A chloride ion corrosion test method coupled with wind load and seawater splash tide

Abstract

The present invention relates to a chloride ion corrosion test method coupled with wind load and seawater splash tide, comprising a test chamber, salt water, test components, and a wave-making assembly arranged in the test chamber; obtaining the daily average value of historical meteorological elements in the test area; the wave-making assembly comprises a fan and a second lifting assembly, the second lifting assembly driving the fan to maintain a constant height between the fan and the salt water level under normal conditions; the wind blown by the fan drives the salt water in the test chamber to fluctuate and simulate ocean waves, and the change of the water level is used to achieve the beating of the test components to form ocean wave impact areas, full immersion areas, and splash areas, etc., which can truly simulate actual working conditions; the salt water level in the test chamber can be controlled by a water pump, achieving dry-wet cycles for the test components, and achieving the purpose of accelerating the corrosion of the test components. At the same time, the simulated ocean waves blown by the wind will change along with the change of the water level, which can truly reflect the actual situation.

Description

Technical Field

[0001] The present invention relates to the technical field of test equipment, and in particular to a chloride ion corrosion test method coupled with wind load and seawater splash tide. Background Art

[0002] The ocean covers over two-thirds of the Earth's surface, and human exploration and utilization of the ocean are steadily deepening. However, the marine environment is also highly corrosive and hazardous, making various materials susceptible to corrosion and degradation. Offshore wind turbine towers are subjected to long-term chloride ion corrosion and dynamic load coupling in marine service environments, leading to material performance degradation, structural corrosion fatigue damage, and even failure, seriously compromising their service safety.

[0003] Currently, marine corrosion testing of materials is primarily conducted at marine corrosion testing stations established near coastlines. However, offshore sampling requires significant manpower, material, and financial resources, and has drawbacks such as long testing cycles, high data dispersion, and inconvenience in real-time observation and testing.

[0004] The strength of the sea breeze is a key factor in the erosion of offshore wind turbine towers. The greater the wind speed and the longer it lasts, the higher the wave height. Furthermore, in shallower waters, waves are restricted by the seabed topography, increasing their height.

[0005] For example, a Chinese patent published on March 19, 2021, with patent publication number CN112525813A, discloses a marine corrosion environment simulation test device that can simulate different marine corrosion environment zones, study the corrosion laws and corrosion mechanisms of materials in different marine corrosion environments, explore the interaction between the marine atmosphere and seawater interface environment, seawater, and sea mud interface materials, and effectively analyze the corrosion mechanism, corrosion resistance and mechanical strength change law and failure mechanism of different materials in different marine environment corrosion zones.

[0006] However, this existing technology uses a push-plate wave maker to generate waves, which cannot effectively reflect the impact of different water depths on wave size. At the same time, the working environment faced by offshore wind turbine towers in different locations is different. Existing simulation experiments cannot truly reflect the erosion of offshore wind turbine towers in a specific location over a certain period of time. Summary of the Invention

[0007] In view of the deficiencies in the prior art, the present invention provides a chloride ion corrosion test method coupled with wind load and seawater splash tide to solve the above problems.

[0008] The present invention provides the following technical solutions:

[0009] A chloride ion corrosion test method includes a test chamber, salt water arranged in the test chamber, a test component, and a wave-making component;

[0010] Obtain the daily average values ​​of historical meteorological elements in the experimental area and simulate one day's conditions in the experimental box with time interval T;

[0011] The wave-making assembly includes a fan and a second lifting assembly, wherein the second lifting assembly drives the fan to maintain a constant height relative to the brine water level under normal conditions;

[0012] The mathematical expression of the final wind speed of the fan is:

[0013] V=V a +V b +V C +V n

[0014] Where V is the final wind speed, V a is the basic wind speed, V b For gusts, V C For gradient style, V n is the random wind volume, basic wind speed V a Take the average wind speed of the day to be simulated;

[0015] The Rafale V b The mathematical modeling formula is:

[0016]

[0017] Among them, V bmax Take 5% of the average wind speed of a certain day to be simulated, t is the length of time in minutes within one hour, and the cycle is repeated once every 1 hour.

[0018] The gradient wind V c The mathematical modeling formula is:

[0019]

[0020] Among them, V Cmax It is 3% of the average wind speed on a certain day to be simulated, V is the length of time in minutes within one hour, and the cycle is repeated once every 1 hour.

[0021] The random wind V n The mathematical modeling formula is:

[0022]

[0023]

[0024] ω i =(i-0.5)Δω

[0025] Among them, ω iis the angular frequency of the i-th random component; Δω is the discrete spacing of the random component, which is 1 rad / s. is a random variable uniformly distributed on [0,2π]; S V (ω i ) is the amplitude of the i-th random component; K N is the surface resistance coefficient or plane expansion coefficient of the wind field in the experimental area to be simulated on a certain day; F is the wind speed fluctuation scale coefficient or turbulence scale factor in the experimental area to be simulated on a certain day; μ is the average wind speed at a height of 10m above sea level in the experimental area to be simulated on a certain day; n is the number of random components, which is 50; when V n When it is greater than 5% of the average wind speed of a certain day in the simulated experimental area, take V n =0.05V a .

[0026] Preferably, it also includes a first lifting component and a pressure plate, the first lifting component is lifted and lowered to drive the pressure plate at its output end to press the experimental component or separate it; the output end of the first lifting component or the pressure plate is provided with a pressure sensor, and the pressure sensor is used to monitor the load size of the experimental component in real time.

[0027] Preferably, the first lifting assembly and the pressure plate cooperate to apply a simple harmonic load to the experimental component, and the mathematical expression of the simple harmonic load is:

[0028]

[0029] Among them, F mean is the average load, F amp is the load amplitude, f is the frequency, is the phase angle, and t is the time.

[0030] Preferably, a chloride ion solubility monitor, a temperature monitor and a heating component are installed in the test box.

[0031] Preferably, the method further comprises a water tank and a pipe connecting the water tank and the test box, wherein a water pump is installed on the pipe, and the water pump drives the salt water to flow bidirectionally between the water tank and the test box through the pipe.

[0032] Preferably, the time T is 1 hour.

[0033] Preferably, the brine water level h in the test box is:

[0034] h=20+10k

[0035]

[0036] Where h is the height of the salt water level, the water level cycle period is one hour, and t is the duration in minutes within the one-hour cycle.

[0037] Preferably, the second lifting assembly drives the lowest point of the fan blades to be adjusted below the surface of the salt water, and the mechanical force of the fan blades scrapes away water to create a salt spray environment.

[0038] Preferably, the lowest point of the fan blades is adjusted to be less than 3 mm from the brine water surface.

[0039] Preferably, the method comprises the following steps:

[0040] Based on neural network technology, a deep neural network wave height prediction model is established;

[0041] Collect ocean wave height model datasets;

[0042] Clean the data needed for the wave height model to obtain the data set for input into the wave height model. Based on the salt water level and wind speed, determine the validity of the data, remove invalid values ​​from the data, and match the salt water level and wind speed with the effective wave height.

[0043] 80% of the dataset is divided into a training set for training the model, and the remaining 20% ​​of the dataset is used as a test set to evaluate the model performance;

[0044] There is a significant causal relationship between wave height, water level and wind speed input. A wave height prediction model is constructed based on a deep neural network model.

[0045] Evaluate the model and analyze whether the wave height prediction model can predict the real data well;

[0046] The significant wave height H is predicted based on the wave height prediction model, which is approximately

[0047]

[0048] Where D is the lowest point of the fan blades, and h is the height of the brine water level; is the predicted value of the ocean waves.

[0049] The present invention has the following beneficial technical effects:

[0050] The wind from the fan drives the salt water in the test chamber to fluctuate and simulate ocean waves, and the change of water level can hit the experimental components to form ocean wave impact zone, full immersion zone and splash zone, which can truly simulate actual working conditions.

[0051] The water pump can control the salt water level in the test chamber, achieving dry-wet cycles for the test components and accelerating their corrosion. Simultaneously, the simulated waves created by the wind will change with the water level, truly reflecting real-world conditions.

[0052] The first lifting assembly and the pressure plate are set to apply pressure / dynamic load to the experimental component to simulate the load borne by the experimental component and further fit the actual working conditions.

[0053] The temperature monitor and heating component are used to simulate the alternation of spring, summer, autumn and winter seasons.

[0054] The use of dynamic loads, temperature and humidity, and simulated seawater splash coupling can well simulate the working conditions of marine engineering in the atmosphere and ocean, providing theoretical support for the engineering application of new composite structure offshore wind power towers, which meets the engineering requirements of high strength, high toughness and high durability of my country's offshore wind power support structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention;

[0056] Figure 2 It is a front cross-sectional view of the present invention;

[0057] Figure 3 This is a schematic diagram of placing experimental components in the test box of the present invention;

[0058] Figure 4 It is a schematic diagram of the deep neural network structure of the present invention;

[0059] Figure 5 This is a flow chart of the present invention for constructing a machine learning model;

[0060] Figure 6 It is the ocean wave waveform diagram of the present invention.

[0061] The reference numerals in the figures are:

[0062] 1. Test chamber; 5. First lifting assembly; 6. Pressure plate; 7. Pressure sensor; 8. Height sensor; 9. Second lifting assembly; 10. Fan; 11. Blades; 12. Experimental components; 14. Pipes; 15. Intelligent water pump console; 16. Water tank inlet; 17. Water pump; 18. Bottom plate; 19. Water tank; 21. Temperature monitor; 22. Chloride ion solubility monitor; 23. Heating assembly; 24. Computer; 25. Fixed component base. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0064] Example 1:

[0065] like Figure 1-Figure 3 As shown, the test box 1 and the water tank 19 are fixed to the top wall of the base plate 18. The test box 1 and the water tank 19 are interconnected by a pipe 14. A water pump 17 is installed on the pipe 14. The water pump 17 can be a metering pump. The water pump 17 is controlled by the intelligent water pump central console 15. The intelligent water pump central console 15 controls the water pump 17 to deliver forward or reverse water, thereby adjusting the water level of the brine in the test box 1. A liquid level sensor is also included, which plays an auxiliary role by monitoring the water level of the brine in the test box 1 in real time. The water pump 17 is used to implement a dry-wet cycle for the experimental component 12, with 12 hours of drying and 12 hours of immersion. The water level in the test box 1, the chloride ion solubility in the test box 1, and the pH value of the solution are adjusted according to the test conditions to achieve the purpose of accelerating the corrosion of the experimental component 12.

[0066] The first lifting assembly 5 can be installed on the top of the test box 1 using a hydraulic cylinder. A pressure plate 6 is installed at the downward output end of the first lifting assembly 5. The first lifting assembly 5 drives the pressure plate 6 to move vertically. A pressure sensor 7 is installed between the output end of the first lifting assembly 5 and the pressure plate 6, or a pressure sensor 7 is installed on the bottom plate of the pressure plate 6. The pressure sensor 7 is used to monitor the load size / load information of the experimental component 12 in real time.

[0067] The height sensor 8 can be fixed on the side wall of the pressing plate 6 by magnetic attraction using a magnetic infrared sensing device. The height sensor 8 is used to monitor the height position of the pressing plate 6 in real time.

[0068] The wave-making assembly includes a fan 10 and a second lifting assembly 9; the second lifting assembly 9 can be installed on the top wall of the test box 1 using an electric push rod, and the fan 10 is installed at the output end of the second lifting assembly 9. The extension and contraction of the second lifting assembly 9 drives the fan 10 to rise and fall vertically. The fan 10 is set horizontally to blow horizontal wind force to create waves on the pool of the test box 1, thereby achieving the purpose of simulating waves scouring the experimental component 12;

[0069] The heating component 23 uses a heating rod installed in the test box 1 to heat the water pool. The test box 1 is also equipped with a temperature monitor 21 to measure the temperature of the water pool. The two cooperate to control the water temperature to simulate the change of spring, summer, autumn and winter seasons within one day.

[0070] A chloride ion solubility monitor 22 is also installed in the test box 1 for real-time monitoring of the chloride ion solubility of the solution in the pool.

[0071] The water tank 19 can be replenished with salt water through the water tank inlet 16 , and a chloride ion solubility monitor 22 can also be provided in the water tank 19 .

[0072] The fixed component base 25 is detachably fixed to the inner wall of the test box 1 by bolts. A plurality of through slots are horizontally provided on the fixed component base 25. The fixed component base 25 limits the horizontal freedom of the experimental component 12 through the through slots. The fixed component base 25 can be replaced according to the different shapes of the experimental component 12.

[0073] The test box 1, the pressing plate 6, the bottom plate 18, the water tank 19 and the fixing member base 25 are made of stainless steel.

[0074] Working principle:

[0075] The experimental component 12 is vertically inserted into the through slot of the fixed component base 25, and the first lifting assembly 5 drives the pressure plate 6 to press down on the top wall of the experimental component 12, and the pressure sensor 7 is used to monitor the load size / load information of the experimental component 12 in real time.

[0076] The wind blown by the fan 10 causes the salt water in the test box 1 to generate waves simulating ocean waves.

[0077] At the same time, pressure is applied to the experimental component 12 to simulate ocean waves, water temperature changes, and water level changes to achieve dry-wet cycles, which well simulate the real working conditions of marine engineering under the atmosphere and ocean.

[0078] By adjusting the chloride ion solubility and pH value of the solution in the water tank 19 / test box 1, the purpose of accelerating the corrosion of the test component 12 can be achieved. The salt water is a 5% sodium chloride solution, and the pH value can be adjusted to acidic, neutral or alkaline according to the test conditions.

[0079] The temperature monitor 21 is used to monitor the water temperature in the test pool in real time. It can be used in conjunction with the heating component 23 (heating rod) to control the water temperature to simulate the alternation of spring, summer, autumn and winter within a single day. For example, the seawater temperature in a certain area is generally 15-20°C in spring, 20-28°C in summer, 18-23°C in autumn, and 12°C in winter. The water temperature can be set to a 24-hour cycle of 6 hours at 17°C, 6 hours at 24°C, 6 hours at 20°C, and 6 hours at 12°C.

[0080] Example 2: includes all the contents of Example 1, except that:

[0081] The first lifting assembly 5 and the pressing plate 6 cooperate to apply a simple harmonic load to the experimental component, so that the experimental component 12 is subjected to a sinusoidal load. The mathematical expression is:

[0082]

[0083] Among them, F mean is the average load, F amp is the load amplitude, f is the frequency, is the phase angle, and t is the time. These parameters can be adjusted to simulate different alternating load conditions according to actual needs.

[0084] Example 3: Contains all the contents of Example 1 / 2, except that:

[0085] The test chamber 1 further includes a computer 24, which is electrically connected to the liquid level sensor, the pressure sensor 7, the fan 10, the second lifting assembly 9, the height sensor 8, the water pump 17, the temperature monitor 21, the chloride ion solubility monitor 22 and the heating assembly 23. The computer 24 can control the salt water height and water temperature in the test chamber 1, the load borne by the experimental component 12, and the speed / height of the waves through the built-in program of the computer 24.

[0086] It also includes an intelligent water pump console 15 , and the computer 24 controls the water pump 17 through the intelligent water pump console 15 .

[0087] Example 4: includes all the contents of Example 3, except that:

[0088] Obtain historical meteorological elements for each day of the experimental area for a consecutive year. These include average values ​​for temperature, wind speed, temperature, and humidity, as well as the wind field surface resistance coefficient or planar expansion coefficient, and the wind speed fluctuation scale coefficient or turbulence scale factor. For example, if the wind speed is used directly instead of the average value, the amount of data to be processed will be very large because the wind speed changes every moment.

[0089] To speed up the test process, the actual day's conditions were simulated for one hour, and the wind turbine 10 was programmed by computer programming to set the wind speed within one hour and the random change. The wind speed change cycle was one hour.

[0090] Final wind speed = basic wind speed + gust + gradual wind + random wind volume

[0091] V=V a +V b +V c +V n

[0092] V is the final wind speed, V a is the basic wind speed, V b For gusts, V C For gradient style, V n is the random air volume.

[0093] Basic wind speed V a Take the average wind speed for the day to be simulated.

[0094] Rafale V b, which is mainly used to reflect the sudden change of wind speed in a short period of time in natural conditions; gusts can also be used to simulate the emergency response of offshore platforms when wind speed changes suddenly or when strong winds strike. The mathematical modeling formula for gusts is:

[0095]

[0096] V b For gusts, V bmax Take 5% of the average wind speed of a certain day to be simulated, t is the length of time in minutes within an hour (for example, 10 is the 10th minute in 60 minutes), and the cycle is repeated for 1 hour.

[0097] Gradual wind is mainly used to reflect the gradual change factor in natural wind speed. The mathematical modeling formula of the gradual wind component is:

[0098]

[0099] V c For gradient style, V Cmax It is 3% of the average wind speed of a certain day to be simulated, t is the length of time in minutes within an hour (for example, 20 is the 20th minute in 60 minutes), and the cycle is repeated once every 1 hour.

[0100] Random wind is mainly used to reflect the randomness of natural wind speed and is the most important component in wind speed dynamic simulation modeling. Usually, the wind field surface resistance coefficient and wind speed fluctuation scale coefficient need to be introduced. The mathematical modeling formula is:

[0101]

[0102]

[0103] ω i =(i-0.5)Δω

[0104] ω i is the angular frequency of the i-th random component; Δω is the discrete spacing of the random component, which is 1 rad / s, ψ i is a random variable uniformly distributed on [0,2π]; random changes are achieved through computer program design, S V (ω i ) is the amplitude of the i-th random component; K N is the surface resistance coefficient or plane expansion coefficient of the wind field in the experimental area to be simulated on a certain day; F is the wind speed fluctuation scale coefficient or turbulence scale factor in the experimental area to be simulated on a certain day; μ is the average wind speed at a height of 10m above sea level in the experimental area to be simulated on a certain day; n is the number of random components, which is 50; when V n When it is greater than 5% of the average wind speed of a certain day in the experimental area, take V n =0.05Va .

[0105] The water level of the brine in the test box 1 is controlled by the water pump 17 to realize the dry-wet cycle of the experimental component 12 and adjust the brine level in the test area according to the test conditions.

[0106] The formula for brine water level change is:

[0107] h=20+10k

[0108]

[0109] h is the water level height, the water level cycle period is one hour, 0≤t<10 means one hour in the cycle, and t is between the 0th minute and the 10th minute.

[0110] With a cycle of 1 hour, the components can be placed in full immersion, half immersion and natural exposure states within the cycle. The water level has an impact on the wave height. When the water level is less than 1 / 2 of the wave height, the friction resistance of the seabed increases, which will hinder the wave speed at the bottom and cause wave breaking.

[0111] The height of the fan 10 is adjusted according to the test requirements through the second lifting assembly 9. The lifting height changes with the change of the water level, so that under normal conditions, the lowest point of the blades 11 of the fan 10 is always 20 cm above the water level (when the water level is 0, the fan height is 20 cm).

[0112] Example 5 includes all the contents of Example 4:

[0113] like Figure 4 、 Figure 5 As shown, a deep neural network wave height prediction model was established using neural network technology. Data samples were collected every 30 seconds for saltwater wave height and wind speed data generated by wind turbine 10. This data inevitably contains missing parameters. Data cleaning is performed to ensure that missing data does not exceed 5% of the total data. Fields containing missing parameters can be deleted, and data with obvious anomalies, such as wave heights greater than 20 cm or negative numbers, can also be directly deleted.

[0114] Based on this, the process of building a wave height prediction model based on deep neural network is as follows:

[0115] The first step is to determine the target of the wave height prediction model, which is to predict the wave height caused by the wind turbine 10 .

[0116] The second step is to collect the wave height model dataset.

[0117] The third step is to clean the data needed for the wave height model to obtain the input data set for the wave height model. The data validity is determined based on the saltwater level and wind speed, invalid values ​​are removed, and the saltwater level and wind speed are matched to the effective wave height. Furthermore, data for the saltwater level, wind speed, and effective wave height are all collected at 30-second intervals, and the input feature parameters and corresponding target parameters are clearly defined.

[0118] The fourth step involves dividing 80% of the dataset into a training set for model training, and the remaining 20% ​​into a test set for evaluating model performance. Wave height has a significant causal relationship with saltwater level and wind speed inputs, so a wave height prediction model was constructed based on a deep neural network.

[0119] The fifth step is to evaluate the model and analyze whether the wave height prediction model can better predict the real data.

[0120]

[0121] Among them, MAE is the mean absolute error, y i is the predicted value, is the true value. We used Bayesian optimization to find hyperparameters, selecting a learning rate of 0.003, a dropout value of 0.4 for the last layer, three hidden layers, and 128, 128, and 64 neurons per layer, respectively. We also increased the number of iterations to 500.

[0122] like Figure 6 As shown, A is the amplitude, which is half of the effective wave height value of H; S is the distance the crest moves forward per second; H is the effective wave height; L is the wavelength.

[0123] The wave height prediction model based on deep neural network predicts the significant wave height H, which is approximately The fan 10 is programmed through computer programming, and the lowest point of the blades 11 of the fan 10 is adjusted to 3 mm below the surface of the wave water. If the depth of the drop is too deep, it cannot simulate the salt spray. The mechanical force of the blades 11 in the fan 10 scrapes up a small amount of water, thereby achieving the effect of a salt spray environment.

[0124]

[0125] D is the lowest point of the blade 11; h is the water level; is the predicted value of sea waves; the unit is cm.

[0126] The wind load generated by the wind angle and wind force can cause waves to the pool in the test area, achieving the purpose of simulating the scouring of components by waves. In addition, the naturally exposed parts of the test components 12 in the test area above the water surface can be quickly dried.

[0127] The above-described embodiments merely represent specific implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

Claims

1. A chloride ion corrosion test method, comprising a test box (1), salt water arranged in the test box (1), a test component (12) and a wave-making component, characterized in that: Obtain the daily average values ​​of historical meteorological elements in the experimental area and simulate the conditions of one day in the test chamber (1) with time interval T; The wave-making assembly comprises a fan (10) and a second lifting assembly (9), wherein the second lifting assembly (9) drives the fan (10) to maintain a constant height relative to the salt water level under normal conditions; The mathematical expression of the final wind speed of the fan (10) is: in, is the final wind speed, is the basic wind speed, For gusts of wind, For the gradual wind, is the random wind volume, basic wind speed Take the average wind speed of the day to be simulated; The gusts The mathematical modeling formula is: in, Take 5% of the average wind speed of a certain day to be simulated, t is the length of time in minutes within an hour, and the cycle is repeated once every 1 hour; The gradient wind The mathematical modeling formula is: in, It is 3% of the average wind speed of a certain day to be simulated, t is the duration in minutes within an hour, and the cycle is repeated once every 1 hour; The random wind The mathematical modeling formula is: in, is the angular frequency of the i-th random component; is the discrete spacing of the random components, is a random variable uniformly distributed on [0,2π]; is the amplitude of the i-th random component; is the surface resistance coefficient or plane expansion coefficient of the wind field in the experimental area to be simulated on a certain day; F is the wind speed fluctuation scale coefficient or turbulence scale factor in the experimental area to be simulated on a certain day; is the average wind speed at a height of 10m above sea level in the experimental area on a certain day; n is the number of random components; When the wind speed in the simulated experimental area is 5% of the average value on a certain day, take =0.05 .

2. A chloride ion corrosion test method according to claim 1, characterized in that: The apparatus further comprises a first lifting assembly (5) and a pressure plate (6), wherein the first lifting assembly (5) is lifted and lowered to drive the pressure plate (6) at its output end to press the experimental component (12) or to separate the experimental component (12); a pressure sensor (7) is provided at the output end of the first lifting assembly (5) or the pressure plate (6), and the pressure sensor (7) is used to monitor the load size of the experimental component (12) in real time.

3. A chloride ion corrosion test method according to claim 2, characterized in that: The first lifting assembly (5) and the pressing plate (6) cooperate to apply a simple harmonic load to the experimental component (12), and the mathematical expression of the simple harmonic load is: in, is the average load, is the load amplitude, is the frequency, is the phase angle, and t is the time.

4. A chloride ion corrosion test method according to claim 1, characterized in that: A chloride ion solubility monitor (22), a temperature monitor (21) and a heating component (23) are installed in the test box (1).

5. A chloride ion corrosion test method according to claim 1, characterized in that: It also includes a water tank (19), a pipe (14) connecting the water tank (19) and the test box (1), a water pump (17) installed on the pipe (14), and the water pump (17) drives the salt water to flow bidirectionally between the water tank (19) and the test box (1) through the pipe (14).

6. A chloride ion corrosion test method according to claim 1, characterized in that: The time T is 1 hour.

7. A chloride ion corrosion test method according to claim 6, characterized in that: The height h of the salt water level in the test chamber (1) is: Where h is the height of the salt water level, the water level cycle period is one hour, and t is the duration in minutes within the one-hour cycle.

8. A chloride ion corrosion test method according to claim 6, characterized in that: The second lifting assembly (9) drives the blades (11) of the fan (10) to move the lowest point below the surface of the salt water, and the mechanical force of the blades (11) in the fan (10) scrapes away water to create a salt spray environment.

9. A chloride ion corrosion test method according to claim 8, characterized in that: The lowest point of the blades (11) of the fan (10) is adjusted to be 3 mm below the surface of the salt water.

10. A chloride ion corrosion test method according to claim 9, characterized in that: The following steps are involved: Based on neural network technology, a deep neural network wave height prediction model is established; Collect ocean wave height model datasets; Clean the data needed for the wave height model to obtain the data set for input into the wave height model. Based on the salt water level and wind speed, determine the validity of the data, remove invalid values ​​from the data, and match the salt water level and wind speed with the effective wave height. 80% of the dataset is divided into a training set for training the model, and the remaining 20% ​​of the dataset is used as a test set to evaluate the model performance; There is a significant causal relationship between wave height, water level and wind speed input. A wave height prediction model is constructed based on a deep neural network model. Evaluate the model and analyze whether the wave height prediction model can predict the real data well; The significant wave height H is predicted based on the wave height prediction model, which is approximately ; D= + -0.3 Wherein, D is the lowest point of the blade (11) of the fan (10), is the height of the salt water level; is the predicted value of the ocean waves.

Images