Methods and systems for corrosion protection of thrusters in deep-sea environments

Abstract

This application relates to the field of data processing technology and discloses a method and system for corrosion protection of thrusters in deep-sea environments. The method includes: collecting seawater corrosion data through a sensor array; performing risk classification using a genetic algorithm; optimizing current distribution using a Lagrange algorithm; constructing an adaptive protective potential field to respond to changes in chloride ion concentration; analyzing corrosion trends based on a predictive model; and implementing corrosion control. This application improves the intelligence level and protection effect of thruster corrosion protection in deep-sea environments.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for corrosion protection of thrusters in deep-sea environments. Background Technology

[0002] In existing technologies, corrosion protection for deep-sea thrusters mainly employs passive protection strategies. This includes coating the thruster surface with anti-corrosion coatings such as epoxy resin or polyurethane coatings, using a single coating system, and installing sacrificial anodes or impressed current cathodic protection devices. These devices prevent electrochemical corrosion by placing the thruster metal in a cathodic state through electrochemical principles. After the equipment is put into service, corrosion status is primarily assessed through regular manual inspections and maintenance. When coating damage or signs of corrosion are found, the equipment needs to be salvaged for coating repair or replacement of the sacrificial anodes.

[0003] Existing technologies have the following shortcomings: the protective effect is not durable enough, and single coating systems are prone to failure such as blistering, peeling and penetration in complex marine environments; there is a lack of targeted design, and uniform anti-corrosion measures are used for different components of the propeller, ignoring the differences in corrosion risks faced by each component; monitoring methods are outdated, mainly relying on manual visual inspection, which cannot monitor the corrosion development in real time; maintenance costs are high, requiring frequent equipment recycling and recoating; and protection parameters are fixed, making it impossible to automatically adjust the anti-corrosion strategy according to changes in actual marine environmental conditions.

[0004] Because current technologies cannot achieve real-time monitoring and data processing of corrosion factors in deep-sea environments, changes in corrosion risk cannot be detected in a timely manner, thus hindering differentiated corrosion risk classification assessments. The lack of differentiated risk assessment further prevents current allocation optimization from being precisely configured for areas with different risk levels, thus preventing the construction of an adaptive protective potential field to respond to environmental changes. Ultimately, due to the lack of predictive analysis capabilities based on data processing, current technologies cannot predict corrosion trends or intelligently optimize protection parameters, limiting them to a passive, reactive maintenance approach. Summary of the Invention

[0005] This application provides a method and system for corrosion protection of thrusters in deep-sea environments, which solves the problems in the prior art that deep-sea thruster corrosion protection cannot achieve real-time monitoring, differentiated risk assessment, adaptive potential field control and predictive parameter optimization, thereby improving the intelligence level and protection effect of thruster corrosion protection in deep-sea environments.

[0006] Firstly, this application provides a method for corrosion protection of a thruster in a deep-sea environment, the method comprising:

[0007] Real-time data acquisition and processing of seawater around the deep-sea thruster was performed using a corrosion monitoring sensor array to obtain an environmental corrosion dataset containing chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential.

[0008] Based on the environmental corrosion dataset, a genetic algorithm was used to calculate the corrosion risk level of the propeller surface, and the corrosion risk levels of the propeller blade area, bearing sealing area and shell stationary area were obtained.

[0009] The corrosion risk level was optimized by using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode.

[0010] The surface potential field of the thruster is reconstructed and controlled based on the current density distribution parameters of each protective electrode to obtain an adaptive protective potential field based on chloride ion concentration.

[0011] The adaptive protective potential field is processed by corrosion trend analysis through a prediction model to obtain protection parameter optimization instructions and execute anti-corrosion control.

[0012] Secondly, this application provides a deep-sea environment propulsion corrosion protection system, the deep-sea environment propulsion corrosion protection system comprising:

[0013] The data acquisition module is used to collect and process real-time data of the seawater around the deep-sea thruster through a corrosion monitoring sensor array, and obtain an environmental corrosion dataset including chloride ion concentration, dissolved oxygen concentration, pH value and corrosion potential.

[0014] The grading module is used to perform corrosion risk grading calculation on the propeller surface based on the environmental corrosion dataset using a genetic algorithm, and to obtain the corrosion risk level of the propeller blade area, bearing sealing area and shell stationary area.

[0015] The allocation module is used to optimize the current allocation of the corrosion risk level using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode.

[0016] The control module is used to reconstruct and control the surface potential field of the thruster according to the current density distribution parameters of each protective electrode, so as to obtain an adaptive protective potential field based on chloride ion concentration.

[0017] The analysis module is used to perform corrosion trend analysis on the adaptive protection potential field through a prediction model, obtain protection parameter optimization instructions, and execute anti-corrosion control.

[0018] Thirdly, a propeller corrosion protection device for deep-sea environments is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the propeller corrosion protection device for deep-sea environments to perform the aforementioned propeller corrosion protection method for deep-sea environments.

[0019] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the aforementioned method for protecting thrusters from corrosion in a deep-sea environment.

[0020] The technical solution provided in this application obtains an environmental corrosion dataset by real-time data acquisition and processing of seawater around the deep-sea thruster through a corrosion monitoring sensor array. This overcomes the limitations of existing technologies that rely on periodic manual inspections, enabling continuous monitoring of key corrosion factors such as chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential. This provides a reliable data foundation for subsequent intelligent corrosion prevention decisions. The application of genetic algorithms in corrosion risk classification calculation optimizes weight parameters through simulated natural selection and genetic mechanisms, making the corrosion risk assessment of the propeller blade area, bearing sealing area, and static area of ​​the outer shell more accurate and objective, avoiding the subjectivity and inconsistency of traditional methods that rely on engineer experience. The application of Lagrange algorithms in current distribution optimization achieves optimal configuration of current density distribution parameters for each protective electrode under the condition of total current conservation through a constrained optimization mathematical model. Compared with the traditional uniform distribution method, it can differentiate current distribution according to the actual protection needs of different areas, ensuring sufficient protection for high-risk areas while avoiding over-protection and energy waste in low-risk areas.

[0021] The adaptive protective potential field construction technology based on chloride ion concentration enables the corrosion protection system to automatically adjust the protection intensity according to the real-time changes in chloride ion concentration in the seawater environment. This breaks through the technical bottleneck of fixed protection parameters in existing technologies, achieving a fundamental shift from passive protection to active adaptation. The application of predictive models in corrosion trend analysis, particularly the introduction of long short-term memory (LSTM) network algorithms, gives the system the ability to learn historical corrosion evolution patterns and predict future corrosion trends. This predictive analysis capability transforms corrosion control from traditional post-disaster repair to pre-disaster prevention, significantly extending the service life of the thruster in the deep-sea environment. The application of particle swarm optimization (PSO) algorithms in protective parameter adjustment automatically finds the optimal current density and potential adjustment amounts through a swarm intelligence optimization mechanism. This allows the entire corrosion protection system to maintain optimal protection in the complex and ever-changing deep-sea environment. Compared to traditional manual adjustment methods, it not only has a faster response speed but also higher adjustment accuracy and better system stability, thus achieving intelligent, automated, and efficient corrosion protection for deep-sea thrusters. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of one embodiment of the anti-corrosion protection method for thrusters in a deep-sea environment as described in this application.

[0024] Figure 2 This is a schematic diagram of one embodiment of the anti-corrosion protection system for thrusters in a deep-sea environment, as described in this application.

[0025] Figure 3 This is a schematic block diagram of the structure of the anti-corrosion protection device for thrusters in a deep-sea environment, as described in this embodiment of the invention. Detailed Implementation

[0026] This application provides a method and system for corrosion protection of thrusters in deep-sea environments. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0027] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the method for corrosion protection of thrusters in a deep-sea environment in this application includes:

[0028] Step S101: Real-time data acquisition and processing of seawater around the deep-sea thruster is performed using a corrosion monitoring sensor array to obtain an environmental corrosion dataset containing chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential.

[0029] Step S102: Based on the environmental corrosion dataset, the corrosion risk classification of the propeller surface is calculated using a genetic algorithm to obtain the corrosion risk levels of the propeller blade area, bearing sealing area, and shell stationary area.

[0030] Step S103: The corrosion risk level is optimized by current distribution using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode.

[0031] Step S104: The surface potential field of the thruster is reconstructed and controlled according to the current density distribution parameters of each protective electrode to obtain an adaptive protective potential field based on chloride ion concentration.

[0032] Step S105: The adaptive protection potential field is processed by corrosion trend analysis through the prediction model to obtain protection parameter optimization instructions and execute anti-corrosion control.

[0033] It is understood that the executing entity of this application can be a propulsion corrosion protection system for deep-sea environments, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiments use a server as an example for illustration.

[0034] Specifically, real-time data acquisition is achieved through a corrosion monitoring sensor array deployed on the thruster surface. This sensor array includes a chloride ion selective electrode sensor, a dissolved oxygen fluorescence sensor, a pH glass electrode sensor, and a corrosion potential reference electrode sensor. The chloride ion selective electrode sensor is based on the principle of ion-selective membrane; when chloride ions in seawater come into contact with the electrode membrane, a potential difference is generated, the magnitude of which is logarithmically related to the chloride ion concentration. The dissolved oxygen fluorescence sensor utilizes the principle of fluorescence quenching; oxygen molecules reduce the fluorescence intensity of a fluorescent dye, and the dissolved oxygen concentration is calculated by measuring the change in fluorescence intensity. The pH glass electrode sensor is based on the principle of glass membrane potential; changes in hydrogen ion concentration cause changes in the potential difference across the glass membrane. The corrosion potential reference electrode sensor uses a silver chloride electrode as a reference to measure the potential of the thruster's metal surface relative to the reference electrode. Data acquisition is performed at 30-second intervals, and the analog signals output by each sensor are converted into digital signals by an analog-to-digital converter. The raw data undergoes Kalman filtering to eliminate noise interference and is then calibrated using a standard solution to obtain accurate values. Outlier removal uses the 3σ criterion; data exceeding the range of the mean plus or minus three standard deviations are marked as outliers and removed. The valid data are combined in time series to form an environmental corrosion dataset. The data structure includes fields such as timestamp, sensor location coordinates, chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential.

[0035] A genetic algorithm was used to calculate the corrosion risk classification of the propeller surface. The genetic algorithm is an optimization algorithm that simulates the biological evolution process, finding the optimal solution through selection, crossover, and mutation operations. First, the environmental corrosion dataset was mapped to three regions according to the propeller geometry: the propeller blade region, the bearing seal region, and the outer shell stationary region. The geometric mapping established a coordinate system based on the three-dimensional model of the propeller, and each sensor data point was assigned to its corresponding region based on its spatial coordinates. A deep-sea pressure correction coefficient was used to correct for the influence of the high-pressure environment on corrosion parameters; the correction formula was that the pressure correction coefficient equals 1 plus a pressure correction factor multiplied by the current depth pressure. The risk index for each region was calculated using a weighted summation method, multiplying chloride ion concentration, dissolved oxygen concentration, pH value, and stress factor by their respective weight coefficients and then summing them. The genetic algorithm population was initialized by randomly generating multiple sets of weight coefficient combinations as initial individuals, and a fitness function was used to evaluate the contribution of each set of weight coefficients to the accuracy of the risk assessment. The selection operation retained individuals with high fitness, the crossover operation combined the weight coefficients of two individuals, and the mutation operation randomly changed some of the weight coefficients. After multiple generations of evolution, the optimal weight parameters are obtained. Based on these weight parameters, the final risk index of each region is calculated and divided into three corrosion risk levels: high, medium, and low, according to preset thresholds.

[0036] The Lagrange multiplier algorithm is used for current allocation optimization. The Lagrange multiplier algorithm is a constrained optimization method used to find the optimal solution of the objective function under certain constraints. The number of protective electrodes is configured according to the corrosion risk level, with more electrodes allocated to high-risk areas to provide sufficient protection. Electrode weight allocation is based on the difference in risk level, with higher weights in high-risk areas, followed by medium-risk areas, and the lowest weights in low-risk areas. The protection area allocation is calculated based on the surface area and risk level of each region, with larger areas and higher risks receiving larger protection areas. A constrained optimization model is established using the Lagrange multiplier method, with the objective function being to minimize the sum of squared currents to reduce energy consumption, and the constraints being the conservation of total current and the minimum protection current requirement for each region. By solving the system of equations where the partial derivatives of the Lagrange function are zero, the optimal current allocation coefficient that satisfies the constraints is obtained. This allocation coefficient is multiplied by the protection area of ​​each region to obtain the current density distribution parameters of each protective electrode, ensuring that each region receives a protection strength commensurate with its corrosion risk.

[0037] The potential field on the thruster surface is reconstructed and controlled based on current density distribution parameters. The potential distribution calculation is based on the principle of electric field superposition; the potential fields generated by each protective electrode are superimposed in space to form the total potential field. The calculation process considers electrode position coordinates, output current intensity, seawater resistivity, and distance factor. Chloride ion concentration gradient correction is achieved by establishing a model relating chloride ion concentration to the potential correction amount. When the chloride ion concentration increases, the protective potential needs to be reduced to maintain effective protection. The correction coefficient is calculated based on the change in chloride ion concentration and the sensitivity coefficient. Each protective electrode dynamically adjusts its output current according to the correction coefficient. The adjustment algorithm uses a proportional-integral control method, with the proportional term responding to the current concentration deviation and the integral term eliminating steady-state errors. The adjusted electrode output current parameters are then recalculated using potential field superposition to generate the reconstructed potential distribution on the thruster surface. The adaptive protective potential field forms a dynamically responsive corrosion protection system by real-time monitoring of chloride ion concentration changes and corresponding adjustment of the potential distribution.

[0038] Corrosion trend analysis is performed using a predictive model. The Long Short-Term Memory (LSTM) network algorithm is a special type of recurrent neural network capable of processing time-series data and long-term dependencies. The network structure includes input gates, forget gates, and output gates, with information flow controlled by a gating mechanism. Adaptive protective potential field data is compared with historical corrosion evolution data over time to extract characteristic parameters of corrosion development trends, including corrosion rate change gradient, potential fluctuation amplitude, and periodicity. The LSM network is trained using historical data to build a corrosion rate prediction model. The input layer receives current potential field data and environmental parameters, the hidden layer learns corrosion development patterns, and the output layer predicts the corrosion rate for the next 24 hours. Remaining protection time is calculated based on the current coating thickness, predicted corrosion rate, and safety threshold, using division to obtain the remaining protection time for each region. A particle swarm optimization algorithm simulates bird foraging behavior, with each particle representing a set of protective parameter adjustment schemes. The algorithm finds the optimal adjustment strategy through velocity and position updates. The algorithm aims to extend remaining protection time and reduce energy consumption, outputting current density adjustment and potential regulation. Each protective electrode adjusts its output parameters according to the optimization results, achieving closed-loop optimization of corrosion control.

[0039] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0040] The chloride ion selective electrode sensor, dissolved oxygen fluorescence sensor, pH glass electrode sensor, and corrosion potential reference electrode sensor are deployed at key locations on the surface of the thruster according to a preset array layout.

[0041] Data was read and processed from each sensor at 30-second intervals to obtain the raw values ​​of chloride ion concentration, dissolved oxygen concentration, pH value and corrosion potential.

[0042] The raw values ​​of chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential are input into the data preprocessing module for filtering and calibration to obtain chloride ion concentration calibration data, dissolved oxygen concentration calibration data, pH value calibration data, and corrosion potential calibration data.

[0043] Outlier removal was performed on the chloride ion concentration calibration data, dissolved oxygen concentration calibration data, pH value calibration data, and corrosion potential calibration data to obtain valid corrosion parameter data;

[0044] The effective corrosion parameter data are combined and processed according to the time series to obtain the environmental corrosion dataset.

[0045] Specifically, the sensor array deployment is based on the geometric characteristics of the thruster and the analysis of corrosion-sensitive areas. The chloride ion selective electrode sensor employs ion-sensitive field-effect transistor technology. Its core is an electrode head containing a chloride ion-sensitive membrane. When chloride ions in seawater come into contact with the sensitive membrane, an ion exchange reaction occurs, generating a potential difference logarithmically related to the chloride ion concentration. This potential difference is converted into a voltage signal output through a high-impedance amplifier. The dissolved oxygen fluorescence sensor operates based on the principle of fluorescence quenching. The sensor contains a fluorescent dye coating. Excitation light irradiates the fluorescent dye to produce fluorescence. When dissolved oxygen molecules in seawater collide with the fluorescent dye, the fluorescence intensity decreases. The dissolved oxygen concentration is calculated by measuring the degree of fluorescence intensity decay. The pH glass electrode sensor uses a special glass membrane as the sensitive element. The difference in hydrogen ion concentration across the glass membrane generates a membrane potential, which is logarithmically proportional to the hydrogen ion activity. The corrosion potential reference electrode sensor uses a silver chloride electrode as a stable reference point, measuring the potential difference between the thruster's metal surface and the reference electrode to reflect the metal's corrosion tendency. The preset array layout deploys chloride ion selective electrode sensors on the leading and trailing edges of the propeller blades, dissolved oxygen fluorescence sensors are installed around the bearing sealing area, pH glass electrode sensors are distributed on the surface of the static area of ​​the housing, and corrosion potential reference electrode sensors are evenly distributed in various areas to form a potential monitoring grid.

[0046] Data acquisition and processing employs a synchronous sampling mode with 30-second intervals. Analog signals from each sensor simultaneously enter a multi-channel analog-to-digital converter (ADC). The ADC discretizes the continuous analog voltage signals into digital values ​​with a 16-bit conversion accuracy and a sampling frequency of 10Hz. The raw chloride ion concentration value is calculated using the potential difference of the ion-selective electrode. The relationship between the potential difference and chloride ion concentration follows the Nernst equation; however, temperature correction factors and electrode constants must be considered in the specific calculation. The raw dissolved oxygen concentration value is based on fluorescence intensity measurements. Fluorescence intensity is inversely proportional to dissolved oxygen concentration, and the fluorescence intensity value is converted to a dissolved oxygen concentration value using a calibration curve. The raw pH value is calculated from the membrane potential of the glass electrode. A linear relationship exists between the membrane potential and pH value, with a slope of approximately 59.16 mV per pH unit. The raw corrosion potential value is directly derived from the potential difference measured by the reference electrode, and the value is expressed in millivolts, representing the electrochemical state of the propeller metal surface.

[0047] The data preprocessing module receives four types of raw values ​​and performs filtering and calibration processing. The filtering process uses a digital low-pass filter to eliminate high-frequency noise interference, with the filter cutoff frequency set to 0.1Hz to retain useful corrosion parameter changes while filtering out electromagnetic interference and mechanical vibration noise. The calibration process establishes calibration equations based on known parameter values ​​of standard solutions. For the chloride ion selective electrode, a calibration relationship between concentration and potential is established using standard seawater samples with known chloride ion concentrations. For the dissolved oxygen fluorescence sensor, zero and full-scale calibration points are determined using nitrogen-saturated and air-saturated water samples. For the pH glass electrode, two-point calibration is performed using a standard buffer solution to determine the slope and intercept. The corrosion potential reference electrode is calibrated using standard electrode potentials to ensure measurement accuracy. The calibration process substitutes the raw values ​​into the calibration equations to calculate chloride ion concentration calibration data, dissolved oxygen concentration calibration data, pH value calibration data, and corrosion potential calibration data.

[0048] Outlier removal is based on statistical methods to identify and remove outlier data points. First, the mean and standard deviation of various calibration data are calculated within a sliding time window, set to 10 minutes and containing 20 data points. Outlier determination uses a three-standard-deviation criterion; a data point is marked as an outlier when its difference from the mean exceeds three times the standard deviation. Causes of outliers include measurement deviations due to factors such as momentary sensor malfunctions, electromagnetic interference, and marine organism attachment. The outlier removal algorithm examines each data point individually, removing those marked as outliers from the data sequence. Simultaneously, the time and location of outlier occurrences are recorded for sensor health assessment. The resulting effective corrosion parameter data after outlier removal exhibits higher reliability and continuity.

[0049] Time-series data aggregation processing arranges and integrates effective corrosion parameter data according to a unified time base. The time base adopts UTC time format to ensure time synchronization of data from different sensors. The data aggregation process first establishes a time index, with each time point corresponding to a complete set of corrosion parameter data, including chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential. When some sensor data is missing at a certain time point, linear interpolation is used to fill in the missing data. The interpolation calculation is based on linear fitting of data values ​​from adjacent time points. Data aggregation also includes the addition of spatial coordinate information. Each data point is labeled with the three-dimensional coordinate position of the corresponding sensor, forming a multi-dimensional data structure containing time, spatial, and parameter dimensions. The environmental corrosion dataset is stored in a relational database format. The data table structure includes timestamp, sensor number, coordinate position, chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential fields. The data table supports query and filtering operations by time range, spatial region, and parameter range.

[0050] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0051] The environmental corrosion dataset was processed by region mapping according to the propeller geometry to obtain the partitioning parameters of the propeller blade region, bearing seal region and shell stationary region.

[0052] Based on the pressure correction factor, environmental correction processing is performed on the partition parameters to obtain the corrected corrosion parameters;

[0053] The risk index for each region is obtained by weighted summation of chloride ion concentration, dissolved oxygen concentration, pH value, and stress factor.

[0054] The risk index is initialized and its fitness is evaluated using a genetic algorithm to obtain the weight parameters.

[0055] The risk index is graded based on weighted parameters to obtain the corrosion risk levels of the propeller blade area, bearing sealing area, and static area of ​​the outer casing.

[0056] Specifically, the region mapping process establishes a spatial coordinate system based on the three-dimensional geometric model of the propeller. The propeller geometry includes the complex curved surfaces of the propeller blades, the cylindrical structure of the bearing sealing area, and the regular surface of the static area of ​​the outer shell. The mapping algorithm first reads the spatial coordinate information of the sensors, and then assigns the sensor data to the corresponding regions according to preset region boundary conditions. The boundary of the propeller blade area is defined as the curved surface range from the leading edge to the trailing edge of the blade; the boundary of the bearing sealing area is defined as the cylindrical area within the radius around the center of the bearing; and the boundary of the static area of ​​the outer shell is defined as the remaining surface of the propeller shell after removing the propeller and bearing areas. The mapping process uses a nearest neighbor algorithm to calculate the Euclidean distance from each sensor location to the center point of each region, and assigns the sensor data to the nearest region. The partitioning parameters include the numerical set of chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential measured by all sensors in each region, while also recording the number and distribution density of sensors in each region. The data aggregation process performs statistical processing on the data from multiple sensors in the same region, calculating statistical characteristic values ​​such as average, maximum, minimum, and standard deviation to form the comprehensive corrosion parameter characteristics of each region.

[0057] The pressure correction factor is calculated based on the influence mechanism of the high-pressure environment in the deep sea on the corrosion process. Increased deep-sea pressure alters the physicochemical properties of seawater, affecting ion diffusion rates and electrochemical reaction rates. The pressure correction factor is calculated using a piecewise linear function, with different correction parameters applied across different depth ranges. For depths less than 1000 meters, the correction factor is 1 plus a linear term multiplied by the depth (0.00005); for depths between 1000 and 3000 meters, it is 1 plus a linear term multiplied by the depth (0.0001); and for depths greater than 3000 meters, it is 1 plus a linear term multiplied by the depth (0.00015). Environmental correction multiplies each corrosion parameter in the partitioned parameters by its corresponding pressure correction factor to obtain corrected corrosion parameters that account for the influence of deep-sea pressure. Chloride ion concentration correction considers changes in ion activity coefficients under high pressure; dissolved oxygen concentration correction considers the effect of pressure on gas solubility; pH ​​correction considers the effect of pressure on the ionization constant of water; and corrosion potential correction considers the effect of pressure on electrode potential. The corrected corrosion parameters more accurately reflect the actual corrosion conditions in the deep-sea environment.

[0058] The risk index calculation employs a weighted summation method to synthesize multiple corrosion factors into a single risk assessment index. Chloride ion concentration, as the primary corrosive medium, has a relatively large weighting coefficient; dissolved oxygen concentration, as the oxidant, has a second-highest weighting coefficient; pH value affects the corrosion reaction rate and has a relatively small weighting coefficient; the stress factor, reflecting the promoting effect of mechanical stress on corrosion, has its weighting coefficient set differently based on regional characteristics. The stress factor calculation is based on stress distribution analysis during propeller operation. The propeller blade area experiences complex dynamic stresses, including centrifugal force, hydrodynamic pressure, and vibration stress; the bearing seal area experiences stress concentration caused by radial and axial loads; and the static area of ​​the outer shell mainly experiences hydrostatic pressure. The weighted summation calculation multiplies the corrected chloride ion concentration by a weighting coefficient of 0.4, the dissolved oxygen concentration by a weighting coefficient of 0.3, the square of the pH value deviation by a weighting coefficient of 0.2, and the stress factor by a weighting coefficient of 0.1. The sum of these four products yields the initial risk index for each region.

[0059] Genetic algorithms, as optimization algorithms that simulate natural selection and genetic mechanisms, are used to find the optimal combination of weight parameters to improve the accuracy of risk assessment. The population initialization process randomly generates multiple individuals, each representing a set of weight parameters. The number of individuals is set to 50, and each individual contains four weight parameters corresponding to chloride ion concentration, dissolved oxygen concentration, pH value, and stress factor. Fitness evaluation defines a fitness function to measure the quality of each set of weight parameters. The fitness function is based on the degree of matching between the risk assessment results and historical corrosion damage data; the higher the matching degree, the higher the fitness value. The fitness calculation process applies the current weight combination to historical data to calculate the risk index, then compares it with the actual corrosion damage level, and calculates the prediction accuracy as the fitness value. The selection operation uses a roulette wheel selection method, where individuals with higher fitness have a higher probability of being selected. The crossover operation recombines the weight parameters of two parent individuals to generate offspring individuals, with the crossover probability set to 0.8. The mutation operation randomly changes some weight parameters in individuals, with the mutation probability set to 0.1. After multiple generations of evolution, the weight parameter combination represented by the individual with the highest fitness is determined as the optimal weight parameters.

[0060] The risk level assessment process recalculates the final risk index for each region based on optimized weight parameters, and then classifies the risk index into different levels according to preset threshold ranges. A three-tier system is used: a risk index less than 0.3 is classified as low risk, a risk index between 0.3 and 0.7 as medium risk, and a risk index greater than 0.7 as high risk. Threshold settings are based on statistical analysis of extensive historical corrosion data; low risk corresponds to regions with slower corrosion rates, medium risk to regions with moderate corrosion rates, and high risk to regions with faster corrosion rates. The assessment process compares the risk index of each region with the threshold range to determine the corrosion risk level of the propeller blade area, bearing seal area, and static outer casing area.

[0061] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0062] The number of protective electrodes is configured according to the corrosion risk level, resulting in a layout scheme for the electrodes in the propeller blade area, the bearing sealing area, and the static area of ​​the outer casing.

[0063] The layout scheme is processed by assigning electrode weights according to the regional risk level to obtain the electrode weights for high-risk areas, medium-risk areas, and low-risk areas.

[0064] Based on the corrosion risk level of each region, the protection area is allocated and calculated to obtain the areas of high-risk, medium-risk, and low-risk zones.

[0065] Based on the protection area, the current constraint is solved using the Lagrange multiplier method to obtain the distribution coefficient that satisfies the total current conservation condition;

[0066] The current density distribution parameters of each protection electrode are obtained by multiplying the distribution coefficient with the area of ​​each region.

[0067] Specifically, the electrode density configuration for protection is determined based on the differentiated requirements of corrosion risk levels, with the density coefficient corresponding to the risk level used for calculation. The propeller blade area, considered a high-risk area, has an electrode density coefficient of 1.5 electrodes per square meter; the bearing seal area, considered a medium-risk area, has an electrode density coefficient of 1.0 electrodes per square meter; and the static outer casing area, considered a low-risk area, has an electrode density coefficient of 0.5 electrodes per square meter. The electrode quantity calculation involves multiplying the surface area of ​​each area by the corresponding density coefficient to obtain the required number of electrodes. The surface area of ​​the propeller blade area includes the complex surface integral calculation results of the front, back, and edge of the blades; the surface area of ​​the bearing seal area is the sum of the side and end areas of the cylindrical sealing surface; and the surface area of ​​the static outer casing is the regular geometric surface area of ​​the propeller casing. The electrode layout scheme considers the uniformity of electrode spacing and the continuity of current distribution, employing a gridded layout to establish a regular electrode distribution grid on the surface of each area. Electrodes in the propeller blade area are uniformly distributed along the radial and circumferential directions of the blades; electrodes in the bearing seal area are uniformly distributed along the circumferential and axial directions; and electrodes in the static outer casing area are distributed in a rectangular grid pattern. The layout scheme also needs to consider the mechanical compatibility between the electrode and the thruster structure to avoid interference between the electrode installation position and the moving parts of the thruster, while ensuring the reasonable routing and fixing method of the electrode connection cables.

[0068] Electrode weighting is determined based on the relative importance of risk levels in each region, reflecting the differences in protective current requirements across different risk levels. High-risk areas have higher weights due to their rapid corrosion rates and severe damage, ensuring adequate electrochemical protection. Medium-risk areas have moderate weights due to their moderate corrosion rates and need for moderate protection, resulting in moderate weights. Low-risk areas have lower weights due to their slow corrosion rates and relatively low protection requirements. The weighting calculation employs normalization to ensure the sum of all electrode weights equals 1. Normalization involves dividing the initial weight of each electrode by the sum of all electrode weights to obtain a standardized weight. Weighting also considers the interaction between electrodes and the superposition effect of the current field; the weighting of adjacent electrodes needs to balance local protection strength and overall protection effectiveness. A dynamic adjustment mechanism for electrode weights fine-tunes the weights based on real-time monitoring of corrosion status changes, adjusting the weights of electrodes in a region accordingly when the corrosion risk level changes.

[0069] The protection area allocation calculation determines the effective area to be protected based on the geometric characteristics and corrosion risk level of each region. The protection area is not equivalent to the region's surface area but rather considers the effective protection range that takes into account corrosion sensitivity and protection requirements. The high-risk area calculation includes the entire surface area of ​​the propeller blades, as the entire surface faces severe corrosion threats under high-speed rotation and complex flow fields. The medium-risk area calculation includes the critical sealing area of ​​the bearing sealing zone, focusing on protecting the sealing interface between the bearing and seawater and stress concentration areas. The low-risk area calculation includes corrosion-sensitive local areas in the static area of ​​the outer casing, such as welded joints, material transition zones, and geometric discontinuities. Area allocation also considers the overlapping coverage of electrode protection radii. The effective protection radius of each electrode is calculated based on current density and seawater resistivity, and the protection ranges of adjacent electrodes need to have appropriate overlap to ensure no protection blind spots. Dynamic adjustment of the protection area is based on feedback from corrosion monitoring data. When the actual corrosion situation in certain areas is found to be inconsistent with expectations, the allocation ratio of the protection area is adjusted promptly.

[0070] The Lagrange multiplier method, a classic mathematical approach for constrained optimization, is used to solve for optimal current distribution. This method seeks a current distribution scheme that minimizes the objective function while satisfying the total current conservation constraint. The objective function for solving the current constraints is the minimization of the sum of the squares of the currents at each electrode. The physical meaning of this objective function is to minimize the total power consumption of the system. The constraints include the total current conservation condition (the sum of all electrode currents equals the total system output current) and the minimum protection current requirement constraint for each region (the current in each region must not be lower than the minimum value required to maintain effective protection). The construction of the Lagrange function combines the objective function and constraints using Lagrange multipliers, forming an unconstrained optimization problem. The solution process involves taking the partial derivatives of the Lagrange function with respect to each electrode current and the Lagrange multiplier, and setting them to zero to obtain a system of linear equations. The linear equations are solved using Gaussian elimination or matrix inversion to obtain the optimal current value for each electrode. The distribution coefficient is calculated by dividing the optimal current value of each electrode by the protection area of ​​the corresponding region, resulting in a current distribution coefficient per unit area. This coefficient reflects the current intensity required per square meter of protection area.

[0071] The calculation of current density distribution parameters involves multiplying the allocation coefficient by the area of ​​each region to obtain the specific current output requirements for each protective electrode. The current density distribution parameters are the direct control commands executed by the electrochemical protection system. The multiplication process involves multiplying the allocation coefficient of each region by the protection area of ​​the corresponding electrode to obtain the total current that the electrode should output, and then dividing by the effective area of ​​the electrode to obtain the current density on the electrode surface. The unit of current density is milliamperes per square meter (m²), and its value directly determines the cathodic protection effect of the electrode. The distribution parameters also include the time-varying characteristics of the current, adjusting the time distribution of the current output according to the dynamic changes in the corrosive environment. The spatial distribution of current density considers the coordination between electrodes; the current density of adjacent electrodes needs to transition smoothly to avoid local overprotection or underprotection. The verification of current density is conducted through potential field simulation calculations to check whether the allocation results meet the protection potential requirements of each region. When the simulation results show that the protection potential of some regions is substandard, the current density distribution parameters of the corresponding electrodes need to be adjusted.

[0072] In one specific embodiment, the process of configuring the number of protective electrodes according to the corrosion risk level may specifically include the following steps:

[0073] The electrode density is calculated according to the corrosion risk level to obtain the electrode density of high-risk area, medium-risk area and low-risk area.

[0074] Based on the electrode density, the number of electrodes in each region is allocated according to the surface area of ​​each region, resulting in the number of electrodes in the propeller blade region, the number of electrodes in the bearing seal region, and the number of electrodes in the static region of the outer casing.

[0075] Based on the geometric features of each region, the electrode positions are optimized and distributed to obtain the positions of the leading edge electrode, the electrode positions around the bearing, and the electrode positions on the outer shell surface.

[0076] The electrode positions are planned according to the current transmission distance to obtain the cable connection path for each electrode.

[0077] Based on the cable connection path, the electrodes are finally integrated into a layout, resulting in a layout scheme for the propeller blade area electrodes, bearing seal area electrodes, and housing stationary area electrodes.

[0078] Specifically, the electrode density calculation process determines the required electrode distribution density for each area based on the correspondence between corrosion risk level and electrode protection strength requirements. Electrode density reflects the number of electrodes needed per unit area to meet the corresponding corrosion protection requirements. For high-risk areas, the electrode density is calculated by multiplying the baseline density by a high-risk coefficient. The baseline density is set to 1 electrode per square meter, and the high-risk coefficient is set to 2.0, resulting in 2 electrodes per square meter for high-risk areas. For medium-risk areas, the electrode density is calculated by multiplying the baseline density by a medium-risk coefficient, set to 1.2, resulting in 1.2 electrodes per square meter. For low-risk areas, the electrode density is calculated by multiplying the baseline density by a low-risk coefficient, set to 0.6, resulting in 0.6 electrodes per square meter. The electrode density setting considers the attenuation characteristics of current transmission in the deep-sea environment and the influence of seawater resistivity. The density calculation also needs to consider the overlap and coverage requirements of the effective protection radius of the electrodes to ensure that there are no blind spots in protection for each area. The dynamic adjustment mechanism for density calculation corrects the density parameters based on corrosion monitoring feedback during actual operation. When it is found that the protection effect in a certain area is not up to standard, the electrode density in that area is appropriately increased.

[0079] Electrode allocation is achieved by multiplying the surface area of ​​each region by its corresponding electrode density to obtain the total number of electrodes required for each region. The allocation must consider the standard specifications of the electrodes and the actual space constraints of installation. The propeller blade area surface area is calculated using 3D modeling software to precisely measure the complex curved surface of the blade, including the entire surface area of ​​the front, back, leading edge, and trailing edge. The calculated area is 3.5 square meters. Multiplying this by the high-risk area electrode density of 2 per square meter yields 7 electrodes needed for the propeller blade area. The bearing seal area surface area includes the cylindrical area of ​​the bearing outer ring and the sealing end area. Geometric calculation yields a surface area of ​​2.1 square meters. Multiplying this by the medium-risk area electrode density of 1.2 per square meter yields 3 electrodes needed for the bearing seal area. Considering the actual installation space constraints, this is rounded up to 3 electrodes. The outer casing stationary area surface area is the area remaining after subtracting the propeller and bearing areas from the regular surface of the propeller casing. The calculated area is 8.0 square meters. Multiplying this by the low-risk area electrode density of 0.6 per square meter yields 5 electrodes needed for the outer casing stationary area. This is rounded up to ensure the integrity of the protective coverage. The quantity allocation also needs to take into account the redundancy of electrodes, with an additional spare electrode added to each area to cope with possible electrode failures during operation.

[0080] The optimal electrode placement is determined based on the geometric characteristics and corrosion sensitivity analysis of each region. This optimization requires balancing multiple factors, including protective effectiveness, installation feasibility, and maintenance convenience. For the blade leading-edge electrode, the optimized placement considers the complex stress distribution and fluid impact experienced by the blade during rotation. The electrode is installed near the stagnation point on the blade leading edge, where the flow velocity is relatively low and the stress concentration is moderate. The blade leading-edge electrode uses an embedded installation method, with the electrode tip flush with the blade surface to minimize impact on hydrodynamic performance. For the bearing perimeter electrode, the optimized placement is based on analysis of key bearing seal components. Electrodes are installed at circumferentially evenly spaced positions on the bearing outer ring to ensure uniform protection of the entire bearing seal area. The bearing perimeter electrode uses a ring array distribution, with three electrodes evenly distributed at 120-degree angles on the bearing outer ring circumference. For the housing surface electrode, a grid layout is adopted, dividing the housing surface into regular rectangular grids, with electrodes installed at the grid intersections. The selection of the housing surface electrode placement also needs to avoid the propeller's inlet, outlet, and other functional openings, while considering the accessibility for electrode maintenance and repair. The location optimization algorithm uses a genetic algorithm to find the combination of electrode locations that achieves the best protection effect. The optimization objective is to minimize the maximum value of the distance from any point in each region to the nearest electrode.

[0081] The connection path planning process determines the cable connection paths between each electrode and the control unit. Path planning needs to consider requirements such as shortest cable length, path safety and reliability, and ease of maintenance. The current transmission distance is calculated based on the three-dimensional spatial distance between the electrode position coordinates and the control unit position coordinates, using the Euclidean distance formula to calculate the straight-line distance as the theoretical shortest path. Actual path planning needs to consider the geometric constraints of the thruster structure. Cables cannot cross moving parts and high-temperature areas and must be laid along the surface of the fixed structure. The path planning algorithm adopts the Dijkstra shortest path algorithm, modeling the thruster structure as a graphical network. Nodes represent passable locations, and edges represent path segments where cables can be laid. The algorithm finds the shortest path from the electrode position to the control unit. The design of the cable connection path also needs to consider the mechanical protection of the cable, adding cable protective sleeves to path segments that are susceptible to impact and wear. The path planning results generate cable routing diagrams for each electrode, marking the cable start point, end point, intermediate nodes, and total length, as well as the critical locations where protective devices need to be installed. The redundant design of the cable path configures a backup cable path for critical electrodes, which can quickly switch to the backup path to maintain protection function when the main path fails.

[0082] The layout integration process comprehensively considers the electrode positions, quantity configurations, and connection paths to form a complete electrode layout scheme and performs overall optimization and adjustment. Layout integration first checks for interference between electrode positions to ensure that electrode installations do not conflict with each other, while also checking the compatibility of the electrodes with other propeller components. The propeller blade area electrode layout integration considers the dynamic balance requirements during blade rotation; the seven electrodes are symmetrically distributed according to the number and angle of the blades, with two electrodes installed at the leading edge of each blade and one electrode installed at the root. The bearing seal area electrode layout integration distributes three electrodes at equal angles on the outer ring of the bearing, with an included angle of 120 degrees between the electrodes, forming a uniform ring protection array. The housing stationary area electrode layout integration uses a rectangular grid distribution; five electrodes are distributed on the housing surface in a 2x3 grid, with row and column spacing determined based on the housing size and protection radius. Layout integration also considers electrode numbering and control loop grouping, assigning functionally related electrodes to the same control loop for centralized control and fault diagnosis. The final layout scheme generates detailed installation drawings, including complete information such as the three-dimensional coordinates of the electrodes, installation methods, cable routing, control connections, and maintenance requirements.

[0083] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0084] The current density distribution parameters of each protective electrode are processed according to the spatial location to calculate the potential distribution, and the initial potential values ​​of each point on the surface of the thruster are obtained.

[0085] Based on chloride ion concentration monitoring data, the initial potential value is corrected by concentration gradient processing to obtain the chloride ion concentration correction coefficient.

[0086] The output current of each protective electrode is dynamically adjusted according to the chloride ion concentration correction coefficient to obtain the adjusted electrode output current parameters.

[0087] The adjusted electrode output current parameters are processed by potential field superposition calculation to obtain the reconstructed potential distribution on the thruster surface;

[0088] By responding to changes in chloride ion concentration based on the reconstructed potential distribution on the thruster surface, an adaptive protective potential field based on chloride ion concentration is obtained.

[0089] Specifically, the potential distribution calculation is based on the principle of potential field superposition in electrochemical theory. The potential fields generated by each protective electrode are mathematically superimposed in three-dimensional space to obtain the potential value at any point on the thruster surface. The potential distribution calculation uses the Green's function method to solve the Laplace equation. Each protective electrode is considered as a point current source, and the potential field it generates follows the point source potential distribution law. The potential calculation for spatial location considers factors such as the three-dimensional coordinates of the electrode, the output current intensity, the conductivity of seawater, and spatial distance. The potential value is directly proportional to the current intensity and inversely proportional to the distance. The numerical calculation of the potential distribution discretizes the thruster surface into triangular grid cells. The grid density is adaptively adjusted according to the degree of change in the potential gradient, with denser grid divisions used in areas of drastic potential changes. The potential value of each grid node is obtained by algebraically summing the potentials contributed by all protective electrodes, taking into account the directionality of the potential and the linearity of superposition. The potential distribution calculation also needs to consider the boundary conditions of the thruster's metallic material. The potential distribution on the metallic surface is affected by the electrochemical properties and geometry of the material. The calculation results of the initial potential values ​​form a potential distribution map on the surface of the thruster. The potential map shows the potential level and gradient changes in each region, providing basic data for subsequent concentration correction and dynamic adjustment.

[0090] Concentration gradient correction is based on the influence mechanism of chloride ion concentration on electrochemical reaction rate and potential distribution, establishing a quantitative relationship between chloride ion concentration and potential correction. Chloride ion concentration monitoring data comes from chloride ion selective electrode sensors distributed on the thruster surface, including the chloride ion concentration values ​​and spatial coordinates at each monitoring point. The concentration gradient is calculated using the finite difference method, obtaining the concentration gradient vector by dividing the concentration difference between adjacent monitoring points by the spatial distance. The effect of chloride ion concentration on potential is based on the Nernst equation, with the potential correction proportional to the logarithm of the chloride ion concentration. The concentration correction coefficient is calculated using a piecewise function method: a negative correction coefficient indicates a need to lower the protection potential when the chloride ion concentration is below the standard seawater concentration, and a positive correction coefficient indicates a need to increase the protection potential when the chloride ion concentration is above the standard seawater concentration. The numerical formula for the correction coefficient is: the correction coefficient equals the concentration sensitivity parameter multiplied by the natural logarithm of the ratio of chloride ion concentration to standard concentration. The concentration sensitivity parameter is determined based on the type of metal material and corrosion mechanism. Concentration gradient correction also needs to consider the spatial non-uniformity of concentration distribution. Spatial interpolation methods are used to extend discrete monitoring point data to the entire thruster surface. The interpolation method uses the Kriging interpolation algorithm to ensure spatial continuity and data accuracy.

[0091] The dynamic adjustment process adjusts the output current of each protective electrode in real time based on the chloride ion concentration correction coefficient. The adjustment algorithm uses a proportional-integral-derivative (PID) controller to achieve precise current control. The adjustment amount of the electrode output current is calculated based on the correction coefficient and the current adjustment range of the electrode. The adjustment amount equals the correction coefficient multiplied by the rated current of the electrode and the adjustment gain coefficient. The proportional control term calculates the instantaneous adjustment amount based on the deviation between the current chloride ion concentration and the set value. The proportional gain coefficient is set according to the system response speed requirements. The integral control term accumulates historical deviations to eliminate steady-state errors. The integral time constant is set according to the system stability requirements. The derivative control term calculates the predictive adjustment amount based on the rate of change of chloride ion concentration. The derivative time constant is set according to the system overshoot suppression requirements. The dynamic adjustment of the electrode output current also needs to consider the physical limitations of the electrodes, including the maximum output current, minimum output current, and current change rate limits. The adjusted electrode output current parameters include the real-time current value of each electrode, the current change trend, and the adjustment status information. The parameter update frequency is set to once per minute to ensure timely response to changes in chloride ion concentration. The safety protection mechanism of the current adjustment monitors the working status of the electrodes. When electrode overload or abnormality is detected, the current output is automatically limited and an alarm is triggered.

[0092] The potential field superposition calculation process re-substitutes the adjusted output currents of each electrode into the potential distribution calculation model, and solves for the updated potential field distribution using numerical methods. The potential field superposition uses the finite element method to establish a three-dimensional potential field model of the seawater region surrounding the thruster. The model boundary conditions include the current density boundary on the thruster surface and the zero potential boundary in the far field. The superposition calculation process first updates the current source intensity of each electrode, and then resolves the Poisson equation to obtain a new potential distribution. The calculation process uses an iterative solution method, obtaining a convergent potential field solution through successive approximations. The convergence criterion is that the potential difference between adjacent iteration steps is less than a set error threshold. The calculation results of the reconstructed potential distribution show the potential changes in each region after adjustment. The potential distribution map intuitively reflects the spatial distribution characteristics of the potential field and the adjustment effect. The potential field superposition also needs to consider the spatial variation of seawater conductivity and the influence of temperature; the non-uniform distribution of conductivity will affect the distribution law of the potential field. The verification of the reconstructed potential distribution is achieved through comparative analysis with measured potential data, verifying the accuracy and reliability of the calculation results.

[0093] The response control process establishes a closed-loop control relationship between the reconstructed potential distribution on the thruster surface and changes in chloride ion concentration, achieving dynamic optimization of the adaptive protection potential field. The response control algorithm monitors the potential values ​​in each region of the reconstructed potential distribution, activating the control mechanism when the potential value deviates from the target protection potential. The target protection potential is set based on the optimal range between the corrosion potential and overprotection potential of the metallic material; different target potentials are set for different regions according to their corrosion risk levels. The control process employs a model predictive control algorithm, predicting future potential changes based on the current potential distribution and chloride ion concentration trends, and adjusting the electrode output in advance to maintain a stable protection effect. The adaptive protection potential field is formed through continuous monitoring, feedback, and adjustment, enabling the potential field distribution to automatically adapt to changes in chloride ion concentration. The adaptive characteristics of the potential field are reflected in its rapid response to environmental changes and automatic optimization of the protection effect; when the chloride ion concentration suddenly increases, the potential field automatically strengthens the protection intensity, and when the concentration decreases, the potential field automatically weakens to save energy. The adaptive control learning mechanism continuously optimizes control parameters by accumulating operational data. The learning algorithm uses a neural network method to establish a nonlinear mapping relationship between chloride ion concentration changes and the optimal potential adjustment amount.

[0094] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0095] By comparing the adaptive protective potential field data with historical corrosion evolution data over time, characteristic parameters of corrosion development trends are obtained.

[0096] The corrosion rate is predicted for the next 24 hours based on the long short-term memory network algorithm to obtain the predicted corrosion rate value.

[0097] The remaining protection time for each area is calculated and evaluated based on the predicted corrosion rate values ​​to obtain the remaining protection time for the propeller blade area, the remaining protection time for the bearing seal area, and the remaining protection time for the static area of ​​the outer casing.

[0098] The remaining protection time is processed by adjusting the protection parameters using a particle swarm optimization algorithm to obtain the current density adjustment amount and the potential adjustment amount.

[0099] The output control of each protective electrode is processed based on the current density adjustment and potential adjustment to obtain the protection parameter optimization command and execute the anti-corrosion control.

[0100] Specifically, time-series comparison processing systematically compares and analyzes current adaptive protective potential field data with stored historical corrosion evolution data to extract the temporal variation patterns and trend characteristics of corrosion development. The adaptive protective potential field data includes parameters such as real-time potential values, potential change rates, potential fluctuation amplitudes, and potential distribution uniformity for each region, with data acquisition frequency of once per hour to form a continuous time series. Historical corrosion evolution data comes from long-term corrosion monitoring records, including information on corrosion rate changes, corrosion depth development, corrosion area expansion, and corrosion morphology evolution, with a data time span covering at least one year of operation. The time-series comparison employs a dynamic time warping algorithm to handle data matching at different time scales, maximizing the similarity between the two time series by finding the optimal time correspondence. The comparison processing identifies the causal relationship between potential field changes and corrosion development, analyzing correlation patterns such as potential decrease and corrosion acceleration, potential fluctuation and corrosion inhomogeneity, and potential gradient and localized corrosion concentration. Extraction of corrosion development trend characteristic parameters includes key indicators such as corrosion rate change gradient, corrosion acceleration, frequency and amplitude of periodic corrosion fluctuations, and spatial propagation speed of corrosion development. The numerical representation of the characteristic parameters uses statistical methods to calculate statistical features such as mean, variance, skewness and kurtosis, while frequency domain analysis is used to extract the periodicity and spectral characteristics of corrosion development.

[0101] Long Short-Term Memory (LSTM) networks, as a neural network architecture in deep learning specifically designed for processing time-series data, possess the ability to learn long-term dependencies and handle sequential change patterns. The network structure comprises an input layer, multiple LSM unit layers, and an output layer. The input layer receives time-series data containing corrosion development trend feature parameters, with each time step containing a multi-dimensional feature vector. The LSM units control the flow of information through gating mechanisms: an input gate, a forget gate, and an output gate. The input gate determines which new information is stored in the cell state, the forget gate determines which old information is discarded, and the output gate determines which part of the cell state is output. The network training process uses historical corrosion data to establish a mapping relationship between feature parameters and future corrosion rates. The training data includes corrosion evolution cases under different marine environmental conditions. The training algorithm employs backpropagation over time, with the loss function being the mean squared error between the predicted and actual corrosion rates. The network hyperparameters include a maximum of 3 hidden layers, 128 neurons per layer, a learning rate of 0.001, and a batch size of 32 samples. The prediction process inputs the corrosion development trend characteristics parameters of the current 24 hours into the trained network model, and the network outputs the predicted corrosion rate values ​​of each region in the next 24 hours. The prediction accuracy is evaluated through cross-validation and reaches an accuracy of over 85%.

[0102] The remaining protection time calculation and assessment, based on the predicted corrosion rate and the current state of the protection system, determines the remaining time that each area can maintain effective protection under existing conditions. The calculation and assessment uses the corrosion allowance divided by the predicted corrosion rate. The corrosion allowance is either the remaining thickness of the protective coating or the allowable corrosion depth of the metal substrate. For the propeller blade area, the corrosion allowance is measured using an ultrasonic thickness gauge; the current remaining thickness is 0.8 mm, and the predicted corrosion rate is 0.02 mm / h, resulting in a calculated remaining protection time of 40 hours. For the bearing seal area, considering sealing performance requirements, the maximum allowable corrosion depth is 0.3 mm, and the predicted corrosion rate is 0.015 mm / h, resulting in a calculated remaining protection time of 20 hours. For the casing stationary area, based on structural strength requirements, the allowable corrosion depth is 1.2 mm, and the predicted corrosion rate is 0.008 mm / h, resulting in a calculated remaining protection time of 150 hours. The calculation of the remaining protection time also needs to consider the uncertainty of the corrosion rate and prediction error; the confidence interval of the prediction results is assessed using Monte Carlo simulation. The introduction of a safety factor multiplies the calculated remaining protection time by a safety factor of 0.8, ensuring reliable protection even under the influence of prediction errors and environmental changes. The remaining protection time is dynamically updated every 6 hours based on changes in real-time monitoring data and prediction results.

[0103] The Particle Swarm Optimization (PSO) algorithm simulates the collective intelligent behavior of a flock of birds foraging to find the optimal protection parameter adjustment scheme to extend the remaining protection time and optimize the protection effect. Algorithm initialization includes setting the particle swarm size to 50 particles, with each particle representing a set of protection parameter adjustment schemes, including current density adjustment and potential regulation for each region. The particle position vector contains 15 dimensions corresponding to parameters such as current density adjustment, potential regulation, and adjustment timing for five regions. The fitness function design comprehensively considers multiple objectives such as the effect of extending the remaining protection time, the degree of energy consumption increase, and system stability. The function expression is: remaining protection time weight multiplied by time extension amount minus energy consumption weight multiplied by energy consumption increase amount minus stability weight multiplied by system fluctuation amount. Particle velocity update adopts the standard PSO algorithm formula, including an inertia term, an individual optimal term, and a global optimal term. The inertia weight is set to 0.9 to represent the degree of maintaining the current direction of motion, and the learning factor is set to 2.0 to represent the learning intensity of the optimal solution. Position update is obtained by adding the velocity vector to the current position vector, and position boundary constraints ensure that the adjustment parameters are within a reasonable range. The optimization iteration process lasts for 100 generations, and the globally optimal solution is selected as the final protection parameter adjustment scheme through fitness evaluation. The output of the adjustment scheme includes a current density adjustment amount, indicating the current output that needs to be increased or decreased for each electrode, and a potential adjustment amount, indicating the adjustment range of the target protection potential for each region.

[0104] The output control processing converts the adjustment parameters obtained from particle swarm optimization into specific electrode control commands, achieving precise adjustment of the output current and potential of each protection electrode. Current density adjustment is executed by regulating the output current of each electrode through a current controller, with an adjustment accuracy of 1 mA and a response time of less than 5 seconds. Potential adjustment is executed by adjusting the electrode's operating potential setpoint, with a potential control accuracy of 1 mV and a control stability error of less than 2%. Protection parameter optimization commands include detailed information such as adjustment target value, adjustment rate, adjustment sequence, and safety limits. The command format uses a standardized digital communication protocol to ensure correct parsing and execution by the control system. The execution process employs a step-by-step adjustment strategy to avoid sudden parameter changes impacting system stability; each adjustment is limited to within 10% of the target value. Closed-loop feedback for corrosion protection control verifies the execution results of the optimization commands by real-time monitoring of the adjusted protection effect. If the adjustment effect is found to be unsatisfactory, a secondary optimization adjustment is initiated. The control system's fault protection mechanism monitors the operating status of each electrode. When an electrode fault or abnormality is detected, it automatically switches to a backup electrode or safety mode to ensure the continuity and reliability of corrosion protection.

[0105] The above describes the method for protecting the thruster from corrosion in a deep-sea environment in the embodiments of this application. The following describes the system for protecting the thruster from corrosion in a deep-sea environment in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the deep-sea environment propulsion corrosion protection system in this application includes:

[0106] The data acquisition module is used to collect and process real-time data of the seawater around the deep-sea thruster through a corrosion monitoring sensor array, and obtain an environmental corrosion dataset including chloride ion concentration, dissolved oxygen concentration, pH value and corrosion potential.

[0107] The grading module is used to perform corrosion risk grading calculation on the propeller surface based on the environmental corrosion dataset using a genetic algorithm, and to obtain the corrosion risk level of the propeller blade area, bearing sealing area and shell stationary area.

[0108] The allocation module is used to optimize the current allocation of the corrosion risk level using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode.

[0109] The control module is used to reconstruct and control the surface potential field of the thruster according to the current density distribution parameters of each protective electrode, so as to obtain an adaptive protective potential field based on chloride ion concentration.

[0110] The analysis module is used to perform corrosion trend analysis on the adaptive protection potential field through a prediction model, obtain protection parameter optimization instructions, and execute anti-corrosion control.

[0111] above Figure 2The corrosion protection system for thrusters in the deep-sea environment in this embodiment of the invention is described in detail from the perspective of modular functional entities. The corrosion protection device for thrusters in the deep-sea environment in this embodiment of the invention is described in detail from the perspective of hardware processing.

[0112] Reference Figure 3 This invention also provides a propulsion corrosion protection device for deep-sea environments. This device can be a server, and its internal structure can be as follows: Figure 3 As shown. This deep-sea environment thruster corrosion protection device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor in this computer design provides computing and control capabilities. The memory of this deep-sea environment thruster corrosion protection device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of this deep-sea environment thruster corrosion protection device stores the data corresponding to this embodiment. The network interface of this deep-sea environment thruster corrosion protection device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0113] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the anti-corrosion protection device for thrusters in deep-sea environments where the present invention is applied.

[0114] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the method for protecting the thruster from corrosion in a deep-sea environment.

[0115] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0116] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a deep-sea environment thruster corrosion protection device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0117] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for corrosion protection of thrusters in deep-sea environments, characterized in that, The method includes: Real-time data acquisition and processing of seawater around the deep-sea thruster was performed using a corrosion monitoring sensor array to obtain an environmental corrosion dataset containing chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential. Based on the environmental corrosion dataset, a genetic algorithm was used to calculate the corrosion risk level of the propeller surface, and the corrosion risk levels of the propeller blade area, bearing sealing area and shell stationary area were obtained. The corrosion risk level is optimized by using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode, including: configuring the number of protective electrodes according to the corrosion risk level to obtain the layout scheme of the propeller blade area electrode, the bearing sealing area electrode and the shell stationary area electrode. The potential field on the thruster surface is reconstructed and controlled based on the current density distribution parameters of each protective electrode to obtain an adaptive protective potential field based on chloride ion concentration. This includes: calculating the potential distribution of the current density distribution parameters of each protective electrode according to their spatial location to obtain the initial potential value at each point on the thruster surface; performing concentration gradient correction processing on the initial potential value based on chloride ion concentration monitoring data to obtain a chloride ion concentration correction coefficient; the chloride ion concentration correction coefficient is calculated using a piecewise function method, where a negative correction coefficient indicates a need to lower the protective potential when the chloride ion concentration is lower than the standard seawater concentration, and a positive correction coefficient indicates a need to increase the protective potential when the chloride ion concentration is higher than the standard seawater concentration; dynamically adjusting the output current of each protective electrode according to the chloride ion concentration correction coefficient to obtain adjusted electrode output current parameters; calculating the adjusted electrode output current parameters through potential field superposition to obtain the reconstructed potential distribution on the thruster surface; and responding to changes in chloride ion concentration based on the reconstructed potential distribution on the thruster surface to obtain the adaptive protective potential field based on chloride ion concentration. The adaptive protective potential field is processed by corrosion trend analysis using a prediction model to obtain optimized protection parameter instructions and execute anti-corrosion control. This includes: comparing the adaptive protective potential field data with historical corrosion evolution data over time to obtain corrosion development trend characteristic parameters; predicting the corrosion rate for the next 24 hours based on a long short-term memory network algorithm to obtain predicted corrosion rate values; calculating and evaluating the remaining protection time for each region based on the predicted corrosion rate values ​​to obtain the remaining protection time for the propeller blade area, the bearing seal area, and the static area of ​​the outer casing; adjusting the protection parameters using a particle swarm optimization algorithm to obtain current density adjustment and potential adjustment; and controlling the output of each protective electrode based on the current density adjustment and potential adjustment to obtain the optimized protection parameter instructions and execute anti-corrosion control.

2. The method for corrosion protection of thrusters in deep-sea environments according to claim 1, characterized in that, The method involves real-time data acquisition and processing of seawater surrounding the deep-sea thruster using a corrosion monitoring sensor array to obtain an environmental corrosion dataset containing chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential, including: The chloride ion selective electrode sensor, dissolved oxygen fluorescence sensor, pH glass electrode sensor, and corrosion potential reference electrode sensor are deployed at key locations on the surface of the thruster according to a preset array layout. Data was read and processed from each sensor at 30-second intervals to obtain the raw values ​​of chloride ion concentration, dissolved oxygen concentration, pH value and corrosion potential. The raw values ​​of chloride ion concentration, dissolved oxygen concentration, pH value, and corrosion potential are input into the data preprocessing module for filtering and calibration to obtain chloride ion concentration calibration data, dissolved oxygen concentration calibration data, pH value calibration data, and corrosion potential calibration data. Outlier removal was performed on the chloride ion concentration calibration data, dissolved oxygen concentration calibration data, pH value calibration data, and corrosion potential calibration data to obtain valid corrosion parameter data. The effective corrosion parameter data are combined and processed according to the time series to obtain the environmental corrosion dataset.

3. The method for corrosion protection of thrusters in deep-sea environments according to claim 1, characterized in that, The corrosion risk classification of the propeller surface is calculated using a genetic algorithm based on the environmental corrosion dataset, resulting in corrosion risk levels for the propeller blade area, bearing sealing area, and static area of ​​the outer casing, including: The environmental corrosion dataset is processed by region mapping according to the propeller geometry to obtain partitioning parameters for the propeller blade region, bearing seal region, and shell stationary region. Based on the pressure correction coefficient, the partition parameters are subjected to environmental correction processing to obtain the corrected corrosion parameters; The risk index for each region is obtained by weighted summation of chloride ion concentration, dissolved oxygen concentration, pH value, and stress factor. The risk index is initialized and its fitness is evaluated using a genetic algorithm to obtain weight parameters. The risk index is graded based on the weighted parameters to obtain the corrosion risk level of the propeller blade area, bearing sealing area and shell stationary area.

4. The method for corrosion protection of thrusters in deep-sea environments according to claim 1, characterized in that, The step of optimizing the current distribution of each protective electrode by applying the Lagrange algorithm to the corrosion risk level, and obtaining the current density distribution parameters of each electrode, includes: The number of protective electrodes is configured according to the corrosion risk level to obtain a layout scheme for the propeller blade area electrodes, bearing seal area electrodes, and housing stationary area electrodes. The layout scheme is processed by assigning electrode weights according to the regional risk level to obtain the electrode weights of high-risk areas, medium-risk areas and low-risk areas. Based on the corrosion risk level of each region, the protection area is allocated and calculated to obtain the areas of high-risk, medium-risk, and low-risk zones. Based on the protection area, the current constraint is solved using the Lagrange multiplier method to obtain the distribution coefficient that satisfies the total current conservation condition; The current density distribution parameters of each protection electrode are obtained by multiplying the distribution coefficient with the area of ​​each region.

5. The method for corrosion protection of thrusters in deep-sea environments according to claim 4, characterized in that, The step of configuring the number of protective electrodes according to the corrosion risk level to obtain a layout scheme for the electrodes in the propeller blade area, the bearing seal area, and the housing stationary area includes: The electrode density is calculated according to the corrosion risk level to obtain the electrode density of high-risk area, medium-risk area and low-risk area. Based on the electrode density, the number of electrodes in each region is allocated according to the surface area of ​​each region to obtain the number of electrodes in the propeller blade region, the number of electrodes in the bearing seal region, and the number of electrodes in the static region of the outer casing. Based on the geometric features of each region, the electrode positions are optimized and distributed to obtain the positions of the leading edge electrode, the electrode positions around the bearing, and the electrode positions on the outer shell surface. The electrode positions are processed according to the current transmission distance to plan the connection path, and the cable connection path of each electrode is obtained. The electrodes are then integrated and laid out according to the cable connection path to obtain the layout scheme of the propeller blade area electrodes, bearing seal area electrodes, and housing stationary area electrodes.

6. A corrosion protection system for thrusters in deep-sea environments, characterized in that, For implementing the method for corrosion protection of thrusters in deep-sea environments as described in any one of claims 1-5, the corrosion protection system for thrusters in deep-sea environments comprises: The data acquisition module is used to collect and process real-time data of the seawater around the deep-sea thruster through a corrosion monitoring sensor array to obtain an environmental corrosion dataset containing chloride ion concentration, dissolved oxygen concentration, pH value and corrosion potential. The grading module is used to perform corrosion risk grading calculation on the propeller surface based on the environmental corrosion dataset using a genetic algorithm, and to obtain the corrosion risk level of the propeller blade area, bearing sealing area and shell stationary area. The allocation module is used to optimize the current allocation of the corrosion risk level using the Lagrange algorithm to obtain the current density distribution parameters of each protective electrode. The control module is used to reconstruct and control the surface potential field of the thruster according to the current density distribution parameters of each protective electrode, so as to obtain an adaptive protective potential field based on chloride ion concentration. The analysis module is used to perform corrosion trend analysis on the adaptive protection potential field through a prediction model, obtain protection parameter optimization instructions, and execute anti-corrosion control.

7. A corrosion protection device for thrusters in deep-sea environments, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method for corrosion protection of a thruster in a deep-sea environment as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to perform the method for protecting the thruster from corrosion in a deep-sea environment as described in any one of claims 1 to 5.

Images