A steel gate corrosion and fatigue accelerated deterioration test bench and a post-deterioration operation state evaluation method

Abstract

The application provides a kind of steel gate accelerated degradation test bench of simultaneous action of corrosion and fatigue and the method for evaluating operating state after degradation, it includes corrosion test device, fatigue test device, data acquisition device and control system;The corrosion test device includes corrosion box, temperature control device, waste liquid collecting device, liquid supplementing device, electrochemical test device and air humidity control device, through liquid level sensor, thermometer and air humidity sensor, realize the accelerated corrosion test of control console to sample;The fatigue test device includes torsional stress loading device and tensile stress loading device, through constant mechanical load loading device, constant force loading and torsional load loading in horizontal direction are carried out to sample;The data acquisition device includes parallel light source, industrial camera, camera fixing frame, plane mirror, control console, remote ultrasonic thickness gauge and stress strain sensor, and the degradation information of sample surface is collected by industrial camera.

Description

Technical Field

[0001] This invention relates to the field of corrosion, fatigue and deterioration assessment, and specifically to an accelerated deterioration test bench for steel gates subjected to simultaneous corrosion and fatigue, and a method for assessing the operational status after deterioration. Background Technology

[0002] Metal components are widely used in various engineering facilities. Among them, the surfaces of hydraulic metal structures are affected by their service environment and operating factors, making corrosion and fatigue significant factors contributing to their failure. Therefore, the corrosion resistance and fatigue resistance of metal surfaces directly affect the safe operation of the entire equipment. Establishing a test bench for simultaneous corrosion and fatigue testing of hydraulic metal components, and analyzing their corrosion resistance and fatigue bending capacity, helps to more clearly understand and quantify the performance of hydraulic metal component samples. Based on this, a more comprehensive and reasonable assessment method for the deterioration damage of hydraulic metal components can be developed, reducing component failures caused by deterioration and ensuring the safe and stable operation of equipment.

[0003] During gate operation, most of the equipment and devices utilize metal materials that operate under specific corrosive environments and stress conditions. Conventional fatigue tests on hydraulic metal components are conducted indoors, failing to simulate fatigue tests in actual corrosive environments. Consequently, the experimental results deviate significantly from actual operating conditions. This not only fails to achieve the goal of rationally selecting and evaluating hydraulic metal components but also leads to substantial economic losses and even accidents resulting in casualties due to the actual operation of the gate. Therefore, an accelerated deterioration test bench for steel gates that simultaneously withstands corrosion and fatigue, along with a method for evaluating the operational status after deterioration, is needed to conduct accelerated deterioration experiments on hydraulic metal parts and assess their operational status after deterioration. Summary of the Invention

[0004] The purpose of this invention is to provide an accelerated deterioration test bench for steel gates under the simultaneous effects of corrosion and fatigue, and a method for evaluating the operational status after deterioration. This test bench is suitable for testing the fatigue and corrosion resistance of hydraulic metal parts, and can better reflect the fatigue and corrosion resistance of different metal samples.

[0005] To achieve the above-mentioned technical features, the present invention aims to provide a steel gate accelerated deterioration test bench that combines corrosion and fatigue, comprising a corrosion test device, a fatigue test device, a data acquisition device, and a control system.

[0006] The corrosion testing device includes a corrosion chamber, a temperature control device, a waste liquid collection device, a liquid replenishment device, an electrochemical testing device, and an air humidity control device. Through a liquid level sensor, a thermometer, and an air humidity sensor, the control console enables accelerated corrosion testing of the sample.

[0007] The fatigue testing apparatus includes a torsional stress loading device and a tensile stress loading device, and applies a constant force and torsional load to the specimen in the horizontal direction through a constant mechanical load loading device.

[0008] The data acquisition device includes a parallel light source, an industrial camera, a camera mount, a plane mirror, a control console, a remote ultrasonic thickness gauge, and a stress-strain sensor. The industrial camera is used to collect information on the deterioration of the sample surface.

[0009] The temperature control device is located on both sides of the corrosion chamber, with multiple parallel thermocouples spaced at intervals along its height. The thermometers are located on both sides of the corrosion chamber. The thermocouples and thermometers are connected to the control console to detect and adjust the temperature inside the corrosion chamber. The corrosion chamber is made of a transparent, corrosion-resistant material.

[0010] The waste liquid collection device includes a liquid outlet valve, a liquid outlet pipe, and a waste liquid collection tank. One end of the liquid outlet pipe is connected to the corrosion tank, and the other end is connected to the waste liquid collection tank. The liquid outlet valve is installed on the liquid outlet pipe and is communicatively connected to the control console.

[0011] The replenishment device includes a first inlet pipe, a second inlet pipe, a peristaltic pump, a corrosion tank, and a liquid level sensor. One end of the first inlet pipe is connected to the peristaltic pump, and the other end is connected to the corrosion tank. One end of the second inlet pipe is connected to the corrosion tank, and the other end is connected to the peristaltic pump. The peristaltic pump is communicatively connected to the control console, and the liquid level sensor is communicatively connected to the control console.

[0012] The electrochemical testing apparatus includes a reference electrode, an auxiliary electrode, a sample, a sample holder, and a constant current instrument. The reference electrode, auxiliary electrode, and sample constitute a three-electrode system. One end of the reference electrode is installed in the corrosion chamber, and the other end of the reference electrode is electrically connected to the constant current instrument. The constant current instrument is communicatively connected to the control console. One end of the auxiliary electrode is installed in the corrosion chamber, and the other end of the auxiliary electrode is electrically connected to the control console. Both ends of the sample are connected to the sample holder, and the sample holder is electrically connected to the control console.

[0013] The air humidity control device includes a humidifier, an air humidity sensor, a water inlet pipe, and a water tank. One end of the humidifier is connected to the corrosion chamber, one end of the water inlet pipe is connected to the humidifier, and the other end of the water inlet pipe is connected to the water tank. The air humidity sensor is communicatively connected to the control console.

[0014] The torsional stress loading device includes a sample clamp, a first clamping ring, a hydraulic actuator, and a hydraulic rod. The sample clamp is located at both ends of the sample. The hydraulic actuator is connected to the hydraulic rod, and the hydraulic rod is connected to the first clamping ring. The hydraulic actuator is communicatively connected to the control console. The hydraulic actuator is a constant mechanical load loading device. The torsional load on the test piece is applied by the extension and retraction of the hydraulic rod. Because the sample is fixed on the clamp, it achieves torsional movement under the force of the first clamping ring, thereby realizing the torsional load loading of the sample by the device. A sealing ring is provided at the connection between the hydraulic actuator and the bottom plate of the corrosion chamber.

[0015] The ends of the first clamping ring are configured as a pair of semi-circular arc structures that match the sample. The pair of semi-circular arc structures are interlocked and sleeved on the outer wall of the sample. The free end of the first clamping ring extends horizontally as a clamping rod. Fastening bolts are provided at corresponding positions of the clamping rods. The first clamping ring is fastened by the bolts. The free end of the clamping rod is connected to the hydraulic rod on the hydraulic device. The height of the hydraulic rod on the hydraulic device can be adjusted according to the actual situation.

[0016] The tensile stress loading device includes a second clamping ring, a fixed plate, a stationary slider, a movable slider, a lead screw, and a distance sensor. The second clamping ring is connected to the sample, the fixed plate is connected to the groove of the corrosion chamber, the movable slider and the stationary slider are connected to the bottom plate of the corrosion chamber and the two sides of the fixed plate, respectively. The two lead screws are connected to the fixed plate and the second clamping ring through lead screw nuts, respectively. The distance sensor is arranged on the fixed plate and is communicatively connected to the control console. The control system detects the signal from the distance sensor to determine the magnitude of the constant force loading value.

[0017] The data acquisition device includes a parallel light source, an industrial camera, a camera mount, a plane mirror, a control console, a remote ultrasonic thickness gauge, and a stress-strain sensor. The industrial camera is connected to the camera mount. Part of the parallel light source is connected to the camera mount, and the other part is located on the bottom plate of the corrosion chamber. The industrial camera is communicatively connected to the control console. The plane mirror is located at the bottom of the corrosion chamber to assist the industrial camera in observing the degradation of the lower surface of the sample. The remote ultrasonic thickness gauge is located on the side surface of the corrosion chamber to measure the degradation inside the sample. The stress-strain sensor is connected to the surface of the sample and is communicatively connected to the control console.

[0018] A method for evaluating the operational status of deteriorated hydraulic metal parts involves first conducting an accelerated deterioration test on the hydraulic metal parts using a steel gate accelerated deterioration test bench with simultaneous corrosion and fatigue effects as described in any one of claims 1-9, and then evaluating the operational status of the deteriorated hydraulic metal parts after the test.

[0019] The accelerated degradation test includes the following steps:

[0020] Step 1.1, Pretreatment of the sample:

[0021] Polish the surface of the metal sample with a grinder, then wipe the surface of the metal sample clean and clean it with acetone and blow it dry. Fix the sample in the corrosion chamber with a sample clamp.

[0022] Step 1.2, conduct the experiment:

[0023] Depending on the experimenter's needs, three different testing requirements can be selectively met: corrosion testing, fatigue testing, and corrosion fatigue testing.

[0024] 1.2-1, Corrosion Test:

[0025] First, the corrosive solution was placed into the replenishment tank and the corrosion chamber, respectively. Then, it was confirmed that all required sensors were functioning correctly. When the level sensor detected that the metal sample was not yet completely immersed in the corrosive solution, corrosion tests were performed on the metal sample under humidified air, alternating wet and dry conditions, and complete immersion. The experimental parameters for the inlet valve, outlet valve, and humidifier were set via the control panel, specifically including air humidity and the frequency of wet and dry alternation. Finally, when the level sensor detected that the metal sample was completely immersed in the corrosive solution, an electrochemical corrosion test was performed on the metal sample. One end of the reference electrode was installed inside the corrosion chamber, and the other end of the reference electrode was electrically connected to the galvanometer, which was communicatively connected to the control panel. One end of the auxiliary electrode is installed in the corrosion tank, and the other end of the auxiliary electrode is electrically connected to the control console. Both ends of the test sample are connected to the sample holder, which is electrically connected to the control console. Electrochemical experimental parameters, including test time and scanning potential parameters, are set through the control console with a data acquisition module. In this way, electrochemical curves of metal samples under fully wet and dry-wet alternating environments can be obtained, including open circuit potential test curves, potentiodynamic polarization curves, and AC impedance Nyquist plots. Through data processing, open circuit potential, self-corrosion current density, and self-corrosion potential parameters can be obtained, and the electrochemical corrosion resistance of the metal sample under multi-factor coupled environment can be evaluated.

[0026] 1.2-2, Fatigue test:

[0027] First, one end of the first clamping ring is fixed to one end of the metal sample, and the other end of the clamping ring is connected to the hydraulic rod. The extension and retraction of the hydraulic rod is controlled by the control console to determine the torsional load on the metal sample. Then, one end of the second clamping ring is fixed to the other end of the metal sample, and the lead screw is connected to the other end of the clamping ring through the fixing plate. The rotation of the lead screw is controlled by the control console, and the tensile load on the metal sample is determined by the distance sensor.

[0028] 1.2-3, Corrosion fatigue test:

[0029] First, fix one end of the first clamping ring to one end of the metal sample, and connect the other end of the clamping ring to the hydraulic rod. Fix one end of the second clamping ring to the other end of the metal sample, and connect the lead screw to the other end of the clamping ring through the fixing plate. Then, put the corrosion solution into the replenishment tank and the corrosion tank respectively. Finally, confirm that all the sensors required for the experiment are working properly. At this time, the fatigue test is carried out. By monitoring the liquid level sensor, it is possible to determine the corrosion test and electrochemical test under what environment. The two tests are carried out simultaneously.

[0030] The method for assessing the deteriorated operating status includes the following steps:

[0031] Step 2.1, Establishing the 3D Model:

[0032] Use modeling tools to create a 3D model of the metal structure, ensuring that the geometry and dimensions of the model are consistent with the actual metal structure;

[0033] Step 2.2, Surface wear and corrosion information collection:

[0034] 2.2-1, Setting the angle of the plane mirror:

[0035] The camera lens is positioned directly above the sample. The distance and angle between the camera lens, the sample, and the plane mirror are determined so that the camera can capture the entire surface of the sample to obtain information on surface wear and corrosion.

[0036] The range of plane mirror angles is:

[0037]

[0038] In the formula: L1 is the distance from the camera lens to the bottom of the sample; L2 is the distance from the bottom of the sample to the bottom of the corrosion chamber; L3 is the distance from the center of the sample to the x-axis of the plane mirror; L4 is the width of the plane mirror; R1 is the radius of the camera lens;

[0039] 2.2-2, Image preprocessing:

[0040] Since the acquired images contain a small amount of background interference and the effect of plane mirror refraction, filtering and denoising methods should be used in the preprocessing stage to remove this interference and crop out the excess background so that the image contains only the metal sample. In addition, enhancement algorithms are used to enhance the details, strengthen the details of the deteriorated parts, highlight the difference between the deteriorated and non-deteriorated areas, thereby improving the detection accuracy and enhancing the image contrast.

[0041] 2.2-3, Measurement of wear and corrosion depth:

[0042] Based on the preprocessed images in 2.2-2, the wear and corrosion depth of the corresponding areas are measured using a remote ultrasonic thickness gauge;

[0043] Step 2.3, Feature Extraction:

[0044] The Canny edge detection algorithm is used to extract preliminary edge information of wear and corrosion. First, the image is converted to grayscale. Then, the Canny edge detection algorithm is used with appropriate parameters to extract the edges of wear and corrosion, resulting in a binary image of wear and corrosion. The specific steps and formulas are as follows:

[0045] 2.3-1, Calculate the gradient:

[0046] Calculate the horizontal gradient of the image: Gx(x,y)=I(x+1,y)-I(x-1,y);

[0047] Calculate the vertical gradient of the image: Gy(x,y)=I(x,y+1)-I(x,y-1);

[0048] Calculate the gradient magnitude: G(x,y) = sqrt(Gx(x,y)) 2 +Gy(x,y) 2 );

[0049] Calculate the gradient direction: θ(x,y)=atan2(Gy(x,y),Gx(x,y));

[0050] In the formula: Gx(x,y) represents the gradient value in the horizontal direction, Gy(x,y) represents the gradient value in the vertical direction, I(x,y) represents the gray value on the original image, and atan2() represents the arctangent function, which is used to calculate the gradient direction;

[0051] 2.3-2, Non-maximum suppression:

[0052] First, for each pixel P(x,y), calculate the positions of the two adjacent pixels of P along the gradient direction. Then, determine whether the gradient magnitude of P is greater than the gradient magnitude of the adjacent pixels. If it is, retain the gradient magnitude of P; otherwise, set the gradient magnitude of P to 0. The formula for calculating the positions of adjacent pixels is as follows:

[0053] (x1,y1)=(x+round(cos(θ(x,y))),y+round(sin(θ(x,y))));

[0054] (x2,y2)=(x-round(cos(θ(x,y))),y-round(sin(θ(x,y))));

[0055] In the formula: (x1, y1) represents the position of the first adjacent pixel along the gradient direction, and (x2, y2) represents the position of the second adjacent pixel along the gradient direction;

[0056] 2.3-3, Dual Threshold Processing:

[0057] For each pixel P(x,y), if G(x,y) is greater than the high threshold (T_high), it is marked as a strong edge; if G(x,y) is less than the low threshold (T_low), it is marked as a non-edge; if G(x,y) is between the low threshold and the high threshold, and is connected to a strong edge, and has strong edge pixels in its 8-neighborhood, it is marked as a weak edge.

[0058] 2.3-4, Edge Connectivity:

[0059] First, for each strong edge pixel, mark it as an edge pixel. Second, for each weak edge pixel P(x,y), check the pixels in its 8-neighborhood. Then, for each pixel Q(a,b) in its 8-neighborhood, if Q(a,b) is a strong edge pixel and has not been marked as an edge pixel, mark Q(a,b) as an edge pixel, and then connect the unmarked edge pixels in the 8-neighborhood of Q(a,b) in turn. Continue processing the next unmarked weak edge pixel, and repeat steps 2.3-2 and 2.3-3 until all weak edge pixels have been processed.

[0060] Step 2.4, Data Labeling and Preparation:

[0061] The collected image data was manually labeled, dividing it into normal areas, worn areas, and corroded areas, where 0 represents normal, 1 represents worn, and 2 represents corroded. The depth data for each type of wear and corrosion was recorded. The labeled image data was then divided into training and testing sets, with 80% of the images used for training and 20% for testing. This ensured that the images in the training and testing sets came from different metal surfaces and were representative. OpenCV was used to augment the images, increasing the number of training samples and enhancing the model's robustness. PyTorch's DataLoader was used to load the training and testing set data. Since PyTorch typically uses integer-encoded category labels, a label encoding dictionary was used to map the category labels to integer codes.

[0062] Step 2.5, Deep Learning Model Training:

[0063] A convolutional neural network (CNN) model is built using PyTorch. The edge features of the input image are used as the input to the model, and the output of the model is the probability distribution of each region category. The CNN model is trained using training set data, and the model weights are optimized through backpropagation and gradient descent to achieve automatic classification of defects such as wear and corrosion.

[0064] Step 2.6, Model Evaluation and Optimization:

[0065] The performance of the trained model is evaluated using test set data, and the accuracy, recall, and F1 score are calculated. Based on the evaluation results, the hyperparameters of the model are adjusted to optimize the model's performance.

[0066] Step 2.7, Defect Identification and Location:

[0067] The trained model is used to predict images and the prediction results are visualized: based on the predicted bounding box information, boxes are drawn on the image to identify defect areas. The bounding boxes are marked with different colors, line widths or styles to highlight the defect areas.

[0068] Step 2.8, Model Optimization and Iteration:

[0069] By providing feedback based on the recognition results, adjusting the model architecture, increasing training data, and trying different data augmentation methods, the performance and generalization ability of the metal surface defect recognition model can be further optimized.

[0070] Step 2.9, Calculation of wear and corrosion degree:

[0071] By calculating the number of pixels in the binary image processed by the neural network in steps 2.2-2.8, the corrosion area can be calculated, and the corrosion area is equal to the number of pixels. Then, by using the remote ultrasonic thickness gauge in steps 2.2-2.3, the corrosion depth of the corresponding area can be calculated, and the length and area of ​​wear, as well as the area and shape characteristics of corrosion can be calculated.

[0072] Step 2.10, Synchronize information on wear and corrosion areas:

[0073] The wear and corrosion information of the metal surface is mapped to the 3D model, and the corresponding areas are marked on the 3D model according to the location, shape and severity of the wear and corrosion.

[0074] Step 2.11, Model Import:

[0075] Import the completed 3D model into ANSYS software, ensuring that the geometric information, material properties, and boundary conditions of the imported model are set correctly.

[0076] Step 2.12, Mechanical Analysis Settings:

[0077] In ANSYS, set the appropriate mechanical analysis type, select the corresponding analysis type according to the actual working environment and loading conditions of the metal structure, including static analysis, dynamic analysis or thermal analysis; set loading conditions, including the force, pressure and temperature applied to the metal structure, to simulate the actual working conditions; set constraint conditions, including fixed supports and constrained displacements, to limit the degree of freedom of the model.

[0078] Step 2.13, Material Property Settings:

[0079] In ANSYS, material properties of metal structures are set, including elastic modulus, Poisson's ratio, and yield strength. Considering the changes in the mechanical properties of the metal structure after deterioration, the material properties are adjusted according to the severity of wear and corrosion information on the metal surface. For areas with different degrees of wear and corrosion, partition settings are made according to their specific material properties.

[0080] Step 14, Simulation of Wear and Corrosion Zones:

[0081] In ANSYS, metal structures with wear and corrosion zones are modeled and simulated. Based on the location, shape, and severity of wear and corrosion, the corresponding areas on the model are modified, and their material properties are set to their corresponding deterioration states. Appropriate meshing techniques are used to mesh the model to ensure accurate modeling of wear and corrosion zones.

[0082] Step 15, Analysis and Judgment:

[0083] After running the mechanical analysis in ANSYS, the stress, strain, and deformation results of the metal structure under actual working conditions are obtained. Based on these parameters, as well as the location and extent of wear and corrosion information, the strength, stability, and reliability of the structure are evaluated. An important evaluation criterion is comparing the relationship between stress and strain and material strength to determine whether the material's yield strength or failure strength has been exceeded. The Mises yield criterion is used to evaluate the structure, and its formula is:

[0084]

[0085] Where: σ eq This refers to the equivalent stress under multiaxial stress, i.e., Mises stress; σ x , σ y , σ z τ represents the normal stress along the x, y, and z axes, respectively; xy , τ yz , τ zx For shear stress, when σ eq When ≤f, f is the design strength value of the steel, that is, the strength condition is met;

[0086] The difference between the load-bearing capacity of a structure and the stress load is defined as the limit state function, as shown in the following formula:

[0087] G(x)=R(x)-S(x)≤0;

[0088] In the formula: R(x) represents the load-bearing capacity of the structure, and S(x) represents the stress load;

[0089] The probability distribution model is chosen using a normal distribution as the input parameter. For each input parameter, including material strength and load magnitude, its mean μ and standard deviation σ are determined. Since the FORM method requires the limit state function to be transformed into a standard normal space for reliability calculation, the inverse transformation of the probability distribution function is used to convert the actual values ​​of the input parameters into standard normal variables. Based on the standard normal variables of the input parameters, the gradient of the limit state function in each direction is calculated, and then the reliability index β is calculated using the inverse normal distribution function. Based on the calculated reliability index β, the reliability of the structure is evaluated. A higher reliability index β indicates that the structure has higher reliability within a given design life.

[0090] The present invention has the following beneficial effects:

[0091] 1. This test bench is suitable for testing the fatigue and corrosion resistance of hydraulic metal parts, and can effectively reflect the fatigue and corrosion resistance of different metal samples. The test bench uses multiple sensors to monitor the environmental information within the test space, and controls various experimental parameters of the test bench through the control console. The corrosion test system and fatigue test system accelerate the deterioration of metal samples. The data acquisition platform uses an industrial camera to collect images of the surface of the metal samples and a remote ultrasonic sensor to collect the internal deterioration of the samples and transmit them to the control console, thus forming an integrated test bench that simultaneously tests corrosion and fatigue.

[0092] 2. This test bench can perform tests on two properties of hydraulic components. The control system involved in this invention can control constant mechanical force, torsional force, corrosive liquid level and air humidity according to requirements. According to the test requirements, it can realize mechanical fatigue performance testing, corrosion resistance testing, and simultaneous corrosion and mechanical fatigue performance testing of hydraulic components. It realizes corrosion experiments on metal samples in a dry and wet corrosive environment. At the same time, according to the test requirements, electrochemical experiments and / or fatigue tests can be selectively performed, satisfying the test under multiple coupled factors.

[0093] 3. This test bench not only considers variables such as corrosion environment, corrosion time, and fatigue load, but also takes into account that hydraulic components work under conditions of simultaneous action of wet corrosion environment, alternating wet and dry corrosion environment, and fatigue alternating load. It can solve the problem that traditional environmental pre-corrosion + fatigue test and corrosion test + fatigue test cannot effectively reflect the real working environment of hydraulic metal components.

[0094] 4. This test bench can provide more realistic and reliable experimental methods for studying the effects of different currents on the electrode reactions of metal samples under alternating wet and dry conditions, which have important reference and application value. Specifically, it studies the effects of different current magnitudes on the electrode reactions of the same metal sample and the same electrolyte solution under alternating wet and dry conditions, in order to evaluate the influence of current on the electrochemical performance and corrosion behavior of metals, which helps to accurately assess the service safety of hydraulic metal parts.

[0095] 5. The Canny edge detection algorithm used in this invention has many advantages and features. First, it can accurately detect edges in images and minimize false detections and missed detections. Second, through non-maximum suppression and double thresholding, it can reduce the false detection rate, clearly mark strong edges, and determine the authenticity of weak edges through edge connectivity.

[0096] 6. The method for evaluating the deteriorated operating status adopted in this invention combines measured data of metal structures, three-dimensional model construction, deterioration information synchronization, finite element analysis of ANSYS, and reliability assessment. It can comprehensively evaluate the operating status of metal structures and provide a scientific basis for decision-making. Attached Figure Description

[0097] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0098] Figure 1 This is a flowchart of the experimental procedure for metal samples according to the present invention.

[0099] Figure 2 This is a top view of the entire test bench of the present invention.

[0100] Figure 3 This is a front view of the entire test bench of the present invention.

[0101] Figure 4 This is a structural diagram of the corrosion chamber of the present invention.

[0102] Figure 5 This is a schematic diagram of the corrosion fatigue testing device of the present invention.

[0103] Figure 6 This is a schematic diagram showing the maximum and minimum values ​​of the angle of a plane mirror.

[0104] In the diagram: 1 Control console, 2 Constant current instrument, 3 Liquid outlet pipe, 4 Etching chamber, 5 Camera mount, 6 First liquid inlet pipe, 7 Peristaltic pump, 8 Second liquid inlet pipe, 9 Etching liquid tank, 10 Humidifier, 11 Water inlet pipe, 12 Humidifier water tank, 13 Waste liquid collection tank, 14 Thermometer, 15 Industrial camera, 16 Thermocouple, 17 Plane mirror, 18 Liquid level sensor, 19 Reference electrode, 20 Parallel light source, 21 First clamping ring, 22 Hydraulic rod, 23 Auxiliary electrode, 24 Hydraulic unit, 25 Lead screw, 26 Fixing plate, 27 Sample, 28 Sample clamp, 29 Second clamping ring, 30 Liquid outlet valve, 31 Air humidity sensor, 32 Distance sensor, 33 Moving slider, 34 Static slider, 35 Remote ultrasonic thickness gauge. Detailed Implementation

[0105] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0106] Example 1:

[0107] As attached Figures 1-5 As shown, a test bench for accelerated deterioration of steel gates under simultaneous corrosion and fatigue is provided. The test bench includes a corrosion testing device, a fatigue testing device, a data acquisition device, and a control system.

[0108] Furthermore, the corrosion testing apparatus includes a corrosion chamber, a temperature control device, a waste liquid collection device, a liquid replenishment device, an electrochemical testing device, and an air humidity control device.

[0109] Furthermore, the temperature control device is provided with multiple parallel thermocouples 16 spaced along the height of both sides of the corrosion chamber 4, and thermometers 14 are located on both sides of the corrosion chamber 4. The thermocouples 16 and thermometers 14 are communicatively connected to the control console 1 to detect and adjust the temperature inside the corrosion chamber 4. The thermometers 14 can be existing glass tube thermometers, which can avoid being corroded by the liquid inside the corrosion chamber 4. The experimenter can observe the temperature inside the corrosion chamber 4 in real time through the control console 1, and thus adjust the switching of the thermocouples 16 to achieve the purpose of adjusting the temperature inside the corrosion chamber 4.

[0110] Furthermore, the waste liquid collection device includes an outlet valve 30, an outlet pipe 3, and a waste liquid collection tank 13. One end of the outlet pipe 3 is connected to the corrosion tank 4, and the other end is connected to the waste liquid collection tank 13. The outlet valve 30 is connected to the outlet pipe 3 and is communicatively connected to the control console 1. The replenishment device includes a first inlet pipe 6, a second inlet pipe 8, a peristaltic pump 7, a corrosion liquid tank 9, and a liquid level sensor 18. One end of the first inlet pipe 6 is connected to the peristaltic pump 7, and the other end is connected to the corrosion tank 4. One end of the second inlet pipe 8 is connected to the corrosion liquid tank 9, and the other end is connected to the peristaltic pump 7. The peristaltic pump 7 is communicatively connected to the control console 1, and the liquid level sensor 18 is communicatively connected to the control console 1. The experimenter can observe the liquid level in the corrosion tank through the control console 1 and control the flow rate of the peristaltic pump 7 and the opening and closing time of the outlet valve 30 as needed, thereby controlling the time when the sample is completely immersed and alternately wet and dry.

[0111] Furthermore, the electrochemical testing apparatus includes a reference electrode 19, an auxiliary electrode 23, a sample 27, a sample holder 28, and a constant current meter 2. The reference electrode 19, auxiliary electrode 23, and sample 27 constitute a three-electrode system. One end of the reference electrode 19 is installed inside the corrosion chamber 4, and the other end of the reference electrode 19 is electrically connected to the constant current meter 2. The constant current meter 2 is communicatively connected to the control console 1. One end of the auxiliary electrode 23 is installed inside the corrosion chamber 4, and the other end of the auxiliary electrode 23 is electrically connected to the control console 1. Both ends of the sample 27 are connected to the sample holder 28, and the sample holder 28 is electrically connected to the control console 1. It should be noted that the electrochemical corrosion experiment needs to be conducted in an environment where the corrosive medium contains a solution. In this embodiment, the three-electrode system can simultaneously measure and control current and voltage. The control console 1, by monitoring the current and voltage, performs an electrochemical corrosion experiment on the sample 27 being corroded, thus improving the ease of use of the corrosion testing equipment.

[0112] Furthermore, the air humidity control device includes a humidifier 10, an air humidity sensor 31, a water inlet pipe 11, and a water tank 12. One end of the humidifier 10 is connected to the corrosion chamber 4, one end of the water inlet pipe 11 is connected to the humidifier 10, and the other end of the water inlet pipe 11 is connected to the water tank 12. The air humidity sensor 31 is communicatively connected to the control console 1.

[0113] Furthermore, the fatigue testing apparatus includes a torsional stress loading device and a tensile stress loading device;

[0114] Furthermore, the torsional stress loading device includes a sample clamp 28, a first clamping ring 21, a hydraulic actuator 24, and a hydraulic rod 22. The sample clamp 28 is located at both ends of the sample 27. The hydraulic actuator 24 is connected to the hydraulic rod 22, and the hydraulic rod 22 is connected to the first clamping ring 21. The hydraulic actuator 24 is communicatively connected to the control console 1. The hydraulic actuator 24 is a constant mechanical load loading device. The torsional load is applied to the test piece by the extension and retraction of the hydraulic rod 22. Since the sample 27 is fixed on the clamp, it achieves torsional movement under the force of the first clamping ring 21, thereby realizing the torsional load on the sample 27 by the device. It should be noted that there is a sealing ring at the connection between the hydraulic actuator 24 and the bottom plate of the corrosion chamber 4.

[0115] The end of the first clamping ring 21 is configured as a pair of semi-circular arc structures that match the sample 27. The pair of semi-circular arc structures are interlocked and sleeved on the outer wall of the sample 27. The free end of the first clamping ring 21 extends horizontally as a clamping rod. Fastening bolts are provided at corresponding positions of the clamping rods. The first clamping ring 21 is fastened by the bolts. The free end of the clamping rod is connected to the hydraulic rod 22 on the hydraulic device 24. The height of the hydraulic rod 22 on the hydraulic device 24 can be adjusted according to the actual situation.

[0116] Furthermore, the tensile stress loading device includes a second clamping ring 29, a fixed plate 26, a stationary slider 34, a movable slider 33, two lead screws 25, and a distance sensor 32. The second clamping ring 29 is connected to the sample 27, the fixed plate 26 is connected to the groove of the corrosion chamber 4, the movable slider 33 and the stationary slider 34 are connected to the bottom plate of the corrosion chamber 4 and both sides of the fixed plate 26, and the two lead screws 25 are connected to the fixed plate 26 and the second clamping ring 29 respectively through lead screw nuts. The distance sensor 32 is arranged on the fixed plate 26 and is communicatively connected to the control console 1. The control system detects the signal of the distance sensor 32 to determine the magnitude of the constant force loading value.

[0117] By applying a constant load to the specimen 27 in the horizontal direction using a constant strain method, the distance between the fixing plate 26 and the second clamping ring 29 can be adjusted to adjust the magnitude of the constant load applied in the horizontal direction by the device. Changing the magnitude of the constant load makes the test process as close as possible to the true load magnitude of the specimen 27, which meets the requirements for fatigue performance testing of hydraulic metal parts. Moreover, the capability of the test piece of the present invention is not limited by the size of the test model.

[0118] Furthermore, the data acquisition device includes a parallel light source 20, an industrial camera 15, a camera mount 5, a plane mirror 17, a control console 1, a remote ultrasonic thickness gauge 35, and a stress-strain sensor. The industrial camera 15 is connected to the camera mount 5. Part of the parallel light source is connected to the camera mount, and the other part is located on the bottom plate of the corrosion chamber 4. The industrial camera 15 is communicatively connected to the control console 1. The plane mirror 17 is located at the bottom of the corrosion chamber 4 to assist the industrial camera 15 in observing the degradation of the lower surface of the sample 27. The remote ultrasonic thickness gauge 35 is located on the side surface of the corrosion chamber 4 to measure the degradation inside the sample 27. The stress-strain sensor is connected to the surface of the sample 27 and is communicatively connected to the control console 1.

[0119] Example 2:

[0120] The accelerated degradation test conducted using the test bench of the present invention includes the following steps:

[0121] Step 1.1, Pretreatment of the sample:

[0122] Polish the surface of the metal sample with a grinder, then wipe the surface of the metal sample clean and clean it with acetone and blow it dry. Fix the sample in the corrosion chamber with a sample clamp.

[0123] Step 1.2, conduct the experiment:

[0124] Depending on the experimenter's needs, three different testing requirements can be selectively met: corrosion testing, fatigue testing, and corrosion fatigue testing.

[0125] 1.2-1, Corrosion Test:

[0126] First, the corrosive solution was placed into the replenishment tank and the corrosion chamber, respectively. Then, it was confirmed that all required sensors were functioning correctly. When the level sensor detected that the metal sample was not yet completely immersed in the corrosive solution, corrosion tests were performed on the metal sample under humidified air, alternating wet and dry conditions, and complete immersion. The experimental parameters for the inlet valve, outlet valve, and humidifier were set via the control panel, specifically including air humidity and the frequency of wet and dry alternation. Finally, when the level sensor detected that the metal sample was completely immersed in the corrosive solution, an electrochemical corrosion test was performed on the metal sample. One end of the reference electrode was installed inside the corrosion chamber, and the other end of the reference electrode was electrically connected to the galvanometer, which was communicatively connected to the control panel. One end of the auxiliary electrode is installed in the corrosion tank, and the other end of the auxiliary electrode is electrically connected to the control console. Both ends of the test sample are connected to the sample holder, which is electrically connected to the control console. Electrochemical experimental parameters, including test time and scanning potential parameters, are set through the control console with a data acquisition module. In this way, electrochemical curves of metal samples under fully wet and dry-wet alternating environments can be obtained, including open circuit potential test curves, potentiodynamic polarization curves, and AC impedance Nyquist plots. Through data processing, open circuit potential, self-corrosion current density, and self-corrosion potential parameters can be obtained, and the electrochemical corrosion resistance of the metal sample under multi-factor coupled environment can be evaluated.

[0127] 1.2-2, Fatigue test:

[0128] First, one end of the first clamping ring is fixed to one end of the metal sample, and the other end of the clamping ring is connected to the hydraulic rod. The extension and retraction of the hydraulic rod is controlled by the control console to determine the torsional load on the metal sample. Then, one end of the second clamping ring is fixed to the other end of the metal sample, and the lead screw is connected to the other end of the clamping ring through the fixing plate. The rotation of the lead screw is controlled by the control console, and the tensile load on the metal sample is determined by the distance sensor.

[0129] 1.2-3, Corrosion fatigue test:

[0130] First, fix one end of the first clamping ring to one end of the metal sample, and connect the other end of the clamping ring to the hydraulic rod. Fix one end of the second clamping ring to the other end of the metal sample, and connect the lead screw to the other end of the clamping ring through the fixing plate. Then, put the corrosion solution into the replenishment tank and the corrosion tank respectively. Finally, confirm that all the sensors required for the experiment are working properly. At this time, the fatigue test is carried out. By monitoring the liquid level sensor, it is possible to determine the corrosion test and electrochemical test under what environment. The two tests are carried out simultaneously.

[0131] Example 3:

[0132] This invention provides a method for evaluating the operational status of hydraulic metal components after deterioration:

[0133] Step 2.1, Establishing the 3D Model:

[0134] Use modeling tools to create a 3D model of the metal structure, ensuring that the geometry and dimensions of the model are consistent with the actual metal structure;

[0135] Step 2.2, Surface wear and corrosion information collection:

[0136] 2.2-1, Setting the angle of the plane mirror:

[0137] The camera lens is positioned directly above the sample. The distance and angle between the camera lens, the sample, and the plane mirror are determined so that the camera can capture the entire surface of the sample to obtain information on surface wear and corrosion.

[0138] like Figure 6 The figures show the maximum and minimum values ​​of the angle of a plane mirror. Therefore, the range of the angle is:

[0139]

[0140] In the formula: L1 is the distance from the camera lens to the bottom of the sample; L2 is the distance from the bottom of the sample to the bottom of the corrosion chamber; L3 is the distance from the center of the sample to the x-axis of the plane mirror; L4 is the width of the plane mirror; R1 is the radius of the camera lens;

[0141] 2.2-2, Image preprocessing:

[0142] Since the acquired images contain a small amount of background interference and the effect of plane mirror refraction, filtering and denoising methods should be used in the preprocessing stage to remove this interference and crop out the excess background so that the image contains only the metal sample. In addition, enhancement algorithms are used to enhance the details, strengthen the details of the deteriorated parts, highlight the difference between the deteriorated and non-deteriorated areas, thereby improving the detection accuracy and enhancing the image contrast.

[0143] 2.2-3, Measurement of wear and corrosion depth:

[0144] Based on the preprocessed images in 2.2-2, the wear and corrosion depth of the corresponding areas are measured using a remote ultrasonic thickness gauge;

[0145] Step 2.3, Feature Extraction:

[0146] The Canny edge detection algorithm is used to extract preliminary edge information of wear and corrosion. First, the image is converted to grayscale. Then, the Canny edge detection algorithm is used with appropriate parameters to extract the edges of wear and corrosion, resulting in a binary image of wear and corrosion. The specific steps and formulas are as follows:

[0147] 2.3-1, Calculate the gradient:

[0148] Calculate the horizontal gradient of the image: Gx(x,y)=I(x+1,y)-I(x-1,y);

[0149] Calculate the vertical gradient of the image: Gy(x,y)=I(x,y+1)-I(x,y-1);

[0150] Calculate the gradient magnitude: G(x,y) = sqrt(Gx(x,y)) 2 +Gy(x,y) 2 );

[0151] Calculate the gradient direction: θ(x,y)=atan2(Gy(x,y),Gx(x,y));

[0152] In the formula: Gx(x,y) represents the gradient value in the horizontal direction, Gy(x,y) represents the gradient value in the vertical direction, I(x,y) represents the gray value on the original image, and atan2() represents the arctangent function, which is used to calculate the gradient direction;

[0153] 2.3-2, Non-maximum suppression:

[0154] First, for each pixel P(x,y), calculate the positions of the two adjacent pixels of P along the gradient direction. Then, determine whether the gradient magnitude of P is greater than the gradient magnitude of the adjacent pixels. If it is, retain the gradient magnitude of P; otherwise, set the gradient magnitude of P to 0. The formula for calculating the positions of adjacent pixels is as follows:

[0155] (x1,y1)=(x+round(cos(θ(x,y))),y+round(sin(θ(x,y))));

[0156] (x2,y2)=(x-round(cos(θ(x,y))),y-round(sin(θ(x,y))));

[0157] In the formula: (x1, y1) represents the position of the first adjacent pixel along the gradient direction, and (x2, y2) represents the position of the second adjacent pixel along the gradient direction;

[0158] 2.3-3, Dual Threshold Processing:

[0159] For each pixel P(x,y), if G(x,y) is greater than the high threshold (T_high), it is marked as a strong edge; if G(x,y) is less than the low threshold (T_low), it is marked as a non-edge; if G(x,y) is between the low threshold and the high threshold, and is connected to a strong edge, and has strong edge pixels in its 8-neighborhood, it is marked as a weak edge.

[0160] 2.3-4, Edge Connectivity:

[0161] First, for each strong edge pixel, mark it as an edge pixel. Second, for each weak edge pixel P(x,y), check the pixels in its 8-neighborhood. Then, for each pixel Q(a,b) in its 8-neighborhood, if Q(a,b) is a strong edge pixel and has not been marked as an edge pixel, mark Q(a,b) as an edge pixel, and then connect the unmarked edge pixels in the 8-neighborhood of Q(a,b) in turn. Continue processing the next unmarked weak edge pixel, and repeat steps 2.3-2 and 2.3-3 until all weak edge pixels have been processed.

[0162] Step 2.4, Data Labeling and Preparation:

[0163] The collected image data was manually labeled, dividing it into normal areas, worn areas, and corroded areas, where 0 represents normal, 1 represents worn, and 2 represents corroded. The depth data for each type of wear and corrosion was recorded. The labeled image data was then divided into training and testing sets, with 80% of the images used for training and 20% for testing. This ensured that the images in the training and testing sets came from different metal surfaces and were representative. OpenCV was used to augment the images, increasing the number of training samples and enhancing the model's robustness. PyTorch's DataLoader was used to load the training and testing set data. Since PyTorch typically uses integer-encoded category labels, a label encoding dictionary was used to map the category labels to integer codes.

[0164] Step 2.5, Deep Learning Model Training:

[0165] A convolutional neural network (CNN) model is built using PyTorch. The edge features of the input image are used as the input to the model, and the output of the model is the probability distribution of each region category. The CNN model is trained using training set data, and the model weights are optimized through backpropagation and gradient descent to achieve automatic classification of defects such as wear and corrosion.

[0166] Step 2.6, Model Evaluation and Optimization:

[0167] The performance of the trained model is evaluated using test set data, and the accuracy, recall, and F1 score are calculated. Based on the evaluation results, the hyperparameters of the model are adjusted to optimize the model's performance.

[0168] Step 2.7, Defect Identification and Location:

[0169] The trained model is used to predict images and the prediction results are visualized: based on the predicted bounding box information, boxes are drawn on the image to identify defect areas. The bounding boxes are marked with different colors, line widths or styles to highlight the defect areas.

[0170] Step 2.8, Model Optimization and Iteration:

[0171] By providing feedback based on the recognition results, adjusting the model architecture, increasing training data, and trying different data augmentation methods, the performance and generalization ability of the metal surface defect recognition model can be further optimized.

[0172] Step 2.9, Calculation of wear and corrosion degree:

[0173] By calculating the number of pixels in the binary image processed by the neural network in steps 2.2-2.8, the corrosion area can be calculated, and the corrosion area is equal to the number of pixels. Then, by using the remote ultrasonic thickness gauge in steps 2.2-2.3, the corrosion depth of the corresponding area can be calculated, and the length and area of ​​wear, as well as the area and shape characteristics of corrosion can be calculated.

[0174] Step 2.10, Synchronize information on wear and corrosion areas:

[0175] The wear and corrosion information of the metal surface is mapped to the 3D model, and the corresponding areas are marked on the 3D model according to the location, shape and severity of the wear and corrosion.

[0176] Step 2.11, Model Import:

[0177] Import the completed 3D model into ANSYS software, ensuring that the geometric information, material properties, and boundary conditions of the imported model are set correctly.

[0178] Step 2.12, Mechanical Analysis Settings:

[0179] In ANSYS, set the appropriate mechanical analysis type, select the corresponding analysis type according to the actual working environment and loading conditions of the metal structure, including static analysis, dynamic analysis or thermal analysis; set loading conditions, including the force, pressure and temperature applied to the metal structure, to simulate the actual working conditions; set constraint conditions, including fixed supports and constrained displacements, to limit the degree of freedom of the model.

[0180] Step 2.13, Material Property Settings:

[0181] In ANSYS, material properties of metal structures are set, including elastic modulus, Poisson's ratio, and yield strength. Considering the changes in the mechanical properties of the metal structure after deterioration, the material properties are adjusted according to the severity of wear and corrosion information on the metal surface. For areas with different degrees of wear and corrosion, partition settings are made according to their specific material properties.

[0182] Step 14, Simulation of Wear and Corrosion Zones:

[0183] In ANSYS, metal structures with wear and corrosion zones are modeled and simulated. Based on the location, shape, and severity of wear and corrosion, the corresponding areas on the model are modified, and their material properties are set to their corresponding deterioration states. Appropriate meshing techniques are used to mesh the model to ensure accurate modeling of wear and corrosion zones.

[0184] Step 15, Analysis and Judgment:

[0185] After running the mechanical analysis in ANSYS, the stress, strain, and deformation results of the metal structure under actual working conditions are obtained. Based on these parameters, as well as the location and extent of wear and corrosion information, the strength, stability, and reliability of the structure are evaluated. An important evaluation criterion is comparing the relationship between stress and strain and material strength to determine whether the material's yield strength or failure strength has been exceeded. The Mises yield criterion is used to evaluate the structure, and its formula is:

[0186]

[0187] Where: σ eq This refers to the equivalent stress under multiaxial stress, i.e., Mises stress; σ x , σ y , σ z τ represents the normal stress along the x, y, and z axes, respectively; xy , τ yz , τ zx For shear stress, when σ eq When ≤f, f is the design strength value of the steel, that is, the strength condition is met;

[0188] The difference between the load-bearing capacity of a structure and the stress load is defined as the limit state function, as shown in the following formula:

[0189] G(x)=R(x)-S(x)≤0;

[0190] In the formula: R(x) represents the load-bearing capacity of the structure, and S(x) represents the stress load;

[0191] The probability distribution model is chosen using a normal distribution as the input parameter. For each input parameter, including material strength and load magnitude, its mean μ and standard deviation σ are determined. Since the FORM method requires the limit state function to be transformed into a standard normal space for reliability calculation, the inverse transformation of the probability distribution function is used to convert the actual values ​​of the input parameters into standard normal variables. Based on the standard normal variables of the input parameters, the gradient of the limit state function in each direction is calculated, and then the reliability index β is calculated using the inverse normal distribution function. Based on the calculated reliability index β, the reliability of the structure is evaluated. A higher reliability index β indicates that the structure has higher reliability within a given design life.

[0192] In summary, by conducting mechanical analysis, strength assessment, deformation analysis, and reliability assessment of metal structures, we can gain a comprehensive understanding of their performance under wear and corrosion conditions, providing a scientific basis and decision support for the operation and maintenance of the structures.

[0193] Example 4:

[0194] This invention provides an experimental method for studying the effect of different currents on the electrode reactions of metal samples under alternating wet and dry conditions. Specifically, it investigates the influence of different current magnitudes on the electrode reactions of the same metal sample and electrolyte solution under alternating wet and dry conditions to evaluate the effect of current on the electrochemical performance and corrosion behavior of metals. The operation process is as follows: First, ensure that all devices within the corrosion chamber 4 are functioning normally. Prepare the same electrolyte solution, ensuring its concentration and pH value meet experimental requirements. Connect the sample 27 to the electrochemical experimental apparatus, ensuring it is in full contact with the electrolyte solution. Set the parameters of the humidifier 10, peristaltic pump 7, and outlet valve 30 to simulate alternating wet and dry environmental conditions. Then, design three control groups, each using a different current magnitude: Control Group 1: using a smaller current magnitude, e.g., low current density; Control Group 2: using a medium current magnitude, e.g., medium current density; Control Group 3: using a larger current magnitude, e.g., high current density. Under alternating wet and dry environmental conditions, wet and dry cycles are performed at set time intervals. Within each cycle, electrochemical tests are run for the three control groups, and the potential, current, and time data are recorded during the experiment. The potential and current response curves of the three control groups were compared to evaluate the effect of different current magnitudes on the electrochemical performance and corrosion behavior of sample 27. Finally, based on the experimental results, the effects of different current magnitudes on the electrode reactions of sample 27 under alternating wet and dry conditions were summarized. The performance differences among the three control groups were compared, and the effects of current on the corrosion behavior, potential evolution, and electrochemical reaction kinetics of sample 27 were discussed.

Claims

1. A method for evaluating the operational status of deteriorated hydraulic metal parts, wherein the evaluation method is implemented using a steel gate accelerated deterioration test bench subjected to simultaneous corrosion and fatigue, the test bench comprising a corrosion test device, a fatigue test device, a data acquisition device, and a control system; The corrosion test device includes a corrosion chamber (4), a temperature control device, a waste liquid collection device, a liquid replenishment device, an electrochemical test device, and an air humidity control device. Through a liquid level sensor (18), a thermometer (14), and an air humidity sensor (31), the control console (1) can perform accelerated corrosion tests on the sample. The fatigue testing apparatus includes a torsional stress loading device and a tensile stress loading device, which apply a constant force and a torsional load to the specimen in the horizontal direction through a constant mechanical load loading device. The data acquisition device includes a parallel light source (20), an industrial camera (15), a camera mount (5), a plane mirror (17), a control console (1), a remote ultrasonic thickness gauge (35), and a stress-strain sensor. The industrial camera (15) is used to collect information on the deterioration of the sample surface. Its features are, The evaluation method first uses a steel gate accelerated deterioration test bench with simultaneous corrosion and fatigue to conduct accelerated deterioration tests on hydraulic metal parts, and then evaluates the operating status of the hydraulic metal parts after deterioration. The accelerated degradation test includes the following steps: Step 1.1, Pretreatment of the sample: Polish the surface of the metal sample with a grinder, then wipe the surface of the metal sample clean and clean it with acetone and blow it dry. Fix the sample in the corrosion chamber with a sample clamp. Step 1.2, conduct the experiment: Depending on the experimenter's needs, three different testing requirements can be selectively met: corrosion testing, fatigue testing, and corrosion fatigue testing. 1.2-1, Corrosion Test: First, the corrosive solution was placed into the replenishment tank and the corrosion chamber, respectively. Then, it was confirmed that all required sensors were functioning correctly. When the level sensor detected that the metal sample was not yet completely immersed in the corrosive solution, corrosion tests were performed on the metal sample under humidified air, alternating wet and dry conditions, and complete immersion. The experimental parameters for the inlet valve, outlet valve, and humidifier were set via the control panel, specifically including air humidity and the frequency of wet and dry alternation. Finally, when the level sensor detected that the metal sample was completely immersed in the corrosive solution, an electrochemical corrosion test was performed on the metal sample. One end of the reference electrode was installed inside the corrosion chamber, and the other end of the reference electrode was electrically connected to the galvanometer, which was communicatively connected to the control panel. One end of the auxiliary electrode is installed in the corrosion tank, and the other end of the auxiliary electrode is electrically connected to the control console. Both ends of the test sample are connected to the sample holder, which is electrically connected to the control console. Electrochemical experimental parameters, including test time and scanning potential parameters, are set through the control console with a data acquisition module. In this way, electrochemical curves of metal samples under fully wet and dry-wet alternating environments can be obtained, including open circuit potential test curves, potentiodynamic polarization curves, and AC impedance Nyquist plots. Through data processing, open circuit potential, self-corrosion current density, and self-corrosion potential parameters can be obtained, and the electrochemical corrosion resistance of the metal sample under multi-factor coupled environment can be evaluated. 1.2-2, Fatigue test: First, one end of the first clamping ring is fixed to one end of the metal sample, and the other end of the clamping ring is connected to the hydraulic rod. The extension and retraction of the hydraulic rod is controlled by the control console to determine the torsional load on the metal sample. Then, one end of the second clamping ring is fixed to the other end of the metal sample, and the lead screw is connected to the other end of the clamping ring through the fixing plate. The rotation of the lead screw is controlled by the control console, and the tensile load on the metal sample is determined by the distance sensor. 1.2-3, Corrosion fatigue test: First, fix one end of the first clamping ring to one end of the metal sample, and connect the other end of the clamping ring to the hydraulic rod. Fix one end of the second clamping ring to the other end of the metal sample, and connect the lead screw to the other end of the clamping ring through the fixing plate. Then, put the corrosion solution into the replenishment tank and the corrosion tank respectively. Finally, confirm that all the sensors required for the experiment are working properly. At this time, the fatigue test is carried out. By monitoring the liquid level sensor, it is possible to determine the corrosion test and electrochemical test under what environment. The two tests are carried out simultaneously. The method for assessing the deteriorated operating status includes the following steps: Step 2.1, Establishing the 3D Model: Use modeling tools to create a 3D model of the metal structure, ensuring that the geometry and dimensions of the model are consistent with the actual metal structure; Step 2.2, Surface wear and corrosion information collection: 2.2-1, Setting the angle of the plane mirror: The camera lens is positioned directly above the sample. The distance and angle between the camera lens, the sample, and the plane mirror are determined so that the camera can capture the entire surface of the sample to obtain information on surface wear and corrosion. The range of plane mirror angles is: ； In the formula: L 1 represents the distance from the camera lens to the bottom of the sample; L 2 represents the distance from the bottom of the sample to the bottom of the corrosion chamber; L 3 represents the distance from the center of the sample to the x-axis of the plane mirror; L 4 represents the width of the plane mirror; R 1 represents the radius of the camera lens; 2.2-2, Image preprocessing: Since the acquired images contain a small amount of background interference and the effect of plane mirror refraction, filtering and denoising methods should be used in the preprocessing stage to remove this interference and crop out the excess background so that the image contains only the metal sample. In addition, enhancement algorithms are used to enhance the details, strengthen the details of the deteriorated parts, highlight the difference between the deteriorated and non-deteriorated areas, thereby improving the detection accuracy and enhancing the image contrast. 2.2-3, Measurement of wear and corrosion depth: Based on the preprocessed images in 2.2-2, the wear and corrosion depth of the corresponding areas are measured using a remote ultrasonic thickness gauge; Step 2.3, Feature Extraction: The Canny edge detection algorithm is used to extract preliminary edge information of wear and corrosion. First, the image is converted to grayscale. Then, the Canny edge detection algorithm is used with appropriate parameters to extract the edges of wear and corrosion, resulting in a binary image of wear and corrosion. The specific steps and formulas are as follows: 2.3-1, Calculate the gradient: Calculate the horizontal gradient of the image: ; Calculate the vertical gradient of the image: ; Calculate the gradient magnitude: ; Calculate the gradient direction: ; In the formula: This represents the gradient value in the horizontal direction. This represents the gradient value in the vertical direction. Represents the grayscale value on the original image. This represents the arctangent function, used to calculate the gradient direction; 2.3-2, Non-maximum suppression: First, for each pixel P(x, y), calculate the positions of the two adjacent pixels of P along the gradient direction. Then, determine whether the gradient magnitude of P is greater than the gradient magnitude of the adjacent pixels. If it is, retain the gradient magnitude of P; otherwise, set the gradient magnitude of P to 0. The formula for calculating the positions of adjacent pixels is as follows: ； ； In the formula: This indicates the position of the first adjacent pixel along the gradient direction. This indicates the position of the second adjacent pixel along the gradient direction; 2.3-3, Dual Threshold Processing: For each pixel ,if Greater than the high threshold Mark it as a strong edge; if Less than the low threshold Mark it as non-edge; if If a pixel is between a low threshold and a high threshold, and is connected to a strong edge, and has a strong edge pixel within an 8-neighborhood, it is marked as a weak edge. 2.3-4, Edge Connectivity: First, for each strong edge pixel, mark it as an edge pixel; second, for each weak edge pixel... Examine the pixels within its 8-neighborhood; then for each pixel within its 8-neighborhood... ,if It is a strong edge pixel and has not been marked as an edge pixel. Mark the pixels as edge pixels, and then sequentially... Connect the unlabeled edge pixels within the 8-neighborhood; continue processing the next unlabeled weak edge pixel, repeating steps 2.3-2 and 2.3-3 until all weak edge pixels have been processed; Step 2.4, Data Labeling and Preparation: The collected image data was manually labeled, dividing it into normal areas, worn areas, and corroded areas, where 0 represents normal, 1 represents worn, and 2 represents corroded. The depth data for each type of wear and corrosion was recorded. The labeled image data was then divided into training and testing sets, with 80% of the images used for training and 20% for testing. This ensured that the images in the training and testing sets came from different metal surfaces and were representative. OpenCV was used to augment the images, increasing the number of training samples and enhancing the model's robustness. PyTorch's DataLoader was used to load the training and testing set data. Since PyTorch typically uses integer-encoded category labels, a label encoding dictionary was used to map the category labels to integer codes. Step 2.5, Deep Learning Model Training: A convolutional neural network (CNN) model is built using PyTorch. The edge features of the input image are used as the input to the model, and the output of the model is the probability distribution of each region category. The CNN model is trained using training set data, and the model weights are optimized through backpropagation and gradient descent to achieve automatic classification of wear and corrosion defects. Step 2.6, Model Evaluation and Optimization: The performance of the trained model is evaluated using test set data, and the accuracy, recall, and F1 score are calculated. Based on the evaluation results, the hyperparameters of the model are adjusted to optimize the model's performance. Step 2.7, Defect Identification and Location: The trained model is used to predict images and the prediction results are visualized: based on the predicted bounding box information, boxes are drawn on the image to identify defect areas. The bounding boxes are marked with different colors, line widths or styles to highlight the defect areas. Step 2.8, Model Optimization and Iteration: By providing feedback based on the recognition results, adjusting the model architecture, increasing training data, and trying different data augmentation methods, the performance and generalization ability of the metal surface defect recognition model can be further optimized. Step 2.9, Calculation of wear and corrosion degree: By calculating the number of pixels in the binary image processed by the neural network in steps 2.2-2.8, the corrosion area can be calculated, and the corrosion area is equal to the number of pixels. Then, by using the remote ultrasonic thickness gauge in steps 2.2-2.3, the corrosion depth of the corresponding area can be calculated, and the length and area of ​​wear, as well as the area and shape characteristics of corrosion, can be calculated. Step 2.10, Synchronize information on wear and corrosion areas: The wear and corrosion information of the metal surface is mapped to the 3D model, and the corresponding areas are marked on the 3D model according to the location, shape and severity of the wear and corrosion. Step 2.11, Model Import: Import the completed 3D model into ANSYS software, ensuring that the geometric information, material properties, and boundary conditions of the imported model are set correctly. Step 2.12, Mechanical Analysis Settings: In ANSYS, set the appropriate mechanical analysis type, select the corresponding analysis type according to the actual working environment and loading conditions of the metal structure, including static analysis, dynamic analysis or thermal analysis; set loading conditions, including the force, pressure and temperature applied to the metal structure, to simulate the actual working conditions; set constraint conditions, including fixed supports and constrained displacements, to limit the degree of freedom of the model. Step 2.13, Material Property Settings: In ANSYS, material properties of metal structures are set, including elastic modulus, Poisson's ratio, and yield strength. Considering the changes in the mechanical properties of the metal structure after deterioration, the material properties are adjusted according to the severity of wear and corrosion information on the metal surface. For areas with different degrees of wear and corrosion, partition settings are made according to their specific material properties. Step 14, Simulation of Wear and Corrosion Zones: In ANSYS, metal structures with wear and corrosion zones are modeled and simulated. Based on the location, shape, and severity of wear and corrosion, the corresponding areas on the model are modified, and their material properties are set to their corresponding deterioration states. Appropriate meshing techniques are used to mesh the model to ensure accurate modeling of wear and corrosion zones. Step 15, Analysis and Judgment: After running the mechanical analysis in ANSYS, the stress, strain, and deformation results of the metal structure under actual working conditions are obtained. Based on these parameters, as well as the location and extent of wear and corrosion information, the strength, stability, and reliability of the structure are evaluated. An important evaluation criterion is comparing the relationship between stress and strain and material strength to determine whether the material's yield strength or failure strength has been exceeded. The Mises yield criterion is used to evaluate the structure, and its formula is: ； In the formula: This is the equivalent stress under multiaxial stress, i.e., Mises stress; , , These are the normal stresses in the x, y, and z axes, respectively; , , For shear stress, when When f is the design strength value of the steel, that is, the strength condition is met; The difference between the load-bearing capacity of a structure and the stress load is defined as the limit state function, as shown in the following formula: ； In the formula: Indicates the load-bearing capacity of the structure. Indicates stress load; The probability distribution model is chosen using a normal distribution as the input parameter. For each input parameter, including material strength and load magnitude, its mean is determined. and standard deviation Since the FORM method requires transforming the limit state function into a standard normal space for reliability calculation, the inverse transformation of the probability distribution function is used to convert the actual values ​​of the input parameters into standard normal variables. Based on the standard normal variables of the input parameters, the gradients of the limit state function in each direction are calculated, and then the reliability index is calculated using the inverse normal distribution function. Based on the calculated reliability index To conduct a reliability assessment of the structure, a higher reliability index is required. This indicates that the structure has high reliability within a given design life.

2. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The temperature control device is located on both sides of the corrosion chamber (4) with multiple parallel thermocouples (16) spaced at intervals along its height direction. The thermometers (14) are located on both sides of the corrosion chamber (4). The thermocouples (16) and the thermometers (14) are connected to the control console (1) for communication to detect and adjust the temperature inside the corrosion chamber (4). The corrosion chamber (4) is made of transparent corrosion-resistant material.

3. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The waste liquid collection device includes a liquid outlet valve (30), a liquid outlet pipeline (3) and a waste liquid collection tank (13). One end of the liquid outlet pipeline (3) is connected to the corrosion tank (4) and the other end is connected to the waste liquid collection tank (13). The liquid outlet valve (30) is connected to the liquid outlet pipeline (3) and is communicatively connected to the control console (1).

4. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The replenishment device includes a first inlet pipe (6), a second inlet pipe (8), a peristaltic pump (7), a corrosion tank (9), and a level sensor (18). One end of the first inlet pipe (6) is connected to the peristaltic pump (7), and the other end is connected to the corrosion tank (4). One end of the second inlet pipe (8) is connected to the corrosion tank (9), and the other end is connected to the peristaltic pump (7). The peristaltic pump (7) is connected to the control console (1), and the level sensor (18) is connected to the control console (1).

5. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The electrochemical test apparatus includes a reference electrode (19), an auxiliary electrode (23), a sample (27), a sample holder (28), and a constant current device (2). The reference electrode (19), the auxiliary electrode (23), and the sample (27) constitute a three-electrode system. One end of the reference electrode (19) is installed in the corrosion chamber (4), and the other end of the reference electrode (19) is electrically connected to the constant current device (2). The constant current device (2) is communicatively connected to the control console (1). One end of the auxiliary electrode (23) is installed in the corrosion chamber (4), and the other end of the auxiliary electrode (23) is electrically connected to the control console (1). Both ends of the sample (27) are connected to the sample holder (28), and the sample holder (28) is electrically connected to the control console (1).

6. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The air humidity control device includes a humidifier (10), an air humidity sensor (31), a water inlet pipe (11), and a water tank (12). One end of the humidifier (10) is connected to the corrosion chamber (4), one end of the water inlet pipe (11) is connected to the humidifier (10), and the other end of the water inlet pipe (11) is connected to the water tank (12). The air humidity sensor (31) is connected to the control console (1).

7. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The torsional stress loading device includes a sample clamp (28), a first clamping ring (21), a hydraulic actuator (24), and a hydraulic rod (22). The sample clamp (28) is located at both ends of the sample (27). The hydraulic actuator (24) is connected to the hydraulic rod (22), and the hydraulic rod (22) is connected to the first clamping ring (21). The hydraulic actuator (24) is connected to the control console (1). The hydraulic actuator (24) is a constant mechanical load loading device. The torsional load is applied to the test piece by the extension and retraction of the hydraulic rod (22). Since the sample (27) is fixed on the clamp, it achieves torsional movement under the force of the first clamping ring (21), thereby realizing the torsional load on the sample (27) by the device. There is a sealing ring at the connection between the hydraulic actuator (24) and the bottom plate of the corrosion chamber (4). The end of the first clamping ring (21) is configured as a pair of semi-circular arc structures that match the sample (27). The pair of semi-circular arc structures are fitted together and sleeved on the outer wall of the sample (27). The free end of the first clamping ring (21) extends horizontally as a clamping rod. Fastening bolts are provided at corresponding positions of the clamping rods. The first clamping ring (21) is fastened by the bolts. The free end of the clamping rod is connected to the hydraulic rod (22) on the hydraulic device (24). The height of the hydraulic rod (22) on the hydraulic device (24) can be adjusted according to the actual situation.

8. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The tensile stress loading device includes a second clamping ring (29), a fixed plate (26), a stationary slider (34), a moving slider (33), a lead screw (25), and a distance sensor (32). The second clamping ring (29) is connected to the sample (27), the fixed plate (26) is connected to the groove of the corrosion chamber (4), the moving slider (33) and the stationary slider (34) are connected to the bottom plate of the corrosion chamber (4) and the two sides of the fixed plate (26), the two lead screws (25) are connected to the fixed plate (26) and the second clamping ring (29) respectively through lead screw nuts, and the distance sensor (32) is arranged on the fixed plate (26). The distance sensor (32) is connected to the control console (1) for communication. The control system detects the signal of the distance sensor (32) to determine the magnitude of the constant force loading value.

9. The method for evaluating the operational status of deteriorated hydraulic metal parts according to claim 1, characterized in that: The data acquisition device includes a parallel light source (20), an industrial camera (15), a camera mount (5), a plane mirror (17), a control console (1), a remote ultrasonic thickness gauge (35), and a stress-strain sensor. The industrial camera (15) is connected to the camera mount (5). One part of the parallel light source (20) is connected to the camera mount (5), and the other part is located on the bottom plate of the corrosion chamber (4). The industrial camera (15) is connected to the control console (1). The plane mirror (17) is located at the bottom of the corrosion chamber (4) to assist the industrial camera (15) in observing the deterioration of the lower surface of the sample (27). The remote ultrasonic thickness gauge (35) is located on the side surface of the corrosion chamber (4) to measure the deterioration inside the sample (27). The stress-strain sensor is connected to the surface of the sample (27) and is connected to the control console (1).

Images