanical transmission efficiency of the linkage mechanism; This is the sum of the frictional resistance torques between each hinge joint of the linkage mechanism and the valve shaft bearing.

[0097] T33. Based on torque timing data and transmission performance parameters under fully closed valve conditions, calculate the maximum reverse resistance torque of the linkage mechanism at the dead point position, which is used as the actual self-locking torque. The formula for calculating the actual self-locking torque is as follows: Based on the actuator output torque and the force amplification effect of the linkage mechanism.

[0098]

[0099] in, This is the actual self-locking holding torque; This is the output locking torque of the actuator when the valve is fully closed.

[0100] Specifically, step T4 includes:

[0101] T41. When the actual self-locking holding torque is greater than or equal to the self-locking critical threshold, and the actual offset of the dead point position is less than the preset offset threshold, the self-locking structure is deemed to be of qualified performance.

[0102] T42. When the actual self-locking holding torque is less than the self-locking critical threshold, or the actual offset of the dead point position is greater than or equal to the preset offset threshold, the self-locking structure performance is determined to be abnormal, and corresponding self-locking failure warning information is generated.

[0103] T43. Integrate transmission performance parameters, dead point position parameters, self-locking holding torque and qualification judgment results to generate self-locking structure performance test results.

[0104] In this embodiment, in step S1, the detection data of the sealing pair includes the light transmission intensity distribution image of the sealing pair, the structured light reflection interference image sequence, the optical speckle image sequence of the opening and closing process, the infrared thermal imaging sequence, and the acoustic emission timing signal; the self-locking performance parameters include the transmission performance parameters, the dead point position parameters, the self-locking critical threshold, and the actual self-locking holding torque.

[0105] Specifically, step S2 includes:

[0106] S21. Based on the light transmission intensity distribution image and the structured light reflection interference image sequence, obtain the spatial distribution of light transmission intensity of the sealing pair and the evolution sequence of the micromorphology of the sealing surface, and establish the spatial reference coordinate system of the sealing pair.

[0107] S22. Construct an elastoplastic contact mechanics model for a metal hard seal pair, clarifying the mapping relationship between the mechanical properties of the sealing surface material and contact deformation and contact pressure; specifically, when the contact stress of the sealing surface is lower than the material's yield strength, the formula for calculating elastic contact deformation is as follows:

[0108]

[0109] in, This refers to the amount of deformation at the sealing surface. is Poisson's ratio of the sealing surface material; E is the elastic modulus of the sealing surface material; a is the equivalent radius of the contact area; denoted as the radial contact pressure distribution function of the contact area; r is the radial distance from any point within the contact area to the contact center.

[0110] The formula for calculating elastoplastic deformation after the contact stress reaches the material's yield strength is as follows:

[0111]

[0112] in, The average contact pressure in the contact area; The yield strength of the sealing surface material; The initial contact equivalent radius is given.

[0113] Quantitative relationship between sealing contact gap and light transmission intensity:

[0114]

[0115] in, Let be the intensity of transmitted light at coordinate (x, y) of the sealed sub-coordinate. The initial intensity of the incident light; The light absorption coefficient of the medium on the sealing surface; This is to seal the contact gap at the sub-coordinate (x, y).

[0116] S23. Based on the aforementioned elastoplastic contact mechanics model, a pre-trained contact pressure inversion convolutional neural network is used to calculate the spatial distribution of light transmission intensity and the evolution sequence of microstructure, outputting a dynamic contact pressure distribution matrix sequence of the sealing pair throughout the entire opening and closing cycle. The contact pressure inversion convolutional neural network employs an Encoder-Decoder architecture, including an encoder module, a residual feature fusion module, and a decoder module. The encoder module extracts deep features from the input data, the residual feature fusion module incorporates prior constraints from the elastoplastic contact mechanics model to correct feature weights, and the decoder module reconstructs the output contact pressure distribution matrix. Specifically, the encoder... Formula for calculating the output feature map of a convolutional layer:

[0117]

[0118] in, For encoder number The output feature map of the layer; For encoder number The convolution kernel weight matrix of the layer; This is a two-dimensional convolution operation; For encoder number -1 layer input feature map (the first layer is a fusion feature map of light transmission intensity and micromorphology); For encoder number Layer bias terms; It is a linear rectification activation function.

[0119] Introducing the residual fusion formula for prior constraints in elastoplastic contact mechanics:

[0120]

[0121] in, The fused feature map; This is the deep feature map output by the encoder; These are the prior constraint weight coefficients, with values ​​ranging from 0.1 to 0.5. This is a mechanical feature map calculated based on the elastoplastic contact model.

[0122] The formula for calculating the output of the k-th deconvolutional layer of the decoder is as follows:

[0123]

[0124] in, This is the output feature map of the k-th layer of the decoder; Let K be the deconvolution kernel weight matrix of the k-th layer of the decoder; This is the input feature map of the (k-1)th layer of the decoder; This is the bias term of the k-th layer of the decoder; the final output of the network is a two-dimensional contact pressure distribution matrix that matches the size of the input image. .

[0125] Specifically, step S3 includes:

[0126] S31. Spatial registration and temporal synchronization of the multi-source detection data of the sealing pair are performed to generate a multimodal feature dataset with a unified spatiotemporal index;

[0127] S32. Using a pre-trained spatiotemporal fusion Transformer defect recognition model, static structural features and dynamic response features are extracted and fused from the multimodal feature dataset, outputting the classification and grading results of sealing joint defects. The spatiotemporal fusion Transformer defect recognition model includes a spatial feature extraction branch, a temporal feature extraction branch, and a cross-modal attention fusion module. The spatial feature extraction branch is used to extract static structural defect features of the sealing joint, the temporal feature extraction branch is used to extract dynamic response anomaly features during the opening and closing process, and the cross-modal attention fusion module is used to achieve adaptive weighted fusion of multimodal features. Specifically, the specific formula for the scaling dot product attention mechanism is shown below:

[0128]

[0129] Where Q is the query matrix; K is the key matrix; and V is the value matrix; Let be the dimension of the key matrix; It is a normalized exponential function.

[0130] The specific formula for cross-modal multi-head attention is as follows:

[0131]

[0132]

[0133] Where h is the number of attention heads; , , Let be the linear transformation weight matrix corresponding to the i-th attention head; The linear transformation weight matrix for the multi-head attention output; This is a feature splicing operation.

[0134] The formula for fusing static spatial features and dynamic temporal features is as follows:

[0135]

[0136] in, The spatiotemporal characteristics after fusion; Extract static structural features from the branch output for spatial features; Extract dynamic response features from the branch outputs for time-series features; This is a layer normalization operation used to stabilize the training process.

[0137] S33. Based on the classification and grading results of the defects, generate a defect distribution map of the sealing pair, mark the areas with leakage risk levels higher than the preset threshold as leakage warning areas, and generate a leakage warning area map.

[0138] Specifically, step S4 includes:

[0139] S41. Using each sealing sub-region of the sealing pair as the first type of node, each transmission component of the linkage mechanism as the second type of node, and the physical relationship between sealing performance and self-locking performance as the edge, construct a heterogeneous graph network of the sealing pair-linkage self-locking structure.

[0140] S42. By coupling the correlation model with a pre-trained graph attention network, the dynamic correlation weights of each node in the heterogeneous graph network are learned, quantifying the contribution of the performance changes of the self-locking structure to the sealing pair defects. Specifically, in the graph node feature linear transformation, a linear transformation is performed on the features of all nodes in the heterogeneous graph network, and the specific formula for unifying the feature dimension is as follows:

[0141]

[0142] in, This is the original feature vector of the i-th node; For shared linear transformation weight matrix; is the node feature vector after linear transformation.

[0143] Attention weight calculation: Calculates the attention weight between adjacent nodes to quantify the strength of the association between nodes.

[0144]

[0145]

[0146] in, is the association attention coefficient between node i and node j; a is the weight vector of a single-layer feedforward neural network; This is a feature vector concatenation operation; It is a linear rectified activation function with leakage; Let i be the set of adjacent nodes of node i; The normalized attention weight represents the degree of influence of node j on node i, i.e., the contribution of the self-locking structure anomaly to the sealing defect.

[0147] Node feature aggregation and output: The features of adjacent nodes are aggregated using attention weights, the node features are updated, and the association results are output.

[0148]

[0149] in, This refers to the aggregated and updated node features; The Sigmoid activation function is used; based on the updated node features and attention weight matrix, a correlation mapping relationship between sealing defects and self-locking structure anomalies is generated.

[0150] S43. Based on the correlation weight and contribution, determine the correlation mapping relationship between sealing defects and self-locking structure anomalies, and generate a coupling anomaly correlation map as the coupling anomaly correlation result.

[0151] In this embodiment, the detection method also includes sealing performance life prediction, the specific steps of which are as follows:

[0152] S6. Obtain historical test data of the eccentric butterfly valve to be tested during multiple cycles of opening and closing, including dynamic contact pressure sequence, defect feature sequence and self-locking performance parameter sequence;

[0153] S7. Using a pre-trained bidirectional LSTM lifetime prediction model, the historical detection data and temporal trends are learned to output the sealing performance degradation trend curve, remaining opening and closing cycles, and remaining service life; specifically, in the LSTM unit gating mechanism, the forget gate calculation formula is as follows:

[0154]

[0155] Input gate calculation formula:

[0156]

[0157]

[0158] Cell state update formula:

[0159]

[0160] Output gate calculation formula:

[0161]

[0162]

[0163] in, , , These represent the outputs of the forget gate, input gate, and output gate at time t, respectively. , These represent the cell state and candidate cell state at time t, respectively. , These are the hidden layer outputs at time t and time t-1, respectively. The input feature vector at time t (contact pressure, defect features, self-locking performance parameters); , , , This is the weight matrix corresponding to the gating; , , , For the bias term corresponding to the gating; is the Sigmoid activation function; tanh is the hyperbolic tangent activation function; This is the Hadamard product (element-level product).

[0164] Bidirectional LSTM feature fusion, the feature fusion formula between forward LSTM and backward LSTM:

[0165]

[0166] in, The fused output features of the bidirectional LSTM at time t; The output of the hidden layer of the feedforward LSTM; This is the hidden layer output of the inverse LSTM.

[0167] The lifetime prediction output is processed through a fully connected layer to output the final lifetime prediction result:

[0168]

[0169] in, The predicted output includes the trend of sealing performance degradation, the remaining number of opening and closing cycles, and the remaining service life; The bidirectional LSTM fusion output features are the features at the last moment of the sequence. This is the output layer weight matrix; This is the output layer bias term.

[0170] S8. Based on the comprehensive test results of sealing performance and the life prediction results, generate operation and maintenance recommendations for the entire life cycle of the butterfly valve.

[0171] The above-described specific embodiments are preferred embodiments of an eccentric butterfly valve with a self-locking structure and a performance testing method of this application, and are not intended to limit the specific scope of this application. The scope of this application includes but is not limited to the specific embodiments described herein. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. An eccentric butterfly valve with a self-locking structure, characterized in that, It includes the valve body, eccentric valve plate, valve shaft, aluminum seat, actuator, and linkage mechanism; The valve shaft is rotatably mounted on the valve body, and the eccentric valve plate is fixedly mounted on the valve shaft and located in the flow channel of the valve body. The eccentric valve plate and the valve shaft are eccentrically mounted. The aluminum base is fixed to the outside of the valve body, and the actuator is fixedly mounted on the aluminum base. The output shaft of the actuator is connected to the valve shaft through the linkage mechanism. The linkage mechanism includes a driving crank, a connecting rod body, and a driven crank. The driving crank is fixed to the output shaft of the actuator, the driven crank is fixed to the extended end of the valve shaft, and the two ends of the connecting rod body are respectively hinged to the driving crank and the driven crank via pins. When the eccentric butterfly valve is in the fully closed state, the hinge center of the connecting rod body and the driven crank, the rotation center of the driven crank and the valve shaft, and the hinge center of the connecting rod body and the driving crank are all on the same straight line. The connecting rod mechanism moves to the mechanical dead point position, forming a reverse self-locking structure.

2. The eccentric butterfly valve with a self-locking structure according to claim 1, characterized in that, A valve seat is fixedly installed inside the flow channel of the valve body. When the eccentric butterfly valve is in the fully closed state, the sealing surface of the eccentric valve plate and the sealing surface of the valve seat are pressed tightly together to form a metal hard seal.

3. The eccentric butterfly valve with a self-locking structure according to claim 1, characterized in that, The rotation center of the eccentric valve plate is eccentrically set with respect to its own geometric center, and the sealing surface of the eccentric valve plate and the sealing surface of the valve seat move along an eccentric circular arc trajectory during the opening and closing process.

4. A method for testing the performance of an eccentric butterfly valve, characterized in that, The method for testing an eccentric butterfly valve with a self-locking structure as described in any one of claims 1-3 includes the following steps: S1. Simultaneously collect the test data of the sealing pair during the entire opening and closing cycle of the eccentric butterfly valve under test, as well as the self-locking performance parameters of the self-locking structure performance test. S2. Based on the elastoplastic contact mechanics model of the metal hard seal pair, calculate the dynamic contact pressure distribution of the seal pair according to the detection data of the seal pair; S3. Multimodal spatiotemporal feature fusion is performed on the static structural characteristics and dynamic response characteristics of the sealing pair during the opening and closing process to identify sealing pair defects and generate a leakage early warning area map. S4. Construct a performance coupling correlation model between the sealing pair and the connecting rod self-locking structure, determine the influence relationship between the performance changes of the self-locking structure and the sealing performance, and generate coupling anomaly correlation results. S5. Combine the leakage early warning area map with the coupling anomaly correlation results to generate a comprehensive test result of the butterfly valve sealing performance.

5. The method for testing the performance of an eccentric butterfly valve according to claim 4, characterized in that, The performance testing of the self-locking structure includes: T1. Collect motion timing data and torque timing data of the linkage mechanism during the entire opening and closing cycle of the eccentric butterfly valve to be tested; T2. Based on the motion timing data and torque timing data, generate the transmission performance parameters and dead point position parameters of the linkage mechanism; T3. Construct a self-locking mechanical model based on the transmission performance parameters, and calculate the self-locking critical threshold and the actual self-locking holding torque; T4. Generate the performance test results of the self-locking structure based on the self-locking critical threshold, the actual self-locking holding torque, and the dead point position parameters.

6. The method for testing the performance of an eccentric butterfly valve according to claim 5, characterized in that, Step T1 includes: T11. The high-speed vision acquisition unit continuously acquires motion trajectory time-series images of the active crank, connecting rod body, driven crank and each hinge point of the linkage mechanism within a preset opening and closing time period, as the motion time-series data. T12. The torque timing data of the actuator output shaft and the load torque timing data of the valve shaft are collected by the torque sensing unit, and the valve opening and closing angle timing data are collected simultaneously, which together serve as the torque timing data.

7. The method for testing the performance of an eccentric butterfly valve according to claim 5, characterized in that, Step T2 includes: T21. Perform subpixel registration on the motion trajectory time image, extract the spatial displacement vector sequence of each hinge point in the entire opening and closing cycle, determine the actual spatial position of the hinge point between the driven crank and the connecting rod body in the fully closed state of the valve, and calculate the actual offset of the dead point position as the dead point position parameter. T22. Based on the output torque timing data, load torque timing data, and opening / closing angle timing data, calculate the transmission ratio, torque amplification factor, and transmission efficiency of the linkage mechanism under different opening / closing angles, and use them as the transmission performance parameters.

8. The method for testing the performance of an eccentric butterfly valve according to claim 5, characterized in that, Step T3 includes: T31. Based on the transmission performance parameters, and combined with the reverse thrust load of the medium on the eccentric valve plate in the fully closed state of the valve, a self-locking mechanical model of the linkage mechanism is constructed. T32. Calculate the minimum self-locking torque to prevent the linkage mechanism from rotating in the opposite direction using the self-locking mechanical model, and use it as the self-locking critical threshold. T33. Based on the torque timing data and transmission performance parameters of the valve in the fully closed state, calculate the maximum reverse resistance torque of the linkage mechanism at the dead point position, which is used as the actual self-locking holding torque.

9. The method for testing the performance of an eccentric butterfly valve according to claim 4, characterized in that, Step S2 includes: S21. Based on the light transmission intensity distribution image and the structured light reflection interference image sequence, obtain the spatial distribution of light transmission intensity of the sealing pair and the evolution sequence of the micromorphology of the sealing surface, and establish the spatial reference coordinate system of the sealing pair. S22. Construct an elastoplastic contact mechanics model for a metal hard seal pair and clarify the mapping relationship between the mechanical properties of the sealing surface material and contact deformation and contact pressure. S23. Based on the elastoplastic contact mechanics model, the spatial distribution of light transmission intensity and the evolution sequence of micromorphology are calculated through a pre-trained contact pressure inversion convolutional neural network, and the dynamic contact pressure distribution matrix sequence of the sealing pair during the entire opening and closing cycle is output.

10. The method for testing the performance of an eccentric butterfly valve according to claim 4, characterized in that, Step S4 includes: S41. Using each sealing sub-region of the sealing pair as the first type of node, each transmission component of the linkage mechanism as the second type of node, and the physical relationship between sealing performance and self-locking performance as the edge, construct a heterogeneous graph network of the sealing pair-linkage self-locking structure. S42. By coupling the correlation model with a pre-trained graph attention network, the dynamic correlation weights of each node in the heterogeneous graph network are learned, and the contribution of the performance change of the self-locking structure to the sealing pair defect is quantified. S43. Based on the correlation weight and contribution, determine the correlation mapping relationship between sealing defects and self-locking structure anomalies, and generate a coupling anomaly correlation map as the coupling anomaly correlation result.

Images