New energy distributed photovoltaic intelligent distribution recommendation and dynamic monitoring system

Abstract

The invention discloses a new energy distributed photovoltaic intelligent distribution recommendation and dynamic monitoring system, and relates to the technical field of photovoltaic layout, and the system comprises a plot function prediction module which is configured to output the function attribute of a surrounding plot and a future construction behavior prediction result; an external impact assessment module configured to generate an external environmental efficiency attenuation factor; a photovoltaic efficiency core module configured to output a photoelectric conversion dynamic efficiency value; the power transmission loss optimization module is configured to calculate the comprehensive line loss rate from the site selection of the target photovoltaic power station to the grid-connected access point; the comprehensive evaluation decision engine is configured to generate a stationing feasibility score and output a recommended site selection map; and the full-life-cycle monitoring platform is configured to monitor the running state of the built new energy photovoltaic facility. The method has the advantages that the shielding pollution hidden danger in the site selection stage is avoided through plot function prediction and dynamic environment risk assessment; and the photovoltaic conversion efficiency and the power grid loss collaborative optimization model are fused, so that the power generation efficiency and the grid-connected economy are improved.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic layout technology, and in particular to a new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system. Background Art

[0002] Distributed photovoltaic power generation, a key area in the clean energy transition, is experiencing rapid growth. However, the industry faces a core contradiction: traditional site selection methods are severely out of sync with the demand for dynamic and refined deployment. On the one hand, existing site selection models rely on static geographic data, lacking the ability to predict the evolving functionalities of surrounding land parcels. These models are unable to effectively assess the erosion of power generation efficiency due to long-term factors such as shading from new high-rise buildings and the spread of industrial pollution, resulting in actual power generation far below expectations after commissioning. On the other hand, grid coordination is weak. The high proportion of distributed photovoltaic power installed leads to issues such as insufficient distribution network capacity and difficulties in optimizing power flows. Existing systems lack a deep integration of grid topology and substation load data, making it difficult to accurately quantify grid connection losses and absorption risks at different sites. Furthermore, operations and maintenance (O&M) suffer from significant lags. Most monitoring solutions only collect basic electrical parameters and lack dynamic compensation mechanisms for module degradation and dust accumulation. Fault diagnosis relies on manual inspections, and response delays result in persistent power generation losses, limiting the overall lifecycle benefits of power plants. Summary of the Invention

[0003] To solve the above technical problems, a new energy distributed photovoltaic intelligent site recommendation and dynamic monitoring system is provided. This technical solution solves the problems that the existing site selection model relies on static geographic data, lacks the ability to predict the functional evolution of surrounding plots, and is difficult to accurately quantify the grid connection losses and absorption risks of different sites.

[0004] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0005] New energy distributed photovoltaic intelligent site recommendation and dynamic monitoring system, including:

[0006] The land parcel function prediction module is configured to collect geographic information data within a preset radius around the target PV power station site. Through land use type identification and planning text analysis models, it outputs the functional attributes of the surrounding land parcels and the prediction results of future construction behavior;

[0007] an external impact assessment module configured to calculate, based on the functional attributes and construction behavior predictions, a shading probability, a pollutant diffusion coefficient, and a land development conflict index, and generate an external environmental efficiency attenuation factor;

[0008] The photovoltaic efficiency core module is configured to obtain the light radiation intensity spectrum, terrain inclination optimization parameters and temperature attenuation characteristics of the target photovoltaic power station site in real time, and output the dynamic efficiency value of photoelectric conversion;

[0009] The transmission loss optimization module is configured to combine the grid topology and substation load data to build a power transmission loss model and calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid connection point;

[0010] A comprehensive evaluation decision engine is configured to input the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value, and the comprehensive line loss rate into a weighted value function, generate a site feasibility score, and output a recommended site selection map;

[0011] The full life cycle monitoring platform is configured to monitor the operating status of completed new energy photovoltaic facilities and dynamically adjust the operating parameters of the new energy photovoltaic facilities based on the monitoring data.

[0012] Preferably, the land parcel function prediction module specifically includes:

[0013] Satellite remote sensing image segmentation unit, used to identify vegetation coverage, building outlines and undeveloped land in surrounding areas;

[0014] Urban planning database interface, real-time access to land use labels and construction project approval lists in regional control plans;

[0015] A time series-based LSTM prediction unit is used to predict the construction behavior of future photovoltaic facilities during their operational lifecycle based on historical land change data.

[0016] Preferably, the calculation of the occlusion probability, the pollutant diffusion coefficient and the land development conflict index based on the functional attributes and construction behavior prediction to generate the external environment efficiency attenuation factor specifically includes:

[0017] Using a sunlight trajectory simulation algorithm, the daily shadow coverage of new buildings on photovoltaic arrays is calculated based on the prediction results of future construction behavior;

[0018] Combining meteorological data with the Gaussian diffusion model, we calculated the impact index of new buildings on component cleanliness based on the pollutant deposition caused by future construction activities.

[0019] Through GIS spatial overlay analysis, the planning conflict levels between cultivated land red lines, ecological protection zones, and the target PV power station site are marked. Based on the planning conflict levels, a planning conflict deduction factor for the target PV power station site is set.

[0020] The external environment efficiency attenuation factor is obtained by normalizing the daily shadow coverage rate of the new building on the photovoltaic array and the impact index on the cleanliness of the components and performing weighted summation.

[0021] Preferably, the combining of the grid topology and substation load data to construct a power transmission loss model and calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid connection point specifically includes:

[0022] Based on the grid topology of the target PV power station site, analyze at least one distribution station connected to the target PV power station site after the PV site is built, which is recorded as the target distribution station;

[0023] Based on the historical operation data of the power grid, the power supply and demand data of the target distribution station is analyzed. Based on the power supply and demand data of the target distribution station and the dynamic efficiency value of the photovoltaic conversion of the target photovoltaic power station site, the power supply data of all target distribution stations after the photovoltaic site is built at the target photovoltaic power station site is determined;

[0024] Based on the power supply data of all target distribution stations after the completed photovoltaic site at the target photovoltaic power station site is completed, and the power supply line loss between the completed photovoltaic site at the target photovoltaic power station site and all target distribution stations, the comprehensive line loss rate from the target photovoltaic power station site to the grid access point is calculated.

[0025] Preferably, the step of inputting the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value, and the comprehensive line loss rate into a weighted value function, generating a site feasibility score, and outputting a recommended site selection map specifically includes:

[0026] Calculate the feasibility scores of all target PV power station sites based on the weighted value function;

[0027] Output target PV power station sites in descending order of site feasibility scores, and construct a recommended site selection map;

[0028] The weighted value function is specifically:

[0029]

[0030] Where V is the feasibility score of the target photovoltaic power station site selection, is the photoelectric conversion dynamic efficiency value of the target photovoltaic power station site selection, is the external environment efficiency attenuation factor for the target photovoltaic power station site selection, The comprehensive line loss rate from the target photovoltaic power station site to the grid access point, is the planning conflict deduction factor for the target PV power station site selection, are all weight coefficients.

[0031] Preferably, the full life cycle monitoring platform includes:

[0032] Inspection unit: Periodically detects component hot spot failure and dust coverage through visible light and infrared images;

[0033] Power quality sensor array: real-time monitoring of harmonic distortion and voltage fluctuation data at the target PV power station grid access point;

[0034] Adaptive cleaning decision maker: Triggering photovoltaic panel cleaning based on dust deposition prediction.

[0035] Preferably, the full life cycle monitoring platform further includes:

[0036] A predictive maintenance warning module is configured to identify the cell degradation rate based on the offset of the component current-voltage characteristic curve and trigger a maintenance work order when the cell degradation rate exceeds a threshold;

[0037] The theoretical power generation value is set based on the historical PV operation data. When the actual power generation is lower than the theoretical power generation value for three consecutive sunny days, a maintenance work order is activated.

[0038] Preferably, the full life cycle monitoring platform adopts a cloud-edge collaborative architecture:

[0039] Edge computing nodes are deployed in the photovoltaic station control room to pre-screen local sensor data;

[0040] The cloud-based analysis platform uses the Spark distributed computing framework to process monitoring data;

[0041] Real-time operation work orders of photovoltaic power stations on mobile terminals.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention significantly improves the technical and economic efficiency of distributed photovoltaics throughout their entire life cycle through a multi-dimensional dynamic coupling mechanism. First, by integrating the prediction of plot function evolution and the sunshine trajectory simulation algorithm, the risks of building shading and pollution deposition are accurately predicted during the site selection stage, effectively avoiding the problem of continuous power generation attenuation caused by dynamic changes in the surrounding environment in traditional solutions, and ensuring that the theoretical power generation efficiency is maintained in a stable range after the power station is put into operation. Secondly, by deeply integrating the grid topology analysis and the real-time data of the substation load, a power generation-transmission collaborative optimization model is constructed. While calculating the dynamic efficiency of photoelectric conversion, the line loss differences of different grid-connected paths are accurately quantified, thereby screening out the optimal layout plan that takes into account both its own power generation efficiency and the economic efficiency of grid absorption, significantly reducing the abandonment rate and reducing the cost of supporting grid transformation. Finally, relying on the cloud-edge collaborative architecture, closed-loop management of the health status of components is achieved, and infrared imaging and current-voltage characteristic curve offset analysis are used to actively identify hot spot failures and hidden attenuation. Predictive cleaning and maintenance are triggered in combination with environmental parameters, reducing fault downtime and eliminating power generation losses caused by the lag of manual inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is the structural diagram of the new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system proposed by the present invention;

[0045] Figure 2A flow chart of generating the external environment efficiency attenuation factor proposed by the present invention;

[0046] Figure 3 This is a flow chart of the present invention for calculating the comprehensive line loss rate from the site selection of the target photovoltaic power station to the grid access point;

[0047] Figure 4 A flowchart of generating a feasibility score for a site and outputting a recommended site selection map proposed by the present invention; DETAILED DESCRIPTION

[0048] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0049] Reference Figure 1 As shown in the figure, the new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system includes:

[0050] The land parcel function prediction module is configured to collect geographic information data within a preset radius around the target PV power station site. Through land use type identification and planning text analysis models, it outputs the functional attributes of the surrounding land parcels and the prediction results of future construction behavior;

[0051] Through the dynamic fusion of satellite remote sensing image segmentation and urban planning data, we can break through the limitations of traditional site selection that relies on static geographic data, accurately predict the expansion of surrounding construction land and the distribution of high-rise buildings during the service life of photovoltaic facilities, and analyze the shadow occlusion risks of surrounding buildings.

[0052] The external impact assessment module is configured to calculate the occlusion probability, pollutant diffusion coefficient and land development conflict index based on functional attributes and construction behavior prediction, and generate the external environmental efficiency attenuation factor;

[0053] The photovoltaic efficiency core module is configured to obtain the light radiation intensity spectrum, terrain inclination optimization parameters and temperature attenuation characteristics of the target photovoltaic power station site in real time, and output the dynamic efficiency value of photoelectric conversion;

[0054] Based on the characteristics of photovoltaic power generation, a comprehensive analysis is conducted on the light radiation intensity spectrum, terrain inclination optimization parameters, and temperature attenuation characteristics of photovoltaic modules at the target photovoltaic power station site. The photoelectric conversion efficiency value of the target photovoltaic power station site is determined to evaluate the photovoltaic power generation status of the target photovoltaic power station site under standard conditions.

[0055] The transmission loss optimization module is configured to combine the grid topology and substation load data to build a power transmission loss model and calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid connection point;

[0056] A comprehensive evaluation decision engine is configured to input the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value, and the comprehensive line loss rate into a weighted value function to generate a site feasibility score and output a recommended site selection map;

[0057] The full life cycle monitoring platform is configured to monitor the operating status of completed new energy photovoltaic facilities and dynamically adjust the operating parameters of the new energy photovoltaic facilities based on the monitoring data.

[0058] The land function prediction module deeply integrates satellite remote sensing and urban planning data to accurately predict the evolution of construction behavior in surrounding land parcels, avoiding the long-term power generation efficiency loss caused by high-rise building obstruction and industrial pollution diffusion in traditional solutions. The external impact assessment module innovatively quantifies three dynamic attenuation factors: shadow coverage, pollutant deposition rate, and land planning conflict level, significantly reducing compliance risks and environmental interference losses. The photovoltaic efficiency core module simultaneously integrates a terrain inclination optimization algorithm and a temperature attenuation compensation mechanism to dynamically evaluate photovoltaic conversion efficiency. The transmission loss optimization module uses ant colony optimization of grid topology and real-time analysis of substation load to significantly reduce line impedance losses and grid transformation costs. The comprehensive evaluation decision engine uses a weighted value function to collaboratively optimize technical and economic indicators to generate high-precision spatial heat maps and cost-per-kilowatt-hour models to support reliable investment decisions. The full life cycle monitoring platform establishes a closed-loop control mechanism of "drone inspection, current characteristic analysis, and sandstorm prediction linkage" to achieve early diagnosis of hidden component faults and adaptive cleaning and maintenance, shortening the system's unplanned downtime to one-tenth of traditional solutions, and comprehensively ensuring the stable and efficient operation of the power station throughout its life cycle.

[0059] Specifically, refer to Figure 1 As shown in Figure 2, the land parcel function prediction module specifically includes:

[0060] Satellite remote sensing image segmentation unit, used to identify vegetation coverage, building outlines and undeveloped land in surrounding areas;

[0061] Urban planning database interface, real-time access to land use labels and construction project approval lists in regional control plans;

[0062] A time series-based LSTM prediction unit is used to predict the construction behavior of future photovoltaic facilities during their operational lifecycle based on historical land change data.

[0063] The land parcel function prediction module uses satellite remote sensing image segmentation units to accurately capture the microscopic morphological characteristics of land parcels, analyze the spatial distribution status of vegetation cover, building outlines, and undeveloped vacant land, and simultaneously utilizes the urban planning database interface to dynamically obtain the legal land use nature and construction project approval lists. Based on this dual-track data fusion, it innovatively introduces a time series-based LSTM prediction unit. Through deep learning, it explores the inherent evolutionary patterns of historical land changes and proactively simulates the dynamic trajectory of surrounding construction activities throughout the life cycle of a photovoltaic power station. This technology combination achieves three-dimensional collaboration from current situation analysis to planning constraints and long-term evolution deduction, overcoming the fundamental flaw of traditional solutions in their lack of foresight into urban expansion dynamics. It enables site selection decisions to proactively avoid long-term efficiency degradation caused by obstruction by newly built high-rise buildings, migration of industrial pollutants, and conflicts in planned land use nature. It also provides a highly consistent temporal and spatial topological analysis basis for subsequent environmental factor calculations, laying a scientific foundation for full-cycle power generation stability.

[0064] Reference Figure 2 As shown in the figure, based on the functional attributes and construction behavior prediction, the shading probability, pollutant diffusion coefficient and land development conflict index are calculated to generate the external environmental efficiency attenuation factor, which includes:

[0065] Using a sunlight trajectory simulation algorithm, the daily shadow coverage of new buildings on photovoltaic arrays is calculated based on the prediction results of future construction behavior;

[0066] Specifically, based on the probability of various building constructions on the surrounding plots of the target photovoltaic power station site predicted by the LSTM prediction unit, the construction probability and the shadow coverage rate of various building facilities on the photovoltaic power station are combined to comprehensively predict the daily shadow coverage rate of the new building on the photovoltaic array. Specifically, the daily shadow coverage rate of the new building on the photovoltaic array can be calculated using the following formula:

[0067]

[0068] Where, is the daily shadow coverage of the new building on the photovoltaic array, A set of possible building types for the surrounding plots of land selected for the target photovoltaic power station site. for The elements in is the probability of building the e-th type of building predicted by the LSTM prediction unit, is the shadow coverage rate of the e-th building type for the photovoltaic power station;

[0069] Combining meteorological data with the Gaussian diffusion model, we calculated the impact index of new buildings on component cleanliness based on the pollutant deposition caused by future construction activities.

[0070] The impact index of component cleanliness is calculated using a method similar to the daily shadow coverage rate. First, the meteorological data of various buildings and facilities at the target PV power station site are analyzed. Then, the Gaussian diffusion model is used to analyze the standard impact index of each group of buildings and facilities on the cleanliness of the PV power station. The probability of various building constructions on the surrounding plots of the target PV power station site predicted by the LSTM prediction unit is then combined. The standard impact index is weighted and summed to obtain the impact index of new buildings on component cleanliness.

[0071] Through GIS spatial overlay analysis, the planning conflict levels between cultivated land red lines, ecological protection zones, and the target PV power station site are marked. Based on the planning conflict levels, a planning conflict deduction factor for the target PV power station site is set.

[0072] The external environment efficiency attenuation factor is obtained by normalizing the daily shadow coverage rate of the new building on the photovoltaic array and the impact index on the cleanliness of the components and performing weighted summation.

[0073] This innovative approach achieves precise quantification of external environmental risks through a multi-physics dynamic coupling algorithm. Based on the construction probability predicted by LSTM units, a spatial probability weighted calculation of building shadow effects is performed, dynamically generating daily shadow trajectory distribution maps covering different seasons. Meteorological parameters are coupled with a Gaussian diffusion model to convert industrial activity emission intensity, dominant wind direction frequency, and particulate deposition characteristics into a standardized pollution impact index. GIS overlay analysis then automatically annotates the spatial conflict level with ecological protection areas and agricultural land.

[0074] Reference Figure 3 As shown in the figure, combining the grid topology and substation load data, a power transmission loss model is constructed to calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid access point. Specifically, the following are included:

[0075] Based on the grid topology of the target PV power station site, analyze at least one distribution station connected to the target PV power station site after the PV site is built, which is recorded as the target distribution station;

[0076] Based on the historical operation data of the power grid, the power supply and demand data of the target distribution stations are analyzed. Based on the power supply and demand data of the target distribution stations and the dynamic efficiency value of the photovoltaic conversion of the target photovoltaic power station site, the power supply data for all target distribution stations is determined;

[0077] Based on the power supply data of all target distribution stations after the completed photovoltaic site at the target photovoltaic power station site is completed, and the power supply line loss between the completed photovoltaic site at the target photovoltaic power station site and all target distribution stations, the comprehensive line loss rate from the target photovoltaic power station site to the grid access point is calculated.

[0078] Specifically, the calculation formula for the comprehensive line loss rate is:

[0079]

[0080] The comprehensive line loss rate from the target photovoltaic power station site to the grid access point, Select the target distribution station set corresponding to the target photovoltaic power station site, is the d-th target distribution station, The daily power generation of the target photovoltaic power station after the photovoltaic site is built, The amount of power distributed to the dth target distribution station after the photovoltaic site is built for the target photovoltaic power station. The power supply line loss to the dth target distribution station after the photovoltaic site is built at the selected target photovoltaic power station site.

[0081] The transmission loss optimization module constructs a dynamic coupling model of grid topology, distribution station load, and power generation. By analyzing the electrical connection paths between the target site and multiple distribution stations and deeply integrating the power supply and demand characteristics in the historical operation data of the grid, it accurately calculates the differentiated power supply of the photovoltaic power station to each target distribution station and the corresponding line losses.

[0082] Reference Figure 4 As shown in the figure, the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value and the comprehensive line loss rate are input into the weighted value function to generate a feasibility score for the site and output a recommended site selection map, which includes:

[0083] Calculate the feasibility scores of all target PV power station sites based on the weighted value function;

[0084] Output target PV power station sites in descending order of site feasibility scores, and construct a recommended site selection map;

[0085] Among them, the weighted value function is specifically:

[0086]

[0087] Where V is the feasibility score of the target photovoltaic power station site selection, is the photoelectric conversion dynamic efficiency value of the target photovoltaic power station site selection, is the external environment efficiency attenuation factor for the target photovoltaic power station site selection, The comprehensive line loss rate from the target photovoltaic power station site to the grid access point, is the planning conflict deduction factor for the target PV power station site selection, are all weight coefficients.

[0088] The comprehensive evaluation decision engine builds a weighted value function through and The negative correlation constraint forces the site selection plan to seek the optimal solution between its own power generation efficiency and grid access loss capacity, and the weighted coefficient Dynamic adjustments based on regional policies are supported, such as increasing the lambda weight in ecological protection areas, ensuring that the output map simultaneously meets both technical feasibility and policy compliance. The resulting 3D thermal rendering of the site selection map provides a quantitative basis for site selection decisions that considers lifecycle power generation revenue, grid transformation costs, and compliance risks, systematically reducing the hidden loss risks associated with traditional empirical site selection.

[0089] Specifically, refer to Figure 1 As shown, the full life cycle monitoring platform includes:

[0090] Inspection unit: Periodically detects component hot spot failure and dust coverage through visible light and infrared images;

[0091] Power quality sensor array: real-time monitoring of harmonic distortion and voltage fluctuation data at the target PV power station grid access point;

[0092] Adaptive cleaning decision maker: Triggering photovoltaic panel cleaning based on dust deposition prediction.

[0093] In some preferred embodiments, the full life cycle monitoring platform further includes:

[0094] A predictive maintenance warning module is configured to identify the cell degradation rate based on the offset of the component current-voltage characteristic curve and trigger a maintenance work order when the cell degradation rate exceeds a threshold;

[0095] The theoretical power generation value is set based on the historical PV operation data. When the actual power generation is lower than the theoretical power generation value for three consecutive sunny days, a maintenance work order is activated.

[0096] Specifically, the full lifecycle monitoring platform adopts a cloud-edge collaborative architecture:

[0097] Edge computing nodes are deployed in the photovoltaic station control room to pre-screen local sensor data;

[0098] The cloud-based analysis platform uses the Spark distributed computing framework to process monitoring data;

[0099] Real-time operation work orders of photovoltaic power stations on mobile terminals.

[0100] The full lifecycle monitoring platform integrates the component-level defect detection capabilities of visible light and infrared imagery through inspection units, and combines this with a cloud-based distributed processing framework to achieve millimeter-level positioning accuracy for hot spot failures and dust accumulation. Simultaneously, a power quality sensor array monitors harmonic distortion and voltage fluctuations at the grid connection point in milliseconds, accurately tracing the source of grid interaction anomalies. The platform innovatively integrates a dust deposition model with an adaptive cleaning decision maker to establish a closed-loop control mechanism that predicts pollution rate and triggers cleaning instructions. Furthermore, component-level attenuation characteristics are identified through offset analysis of current-voltage characteristic curves, automatically activating predictive maintenance work orders when power generation on sunny days deviates abnormally from theoretical values. Relying on the intelligent collaborative model of the cloud-edge collaborative architecture, edge nodes perform real-time pre-screening and compression of sensor data to reduce transmission delays. The cloud platform utilizes a big data engine to conduct cross-site energy efficiency degradation correlation analysis, and mobile terminals simultaneously push executable operation and maintenance instructions. This forms a full-chain closed-loop management system of "data perception, intelligent diagnosis, and dynamic response." This upgrades the traditional passive operation and maintenance model to predictive dynamic intervention based on spatiotemporal big data, significantly eliminating power generation losses caused by hidden component attenuation and sudden failures.

[0101] In summary, the advantages of the present invention are: significantly improving the technical and economic efficiency of distributed photovoltaics throughout their entire life cycle through a multi-dimensional dynamic coupling mechanism. First, by integrating the prediction of plot function evolution and the sunshine trajectory simulation algorithm, the risks of building shading and pollution deposition can be accurately predicted in the site selection stage, effectively avoiding the problem of continuous power generation attenuation caused by dynamic changes in the surrounding environment in traditional solutions, and ensuring that the theoretical power generation efficiency is maintained in a stable range after the power station is put into operation. Secondly, by deeply integrating the grid topology analysis and the real-time data of the substation load, a power generation-transmission collaborative optimization model is constructed. While calculating the dynamic efficiency of photoelectric conversion, the line loss differences of different grid-connected paths are accurately quantified, thereby screening out the optimal layout plan that takes into account both its own power generation efficiency and the economic efficiency of grid absorption, significantly reducing the abandonment rate and reducing the cost of supporting grid transformation. Finally, relying on the cloud-edge collaborative architecture, closed-loop management of the health status of components is achieved, and infrared imaging and current-voltage characteristic curve offset analysis are used to actively identify hot spot failures and hidden attenuation. Predictive cleaning and maintenance are triggered in combination with environmental parameters, reducing fault downtime and eliminating power generation losses caused by the lag of manual inspections.

[0102] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. New energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system, characterized by: include: The land parcel function prediction module is configured to collect geographic information data within a preset radius around the target PV power station site. Through land use type identification and planning text analysis models, it outputs the functional attributes of the surrounding land parcels and the prediction results of future construction behavior; an external impact assessment module configured to calculate, based on the functional attributes and construction behavior predictions, a shading probability, a pollutant diffusion coefficient, and a land development conflict index, and generate an external environmental efficiency attenuation factor; The photovoltaic efficiency core module is configured to obtain the light radiation intensity spectrum, terrain inclination optimization parameters and temperature attenuation characteristics of the target photovoltaic power station site in real time, and output the dynamic efficiency value of photoelectric conversion; The transmission loss optimization module is configured to combine the grid topology and substation load data to build a power transmission loss model and calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid connection point; A comprehensive evaluation decision engine is configured to input the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value, and the comprehensive line loss rate into a weighted value function, generate a site feasibility score, and output a recommended site selection map; The full life cycle monitoring platform is configured to monitor the operating status of completed new energy photovoltaic facilities and dynamically adjust the operating parameters of the new energy photovoltaic facilities based on the monitoring data.

2. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 1 is characterized in that: The land parcel function prediction module specifically includes: Satellite remote sensing image segmentation unit, used to identify vegetation coverage, building outlines and undeveloped land in surrounding areas; Urban planning database interface, real-time access to land use labels and construction project approval lists in regional control plans; A time series-based LSTM prediction unit is used to predict the construction behavior of future photovoltaic facilities during their operational lifecycle based on historical land change data.

3. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 2 is characterized in that: The calculation of the shading probability, the pollutant diffusion coefficient and the land development conflict index based on the functional attributes and construction behavior prediction to generate the external environment efficiency attenuation factor specifically includes: Using a sunlight trajectory simulation algorithm, the daily shadow coverage of new buildings on photovoltaic arrays is calculated based on the prediction results of future construction behavior; Combining meteorological data with the Gaussian diffusion model, we calculated the impact index of new buildings on component cleanliness based on the pollutant deposition caused by future construction activities. Through GIS spatial overlay analysis, the planning conflict levels between cultivated land red lines, ecological protection zones, and the target PV power station site are marked. Based on the planning conflict levels, a planning conflict deduction factor for the target PV power station site is set. The external environment efficiency attenuation factor is obtained by normalizing the daily shadow coverage rate of the new building on the photovoltaic array and the impact index on the cleanliness of the components and performing weighted summation.

4. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 3 is characterized in that: The power transmission loss model is constructed by combining the grid topology and substation load data to calculate the comprehensive line loss rate from the target photovoltaic power station site to the grid connection point. Specifically, the following steps are involved: Based on the grid topology of the target PV power station site, analyze at least one distribution station connected to the target PV power station site after the PV site is built, which is recorded as the target distribution station; Based on the historical operation data of the power grid, the power supply and demand data of the target distribution station is analyzed. Based on the power supply and demand data of the target distribution station and the dynamic efficiency value of the photovoltaic conversion of the target photovoltaic power station site, the power supply data of all target distribution stations after the photovoltaic site is built at the target photovoltaic power station site is determined; Based on the power supply data of all target distribution stations after the completed photovoltaic site at the target photovoltaic power station site is completed, and the power supply line loss between the completed photovoltaic site at the target photovoltaic power station site and all target distribution stations, the comprehensive line loss rate from the target photovoltaic power station site to the grid access point is calculated.

5. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 4 is characterized in that: The step of inputting the external environment efficiency attenuation factor, the photoelectric conversion dynamic efficiency value, and the comprehensive line loss rate into a weighted value function to generate a site feasibility score and output a recommended site selection map specifically includes: Calculate the feasibility scores of all target PV power station sites based on the weighted value function; Output the target PV power station sites in descending order of site feasibility scores and construct a recommended site selection map; The weighted value function is specifically: Where V is the feasibility score of the target photovoltaic power station site selection, is the photoelectric conversion dynamic efficiency value of the target photovoltaic power station site selection, is the external environment efficiency attenuation factor for the target photovoltaic power station site selection, The comprehensive line loss rate from the site selection of the target photovoltaic power station to the grid access point, is the planning conflict deduction factor for the target PV power station site selection, are all weight coefficients.

6. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 5 is characterized in that: The full life cycle monitoring platform includes: Inspection unit: Periodically detects component hot spot failure and dust coverage through visible light and infrared images; Power quality sensor array: real-time monitoring of harmonic distortion and voltage fluctuation data at the target PV power station grid access point; Adaptive cleaning decision maker: Triggering photovoltaic panel cleaning based on dust deposition prediction.

7. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 6 is characterized in that: The full life cycle monitoring platform also includes: A predictive maintenance warning module is configured to identify the cell degradation rate based on the offset of the component current-voltage characteristic curve and trigger a maintenance work order when the cell degradation rate exceeds a threshold; The theoretical power generation value is set based on the historical PV operation data. When the actual power generation is lower than the theoretical power generation value for three consecutive sunny days, a maintenance work order is activated.

8. The new energy distributed photovoltaic intelligent layout recommendation and dynamic monitoring system according to claim 7 is characterized in that: The full life cycle monitoring platform adopts a cloud-edge collaborative architecture: Edge computing nodes are deployed in the photovoltaic station control room to pre-screen local sensor data; The cloud-based analysis platform uses the Spark distributed computing framework to process monitoring data; Real-time operation work orders of photovoltaic power stations on mobile terminals.

Images