Wind power construction safety risk assessment method based on multi-parameter fusion

Abstract

The present application provides a wind power construction safety risk assessment method based on multi-parameter fusion, which relates to the field of safety risk assessment technology, including: obtaining a wind power layout plan, performing sensor array layout, determining component stacking areas, calling continuous meteorological monitoring data, determining a wind turbine construction sequence, performing serialized weight fusion analysis on K settlement monitoring sensor clusters, obtaining K sequence settlement deviation coefficients, calling meteorological monitoring data sequences, performing risk assessment, and obtaining target construction safety risk assessment results. This application can solve the technical problem of inaccurate wind power construction safety risk assessment in the prior art due to the failure to fully consider various factors affecting safety. By monitoring the geological settlement safety risk in the region and combining the environmental factors and personnel safety factors in the hoisting process for evaluation, the construction safety risk is obtained, thereby improving the accuracy and comprehensiveness of the wind farm construction safety risk assessment.

Description

Technical Field

[0001] The present application relates to the technical field of safety risk assessment, and in particular to a wind power construction safety risk assessment method based on multi-parameter fusion. Background Art

[0002] With the increasing global demand for renewable energy, the development and utilization of wind energy, a clean and renewable energy source, has garnered widespread attention. The construction and operation of wind farms are critical to wind energy utilization, and ensuring safe construction is fundamental to ensuring efficient and stable operation. During wind farm construction, the hoisting of wind turbines is a core step that is technically complex and carries significant safety risks. In recent years, with the continued expansion and rapid growth of new energy services, the safety management of wind turbine hoisting operations has faced unprecedented challenges. While existing safety management methods can ensure the smooth progress of hoisting operations to a certain extent, they suffer from significant shortcomings in risk identification, management methods, and risk response capabilities. The frequent occurrence of workplace safety accidents highlights deficiencies in hoisting safety management, severely impacting the construction and operation of wind farms. Currently, existing assessment methods fail to fully reflect the settlement risk of each wind turbine during construction, limiting the comprehensiveness and accuracy of risk management decisions.

[0003] In summary, the existing technology has a technical problem of inaccurate wind power construction safety risk assessment due to failure to fully consider various factors affecting safety. Summary of the Invention

[0004] The purpose of this application is to provide a wind power construction safety risk assessment method based on multi-parameter fusion to solve the technical problem in the existing technology that the wind power construction safety risk assessment is inaccurate due to the failure to fully consider various factors affecting safety.

[0005] In view of the above problems, the present application provides a wind power construction safety risk assessment method based on multi-parameter fusion, wherein the wind power construction safety risk assessment method based on multi-parameter fusion includes: obtaining a wind power layout plan of K wind turbines in a target wind farm, wherein the wind power layout plan includes K wind turbine positions and K wind turbine specifications; arranging a sensor array of the target wind farm according to the K wind turbine positions and the K wind turbine specifications to obtain a settlement monitoring sensor array, wherein the settlement monitoring sensor includes M settlement monitoring sensors and M sensor positions; determining K component stacking areas of the K wind turbines, combining the M sensor positions and the K wind turbines Sensors are divided at different positions to obtain K settlement monitoring sensor clusters, wherein each settlement monitoring sensor cluster corresponds to a wind turbine; meteorological continuous monitoring data is called to obtain a meteorological monitoring data sequence; according to the wind power construction plan of the target wind farm, the wind turbine construction sequence is determined, and according to the wind turbine construction sequence, a serialized weight fusion analysis is performed on the K settlement monitoring sensor clusters to obtain K sequence settlement deviation coefficients; based on the K sequence settlement deviation coefficients, the meteorological monitoring data sequence is called to obtain K simultaneous meteorological monitoring data sets; risk assessment is performed on the K sequence settlement deviation coefficients and the K simultaneous meteorological monitoring data sets to obtain a target construction safety risk assessment result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] By obtaining a wind power layout plan of K wind turbines of a target wind farm, wherein the wind power layout plan includes K wind turbine positions and K wind turbine specifications; laying out a sensor array of the target wind farm according to the K wind turbine positions and the K wind turbine specifications, and obtaining a settlement monitoring sensor array, wherein the settlement monitoring sensor includes M settlement monitoring sensors and M sensor positions; determining K component stacking areas of the K wind turbines, dividing the sensors in combination with the M sensor positions and the K wind turbine positions, and obtaining K settlement monitoring sensor clusters, wherein each settlement A monitoring sensor cluster corresponds to a wind turbine; continuous meteorological monitoring data is called to obtain a meteorological monitoring data sequence; the wind turbine construction sequence is determined based on the wind turbine construction plan of the target wind farm, and a serialized weight fusion analysis is performed on the K settlement monitoring sensor clusters based on the wind turbine construction sequence to obtain K sequence settlement deviation coefficients; based on the K sequence settlement deviation coefficients, the meteorological monitoring data sequence is called to obtain K simultaneous meteorological monitoring data sets; a risk assessment is performed on the K sequence settlement deviation coefficients and the K simultaneous meteorological monitoring data sets to obtain the target construction safety risk assessment result. In other words, by monitoring the geological settlement safety risk of the region and combining it with the environmental factors and personnel safety factors during the hoisting process for evaluation, the construction safety risk is obtained, thereby improving the accuracy and comprehensiveness of the wind farm construction safety risk assessment.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0010] Figure 1 This is a flow chart of a wind power construction safety risk assessment method based on multi-parameter fusion in this application;

[0011] Figure 2This is a flow chart of obtaining the settlement deviation coefficient in the wind power construction safety risk assessment method based on multi-parameter fusion in this application. DETAILED DESCRIPTION

[0012] This application addresses the existing technical problem of inaccurate wind power construction safety risk assessments due to a failure to fully consider various factors affecting safety by providing a multi-parameter fusion-based wind power construction safety risk assessment method. By monitoring regional geological subsidence safety risks and evaluating them in conjunction with environmental factors and personnel safety factors during the hoisting process, the construction safety risk is determined, improving the accuracy and comprehensiveness of wind farm construction safety risk assessments.

[0013] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0014] For examples, please see the attached Figure 1 The present application provides a wind power construction safety risk assessment method based on multi-parameter fusion, wherein the wind power construction safety risk assessment method based on multi-parameter fusion specifically includes the following steps:

[0015] Step 1: Obtain a wind power layout plan for K wind turbines in a target wind farm, wherein the wind power layout plan includes K wind turbine positions and K wind turbine specifications.

[0016] Specifically, various information related to the target wind farm is collected, including topographic maps, wind direction and speed data, grid access point locations, and environmental restrictions. At the same time, the mutual influence between wind turbines, such as wake effects, is considered. The layout of the wind turbines in the target wind farm is determined, and an efficient and safe wind power deployment plan is determined. The specific location of each wind turbine in the wind farm is determined based on factors such as topography, wind direction, wind speed, and environmental restrictions. The specifications of the wind turbines, including the size, capacity, and other technical parameters of each wind turbine, are determined based on power generation requirements, the scale of the wind farm, local wind resources, and other technical and economic factors. Reasonable location and specification selection can maximize power generation efficiency while ensuring the safety and economy of construction and operation.

[0017] Step 2: Deploy a sensor array of the target wind farm according to the K wind turbine positions and the K wind turbine specifications to obtain a settlement monitoring sensor array, wherein the settlement monitoring sensor includes M settlement monitoring sensors and M sensor positions.

[0018] Specifically, the effective coverage area of each subsidence monitoring sensor is determined based on the sensor type, performance, and deployment environment. The effective coverage area refers to the range that a sensor can accurately monitor, and determining this range is crucial for subsequent sensor deployment. An initial sensor array deployment plan is generated based on the effective coverage area of each sensor and the location and specifications of each wind turbine. To improve subsidence monitoring coverage in the area surrounding the wind turbines and reduce the cost of deploying subsidence monitoring sensors, the initial deployment plan is optimized and adjusted, adjusting the number and location of sensors. The deployment plan is repeatedly adjusted and evaluated, ultimately resulting in a target deployment plan consisting of M subsidence monitoring sensors and M sensor locations. Sensors are deployed across the wind farm according to the target deployment plan, generating a subsidence monitoring sensor array. The subsidence monitoring sensor array is deployed across the target wind farm based on the location and specifications of the wind turbines. This optimized deployment plan improves monitoring coverage, enabling more effective monitoring of subsidence around the wind turbines.

[0019] Step 3: Determine K component stacking areas of the K wind turbines, divide the sensors based on the M sensor positions and the K wind turbine positions, and obtain K settlement monitoring sensor clusters, where each settlement monitoring sensor cluster corresponds to one wind turbine.

[0020] Specifically, the design drawings and specifications of the wind turbine are analyzed to determine the size, weight, and installation requirements of each component. The component stacking area is determined, taking into account the size and shape of the construction site, as well as the availability of transportation and lifting equipment. The safety of the stacking area is ensured, including protecting it from wind damage and ensuring a reasonable distance between the stacking area and the construction area. Sensors are divided based on the locations of the M sensors and the K wind turbines, with the M sensors allocated to the K wind turbines to ensure adequate monitoring coverage for each turbine. This division generates K settlement monitoring sensor clusters, one for each wind turbine. Each sensor cluster consists of a group of sensors positioned to effectively monitor the foundation settlement of the corresponding turbine, ensuring that each sensor cluster covers the monitoring area of the wind turbine. By rationally dividing the sensors, key areas of each turbine are effectively monitored. Each sensor cluster corresponds to a specific wind turbine, ensuring targeted and accurate monitoring.

[0021] Step 4: Call the continuous meteorological monitoring data to obtain the meteorological monitoring data sequence.

[0022] Specifically, data from meteorological stations or meteorological monitoring equipment near wind farms is obtained, including key meteorological parameters such as wind speed, wind direction, temperature, humidity, precipitation, and visibility. Meteorological conditions have a significant impact on the safety of wind power construction. For example, extreme weather such as strong winds and heavy rains may increase construction risks and affect construction progress and quality. Continuous monitoring of data means that data is collected continuously over a period of time, which helps capture the changing trends and periodic characteristics of meteorological conditions. These continuously monitored data series can be obtained by calling a meteorological monitoring system or service. The data series can be arranged in chronological order, with each data point corresponding to a specific time point, forming a continuous time series for subsequent risk assessment and analysis.

[0023] Step 5: Determine the wind turbine construction sequence according to the wind turbine construction plan of the target wind farm, and perform serialized weight fusion analysis on the K settlement monitoring sensor clusters according to the wind turbine construction sequence to obtain K sequence settlement deviation coefficients.

[0024] Specifically, the construction sequence of wind turbines is determined based on the wind turbine construction plan, taking into account various factors such as terrain, turbine type, construction resources, and safety requirements. Based on the construction sequence, a sequenced weighted fusion analysis is performed on K settlement monitoring sensor clusters. During the construction of each wind turbine, its associated settlement monitoring sensor cluster is taken into account. A weight is assigned to each wind turbine based on its position and importance in the construction sequence, and a safety risk assessment is performed during the construction of each wind turbine. The settlement deviation value of each settlement monitoring sensor cluster is multiplied by its corresponding weight to perform a sequenced weighted fusion analysis. This sequenced weighted fusion analysis generates a sequenced settlement deviation coefficient for each wind turbine. This sequenced weighted fusion analysis allows for a more accurate assessment of the settlement risk of each wind turbine during construction, enabling the implementation of appropriate risk control measures.

[0025] Step 6: Call the meteorological monitoring data sequence based on the K sequence sedimentation deviation coefficients to obtain K simultaneous meteorological monitoring data sets.

[0026] Specifically, according to the timestamps of the wind turbine construction sequence and the settlement monitoring data sequence, the corresponding meteorological monitoring data are matched to ensure that the settlement monitoring data in each construction sequence corresponds to the corresponding meteorological monitoring data in time. Based on the sequence settlement deviation coefficient, the meteorological monitoring data of the same sequence is called, which means that the meteorological monitoring data in the time period corresponding to the settlement deviation coefficient of each sequence will be called to form a set of meteorological monitoring data of the same sequence. K sets of meteorological monitoring data of the same sequence are obtained, each set contains a set of settlement deviation coefficients and a set of meteorological monitoring data, which are synchronized in time. By combining the sequence settlement deviation coefficient and the meteorological monitoring data of the same sequence, the safety risks in the construction process can be assessed more accurately to ensure the safe progress of the construction.

[0027] Step 7: Perform risk assessment on the K sequential settlement deviation coefficients and the K simultaneous meteorological monitoring data sets to obtain target construction safety risk assessment results.

[0028] Specifically, the sequential settlement deviation coefficient reflects the settlement trend of each wind turbine during construction, while the sequential meteorological monitoring data provides information on the meteorological conditions during construction. The combination of these two allows for a more comprehensive assessment of safety risks during construction. A risk assessment model is developed that integrates the sequential settlement deviation coefficient and meteorological monitoring data to assess construction risks. The model incorporates statistical analysis, machine learning, or deep learning techniques to identify potential safety risks. The integrated dataset is analyzed using the risk assessment model to generate a target construction safety risk assessment result, including indicators such as risk level, risk probability, risk impact, potential risk factors, and risk prediction. Based on the risk assessment results, appropriate risk management strategies are formulated, including adjustments to construction plans, strengthening safety measures, and preparing emergency response plans. Through risk assessment, safety risks during construction are identified and quantified, and preventive measures are implemented to avoid potential risk events and protect the safety of people and property.

[0029] Furthermore, step 2 of this application includes:

[0030] Acquire the effective coverage area of the settlement monitoring sensors; determine the initial layout plan of the sensor array according to the effective coverage area, the positions of the K wind turbines, and the specifications of the K wind turbines, and generate an initial layout plan; with improving the settlement monitoring coverage rate of the area around the wind turbines and reducing the layout cost of the settlement monitoring sensors as the optimization goals, optimize the direction of the initial layout plan and generate a target layout plan, wherein the target layout plan includes M settlement monitoring sensors and M sensor positions; and layout the sensor array based on the target layout plan to generate the settlement monitoring sensor array.

[0031] Specifically, different types of settlement monitoring sensors have different detection principles and technical parameters. The range determines the maximum distance or change a sensor can monitor, while accuracy affects the reliability of the monitoring data. Based on the sensor's technical parameters, a theoretical maximum coverage area is calculated. In practice, the monitoring environment and requirements may result in the actual effective coverage area being smaller than the theoretical coverage area. Topography and landforms influence sensor deployment and coverage. For example, in complex terrain, more sensors are required to ensure comprehensive coverage. Furthermore, climatic conditions such as temperature, humidity, and rain and snow can affect sensor performance and data accuracy. Monitoring frequency and duration are determined based on different monitoring requirements, such as monitoring wind turbine settlement or the overall stability of a wind farm. For example, in areas with complex terrain or poor climatic conditions, a denser deployment is required to ensure effective coverage. The effective coverage area is ultimately determined based on the sensor's monitoring range and site conditions.

[0032] Based on the geographic information of the wind farm, the location and specifications of the wind turbines, and the effective coverage area of the sensors, an appropriate model, such as a neural network model, is selected to initially determine the layout plan. The neural network model is trained using historical data, with input parameters including the location and specifications of the wind turbines and the effective coverage area of the sensors. The output is the initial layout location of the sensors. The training process is iterated, and the model's weights and biases are continuously adjusted to achieve a satisfactory performance level. The trained model is used to generate an initial layout plan for the sensor array. The initial layout plan should be based on layout principles, such as uniform layout and encryption in key areas, to ensure the accuracy and comprehensiveness of the monitoring data.

[0033] Set the optimization goal: improve the settlement monitoring coverage rate around the wind turbine and reduce the cost of deploying settlement monitoring sensors. Select an appropriate optimization algorithm, such as particle swarm optimization. Based on the optimization algorithm, randomly adjust the initial deployment plan, taking into account monitoring coverage and cost-effectiveness, and adjust the number and location of sensors. Generate multiple adjusted deployment plans, calculate the fitness of each deployment plan, compare them, and optimize the plans. Ultimately, determine the target deployment plan, which should have the best fitness. The target deployment plan includes M settlement monitoring sensors and M sensor locations. Based on the target deployment plan, determine the actual sensor deployment locations. Consider site conditions, such as topography, soil type, and surrounding environment, to ensure the rationality and feasibility of the sensor deployment. Install M settlement monitoring sensors at appropriate locations within the wind farm, ensuring they are securely fixed and protected from external interference. After installation, test the settlement monitoring system to ensure that all sensors are functioning properly and that data transmission is accurate. After sensor deployment and system testing are complete, generate a settlement monitoring sensor array to monitor settlement of the wind turbine foundation and surrounding ground in real time or periodically. By rationally planning the number and location of sensors, unnecessary sensor deployment can be reduced, lowering the construction and maintenance costs of the settlement monitoring system. The settlement monitoring sensor array generated based on the target deployment plan can monitor the settlement of wind turbine foundations and surrounding ground in real time or periodically, providing data support for wind farm construction safety risk assessments.

[0034] Furthermore, the present application further comprises the following steps:

[0035] The initial layout plan is retrieved using the number of sensors and the position of sensors as indexes to generate the initial number of sensors and the initial sensor positions; the layout fitness analysis of the initial layout plan is performed based on the initial number of sensors and the initial sensor positions to obtain the initial layout fitness; a first adjustment scale is obtained based on the initial layout fitness, wherein the first adjustment scale is the size of a single adjustment to the number of initial sensors and the displacement amplitude of a single movement of the initial sensor positions when performing directional optimization adjustment; the initial layout plan is randomly adjusted multiple times based on the first adjustment scale to generate multiple adjusted layout plans; the multiple adjusted layout plans are directional optimized and adjusted to generate the target layout plan.

[0036] Specifically, the two key parameters, the number of sensors and the sensor positions, are used to quickly locate and identify the initial deployment plan, determine the initial number of sensors used in this deployment plan, and the positions of these initial sensors. A fitness analysis unit is constructed, and the initial number of sensors, initial sensor positions, and effective coverage area are input into the fitness analysis unit for evaluation. Based on the evaluation results, the initial deployment fitness is obtained. Based on the initial deployment fitness, the degree of optimization required is determined. A first adjustment scale is set, including the size of a single adjustment of the number of sensors and the displacement amplitude of the sensor position. Based on the first adjustment scale, the initial deployment plan is randomly adjusted to increase or decrease the number of sensors, as well as to change the position of the sensors. After each random adjustment, a new adjusted deployment plan is generated.

[0037] Use the particle optimization algorithm to optimize the adjustment layout scheme, use the fitness analysis unit to perform fitness analysis on multiple adjustment layout schemes, calculate the fitness value of each particle, that is, the objective function value corresponding to the particle position, and obtain multiple adjustment layout fitnesses. Compare the fitness of these adjustment layout schemes to find the adjustment layout scheme with the largest fitness. Compare the maximum fitness with the initial layout fitness value. If the maximum value among the multiple adjustment layout fitnesses is greater than or equal to the initial layout fitness, then the adjustment layout scheme corresponding to this maximum value is used as the optimization direction. Otherwise, two random numbers are randomly generated. If the first random number is greater than or equal to the second random number, the initial layout scheme is used as the optimization direction for optimization adjustment. If the first random number is less than the second random number, the adjustment layout scheme corresponding to the maximum value is used as the optimization direction, and other schemes are optimized and adjusted.

[0038] Each solution is evaluated, and the number and location of sensors are adjusted based on the results. The goal of optimization is to improve monitoring coverage and reduce costs while ensuring layout rationality. Through multiple iterations and optimization adjustments, the fitness score of the deployment plan can be gradually improved, resulting in a higher-quality deployment plan. The Particle Optimization Algorithm (PSO) is an optimization algorithm based on swarm intelligence. Its basic concept is to find the optimal solution to a problem through the interactions of individuals in a swarm of particles. In the PSO, each particle represents a candidate solution to the problem and has a corresponding objective function value, namely, a fitness value. Each particle in the swarm updates its position by tracking two "extremes": the optimal solution found by the particle to date, called the individual extreme value; the optimal solution found by the entire swarm to date, called the global extreme value. The introduction of randomness and the PSO algorithm prevents the optimization process from falling into local optimal solutions and ensures the possibility of finding the global optimal solution. By optimizing the deployment plan, monitoring coverage can be maximized while reducing the number of unnecessary sensors, thereby improving monitoring efficiency.

[0039] Furthermore, the present application further comprises the following steps:

[0040] Constructing a fitness analysis unit; inputting the initial number of sensors, initial sensor positions and the effective coverage area into the fitness analysis unit to obtain the initial layout fitness.

[0041] Specifically, the metrics for evaluating deployment plans are clearly defined, and the specific form of the fitness function is determined, such as a linear combination, weighted sum, or other complex function. Mathematical modeling methods, such as linear programming, integer programming, or heuristic algorithms, are used to establish the evaluation model. Within the model, the evaluation metric serves as the objective function, while also considering constraints such as the effective coverage area of the sensors, the location and specifications of the wind turbines, and so on. Simulation software or algorithms are used to simulate the sensor array deployment plan and evaluate its performance, including simulating sensor data and analyzing monitoring coverage and cost-effectiveness. The fitness analysis unit is an integrated system used to analyze the sensor array deployment plan to determine its fitness. It typically includes an input module, a simulation and evaluation module, an optimization algorithm module, and an output module. The initial number of sensors, initial sensor locations, and effective coverage area are input into the fitness analysis unit for evaluation. Based on the evaluation results, the fitness of the initial deployment is determined. This fitness is a comprehensive metric used to measure the quality of the deployment plan. A plan with a higher fitness indicates that it better meets monitoring requirements and has better performance. Fitness analysis can determine the performance of the initial deployment plan and provide a basis for subsequent optimization and adjustment.

[0042] Furthermore, the present application further comprises the following steps:

[0043] Utilize the fitness analysis unit to perform fitness analysis on the multiple adjustment layout schemes to obtain multiple adjustment layout fitnesses; determine whether the maximum value among the multiple adjustment layout fitnesses is greater than or equal to the initial layout fitness; if so, use the adjustment layout scheme corresponding to the maximum value among the multiple adjustment layout fitnesses as the optimization direction, and perform directional optimization adjustment on the remaining multiple adjustment layout schemes according to the first adjustment scale until the preset optimization times are met to obtain the target layout scheme; if not, call a random number generator to generate a first random number and a second random number; when the first random number is greater than or equal to the second random number, perform directional optimization adjustment with the initial layout scheme as the optimization direction until the preset optimization times are met to obtain the target layout scheme.

[0044] Furthermore, the present application further comprises the following steps:

[0045] When the first random number is smaller than the second random number, the adjustment layout scheme corresponding to the maximum value among multiple adjustment layout fitnesses is accepted as the optimization direction, and the remaining multiple adjustment layout schemes are optimized and adjusted in direction according to the first adjustment scale until the preset optimization times are met to obtain the target layout scheme.

[0046] Specifically, the number of sensors, sensor locations, and effective coverage area for each adjusted layout are input into the fitness analysis unit. Based on this input data, the simulation and evaluation module simulates the sensor array layout and evaluates its performance. The evaluation model outputs the fitness of each layout, a comprehensive metric used to measure the quality of the layout. The multiple adjusted layout fitnesses are sorted, the maximum value is found, and this value is compared with the initial layout fitness. If the maximum value among the multiple adjusted layout fitnesses is greater than the initial layout fitness, it indicates that the adjusted layout corresponding to this maximum value is better than the initial layout, and this value is used as the optimization direction. Based on the first adjustment scale, the remaining adjusted layout plans are optimized, such as by increasing or decreasing the number of sensors or fine-tuning sensor locations. The preset optimization number is a pre-set limit on the number of iterations during the optimization process, ensuring that the optimization process does not continue indefinitely but is completed within a finite number of times. These steps are repeated for multiple iterations. In each iteration, the direction is optimized based on the first adjustment scale, and the fitness of each new plan is re-evaluated. Continue optimizing until the preset number of optimizations is met, and use the current adjusted layout plan as the target layout plan.

[0047] If the maximum value among multiple adjustment layout fitness is less than or equal to the initial layout fitness, it means that the adjustment layout schemes generated after these random adjustments are not better than the initial scheme. At this time, it is necessary to call the random number generator to generate two random numbers, namely the first random number and the second random number. The RAND function can be used to generate random numbers. If the first random number is greater than or equal to the second random number, the initial layout scheme is still used as the optimization direction for directional optimization adjustment. This process is iterated until the preset number of optimizations is met and the target layout scheme is obtained. On the contrary, if the first random number is less than the second random number, the adjustment layout scheme corresponding to the maximum fitness value is used as the optimization direction, and the remaining multiple adjustment layout schemes are optimized according to the first adjustment scale. Repeat this process until the preset number of optimizations is met and the target layout scheme is finally determined.

[0048] In a specific example, suppose a wind farm needs to deploy subsidence monitoring sensors. The initial deployment plan includes 10 sensors distributed at specific locations, with each sensor covering an effective area of 100 square meters. The fitness analysis unit calculates the initial deployment fitness of 0.75. This plan is then randomly adjusted, generating five modified deployment plans, each adjusting the number and location of sensors. Fitness analysis of these plans yields the following fitness scores: 0.80, 0.78, 0.76, 0.74, and 0.79. The maximum fitness value, 0.80, exceeds the initial deployment fitness, so this plan is selected for optimization. The remaining plans are adjusted according to the first adjustment scale, such as increasing or decreasing the number of sensors or relocating them. This process is repeated 10 times until the optimal deployment plan is found. However, in one iteration, the maximum fitness score does not exceed the initial deployment fitness. Therefore, a first random number is randomly generated, with the first random number being 0.5 and the second being 0.6. Because the first random number is smaller than the second, the layout adjustment with the highest fitness of 0.80 is accepted as the new optimization direction. Based on the first adjustment scale, the other layout adjustments are optimized. This process is repeated 10 times to determine the target layout.

[0049] In summary, the fitness analysis unit optimizes the adjusted deployment plans and determines which ones outperform the initial ones. Introducing randomness prevents the optimization process from falling into local optimal solutions and ensures the possibility of finding a global optimal solution. By continuously optimizing the deployment plans to adapt to different environmental conditions, we ultimately achieve an efficient and economical deployment plan, which helps improve the adaptability and robustness of the monitoring system.

[0050] Further, as attached Figure 2 As shown, step five of this application includes:

[0051] Extract the first wind turbine from the wind turbine construction sequence; with the first wind turbine as the center, perform weight allocation according to the distances from the K settlement monitoring sensor clusters to the first wind turbine to obtain K first weights; obtain a first construction window for the first wind turbine, and retrieve K settlement monitoring data sets of the K settlement monitoring sensor clusters based on the first construction window; perform difference calculations on the K settlement monitoring data sets and K initial monitoring data to generate K settlement deviation value sets; perform centralized value screening on the K settlement deviation value sets to generate K centralized settlement deviation values; perform serialized weight fusion analysis on the K centralized settlement deviation values using the K first weights to generate K first settlement deviation coefficients; and use the K first settlement deviation coefficients as the K sequence settlement deviation coefficients.

[0052] Specifically, the first wind turbine, typically the first installed turbine, is extracted from the determined wind turbine construction sequence. With the first wind turbine as the core, weights are assigned based on the distance between the settlement monitoring sensor cluster and the first wind turbine, reflecting the degree of impact of different settlement monitoring sensor clusters on the construction safety of the first wind turbine. Sensor clusters closer to the first wind turbine have a greater impact on the safety of the first wind turbine and should therefore be assigned higher weights. Based on the weight assignment rules, a first weight is calculated for each settlement monitoring sensor cluster. The first construction window for the first wind turbine is determined based on the construction plan, weather forecast, and other relevant factors. A construction window refers to a time period suitable for wind turbine construction, typically determined by factors such as weather conditions, construction resources, and safety requirements. The first construction window refers to the construction time window for the first wind turbine. Based on the determined first construction window, the corresponding settlement monitoring sensor cluster data set is retrieved, including all settlement monitoring data collected from each sensor cluster during the first construction window.

[0053] Initial monitoring data refers to settlement monitoring data collected before construction begins or at a reference time point. This data serves as a baseline for comparison with data collected during construction. For each settlement monitoring sensor cluster, a difference calculation is performed between the data set collected during the first construction window and the initial monitoring data. This difference calculation determines the change in settlement monitoring data over a period of time, known as the settlement deviation value. This difference calculation typically involves a subtraction operation: subtracting the initial data from the new data. Through this difference calculation, a set of settlement deviation values is generated for each settlement monitoring sensor cluster. These sets contain the settlement change for each sensor cluster during the first construction window relative to the initial state. A central value screening is performed on the K settlement deviation value sets. The mode is extracted from the multiple settlement deviation value sets, and the central value screening is performed using the mean shift algorithm. This screening removes outliers or noise, improving data reliability and accuracy. The mean shift algorithm generates a central settlement deviation value for each settlement deviation value set, representing the main trends and characteristics of the set. After screening, K central settlement deviation values are generated.

[0054] A serialized weighted fusion analysis is performed on K concentrated settlement deviation values using K first weights. The importance of different settlement monitoring points and their settlement deviation values are combined to generate a comprehensive settlement deviation coefficient. Each settlement deviation value is multiplied by its corresponding first weight for serialized weighted fusion analysis. This analysis method takes into account the importance of different monitoring points to generate a comprehensive settlement deviation coefficient. Through serialized weighted fusion analysis, a first settlement deviation coefficient is generated for each monitoring point, resulting in K first settlement deviation coefficients, reflecting the settlement risk level of each wind turbine relative to other turbines during construction. These K first settlement deviation coefficients are used as K serialized settlement deviation coefficients. By performing difference calculation, concentrated value screening, and serialized weighted fusion analysis on settlement monitoring data, the settlement of wind turbines during construction can be accurately assessed, settlement anomalies can be detected promptly, and appropriate measures can be taken to prevent potential safety risks and reduce construction risks.

[0055] Furthermore, the present application further comprises the following steps:

[0056] The modes are extracted from the K settlement deviation value sets respectively to obtain K settlement deviation value modes; taking the K settlement deviation value modes as the starting point, the mean shift algorithm is used to perform centralized value screening in the K settlement deviation value sets to generate the K centralized settlement deviation values.

[0057] Specifically, the mode (the value with the most occurrences) is extracted from a set of K settlement deviation values, resulting in the K modes of these values. Using this extracted mode as a starting point, the mean-shift algorithm is used to filter the set of settlement deviation values for a central value. During this process, the algorithm gradually adjusts the central point to reflect the main distribution characteristics of the data. The mean-shift algorithm is a nonparametric density estimation method that uses an iterative process to find patterns in the data distribution. The basic idea is to move along the ascending density gradient in feature space until reaching a local maximum of the density function. These maxima typically represent cluster centers in the dataset. A random point is selected as the central point, and the weighted average position of all points within a certain range around the current central point is calculated. This position is called the shifted mean. The weight is typically inversely proportional to the distance from the point to the central point, meaning that closer points have a greater influence on the shifted mean. Through the mean-shift algorithm, a central settlement deviation value is generated for each set of settlement deviation values. This value should best represent the main trends and characteristics of the set. The mean-shift algorithm can improve data reliability and accuracy, ensuring the quality of monitoring data.

[0058] In summary, the wind power construction safety risk assessment method based on multi-parameter fusion provided in this application has the following technical effects:

[0059] By obtaining a wind power layout plan of K wind turbines of a target wind farm, wherein the wind power layout plan includes K wind turbine positions and K wind turbine specifications; laying out a sensor array of the target wind farm according to the K wind turbine positions and the K wind turbine specifications, and obtaining a settlement monitoring sensor array, wherein the settlement monitoring sensor includes M settlement monitoring sensors and M sensor positions; determining K component stacking areas of the K wind turbines, dividing the sensors in combination with the M sensor positions and the K wind turbine positions, and obtaining K settlement monitoring sensor clusters, wherein each settlement A monitoring sensor cluster corresponds to a wind turbine; continuous meteorological monitoring data is called to obtain a meteorological monitoring data sequence; the wind turbine construction sequence is determined based on the wind turbine construction plan of the target wind farm, and a serialized weight fusion analysis is performed on the K settlement monitoring sensor clusters based on the wind turbine construction sequence to obtain K sequence settlement deviation coefficients; based on the K sequence settlement deviation coefficients, the meteorological monitoring data sequence is called to obtain K simultaneous meteorological monitoring data sets; a risk assessment is performed on the K sequence settlement deviation coefficients and the K simultaneous meteorological monitoring data sets to obtain the target construction safety risk assessment result. In other words, by monitoring the geological settlement safety risk of the region and combining it with the environmental factors and personnel safety factors during the hoisting process for evaluation, the construction safety risk is obtained, thereby improving the accuracy and comprehensiveness of the wind farm construction safety risk assessment.

[0060] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0061] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A wind power construction safety risk assessment method based on multi-parameter fusion is characterized by: include: Obtaining a wind power layout plan for K wind turbines of a target wind farm, wherein the wind power layout plan includes K wind turbine positions and K wind turbine specifications; Deploying a sensor array of the target wind farm according to the K wind turbine positions and the K wind turbine specifications to obtain a subsidence monitoring sensor array, wherein the subsidence monitoring sensor array includes M subsidence monitoring sensors and M sensor positions; Determine K component stacking areas of the K wind turbines, perform sensor division based on the M sensor positions and the K wind turbine positions, and obtain K settlement monitoring sensor clusters, wherein each settlement monitoring sensor cluster corresponds to one wind turbine; Call the meteorological continuous monitoring data to obtain the meteorological monitoring data sequence; Determine a wind turbine construction sequence according to the wind turbine construction plan of the target wind farm, and perform a sequenced weight fusion analysis on the K settlement monitoring sensor clusters according to the wind turbine construction sequence to obtain K sequence settlement deviation coefficients; Calling the meteorological monitoring data sequence based on the K sequence sedimentation deviation coefficients to obtain K simultaneous meteorological monitoring data sets; Performing risk assessment on the K sequential settlement deviation coefficients and the K simultaneous meteorological monitoring data sets to obtain a target construction safety risk assessment result; Extracting a first wind turbine generator set from the wind turbine generator set construction sequence; Taking the first wind turbine generator set as the center, weights are assigned according to the distances from the K settlement monitoring sensor clusters to the first wind turbine generator set to obtain K first weights; Acquire a first construction window of the first wind turbine generator set, and retrieve K settlement monitoring data sets of the K settlement monitoring sensor clusters based on the first construction window; Perform difference calculations on the K settlement monitoring data sets and the K initial monitoring data to generate K settlement deviation value sets; Performing concentrated value screening on the K settlement deviation value sets respectively to generate K concentrated settlement deviation values; Performing a serialized weight fusion analysis on the K concentrated settlement deviation values using the K first weights to generate K first settlement deviation coefficients; The K first sedimentation deviation coefficients are used as the K sequence sedimentation deviation coefficients.

2. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 1, characterized in that: include: Obtain the effective coverage area of the settlement monitoring sensor; Determine an initial layout plan of the sensor array according to the effective coverage area, the positions of the K wind turbines, and the specifications of the K wind turbines, and generate an initial layout plan; With the optimization goals of increasing the settlement monitoring coverage rate of the area around the wind turbine and reducing the cost of deploying settlement monitoring sensors, the initial deployment plan is directional-optimized and adjusted to generate a target deployment plan, wherein the target deployment plan includes M settlement monitoring sensors and M sensor positions; The sensor array is arranged based on the target arrangement scheme to generate the settlement monitoring sensor array.

3. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 2, characterized in that: include: Using the number of sensors and the sensor positions as indexes, searching the initial layout plan to generate an initial number of sensors and an initial sensor position; performing a layout fitness analysis of the initial layout scheme according to the initial number of sensors and the initial sensor positions to obtain an initial layout fitness; Obtaining a first adjustment scale according to the initial layout adaptability, wherein the first adjustment scale is the size of a single adjustment of the number of initial sensors and the displacement amplitude of a single movement of the initial sensor position when performing direction optimization adjustment; Performing multiple random adjustments on the initial layout plan based on the first adjustment scale to generate multiple adjusted layout plans; Direction optimization adjustment is performed on the multiple adjustment layout schemes to generate the target layout scheme.

4. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 3 is characterized in that: include: Construct a fitness analysis unit; The initial number of sensors, initial sensor positions, and the effective coverage area are input into the fitness analysis unit to obtain the initial layout fitness.

5. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 4 is characterized in that: include: Performing fitness analysis on the multiple adjustment layout schemes using the fitness analysis unit to obtain multiple adjustment layout fitnesses; Determine whether the maximum value among the multiple adjusted layout fitnesses is greater than or equal to the initial layout fitness; if so, use the adjusted layout scheme corresponding to the maximum value among the multiple adjusted layout fitnesses as the optimization direction, and perform directional optimization adjustment on the remaining multiple adjusted layout schemes according to the first adjustment scale until a preset number of optimization times is met, thereby obtaining the target layout scheme; If not, call the random number generator to generate a first random number and a second random number. When the first random number is greater than or equal to the second random number, perform directional optimization adjustment with the initial layout plan as the optimization direction until the preset optimization times are met to obtain the target layout plan.

6. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 5, characterized in that: include: When the first random number is smaller than the second random number, the adjustment layout scheme corresponding to the maximum value among multiple adjustment layout fitnesses is accepted as the optimization direction, and the remaining multiple adjustment layout schemes are optimized and adjusted in direction according to the first adjustment scale until the preset optimization times are met to obtain the target layout scheme.

7. The wind power construction safety risk assessment method based on multi-parameter fusion according to claim 1, characterized in that: include: Extracting modes from the K settlement deviation value sets respectively to obtain K settlement deviation value modes; Taking the modes of the K settlement deviation values as a starting point, a mean shift algorithm is used to perform concentrated value screening in the K settlement deviation value set to generate the K concentrated settlement deviation values.

Images