A dynamic self-adaptive point distribution method and system for a photovoltaic power station microclimate observation device

Abstract

The application discloses a kind of photovoltaic power station microclimate observation device dynamic self-adapting point distribution method and system, belong to photovoltaic power station meteorological monitoring technical field, comprising: based on the topography and shelter information of photovoltaic power station, construct four-dimensional space-time microclimate field reconstruction model of fusion time attenuation weight;Based on four-dimensional space-time microclimate field reconstruction model, establish multi-objective fitness function;Using the swarm intelligence optimization algorithm of fusion quantum behavior mechanism and reverse learning strategy, the multi-objective fitness function is iteratively solved, so as to obtain the initial optimal measurement point distribution scheme;Based on the dynamic adjustment mechanism constructed by digital twin, the performance of the initial optimal measurement point distribution scheme is evaluated and adaptively adjusted according to the real-time collected observation data.The application comprehensively improves the accuracy of microclimate observation, the environmental adaptability and long-term operation stability of the system, and provides a solid and reliable technical foundation for the power generation efficiency optimization and intelligent operation and maintenance management of photovoltaic power station.

Description

Technical Field

[0001] This invention belongs to the field of meteorological monitoring technology for photovoltaic power plants, and particularly relates to a dynamic adaptive deployment method and system for micro-meteorological observation devices in photovoltaic power plants. Background Technology

[0002] The power generation efficiency of photovoltaic power plants is significantly affected by local micrometeorological conditions. Accurate micrometeorological observation is a key technical means to optimize power plant operation strategies and improve power generation efficiency. In terms of application scenarios, traditional methods for deploying micrometeorological observation devices mainly adopt uniform grid deployment or rely on manual experience for deployment, and often use classic optimization algorithms such as genetic algorithms and particle swarm optimization to solve the deployment scheme.

[0003] However, these existing technologies have the following prominent drawbacks: 1) The static point placement method is rigid. Existing technologies mostly use uniform grid placement or rely on manual experience for point placement, without fully considering complex environmental factors such as the terrain undulations of photovoltaic power stations, buildings, and photovoltaic module shading, resulting in the observation data not being able to fully reflect the spatial distribution differences of the micro-meteorological field; 2) The optimization objective is singular and limited, focusing only on the coverage of measurement points or the control of equipment costs, ignoring the spatiotemporal evolution characteristics of the micro-meteorological field and key features such as abrupt changes in irradiance gradient, resulting in insufficient observation accuracy; 3) The algorithm performance has shortcomings. Traditional genetic algorithms and particle swarm algorithms are prone to getting stuck in local optima when solving point placement optimization problems, and their convergence speed is slow and their computational efficiency is low, making it difficult to adapt to the point placement requirements of large-scale photovoltaic power stations; 4) There is a lack of dynamic adaptive capability. Once the point placement scheme is determined, it remains unchanged and cannot be dynamically adjusted according to factors such as seasonal changes, equipment aging, and extreme weather, resulting in a decline in observation performance and even data loss during long-term operation.

[0004] Therefore, there is an urgent need to propose a method for optimizing the deployment of micro-meteorological observation devices that combines high precision, strong adaptability, and high efficiency, in order to solve the problems existing in the current technology. Summary of the Invention

[0005] To fill the gaps in existing technologies regarding the lack of dynamic adaptive optimization for the placement of micrometeorological observation devices, the neglect of spatiotemporal evolution characteristics, and the susceptibility of algorithms to local optima, this invention provides a dynamic adaptive placement method and system for micrometeorological observation devices in photovoltaic power plants. Utilizing a four-dimensional spatiotemporal micrometeorological field reconstruction model, a multi-objective fitness function, and a digital twin real-time feedback mechanism, combined with a collaborative optimization strategy for fixed and mobile measuring points, this method achieves high precision in placement quantification, strong environmental adaptability, and high computational efficiency by utilizing only existing terrain, shading data, and conventional environmental parameter measurements from the photovoltaic power plant. This enables accurate coverage and dynamic optimization of micrometeorological observations, significantly reducing the investment in measuring points and improving field reconstruction accuracy, thus solving the problems of low precision, poor adaptability, and insufficient optimization efficiency in traditional placement methods.

[0006] This invention provides a dynamic adaptive deployment method for micro-meteorological observation devices in photovoltaic power plants, comprising the following steps: Based on the terrain and shading information of photovoltaic power plants, a four-dimensional spatiotemporal micro-meteorological field reconstruction model with time decay weights is constructed. Based on the aforementioned four-dimensional spatiotemporal micro-meteorological field reconstruction model, a multi-objective fitness function is established that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. A swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and reverse learning strategies is used to iteratively solve the multi-objective fitness function, thereby obtaining the initial optimal measurement point layout scheme. The dynamic adjustment mechanism based on digital twins evaluates and adaptively adjusts the initial optimal measurement point layout scheme based on real-time collected observation data. In the adaptively adjusted point layout scheme, a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the track system is introduced to realize the observation of the area to be observed.

[0007] Optionally, based on the terrain and shading information of the photovoltaic power station, a four-dimensional spatiotemporal micro-meteorological field reconstruction model with time decay weights is constructed. The specific process includes: A time decay weighting factor is constructed based on the decay time constant and the seasonal cycle; A four-dimensional spatiotemporal micro-meteorological field reconstruction model is constructed based on the time decay weighting factor, the trend term fitted by the spatiotemporal anisotropic kernel function, and the random residual term that follows a normal distribution.

[0008] Optionally, a multi-objective fitness function is established that integrates coverage energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. The specific process includes: The weighted distance is obtained based on the distance between the measuring point and the field point, as well as the weights determined by the terrain slope and shading factor. Based on the weighted distance, a coverage energy density sub-target is constructed to characterize spatial coverage capability; The observation value of the corresponding point is calculated based on the Laplace operator of the irradiance field at the measuring point. Based on the observed value, a gradient-sensitive sub-target is constructed for priority placement of points in areas with drastic changes in irradiance gradient. Based on the preset influence radius of the measuring point, a robust sub-target is constructed to ensure that the remaining measuring points can still maintain basic observation functions after any measuring point fails. The root mean square error is calculated based on historical observation data and field reconstruction results. Based on the root mean square error, an observation accuracy sub-objective is constructed to minimize the field reconstruction error; The correlation coefficient is calculated based on the correlation between the observation data of each measuring point. Based on the correlation coefficient, a redundancy sub-objective is constructed to reduce the degree of information overlap; Based on equipment procurement, installation, operation and maintenance, and energy consumption costs, construct cost sub-objectives to control total costs; The multi-objective fitness function is obtained by weighted summation of the coverage energy density sub-objective, gradient sensitivity sub-objective, robustness sub-objective, observation accuracy sub-objective, redundancy sub-objective, and cost sub-objective.

[0009] Optionally, a swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and reverse learning strategies is used for iterative solution. The specific process includes: A quantum behavior-based firefly swarm optimization algorithm is adopted, which updates the quantum position of fireflies based on the population centroid and quantum feature length, and determines the sign according to the fitness value during the iteration process; In each iteration, a backward learning operation is performed on the individuals in the current population to generate the corresponding backward solution; Compare the fitness values ​​of the reverse solution and the original solution. If the reverse solution is better, then replace the original solution with the reverse solution.

[0010] Optionally, a dynamic adjustment mechanism based on digital twins is used to evaluate and adaptively adjust the deployment scheme. The specific process includes: Based on real-time observation data, the Kriging interpolation error at each measuring point is calculated daily, and the observation error is evaluated in real time. Calculate the reduction in information entropy for each measurement point, and dynamically rank all measurement points based on the reduction in information entropy; When the decrease in information entropy at a certain measuring point is lower than a preset threshold, a command is triggered to move the measuring point to another location.

[0011] Optionally, the dynamic adjustment mechanism based on digital twins also includes: Switch between different preset deployment modes according to seasonal changes; When switching to summer mode, increase the density of measurement points between photovoltaic modules; When switching to winter mode, increase the density of measurement points in the shaded area on the north side of the photovoltaic array.

[0012] Optionally, a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the orbital system is introduced. The specific process includes: The ratio of fixed measuring points to mobile measuring points is set at three to two. Fixed measuring points are set up at key locations in the substation or inverter room; The mobile measuring point is installed on the low-altitude cable track system.

[0013] Optionally, it also includes planning periodic survey trajectories for the moving measurement points, the specific process of which includes: The trajectory planning problem of moving measurement points is modeled as a spatiotemporal traveling salesman problem with time window constraints; An improved ant colony algorithm is used to solve the spacetime traveling salesman problem to obtain an optimized survey trajectory; In trajectory planning, acceleration constraints, velocity constraints, and track offset compensation caused by wind loads are added for moving measurement points.

[0014] This invention also proposes a dynamic adaptive deployment system for micrometeorological observation devices in photovoltaic power plants, used to implement the method, including: The data acquisition module is used to collect micro-meteorological data within the photovoltaic power station. It includes fixed measuring point units deployed at key locations within the power station and mobile measuring point units installed on the track system. The central processing module, connected to the data acquisition module, is used to construct and update a four-dimensional spatiotemporal micro-meteorological field reconstruction model based on the micro-meteorological data and integrate time decay weights. It establishes and solves a multi-objective fitness function for the fusion coverage energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators to obtain the optimal measurement point layout scheme and movement trajectory. It also dynamically evaluates and adjusts the layout scheme based on the digital twin model. A motion control module, connected to the central processing module and the moving measuring point unit, is used to receive and execute the movement trajectory command and drive the moving measuring point unit to move along the track system. A data communication module connects the data acquisition module, the central processing module, and the motion control module, and is used to transmit data and control commands between the modules. An energy supply module is used to supply power to the data acquisition module, the central processing module, the motion control module, and the data communication module.

[0015] The present invention also proposes an electronic device, including a memory, a processor, and a computing program stored in the memory and executable on the processor, wherein the processor implements the method when executing the computing program.

[0016] Compared with the prior art, the present invention has the following advantages and technical effects: This invention overcomes the shortcomings of traditional static models in characterizing the spatiotemporal evolution of micrometeorological parameters by constructing a four-dimensional spatiotemporal micrometeorological field reconstruction model that integrates time decay weights. It achieves a leap from single-objective optimization to multi-dimensional collaborative trade-offs in site selection by establishing a multi-objective fitness function that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost. Furthermore, it significantly improves the efficiency and global convergence capability of optimal solution search in large-scale complex scenarios by introducing a swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and reverse learning strategies. Finally, it utilizes digital twin technology to construct a dynamic adjustment mechanism with real-time evaluation and feedback, enabling the site selection scheme to adaptively optimize according to environmental changes and operating conditions. Finally, through a collaborative observation network architecture combining fixed and orbital mobile measurement points, it achieves flexible, accurate, and persistent coverage of key areas throughout the photovoltaic power plant.

[0017] The systematic integration of this series of technologies fundamentally solves the problems of low accuracy, poor adaptability, and insufficient optimization efficiency in traditional point-of-sight methods, comprehensively improving the accuracy of micro-meteorological observation, the environmental adaptability of the system, and the long-term operational stability, providing a solid and reliable technical foundation for the optimization of photovoltaic power plant power generation efficiency and intelligent operation and maintenance management. Attached Figure Description

[0018] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram illustrating the four-dimensional spatiotemporal micro-meteorological field reconstruction principle of the dynamic adaptive deployment method and system for photovoltaic power station micro-meteorological observation devices in this embodiment of the invention. Figure 2 This is a flowchart of the quantum behavior firefly swarm optimization algorithm for the dynamic adaptive deployment method and system of the photovoltaic power station micrometeorological observation device in this embodiment of the invention. Figure 3 This is a structural diagram of the moving measuring point track system of the dynamic adaptive deployment method and system for the micro-meteorological observation device of a photovoltaic power station in an embodiment of the present invention; Figure 4 This is a framework diagram of the dynamic adaptive deployment method and dynamic adaptive adjustment mechanism of the photovoltaic power station micro-meteorological observation device in this embodiment of the invention. Figure 5 This is a schematic diagram of the actual photovoltaic power station layout scheme for the dynamic adaptive deployment method and system of micrometeorological observation devices in a photovoltaic power station, as described in this embodiment of the invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0021] Example 1 like Figure 1 As shown, this embodiment provides a dynamic adaptive deployment method for micrometeorological observation devices in photovoltaic power plants, including the following steps: Based on the terrain and shading information of photovoltaic power plants, a four-dimensional spatiotemporal micro-meteorological field reconstruction model with time decay weights is constructed. Based on the aforementioned four-dimensional spatiotemporal micro-meteorological field reconstruction model, a multi-objective fitness function is established that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. A swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and reverse learning strategies is used to iteratively solve the multi-objective fitness function, thereby obtaining the initial optimal measurement point layout scheme. The dynamic adjustment mechanism based on digital twins evaluates and adaptively adjusts the initial optimal measurement point layout scheme based on real-time collected observation data. In the adaptively adjusted point layout scheme, a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the track system is introduced to realize the observation of the area to be observed.

[0022] A feasible four-dimensional spatiotemporal micro-meteorological field reconstruction model, incorporating time decay weights, is constructed based on the terrain and shading information of photovoltaic power plants. The specific process includes: A time decay weighting factor is constructed based on the decay time constant and the seasonal cycle; A four-dimensional spatiotemporal micro-meteorological field reconstruction model is constructed based on the time decay weighting factor, the trend term fitted by the spatiotemporal anisotropic kernel function, and the random residual term that follows a normal distribution.

[0023] As a specific implementation method, the process of constructing a four-dimensional spatiotemporal micro-meteorological field reconstruction model includes: A four-dimensional spatiotemporal micrometeorological field reconstruction model based on Kriging interpolation is constructed, fully considering the spatiotemporal correlation of micrometeorological parameters. The expression is as follows: ; in, Representing spacetime coordinates The micrometeorological parameters at the location, such as irradiance, temperature, humidity, and wind speed; The trend term is represented by a spatiotemporal anisotropic kernel function fitting, which accurately describes the overall trend of micrometeorological parameters. This represents the random residual term, which follows a normal distribution and reflects local random fluctuations.

[0024] To accurately capture the temporal evolution characteristics of micrometeorological parameters, a time decay weighting factor is introduced, causing the contribution of historical observation data to the current prediction to decay exponentially over time. The formula is as follows: ; in, The decay time constant is determined based on the parameter characteristics, and the irradiance is... =24h, temperature =72h, The seasonal cycle is typically 365 days. This formula can capture both daily variations and seasonal fluctuations, improving the spatiotemporal prediction accuracy of the model.

[0025] It is feasible to establish a multi-objective fitness function that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. The specific process includes: The weighted distance is obtained based on the distance between the measuring point and the field point, as well as the weights determined by the terrain slope and shading factor. Based on the weighted distance, a coverage energy density sub-target is constructed to characterize spatial coverage capability; The observation value of the corresponding point is calculated based on the Laplace operator of the irradiance field at the measuring point. Based on the observed value, a gradient-sensitive sub-target is constructed for priority placement of points in areas with drastic changes in irradiance gradient. Based on the preset influence radius of the measuring point, a robust sub-target is constructed to ensure that the remaining measuring points can still maintain basic observation functions after any measuring point fails. The root mean square error is calculated based on historical observation data and field reconstruction results. Based on the root mean square error, an observation accuracy sub-objective is constructed to minimize the field reconstruction error; The correlation coefficient is calculated based on the correlation between the observation data of each measuring point. Based on the correlation coefficient, a redundancy sub-objective is constructed to reduce the degree of information overlap; Based on equipment procurement, installation, operation and maintenance, and energy consumption costs, construct cost sub-objectives to control total costs; The multi-objective fitness function is obtained by weighted summation of the coverage energy density sub-objective, gradient sensitivity sub-objective, robustness sub-objective, observation accuracy sub-objective, redundancy sub-objective, and cost sub-objective.

[0026] As a specific implementation method, the construction process of the multi-objective fitness function includes: A fitness function that comprehensively considers six dimensions is constructed to achieve multi-objective optimization balance, and its expression is: ; in, These are weighting coefficients, which can be flexibly adjusted according to the power plant scenario. The default setting satisfies... .

[0027] Definition of each key sub-objective: (1) Coverage energy density The ability of a measuring point to cover the microclimate is characterized by the following expression: ; in, As a venue to the measuring point The weighted distance is determined by terrain slope and shading factor. This is the distance attenuation coefficient.

[0028] (2) Observation accuracy The root mean square error (RMSE) is defined based on historical data and interpolation results. The smaller the RMSE, the better. The higher.

[0029] (3) Redundancy The correlation coefficient matrix of measurement point data is used to measure the degree of information overlap. The higher the correlation coefficient, the greater the redundancy. The lower.

[0030] (4) Cost Calculate the costs of equipment procurement, installation, operation and maintenance, and energy consumption. The lower the cost, the better. The higher.

[0031] (5) Robustness index To ensure the system continues to function even in the event of a single point of failure, the expression is: ; in, The radius of influence of the measuring point.

[0032] (6) Gradient sensitivity Prioritize the placement of sampling points in areas with severe irradiance gradients, as expressed in the following formula: ; in, Let Laplace operator be the irradiance field at the measurement point. The observation value function for this point can reduce the number of measurement points by 1 / 3 while maintaining the same accuracy.

[0033] A feasible approach is to use a swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and a reverse learning strategy for iterative solution. The specific process includes: A quantum behavior-based firefly swarm optimization algorithm is adopted, which updates the quantum position of fireflies based on the population centroid and quantum feature length, and determines the sign according to the fitness value during the iteration process; In each iteration, a backward learning operation is performed on the individuals in the current population to generate the corresponding backward solution; Compare the fitness values ​​of the reverse solution and the original solution. If the reverse solution is better, then replace the original solution with the reverse solution.

[0034] like Figure 2 As shown, as a specific implementation method, the process of solving the problem using the quantum behavior firefly swarm-back learning hybrid optimization algorithm includes: Integrating quantum behavior mechanisms with reverse learning strategies to improve algorithm performance: (1) Quantum position update: Quantum tunneling is achieved by introducing quantum superposition states, as shown in the formula: ; in, For population mass center, , The fluorescence intensity weight is used; For quantum characteristic length, It is a uniform random number in (0,1), and the sign of ± is determined by the fitness value.

[0035] (2) Reverse learning strategy: Perform reverse operation on the current optimal solution. Explore symmetric regions in the solution space to enhance global search capabilities.

[0036] (3) Adaptive Attraction: Considering terrain obstacles, the formula is: ; in, For path accessibility function, This is a function of obstacle height, enabling automatic obstacle avoidance.

[0037] A feasible, dynamic adjustment mechanism based on digital twins can be used to evaluate and adaptively adjust the deployment plan. The specific process includes: Based on real-time observation data, the Kriging interpolation error at each measuring point is calculated daily, and the observation error is evaluated in real time. Calculate the reduction in information entropy for each measurement point, and dynamically rank all measurement points based on the reduction in information entropy; When the decrease in information entropy at a certain measuring point is lower than a preset threshold, a command is triggered to move the measuring point to another location.

[0038] The complete framework of this dynamic adaptive adjustment mechanism is as follows: Figure 4 As shown, it includes modules such as data acquisition, error assessment, importance ranking, mode switching and instruction execution, forming a closed-loop feedback optimization system.

[0039] Furthermore, the dynamic adjustment mechanism built upon digital twins also includes: Switch between different preset deployment modes according to seasonal changes; When switching to summer mode, increase the density of measurement points between photovoltaic modules; When switching to winter mode, increase the density of measurement points in the shaded area on the north side of the photovoltaic array.

[0040] As a specific implementation method, the dynamic adaptive adjustment process includes: Building a real-time feedback system based on digital twin technology: (1) Real-time assessment of observation error: Daily calculation of Kriging interpolation error ;in, This represents the absolute interpolation error at that position, also known as the Kriging interpolation error. This represents the actual observed value at that location. This represents the Kriging prediction value for that location; (2) Ranking of measurement points by importance: Introducing the reduction in information entropy , < threshold for migration measurement points; where, This represents the information entropy of the system when the measurement point is not introduced. Indicates a known measuring point Conditional information entropy of the system under data conditions.

[0041] (3) Seasonal mode switching: The summer mode increases the density of measuring points between components to monitor the hot spot effect, while the winter mode increases the density of measuring points in the north shadow area to monitor snow cover.

[0042] It is feasible to introduce a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the orbital system. The specific process includes: The ratio of fixed measuring points to mobile measuring points is set at three to two. Fixed measuring points are set up at key locations in the substation or inverter room; The mobile measuring point is installed on the low-altitude cable track system.

[0043] The system structure is as follows Figure 3 As shown, it includes a low-altitude cable track, a moving measuring point unit, a drive mechanism, and a positioning module, which can realize the flexible movement and precise positioning of the measuring point above the photovoltaic array.

[0044] Furthermore, it also includes planning periodic survey trajectories for moving measurement points, the specific process of which includes: The trajectory planning problem of moving measurement points is modeled as a spatiotemporal traveling salesman problem with time window constraints; An improved ant colony algorithm is used to solve the spacetime traveling salesman problem to obtain an optimized survey trajectory; In trajectory planning, acceleration constraints, velocity constraints, and track offset compensation caused by wind loads are added for moving measurement points.

[0045] As a specific implementation method, the process of collaborative optimization of mobile measurement points includes: (1) System architecture: Fixed measuring points are arranged in key locations such as substations and inverter rooms, while mobile measuring points are installed on the low-altitude cable track system; (2) Trajectory optimization: The point placement is transformed into a time-space traveling salesman problem, which is solved by improving the ant colony algorithm to meet the time window constraint; (3) Dynamic constraints: The movement trajectory is solved by an improved ant colony algorithm, so that the moving measurement point periodically patrols the high-value area according to the optimized trajectory, while considering acceleration. ( Value ),speed ( Values And wind load track offset compensation.

[0046] On the other hand, such as Figure 2 As shown, this embodiment also proposes a dynamic adaptive deployment system for micrometeorological observation devices in photovoltaic power plants, used to implement the method described, including: The data acquisition module is used to collect micro-meteorological data within the photovoltaic power station. It includes fixed measuring point units deployed at key locations within the power station and mobile measuring point units installed on the track system. The central processing module, connected to the data acquisition module, is used to construct and update a four-dimensional spatiotemporal micro-meteorological field reconstruction model based on the micro-meteorological data and integrate time decay weights. It establishes and solves a multi-objective fitness function for the fusion coverage energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators to obtain the optimal measurement point layout scheme and movement trajectory. It also dynamically evaluates and adjusts the layout scheme based on the digital twin model. A motion control module, connected to the central processing module and the moving measuring point unit, is used to receive and execute the movement trajectory command and drive the moving measuring point unit to move along the track system. A data communication module connects the data acquisition module, the central processing module, and the motion control module, and is used to transmit data and control commands between the modules. An energy supply module is used to supply power to the data acquisition module, the central processing module, the motion control module, and the data communication module.

[0047] This embodiment provides a dynamic adaptive deployment method and system for micrometeorological observation devices in photovoltaic power plants. Through a comprehensive design that incorporates four-dimensional spatiotemporal micrometeorological field reconstruction, multi-objective optimization algorithms, dynamic adaptive adjustment, and collaborative optimization of moving measurement points, it achieves precise, efficient, and intelligent deployment of micrometeorological observation points.

[0048] On the other hand, this embodiment also proposes an electronic device, including a memory, a processor, and a computing program stored in the memory and executable on the processor, wherein the processor implements the method when executing the computing program.

[0049] Compared to traditional uniform grid placement, empirical placement, and single algorithm optimization schemes, this embodiment has significant technical advantages and application value: On the one hand, by using a gradient-sensitive placement strategy and a fixed-moving measurement point collaborative architecture, it significantly reduces equipment investment and maintenance costs; on the other hand, by integrating a hybrid optimization algorithm that combines quantum tunneling mechanism and reverse learning, and combining a four-dimensional spatiotemporal model with a time decay weight factor, it improves the accuracy of field reconstruction; at the same time, based on the dynamic adaptive adjustment mechanism and seasonal mode switching function of digital twin, it can flexibly cope with complex scenarios such as terrain undulations, occlusion changes, seasonal changes, and extreme weather, improving the system's adaptability and robustness.

[0050] This embodiment effectively solves the core problems of low accuracy, poor adaptability, and insufficient optimization efficiency of traditional point deployment methods, and provides a brand-new technical solution for precise micro-meteorological monitoring of photovoltaic power plants. It has important guiding significance for optimizing power plant operation strategies, improving power generation efficiency, and promoting the development of intelligent operation and maintenance of photovoltaic power plants.

[0051] Example 2 This embodiment applies to mountainous photovoltaic power station scenarios with significant terrain undulations and localized shading. Taking a 100MW mountainous photovoltaic power station as an example, the specific implementation steps are as follows: (1) Data acquisition and modeling: First, aerial photography of the mountain photovoltaic power station was carried out using UAV oblique photography technology to obtain high-precision image data, and a three-dimensional real-scene model of the power station area was established based on this data. From the model, the terrain elevation, slope, aspect, and shading information caused by photovoltaic arrays, buildings, etc. were accurately extracted.

[0052] (2) Algorithm parameter settings: Initialize the key parameters of the optimization algorithm, set the initial population size to 50, the initial value of the quantum feature length to 0.8, and the maximum number of iterations to 200. At the same time, set the weight coefficients of each sub-objective in the multi-objective fitness function as follows: coverage energy density weight 0.25, gradient sensitivity weight 0.30, observation accuracy weight 0.15, robustness weight 0.10, redundancy weight 0.10, and cost weight 0.10.

[0053] (3) Optimization and Layout Scheme Generation: Based on the above model and parameters, the quantum behavior firefly swarm optimization algorithm-reverse learning hybrid optimization algorithm was run to solve the problem. The final optimal layout scheme is as follows: a total of 60 observation points are deployed within the power station area, including 36 fixed measurement points located in key locations such as the substation and inverter room, and 24 mobile measurement points. These mobile measurement points are assigned to 4 pre-erected low-altitude cable tracks. The spatial distribution of this optimal layout scheme is as follows: Figure 5 As shown, the locations of fixed measuring points (such as triangular markers) and moving measuring points (such as circular markers) within the power station area and their corresponding track paths are clearly identifiable.

[0054] (4) Dynamic adaptive adjustment: After the system was put into operation, it was continuously monitored and evaluated through the digital twin platform. After about 3 months of operation, based on the daily calculated Kriging interpolation error and information entropy reduction indicators, the system determined that the contribution value of measuring point number 8 had fallen below the preset threshold. Therefore, the system automatically generated an adjustment command to move the moving measuring point from its original location to the newly appearing shaded area on the east side of the power station due to seasonal changes, thereby maintaining and optimizing the performance of the overall observation network.

[0055] Example 3 This embodiment is applied to a floating photovoltaic power station scenario characterized by strong water surface reflection and drastic wind speed changes. Taking a 50MW floating photovoltaic power station as an example, the implementation process was specifically adjusted for the unique water surface environment, with the following highlights: (1) Adaptive correction of model and algorithm: In order to cope with the interference of water surface fluctuations on the measurement of micro-meteorological parameters (especially irradiance), a water surface fluctuation correction term is specially added to the multi-objective fitness function adopted in this embodiment to improve the robustness and observation accuracy of the point layout scheme in such environments.

[0056] (2) Special deployment of the mobile measuring point system: The mobile measuring point unit is installed on a specially designed floating track system. The track system can automatically adjust its overall height according to the rise and fall of the water level, thereby ensuring that the sensor of the mobile measuring point unit is always kept at a suitable observation height and avoiding observation interruption or data distortion due to water level changes.

[0057] (3) Predictive Maintenance and Strategy Pre-adjustment: An predictive maintenance algorithm module is integrated. This module can comprehensively analyze meteorological forecast data and power plant historical operation data to predict the probability of extreme weather events (such as strong winds and rainstorms) up to 72 hours in advance. When high-risk weather is predicted, the system will optimize the patrol trajectory of the mobile measuring point in advance or retrieve it to a safe location, and pre-adjust the weight of the point deployment strategy in the algorithm (such as temporarily increasing the robustness weight) to proactively ensure the safety of the observation system and the continuity of data.

[0058] In summary, this embodiment constructs a four-dimensional spatiotemporal micro-meteorological field reconstruction model and a multi-objective fitness function, employs an improved quantum behavior firefly swarm optimization algorithm to solve for the optimal point placement scheme, and combines digital twin technology to achieve dynamic adaptive adjustment. It also introduces a track-based mobile measuring point system to improve observation flexibility and coverage efficiency. This solves the problems of low accuracy, poor adaptability, and insufficient optimization efficiency of traditional point placement methods. While maintaining observation accuracy, it reduces the number of measuring points by 60%, improves field reconstruction accuracy by 40%, and increases computational efficiency by 5 times. This reduces monitoring costs and improves the power generation efficiency of photovoltaic power plants, making it suitable for photovoltaic power plants in various scenarios such as mountainous areas and water surfaces.

[0059] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A dynamic adaptive deployment method for micrometeorological observation devices in photovoltaic power plants, characterized in that, Includes the following steps: Based on the terrain and shading information of photovoltaic power plants, a four-dimensional spatiotemporal micro-meteorological field reconstruction model with time decay weights is constructed. Based on the aforementioned four-dimensional spatiotemporal micro-meteorological field reconstruction model, a multi-objective fitness function is established that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. A swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and reverse learning strategies is used to iteratively solve the multi-objective fitness function, thereby obtaining the initial optimal measurement point layout scheme. The dynamic adjustment mechanism based on digital twins evaluates and adaptively adjusts the initial optimal measurement point layout scheme based on real-time collected observation data. In the adaptively adjusted point layout scheme, a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the track system is introduced to realize the observation of the area to be observed.

2. The method according to claim 1, characterized in that, Based on the terrain and shading information of photovoltaic power plants, a four-dimensional spatiotemporal micro-meteorological field reconstruction model with time decay weights is constructed. The specific process includes: A time decay weighting factor is constructed based on the decay time constant and the seasonal cycle; A four-dimensional spatiotemporal micro-meteorological field reconstruction model is constructed based on the time decay weighting factor, the trend term fitted by the spatiotemporal anisotropic kernel function, and the random residual term that follows a normal distribution.

3. The method according to claim 1, characterized in that, A multi-objective fitness function is established that comprehensively covers energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators. The specific process includes: The weighted distance is obtained based on the distance between the measuring point and the field point, as well as the weights determined by the terrain slope and shading factor. Based on the weighted distance, a coverage energy density sub-target is constructed to characterize spatial coverage capability; The observation value of the corresponding point is calculated based on the Laplace operator of the irradiance field at the measuring point. Based on the observed value, a gradient-sensitive sub-target is constructed for priority placement of points in areas with drastic changes in irradiance gradient. Based on the preset influence radius of the measuring point, a robust sub-target is constructed to ensure that the remaining measuring points can still maintain basic observation functions after any measuring point fails. The root mean square error is calculated based on historical observation data and field reconstruction results. Based on the root mean square error, an observation accuracy sub-objective is constructed to minimize the field reconstruction error; The correlation coefficient is calculated based on the correlation between the observation data of each measuring point. Based on the correlation coefficient, a redundancy sub-objective is constructed to reduce the degree of information overlap; Based on equipment procurement, installation, operation and maintenance, and energy consumption costs, construct cost sub-objectives to control total costs; The multi-objective fitness function is obtained by weighted summation of the coverage energy density sub-objective, gradient sensitivity sub-objective, robustness sub-objective, observation accuracy sub-objective, redundancy sub-objective, and cost sub-objective.

4. The method according to claim 1, characterized in that, An iterative solution is achieved using a swarm intelligence optimization algorithm that integrates quantum behavior mechanisms and a reverse learning strategy. The specific process includes: A quantum behavior-based firefly swarm optimization algorithm is adopted, which updates the quantum position of fireflies based on the population centroid and quantum feature length, and determines the sign according to the fitness value during the iteration process; In each iteration, a backward learning operation is performed on the individuals in the current population to generate the corresponding backward solution; Compare the fitness values ​​of the reverse solution and the original solution. If the reverse solution is better, then replace the original solution with the reverse solution.

5. The method according to claim 1, characterized in that, The dynamic adjustment mechanism based on digital twins is used to evaluate and adaptively adjust the deployment scheme. The specific process includes: Based on real-time observation data, the Kriging interpolation error at each measuring point is calculated daily, and the observation error is evaluated in real time. Calculate the reduction in information entropy for each measurement point, and dynamically rank all measurement points based on the reduction in information entropy; When the decrease in information entropy at a certain measuring point is lower than a preset threshold, a command is triggered to move the measuring point to another location.

6. The method according to claim 5, characterized in that, The dynamic adjustment mechanism built upon digital twins also includes: Switch between different preset deployment modes according to seasonal changes; When switching to summer mode, increase the density of measurement points between photovoltaic modules; When switching to winter mode, increase the density of measurement points in the shaded area on the north side of the photovoltaic array.

7. The method according to claim 1, characterized in that, The process of introducing a collaborative observation network consisting of fixed measuring points and mobile measuring points installed on the orbital system includes: The ratio of fixed measuring points to mobile measuring points is set at three to two. Fixed measuring points are set up at key locations in the substation or inverter room; The mobile measuring point is installed on the low-altitude cable track system.

8. The method according to claim 7, characterized in that, It also includes planning periodic survey trajectories for moving measurement points, the specific process of which includes: The trajectory planning problem of moving measurement points is modeled as a spatiotemporal traveling salesman problem with time window constraints; An improved ant colony algorithm is used to solve the spacetime traveling salesman problem to obtain an optimized survey trajectory; In trajectory planning, acceleration constraints, velocity constraints, and track offset compensation caused by wind loads are added for moving measurement points.

9. A dynamic adaptive deployment system for micrometeorological observation devices in photovoltaic power plants, used to implement the method described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to collect micro-meteorological data within the photovoltaic power station. It includes fixed measuring point units deployed at key locations within the power station and mobile measuring point units installed on the track system. The central processing module, connected to the data acquisition module, is used to construct and update a four-dimensional spatiotemporal micro-meteorological field reconstruction model based on the micro-meteorological data and integrate time decay weights. It establishes and solves a multi-objective fitness function for the fusion coverage energy density, gradient sensitivity, robustness, observation accuracy, redundancy, and cost indicators to obtain the optimal measurement point layout scheme and movement trajectory. It also dynamically evaluates and adjusts the layout scheme based on the digital twin model. A motion control module, connected to the central processing module and the moving measuring point unit, is used to receive and execute the movement trajectory command and drive the moving measuring point unit to move along the track system. A data communication module connects the data acquisition module, the central processing module, and the motion control module, and is used to transmit data and control commands between the modules. An energy supply module is used to supply power to the data acquisition module, the central processing module, the motion control module, and the data communication module.

10. An electronic device comprising a memory, a processor, and a computing program stored in the memory and executable on the processor, characterized in that, When the processor executes the computing program, it implements the method of any one of claims 1-8.

Images