Marine photovoltaic power generation facility development trend prediction method and device, and storage medium

Abstract

The present application relates to the technical field of remote sensing, in particular to a method and device for predicting the development trend of offshore photovoltaic power generation facilities, equipment and computer storage medium. The present application collects and integrates multi-source remote sensing data, including data from different remote sensing platforms, fuses remote sensing data with geographic information data, establishes a unified geographic spatial database, and performs spatio-temporal dynamic analysis on offshore photovoltaic power generation facilities. By introducing long time series remote sensing data and spatio-temporal dynamic analysis method, the present application realizes comprehensive dynamic monitoring of the whole process from construction to operation of offshore photovoltaic power generation facilities. This not only can accurately capture the subtle changes in the construction process, but also can continuously monitor the expansion and operation status of the facilities, so as to timely discover and warn potential problems. This comprehensive dynamic monitoring capability helps to improve the management level of the facilities, reduce the maintenance cost, and ensure the long-term stable operation of the facilities.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote sensing, in particular to a method, device and equipment for predicting the development trend of offshore photovoltaic power generation facilities and a computer storage medium. BACKGROUND

[0002] With the continuous growth of global demand for clean energy, offshore photovoltaic power generation, as a highly potential renewable energy utilization method, is gradually becoming a research hotspot in the energy field. The ocean contains abundant solar energy resources, and offshore photovoltaic power generation has the advantages of not occupying land resources and relatively high power generation efficiency. However, the construction and operation of offshore photovoltaic power generation facilities face many complex challenges. The marine environment is complex and changeable, including high humidity, high salinity, strong wind and wave, and complex marine ecosystems, which puts higher requirements on the design, construction, maintenance and management of photovoltaic power generation facilities.

[0003] However, in the existing technology, for the monitoring and analysis of offshore photovoltaic power generation facilities, especially the spatio-temporal dynamic analysis, the existing remote sensing technology only relies on optical remote sensing data when acquiring offshore photovoltaic power generation facility data, which affects the time continuity and completeness of the data. Secondly, in the aspect of feature extraction and analysis, the existing method is difficult to accurately extract the subtle change characteristics of offshore photovoltaic power generation facilities in the construction, expansion and operation process, and it is difficult to accurately evaluate the change trend of facility power generation efficiency over time and the influence on the surrounding marine ecological environment.

[0004] In summary, the existing technical means has many defects in the spatio-temporal dynamic analysis of offshore photovoltaic power generation facilities, and cannot meet the needs of the rapid development of the offshore photovoltaic power generation industry. SUMMARY

[0005] Therefore, the technical problem to be solved by the present application is to overcome the problem of inaccurate spatio-temporal dynamic analysis and facility development trend prediction in the prior art.

[0006] To solve the above technical problems, the present application provides a method for predicting the development trend of offshore photovoltaic power generation facilities, comprising:

[0007] Collecting multi-source remote sensing data and fusing geographic information data to establish a geographic spatial database;

[0008] Performing spatio-temporal dynamic analysis on offshore photovoltaic power generation facilities based on the geographic spatial database, wherein the spatio-temporal dynamic analysis includes construction phase analysis, expansion phase analysis and operation state analysis, and the construction progress, expansion situation and operation state are obtained;

[0009] Based on the geographic spatial database, multi-dimensional feature variables related to the development trend of offshore photovoltaic power generation facilities are extracted from long-time sequence remote sensing data and auxiliary data.

[0010] constructing a prediction model based on the multi-dimensional feature variables, in combination with the geospatial database, the expansion situation and the operation state;

[0011] using the trained prediction model to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility, and integrating the prediction output result with the construction progress, the expansion situation and the operation state to obtain a target prediction result.

[0012] Preferably, the collecting multi-source remote sensing data and fusing geographic information data to establish a geospatial database comprises:

[0013] collecting multi-source remote sensing data from different remote sensing platforms and pre-processing the multi-source remote sensing data;

[0014] fusing the pre-processed multi-source remote sensing data with geographic information data to establish a geospatial database.

[0015] Preferably, the spatio-temporal dynamic analysis of the offshore photovoltaic power generation facility based on the geospatial database comprises construction phase analysis, expansion phase analysis and operation state analysis to obtain the construction progress, the expansion situation and the operation state, and further comprises:

[0016] extracting feature information of remote sensing images at different periods in the facility construction process based on the geospatial database, and analyzing the difference between the actual construction situation and the design planning data according to the feature information to evaluate the construction progress;

[0017] comparing remote sensing images at different periods based on the geospatial database, and determining the direction, range and speed of future expansion in combination with the historical expansion situation to obtain the future expansion situation;

[0018] extracting spectral feature information of remote sensing images at different periods in the facility operation process based on the geospatial database, and monitoring the surface state of the photovoltaic panels according to the spectral feature information, in combination with the historical facility operation state to determine the trend of the facility power generation efficiency over time, thereby determining the future facility operation state.

[0019] Preferably, the spatio-temporal dynamic analysis of the offshore photovoltaic power generation facility based on the geospatial database comprises construction phase analysis, expansion phase analysis and operation state analysis to obtain the construction progress, the expansion situation and the operation state, and further comprises:

[0020] analyzing the construction deviation reasons according to the construction progress in combination with marine engineering knowledge and on-site construction conditions, and outputting improvement suggestions according to the construction deviation reasons;

[0021] According to the expansion situation, the rationality of the expansion scheme is analyzed in combination with marine environmental factors and energy demand changes, and the spatial analysis technology is used to analyze the mutual relationship between the expanded facility and the surrounding marine ecological environment, and the expansion scheme is adjusted according to the analysis result;

[0022] In combination with the operation situation, the marine ecological environment change around the facility is monitored, and the real-time monitoring result is analyzed, and the corresponding improvement measures for facility construction or operation are output.

[0023] Preferably, the constructing a prediction model based on the multi-dimensional feature variables in combination with the geographic spatial database, the expansion situation and the operation state comprises:

[0024] selecting and reducing the multi-dimensional feature variables;

[0025] constructing a prediction model based on the selected multi-dimensional feature variables in combination with the geographic spatial database, the expansion situation and the operation state.

[0026] Preferably, the using the trained prediction model to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility, and integrating the prediction output result with the construction progress, the expansion situation and the operation state to obtain a target prediction result further comprises:

[0027] performing anomaly detection on the target prediction result according to a preset threshold;

[0028] when an anomaly is detected, triggering an alarm.

[0029] Preferably, when an anomaly is detected, triggering an alarm comprises:

[0030] determining an anomaly degree score of the detected anomaly;

[0031] based on the anomaly degree score, determining an alarm level corresponding to the detected anomaly according to a preset severity threshold;

[0032] triggering an alarm of a corresponding type according to the alarm level.

[0033] The application also provides an offshore photovoltaic power generation facility development trend prediction device, comprising:

[0034] a data collection and fusion module for collecting multi-source remote sensing data and fusing geographic information data to establish a geographic spatial database;

[0035] a space-time dynamic analysis module for performing space-time dynamic analysis on the offshore photovoltaic power generation facility based on the geographic spatial database, wherein the space-time dynamic analysis comprises construction phase analysis, expansion phase analysis and operation state analysis, and the construction progress, the expansion situation and the operation state are obtained;

[0036] a multi-dimensional feature extraction module configured to extract multi-dimensional feature variables related to the development trend of the offshore photovoltaic power generation facility from long-time series remote sensing data and auxiliary data based on the geospatial database;

[0037] a prediction model construction module configured to construct a prediction model based on the multi-dimensional feature variables in combination with the geospatial database, the expansion situation and the operation state;

[0038] a prediction result output module configured to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility by using the trained prediction model, and integrate the prediction output result with the construction progress, the expansion situation and the operation state to obtain a target prediction result.

[0039] The application further provides an offshore photovoltaic power generation facility development trend prediction device, comprising:

[0040] a memory configured to store a computer program;

[0041] a processor configured to implement the steps of the above-mentioned offshore photovoltaic power generation facility development trend prediction method when executing the computer program.

[0042] The application further provides a computer readable storage medium having a computer program stored thereon, and the computer program is configured to implement the steps of the above-mentioned offshore photovoltaic power generation facility development trend prediction method when executed by a processor.

[0043] The above-mentioned technical solution of the application has the following advantages compared with the prior art:

[0044] The offshore photovoltaic power generation facility development trend prediction method of the application realizes comprehensive dynamic monitoring of the whole process from construction to operation of the offshore photovoltaic power generation facility by introducing long-time series remote sensing data and spatio-temporal dynamic analysis method, which not only can accurately capture subtle changes in the construction process, but also can continuously monitor the expansion situation and operation state of the facility, so as to timely discover and warn potential problems, and this comprehensive dynamic monitoring capability helps to improve the management level of the facility, reduce the maintenance cost and ensure the long-term stable operation of the facility; the prediction model constructed based on multi-dimensional feature variables can accurately predict the development trend of the facility in the future period of time, and through multi-scenario prediction, the facility scale growth, technology update direction, power generation efficiency change and influence on the surrounding marine ecological environment can be evaluated, the prediction result is presented in the form of intuitive charts and reports, which provides a scientific basis for decision makers, at the same time, the warning mechanism can automatically issue an alarm when detecting an abnormal situation, helping relevant personnel to take timely response measures, thereby improving the accuracy and response speed of decision making. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to make the content of the present application more easily understood, the present application will be further described in detail below according to specific embodiments of the present application and in conjunction with the accompanying drawings, in which:

[0046] Figure 1 is an implementation flowchart of a marine photovoltaic power generation facility development trend prediction method provided by the present application;

[0047] Figure 2 is an implementation flowchart of a marine photovoltaic power generation facility development trend prediction method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0048] The core of the present application is to provide a marine photovoltaic power generation facility development trend prediction method, device, equipment and computer storage medium, which can provide comprehensive and accurate data support and decision basis for the planning, construction, operation and maintenance of marine photovoltaic power generation facilities through in-depth spatio-temporal dynamic analysis and accurate prediction model, and can develop more scientific and reasonable development strategies according to the historical development situation and future trend prediction of the facility, avoid blind investment and resource waste, and improve the success rate and economic benefit of the project.

[0049] In order to make the person skilled in the art better understand the present application, the present application will be further described in detail below in conjunction with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor belong to the scope of protection of the present application.

[0050] Please refer to Figure 1 , Figure 1 is an implementation flowchart of a marine photovoltaic power generation facility development trend prediction method provided by the present application;The specific operation steps are as follows:

[0051] S101: Collect multi-source remote sensing data, and fuse geographic information data to establish a geographic space database;

[0052] S102: Based on the geographic space database, perform spatio-temporal dynamic analysis on the marine photovoltaic power generation facility, wherein the spatio-temporal dynamic analysis includes construction phase analysis, expansion phase analysis and operation state analysis, and the construction progress, expansion situation and operation state are obtained;

[0053] S103: Based on the geographic space database, extract multi-dimensional feature variables related to the development trend of the marine photovoltaic power generation facility from long-time sequence remote sensing data and auxiliary data;

[0054] S104: Based on the multi-dimensional feature variables, a prediction model is constructed by combining the geospatial database, the expansion status, and the operational status;

[0055] S105: Utilize the trained prediction model to predict the development trend of offshore photovoltaic power generation facilities in multiple scenarios, and integrate the prediction output with the construction progress, expansion status, and operational status to obtain the target prediction result.

[0056] Based on the above embodiments, this embodiment will provide a detailed description of step S101:

[0057] • Collect multi-source remote sensing data from different remote sensing platforms and preprocess the multi-source remote sensing data;

[0058] Multi-source remote sensing data includes data from different remote sensing platforms;

[0059] In the first specific application, the required remote sensing data sources are determined, including optical satellites, radar satellites, and hyperspectral satellites. For example, optical satellites with high spatial resolution are selected to acquire information on the appearance and layout of offshore photovoltaic power generation facilities, radar satellites are used to acquire facility data under adverse weather conditions by utilizing their ability to penetrate clouds and fog, and hyperspectral satellites provide rich spectral information for analyzing the material properties of the facilities.

[0060] In the second specific application, multiple data sources are selected, including optical satellites, radar satellites, and hyperspectral satellites. Optical satellite data is used to acquire intuitive image information about offshore photovoltaic power generation facilities and the surrounding marine environment, such as the location and scale of the facilities, as well as some basic features of the surrounding ocean surface. Radar satellite data can penetrate clouds and fog, providing information about the structure of the facilities and the condition of the ocean surface under adverse weather conditions, which is crucial for continuous monitoring of the facilities and the marine environment. Hyperspectral satellite data provides rich spectral information, which can be used to analyze the details of the marine ecological environment in depth, such as water quality and the distribution of marine biological communities. Preprocessing operations are performed on the various types of remote sensing data collected. For optical satellite data, radiometric calibration is performed to ensure the physical accuracy of the data, atmospheric correction is performed to eliminate atmospheric interference with the spectrum, and geometric correction is performed to ensure that the data accurately corresponds in geospatial space. Similarly, corresponding preprocessing steps are performed on radar satellite and hyperspectral satellite data.

[0061] • The preprocessed multi-source remote sensing data is fused with geographic information data to establish a geospatial database.

[0062] Specifically, the collected remote sensing data is preprocessed and fused to combine remote sensing data from different sources and of different types to obtain D. integrated = f(D1, D2, ..., D n ;G), where Dintegrated Dintegrated represents the integrated geospatial database, D i D represents the data set of the i-th remote sensing platform, G represents the geographic information data, f is a comprehensive function for fusing multiple data sources and geographic information data, and D integrated The specific form is: where (x, y) represents the geographic coordinates, t represents the time point, w i D represents the weight coefficient of the i-th remote sensing data source, used to adjust the contribution of different data sources in the fusion process, and a represents the weight coefficient of the geographic information data, used to adjust the contribution of the geographic information data in the fusion process.

[0063] In the first specific application, for optical satellite data, radiation calibration converts the digital quantization value obtained by the sensor into a physically meaningful radiance or reflectivity value, atmospheric correction eliminates the absorption and scattering of electromagnetic wave radiation by the atmosphere, restores the true spectral characteristics of the ground object, and geometric correction corrects the geometric deformation of the image by establishing a mapping relationship between the image coordinates and the geographic coordinates. Similarly, corresponding preprocessing operations are also performed on radar satellite and hyperspectral satellite data, and data fusion is performed on the preprocessed data. Let the optical satellite data be D1, the radar satellite data be D2, the hyperspectral satellite data be D3, and the geographic information data be G. According to the formula D i integrated = f (D1, D2, D3; G), where D integrated Dintegrated represents the integrated geospatial database, and f is a comprehensive function, the specific form of which is Here (x, y) represents the geographic coordinates, t represents the time point, w1, w2, w3 are the weight coefficients of the optical satellite, radar satellite, and hyperspectral satellite data sources, respectively, and a is the weight coefficient of the geographic information data. For example, according to the data quality and the importance of the facility analysis, w1 = 0.4, w2 = 0.3, w3 = 0.2, and a = 0.1 are determined. In this way, different sources and different types of remote sensing data and geographic information data are fused to establish a unified geospatial database, providing a comprehensive data foundation for subsequent analysis.

[0064] In the second specific application, the collected data is fused, i.e., let the optical satellite data be D1, the radar satellite data be D2, the hyperspectral satellite data be D3, and the geographic information data be G. According to the formula D integrated integrated = f (D1, D2, D3; G), where D integrated Dintegrated represents the integrated geospatial database, and f is a comprehensive function, the specific form of which is Here (x, y) represents the geographical location coordinates, t represents the time point, according to the characteristics of the scene and data requirements, determine w1=0.3, w2=0.3, w3=0.2, α=0.2, in this way, the data of different sources are integrated into a comprehensive geospatial database, which provides a solid data foundation for subsequent analysis.

[0065] Based on the above embodiment, the step S102 is described in detail:

[0066] · Based on the geospatial database, the feature information of remote sensing images at different stages of facility construction is extracted, and the difference between the actual construction and the design planning data is analyzed according to the feature information to evaluate the construction progress;

[0067] Specifically, from the integrated geospatial database D integratod (x, y, t), the feature information of the facility construction process is extracted, and the difference between the actual construction and the design planning data is calculated to evaluate the construction progress, that is, Where C progress (x, y, t) represents the construction progress at geographical location (x, y) and time point t, D integratod (x, y, t) represents the integrated geospatial database, P design (x, y) represents the design planning data, ki represents the weight coefficient of the i-th feature, which is used to adjust the importance of different features in the evaluation, and m represents the number of features.

[0068] In the first specific application, from the integrated geospatial database D integratod (x, y, t), the feature information of the facility construction process is extracted, by comparing remote sensing images at different times, using image recognition and analysis technology, for the basic framework (such as support pile), the position, shape, size and other parameters are determined, for the photovoltaic panel, the installation angle and layout change are identified, and the construction progress is evaluated, according to the formula Where C progress (x, y, t) represents the construction progress at geographical location (x, y) and time point t, D integratod (x, y, t) represents the integrated geospatial database, P desigm (x, y) represents the design planning data, k i represents the weight coefficient of the i-th feature. For example, assuming that the weight coefficient of the basic framework position k1=0.3, the shape weight coefficient k2=0.2, the size weight coefficient k3=0.2, the photovoltaic panel installation angle weight coefficient k4=0.2, and the layout change weight coefficient k5=0.1. By calculating the difference between the actual construction and the design planning data, a quantitative index of construction progress is obtained.

[0069] In the second specific application, during the facility construction phase, by comparing remote sensing images at different times, using image recognition and analysis techniques, not only the feature information extraction of the facility construction progress itself is concerned, but also the impact of the construction process on the surrounding marine ecological environment is considered. For example, during the construction of the foundation structure (such as support piles), it is analyzed whether the marine biological activities around the construction area are disturbed, and the distribution changes of marine organisms and the disturbance of the water body are monitored to evaluate the impact. For the installation of photovoltaic panels, the impact of the change of light conditions in the surrounding sea area on marine organisms, such as the photosynthesis of certain plankton, may be affected by the change of light intensity and time, and when evaluating the construction progress, The idea is to consider ecological impact factors in the consideration range.

[0070] • Based on the geospatial database, compare remote sensing images at different periods, and combine historical expansion to determine the direction, range and speed of future expansion, and obtain the future expansion situation;

[0071] Specifically, by comparing remote sensing images at different periods, the direction, range and speed of expansion are determined, i.e. Ee xpansion (x, y, t) = AD integrated (x, y, t) + λ·E prev (x, y, t-1), where E expansion (x, y, t) represents the expansion situation at geographical location (x, y) and time point t, AD integrated (x, y, t) represents the data difference between the current time point and the previous time point, E prev (x, y, t-1) represents the expansion situation at the previous time point, and λ represents the smoothing coefficient of the expansion trend.

[0072] In the first specific application, the expansion of the facility is continuously monitored, and by comparing remote sensing images at different periods, the direction, range and speed of expansion are determined. Let the current time point be t and the previous time point be t-1, according to the formula E expansion (x, y, t) = AD integrated (x, y, t) + λ·E prev (x, y, t-1), where E expansiom (x, y, t) represents the expansion situation at geographical location (x, y) and time point t, AD integrated (x, y, t) represents the data difference between the current time point and the previous time point, E prev (x, y, t-1) represents the expansion situation at the previous time point, and λ represents the smoothing coefficient of the expansion trend. For example, when λ = 0.5, if AD integrated (x, y, t) shows that the facility area in a certain area has increased by 1000 square meters in this time period, Eprev (x, y, t-1) represents the expansion of the region at the last time period, which is an increase of 500 square meters, then E expansiom (x, y, t) = 1000 + 0.5 * 500 = 1250 square meters, that is, the expansion of the region at this time period is an increase of 1250 square meters.

[0073] • Extracting spectral feature information of remote sensing images at different periods of facility operation based on the geospatial database, and monitoring the surface state of the facility photovoltaic panel according to the spectral feature information, combining the historical facility operation state, to determine the trend of the facility power generation efficiency over time, thereby determining the future facility operation state.

[0074] Among them, the surface state of the facility photovoltaic panel includes the dirt, aging, and damage of the photovoltaic panel surface;

[0075] Specifically, based on the spectral features of remote sensing data, the dirt, aging, and damage of the photovoltaic panel surface are monitored, and the trend of the facility power generation efficiency over time is evaluated, that is, O status (x, y, t) = μ·D integrated (x, y, t) + (1-μ)·O prev (x, y, t-1), O status (x, y, t) represents the operation state at geographical location (x, y) and time point t, μ represents the weight coefficient of the current data, used to adjust the influence of the current data and the historical data, O prev (x, y, t-1) represents the operation state at the previous time point.

[0076] In the first specific application, based on the spectral features of remote sensing data, the dirt, aging, and damage of the photovoltaic panel surface are monitored. For example, through hyperspectral remote sensing data, the reflectivity changes of the photovoltaic panel at different wavebands are analyzed. Dirt will cause the reflectivity of the photovoltaic panel to rise at certain wavebands, and aging and damage may cause abnormal changes in reflectivity at other wavebands, and the trend of the facility power generation efficiency over time is evaluated in combination with time series data. According to the formula, the trend of the facility power generation efficiency over time is evaluated. According to the formula O status (x, y, t) = μ·D integrated (x, y, t) + (1-μ)·O prev (x, y, t-1), where O status (x, y, t) represents the operation state at geographical location (x, y) and time point t, μ represents the weight coefficient of the current data, O prev (x, y, t-1) represents the operation state at the previous time point, for example, set μ = 0.6, if the current D integrated(x, y, t) shows that the power generation efficiency of the photovoltaic panel under the comprehensive influence of spectral characteristics and environmental factors can be 80%, and the operating status O of the last time period prev (x, y, t-1) shows that the power generation efficiency is 82%, so O status (x, y, t) = 0.6 x 80% + (1-0.6) x 82% = 80.8%, i.e. the current estimated value of power generation efficiency is 80.8%.

[0077] • According to the construction progress, combined with marine engineering knowledge and on-site construction conditions, analyze the reasons for the construction deviation, and output improvement suggestions according to the construction deviation reasons;

[0078] In the first specific application, according to the construction progress, combined with marine engineering knowledge and on-site construction conditions, analyze the reasons for the construction deviation, for example, if it is found that the construction progress of the support pile in this area is lagging behind, analyze whether the seabed geological conditions are complex, whether the construction equipment is malfunctioning, whether it is affected by adverse weather (such as typhoon, heavy rain, etc.), whether the technical level and manpower arrangement of the construction personnel are reasonable, etc. According to the analysis results, provide improvement suggestions, such as adjusting the construction equipment, optimizing the construction scheme or increasing the construction personnel, etc.

[0079] In the second specific application, if it is found that the construction process has caused great disturbance to an important marine ecological region, even if the construction progress of the facility meets the design planning, it is necessary to adjust the construction scheme to reduce the impact on the ecological environment. Through this comprehensive evaluation, it is ensured that the facility construction meets the engineering progress requirements while protecting the marine ecological environment to the greatest extent.

[0080] • According to the expansion situation, combined with marine environmental factors and changes in energy demand, analyze the rationality of the expansion scheme, and use spatial analysis technology to analyze the mutual relationship between the expanded facility and the surrounding marine ecological environment, and adjust the expansion scheme according to the analysis results;

[0081] In the first specific application, the rationality of the expansion plan is evaluated considering the marine environmental factors (such as ocean currents, wind speed, waves, etc.) and changes in energy demand. For example, if the facility expands in a direction that may be impacted by strong ocean currents, the impact of the currents on the facility structure and the efficiency of power generation is analyzed. At the same time, combined with energy demand forecasts, if the energy demand in this region grows rapidly in the future and the expansion direction meets the requirements of marine environmental conditions, the expansion plan has certain rationality; otherwise, if the expansion direction will cause great damage to the marine ecological environment and the energy demand growth is not obvious, the expansion plan needs to be re-evaluated to analyze the relationship between the expanded facility and the surrounding marine ecological environment. Using spatial analysis techniques, the distance and spatial overlap between the facility and important ecological areas such as marine protected areas, fish migration channels, and coral reefs are calculated to ensure ecological balance. If it is found that the expanded facility overlaps with a marine protected area, the expansion plan needs to be adjusted to avoid damaging the ecological environment.

[0082] · In combination with the operation, monitor the changes in the marine ecological environment around the facility and analyze the real-time monitoring results to output corresponding improvement measures for the construction or operation of the facility.

[0083] In the first specific application, the changes in the marine ecological environment around the facility are monitored simultaneously, such as water quality parameters (temperature, salinity, pH, dissolved oxygen, etc.) and marine biological community structure and diversity, etc. High-spectral remote sensing data and biological index analysis methods are used, for example, by analyzing the spectral characteristics of seawater to retrieve water quality parameters and by analyzing marine biological survey data to evaluate changes in biological community structure and diversity. If it is found that the water quality in the surrounding sea area has deteriorated or the biological community structure has changed significantly after the operation of the facility, further analysis is needed to find the cause, which may be due to the release of harmful substances by the materials of the photovoltaic panels or the changes in ocean currents and light conditions caused by the construction and operation of the facility, etc. Corresponding measures are taken to improve the situation.

[0084] In the second specific application, the spectral characteristics of remote sensing data are used to monitor changes in the marine ecological environment: the rich spectral information of hyperspectral remote sensing data is used to analyze water quality parameters (such as temperature, salinity, pH, dissolved oxygen, etc.). By establishing a relationship model between spectral characteristics and water quality parameters, water quality information can be retrieved from remote sensing data. For example, the spectral reflectance of certain bands has a specific relationship with the dissolved oxygen content in water. By monitoring the changes in reflectance of these bands, the changes in dissolved oxygen content can be understood in real time. At the same time, biological indicator analysis methods are used in combination with remote sensing data to monitor marine biological community structure and diversity. For example, by analyzing the spectral characteristics of the ocean surface and some specific spectral markers related to the distribution of marine organisms, the distribution range and changes in the composition of marine biological communities can be inferred. Based on remote sensing data, the impact of facility operation on the marine ecological environment is monitored: by comparing remote sensing data before and after the construction of facilities and during different operation stages, the impact of facility operation on the marine ecological environment is analyzed. For example, by observing the changes in spectral characteristics of the sea area around the facility, if it is found that the reflectance of certain bands has changed significantly after the operation of the facility, it may mean that the water quality or marine biological community in that area has been affected. By combining time series data, the trend of this impact over time is further analyzed, and the O status (x, y, t) = μ · D integrated (x, y, t) + (1 - μ) · O prev (x, y, t - 1), the dynamic evaluation idea of the operation state is embodied, which can more accurately grasp the impact of facility operation on the ecological environment.

[0085] Based on the above embodiments, step S103 is described in detail:

[0086] Specifically, multi-dimensional feature variables F integrated related to the development trend of offshore photovoltaic power generation facilities are extracted from long time series of remote sensing data D i and auxiliary data.

[0087] In the first specific application, multi-dimensional feature variables F integrated related to the development trend of offshore photovoltaic power generation facilities are extracted from long time series of remote sensing data D i and auxiliary data (such as meteorological data, marine environmental data, energy market data, etc.). For example, spatial distribution change features, image texture features, wind speed, wind direction, rainfall, solar radiation intensity in meteorological data, ocean temperature, salinity, wave height, ocean current speed and direction in marine environmental data, and electricity price, market demand, changes in policies and regulations in energy market data are extracted.

[0088] Based on the above embodiments, step S104 is described in detail:

[0089] • selecting and reducing dimensionality of the multi-dimensional feature variables;

[0090] • constructing a prediction model based on the selected multi-dimensional feature variables, in combination with the geospatial database, the expansion situation, and the operation status.

[0091] Specifically, based on the selected feature variables F i in combination with the integrated geospatial database D integrated , the expansion situation E expansion , and the operation status O status , a prediction model M predict is constructed, i.e. where M predict denotes the prediction model, F i denotes the i-th feature variable, D integrated (x, y, t) denotes the integrated geospatial database, E expansion (x, y, t) denotes the expansion situation, O status (x, y, t) denotes the operation status, θ i denotes the weight coefficient of the i-th feature variable, η, ζ, ξ respectively denote the weight coefficients of the integrated geospatial database, the expansion situation, and the operation status in the prediction model, and t denotes the prediction time. The influences of various feature variables and data sources are integrated through weighted summation to obtain the prediction results at different times.

[0092] In the first specific application, based on the selected feature variables F i in combination with the integrated geospatial database D integrated , the expansion situation E expansion , and the operation status O status , a prediction model M predict is constructed, according to the formula M predict denotes the prediction model, F i denotes the i-th feature variable, θ i denotes the weight coefficient of the i-th feature variable, η, ζ, ξ respectively denote the weight coefficients of the integrated geospatial database, the expansion situation, and the operation status in the prediction model, and t denotes the prediction time. For example, by analyzing the data features and the influence degree on the prediction results, θ1=0.2, η=0.3, ζ=0.2, and ξ=0.3 are determined.

[0093] In the second specific application, multi-dimensional feature variables related to offshore photovoltaic power generation facilities and marine ecological environment are extracted from long-term sequence remote sensing data and auxiliary data such as marine ecological research reports, marine environmental monitoring data, etc. These variables include the construction progress of the facility, the impact indicators of the operation state on the ecological environment, and the change characteristics of the marine ecological environment itself (such as water quality parameter change trend, biological community diversity change, etc.). Based on these feature variables, a decision model is constructed to evaluate the feasibility and effectiveness of different ecological protection measures. For example, a cost-benefit analysis-based model is constructed to consider the implementation cost of ecological protection measures (such as the construction cost of artificial reefs, the engineering cost of adjusting the layout of the facility, etc.) and the expected benefits of improving the marine ecological environment (such as the degree of restoration of biodiversity, the improvement of water quality, etc.). Although there is no direct corresponding formula, the construction idea of the decision model is similar to that of the prediction model, considering the influence of multiple factors on the decision result, and through the decision model, various possible ecological protection measures are evaluated and compared to select the optimal protection measure scheme.

[0094] Based on the above embodiments, the step S105 is described in detail:

[0095] Specifically, the trained prediction model M predict The development trend of the facility is predicted in multiple scenarios, including facility size growth, technology update direction, power generation efficiency change, and impact on the surrounding marine ecological environment, i.e. P trend (t+Δt)=M predict (t+Δt)+κ·C progress (x,y,t)+v·E expansion (x,y,t)+ω·O status (x,y,t), where P trend (t+Δt) represents the development trend prediction result at time point t+Δt, M predict (t+Δt represents the output result of the prediction model, κ, v, ω respectively represent the weight coefficients of construction progress, expansion and operation state in the prediction result, and the prediction result is presented in the form of charts and reports to better understand and use the prediction result.

[0096] In the first specific application, the trained prediction model is used to predict the development trend of the facility in multiple scenarios. The prediction content includes facility size growth, technology update direction, power generation efficiency change, and impact on the surrounding marine ecological environment, according to the formula P trend (t+Δt)=M predict (t+Δt)+κ·C progress (x,y,t)+v·E expansion (x,y,t)+ω·O status(x, y, t), where P trend (t+Δt) represents the predicted trend at time point t+Δt, M predict (t+Δt represents the output of the prediction model, and κ, v, and ω represent the weighting coefficients of construction progress, expansion, and operational status in the prediction results, respectively. Finally, the prediction results are presented in the form of charts and reports. For example, for facility scale growth, a line graph can be drawn, with time on the horizontal axis and facility scale (such as installed capacity, land area, etc.) on the vertical axis; for technology update direction, a bar chart can be used to represent the possibility of different technologies in different scenarios; for changes in power generation efficiency, a curve can be drawn to show the trend of power generation efficiency over time in different scenarios; for the impact on the surrounding marine ecological environment, charts can be used to show changes in water quality parameters, changes in biodiversity index, etc.)

[0097] In the second specific application, a scientific sustainable development assessment indicator system is established, comprehensively considering both the power generation benefits and the impact on the marine ecological environment of offshore photovoltaic power generation facilities. This system includes indicators such as the ratio of energy output to ecological and environmental impact, the marine biodiversity retention rate, and the water quality compliance rate. Through long-term monitoring and analysis of these indicators, the sustainable development status of the project is assessed. Regular assessments of the project's sustainable development status are conducted, and adjustments and optimizations are made to the facility's operation and ecological protection measures based on the assessment results. For example, if the ratio of energy output to ecological and environmental impact is found to be continuously declining, it indicates that the project's sustainable development faces challenges, requiring further analysis of the reasons. These reasons may include excessive impact on the ecological environment during facility operation or a decline in power generation efficiency. Corresponding measures are then taken to address these issues, such as optimizing facility operation strategies and strengthening ecological protection measures. A reassessment is then conducted, forming a dynamic assessment and feedback mechanism to ensure the project achieves sustainable development.

[0098] Based on the above embodiments, step S105 further includes:

[0099] • Perform anomaly detection on the target prediction results based on a preset threshold;

[0100] Specifically, by setting a threshold T threshold Detection prediction result P trend The abnormal situation in (t+Δt), i.e. Among them W alert (t+Δt) represents the alarm state at time point t+Δt, P trend (t+Δt) represents the predicted trend at time point t+Δt. This is an anomaly detection function.

[0101] • When an anomaly is detected, trigger an alarm:

[0102] Specifically, for the alarm state When the prediction result is greater than a set threshold, it is marked as abnormal, otherwise it is marked as normal, that is, when an abnormal situation is detected, W alert (t+Δt) = 1, an automatic alarm is issued to inform the relevant personnel.

[0103] In another specific embodiment, the alarm is divided into different levels according to the severity of the anomaly, specifically:

[0104] • Determine the anomaly degree score S severity of the detected anomaly.

[0105] • Based on the anomaly degree score, determine the alarm level corresponding to the detected anomaly according to a preset severity threshold; trigger an alarm of the corresponding type according to the alarm level.

[0106] Specifically, according to the severity S severity of the anomaly, the alarm is divided into different levels, that is, Where W level (t+Δt) represents the alarm level at time point t+Δt, represents a warning level division function, and the discriminant of the warning level division function is: Where S1 and S2 are severity thresholds for dividing different alarm levels, and for W level (t+Δt) = 1, it is a low-level warning, for W level (t+Δt) = 2, it is a medium-level warning, and for W level (t+Δt) = 3, it is a high-level warning.

[0107] As Figure 2 shown, Figure 2 is an implementation process framework diagram of a marine photovoltaic power generation facility development trend prediction method provided by an embodiment of the present application.

[0108] In summary, in the first specific application, the application process of the offshore photovoltaic power generation facility development trend prediction method in the offshore photovoltaic power plant construction and monitoring scene is demonstrated in detail, which effectively improves the management level of the offshore photovoltaic power generation facility, ensures the smooth progress and long-term stable operation of the project, and also provides strong technical support for the sustainable development of the offshore photovoltaic power generation industry. In the second specific application, the comprehensive application of the offshore photovoltaic power generation facility development trend prediction method in the coordinated development of offshore photovoltaic power generation and marine ecological protection is demonstrated. Through effective integration of multi-source remote sensing data and spatio-temporal dynamic analysis, accurate monitoring and evaluation of the ecological environmental impact during the construction and operation of the facility are realized. Based on these analysis results, a decision-making model is constructed to develop ecological protection measures, and through a sustainable development evaluation index system and feedback mechanism, the development strategy of the project is continuously optimized to ensure that offshore photovoltaic power generation and marine ecological protection can progress together and achieve the goal of sustainable development. This method provides strong technical support and practical guidance for the offshore photovoltaic power generation industry in the field of ecological environmental protection.

[0109] The present application realizes comprehensive dynamic monitoring of the whole process from construction to operation of offshore photovoltaic power generation facilities by introducing long time series remote sensing data and spatio-temporal dynamic analysis method, which not only can accurately capture the subtle changes in the construction process, but also can continuously monitor the expansion and operation status of the facilities, so as to timely discover and warn potential problems. This comprehensive dynamic monitoring capability helps to improve the management level of the facilities, reduce maintenance costs, and ensure the long-term stable operation of the facilities.

[0110] The prediction model constructed based on multi-dimensional feature variables can accurately predict the development trend of the facilities in the future period of time. Through multi-scenario prediction, the facility scale growth, technology update direction, power generation efficiency change and influence on the surrounding marine ecological environment can be evaluated. The prediction results are presented in the form of intuitive charts and reports, providing scientific basis for decision makers. At the same time, the early warning mechanism can automatically issue an alarm when an abnormal situation is detected, helping relevant personnel to take timely measures, thereby improving the accuracy and response speed of decision-making.

[0111] The embodiment of the present application provides a kind of offshore photovoltaic power generation facility development trend prediction device;Specific device can include:

[0112] Data collection and fusion module is used to collect multi-source remote sensing data, and fuse geographic information data to establish geographic spatial database;

[0113] Spatio-temporal dynamic analysis module is used to carry out spatio-temporal dynamic analysis on offshore photovoltaic power generation facility based on the geographic spatial database, and the spatio-temporal dynamic analysis includes construction phase analysis, expansion phase analysis and operation state analysis, to obtain construction progress, expansion and operation state;

[0114] a multi-dimensional feature extraction module configured to extract multi-dimensional feature variables related to the development trend of the offshore photovoltaic power generation facility from the long-time series of remote sensing data and auxiliary data based on the geospatial database;

[0115] a prediction model construction module configured to construct a prediction model based on the multi-dimensional feature variables in combination with the geospatial database, the expansion condition and the operation state;

[0116] a prediction result output module configured to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility by using the trained prediction model, and integrate the prediction output result with the construction progress, the expansion condition and the operation state to obtain a target prediction result.

[0117] The offshore photovoltaic power generation facility development trend prediction device of the embodiment is configured to implement the offshore photovoltaic power generation facility development trend prediction method described above, and thus the specific embodiments of the offshore photovoltaic power generation facility development trend prediction device can refer to the description of the embodiment part of the offshore photovoltaic power generation facility development trend prediction method, for example, the data collection and fusion module, the spatio-temporal dynamic analysis module, the multi-dimensional feature extraction module, the prediction model construction module, and the prediction result output module are respectively configured to implement steps S101, S102, S103, S104 and S105 of the offshore photovoltaic power generation facility development trend prediction method described above, and thus the specific embodiments can refer to the description of the corresponding embodiment part, which will not be described herein again.

[0118] The embodiment of the present application further provides an offshore photovoltaic power generation facility development trend prediction device, which comprises a memory configured to store a computer program and a processor configured to implement the steps of the offshore photovoltaic power generation facility development trend prediction method described above when executing the computer program.

[0119] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is configured to implement the steps of the offshore photovoltaic power generation facility development trend prediction method described above when executed by a processor.

[0120] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0121] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0122] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0123] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0124] Obviously, the above-described embodiments are only examples for clarity of description and are not limiting on the embodiments. Based on the above description, one of ordinary skill in the art can further make other different forms of changes or modifications. Here, all the embodiments are not required to be enumerated, and the obvious changes or modifications derived therefrom are still within the protection scope of the present application.

Claims

1. A method of predicting the development trend of a marine photovoltaic power generation facility, characterized by, The method comprises the following steps: collecting multi-source remote sensing data and fusing geographic information data to establish a geographic spatial database; extracting feature information of remote sensing images at different stages of facility construction based on the geographic spatial database, and analyzing the differences between actual construction and design planning data according to the feature information to evaluate the construction progress, comparing remote sensing images at different stages based on the geographic spatial database, and combining historical expansion conditions to determine the direction, range and speed of future expansion to obtain future expansion conditions, extracting spectral feature information of remote sensing images at different stages of facility operation based on the geographic spatial database, and monitoring the surface state of the facility photovoltaic panel according to the spectral feature information, combining the historical facility operation state to determine the change trend of the facility power generation efficiency with time, thereby determining the future facility operation state; based on the geographic spatial database, extracting multi-dimensional feature variables related to the development trend of offshore photovoltaic power generation facilities from long-time sequence remote sensing data and auxiliary data; Based on the multi-dimensional characteristic variable, a prediction model is constructed in combination with the geographic space database, the expansion condition and the operation state, that is wherein M predict represents the prediction model, F i represents the i-th characteristic variable, D integrated (x, y, t) represents the integrated geographic space database, E expansion (x, y, t) represents the expansion condition, O status (x, y, t) represents the operation state, θ i represents the weight coefficient of the i-th characteristic variable, η, ζ and ξ respectively represent the weight coefficients of the integrated geographic space database, the expansion condition and the operation state in the prediction model, t represents the prediction time, and p is the total number of characteristic variables. The trained prediction model is used to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility, i.e., P trend (t+Δt) = M predict (t+Δt) + κ·C progress (x,y,t) + v·E expansion (x,y,t) + ω·O status (x,y,t), wherein P trend (t+Δt) represents the development trend prediction result at the time point t+Δt, M predict (t+Δt) represents the output result of the prediction model, κ, v, and ω respectively represent the weight coefficients of the construction progress, the expansion situation, and the operation state in the prediction result, and the prediction output result is integrated with the construction progress, the expansion situation, and the operation state to obtain a target prediction result.

2. The method according to claim 1, wherein the collection of multi-source remote sensing data and the fusion of geographic information data to establish a geographic spatial database comprises: collecting multi-source remote sensing data from different remote sensing platforms and preprocessing the multi-source remote sensing data; fuse the preprocessed multi-source remote sensing data with geographic information data to establish a geographic spatial database.

3. The method according to claim 1, wherein The spatio-temporal dynamic analysis of offshore photovoltaic power generation facilities based on the geographic spatial database comprises construction stage analysis, expansion stage analysis and operation state analysis, and after obtaining the construction progress, expansion conditions and operation state, it further comprises: According to the construction progress, combined with marine engineering knowledge and on-site construction conditions, analyze the construction deviation reasons, and output improvement suggestions according to the construction deviation reasons; According to the expansion conditions, combined with marine environmental factors and energy demand changes, analyze the rationality of the expansion scheme, and use spatial analysis technology to analyze the mutual relationship between the expanded facilities and the surrounding marine ecological environment, and adjust the expansion scheme according to the analysis results; combined with the operation conditions, monitor the changes of marine ecological environment around the facility, and analyze the real-time monitoring results, and output the corresponding improvement measures of facility construction or operation.

4. The method according to claim 1, wherein The construction of a prediction model based on the multi-dimensional feature variables, combined with the geographic spatial database, the expansion conditions and the operation state comprises: select and reduce the multi-dimensional feature variables; based on the selected multiple feature variables, combined with the geographic spatial database, the expansion conditions and the operation state, a prediction model is constructed.

5. The method according to claim 1, wherein After integrating the prediction output results with the construction progress, the expansion conditions and the operation state to obtain the target prediction results, the method further comprises the following steps: According to the preset threshold, the target prediction result is detected for anomaly; when detecting an anomaly, triggering an alarm.

6. The method according to claim 1, wherein When detecting an anomaly, triggering an alarm comprises: determine the anomaly degree score of the detected anomaly; based on the anomaly degree score, according to the preset severity threshold, determine the alarm level corresponding to the detected anomaly; According to the alarm level, a corresponding type of alarm is triggered.

7. An offshore photovoltaic power plant development trend prediction device characterized by comprising: The method comprises the steps of: a data collection and fusion module for collecting multi-source remote sensing data and fusing geographic information data to establish a geospatial database; a space-time dynamic analysis module for extracting feature information of remote sensing images at different periods in the facility construction process based on the geospatial database, analyzing the difference between the actual construction and the design planning data according to the feature information, and evaluating the construction progress, based on the geospatial database, comparing remote sensing images at different periods and combining historical expansion conditions to determine the direction, range and speed of future expansion, and obtaining future expansion conditions, based on the geospatial database, extracting spectral feature information of remote sensing images at different periods in the facility operation process, monitoring the facility photovoltaic panel surface state according to the spectral feature information, combining the historical facility operation state, determining the change trend of the facility power generation efficiency with time, and determining the future facility operation state; a multi-dimensional feature extraction module for extracting multi-dimensional feature variables related to the development trend of the offshore photovoltaic power generation facility from long-time sequence remote sensing data and auxiliary data based on the geospatial database. a prediction model construction module, configured to construct a prediction model based on the multi-dimensional feature variables, in combination with the geospatial database, the expansion condition and the operation state, i.e. wherein M predict represents the prediction model, F i represents the i-th feature variable, D integrated (x, y, t) represents the integrated geospatial database, E expansion (x, y, t) represents the expansion condition, O status (x, y, t) represents the operation state, θ i represents the weight coefficient of the i-th feature variable, η, ζ and ξ respectively represent the weight coefficients of the integrated geospatial database, the expansion condition and the operation state in the prediction model, t represents the prediction time, and p is the total number of feature variables. a prediction result output module configured to utilize the trained prediction model to perform multi-scenario prediction on the development trend of the offshore photovoltaic power generation facility, i.e., P trend (t+Δt) = M predict (t+Δt) + k·C progress (x,y,t) + v·E expansion (x,y,t) + ω·O status (x,y,t), wherein P trend (t+Δt) represents the development trend prediction result at the time point t+Δt, M predict (t+Δt) represents the output result of the prediction model, k, v, and ω represent the weight coefficients of the construction progress, the expansion situation, and the operation state in the prediction result, respectively, and the prediction output result is integrated with the construction progress, the expansion situation, and the operation state to obtain a target prediction result.

8. An offshore photovoltaic power plant development trend prediction device characterized by comprising: The method comprises the steps of: a memory for storing a computer program; a processor for executing the computer program to realize the steps of the offshore photovoltaic power generation facility development trend prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the steps of the offshore photovoltaic power generation facility development trend prediction method according to any one of claims 1 to 6.

Images