Optimal arrangement method of photovoltaic monitoring sensor on offshore semi-submersible platform, computer and storage medium

Abstract

The application discloses a kind of offshore semi-submersible platform photovoltaic monitoring sensor optimization arrangement method, computer and storage medium, first establish platform fine three-dimensional model, determine hydrodynamic and thermal condition;Through the multi-physical field simulation such as AQWA, Fluent obtains the full platform structure response and temperature field data;Based on stress, temperature threshold preliminary screening candidate point;Comprehensive monitoring value evaluation model is built including risk coefficient, information non-redundancy coefficient and operation availability coefficient;With the maximum goal of comprehensive value index, using the improved greedy algorithm iteration optimization of coverage constraint priority, correlation dynamic update and redundant point inhibition, obtain the optimal arrangement point of sensor.The application significantly reduces the number of sensors, reduces installation and operation cost under the premise of ensuring full coverage and monitoring accuracy in high-risk and high-temperature areas.

Description

Technical Field

[0001] This invention relates to marine new energy equipment monitoring technology, specifically to an optimized arrangement method for photovoltaic monitoring sensors on a semi-submersible floating platform, a computer, and a storage medium. Background Technology

[0002] With the continuous development of offshore floating photovoltaic technology, semi-submersible platforms have become the mainstream carrier for deep-sea projects. Offshore semi-submersible platforms operate in complex marine environments for extended periods. On the one hand, they are subjected to dynamic loads such as waves and currents, which can easily lead to structural fatigue and localized damage. On the other hand, the high humidity and heat environment and high irradiance at sea significantly reduce the photoelectric conversion efficiency and lifespan of photovoltaic modules. Therefore, it is necessary to deploy sensors to monitor the platform structure and the status of photovoltaic modules in real time.

[0003] Currently, the deployment of sensors on offshore platforms largely relies on engineering experience or uniform grid layout, which has the following main drawbacks:

[0004] High redundancy and high cost: Experience-based deployment often involves installing sensors in a large number of locations, but the data from many of these locations are highly correlated, resulting in wasted procurement, installation, and maintenance costs.

[0005] The monitoring is not targeted and has many blind spots: it does not take into account the actual load distribution and temperature field distribution of the platform, and it is easy to miss high-risk and high-temperature areas.

[0006] Lack of quantitative optimization basis: It does not comprehensively consider multi-dimensional indicators such as structural risk, information redundancy and accessibility for offshore operation and maintenance.

[0007] The unique characteristics of offshore operations and maintenance have not been taken into account: In the deep-sea environment, the difficulty and cost of sensor installation, replacement, and data acquisition are far greater than on land, and existing methods lack quantitative evaluation of operational accessibility.

[0008] Existing patents cover sensor optimization placement for fixed marine platforms or structures, but research on optimization methods for semi-submersible platforms at sea, integrating hydrodynamic-structural-thermal multiphysics simulation and multi-objective value evaluation, is still insufficient. Therefore, there is an urgent need for a sensor optimization placement method that balances accuracy, cost, and engineering feasibility. Summary of the Invention

[0009] The purpose of this invention is to propose an optimized arrangement method for photovoltaic monitoring sensors on a semi-submersible offshore platform, as well as a computer and storage medium.

[0010] The technical solution to achieve the purpose of this invention is: a method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform at sea, comprising the following steps:

[0011] S1: Establish a detailed three-dimensional model of the target semi-submersible floating platform at sea, and determine the hydrodynamic and thermal calculation conditions during the platform's operation.

[0012] S2: Obtain the structural response data and photovoltaic module temperature field data of the entire platform through multi-physics field collaborative simulation. The structural response data includes peak equivalent stress data.

[0013] S3: Based on preset stress thresholds and temperature thresholds, candidate points for structural safety monitoring and photovoltaic performance monitoring are selected from the structural response data and temperature field data, respectively. After merging and deduplication, a total set of candidate points U is obtained.

[0014] S4: Construct a comprehensive monitoring value evaluation model that includes risk coefficient, information non-redundancy coefficient, and operation and maintenance accessibility coefficient, and calculate the comprehensive monitoring value index of each candidate point in the total candidate point set; the information non-redundancy coefficient is dynamically calculated based on the current set A in S5, and takes a value of 1 when the current set A is empty in the initial state;

[0015] S5: With the goal of maximizing the comprehensive monitoring value index, an improved greedy algorithm with coverage constraint priority, dynamic correlation update and redundant point suppression is adopted. Combined with the preset sensor number constraint and monitoring coverage constraint, the candidate points are iteratively optimized and screened to obtain the final set of optimal sensor placement points A.

[0016] S6: Verify the monitoring accuracy of the optimal set of locations, and output the final sensor layout scheme after the verification is successful.

[0017] Furthermore, in S1, the hydrodynamic calculation conditions are divided into normal operating sea state and extreme design sea state, including design wave height, wave period, current velocity and current direction; the thermal calculation conditions are divided into normal operating temperature conditions and extreme high temperature conditions, including ambient temperature, solar irradiance and sea surface wind speed.

[0018] Furthermore, S2 specifically includes:

[0019] S21: Hydrodynamic Simulation Calculation: A hydrodynamic numerical model of the target platform is established using AQWA software. Hydrodynamic surface element models are established for the main submerged structures of the pontoons and columns. Morison beam element models are established for the underwater tubular floats, underwater slender connecting rods, and mooring connection components. Beam elements or shell elements are established for the upper truss and photovoltaic support in the structural finite element model. Wet surface mesh, mass parameters, center of gravity position, added mass, radiation damping, hydrostatic restoring force, mooring constraints, wave boundary conditions, and ocean current boundary conditions are set. The six-degree-of-freedom motion response of the platform, wet surface hydrodynamic pressure, Morison beam element hydrodynamic load, wave load, ocean current load, and mooring point tension time history data are calculated under various hydrodynamic conditions to form a hydrodynamic load dataset.

[0020] S22: Structural stress calculation: Based on the unidirectional hydrodynamic-structural load transfer method combining Morison beam elements and finite element method, the hydrodynamic load data obtained in S21, the platform motion response, and the mooring point tension are applied to the structural calculation model to calculate the axial force, shear force, bending moment, torque, nodal reaction force, and peak equivalent stress data of the main strut structure of the target platform; and local finite element sub-models are established for connection nodes, mooring interfaces, weld areas, and photovoltaic bracket connection parts with large responses to further calculate the local structural stress distribution and the maximum stress data of key nodes, forming a structural response dataset for screening candidate points for structural safety monitoring;

[0021] S23: Photovoltaic module thermal simulation calculation: Using Fluent software, combined with the thermal calculation conditions determined in S1, a thermal numerical model of photovoltaic modules and photovoltaic arrays is established. The thermal property parameters of photovoltaic module materials, solar irradiance heat source, heat source reduction coefficient corresponding to photoelectric conversion efficiency, and external flow field inlet conditions or convective heat transfer boundary conditions determined based on sea surface wind speed are set. The temperature distribution cloud map of photovoltaic array, average operating temperature data and maximum temperature data of each photovoltaic module are calculated to form a temperature field dataset for photovoltaic performance monitoring point selection.

[0022] Furthermore, S3 specifically includes:

[0023] S31: Initial Screening of Candidate Sites for Structural Safety Monitoring: Based on the structural response data and local stress data obtained in S2, preset structural response thresholds and local stress thresholds are used to screen beam elements, connection nodes, and local high-stress areas whose peak equivalent stress, bending moment, nodal reaction force, mooring tension, or the maximum stress data of key nodes calculated by the local finite element sub-model exceeds the corresponding threshold. These are selected as candidate sites for structural safety monitoring, forming a candidate set of structures. The local stress threshold is 50%-70% of the allowable stress or design strength of the corresponding structural material.

[0024] S32: Initial Screening of Photovoltaic Performance Monitoring Candidate Sites: Based on the photovoltaic module temperature field dataset obtained in S2, a preset temperature threshold is used to screen photovoltaic modules or local areas whose average operating temperature or maximum temperature exceeds the threshold, forming a temperature candidate set. The temperature threshold is 70%-90% of the rated maximum operating temperature of the photovoltaic module.

[0025] S33: Merge and remove duplicates: Combine the candidate structure set With temperature candidate set Merging: Candidate points with overlapping spatial locations and the same monitoring type are deduplicated; candidate points with overlapping spatial locations but different monitoring types are merged into composite monitoring candidate points while retaining their corresponding monitoring parameter types, resulting in the total set of candidate points. .

[0026] Furthermore, in S4, the calculation formula for the comprehensive monitoring value evaluation model is as follows:

[0027] ;

[0028] In the formula, For the first The comprehensive monitoring value index of each candidate site; For the first Risk coefficient of each candidate site; For the first Information redundancy coefficient of each candidate point; For the first The operational reachability coefficient of each candidate site; , , These are the weighting coefficients, and .

[0029] Furthermore, the risk coefficient The specific calculation method is as follows:

[0030] ;

[0031] in Indicates the first The peak equivalent stress of each candidate point for structural safety monitoring includes the maximum stress data of key nodes calculated by the local finite element sub-model. This indicates the allowable stress value of the platform structure material; Indicates the first Each photovoltaic performance monitoring candidate point corresponds to the highest operating temperature of the photovoltaic module or local area; This indicates the rated maximum operating temperature of the photovoltaic module;

[0032] The information non-redundancy coefficient The calculation method is as follows:

[0033] ;

[0034] ;

[0035] in, Let represent the Pearson correlation coefficient between the i-th candidate point and the j-th monitoring point of the same type in the currently selected point set (i.e., the dynamically updated set A during the S5 iteration). When the current set A contains no monitoring points of the same type, let . Pearson correlation coefficient The calculation formula is:

[0036] ;

[0037] Where Xi={x i,1 ,x i,2 ,…,x i,m} represents the monitoring response sequence of the i-th candidate point under the same computational conditions, Xj={x j,1 ,x j,2 ,…,x j,m} represents the monitoring response sequence of the same type of monitoring points in the current set A under the same computational conditions, x̄ i and x̄ j , where are the average values ​​of the corresponding response sequences, and m is the number of data points in the response sequence. The maximum absolute value is taken as . ; The larger the value, the more irreplaceable the monitoring data at that location;

[0038] The maintenance accessibility coefficient The range of values ​​for is:

[0039] ;

[0040] in, The larger the value, the better the accessibility for installation and maintenance of that location;

[0041] The weighting coefficients satisfy:

[0042] .

[0043] Furthermore, S5 specifically includes:

[0044] S51: Preset upper limit N for the number of sensors and correlation threshold and monitoring coverage constraints; the upper limit N of the number of sensors is not less than the number of mandatory coverage points, and the correlation threshold The monitoring coverage constraint is 0.8, which includes: 100% coverage of high-risk areas with a risk coefficient exceeding 0.8 among the candidate sites for structural safety monitoring; and 100% coverage of high-temperature areas where the photovoltaic module temperature exceeds 90% of the rated maximum operating temperature.

[0045] S52: Add the risk coefficient R from the total candidate site set U. i Candidate points for structural safety monitoring with a value exceeding 0.8 and candidate points for photovoltaic performance monitoring with a temperature exceeding 90% of the rated maximum operating temperature are marked as mandatory coverage points, while the remaining candidate points are marked as optional points. With the optimization objective of maximizing the comprehensive monitoring value index, the candidate points in U are sorted from high to low according to the currently calculated comprehensive monitoring value index.

[0046] S53: Initialize the current set A to empty. Examine the candidate points i that are currently ranked first. If candidate point i is a mandatory coverage point and its corresponding high-risk or high-temperature area has not yet been covered, add candidate point i to the current set A. If candidate point i is an optional point, calculate the absolute value of the Pearson correlation coefficient between candidate point i and the monitoring data of the same type of monitoring points in the current set A, and take the maximum value. ,like Greater than If the high-risk or high-temperature area corresponding to candidate point i is already covered by the current set A, then candidate point i is considered a redundant point and excluded; otherwise, it is added to the current set A. After each candidate point is added to the current set A, the information non-redundancy coefficient and comprehensive monitoring value index of the remaining candidate points in U are recalculated, and the remaining candidate points are reordered.

[0047] S54: Repeat S53 until the current set A reaches the preset upper limit of the number of sensors N, or all high-risk and high-temperature points that meet the mandatory coverage constraints have been covered and no more candidate points can be added. Output the current set A as the final set of optimal sensor placement points.

[0048] Furthermore, in S6, the standard for verifying the monitoring accuracy is as follows: based on the simulation response data corresponding to the selected point, the representativeness of the full platform structural response data, local stress data, and temperature field data is verified, and the maximum relative error obtained from the verification does not exceed 8%, and meets the preset requirements for coverage of high-risk structural points and high-temperature photovoltaic points.

[0049] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible offshore platform.

[0050] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible offshore platform.

[0051] Compared with the prior art, the significant advantages of this invention are:

[0052] 1) This invention integrates AQWA hydrodynamic simulation, Morison beam element structural response calculation, local finite element stress calculation and Fluent thermal simulation to construct a hydrodynamic-structural-thermal multi-physics field data system. It covers two major monitoring needs: the structural safety of semi-submersible platforms and the performance of photovoltaic modules. It solves the problem of blind spots in existing single-dimensional deployment monitoring and is suitable for special working conditions of offshore semi-submersible floating platforms.

[0053] 2) This invention adopts a three-step strategy of “initial screening of candidate sites - comprehensive value quantification - iterative optimization screening” to eliminate redundant sites while meeting the preset coverage constraints of high-risk structural sites and high-temperature photovoltaic sites.

[0054] 3) The comprehensive monitoring value evaluation model of this invention simultaneously incorporates risk coefficients, information non-redundancy coefficients, and operation and maintenance accessibility coefficients, ensuring monitoring accuracy while quantifying the unique difficulties and costs associated with the installation and maintenance of semi-submersible floating platforms. The optimized solution achieves a balance between monitoring performance and engineering economics, demonstrating superior practicality and feasibility compared to existing methods.

[0055] 4) The optimized process standards and modules of this invention are clear and can be easily adapted to semi-submersible floating platforms of different sizes and structural forms. It can also be extended to the design of monitoring systems for other floating marine structures, and can directly guide engineering design and construction. Attached Figure Description

[0056] Figure 1 This is a flowchart of the method of the present invention.

[0057] Figure 2 This is a flowchart of the multiphysics field collaborative simulation data acquisition process of the present invention.

[0058] Figure 3 This is a flowchart illustrating the initial screening and comprehensive monitoring value quantification process for candidate sites in this invention.

[0059] Figure 4 This is a flowchart of the iterative selection process for the optimal location of the sensor in this invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0061] This invention provides a method for optimizing the layout of photovoltaic monitoring sensors on a semi-submersible floating platform, along with a computer and storage medium, aiming to solve the problems of high redundancy, high cost, numerous monitoring blind spots, and poor targeting in existing sensor deployments. This method, by establishing a refined 3D model of the platform and combining multiphysics simulation, a comprehensive evaluation model, and iterative optimization algorithms, significantly reduces the number of sensors required while ensuring monitoring accuracy and full coverage of high-risk locations.

[0062] like Figure 1 As shown, the present invention provides a method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform at sea, comprising the following steps. Figures 2 to 4 These are sub-flowcharts for each key step.

[0063] S1, 3D modeling and calculation conditions determination of the semi-submersible floating platform at sea: Establish a refined 3D model of the target semi-submersible floating platform at sea. The model covers the main floating structure of the platform (floating cylinders, columns), the upper truss structure, photovoltaic support and photovoltaic modules. The non-stressed and non-heat-conducting auxiliary structures are reasonably simplified to reduce the amount of calculation while ensuring the simulation accuracy.

[0064] Based on the hydrological and meteorological data of the target platform's service area, two types of calculation conditions were determined:

[0065] ① Hydrodynamic calculation conditions, including normal operating sea state and extreme design sea state, covering core hydrological parameters such as design wave height, wave period, current velocity, and current direction;

[0066] ② Thermal calculation conditions, including normal operating temperature conditions and extreme high temperature conditions, covering core thermal environment parameters such as ambient temperature, solar irradiance, and sea surface wind speed.

[0067] S2, multi-physics collaborative simulation calculation, obtains multi-dimensional response data across the entire platform: such as... Figure 2 As shown, multiphysics co-simulation specifically includes three sub-steps: hydrodynamic simulation, structural stress calculation, and thermal simulation.

[0068] S21, Hydrodynamic Simulation Calculation: A hydrodynamic numerical model of the target platform is established using AQWA software. Hydrodynamic surface element models are established for the main submerged structures of the pontoons and columns. Morison beam element models are established for the underwater tubular floats, underwater slender connecting rods, and mooring connection components. Beam elements or shell elements are established for the upper truss and photovoltaic support in the structural finite element model. Wet surface mesh, mass parameters, center of gravity position, added mass, radiation damping, hydrostatic restoring force, mooring constraints, wave boundary conditions, and ocean current boundary conditions are set. The time history data of the platform's six-degree-of-freedom motion response (roll, pitch, heave, sway, yaw, bow roll), wave loads, and ocean current loads are calculated under each hydrodynamic condition determined in S1.

[0069] S22, Structural stress calculation: Based on the unidirectional hydrodynamic-structural load transfer method combining Morison beam elements and local finite element method, the hydrodynamic load data, platform motion response and mooring point tension obtained in S21 are imported into the structural calculation model. The material properties, mesh generation and constraint conditions of the platform structure are set, and the structural dynamic response is calculated to obtain the axial force, shear force, bending moment, torque, nodal reaction force and peak equivalent stress data of the main rod structure of the target platform (i.e. the maximum value or envelope value of the equivalent stress time history under each working condition). Local finite element sub-models are established for the connection nodes, mooring interfaces, weld areas and photovoltaic bracket connection parts with large responses to obtain the local structural stress distribution and the maximum stress data of key nodes.

[0070] S23, Thermal simulation calculation of photovoltaic modules: Using Fluent software, combined with the thermal calculation conditions determined in S1, a thermal numerical model of photovoltaic modules is established. The photoelectric conversion efficiency, thermal property parameters, external flow field inlet conditions or convective heat transfer boundary conditions determined based on sea surface wind speed, and solar radiation heat source are set. The temperature distribution cloud map of the photovoltaic array and the operating temperature data of each photovoltaic module are calculated.

[0071] S3, Initial screening of candidate monitoring sites: such as Figure 3 As shown, the initial screening and comprehensive monitoring value quantification process for candidate sites includes threshold screening, merging and deduplication, and weighted calculation. Based on the stress distribution data and temperature distribution data obtained from S2, high-value candidate sites for structural safety monitoring and photovoltaic performance monitoring are screened respectively, and low-value areas are eliminated through threshold screening to avoid excessive computation caused by full-platform, full-domain traversal. Specifically, this includes:

[0072] S31, Initial Screening of Candidate Points for Structural Safety Monitoring: A preset stress threshold is established, which is 50%-70% of the allowable stress of the platform structural materials. Points whose peak equivalent stress, bending moment, nodal reaction force, mooring tension, or the maximum stress data of key nodes calculated by the local finite element sub-model exceed the corresponding threshold are selected from structural response data and local stress data. Special attention is paid to stress-concentration sensitive areas such as nodal connection parts, weld areas, and abrupt changes in cross-section. These points are selected as candidate points for structural safety monitoring, forming a candidate structural set. ;

[0073] S32, Initial Screening of Photovoltaic Performance Monitoring Candidate Sites: A preset temperature threshold is set, which is 70%-90% of the rated maximum operating temperature of the photovoltaic module. Photovoltaic modules or local areas with average or maximum operating temperatures exceeding this threshold are selected from the photovoltaic module temperature field data. Special attention is paid to array areas with poor ventilation and concentrated irradiance. These sites are selected as candidate sites for photovoltaic performance monitoring, forming a temperature candidate set. ;

[0074] S33, Merging and Deduplication: Merge the structural candidate set S and the temperature candidate set T. Deduplicate candidate points that have overlapping spatial locations and the same monitoring type. For candidate points that have overlapping spatial locations but different monitoring types, merge them into composite monitoring candidate points and retain their corresponding monitoring parameter types to obtain the total candidate point set. .

[0075] S4, Quantitative Calculation of the Comprehensive Monitoring Value of Candidate Points: Construct a comprehensive monitoring value evaluation model, comprehensively considering three core dimensions: risk priority, data redundancy, and operation and maintenance cost, to calculate the comprehensive monitoring value index of each candidate point in the total candidate point set. The dynamically updated set of selected sensor points during the iterative optimization and screening process is denoted as the current selected point set (formally output in S5, hereinafter referred to as the current set A). The calculation formula is as follows:

[0076]

[0077] In the formula, For the first The comprehensive monitoring value index of each candidate point is used; the higher the value, the higher the monitoring value, and the more priority is given to deploying sensors there.

[0078] For the first The risk coefficient of each candidate point represents the failure risk level of the point: if it is a structural safety monitoring point, This is the ratio of the peak equivalent stress at that point to the allowable stress of the structural material; if it is a photovoltaic performance monitoring point... This is the ratio of the highest operating temperature at this point to the rated highest operating temperature of the photovoltaic module.

[0079] The information non-redundancy coefficient for the i-th candidate point represents the irreplaceability of the point monitoring data. It is initially set to 1 and is dynamically calculated during the S5 iterative optimization process. , where is the Pearson correlation coefficient between the i-th candidate point and the monitoring data of all points of the same type in the current set A (structural points are compared only with structural response data, and temperature points are compared only with temperature data). Maximum absolute value The value ranges from 0 to 1; The larger the value, the higher the data correlation between the candidate point and the points in the current set A, and the lower the information non-redundancy. The smaller; if the current set A contains no monitoring points of the same type, let ;

[0080] For the first The operational accessibility coefficient of each candidate site quantifies the installation and maintenance costs of the site. The value ranges from 0 to 1, with lower installation difficulty and better operational accessibility. The larger the value;

[0081] , , These are the weighting coefficients, and The value was determined using the Analytic Hierarchy Process (AHP). In one optional implementation, the recommended value is... , , .

[0082] During the S5 iteration process, for each newly selected point, the remaining candidate points will be recalculated. And then update .

[0083] S5, sensor optimal location iterative optimization and selection: such as Figure 4 As shown, with the goal of maximizing the comprehensive monitoring value index, and combining preset constraints on the number of sensors and monitoring coverage, the candidate locations are iteratively optimized and screened, specifically including:

[0084] S51, Preset constraints:

[0085] ①Quantity constraint: Based on the project budget and monitoring requirements, a maximum number of sensors, N, is preset;

[0086] ② Coverage constraints: 100% coverage for high-risk areas where structural stress exceeds 80% of allowable stress, and 100% coverage for high-temperature areas where photovoltaic module temperature exceeds 90% of rated maximum operating temperature;

[0087] S52, Location Sorting: Based on Comprehensive Monitoring Value Index Using this as the core indicator, all points in the total candidate point set U are sorted from highest to lowest.

[0088] S53, Iterative Screening: An improved greedy algorithm, prioritizing coverage constraints, dynamically updating correlations, and suppressing redundant points, is used to examine candidate points ranked higher in sequence. Points that meet the mandatory coverage requirements for high-risk or high-temperature locations and are not yet covered are added to the current set A. For the remaining points, the absolute value of the Pearson correlation coefficient between the point and the monitoring data of similar points in the current set A is calculated. If the absolute value exceeds the preset correlation threshold, the point is considered for inclusion in the list. Furthermore, it cannot supplement high-risk or high-temperature areas not yet covered by the current set A (i.e., if at least one corresponding monitoring point exists in the current set A for the high-risk or high-temperature area corresponding to the candidate point, it is excluded as a redundant point; otherwise, it is added to the current set A; after selecting a point, the information non-redundancy coefficient of the remaining candidate points is updated). Comprehensive monitoring value index To avoid duplicate monitoring;

[0089] S54, Termination Condition: Repeat the iteration process of S53 until the current set A reaches the preset upper limit of the number of sensors N, or all mandatory coverage points that meet the monitoring coverage constraints have been covered and no more candidate points can be added. Then, terminate the iteration and output the current set A as the final optimal set of sensor placement points. Here, "covered" means that at least one monitoring point already exists in the current set A in the high-risk or high-temperature area corresponding to the candidate point.

[0090] S6, Solution Verification and Output:

[0091] The simulation monitoring data corresponding to the optimal layout point set A obtained from S5 is compared with the full-domain simulation data of the entire platform to verify the monitoring accuracy. The verification standard is as follows:

[0092] The maximum relative error between the monitoring data of the selected points and the simulation data of the whole platform does not exceed 8%, and meets the preset monitoring coverage requirements;

[0093] After successful verification, the final sensor layout diagram, installation location parameters, monitoring parameter configuration, and operation and maintenance requirements will be output.

[0094] Furthermore, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible platform at sea.

[0095] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible offshore platform.

[0096] Example

[0097] This embodiment applies to a 400kW bamboo-based composite semi-submersible floating demonstration platform in a certain sea area. The platform is designed for a water depth of 16.8m. The main structure adopts a modular floating structure composed of bamboo-wound composite pipes and steel nodes. It consists of four individual floating photovoltaic power generation platforms arranged in a straight line. The main dimensions of each platform are 22.2m × 22.2m × 7m, with a design draft of 1.8m. Several N-type double-sided double-glass photovoltaic modules are arranged on the upper steel truss of the platform to form a floating photovoltaic power generation unit. The main body of the platform float is made of bamboo-based composite material with a tensile strength of 25.41MPa along the grain and 20.99MPa across the grain. The steel structure nodes and trusses are made of Q355 steel with an allowable stress of 235MPa. The rated maximum operating temperature of the photovoltaic modules is 45℃.

[0098] The optimized arrangement method for photovoltaic monitoring sensors on a semi-submersible floating platform at sea, as described in this embodiment, includes the following specific steps:

[0099] S1 3D Modeling and Calculation Condition Determination

[0100] A detailed 3D model of the target platform was built using SolidWorks. The model includes the main floating structure, the superstructure steel truss, photovoltaic support structures, and photovoltaic modules. It fully recreates the structural forms of the bamboo-wound composite pipe floating body, steel connection nodes, superstructure steel truss, photovoltaic support structures, and photovoltaic modules. Core load-bearing structures such as mooring interfaces and pipe-steel connection nodes are fully preserved, while non-core auxiliary structures that do not affect mechanical transfer and heat conduction are reasonably simplified. The calculation conditions are determined based on hydrological and meteorological data from the target sea area.

[0101] Hydrodynamic calculation conditions:

[0102] Normal operating conditions: significant wave height 3.75m, spectral peak period 8.4s, surface velocity 2.14m / s, flow direction 0°;

[0103] Extreme design conditions: significant wave height 5.69m, maximum wave height 10.93m, spectral peak period 8.8s, surface velocity 2.88m / s, flow direction 45°.

[0104] Thermal calculation conditions:

[0105] Normal operating conditions: ambient temperature 25℃, solar irradiance 800W / ㎡, sea surface wind speed 3m / s;

[0106] Extreme high temperature conditions: ambient temperature 38℃, solar irradiance 1000W / ㎡, sea surface wind speed 5m / s.

[0107] S2 Multiphysics Co-simulation Calculation

[0108] S21 Hydrodynamic Simulation Calculation: A hydrodynamic numerical model of the platform was established using AQWA software. Hydrodynamic surface element models were established for the main submerged structures of the pontoons and columns. Morison beam element models were established for the underwater tubular floats, underwater slender connecting rods, and mooring connection components. Beam elements or shell elements were established for the upper truss and photovoltaic support in the structural finite element model. Wet surface mesh, mass parameters, center of gravity position, added mass, radiation damping, hydrostatic restoring force, mooring constraints, and environmental boundary conditions such as waves and ocean currents were set. Frequency domain and time domain analysis were performed to calculate the six-degree-of-freedom motion response of the platform, wet surface hydrodynamic pressure, Morison beam element hydrodynamic load, wave load, ocean current load, and mooring point tension time history data under two working conditions, forming a hydrodynamic load dataset.

[0109] S22 Structural Stress Calculation: Based on the unidirectional hydrodynamic-structural load transfer method combining Morison beam elements and finite element method, the hydrodynamic load data obtained in S21, the platform motion response, and the mooring point tension are applied to the structural calculation model. The material properties of bamboo-based composite materials and Q355 steel are set respectively, and the axial force, shear force, bending moment, torque, nodal reaction force, and peak equivalent stress data of the main rod structure of the target platform are calculated. Local finite element sub-models are established for pipe-steel connection nodes, mooring interfaces, weld areas, and photovoltaic bracket connection parts with large responses. The local structural stress distribution and the maximum stress data of key nodes are further calculated to form a structural response dataset for screening candidate points for structural safety monitoring.

[0110] S23 Photovoltaic Module Thermal Simulation Calculation: Using Fluent software and combining the thermal calculation conditions determined in S1, a thermal numerical model of the photovoltaic module and photovoltaic array is established. The thermal property parameters of the photovoltaic module material, the solar irradiance heat source, the heat source reduction coefficient corresponding to the photoelectric conversion efficiency, and the external flow field inlet conditions or convective heat transfer boundary conditions determined based on the sea surface wind speed are set. The temperature distribution cloud map of the photovoltaic array, the average operating temperature data and the maximum temperature data of each photovoltaic module are calculated to form a temperature field dataset for the selection of photovoltaic performance monitoring points. The high temperature area is concentrated in the poorly ventilated area in the middle of the array.

[0111] Initial screening of candidate monitoring sites for S3

[0112] The stress threshold for bamboo-based composite materials was set at 60% of the design tensile strength, i.e., 15.25 MPa parallel to the grain and 12.59 MPa transverse to the grain; the stress threshold for steel structures was set at 60% of the allowable stress, i.e., 141 MPa. Based on the structural response and local stress data obtained from S2, several candidate points for structural safety monitoring were selected from beam elements, connection nodes, and local high-stress areas where the peak equivalent stress, bending moment, nodal reaction force, mooring tension, or local maximum stress of beam elements exceeded the corresponding thresholds, forming a candidate structural set. The focus is on covering stress concentration areas such as pipe-steel connection nodes, mooring interfaces, and purlin welds.

[0113] The temperature threshold is set to 80% of the rated maximum operating temperature, i.e., 36℃. Based on the photovoltaic module temperature field dataset obtained from S2, several candidate photovoltaic performance monitoring points are selected from photovoltaic modules or local areas where the average operating temperature or maximum temperature exceeds 36℃, forming a temperature candidate set. .

[0114] The candidate sets for structures and temperatures are merged. Candidate points with overlapping spatial locations and the same monitoring type are deduplicated. Candidate points with overlapping spatial locations but different monitoring types are merged into composite monitoring candidate points, retaining their corresponding monitoring parameter types. In this embodiment, the total candidate point set U is obtained after merging and deduplication, generally covering both high-risk structural areas and high-temperature photovoltaic areas.

[0115] Quantitative Calculation of the Comprehensive Monitoring Value of Candidate Sites S4

[0116] A comprehensive monitoring value evaluation model is constructed, which includes risk coefficient, information non-redundancy coefficient, and operation and maintenance accessibility coefficient. The weight coefficients are determined by the analytic hierarchy process (AHP). =0.5, =0.3, =0.2, calculate parameters for each candidate point:

[0117] Risk coefficient For structural safety monitoring points, the ratio of the peak equivalent stress at that point to the allowable stress or design strength of the corresponding structural material is taken; for photovoltaic performance monitoring points, the ratio of the highest operating temperature of the corresponding photovoltaic module or local area at that point to the rated maximum operating temperature of the photovoltaic module, 45℃, is taken.

[0118] Information non-redundancy coefficient Initially set to 1, it is dynamically updated in iteration S5 to 1 minus the maximum absolute value of the Pearson correlation coefficient between the candidate point and the monitoring data of the same type of monitoring points in the current set A. When there are no monitoring points of the same type in the current set A, let ; .

[0119] Operation and maintenance accessibility coefficient The platform photovoltaic panel reachable points are set at 1.0, the upper purlin points at 0.7, the floating body pipe-steel connection points at 0.5, and the underwater mooring points at 0.4.

[0120] The comprehensive monitoring value index for each location was calculated using a comprehensive monitoring value evaluation model. And sort them from highest to lowest.

[0121] S5 sensor optimal location iteration optimization screening

[0122] Preset constraints: The upper limit N of the number of sensors is set according to the project budget and monitoring requirements. In this embodiment, N is no more than 30. The coverage constraint is that the points with steel structure stress exceeding 188MPa (80% of allowable stress), the points with bamboo-based material stress exceeding 80% of the corresponding design tensile strength are 100% covered, and the points with temperature exceeding 40.5℃ (90% of the rated maximum working temperature) are 100% covered.

[0123] A dynamically improved greedy algorithm is adopted, examining candidate monitoring points in descending order of their comprehensive monitoring value index. Points meeting the mandatory coverage constraints for high-risk or high-temperature points are preferentially added to the current set A. For the remaining points, the absolute value of the Pearson correlation coefficient between the point and the monitoring data of similar points in the current set A is calculated. If the maximum value is... Greater than If the high-risk or high-temperature area corresponding to the candidate point already has at least one corresponding monitoring point in the current set A, it is considered a redundant point and excluded; otherwise, it is added to the current set A. After each newly selected point is added, the information non-redundancy coefficient of the remaining candidate points is recalculated. and comprehensive monitoring value index And reorder them.

[0124] After the iteration terminates, the current set A is output as the final optimal sensor placement set. The final optimal sensor placement set includes both structural safety monitoring points and photovoltaic performance monitoring points, and covers key locations such as bamboo-based structures, steel structures and mooring interfaces, and high-temperature areas of photovoltaic arrays. Compared with the traditional empirical uniform placement, this embodiment can significantly reduce the number of sensors and lower installation and subsequent maintenance costs while meeting coverage constraints and monitoring accuracy requirements.

[0125] S6 Solution Verification and Output

[0126] Based on the simulation response data corresponding to the selected points in the final optimal sensor placement set A, the monitoring data of the selected points are compared with the structural response data, local stress data and temperature field data of the entire platform for representative verification. The verification results show that the maximum relative error of the structural stress data and temperature data does not exceed 8%, and meets the preset requirements for coverage of high-risk structural points and high-temperature photovoltaic points. After the verification is passed, the sensor placement map, installation area or coordinate range, monitoring parameter configuration and operation and maintenance requirements are output.

[0127] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform at sea, characterized in that, Includes the following steps: S1: Establish a detailed three-dimensional model of the target semi-submersible floating platform at sea, and determine the hydrodynamic and thermal calculation conditions during the platform's operation. S2: Obtain the structural response data and photovoltaic module temperature field data of the entire platform through multi-physics field collaborative simulation. The structural response data includes peak equivalent stress data. S3: Based on preset stress thresholds and temperature thresholds, candidate points for structural safety monitoring and photovoltaic performance monitoring are selected from the structural response data and temperature field data, respectively. After merging and deduplication, a total set of candidate points U is obtained. S4: Construct a comprehensive monitoring value evaluation model that includes risk coefficient, information non-redundancy coefficient, and operation and maintenance accessibility coefficient, and calculate the comprehensive monitoring value index of each candidate point in the total candidate point set; the information non-redundancy coefficient is dynamically calculated based on the current set A in S5, and takes a value of 1 when the current set A is empty in the initial state; S5: With the goal of maximizing the comprehensive monitoring value index, an improved greedy algorithm with coverage constraint priority, dynamic correlation update and redundant point suppression is adopted. Combined with the preset sensor number constraint and monitoring coverage constraint, the candidate points are iteratively optimized and screened to obtain the final set of optimal sensor placement points A. S6: Verify the monitoring accuracy of the optimal set of locations, and output the final sensor layout scheme after the verification is passed; In S4, the calculation formula for the comprehensive monitoring value evaluation model is as follows: ； In the formula, For the first The comprehensive monitoring value index of each candidate site; For the first Risk coefficient of each candidate site; For the first Information redundancy coefficient of each candidate point; For the first The operational reachability coefficient of each candidate site; , , These are the weighting coefficients, and ; The risk coefficient The specific calculation method is as follows: ； in Indicates the first The peak equivalent stress of each candidate point for structural safety monitoring includes the maximum stress data of key nodes calculated by the local finite element sub-model. This indicates the allowable stress value of the platform structure material; Indicates the first Each photovoltaic performance monitoring candidate point corresponds to the highest operating temperature of the photovoltaic module or local area; This indicates the rated maximum operating temperature of the photovoltaic module; The information non-redundancy coefficient The calculation method is as follows: ； ； in, Let represent the Pearson correlation coefficient between the i-th candidate point and the j-th monitoring point of the same type in the currently selected point set (i.e., the dynamically updated set A during the S5 iteration). When the current set A contains no monitoring points of the same type, let . Pearson correlation coefficient The calculation formula is: ； Where Xi={x i,1 ,x i,2 ,…,x i,m } represents the monitoring response sequence of the i-th candidate point under the same computational conditions, Xj={x j,1 ,x j,2 ,…,x j,m } represents the monitoring response sequence of the same type of monitoring points in the current set A under the same computational conditions, x̄ i and x̄ j , where are the average values ​​of the corresponding response sequences, and m is the number of data points in the response sequence. The maximum absolute value is taken as . ; The larger the value, the more irreplaceable the monitoring data at that location; The maintenance accessibility coefficient The range of values ​​for is: ； in, The larger the value, the better the accessibility for installation and maintenance of that location; The weighting coefficients satisfy: 。 2. The method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform according to claim 1, characterized in that, In S1, the hydrodynamic calculation conditions are divided into normal operating sea state and extreme design sea state, including design wave height, wave period, current velocity and current direction; the thermal calculation conditions are divided into normal operating temperature conditions and extreme high temperature conditions, including ambient temperature, solar irradiance and sea surface wind speed.

3. The method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform according to claim 1, characterized in that, S2 specifically includes: S21: Hydrodynamic Simulation Calculation: A hydrodynamic numerical model of the target platform is established using AQWA software. Hydrodynamic surface element models are established for the main submerged structures of the pontoons and columns. Morison beam element models are established for the underwater tubular floats, underwater slender connecting rods, and mooring connection components. Beam elements or shell elements are established for the upper truss and photovoltaic support in the structural finite element model. Wet surface mesh, mass parameters, center of gravity position, added mass, radiation damping, hydrostatic restoring force, mooring constraints, wave boundary conditions, and ocean current boundary conditions are set. The six-degree-of-freedom motion response of the platform, wet surface hydrodynamic pressure, Morison beam element hydrodynamic load, wave load, ocean current load, and mooring point tension time history data are calculated under various hydrodynamic conditions to form a hydrodynamic load dataset. S22: Structural stress calculation: Based on the unidirectional hydrodynamic-structural load transfer method combining Morison beam elements and finite element method, the hydrodynamic load data obtained in S21, the platform motion response, and the mooring point tension are applied to the structural calculation model to calculate the axial force, shear force, bending moment, torque, nodal reaction force, and peak equivalent stress data of the main strut structure of the target platform; and local finite element sub-models are established for connection nodes, mooring interfaces, weld areas, and photovoltaic bracket connection parts with large responses to further calculate the local structural stress distribution and the maximum stress data of key nodes, forming a structural response dataset for screening candidate points for structural safety monitoring; S23: Photovoltaic module thermal simulation calculation: Using Fluent software, combined with the thermal calculation conditions determined in S1, a thermal numerical model of photovoltaic modules and photovoltaic arrays is established. The thermal property parameters of photovoltaic module materials, solar irradiance heat source, heat source reduction coefficient corresponding to photoelectric conversion efficiency, and external flow field inlet conditions or convective heat transfer boundary conditions determined based on sea surface wind speed are set. The temperature distribution cloud map of photovoltaic array, average operating temperature data and maximum temperature data of each photovoltaic module are calculated to form a temperature field dataset for photovoltaic performance monitoring point selection.

4. The method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform according to claim 1, characterized in that, S3 specifically includes: S31: Initial Screening of Candidate Sites for Structural Safety Monitoring: Based on the structural response data and local stress data obtained in S2, preset structural response thresholds and local stress thresholds are used to screen beam elements, connection nodes, and local high-stress areas whose peak equivalent stress, bending moment, nodal reaction force, mooring tension, or the maximum stress data of key nodes calculated by the local finite element sub-model exceeds the corresponding threshold. These are selected as candidate sites for structural safety monitoring, forming a candidate set of structures. The local stress threshold is 50%-70% of the allowable stress or design strength of the corresponding structural material. S32: Initial Screening of Photovoltaic Performance Monitoring Candidate Sites: Based on the photovoltaic module temperature field dataset obtained in S2, a preset temperature threshold is used to screen photovoltaic modules or local areas whose average operating temperature or maximum temperature exceeds the threshold, forming a temperature candidate set. The temperature threshold is 70%-90% of the rated maximum operating temperature of the photovoltaic module. S33: Merge and remove duplicates: Combine the candidate structure set With temperature candidate set Merging involves removing duplicate candidate points that are spatially overlapping and have the same monitoring type; for candidate points that are spatially overlapping but have different monitoring types, merging them into composite monitoring candidate points while retaining their corresponding monitoring parameter types, resulting in the total set of candidate points. .

5. The method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform according to claim 1, characterized in that, S5 specifically includes: S51: Preset upper limit N for the number of sensors and correlation threshold and monitoring coverage constraints; the upper limit N of the number of sensors is not less than the number of mandatory coverage points, and the correlation threshold The monitoring coverage constraint is 0.8, which includes: 100% coverage of high-risk areas with a risk coefficient exceeding 0.8 among the candidate sites for structural safety monitoring; and 100% coverage of high-temperature areas where the photovoltaic module temperature exceeds 90% of the rated maximum operating temperature. S52: Add the risk coefficient R from the total candidate site set U. i Candidate points for structural safety monitoring with a value exceeding 0.8 and candidate points for photovoltaic performance monitoring with a temperature exceeding 90% of the rated maximum operating temperature are marked as mandatory coverage points, while the remaining candidate points are marked as optional points. With the optimization objective of maximizing the comprehensive monitoring value index, the candidate points in U are sorted from high to low according to the currently calculated comprehensive monitoring value index. S53: Initialize the current set A to empty. Examine the candidate points i that are currently ranked first. If candidate point i is a mandatory coverage point and its corresponding high-risk or high-temperature area has not yet been covered, add candidate point i to the current set A. If candidate point i is an optional point, calculate the absolute value of the Pearson correlation coefficient between candidate point i and the monitoring data of the same type of monitoring points in the current set A, and take the maximum value. ,like Greater than If the high-risk or high-temperature area corresponding to candidate point i is already covered by the current set A, then candidate point i is considered a redundant point and excluded; otherwise, it is added to the current set A. After each candidate point is added to the current set A, the information non-redundancy coefficient and comprehensive monitoring value index of the remaining candidate points in U are recalculated, and the remaining candidate points are reordered. S54: Repeat S53 until the current set A reaches the preset upper limit of the number of sensors N, or all high-risk and high-temperature points that meet the mandatory coverage constraints have been covered and no more candidate points can be added. Output the current set A as the final set of optimal sensor placement points.

6. The method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible floating platform according to claim 1, characterized in that, In S6, the standard for verifying the monitoring accuracy is as follows: based on the simulation response data corresponding to the selected point, the representativeness of the full platform structural response data, local stress data and temperature field data is verified. The maximum relative error obtained from the verification does not exceed 8%, and it meets the preset requirements for coverage of high-risk structural points and high-temperature photovoltaic points.

7. A computer device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible platform as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the method for optimizing the arrangement of photovoltaic monitoring sensors on a semi-submersible platform as described in any one of claims 1-6.

Images