Water platform remote operation and maintenance method and system based on time sequence load

Abstract

The invention provides an overwater platform remote operation and maintenance method and system based on time sequence loads, and the method comprises the steps: dividing a buoy stress region based on structural characteristics and working condition simulation, and constructing a sensor layout set and load mapping model; acquiring load signals of the sensor arrangement set in real time, and outputting a filtered load signal set; estimating a regional total load based on the filtered load signal set, and constructing a spatial load matrix; an overload early warning threshold value is set based on the platform bearing capacity and the real-time load, and a load state vector is obtained; dividing risk levels based on time sequence evolution and fatigue prediction of a spatial load matrix; and evaluating platform safety based on the load state vector and the accumulated fatigue risk index, and visualizing a load thermodynamic diagram, a fatigue risk distribution diagram and a comprehensive analysis trend chart. Structural simulation and sequential load evolution analysis are combined, the dynamic coupling relation between the platform and the environment is constructed, and the method is suitable for remote monitoring, operation and maintenance of the water platform.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance technology for floating platforms, and in particular to a method and system for remote operation and maintenance of floating platforms based on time-series loads. Background Technology

[0002] With the rapid development of emerging industries such as water tourism, entertainment, and transportation, modular and interlocking floating platforms are widely used in scenarios such as dock construction, floating bridge laying, and water entertainment facilities, becoming an important carrier for expanding and multifunctionally utilizing nearshore water space. The structural safety of these platforms has become a key factor in ensuring their long-term stable operation. How to perceive real-time stress changes in local areas of the platform under complex dynamic load environments, identify potential structural fatigue and overload risks, and provide decision-making basis for remote operation and maintenance management and intelligent diagnosis has become a crucial problem that urgently needs to be solved in the field of intelligent management of floating platforms.

[0003] Chinese patent application CN119239854A discloses a method for a floating intelligent health and leisure platform. The method includes a floating mechanism comprising a float, a protrusion located on one outer edge of the float, a fitting portion located on the other outer edge of the float, a cover plate on the protrusion, and a connector on the fitting portion; and a connecting mechanism comprising a positioning rod disposed inside the protrusion, a control component sleeved on the outside of the positioning rod, a rotating component sleeved on the outside of the control component, and a lifting component on the top of the control component. This invention, through the cooperation between the control component and the rotating component, allows the rotating component to rotate by moving the sleeve up and down, thereby achieving a snap-fit ​​engagement between the buckle inside the protrusion and the connector on the fitting portion. The splicing of multiple floats does not require external tools, making the construction of the floating platform more convenient and faster. Users can customize the actual size of the floating platform according to their needs.

[0004] However, current technology still faces many challenges. In actual use, the structural load distribution of current floating platforms exhibits significant dynamic changes, especially during peak tourist seasons or special periods. Frequent occurrences of dense crowds, jumping, and stacking can easily lead to overloading or stress concentration in localized areas, potentially causing structural fatigue, deformation, or even fracture. Currently, most floating platforms rely on static structural design and regular manual inspections, lacking real-time dynamic perception of localized load distribution. This makes it difficult to identify hidden fatigue zones or sudden overload events, thus hindering data support for intelligent operation and maintenance. Especially in sudden situations such as rapid gatherings of people or violent disturbances, the platform's stress state can change drastically. Failure to promptly detect and respond to overload risks can easily lead to accelerated structural fatigue, localized collapse, or even fracture. Therefore, there is an urgent need for a real-time perception system for zoned load distribution and intelligent identification of regional overloads for floating platforms. This system would enable online monitoring and risk warning of platform structural health, thereby improving the intelligence level of remote operation and maintenance and the safety of platform use. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of existing technologies, this invention provides a method and system for remote operation and maintenance of floating platforms based on time-series loads. This method deploys a sensor array to continuously collect load data on key structural areas of the platform, such as deck edges, connection nodes, and pontoon support components. Through modeling and filtering of each stress sub-zone, the system reconstructs the stress distribution changes in different areas of the platform in real time. Furthermore, a fatigue sensitivity model is introduced to track the load evolution process under typical scenarios such as crowd gatherings during holidays, strenuous activities like jumping and dancing, and multi-point concentrated loads, such as crowds moving on the same side, to determine the fatigue accumulation trend of local platform structures. By establishing time-series overload response records and state transition maps, the system can issue early warnings when initial signs of local support deformation appear. This method is not only applicable to public facilities with large crowds, such as sightseeing floating bridges, water performance platforms, and lake tour platforms, but also provides auxiliary judgment criteria for load responses under sudden weather conditions (such as gusts and heavy rain), enabling maintenance intervention before accidents occur and preventing major safety incidents such as platform breakage and capsizing. Simultaneously, it provides relevant management departments with historical load archives and structural health assessment reports.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for remote operation and maintenance of a surface platform based on time-series loads includes:

[0008] Based on structural features and working condition simulation, the stress area of ​​the entire pontoon module is divided, and a model is constructed to map the overall sensor layout scheme S with the electrical characteristics and stress of the sensors.

[0009] Real-time acquisition of load signals from the overall sensor array layout scheme set S, and output of filtered sensor load signal set.

[0010] Based on the filtered sensor load signal set The total load of the floating platform area is estimated and the spatial load matrix is ​​constructed, and the spatial load matrix L(t) is constructed.

[0011] Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, the stress sub-regions d of the floating platform are analyzed. i Overload warning threshold setting and intelligent warning discrimination are performed to obtain the real-time load state vector P(t);

[0012] Based on the load time-series evolution analysis and cumulative fatigue risk prediction of the spatial load matrix L(t), fatigue risk levels are classified.

[0013] Based on the real-time load state vector P(t) and each force sub-region d i Cumulative fatigue risk index over historical time series Conduct a comprehensive platform safety assessment and remote visualization, and construct load heat maps, fatigue risk distribution maps, and comprehensive analysis trend maps.

[0014] Furthermore, the method for constructing the overall sensor layout scheme set S and the sensor electrical characteristics and force mapping relationship model includes:

[0015] Based on structural features and working condition simulation, the pontoon module is divided into multiple stress sub-regions, and the set of stress regions D of the pontoon module is constructed.

[0016] Based on the set of force-bearing regions D, design a sensor array layout scheme and construct an overall sensor display layout scheme set S;

[0017] Based on the overall sensor layout scheme set S, a model of sensor electrical characteristics and force mapping relationship is constructed.

[0018] The sensor array deployment scheme includes, based on the set of stress regions D, and for each stress sub-region d i This involves sequentially adjusting the deployment density driven by the area, prioritizing deployment guided by stress intensity, defining the sensor array set, and controlling the spacing constraints.

[0019] Furthermore, the filtered sensor load signal set The output steps include:

[0020] Based on the overall sensor array layout scheme set S, multi-channel sensor signals are acquired in real time in parallel to obtain the sensor voltage signal set V.

[0021] Based on the sensor electrical characteristics and force mapping model, the sensor voltage signal set V is linearly transformed to obtain the sensor load signal set F.

[0022] Low-pass filtering and noise reduction are performed on the sensor load signal set F, outputting the filtered sensor load signal set.

[0023] Furthermore, the step of constructing the spatial load matrix L(t) at time t includes:

[0024] Based on the filtered sensor load signal set For the force sub-partition d i Internal load estimates By performing weighted superposition, the force sub-partition d of each region is calculated. i Total load value at the current time t

[0025] Based on each force-affected sub-partition d i Total load value at the current time t Construct the spatial load matrix L(t) at time t;

[0026] For the force sub-partition d i Internal load estimates The design of weighting coefficients in weighted superposition includes three factors: sensor spatial layout, load representativeness, and the coupling relationship between partition area and sensor density.

[0027] Furthermore, the step of obtaining the real-time load state vector P(t) at time t includes:

[0028] By integrating structural design parameters, material mechanical properties, and finite element simulation results, the stress sub-partition d is obtained. i Theoretical maximum load value The maximum total load value S of the entire floating platform structure is obtained by summing up the values. all ;

[0029] Based on the maximum total load value S of the entire floating platform structure all Medium force sub-partition d i Theoretical maximum load value Set the corresponding overload warning threshold.

[0030] Based on force-dependent sub-partitions d i Total load value Overload warning threshold Intelligent assessment of overload risk is achieved by constructing a Boolean discriminant function for real-time identification of overloaded areas on the water platform, generating a real-time load state vector P(t) at time t.

[0031] Furthermore, the steps of load time-series evolution analysis and cumulative fatigue risk prediction based on the spatial load matrix L(t) include:

[0032] Based on the spatial load matrix L(t), the load time series data at K consecutive time points are extracted by serialization along the time axis.

[0033] Based on load time series data of K consecutive time points Construct a fatigue sensitivity model and calculate the stress sub-partition d. i Cumulative fatigue risk index over K consecutive historical time series

[0034] Furthermore, the steps for classifying the fatigue risk level include:

[0035] Based on force-dependent sub-partitions d i Cumulative fatigue risk index over K consecutive historical time series Three different fatigue risk level thresholds θ1, θ2, and θ3 are set, and the stress sub-region d is judged according to the rules. i Classify fatigue risk levels;

[0036] The rule judgment includes: if the cumulative fatigue risk index If the stress sub-region d is less than the fatigue risk level threshold θ1, then... i Low risk; if the cumulative fatigue risk index If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ1 and less than the fatigue risk level threshold θ2, then... i The risk level is medium; if the cumulative fatigue risk index is... If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ2 and less than the fatigue risk level threshold θ3, then... i High risk; if the cumulative fatigue risk index If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ3, then... i This is extremely high risk.

[0037] Furthermore, the steps for conducting the overall security assessment and remote visualization of the platform include:

[0038] Based on the real-time load state vector P(t) and the force sub-region d i Cumulative fatigue risk index over K consecutive historical time series Construct the overall safety state function of the floating platform at time t;

[0039] Based on the real-time load state vector P(t) and each force sub-partition d i The cumulative fatigue risk index over K consecutive time periods and the overall safety state function S at time t platform (t) is used for remote visualization design, and load heat map, fatigue risk distribution map and comprehensive analysis trend map are constructed respectively.

[0040] Furthermore, the steps for constructing the load heat map, fatigue risk distribution map, and comprehensive analysis trend map include:

[0041] Based on the force sub-partition d in the real-time load state vector P(t) i Real-time load status value A discrete color-level rendering strategy is used to construct a load heatmap of the water platform;

[0042] Based on the force sub-partition of the water platform d i The cumulative fatigue risk index over K consecutive time periods A four-level risk level color scheme was used to construct a fatigue risk distribution map for the water platform.

[0043] Based on the overall safety state function S of the water platform at time t platform (t), with time series as the horizontal axis and comprehensive risk index as the vertical axis, load peak points are marked, automatic alarm records are recorded, and manual maintenance time points are recorded in sequence to construct a comprehensive analysis trend chart of the floating platform changing over time;

[0044] The discrete color level rendering strategy includes: if the force-affected sub-partition d i Real-time load status value Denotes the i-th force-bearing sub-partition d i In an overloaded state, it is highlighted in red in the diagram; if the force sub-region d i Real-time load status value Denotes the i-th force-bearing sub-partition d i Under normal operating conditions, it is marked in gray in the diagram;

[0045] The four-level risk level color matching strategy includes: if the cumulative fatigue risk index If the stress sub-region d is less than the fatigue risk level threshold θ1, then... i Low risk is indicated by green in the graph; if the cumulative fatigue risk index is low... If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ1 and less than the fatigue risk level threshold θ2, then... i Medium risk is indicated by yellow in the chart; if the cumulative fatigue risk index... If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ2 and less than the fatigue risk level threshold θ3, then... i High-risk areas are marked in orange on the chart; if the cumulative fatigue risk index... If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ3, then... i Extremely high risk, marked in red on the map;

[0046] The steps of sequentially performing load peak point marking, automatic system alarm recording, and manual maintenance time point recording include: Load peak point marking refers to the real-time load state values ​​of multiple stress sub-regions at a given time t. Simultaneously satisfy The point is highlighted with an orange dot on the trend curve of the comprehensive analysis chart to indicate a short-term concentrated load event; the system's automatic alarm record indicates that if at time t, the overall safety state function S platform (t) Exceeds the set alarm threshold S thresh If a point is found to be faulty, it will be highlighted with a red dot on the curve of the comprehensive analysis trend chart to remind maintenance personnel to pay close attention. The manual maintenance time point recording operation means that if there are manual maintenance or reinforcement operations in the history of the water platform, the system can import the relevant data and highlight the corresponding time point with a blue dot to facilitate the subsequent tracking and evaluation of the maintenance effect.

[0047] A system for remote operation and maintenance of a surface platform based on time-series loads, used to implement the aforementioned method for remote operation and maintenance of a surface platform based on time-series loads, the system comprising:

[0048] Force perception modeling module: Based on structural features and working condition simulation, the force-bearing area of ​​the entire pontoon module is divided, and a set of overall sensor layout schemes and sensor electrical characteristics and force mapping relationship models are constructed.

[0049] Sensor load signal acquisition and preprocessing module: Used to acquire load signals from the overall sensor array layout scheme set S in real time, and output a filtered set of sensor load signals.

[0050] Space payload estimation and matrix construction module: based on the filtered sensor payload signal set The total load of the floating platform area is estimated and the spatial load matrix is ​​constructed, and the spatial load matrix L(t) is constructed.

[0051] Structural bearing capacity assessment and overload early warning module: Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, it assesses the load-bearing capacity of each stress sub-zone d of the floating platform. i Overload warning threshold setting and intelligent warning discrimination are performed to obtain the real-time load state vector P(t);

[0052] Load-time fatigue risk prediction module: Based on the spatial load matrix L(t), load-time evolution analysis and cumulative fatigue risk prediction are performed, and fatigue risk levels are classified.

[0053] Platform overall safety assessment and remote visualization module: based on real-time load state vector P(t) and each force sub-partition d i Cumulative fatigue risk index over historical time series Conduct a comprehensive platform safety assessment and remote visualization, and construct load heat maps, fatigue risk distribution maps, and comprehensive analysis trend maps.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0055] This invention avoids the problems of monitoring redundancy or blind spots in key areas caused by traditional uniform sensor deployment by dividing the float module structure into refined stress sub-regions based on structural features and working condition simulation. Furthermore, based on the stress characteristics of each stress sub-region and sensor deployment rules, a sensor arrangement scheme for mechanically sensitive areas is formulated to improve the sensitivity and resource utilization of sensor deployment. At the same time, by combining the electrical response characteristics of sensors with the quantitative modeling of structural stress state, a force and electrical co-inversion mechanism is formed, which can realize the dynamic tracking and spatial difference modeling of local stress state under complex working conditions such as wind, waves, and sudden load changes, thereby realizing intelligent monitoring of the floating platform.

[0056] This application introduces a multi-channel parallel acquisition mechanism to achieve real-time synchronous acquisition of voltage signals from various sensors, avoiding the time delay and information distortion problems existing in traditional polling acquisition. Furthermore, based on the linear mapping relationship between the electrical response characteristics of the sensors and the mechanical load, the sensor voltage signals are linearly transformed to eliminate the load estimation error caused by the inconsistency of responses between different sensors. By introducing low-pass filtering and noise suppression algorithms, the signal fluctuation problem caused by high-frequency interference in complex environments is improved, and a stable load signal output is achieved.

[0057] This application avoids the force estimation deviation caused by the dispersion of signal sampling in traditional methods by performing regional weighted superposition processing on the filtered sensor load signal, thereby improving the spatial consistency and boundary continuity of load estimation for each force sub-zone of the pontoon. Furthermore, by constructing a spatial load matrix of the load values ​​of each force sub-zone on a unified time scale, the synchronous evolution characteristics of the structural stress state in the spatial and temporal dimensions are preserved, thus improving the problem that traditional static load assessment methods cannot reflect dynamic force distribution changes.

[0058] This application obtains the theoretical maximum load capacity of each stress sub-zone by integrating the structural design parameters, material mechanical properties, and finite element simulation results of the floating platform, avoiding the load capacity error caused by traditional single parameter estimation. Furthermore, it sets overload warning thresholds for each stress sub-zone based on the maximum load capacity of each stress sub-zone, thereby effectively distinguishing between local overload and overall load-bearing risk. By constructing a Boolean discriminant function for real-time identification of overload risk areas of the floating platform, the accuracy and response speed of overload warning for the floating platform structure are improved.

[0059] This application addresses the problem that traditional load analysis cannot effectively reflect dynamic load changes by performing time-series processing on the spatial load matrix, thereby enhancing the ability to capture the temporal characteristics of the load. Furthermore, based on the fatigue sensitivity model and combined with load data from continuous historical moments, it calculates the cumulative fatigue risk index of the historical time series of each stress sub-partition, thereby achieving a quantitative assessment of fatigue damage. By setting graded fatigue risk standards, it achieves dynamic hierarchical identification of fatigue risk.

[0060] This application establishes a dynamic overall safety status assessment model by integrating the real-time load state vector with the cumulative fatigue risk index of the historical time sequence of each stress sub-zone. This avoids the limitations of traditional single indicators that cannot accurately reflect the complex loads and fatigue states of the platform. Subsequently, remote visualization design is carried out to realize the synchronous dynamic display of the load distribution heat map, fatigue risk area map and trend analysis map of the platform, which improves the intuitiveness and real-time nature of structural safety information. Through dynamic visualization, the ability to perceive the safety situation and provide early warning of key stress parts of the platform is enhanced, and the decision-making process for structural safety management and maintenance is optimized. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the 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. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating the principle of a remote operation and maintenance method for a surface platform based on time-series loads according to the present invention.

[0063] Figure 2 This is a flowchart of a method for constructing an overall sensor array and deployment scheme set and a model of sensor electrical characteristics and force mapping relationship in a method for remote operation and maintenance of a water platform based on time-series loads according to the present invention.

[0064] Figure 3This is a flowchart of a method for outputting a filtered sensor load signal set in a method for remote operation and maintenance of a floating platform based on time-series loads according to the present invention;

[0065] Figure 4 This is a flowchart of the method for constructing a spatial load matrix in a method for remote operation and maintenance of a surface platform based on time-series loads according to the present invention;

[0066] Figure 5 This is a flowchart of a method for obtaining a real-time load state vector in a method for remote operation and maintenance of a floating platform based on time-series loads according to the present invention.

[0067] Figure 6 This is a flowchart of a method for classifying fatigue risk levels in a remote operation and maintenance method for a surface platform based on time-series load, according to the present invention.

[0068] Figure 7 This is a flowchart of a method for constructing a load heat map, fatigue risk distribution map, and comprehensive analysis trend map of a water platform in a remote operation and maintenance method based on time-series loads according to the present invention.

[0069] Figure 8 This is a functional block diagram of a system for remote operation and maintenance of a floating platform based on time-series load, according to the present invention. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described 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.

[0071] Example 1:

[0072] Please see Figure 1 As shown, this embodiment provides a method for remote operation and maintenance of a surface platform based on time-series loads, including:

[0073] Step S1000: Based on structural features and working condition simulation, the stress area of ​​the entire pontoon module is divided, and a model is constructed to map the overall sensor layout scheme set S with the electrical characteristics and stress of the sensors.

[0074] Specifically, this step aims to divide the pontoon module into key stress areas by analyzing the structural mechanical characteristics and working condition simulation results, deploy flexible piezoresistive sensor arrays to formulate an overall sensor layout scheme set S, and establish a sensor electrical characteristics and force mapping relationship model to realize the modular integration of the pontoon structure's self-sensing capability, providing a data foundation for subsequent structural health monitoring and intelligent feedback control.

[0075] Furthermore, such as Figure 2 As shown, step S1000 includes:

[0076] Step S1100: Based on structural features and working condition simulation, the pontoon module is divided into several stress sub-regions to construct the set of stress regions D of the entire pontoon module.

[0077] Specifically, this step aims to systematically evaluate the overall stress distribution characteristics of the pontoon module based on its structural configuration and functional attributes, combined with historical load data and finite element simulation analysis results. Based on this, the entire pontoon module is divided into several representative stress sub-regions, which serve as the basic units of the sensors, in order to delineate and cover the sensing area.

[0078] In the specific implementation process, firstly, based on the usage scenarios and structural functional characteristics of the pontoon module, five typical stress areas are identified, including: the central load-bearing area d center The central region of the module primarily bears the concentrated loads on the platform; the edge load-bearing area d edge The outer area of ​​the module is frequently subjected to external forces such as diving and edge manipulation; the fixed area of ​​the apparatus d fixture Structural area used for installing anchors, planks, handrails, and other components; connecting splicing area d joint It is the interface between modules, where there is a tendency for stress concentration; the passage area d passage It is a high-frequency passageway for people, and although the force is dispersed, it changes frequently.

[0079] Based on the preliminary identification of the stress-bearing areas, the stress distribution of the pontoon structure under typical working conditions (such as personnel load, wave impact, equipment installation, etc.) is simulated using the finite element method. Supplemented by historical measured data, high stress concentration areas, stress fluctuation areas, and low stress areas are further identified. Based on this, the number and boundary delineation of each sub-zone are dynamically adjusted to optimize the sub-zone morphology, improve the stress representativeness of each area, and balance the density control of sensor deployment with the effectiveness of data acquisition. The specific formula is as follows:

[0080] D={d i |i=1,2,...,n}

[0081] Where D represents the set of stress-bearing areas of the entire pontoon module; di This represents the i-th stress-bearing sub-section in the pontoon module; i represents the number of the stress-bearing sub-section; n represents the total number of stress-bearing sub-sections in the pontoon structure.

[0082] Each force-bearing sub-section d in the float module i Satisfy d i ∈{d center ,d center ,d fixture ,d joint ,d passage Furthermore, each force-bearing sub-partition satisfies the following geometrical logical relationship: and Where ∈ represents an element belonging to the set; ∪ represents the union symbol, which is the set of multiple force sub-partitions; ∩ represents the intersection symbol, which is the intersection between the force sub-partitions; This indicates an empty set, meaning there is no overlap between the force-bearing sub-partitions.

[0083] Step S1200: Design a sensor array layout scheme based on the set of force-bearing areas D, and construct an overall sensor array layout scheme set S.

[0084] Specifically, this step aims to formulate a targeted flexible piezoresistive sensor array deployment strategy based on the set of stress regions D defined in step S1100, taking into account the area, stress characteristic distribution and deployment feasibility of each stress sub-region, so as to ensure the effective perception of key stress characteristics and the integrity of data acquisition coverage.

[0085] In the specific implementation process, based on the set of force-bearing regions D, for each force-bearing sub-region d i The sensor layout design should be based on the following principles:

[0086] Firstly, the deployment density is adjusted based on the area. For larger sub-regions subject to stress, considering their large spatial range and complex mechanical distribution, the sensor deployment density should be increased accordingly to avoid monitoring blind spots. Deployment density ρ i With area A(d) i They are directly proportional, i.e., ρ i ∝A(d i ), where ρ i Denotes the i-th force-bearing sub-partition d i The sensor deployment density, i.e., the number of sensors per unit area, can be preset based on factors such as sensor physical size, coverage area, and data accuracy, or determined through simulation calibration; A(d i ) represents the i-th force-bearing sub-partition d iThe area can be automatically obtained through CAD design drawings, BIM models, or geometric modeling of the pontoon structure; ∝ represents a proportional relationship, used to reflect the proportional relationship between variables.

[0087] Secondly, the key layout guided by stress intensity. To improve the detection effect, historical stress detection data and finite element simulation results are used to identify stress sub-regions d. i The region where the stress distribution is most concentrated, i.e., the high-stress center area. Furthermore, the sensor density within this area is appropriately increased to form a locally denser deployment strategy, enabling the collection of more critical data with fewer sensors. (The high-stress center area...) This refers to a localized area where the stress value reaches its peak or fluctuates frequently under typical load conditions. This area is often a high-risk point for potential fatigue, damage, or fracture. Enhancing the area helps improve early warning of faults.

[0088] Thirdly, the definition of the sensor array set. To clearly define the force-bearing sub-region d... i Based on the sensor deployment, construct the sensor set S for the force-bearing sub-region. i The specific formula is as follows:

[0089] S i ={s ij |i=1,2,...,n∧j=1,2,...,m i}

[0090] Among them, S i Denotes the i-th force-bearing sub-partition d i The corresponding sensor set is usually m i A collection of sensors; s ij Denotes the i-th force-bearing sub-partition d i The j-th sensor; m i Denotes the i-th force-bearing sub-partition d i The number of sensors deployed internally; ∧ represents the logical symbol connecting multiple conditions.

[0091] Force sub-partition d i Number of sensors deployed internally (m) i Taking into account factors such as the sensor deployment density of the stress-bearing sub-zone, the area of ​​the stress-bearing sub-zone, and the area of ​​the high-stress center zone, the specific process formula is as follows:

[0092]

[0093] in, This indicates rounding up to ensure an integer number of sensor devices; ρ i ·A(d i ) represents the i-th force-bearing sub-partition d iThe number of sensors; λ i This represents the local density factor, used to adjust the high-stress center area. The deployment density is usually λ i >1; Indicates the high stress center area The area of ​​the layout density can be automatically calculated from structural CAD drawings or finite element mesh elements; Denotes the i-th force-bearing sub-partition d i The number of sensors in the high-stress center area; k i ∈(0,1] represents the correction factor, which is set empirically or obtained through on-site constraint setting and simulation optimization;

[0094] Fourth, spacing constraint control. To prevent data redundancy or interference caused by sensors being too close together, or detection blind spots caused by sensors being too far apart, the spacing δ between sensors within each sub-zone needs to be controlled. ij The specific process for setting constraints is as follows:

[0095]

[0096] Where, δ ij Represents any two sensors s ij and s ik The spacing between sensors refers to the spacing between any two sensors within a given stress sub-zone; δ min This represents the minimum spacing between sensors. If the spacing is less than this value, it may lead to signal redundancy or mutual interference, resulting in a decrease in data acquisition efficiency, such as when two piezoresistive sensors overlap for sampling; δ max This indicates the maximum spacing between sensors. A spacing greater than this may result in a detection blind zone, where the central area cannot effectively sense changes in force. ij ,s ik Represent the i-th force-bearing sub-partition d respectively i The j-th and k-th sensors, k = 1, 2, ..., m i .

[0097] Finally, based on the above layout principles and constraints, the force-bearing sub-partition d is... i The corresponding sensor set S i A comprehensive set of sensor display and deployment schemes was constructed.

[0098] Step S1300: Based on the overall sensor layout scheme set S, construct a model of sensor electrical characteristics and force mapping relationship.

[0099] Specifically, this step aims to, based on the overall sensor layout scheme set S generated in step S1200, target each force-bearing sub-zone d. i For the flexible piezoresistive sensors already deployed in China, a quantitative mapping model between their output voltage and actual force is constructed, providing an electrical signal conversion basis for real-time detection and identification of structural stress states. Based on the physical mechanism that the internal resistance of the flexible piezoresistive sensor changes under varying stress conditions, thereby causing changes in output voltage, a mathematical relationship between voltage and force is established.

[0100] In the specific implementation process, based on experimental calibration results and engineering application requirements, the sensor output voltage The force it perceives It exhibits a good linear relationship within a certain working range, and the specific process formula is as follows:

[0101]

[0102] in, Denotes the i-th force-bearing sub-partition d i The j-th sensor s ij The sensed force is expressed in N; k represents the voltage-force linear gain coefficient of the sensor, expressed in N / V, which reflects the sensitivity of the voltage output to changes in force and is obtained through calibration experiments. Denotes the i-th force-bearing sub-partition d i The j-th sensor s ij The output voltage value is expressed in V; b represents the bias term of the linear model, which is the basic force value under zero-point voltage, and is usually related to the sensor's factory condition, circuit design, and initial mechanical state.

[0103] Step S2000: Real-time acquisition of load signals from the overall sensor array layout scheme set S, and output of the filtered sensor load signal set.

[0104] Specifically, this step aims to perform real-time signal acquisition and preliminary filtering of the flexible piezoresistive sensors deployed in the overall sensor array layout scheme set S constructed in step S1200. By preprocessing the raw output signals of the sensors, environmental noise and invalid disturbances are filtered out, thereby extracting stable and reliable force data, laying a data foundation for subsequent load distribution analysis and force identification.

[0105] Furthermore, such as Figure 3 As shown, step S2000 includes:

[0106] Step S2100: Based on the overall sensor array layout scheme set S, perform real-time parallel acquisition of multi-channel sensor signals to obtain the sensor voltage signal set V.

[0107] Specifically, this step, based on the overall sensor layout scheme set S constructed in step S1200, involves all the sensor sets S of the force-bearing sub-zones. i ={s ij |i=1,2,...,n; j=1,2,...,m i} as the object of signal acquisition.

[0108] In practice, by deploying a multi-channel signal acquisition module, the system can monitor each sensor within each force-bearing sub-zone. ij Instantaneous voltage signal output at the same time t Parallel data acquisition is performed. The multi-channel signal acquisition module supports the parallel acquisition of n×m signals and ensures that the sampling time of each channel is consistent, realizing synchronous data acquisition of the entire array.

[0109] During the data acquisition process, each sensor s within each force-bearing sub-region... ij Output instantaneous voltage signal Recording is completed at the same timestamp t, resulting in a set of sensor voltage signals containing the voltage outputs of all channels.

[0110] Where i represents the number of the force-bearing sub-region, n is the total number of force-bearing sub-regions; j represents the number of the flexible piezoresistive sensor in the i-th force-bearing sub-region, and there are m such sensors. i A flexible piezoresistive sensor.

[0111] Step S2200: Based on the sensor electrical characteristics and force mapping relationship model, perform linear transformation on the sensor voltage signal set V to obtain the sensor load signal set F.

[0112] Specifically, this step aims to process the sensor voltage signal set obtained in step S2100 based on the sensor electrical characteristics and force mapping relationship model established in step S1300. A linear transformation is performed to convert the voltage signals of each sensor into corresponding load signals, thereby enabling a preliminary quantitative estimation of the load signals.

[0113] In the specific implementation process, for the i-th force-bearing sub-partition d i The j-th flexible piezoresistive sensor s ij The instantaneous output voltage signal value at time t Based on its corresponding linear calibration model, its instantaneous load estimate can be calculated. The specific process formula is as follows:

[0114]

[0115] in, Denotes the i-th force-bearing sub-partition d i The j-th sensor s ij The estimated instantaneous load at time t, in N; Denotes the i-th force-bearing sub-partition d i The j-th sensor s ij The instantaneous output voltage value at time t; k and b represent the linear calibration coefficients of the sensor, which are the sensitivity (slope) and bias, respectively; t represents the time variable, which is the current sampling time point.

[0116] Through the above linear transformation process, the system can obtain the set of sensor load signals for all sensors at each time.

[0117] Step S2300: Perform low-pass filtering and noise reduction processing on the sensor load signal set F, and output the filtered sensor load signal set.

[0118] Specifically, this step aims to estimate the instantaneous load at each time point t based on the sensor load signal set F obtained in step S2200. Further low-pass filtering and noise reduction processing of the signal are performed to further suppress high-frequency noise and external environmental interference in the load signal, thereby improving the stability and physical availability of the signal.

[0119] In the specific implementation process, for the i-th force-bearing sub-partition d i The j-th sensor s ij Estimated instantaneous load at time t The signal is filtered using a low-pass filter (LPE) operator, and the filtered sensor load signal set is output. The specific process formula is as follows:

[0120]

[0121] in, LPE(·) represents the filtered instantaneous load estimate, where the load signal is stable and effective. LPE(·) represents the low-pass filter operator, and the specific implementation method can be selected according to the requirements of the actual application scenario, including but not limited to first-order RC filter, Kalman filter, sliding window averaging and other algorithms.

[0122] Step S3000, based on the filtered sensor load signal set The total load of the floating platform area is estimated and the spatial load matrix is ​​constructed, and the spatial load matrix L(t) is constructed.

[0123] Specifically, this step aims to utilize the filtered sensor load signal set obtained in step S2300. According to the preset spatial partitioning rules, the total load of each stress-bearing sub-region of the water platform is dynamically estimated, and the overall load distribution matrix of the platform at the current time t is constructed. The overall load distribution matrix of the platform is used to depict the regional load status of the platform in real time, track its dynamic evolution process, and realize overload trend early warning, providing basic data support for subsequent structural risk assessment and load scheduling strategies.

[0124] Furthermore, such as Figure 4 As shown, step S3000 includes:

[0125] Step S3100, based on the filtered sensor load signal set For the force sub-partition d i Internal load estimates By performing weighted superposition, the force sub-partition d of each region is calculated. i Total load value at the current time t

[0126] Specifically, this step is based on the filtered set of sensor load signals. For each force-bearing sub-region d i Instantaneous load estimates of all internally deployed flexible piezoresistive sensors By performing weighted summation, the total load estimate for each sub-partition at the current time t can be calculated.

[0127] In the specific implementation process, at time t, the system performs calculations for each force-bearing sub-partition d. i (i = 1, 2, ..., n), divide each force sub-partition d i Estimated load values ​​after filtering for all flexible piezoresistive sensors By performing weighted superposition, the force sub-partition d of each region is calculated. i Total load value at the current time t The specific process formula is as follows:

[0128]

[0129] in, Denotes the i-th force-bearing sub-partition d i The total load value at time t, in N; j represents the j-th flexible piezoresistive sensor within the i-th force-bearing sub-region, with a total of m. i A flexible piezoresistive sensor; w ijLet represent the weighting coefficient of the j-th flexible piezoresistive sensor within the i-th force-bearing sub-region. This coefficient reflects the representativeness of the sensor within the sub-region and satisfies the normalization constraint condition, i.e. Denotes the i-th force-bearing sub-partition d i The j-th sensor s ij The estimated instantaneous load at time t.

[0130] The weighting coefficient w ij The design takes into account the following three factors:

[0131] Firstly, sensor spatial layout factors: the spatial geometric distribution density of sensors within sub-regions, prioritizing higher weighting for central areas or key boundary points;

[0132] Secondly, load representativeness factors: typical force response characteristics of the local area where the sensor is located, with priority given to whether it is a high-response sensitive area or a critical load-bearing location;

[0133] Third, the coupling relationship between partition area and sensor density: By combining the size of the partition area with the density of sensor deployment within it, the overall weight contribution of sensors in different areas is dynamically adjusted.

[0134] Step S3200, based on each force sub-partition d i Total load value at the current time t Construct the spatial load matrix L(t) at time t.

[0135] Specifically, based on the total load estimation of each stress sub-region, and based on the current time t, each stress sub-region d i Total load value Construct the spatial load matrix L(t) of the water platform at time t to reflect the two-dimensional load distribution characteristics of the overall structure.

[0136] In the specific implementation process, at time t, each force-bearing sub-partition d is divided. i Total load value The data is then aggregated to construct the regional load set L'(t) of the platform at time t. The specific process and formula are as follows:

[0137]

[0138] Where i represents the number of the stress-bearing sub-section; n represents the total number of stress-bearing sub-sections in the pontoon structure.

[0139] Furthermore, the regional load set L'(t) at time t is reconstructed into a matrix to construct the spatial load matrix L(t) at time t. The specific expression formula is as follows:

[0140]

[0141] Where L(t) represents the spatial load matrix of the water platform at time t; Let vector L(t) be an n-row, 1-column real matrix space.

[0142] Step S4000: Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, analyze the stress sub-regions d of the floating platform. i Overload warning threshold setting and intelligent warning discrimination are performed to obtain the real-time load state vector P(t).

[0143] Specifically, this step aims to analyze the load-bearing capacity and real-time load distribution information of each stress sub-zone d of the floating platform based on the theoretical bearing capacity and real-time load distribution information of the platform structure. i Total load value Dynamic threshold determination and intelligent overload early warning analysis are performed to achieve dynamic identification, early warning triggering and status feedback of regional structural overload risks.

[0144] Furthermore, such as Figure 5 As shown, step S4000 includes:

[0145] Step S4100: Integrate structural design parameters, material mechanical property indicators, and finite element simulation results to obtain the stress sub-partition d. i Theoretical maximum load value The maximum total load value S of the entire floating platform structure is obtained by summing up the values. all .

[0146] Specifically, this step aims to establish the structural bearing capacity limit model for each stress-bearing sub-zone of the floating platform, which will serve as the basic reference index for subsequent overload assessment. The system extracts the maximum bearing capacity of each stress-bearing sub-zone under normal operating conditions by integrating structural design parameters, material mechanical performance indicators, and finite element simulation results.

[0147] In the specific implementation process, the structural design drawings and mechanical modeling files of the floating platform are first analyzed to obtain the force sub-partitions d in the platform structure. i The system then uses the spatial range (i = 1, 2, ..., n) and component layout information to extract the mechanical performance indicators corresponding to the main structural components of each load-bearing sub-zone, including but not limited to compressive strength, elastic modulus, and yield strength, in conjunction with a material parameter database. Finally, based on the above information, the system can obtain the theoretical maximum load value of each sub-zone through one or more of the following methods in combination. Including but not limited to: static analysis methods, finite element simulation (FEM), empirical regression models, and reverse engineering based on measured data from prototype platforms.

[0148] Furthermore, after completing the structural bearing capacity modeling of each stress-bearing sub-region, the i-th stress-bearing sub-region d is obtained. i Maximum load that can be withstood under ideal working conditions The unit is N, which is used to characterize the maximum load that the stress sub-region can withstand without structural damage.

[0149] Finally, all the force-bearing sub-partitions d i Maximum load value These parameters are uniformly stored in a structural parameter mapping table, and are then aggregated to form the maximum total load value S for the entire floating platform structure. all The specific formula is as follows:

[0150]

[0151] Step S4200, based on the maximum total load value S of the entire floating platform structure all Each force-bearing sub-partition d i Theoretical maximum load value Set the corresponding overload warning threshold.

[0152] Specifically, this step aims to determine the maximum total load value S of the entire floating platform structure. all Each force-bearing sub-partition d i Theoretical maximum load value For each force sub-zone d of the floating platform i Set the appropriate overload warning threshold. By introducing a safety factor mechanism, the platform can make early judgments on the actual load status and provide risk warnings, thereby improving the platform's overall safety redundancy and risk resistance capabilities.

[0153] In the specific implementation process, for each force-bearing sub-partition d i Set the corresponding overload warning threshold. The specific process formula is as follows:

[0154]

[0155] in, Denotes the i-th force-bearing sub-partition d i The overload warning threshold is in N; α represents the safety factor, which ranges from (0,1) and is usually set according to the purpose, grade, material grade and operating environment of the platform. The recommended value is 0.6 to 0.9 to reserve structural redundancy space. Denotes the i-th force-bearing sub-partition d i The maximum load that can be withstood under ideal working conditions.

[0156] For example, taking the area d of a certain water platform i For example, if the region d i The theoretical maximum load value under ideal conditions Given a current of 5000N and a safety factor α of 0.8, the value of d in this region is... i Overload warning threshold This means that if the region d i When the load value exceeds 4000N, the system will trigger an early warning system, thus reserving a 20% structural safety redundancy for actual operation, effectively ensuring the stability and safety of the water platform under sudden overload conditions.

[0157] Step S4300, based on the force sub-partition d i Total load value Overload warning threshold Intelligent assessment of overload risk is achieved by constructing a Boolean discriminant function for real-time identification of overloaded areas on the water platform, generating a real-time load state vector P(t) at time t.

[0158] Specifically, this step aims to construct a Boolean discriminant function to achieve real-time intelligent identification of overload risk areas in each stress sub-zone of the water platform.

[0159] In the specific implementation process, this step is based on each force sub-partition d in the spatial load matrix L(t) obtained at a certain time t in step S3200. i Total load value And in conjunction with the force sub-regions d set in step S4200 i Overload warning threshold For each force-bearing sub-region d i The system intelligently determines the load status, thereby enabling real-time identification and recording of overloaded areas. The specific process formula is as follows:

[0160]

[0161] in, Denotes the i-th force-bearing sub-partition d i The real-time load state value, i.e., the Boolean judgment output, if the i-th force sub-partition d i Total load value Greater than the set overload warning threshold When the overload condition is set to 1, it indicates that the stressed sub-partition is overloaded; conversely, a value of 0 indicates that the stressed sub-partition is within the safe range and is not overloaded.

[0162] Furthermore, at time t, construct a sub-partition d containing all force-bearing components. i The real-time load state vector is expressed by the following formula:

[0163]

[0164] Where P(t) represents the real-time load state vector at time t, which is an n-dimensional Boolean vector.

[0165] Step S5000: Based on the load time-series evolution analysis and cumulative fatigue risk prediction of the spatial load matrix L(t), fatigue risk level classification is performed.

[0166] Specifically, this step aims to analyze the stress sub-region d based on historical time-series load data. i Long-term load variation trend analysis and fatigue risk assessment are conducted to identify potential structural fatigue hazards, thereby providing scientific support for subsequent structural maintenance optimization, reinforcement decisions and life cycle management.

[0167] Furthermore, such as Figure 6 As shown, step S5000 includes:

[0168] Step S5100: Based on the spatial load matrix L(t), perform serialization processing along the time axis to extract load time series data at K consecutive time points.

[0169] Specifically, this step aims to extract and construct each force-bearing sub-partition d. i The load time series within the historical operating cycle provides continuous and standardized input data for subsequent fatigue analysis.

[0170] In the specific implementation process, this step is based on the spatial load matrix of the water platform at time t constructed in step S3200. It is serialized along the time axis. Using equal-interval sampling, each force sub-partition d in the water platform is extracted along the time dimension. i From the load data at K consecutive time points, obtain the i-th force sub-partition d. i Load time series data at K consecutive time points in, Indicates at time t k When the i-th force-bearing sub-partition d i The total load value; K represents the total number of time points in the load time series data; T is... The superscript indicates the transpose of the matrix.

[0171] Step S5200: Based on load time series data over K consecutive time periods Construct a fatigue sensitivity model and calculate d for each stress sub-region. i Cumulative fatigue risk index over K consecutive historical time series

[0172] Specifically, this step aims to utilize the load time series data obtained in step S5100 over K consecutive time periods. For each force-bearing sub-partition d i Fatigue sensitivity modeling is performed, and its cumulative fatigue risk index is calculated to quantify the potential fatigue risk level it faces over historical periods.

[0173] In the specific implementation process, in order to reasonably assess the actual impact of load on structural fatigue, a fatigue weighting index m is first set, and its value should satisfy m>1. It is usually recommended that the value range be 2 to 4, which is used to enhance the weight expression of load peak in the risk accumulation process, thereby improving the sensitivity to occasional high load events.

[0174] At the same time, the i-th force-bearing sub-partition d is set. i Theoretical fatigue tolerance threshold As a normalization standard, it is used to measure the degree of risk of the current load level relative to the structural fatigue limit.

[0175] Based on the above parameter settings, calculate the i-th force-bearing sub-partition d. i Cumulative fatigue risk index over K consecutive historical time series The specific process formula is as follows:

[0176]

[0177] in, Indicates at time t k When the i-th force-bearing sub-partition d i Total load value; Denotes the i-th force-bearing sub-partition d i The theoretical fatigue tolerance threshold; Represents the current time t k Force sub-partition d i The total load value is the ratio of the theoretical fatigue tolerance threshold, i.e., the load ratio. The closer it is to 1 or greater than 1, the closer it is to the fatigue limit or the load condition has been exceeded. m represents the fatigue weighting index, which is usually >1 and is used to highlight the aggravating effect of high load on fatigue loss.

[0178] Step S5300, based on the force sub-partition d i Cumulative fatigue risk index over K consecutive historical time series For the force sub-partition d i Classify fatigue risk levels.

[0179] Specifically, the i-th force-bearing sub-partition d is obtained according to step S5200. i Cumulative fatigue risk index over K consecutive historical time series The system applies force to the i-th sub-partition d.i The fatigue risk level is classified, and the risk assessment results are output for subsequent processing.

[0180] In the specific implementation process, based on the cumulative fatigue risk index Three different fatigue risk level thresholds, θ1, θ2, and θ3, are set, and the following rules are applied:

[0181] If the cumulative fatigue risk index If the stress level is less than the fatigue risk level threshold θ1, then the stress sub-region d i Low risk;

[0182] If the cumulative fatigue risk index If the stress level is greater than or equal to the fatigue risk level threshold θ1 and less than the fatigue risk level threshold θ2, then the stress sub-region d i Medium risk;

[0183] If the cumulative fatigue risk index If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ2 and less than the fatigue risk level threshold θ3, then... i High risk;

[0184] If the cumulative fatigue risk index If the stress level is greater than or equal to the fatigue risk level threshold θ3, then the stress sub-region d i This poses an extremely high risk and should be reinforced or its use restricted in a timely manner.

[0185] Step S6000, based on the real-time load state vector P(t) and each force sub-partition d i Cumulative fatigue risk index over historical time series Conduct a comprehensive platform safety assessment and remote visualization, and construct load heat maps, fatigue risk distribution maps, and comprehensive analysis trend maps.

[0186] Specifically, this step aims to determine the real-time load state P(t) of the floating platform at time t based on step S4300, and the force sub-partition d of the floating platform obtained in step S5200. i Cumulative fatigue risk index over K consecutive historical time series Establish an overall safety status assessment function for the water platform, generate a unified risk level output, and present it graphically and provide intelligent early warnings through a remote visual management platform, providing operation and maintenance personnel with intuitive and comprehensive decision support.

[0187] Furthermore, such as Figure 7 As shown, step S6000 includes:

[0188] Step S6100, based on the real-time load state vector P(t) and the force sub-partition d iCumulative fatigue risk index over K consecutive historical time series Construct the overall safety state function of the water platform at time t.

[0189] In the specific implementation process, the real-time load state P(t) of the water platform at time t, constructed based on step S4300, and the force sub-partition d of the water platform obtained in step S5200 are compared. i The cumulative fatigue risk index over K consecutive time periods The comprehensive safety state function of the floating platform at any time t is constructed, and the specific formula is as follows:

[0190] S platform (t)=f(P'(t),FRI)

[0191] Among them, S platform P'(t) represents the overall safety state function of the surface platform at time t. It can be in continuous numerical form (e.g., risk score 0-1) or discrete levels (e.g., "low risk", "medium risk", "high risk") to reflect the comprehensive health level of the surface platform at the current time point, supporting operation and maintenance decisions and early warning responses. f(·) represents the multivariate fusion function, which is used to weight and combine the real-time load state vector P'(t) with the historical fatigue index vector FRI to generate a unified structural state assessment output. P'(t) = P(t) T The transpose of the vector representing the real-time load state of the platform at time t is an n-dimensional Boolean vector used to reflect the actual load currently borne by each load sub-partition. It is an important input variable for real-time detection of structural status. The force sub-regions d of the floating platform are represented by this symbol. i The transpose of the cumulative fatigue risk index over K consecutive time periods is used to reflect the potential fatigue failure trend under historical loads.

[0192] Step S6200, based on the real-time load state vector P(t) and the force sub-partition d i The cumulative fatigue risk index over K consecutive time periods and the overall safety state function S at time t platform (t) is used for remote visualization design, and load heat map, fatigue risk distribution map and comprehensive analysis trend map of the water platform are constructed respectively.

[0193] Specifically, this step aims to design and deploy a remote visualization interface system to dynamically and interactively represent and analyze multiple core indicators involved in the health assessment of the water platform, enabling real-time monitoring, trend identification, and decision support.

[0194] In the specific implementation process, this step maps the calculated security assessment results into graphical elements. These elements are then displayed in real-time and intuitively in a visualization interface, primarily including the following three core functions:

[0195] Firstly, the load heatmap is displayed. This load heatmap reflects the real-time load status of each stress sub-zone of the platform, aiding in the identification of areas of concentrated stress. The stress sub-zone d in the P(t) vector constructed in step S4300 is... i Real-time load status value Based on this, a load heat map of the floating platform is constructed.

[0196] If a discrete color level rendering strategy is adopted, Denotes the i-th force-bearing sub-partition d i If it is in an overloaded state, it will be highlighted in red on the diagram; if Denotes the i-th force-bearing sub-partition d i Under normal operating conditions, it is marked in gray in the diagram. It supports zooming, partition positioning, and time-series playback functions, dynamically reviewing the load state evolution process, making it easy for users to intuitively grasp the dynamic force distribution of the platform.

[0197] For example, if, under specific wind and wave conditions or when multiple people are working, the load on multiple support sections on one side of a floating platform frequently exceeds the limit, the system will display that area as a consistently highlighted red area on the heat map. Maintenance personnel can then use this information to identify areas with frequent high stress concentrations, indicating that the area has become a hidden fatigue zone and requires structural reinforcement or personnel relocation, thereby effectively mitigating the potential risks of long-term cumulative damage or overall stability degradation of the horizontal platform.

[0198] Secondly, the Fatigue Risk Map. This map displays the distribution of fatigue risk levels in each stress-bearing sub-section of the platform over long-term accumulation, supporting structural strengthening and maintenance decisions. The stress-bearing sub-section d of the platform obtained in step S5200... i The cumulative fatigue risk index over K consecutive time periods Based on this, it is mapped onto the topology of the floating platform to generate a fatigue risk distribution map of the floating platform.

[0199] The fatigue risk distribution map uses a four-level risk level color scheme. If the cumulative fatigue risk index... If the stress level is less than the fatigue risk level threshold θ1, then the stress sub-region d i Low risk is indicated by green in the image;

[0200] If the cumulative fatigue risk index If the stress level is greater than or equal to the fatigue risk level threshold θ1 and less than the fatigue risk level threshold θ2, then the stress sub-region d i Medium risk is indicated by yellow in the image;

[0201] If the cumulative fatigue risk index If the stress sub-region d is greater than or equal to the fatigue risk level threshold θ2 and less than the fatigue risk level threshold θ3, then... i High-risk areas are marked in orange on the map;

[0202] If the cumulative fatigue risk index If the stress level is greater than or equal to the fatigue risk level threshold θ3, then the stress sub-region d i This is an extremely high risk, indicating that it should be inspected immediately or used under limited load. It is marked in red in the picture.

[0203] For example, taking a floating platform as an example, over several consecutive operating days, the southern area of ​​the platform serves as a high-frequency use area for activities such as slides and diving. The cumulative fatigue risk index of its supporting structure over these multiple time periods is significant. The risk level has remained at the orange level (high risk) for an extended period, with the trend value showing an upward trend. Based on this, maintenance personnel can arrange for reinforcement of the supports in this area or restrict high-intensity activities to prevent structural fractures or localized platform subsidence caused by prolonged repeated loads, thereby effectively reducing the probability of safety accidents.

[0204] Third, a comprehensive trend analysis chart. This chart displays the dynamic changes in the overall safety status of the platform over time, aiding in the early identification of potential risk abrupt changes and maintenance intervention opportunities. The overall safety status function S of the platform at time t, obtained in step S6100, is... platform Based on (t), a comprehensive analysis trend chart of the water platform is constructed to change over time, that is, the horizontal axis is the time series and the vertical axis is the comprehensive risk index, presenting a dynamic curve.

[0205] First, the load peak point is marked (orange). If at a certain time t, there are multiple stress sub-regions d i Simultaneously satisfy The point on the curve will be highlighted with an orange dot to indicate a short-term concentrated load event.

[0206] Next, the system automatically records alarms (in red). If at a certain time t, the overall safety state function S... platform (t) Exceeds the set alarm threshold S thresh If so, that time point will be highlighted with a red dot on the curve to remind maintenance personnel to pay close attention;

[0207] Finally, the system records the time points for manual maintenance (in blue). If there are manual maintenance or reinforcement operations in the history of the floating platform, the system can import the relevant data and highlight the corresponding time points with blue dots, which facilitates the tracking and evaluation of the maintenance results.

[0208] For example, taking a floating water recreation platform as an example, around the 120th day of operation, the system detected high-frequency load peaks in the central main load-bearing area of ​​the platform at multiple times, and the comprehensive analysis curve showed a cluster of red dots at several consecutive points (system alarm). Subsequently, on the 125th day, the platform management arranged for the reinforcement of the pontoons and the replacement of the supports in this area, and the system simultaneously recorded blue dots to mark the maintenance actions. After the maintenance was completed, the trend chart showed a significant decrease in the comprehensive risk index, and the fluctuation range of the curve tended to stabilize, indicating that the maintenance measures achieved good results and effectively alleviated structural risks.

[0209] Example 2:

[0210] This embodiment, based on Embodiment 1, provides a system for remote operation and maintenance of a floating platform based on time-series loads, such as... Figure 8 As shown, it includes:

[0211] Force perception modeling module: Based on structural features and working condition simulation, the force-bearing area of ​​the entire pontoon module is divided, and a set of overall sensor layout schemes and sensor electrical characteristics and force mapping relationship models are constructed.

[0212] Sensor load signal acquisition and preprocessing module: Used to acquire load signals from the overall sensor array layout scheme set S in real time, and output a filtered set of sensor load signals.

[0213] Space payload estimation and matrix construction module: based on the filtered sensor payload signal set The total load of the floating platform area is estimated and the spatial load matrix is ​​constructed, and the spatial load matrix L(t) is constructed.

[0214] Structural bearing capacity assessment and overload early warning module: Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, it assesses the load-bearing capacity of each stress sub-zone d of the floating platform. i Overload warning threshold setting and intelligent warning discrimination are performed to obtain the real-time load state vector P(t);

[0215] Load-time fatigue risk prediction module: Based on the spatial load matrix L(t), load-time evolution analysis and cumulative fatigue risk prediction are performed, and fatigue risk levels are classified.

[0216] Platform overall safety assessment and remote visualization module: based on real-time load state vector P(t) and each force sub-partition d iCumulative fatigue risk index over historical time series Conduct a comprehensive platform safety assessment and remote visualization, and construct load heat maps, fatigue risk distribution maps, and comprehensive analysis trend maps.

[0217] The parts of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0218] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for remote operation and maintenance of a surface platform based on time-series loads, characterized in that, include: Based on structural features and working condition simulation, the stress area of ​​the entire pontoon module is divided, and a set of overall sensor layout schemes and sensor electrical characteristics and stress mapping relationship models are constructed. The system collects load signals from the entire sensor array and layout scheme in real time, and outputs a filtered set of sensor load signals. Based on the filtered sensor load signal set, the total load of the floating platform area is estimated and the spatial load matrix is ​​constructed. Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, overload warning thresholds are set and intelligent warnings are made for each stress sub-zone of the floating platform to obtain the real-time load state vector. Based on the load time-series evolution analysis and cumulative fatigue risk prediction of the spatial load matrix, fatigue risk levels are classified. Based on the real-time load state vector and the cumulative fatigue risk index of each stress sub-region over historical time, the platform's overall safety assessment and remote visualization are performed, and load heat map, fatigue risk distribution map and comprehensive analysis trend map are constructed.

2. The method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The method for constructing the overall sensor layout scheme set and the sensor electrical characteristics and force mapping relationship model includes: Based on structural features and working condition simulation, the pontoon module is divided into multiple stress sub-regions to construct a set of stress regions for the pontoon module; Based on the set of stress-bearing regions, design sensor array layout schemes and construct a set of overall sensor display layout schemes; Based on a set of overall sensor layout schemes, a model of sensor electrical characteristics and force mapping relationship is constructed. The sensor array deployment scheme includes, based on the set of stress-bearing regions, sequentially adjusting the deployment density driven by the region area, deploying according to stress intensity, defining the sensor array set, and controlling the spacing constraints for each stress-bearing sub-region.

3. The method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The method for outputting the filtered sensor load signal set includes: Based on the overall sensor array layout scheme, multi-channel sensor signals are acquired in real time in parallel to obtain the sensor voltage signal set. Based on the sensor electrical characteristics and force mapping model, the sensor voltage signal set is linearly transformed to obtain the sensor load signal set. Low-pass filtering and noise reduction are performed on the sensor load signal set to output the filtered sensor load signal set.

4. The method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The method for constructing the space load matrix includes: Based on the filtered set of sensor load signals, the estimated load values ​​in each stress sub-region are weighted and superimposed to calculate the total load value of each stress sub-region at the current time. Construct a spatial load matrix based on the total load value of each stress sub-region at the current moment; The design of the weighting coefficients in the weighted superposition of the load estimates in each stress sub-region includes three factors: sensor spatial layout factors, load representativeness factors, and the coupling relationship between the region area and sensor density.

5. A method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The steps for obtaining the real-time load state vector include: By integrating structural design parameters, material mechanical performance indicators, and finite element simulation results, the theoretical maximum load value of each stress sub-zone is obtained, and the maximum total load value of the entire floating platform structure is obtained by summing them up. Based on the theoretical maximum load value of each stress sub-zone in the maximum total load value of the entire floating platform structure, set the corresponding overload warning threshold. Based on the total load value of each stress sub-zone and the overload warning threshold, the overload risk is intelligently judged, and a Boolean discriminant function is constructed for real-time identification of overload areas of the water platform to generate a real-time load state vector.

6. A method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The steps of load time-series evolution analysis and cumulative fatigue risk prediction based on the spatial load matrix include: Based on the spatial load matrix, the load time series data at continuous moments are extracted by serialization processing along the time axis. Based on load time series data at continuous time points, a fatigue sensitivity model is constructed to calculate the cumulative fatigue risk index of each stress sub-region over continuous historical time series.

7. A method for remote operation and maintenance of a surface platform based on time-series loads according to claim 6, characterized in that, The steps for classifying fatigue risk levels include: Based on the cumulative fatigue risk index of each stress sub-region over a continuous historical time series, three different fatigue risk level thresholds θ1, θ2 and θ3 are set, and the fatigue risk level of each stress sub-region is classified according to the rules. The rule-based judgment includes: if the cumulative fatigue risk index is less than the fatigue risk level threshold θ1, the stress sub-region is low risk; if the cumulative fatigue risk index is greater than or equal to the fatigue risk level threshold θ1 and less than the fatigue risk level threshold θ2, the stress sub-region is medium risk; if the cumulative fatigue risk index is greater than or equal to the fatigue risk level threshold θ2 and less than the fatigue risk level threshold θ3, the stress sub-region is high risk; if the cumulative fatigue risk index is greater than or equal to the fatigue risk level threshold θ3, the stress sub-region is extremely high risk.

8. A method for remote operation and maintenance of a surface platform based on time-series loads according to claim 7, characterized in that, The steps for conducting the overall platform security assessment and remote visualization include: Based on the real-time load state vector and the cumulative fatigue risk index of each stress sub-partition in continuous historical time series, the overall safety state function of the water platform is constructed. Based on the real-time load state vector, the cumulative fatigue risk index of each stress sub-region over continuous historical time series, and the overall safety state function, remote visualization design is carried out to construct load heat map, fatigue risk distribution map, and comprehensive analysis trend map respectively.

9. A method for remote operation and maintenance of a surface platform based on time-series loads according to claim 1, characterized in that, The steps for constructing the load heat map, fatigue risk distribution map, and comprehensive analysis trend map include: Based on the real-time load state values ​​of each force sub-partition in the real-time load state vector, a discrete color-level rendering strategy is adopted to construct a load heat map of the water platform. Based on the cumulative fatigue risk index of each stress sub-zone of the water platform over a continuous historical time series, a fatigue risk distribution map of the water platform is constructed using a four-level risk level color scheme. Based on the overall safety state function of the water platform, with time series as the horizontal axis and comprehensive risk index as the vertical axis, load peak points are marked, automatic system alarms are recorded, and manual maintenance time points are recorded in sequence to construct a comprehensive analysis trend chart of the water platform changing over time.

10. A system for remote operation and maintenance of a surface platform based on time-series loads, used to implement a method for remote operation and maintenance of a surface platform based on time-series loads according to any one of claims 1-9, characterized in that... The system includes: Force perception modeling module: Based on structural features and working condition simulation, the force-bearing area of ​​the entire pontoon module is divided, and a set of overall sensor layout schemes and sensor electrical characteristics and force mapping relationship models are constructed. Sensor load signal acquisition and preprocessing module: used to acquire load signals of the entire sensor array layout scheme in real time and output a filtered set of sensor load signals; Spatial load estimation and matrix construction module: Based on the filtered sensor load signal set, the module estimates the total load of the floating platform area and constructs the spatial load matrix. Structural bearing capacity assessment and overload early warning module: Based on the theoretical bearing capacity and real-time load distribution information of the floating platform structure design, the module sets overload early warning thresholds and makes intelligent early warning judgments for each stress sub-zone of the floating platform to obtain the real-time load state vector. Load-time fatigue risk prediction module: Based on the spatial load matrix, load-time evolution analysis and cumulative fatigue risk prediction are performed, and fatigue risk levels are classified. Platform overall safety assessment and remote visualization module: Based on the real-time load state vector and the cumulative fatigue risk index of each stress sub-partition in the historical time series, the module performs an overall platform safety assessment and remote visualization, and constructs a load heat map, fatigue risk distribution map and comprehensive analysis trend map.

Images