Ship coating remote monitoring and data management method and system

Abstract

The present application relates to the technical field of ship maintenance, in particular to a ship coating remote monitoring and data management method and system, which establishes a digital gene file for the ship coating; through a distributed sensing network deployed on the ship body, coating state and environmental data are collected in real time and edge side anomaly diagnosis is performed; multi-source data is fused, and the coating health trend and failure risk are analyzed and predicted based on a physical and data-driven hybrid model; based on the prediction results, the sailing plan, the resource and the cost model, predictive maintenance decisions are optimized and generated; and through technologies such as block chain, full-cycle data asset management is realized. The corresponding system includes a shipborne sensing network, a cloud analysis platform, a block chain data service and a client application. The present application realizes a paradigm change from passive repair to active health management, significantly prolongs the early warning period, optimizes the maintenance decision, reduces the whole life cycle cost, and forms a data value closed loop that can be fed back to the research and development, manufacturing and insurance links.

Description

Technical Field

[0001] This invention relates to the field of ship maintenance technology, and in particular to a method and system for remote monitoring and data management of ship painting. Background Technology

[0002] The condition of a ship's coating directly affects its structural safety, operational efficiency, and maintenance costs. Currently, monitoring ship coatings mainly relies on periodic manual inspections or dockside checks. This approach suffers from significant delays, strong subjectivity, fragmented data, and an inability to provide early warnings. Although some remote monitoring attempts have emerged, such as using fixed sensors to monitor coating resistance or thickness, these methods are typically single-function and provide isolated data, offering only limited and fragmented status information. They cannot construct a full lifecycle health profile of the coating, let alone enable predictive maintenance and data-driven decision optimization. Coating failure often results in irreversible substrate corrosion by the time it becomes visible to the naked eye, leading to high maintenance costs and unplanned downtime.

[0003] Therefore, there is an urgent need for a systematic solution that can deeply integrate perception, analysis, prediction and management to achieve intelligent, precise and forward-looking management of ship coating. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a remote monitoring and data management method and system for ship coating. By integrating multi-dimensional data and using hybrid model prediction, the early warning time for coating failure is significantly reduced from several months in traditional methods, enabling true predictive maintenance. It constructs a full-cycle data chain from coating selection, construction monitoring, in-service maintenance to scrap analysis, forming a value loop that can provide feedback to optimize upstream processes. This significantly improves the efficiency of maintenance resource utilization, reduces the total lifecycle maintenance cost, and greatly enhances the transparency and precision of ship asset management.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] In a first aspect, the present invention provides a method for remote monitoring and data management of ship painting, comprising the following steps:

[0007] Collect batch composition data, construction process parameters and initial environmental data of the coatings used in the coating process of the target ship, process and store them in association, and generate a unique digital gene file of coating for each independent coating area of ​​the ship.

[0008] Sensing nodes are deployed in the coating area of ​​the ship to periodically collect multidimensional state data and local microenvironment data of the coating in the area, and perform local preprocessing and anomaly diagnosis.

[0009] Data from multiple sources, including sensing nodes, Automatic Identification System (AIS), and navigation weather services, is aggregated and spatiotemporally aligned and fused with the corresponding region's coating digital gene archive. Based on the fused dataset, a cloud-based analysis engine is used to comprehensively assess the coating's health status and construct a performance degradation trend model.

[0010] Based on the output of the performance degradation trend model, the failure risk probability of the target coating area within a preset time period is predicted; based on the failure risk probability, the ship's scheduled voyage plan, global port and ship repair resource information, and maintenance cost model, a comprehensive evaluation and optimization are performed to generate at least one recommended maintenance decision scheme.

[0011] By linking and storing monitoring data, analysis results, and maintenance operation records throughout the entire coating lifecycle, and conducting data mining, a data closed loop is formed to optimize coating research and development, coating processes, maintenance strategies, and asset assessment.

[0012] As a further embodiment of the present invention, the coating digital gene profile includes at least the following data fields: a unique material identifier code of the coating resin system, a pigment volume concentration value, a curing temperature-time curve, a substrate surface roughness during construction, a temperature and humidity sequence of the construction environment, and an initial dry film thickness distribution map collected by an ultrasonic thickness gauge.

[0013] As a further aspect of the present invention, the sensing node employs a negative learning strategy for preliminary anomaly diagnosis, specifically including the following sub-steps:

[0014] During the healthy service life of the coating, the multidimensional state data is continuously collected, time domain and frequency domain features are extracted, and a normal state feature library of the coating is constructed for the sensing nodes.

[0015] In subsequent periodic monitoring, the feature vectors collected and extracted in real time will be compared with the historical features in the coating normal state feature library for similarity.

[0016] When the deviation between real-time features and historical features exceeds a preset threshold, it is determined to be a preliminary anomaly, triggering the sensing node to upload encrypted detailed monitoring data packets to the cloud.

[0017] As a further aspect of the present invention, the cloud-based analytics engine employs a hybrid modeling method combining physical models and data-driven models, and the construction of the performance degradation trend model includes:

[0018] The effect of temperature on the coating aging rate is simulated based on the Arrhenius formula, and the moisture penetration process is simulated based on Fick's diffusion law. The output of the physical driving sub-model is the physical degradation fraction. ;

[0019] Using historical data on coating condition, environmental stress, and navigation conditions as input features, and measured performance degradation status as labels, a machine learning model is trained. The output of the data-driven sub-model is the data degradation score. ;

[0020] The outputs of the physics-driven sub-model and the data-driven sub-model are weighted and fused to obtain the comprehensive coating health index. The calculation formula is: ,in, and These are weight coefficients that are dynamically adjusted based on the model confidence level, and .

[0021] As a further aspect of the present invention, generating at least one recommended maintenance decision scheme includes the following steps:

[0022] Define the objective function as minimizing the total expected cost. , ,in, For direct maintenance costs, Costs of ship downtime This is based on the probability of failure and the estimated risk cost caused by coating failure;

[0023] Define decision variables and constraints. Decision variables include the maintenance time window. Repair Location Repair methods The constraints include: the probability of failure is below the safety threshold, the maintenance time window is compatible with the ship's schedule, and the maintenance location has the necessary resources.

[0024] Based on the failure risk probability prediction results, feasible combinations of decision variables are enumerated under the constraints, and the total expected cost corresponding to each combination is calculated. ;

[0025] Select the option that maximizes the total expected cost The combination of decision variables with the smallest minimum value is taken as the optimal recommended solution, and the maintenance method is described above. This includes targeted micro-repairs and planned dry-dock overhauls.

[0026] As a further aspect of the present invention, the targeted micro-repair includes: after the decision is determined, automatically generating a work order containing the geographical coordinates of the area to be repaired, the damage type, and recommended repair materials, and sending it through an application programming interface to the control system of an underwater robot or a wall-climbing robot that is waiting at a designated port or can be carried by a mother ship.

[0027] As a further aspect of the present invention, when storing and mining monitoring data, analysis results, and maintenance operation records throughout the entire life cycle of the coating, blockchain is used to generate data hashes for key data of each monitoring alarm, each health assessment result, and each maintenance operation. The data hash value, along with the timestamp and data identifier, is written into an immutable distributed ledger to form a digital health passport for the coating area.

[0028] Secondly, the present invention also provides a remote monitoring and data management system for ship painting, comprising:

[0029] The shipborne sensing network subsystem consists of multiple sensing nodes deployed on the ship's hull, a shipborne gateway, and an edge computing unit. It is used to deploy sensing nodes in the coating area of ​​the ship, periodically collect multi-dimensional state data and local micro-environment data of the coating in the area, and perform local preprocessing and anomaly diagnosis.

[0030] The cloud-based intelligent analysis platform, comprising a data lake warehouse, a digital twin engine, a hybrid AI analysis center, and a decision optimization module, is used to collect batch composition data, construction process parameters, and initial environmental data of the coatings used during the target ship's painting process. This data is processed, correlated, and stored to generate a unique digital genetic profile for each independent coating area of ​​the ship. The platform aggregates multi-source data from sensing nodes, the Automatic Identification System (AIS), and navigation weather services, and performs spatiotemporal alignment and data fusion with the corresponding region's digital genetic profile. Based on the fused dataset, the cloud-based analysis engine comprehensively assesses the coating's health status and constructs a performance degradation trend model. According to the output of the performance degradation trend model, the platform predicts the failure probability of the target coating area within a predetermined time period. Based on the failure probability, the ship's scheduled voyage plan, global port and repair resource information, and a maintenance cost model, a comprehensive evaluation and optimization are performed to generate at least one recommended maintenance decision plan.

[0031] The blockchain data service subsystem is used to provide data storage and traceability services, and supports the association and storage of monitoring data, analysis results and maintenance operation records throughout the entire life cycle of the coating, as well as data mining, to form a data closed loop for optimizing coating research and development, coating process, maintenance strategy and asset assessment.

[0032] The client-side interactive application provides administrators with a visual interface for monitoring status, receiving alarms, reviewing reports, and issuing maintenance instructions.

[0033] As a further embodiment of the present invention, the sensing node includes at least a microprocessor, a self-powered module, a sensor array, and a low-power communication module; the sensor array includes a pulsed eddy current sensor for detecting early corrosion under the coating, a miniature fiber optic grating sensor for monitoring coating adhesion strain, and an electrochemical sensor for detecting surface salt and contaminant ion concentration.

[0034] As a further embodiment of the present invention, the decision optimization module is connected to an external maintenance resource scheduling platform through a standardized interface. When the decision scheme is targeted micro-repair, the decision optimization module automatically sends a service request to the maintenance resource scheduling platform. The service request includes at least the repair location, time window, and robot operation instructions.

[0035] Compared with existing technologies, the remote monitoring and data management method and system for ship painting provided by this invention has the following advantages:

[0036] The remote monitoring and data management method and system for ship coatings of this invention transforms ship coatings from static, consumable anti-corrosion materials into digital management of the entire coating lifecycle by establishing digital genetic files and a real-time sensing network. By combining material genes, hybrid models, and multi-source sensing, the early warning time of coating failure is shortened, facilitating precise targeted micro-repair and reducing maintenance costs. Through keen detection of early corrosion under the coating, potential safety accidents caused by structural corrosion are effectively prevented. The distributed ganglion architecture and negative learning strategy adopted by the sensing network enable the system to have edge intelligence and resilience. A single point of failure does not affect the whole system, and the amount of data transmission is reduced. It is particularly suitable for the high bandwidth cost and low power consumption application scenarios of ocean-going vessels, realizing real-time status perception, accurate failure prediction, optimized maintenance decisions, and maximized asset value.

[0037] These or other aspects of the invention will become more apparent from the following description of embodiments. It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. In the drawings:

[0039] Figure 1 This is a flowchart of a remote monitoring and data management method for ship painting according to the present invention.

[0040] Figure 2This is a flowchart illustrating the preliminary anomaly diagnosis of a sensing node in a remote monitoring and data management method for ship painting according to the present invention.

[0041] Figure 3 This is a flowchart illustrating the construction of a performance degradation trend model in a remote monitoring and data management method for ship painting according to the present invention.

[0042] Figure 4 This is a flowchart illustrating the generation of at least one recommended maintenance decision scheme in a remote monitoring and data management method for ship painting according to the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0044] The technical solutions in the exemplary embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described exemplary embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Specifically, the embodiments of this application will be further described below with reference to the accompanying drawings.

[0046] See Figure 1 As shown in the figure, embodiments of this application also provide a method for remote monitoring and data management of ship painting, including the following steps:

[0047] Step S10: Collect batch composition data, construction process parameters and initial environmental data of the paint used in the coating process of the target ship, process and store them together to generate a unique coating digital gene file for each independent coating area of ​​the ship.

[0048] In this step, the coating digital gene profile includes at least the following data fields: unique material identification code of the coating resin system, pigment volume concentration value, curing temperature-time curve, substrate surface roughness during construction, temperature and humidity sequence of the construction environment, and initial dry film thickness distribution map collected by an ultrasonic thickness gauge.

[0049] During data collection, IoT devices automatically collect temperature, humidity, and dew point data of the construction environment; batch information on paint cans is read using QR code / RFID scanners and linked to Material Safety Data Sheets (MSDS) and formulation data in the backend database; an automated spraying robot equipped with LiDAR and a high-precision temperature sensing array records the spraying path, speed, film thickness distribution, and curing temperature curve. When generating the file, a unique digital file is created on the cloud platform for each area with independent coating design requirements, such as area A101 below the port waterline. The collected structured and unstructured data are linked and stored, such as film thickness distribution heatmaps, and the theoretical life curve of the coating system is calculated using a material gene library.

[0050] For example, consider the ballast tank painting of a 300,000-ton oil tanker (VLCC).

[0051] Before construction, a unique QR code is generated for each compartment. When the paint can is opened, the worker scans the QR code on the can, and the system automatically associates the following information for that batch of epoxy paint: resin type: bisphenol A epoxy, curing agent type, pigment volume concentration: PVC=35%, theoretical solids content.

[0052] During construction, environmental sensors installed inside the cabin continuously record: temperature, 22-25℃; humidity, 45-50%. Each time the automated spraying robot completes a spraying sector, it uploads the measured average film thickness of 320μm and uniformity data for that area.

[0053] After construction, the system automatically integrates all data to generate a "coating digital gene profile" for the ballast tank, with the profile ID "VLCC001-BT-No.5-P". This profile contains all the aforementioned data fields, as well as a theoretical performance degradation curve under standard conditions, which is initially calculated based on the Arrhenius formula.

[0054] Step S20: Deploy sensing nodes in the coating area of ​​the ship to periodically collect multi-dimensional state data and local micro-environment data of the coating in the area, and perform local preprocessing and anomaly diagnosis.

[0055] In this step, during hardware deployment, the sensing nodes are designed to be explosion-proof and are fixed to the coating surface or interior via magnetic bases. The sensing nodes integrate a microprocessor, a terahertz reflective array sensor for detecting under-corrosion, an impedance spectrum sensor for assessing coating water absorption, a temperature / humidity sensor, and a low-power wide-area network communication module. See also... Figure 2 As shown, the perception node uses a negative learning strategy for preliminary anomaly diagnosis, which includes the following sub-steps:

[0056] Step S201: During the healthy service period of the coating, continuously collect the multi-dimensional state data, extract time-domain and frequency-domain features, and construct a normal state feature library of the sensing node.

[0057] During the ship's health period (three months prior to commissioning), raw sensor signals were collected hourly. At the edge, a Fast Fourier Transform was performed on the signals to extract the frequency domain features of the dominant frequency amplitude, and the time domain statistical features of the root mean square value were calculated. The feature vector set under normal operating conditions was stored as a local feature library.

[0058] Step S202: In subsequent periodic monitoring, the feature vectors collected and extracted in real time are compared with the historical features in the coating normal state feature library for similarity.

[0059] During monitoring, the average Euclidean distance of the 100 most recent historical vectors extracted in real time from the 6-dimensional feature vector feature library is calculated.

[0060] Step S203: When the deviation between real-time features and historical features exceeds a preset threshold, it is determined to be a preliminary anomaly, triggering the sensing node to upload the encrypted detailed monitoring data packet to the cloud.

[0061] By training with historical data and setting a threshold T=1.5, when the average Euclidean distance is greater than the set threshold T, it is judged as abnormal. The node immediately starts the high-definition diagnostic mode, collecting 10 seconds of data once per second, and uploading it to the cloud along with the timestamp and location information via satellite communication link.

[0062] The aforementioned 300,000-ton VLCC was sailing in the high-temperature, high-salinity region of the Indian Ocean. Node "SN-202," deployed near the starboard waterline, detected a slow but continuous drift in the coating capacitance value using its impedance spectroscopy sensor. Over three consecutive periods, the calculated deviations between the real-time feature vector and the historical feature database were 1.2, 1.6, and 1.8, respectively. When the calculated deviation exceeded the threshold of 1.5 for the third time, the node was flagged as abnormal and immediately uploaded an encrypted data packet to the cloud platform.

[0063] Step S30: Aggregate multi-source data from sensing nodes, automatic identification systems for ships, and navigation weather services, and perform spatiotemporal alignment and data fusion with the coating digital gene archives of the corresponding regions. Based on the fused dataset, conduct a comprehensive assessment of the coating health status through a cloud-based analysis engine, and construct a performance degradation trend model.

[0064] During data fusion, the cloud platform receives abnormal data packets uploaded by nodes and automatically obtains the ship's current AIS position, speed, and heading via API, as well as real-time seawater temperature, salinity, and wind speed data from meteorological service providers. Based on timestamps and location information, the platform spatiotemporally aligns all data with the coating's digital genetic profile.

[0065] In this embodiment, see Figure 3 As shown, the cloud-based analytics engine employs a hybrid modeling approach combining physical models and data-driven models. The construction of the performance degradation trend model includes:

[0066] Step S301: Simulate the effect of temperature on the coating aging rate based on the Arrhenius formula, and simulate the moisture penetration process based on Fick's diffusion law. The output of the physical driving sub-model is the physical degradation fraction. .

[0067] The process involves calling up material parameters from the coating archive, substituting the current environmental stress (temperature 25°C, salinity 3.5%) into a preset physical aging equation, calculating the theoretical aging rate of the coating under the current conditions, and outputting the physical degradation score. The physical degradation score ranges from 0 to 1, where 1 represents health.

[0068] Step S302: Using coating condition, environmental stress, and navigation conditions from historical datasets as input features, and measured performance degradation status as labels, train a machine learning model. The output of the data-driven sub-model is the data degradation score. Among them, the data degradation score

[0069] Step S303: Weight and fuse the outputs of the physics-driven sub-model and the data-driven sub-model to obtain the comprehensive coating health index. The calculation formula is: ,in, and These are weight coefficients that are dynamically adjusted based on the model confidence level, and .

[0070] Set weights , The overall coating health index The system determines the health status as "warning", with preset thresholds: >0.8 for healthy, 0.6-0.8 for attentive, and <0.6 for warning.

[0071] In response to the anomaly at node "SN-202", the cloud platform initiated a hybrid model analysis. The physical model, based on diffusion theory, calculated that the accelerated loss of plasticizer within the coating at the current high temperature led to impedance changes, predicting a performance score S_p = 0.65. The data model, matched against historical case databases, determined that under similar flight paths and sensor signal patterns, there was a 70% probability of microcracks appearing in the coating, outputting S_d = 0.45. After fusion, HI = 0.59, and the system generated a warning: "VLCC001 starboard waterline R14 area, coating health index 0.59, the risk of performance deteriorating below the critical value within 90 days is 75%."

[0072] Step S40: Based on the output of the performance degradation trend model, predict the failure risk probability of the target coating area within a preset time period in the future; based on the failure risk probability, the ship's scheduled voyage plan, global port and ship repair resource information, and maintenance cost model, conduct a comprehensive evaluation and optimization to generate at least one recommended maintenance decision scheme.

[0073] In this step, see Figure 4 As shown, generating at least one recommended maintenance decision scheme includes the following steps:

[0074] Step S401: Define the objective function as minimizing the total expected cost. , ,in, For direct maintenance costs, Costs of ship downtime This is based on the probability of failure and the estimated risk cost caused by coating failure;

[0075] Step S402: Define decision variables and constraints. Decision variables include the maintenance time window. Repair Location Repair methods The constraints include: the probability of failure is lower than the safety threshold, the maintenance time window is compatible with the ship's schedule, and the maintenance location has the necessary resources.

[0076] Step S403: Based on the failure risk probability prediction results, enumerate the feasible decision variable combinations under the constraints, and calculate the total expected cost corresponding to each combination. ;

[0077] Step S404: Select the option that maximizes the total expected cost. The minimum combination of decision variables is taken as the optimal recommended solution, and the maintenance method is... This includes targeted micro-repairs and planned dry-dock overhauls.

[0078] Targeted micro-repair includes: after the decision is made, automatically generating a work order containing the geographical coordinates of the area to be repaired, the damage type, and recommended repair materials, and sending it through an application programming interface to the control system of an underwater robot or wall-climbing robot that is waiting at a designated port or can be carried by a mother ship.

[0079] During cost optimization calculations, the decision optimization module received an alert: Region R14, 75% risk over 90 days. The module automatically retrieved the ship's voyage plan for the next 6 months: for example, planned to arrive in Singapore in 60 days and in Rotterdam in 120 days. Then, it enumerated feasible options:

[0080] Option A: Deploy an underwater robot for targeted micro-repair at the Singapore anchorage. Estimated costs: Robot service fee 5,000, paint 500, no ship schedule loss. Risk cost: 75% risk * (10,000 potential corrosion loss) = 7,500. Total expected cost. =$5,000 + $0 + $7,500 = $12,500.

[0081] Option B: Dry dock repairs in Rotterdam. Estimated costs: Dry dock repair fee 50,000, paint 1,000, loss of 100,000 for 2 days of downtime. At this point, the risk has occurred; the risk cost is set at 10,000. =$51,000 + $100,000 + $10,000 = $161,000.

[0082] Option C: Take no action and continue sailing. This option carries extremely high risk and cost. ≈75% * (structural corrosion loss $500,000 + accidental shutdown loss $200,000) = $525,000.

[0083] Therefore, Option A has the lowest total cost. The system automatically generates a decision recommendation: "It is recommended to use an underwater robot to perform targeted micro-repairs on area R14 when the vessel arrives in Singapore in 60 days," along with a detailed cost analysis. The system generates a detailed work order for Option A and sends it to the scheduling platform of the partner service provider "RoboMarine" in Singapore via API. The work order includes: vessel MMSI, appointment time window, GPS coordinates of the work area, damage type, recommended repair coating type and dosage.

[0084] Step S50: Link and store monitoring data, analysis results and maintenance operation records throughout the entire life cycle of the coating and perform data mining to form a data closed loop for optimizing coating research and development, coating process, maintenance strategy and asset assessment.

[0085] In this step, when the monitoring data, analysis results and maintenance operation records throughout the entire life cycle of the coating are associated, stored and data-mined, the blockchain is used to generate data hashes for key data of each monitoring alarm, each health assessment result and each maintenance operation. The data hash value, along with the timestamp and data identifier, is written into an immutable distributed ledger to form a digital health passport for the coating area.

[0086] For the critical event of "R14 Area Warning," the system writes a hash value generated from a timestamp, vessel ID, area ID, health index (HI), and decision recommendations into a permissioned blockchain network. Shipowners, classification societies, and insurance companies, acting as network nodes, jointly maintain the ledger, preventing any single party from unilaterally altering the records. After accumulating sufficient VLCC ballast tank coating data, the data analysis center discovered that on specific routes, Brand B coatings had an average lifespan 20% longer than Brand A coatings. This analysis report can be sold to coating manufacturer C to improve their products. Simultaneously, the vessel's excellent "coating health passport" can be used to apply for premium discounts from insurance companies. For example, a year later, if the shipowner plans to sell the 300,000-ton VLCC, the buyer can authorize access to the blockchain summary of the vessel's coating data. Clear and tamper-proof maintenance records demonstrate the excellent condition of the vessel's coatings, thereby enhancing the vessel's asset valuation and transaction confidence.

[0087] This invention completes holographic digital modeling of coating information at the source of construction, laying the data foundation for full lifecycle management; by deploying an intelligent sensor network with edge computing capabilities, it uses negative learning to achieve real-time, sensitive, and low-power monitoring of coating status; by aggregating multi-dimensional data and using a hybrid model of physical mechanisms and data-driven approaches, it penetrates the phenomenon to the essence, achieving a leap from monitoring to prediction; by deeply integrating the prediction results with business rules, it outputs executable and optimal maintenance action plans through a global optimization algorithm; by using blockchain to ensure data trustworthiness, and through data mining and sharing, it drives collaborative optimization of upstream and downstream links in coating R&D, maintenance processes, and insurance finance, forming a sustainable industrial intelligent ecosystem.

[0088] See Figure 2 As shown in the embodiments of this application, a remote monitoring and data management system for ship painting is also provided. The system includes:

[0089] The shipborne sensing network subsystem consists of multiple sensing nodes deployed on the ship's hull, a shipborne gateway, and an edge computing unit. It is used to execute step S20 of the aforementioned ship coating remote monitoring and data management method: deploying sensing nodes in the coating area of ​​the ship, periodically collecting multi-dimensional status data and local micro-environment data of the coating in the area, and performing local preprocessing and anomaly diagnosis.

[0090] The cloud-based intelligent analysis platform, comprising a data lake warehouse, a digital twin engine, a hybrid AI analysis center, and a decision optimization module, is used to execute steps S10, S30, and S40 of the aforementioned remote monitoring and data management method for ship coating: It collects batch composition data, construction process parameters, and initial environmental data of the coatings used during the target ship's coating process; processes and stores these data in association; and generates a unique digital gene profile for each independent coating area of ​​the ship. It aggregates multi-source data from sensing nodes, the Automatic Identification System (AIS), and navigation weather services, and performs spatiotemporal alignment and data fusion with the corresponding region's digital gene profile. Based on the fused dataset, it comprehensively assesses the coating's health status through the cloud-based analysis engine and constructs a performance degradation trend model. Based on the output of the performance degradation trend model, it predicts the failure risk probability of the target coating area within a preset time period. Based on the failure risk probability, the ship's planned voyage, global port and ship repair resource information, and a maintenance cost model, it performs a comprehensive evaluation and optimization to generate at least one recommended maintenance decision scheme.

[0091] The blockchain data service subsystem is used to provide data storage and traceability services, and supports step S50 of the aforementioned remote monitoring and data management method for ship coating: to associate and store monitoring data, analysis results and maintenance operation records throughout the entire life cycle of the coating and to perform data mining to form a data closed loop for optimizing coating research and development, coating process, maintenance strategy and asset assessment.

[0092] The client-side interactive application provides administrators with a visual interface for monitoring status, receiving alarms, reviewing reports, and issuing maintenance instructions.

[0093] The remote monitoring and data management method and system for ship coatings of this invention transforms ship coatings from static, consumable anti-corrosion materials into digital management of the entire coating lifecycle by establishing digital genetic files and a real-time sensing network. By combining material genes, hybrid models, and multi-source sensing, the early warning time of coating failure is shortened, facilitating precise targeted micro-repair and reducing maintenance costs. Through keen detection of early corrosion under the coating, potential safety accidents caused by structural corrosion are effectively prevented. The distributed ganglion architecture and negative learning strategy adopted by the sensing network enable the system to have edge intelligence and resilience. A single point of failure does not affect the whole system, and the amount of data transmission is reduced. It is particularly suitable for the high bandwidth cost and low power consumption application scenarios of ocean-going vessels, realizing real-time status perception, accurate failure prediction, optimized maintenance decisions, and maximized asset value.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for remote monitoring and data management of ship painting, characterized in that, Includes the following steps: Collect batch composition data, construction process parameters and initial environmental data of the coatings used in the coating process of the target ship, process and store them in association, and generate a unique digital gene file of coating for each independent coating area of ​​the ship. Sensing nodes are deployed in the coating area of ​​the ship to periodically collect multidimensional state data and local microenvironment data of the coating in the area, and perform local preprocessing and anomaly diagnosis. Data from multiple sources, including sensing nodes, Automatic Identification System (AIS), and navigation weather services, is aggregated and spatiotemporally aligned and fused with the corresponding region's coating digital gene archive. Based on the fused dataset, a cloud-based analysis engine is used to comprehensively assess the coating's health status and construct a performance degradation trend model. Based on the output of the performance degradation trend model, the probability of failure of the target coating area within a preset time period is predicted. Based on the failure risk probability, the ship's scheduled voyage plan, global port and ship repair resource information, and maintenance cost model, a comprehensive evaluation and optimization are conducted to generate at least one recommended maintenance decision scheme. By linking and storing monitoring data, analysis results, and maintenance operation records throughout the entire coating lifecycle, and conducting data mining, a data closed loop is formed to optimize coating research and development, coating processes, maintenance strategies, and asset assessment.

2. The method for remote monitoring and data management of ship painting as described in claim 1, characterized in that, The coating digital gene profile includes at least the following data fields: unique material identifier of the coating resin system, pigment volume concentration value, curing temperature-time curve, substrate surface roughness during construction, temperature and humidity sequence of the construction environment, and initial dry film thickness distribution map collected by an ultrasonic thickness gauge.

3. The method for remote monitoring and data management of ship painting as described in claim 2, characterized in that, The sensing node employs a negative learning strategy for preliminary anomaly diagnosis, which includes the following sub-steps: During the healthy service life of the coating, the multidimensional state data is continuously collected, time domain and frequency domain features are extracted, and a normal state feature library of the coating is constructed for the sensing nodes. In subsequent periodic monitoring, the feature vectors collected and extracted in real time will be compared with the historical features in the coating normal state feature library for similarity. When the deviation between real-time features and historical features exceeds a preset threshold, it is determined to be a preliminary anomaly, triggering the sensing node to upload encrypted detailed monitoring data packets to the cloud.

4. The method for remote monitoring and data management of ship painting as described in claim 1, characterized in that, The cloud-based analytics engine employs a hybrid modeling approach combining physical and data-driven models. The construction of the performance degradation trend model includes: The effect of temperature on the coating aging rate is simulated based on the Arrhenius formula, and the moisture penetration process is simulated based on Fick's diffusion law. The output of the physical driving sub-model is the physical degradation fraction. ; Using historical data on coating condition, environmental stress, and navigation conditions as input features, and measured performance degradation status as labels, a machine learning model is trained. The output of the data-driven sub-model is the data degradation score. ; The outputs of the physics-driven sub-model and the data-driven sub-model are weighted and fused to obtain the comprehensive coating health index. The calculation formula is: ,in, and These are weight coefficients that are dynamically adjusted based on the model confidence level, and .

5. The method for remote monitoring and data management of ship painting as described in claim 4, characterized in that, Generate at least one recommended maintenance decision scheme, including the following steps: Define the objective function as minimizing the total expected cost. , ,in, For direct maintenance costs, Costs of ship downtime This is based on the probability of failure and the estimated risk cost caused by coating failure; Define decision variables and constraints. Decision variables include the maintenance time window. Repair Location Repair methods The constraints include: the probability of failure is below the safety threshold, the maintenance time window is compatible with the ship's schedule, and the maintenance location has the necessary resources. Based on the failure risk probability prediction results, feasible combinations of decision variables are enumerated under the constraints, and the total expected cost corresponding to each combination is calculated. ; Select the option that maximizes the total expected cost The minimum combination of decision variables is taken as the optimal recommended solution, and the maintenance method is... This includes targeted micro-repairs and planned dry-dock overhauls.

6. The method for remote monitoring and data management of ship painting as described in claim 5, characterized in that, The targeted micro-repair includes: after the decision is made, automatically generating a work order containing the geographical coordinates of the area to be repaired, the damage type, and recommended repair materials, and sending it through an application programming interface to the control system of an underwater robot or a wall-climbing robot that is waiting at a designated port or can be carried by a mother ship.

7. The method for remote monitoring and data management of ship painting as described in claim 1, characterized in that, When linking and storing monitoring data, analysis results, and maintenance operation records throughout the entire life cycle of the coating and conducting data mining, blockchain is used to generate data hashes for key data of each monitoring alarm, each health assessment result, and each maintenance operation. The data hash value, along with the timestamp and data identifier, is written into an immutable distributed ledger to form a digital health passport for the coating area.

8. A remote monitoring and data management system for ship painting, characterized in that, The system is used to perform the steps of the remote monitoring and data management method for ship painting according to any one of claims 1-7, the system comprising: The shipborne sensing network subsystem consists of multiple sensing nodes deployed on the ship's hull, a shipborne gateway, and an edge computing unit. It is used to deploy sensing nodes in the coating area of ​​the ship, periodically collect multi-dimensional state data and local micro-environment data of the coating in the area, and perform local preprocessing and anomaly diagnosis. The cloud-based intelligent analysis platform, comprising a data lake warehouse, a digital twin engine, a hybrid AI analysis center, and a decision optimization module, is used to collect batch composition data, construction process parameters, and initial environmental data of the coatings used during the target ship's painting process. This data is processed, correlated, and stored to generate a unique digital genetic profile for each independent coating area of ​​the ship. The platform aggregates multi-source data from sensing nodes, the Automatic Identification System (AIS), and navigation weather services, and performs spatiotemporal alignment and data fusion with the corresponding region's digital genetic profile. Based on the fused dataset, the cloud-based analysis engine comprehensively assesses the coating's health status and constructs a performance degradation trend model. According to the output of the performance degradation trend model, the platform predicts the failure probability of the target coating area within a predetermined time period. Based on the failure probability, the ship's scheduled voyage plan, global port and repair resource information, and a maintenance cost model, a comprehensive evaluation and optimization are performed to generate at least one recommended maintenance decision plan. The blockchain data service subsystem is used to provide data storage and traceability services, and supports the association and storage of monitoring data, analysis results and maintenance operation records throughout the entire life cycle of the coating, as well as data mining, to form a data closed loop for optimizing coating research and development, coating process, maintenance strategy and asset assessment. The client-side interactive application provides administrators with a visual interface for monitoring status, receiving alarms, reviewing reports, and issuing maintenance instructions.

9. The remote monitoring and data management system for ship painting as described in claim 8, characterized in that, The sensing node includes at least a microprocessor, a self-powered module, a sensor array, and a low-power communication module; the sensor array includes a pulsed eddy current sensor for detecting early corrosion under the coating, a miniature fiber optic grating sensor for monitoring coating adhesion strain, and an electrochemical sensor for detecting surface salt and contaminant ion concentration.

10. The remote monitoring and data management system for ship painting as described in claim 9, characterized in that, The decision optimization module is connected to an external maintenance resource scheduling platform through a standardized interface. When the decision scheme is targeted micro-repair, the decision optimization module automatically sends a service request to the maintenance resource scheduling platform. The service request includes at least the repair location, time window, and robot operation instructions.

Images