A photovoltaic power station alarm management system based on drones

Abstract

The present invention provides a photovoltaic power station alarm management system based on unmanned aerial vehicles, which belongs to the field of data processing technology, and includes: a monitoring module: the unmanned aerial vehicle inspects the photovoltaic power station based on the inspection route, and collects the real-time operation data of the photovoltaic power station; an analysis module: the real-time operation data is transmitted to the system center, the real-time operation data is analyzed, and a real-time operation report of the photovoltaic power station is generated; an alarm module: the real-time fault data of the photovoltaic power station is determined based on the real-time operation report, and an alarm is issued based on the real-time fault data and the alarm strategy; a display module: the historical alarm data of the photovoltaic power station within a specified time period stored in the system center is extracted, and a visual display is performed. The inspection efficiency and data accuracy can be improved, the fault trend and potential problems can be quickly identified, early warnings can be issued in time, maintenance decisions can be optimized, the fault response speed and management efficiency can be improved, and the intelligence and safety of photovoltaic power stations can be promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a photovoltaic power station alarm management system based on an unmanned aerial vehicle. Background Art

[0002] With the rapid development of renewable energy, photovoltaic power stations have been widely used around the world. However, the efficient operation of photovoltaic power stations depends on timely fault monitoring and maintenance. The traditional manual inspection method has problems such as low efficiency, slow response, and high risk of missed inspection.

[0003] Existing monitoring technologies mainly rely on static sensors and manual inspections, lacking the ability to collect real-time data and conduct dynamic analysis. In addition, fault warning and alarm mechanisms are often based on experience and are difficult to adapt to complex and changing operating environments.

[0004] Therefore, the present invention provides a photovoltaic power station alarm management system based on unmanned aerial vehicle. Summary of the invention

[0005] The present invention provides a photovoltaic power station alarm management system based on unmanned aerial vehicles. The unmanned aerial vehicles are used to patrol the photovoltaic power station, collect real-time operation data, determine the real-time operation report of the photovoltaic power station and real-time fault data, issue alarms according to the real-time fault data and alarm strategies, and visualize the alarm data within a specified time period. This can improve patrol efficiency and data accuracy, quickly identify fault trends and potential problems, issue early warnings in a timely manner, optimize maintenance decisions, improve fault response speed and management efficiency, and promote the intelligence and safety of photovoltaic power stations.

[0006] The present invention provides a photovoltaic power station alarm management system based on an unmanned aerial vehicle, comprising:

[0007] Monitoring module: The drone inspects the photovoltaic power station based on the inspection route and collects real-time operation data of the photovoltaic power station;

[0008] Analysis module: transmits real-time operation data to the system center, analyzes the real-time operation data, and generates a real-time operation report of the photovoltaic power station;

[0009] Alarm module: determines the real-time fault data of the photovoltaic power station based on the real-time operation report, and issues an alarm based on the real-time fault data and alarm strategy;

[0010] Display module: extracts the historical alarm data of the photovoltaic power station within a specified time period stored in the system center and displays it visually.

[0011] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the monitoring module includes:

[0012] First patrol route unit: The system center determines the first patrol route of the drone based on the geographic information of the photovoltaic power plant, the layout of all components, and the location of all equipment;

[0013] Second patrol route unit: the system center obtains patrol environment data in real time, optimizes the first patrol route based on the patrol environment data, and determines the second patrol route;

[0014] Real-time patrol parameter unit: extracts all component requirements and all equipment requirements of the photovoltaic power station, determines patrol demand data, and configures the real-time patrol parameters of the drone based on patrol environment data and patrol demand data;

[0015] Inspection unit: uploads the second inspection route to the navigation system of the UAV, and the UAV inspects the photovoltaic power station based on the second inspection route and real-time inspection parameters.

[0016] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the monitoring module further includes:

[0017] Flight route unit: locates the drone in real time based on GPS and ground station signals to determine the flight route of the drone;

[0018] Deviation unit: if the flight route deviates from the second patrol route, the adjustment patrol parameters of the drone are determined based on the deviation route that the drone has deviated from and the second patrol route, and the drone continues to patrol the photovoltaic power station based on the second patrol route and the adjusted patrol parameters;

[0019] Real-time operation data unit: The UAV collects the real-time operation sub-data of each component or device of the photovoltaic power station along the second patrol route, and determines the real-time operation data based on the real-time operation sub-data of all components and all devices.

[0020] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the analysis module includes:

[0021] Transmission unit: The drone transmits the collected real-time operation data to the system center, and the system center pre-processes the received real-time operation data;

[0022] A first classification unit: classifies the real-time operation data based on the device characteristics or component characteristics of each real-time operation sub-data in the preprocessed real-time operation data, determines that all the real-time operation sub-data with the same device characteristics or component characteristics are first sub-categories, and determines the first category based on all the first sub-categories;

[0023] Key operation sub-data unit: extracting key operation sub-data of each real-time operation sub-data in the first sub-category based on the device characteristics or component characteristics of each first sub-category in the first category;

[0024] An operation feature vector unit: extracting features of key operation sub-data of each real-time operation sub-data in the first sub-category, and determining an operation feature vector of a device or component corresponding to each real-time operation sub-data;

[0025] An operation characteristic matrix unit: determining an operation characteristic matrix of each first subcategory based on the operation characteristic vectors of all real-time operation sub-data in each first subcategory respectively;

[0026] A feature range unit: determining a feature range of each feature in the first subcategory based on the running feature matrix of each first subcategory;

[0027] Anomaly scoring unit: calculating anomaly scores for each device or component in the first subcategory based on an operation feature vector of the real-time operation subdata of each device or component in the first subcategory and a feature range of all features in the first subcategory;

[0028] Real-time operation reporting unit: generates a real-time operation report based on the real-time operation sub-data, operation feature vectors, anomaly scores of all devices or components in all first sub-categories in the first category, and feature ranges of all features in all first sub-categories in the first category.

[0029] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the characteristic range unit includes:

[0030] The lower range limit and upper range limit subunits are configured to determine the lower range limit and the upper range limit of each feature in the first subcategory based on the operating feature matrix of each first subcategory;

[0031]

[0032] MD j =Median(|fe ij -Med(M j )|);

[0033] Among them, FL j represents the lower limit of the range of the jth feature in the first subcategory, FU j represents the upper limit of the range of the jth feature in the first subcategory, represents the α quantile of all devices or components in the first subcategory on the jth feature, represents the 1-α quantile of all devices or components in the first subcategory on the jth feature, MD j represents the absolute median deviation of the jth feature in the first subcategory, δ j represents the offset parameter of the jth feature in the first subcategory, γj represents the concentration parameter of the jth feature in the first subcategory, Med(M j ) represents the median of the jth feature in the first subcategory, ε1 j represents the first adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, ε2 j represents the second adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, fe ij represents the characteristic value of the i-th device or component in the first subcategory on the j-th characteristic, μ j represents the average value of the jth feature in the first subcategory, σ j represents the variance of the jth feature in the first subcategory, and N1 represents the number of devices or components in the first subcategory;

[0034] Feature range subunit: determines a feature range of each feature in the first subcategory based on a lower range limit and an upper range limit of each feature in the first subcategory.

[0035] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the abnormality scoring unit includes:

[0036] Among them, A ij represents the outlier value of the i-th device or component in the first subcategory on the j-th feature, represents the third adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, represents the fourth adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, S i represents the anomaly score of the i-th device or component in the first subcategory.

[0037] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the alarm module includes:

[0038] Extraction unit: extracts the real-time operation sub-data and abnormality scores of all devices or components whose abnormality scores are not 0 in each first sub-category of the real-time operation report, and determines the real-time fault data;

[0039] Alarm unit: determines the alarm level of each device or component in the real-time fault data based on the alarm strategy and the abnormal score of each device or component in the real-time fault data;

[0040] Maintenance unit: issues warnings based on the alarm levels of all devices in real-time fault data, and performs maintenance on faulty devices in the PV power station based on the alarms;

[0041] Real-time alarm data unit: determines real-time alarm data based on maintenance results and real-time fault data.

[0042] According to a photovoltaic power station alarm management system based on a drone provided by the present invention, the display module includes:

[0043] Statistical display unit: counts the number of alarms of each alarm level in the historical alarm data, and visualizes the statistical results of all alarm levels within a specified time period.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] By using drones to patrol photovoltaic power stations, collect real-time operation data, determine the real-time operation report and real-time fault data of the photovoltaic power station, issue alarms based on real-time fault data and alarm strategies, and visualize the alarm data within a specified time period, it can improve patrol efficiency and data accuracy, quickly identify fault trends and potential problems, issue early warnings in a timely manner, optimize maintenance decisions, improve fault response speed and management efficiency, and promote the intelligence and safety of photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 It is a structural schematic diagram of a photovoltaic power station alarm management system based on a drone provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Embodiment 1:

[0050] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, such as Figure 1 As shown, including:

[0051] Monitoring module: The drone inspects the photovoltaic power station based on the inspection route and collects real-time operation data of the photovoltaic power station;

[0052] Analysis module: transmits real-time operation data to the system center, analyzes the real-time operation data, and generates a real-time operation report of the photovoltaic power station;

[0053] Alarm module: determines the real-time fault data of the photovoltaic power station based on the real-time operation report, and issues an alarm based on the real-time fault data and alarm strategy;

[0054] Display module: extracts the historical alarm data of the photovoltaic power station within a specified time period stored in the system center and displays it visually.

[0055] In this embodiment, the drone performs regular inspections of the photovoltaic power station according to a preset inspection route and collects operating data of the photovoltaic power station in real time.

[0056] In this embodiment, the collected real-time operation data will be transmitted to the system center for data analysis to generate a real-time operation report of the photovoltaic power station.

[0057] In this embodiment, based on the real-time operation report, the alarm module determines the real-time fault data of the photovoltaic power station. If the system detects an abnormality (such as equipment failure or performance degradation), an alarm is issued to notify relevant personnel based on a preset alarm strategy.

[0058] In this embodiment, historical alarm data stored in the system center is extracted and visualized. Through charts and dashboards, managers can intuitively understand the alarm conditions in each time period, analyze fault trends and equipment performance, and formulate corresponding maintenance strategies.

[0059] In this embodiment, the drone is an unmanned aerial vehicle used for inspecting photovoltaic power stations and has automatic flight and data collection capabilities.

[0060] In this embodiment, the alarm strategy represents preset rules and conditions for determining when to issue an alarm.

[0061] The beneficial effects of the above technical solution are as follows: by patrolling the photovoltaic power station with drones, collecting real-time operation data, determining the real-time operation report and real-time fault data of the photovoltaic power station, issuing alarms based on the real-time fault data and alarm strategies, and visually displaying the alarm data within a specified time period, the inspection efficiency and data accuracy can be improved, fault trends and potential problems can be quickly identified, early warnings can be issued in a timely manner, maintenance decisions can be optimized, fault response speed and management efficiency can be improved, and the intelligence and safety of photovoltaic power stations can be promoted.

[0062] Embodiment 2:

[0063] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the monitoring module comprises:

[0064] First patrol route unit: The system center determines the first patrol route of the drone based on the geographic information of the photovoltaic power plant, the layout of all components, and the location of all equipment;

[0065] Second patrol route unit: the system center obtains patrol environment data in real time, optimizes the first patrol route based on the patrol environment data, and determines the second patrol route;

[0066] Real-time patrol parameter unit: extracts all component requirements and all equipment requirements of the photovoltaic power station, determines patrol demand data, and configures the real-time patrol parameters of the drone based on patrol environment data and patrol demand data;

[0067] Inspection unit: uploads the second inspection route to the navigation system of the UAV, and the UAV inspects the photovoltaic power station based on the second inspection route and real-time inspection parameters.

[0068] In this embodiment, the system center determines the initial patrol route of the drone based on the geographic information of the photovoltaic power station (such as terrain, buildings, etc.), the layout of all components (such as photovoltaic panels, inverters, etc.) and the location of each device. This process ensures that the drone can effectively cover the entire power station and collect necessary operating data.

[0069] In this embodiment, when conducting actual inspections, the system center obtains inspection environment data in real time, such as weather conditions, light intensity, wind speed, etc. These environmental data will affect the flight path and effect of the drone. Based on these data, the system will optimize the first inspection route and generate a second inspection route to ensure the efficiency and safety of the inspection.

[0070] In this embodiment, the requirements of all components and equipment of the photovoltaic power station are extracted, such as the operating temperature of each component or equipment, power generation efficiency monitoring, etc., to determine the patrol demand data. At the same time, combined with the patrol environment data, the patrol parameters of the drone (such as flight altitude, speed, shooting angle, etc.) are configured to meet the current patrol needs.

[0071] In this embodiment, the optimized second patrol route is uploaded to the navigation system of the UAV, and the UAV patrols the photovoltaic power station according to this route and real-time patrol parameters. This process ensures that the UAV can accurately complete the patrol task according to the predetermined route and collect the required data.

[0072] The beneficial effects of the above technical solution are as follows: the UAV can inspect the photovoltaic power station based on the inspection route, which can improve the flexibility and adaptability of the UAV inspection, ensure the accuracy and efficiency of the inspection, and realize intelligent and refined operation and maintenance management.

[0073] Embodiment 3:

[0074] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the monitoring module further includes:

[0075] Flight route unit: locates the drone in real time based on GPS and ground station signals to determine the flight route of the drone;

[0076] Deviation unit: if the flight route deviates from the second patrol route, the adjustment patrol parameters of the drone are determined based on the deviation route that the drone has deviated from and the second patrol route, and the drone continues to patrol the photovoltaic power station based on the second patrol route and the adjusted patrol parameters;

[0077] Real-time operation data unit: The UAV collects the real-time operation sub-data of each component or device of the photovoltaic power station along the second patrol route, and determines the real-time operation data based on the real-time operation sub-data of all components and all devices.

[0078] In this embodiment, the UAV is positioned in real time using GPS (Global Positioning System) and ground station signals. With precise location information, the system can dynamically determine the flight route of the UAV, ensuring that the UAV can fly according to the predetermined second patrol route.

[0079] In this embodiment, if the flight route of the UAV deviates from the second patrol route, the deviation unit will determine the degree of deviation based on the route from which the UAV has deviated and the second patrol route, and adjust the patrol parameters of the UAV (such as flight altitude, speed, etc.) to ensure that the UAV can correct the deviation and continue to effectively patrol the photovoltaic power station along the second patrol route.

[0080] In this embodiment, during the patrol process, the drone collects the operating sub-data (such as power generation, temperature, status, etc.) of each component or device of the photovoltaic power station in real time along the second patrol route, and the real-time operating sub-data of all components and devices are aggregated to generate complete real-time operating data.

[0081] The beneficial effects of the above technical solution are: collecting real-time operating data of photovoltaic power stations can ensure continuous and accurate monitoring, significantly improve patrol efficiency and data accuracy, enhance the system's early warning capability for potential faults, and provide strong support for the intelligent operation and maintenance of photovoltaic power stations.

[0082] Embodiment 4:

[0083] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the analysis module comprises:

[0084] Transmission unit: The drone transmits the collected real-time operation data to the system center, and the system center pre-processes the received real-time operation data;

[0085] A first classification unit: classifies the real-time operation data based on the device characteristics or component characteristics of each real-time operation sub-data in the preprocessed real-time operation data, determines that all the real-time operation sub-data with the same device characteristics or component characteristics are first sub-categories, and determines the first category based on all the first sub-categories;

[0086] Key operation sub-data unit: extracting key operation sub-data of each real-time operation sub-data in the first sub-category based on the device characteristics or component characteristics of each first sub-category in the first category;

[0087] An operation feature vector unit: extracting features of key operation sub-data of each real-time operation sub-data in the first sub-category, and determining an operation feature vector of a device or component corresponding to each real-time operation sub-data;

[0088] An operation characteristic matrix unit: determining an operation characteristic matrix of each first subcategory based on the operation characteristic vectors of all real-time operation sub-data in each first subcategory respectively;

[0089] A feature range unit: determining a feature range of each feature in the first subcategory based on the running feature matrix of each first subcategory;

[0090] Anomaly scoring unit: calculating anomaly scores for each device or component in the first subcategory based on an operation feature vector of the real-time operation subdata of each device or component in the first subcategory and a feature range of all features in the first subcategory;

[0091] Real-time operation reporting unit: generates a real-time operation report based on the real-time operation sub-data, operation feature vectors, anomaly scores of all devices or components in all first sub-categories in the first category, and feature ranges of all features in all first sub-categories in the first category.

[0092] In this embodiment, the real-time operation data collected by the drone during the patrol process will be transmitted to the system center via wireless signals. After receiving the data, the system center will perform preprocessing, such as data format conversion, error checking and preliminary analysis, in preparation for subsequent in-depth analysis.

[0093] In this embodiment, the preprocessed real-time operation data is classified, and based on the device characteristics or component characteristics of each real-time operation sub-data, the sub-data with similar characteristics are classified into the same category to form a first sub-category, which can be a photovoltaic component sub-category, an inverter sub-category, a distribution cabinet sub-category, a battery sub-category, etc.

[0094] In this embodiment, key operation sub-data are extracted from each first sub-category and are mainly screened based on device characteristics or component characteristics.

[0095] In this embodiment, feature extraction is performed on key operation sub-data of each real-time operation sub-data in the first sub-category to form a corresponding operation feature vector.

[0096] In this embodiment, based on the operation feature vectors of the real-time operation sub-data of all devices in each first sub-category, a corresponding operation feature matrix is ​​constructed, and the operation feature matrix includes the operation feature vectors of all devices or components in the corresponding first sub-category.

[0097] In this embodiment, based on the real-time operation data, operation feature vectors, abnormal scores and feature ranges of all first subcategories in the first category, a real-time operation report is generated to summarize the overall operation status and problems of the power station.

[0098] The beneficial effects of the above technical solution are: transmitting real-time operation data to the system center, analyzing real-time operation data, and generating real-time operation reports of photovoltaic power stations, which can realize accurate monitoring and early warning of potential faults, improve decision-making efficiency and safety, and promote the process of intelligent management.

[0099] Embodiment 5:

[0100] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the characteristic range unit includes:

[0101] The lower range limit and upper range limit subunits are configured to determine the lower range limit and the upper range limit of each feature in the first subcategory based on the operating feature matrix of each first subcategory;

[0102]

[0103] MD j =Median(|fe ij -Med(M j )|);

[0104] Among them, FL j represents the lower limit of the range of the jth feature in the first subcategory, FU j represents the upper limit of the range of the jth feature in the first subcategory, represents the α quantile of all devices or components in the first subcategory on the jth feature, represents the 1-α quantile of all devices or components in the first subcategory on the jth feature, MD j represents the absolute median deviation of the jth feature in the first subcategory, δ j represents the offset parameter of the jth feature in the first subcategory, γ j represents the concentration parameter of the jth feature in the first subcategory, Med(M j) represents the median of the jth feature in the first subcategory, ε1 j represents the first adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, ε2 j represents the second adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, fe ij represents the characteristic value of the i-th device or component in the first subcategory on the j-th characteristic, μ j represents the average value of the jth feature in the first subcategory, σ j represents the variance of the jth feature in the first subcategory, and N1 represents the number of devices or components in the first subcategory;

[0105] Feature range subunit: determines a feature range of each feature in the first subcategory based on a lower range limit and an upper range limit of each feature in the first subcategory.

[0106] In this embodiment, the α quantile indicates that the proportion of values ​​smaller than this number in the data to the total data is α, and the α quantile is the value of the data at the α percentage position. Assume that there is a data set, if Q α represents the α quantile of this set, then α×100% of the values ​​in the data are less than or equal to Q α .

[0107] In this embodiment, the 50% quantile, i.e., the 0.5 quantile, corresponds to the median of the data set, which means that 50% of the data are less than or equal to this value; the 25% quantile (i.e., the one-quarter quantile, or the 0.25 quantile) indicates that 25% of the data are less than or equal to this value; and the 95% quantile indicates that 95% of the values ​​in the data are less than or equal to this value.

[0108] In this embodiment, the 1-α quantile indicates that the proportion of values ​​smaller than this number in the data to the total data is 1-α, and the 1-α quantile is the value of the data at the 1-α percentage position.

[0109] In this embodiment, the lower limit and the upper limit of the range of each feature in the first subcategory are determined according to the feature value of each feature of all devices or components in the operation feature matrix of the first subcategory.

[0110] In this embodiment, the feature range of each feature in the first subcategory is determined for use in calculating the abnormality score of each device or component in the first subcategory.

[0111] The beneficial effect of the above technical solution is: based on the operating feature matrix of each first subcategory, the feature range of each feature in the first subcategory is determined, thereby improving the data basis for calculating the abnormal score of each device or component in the first subcategory.

[0112] Embodiment 6:

[0113] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the abnormality scoring unit comprises:

[0114] Among them, A ij represents the outlier value of the i-th device or component in the first subcategory on the j-th feature, represents the third adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, represents the fourth adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, S i represents the anomaly score of the i-th device or component in the first subcategory.

[0115] In this embodiment, if the feature value fe of the i-th device or component in the first subcategory on the j-th feature ij Between the corresponding lower and upper limits of the range, its outlier value is 0; if the feature value fe of the i-th device or component in the first subcategory on the j-th feature ij is less than the corresponding range lower limit, according to Calculate its abnormal value. If the feature value fe of the i-th device or component in the first subcategory on the j-th feature ij Greater than the upper limit of the corresponding range, according to Calculate its outlier value.

[0116] The beneficial effects of the above technical solution are as follows: based on the operation feature vector of the real-time operation sub-data of each device or component in the first subcategory and the feature range of all features in the first subcategory, the abnormal score of each device or component in the first subcategory is calculated, thereby improving the data basis for generating real-time operation reports of photovoltaic power stations, thereby improving decision-making efficiency and safety and promoting the process of intelligent management.

[0117] Embodiment 7:

[0118] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the alarm module comprises:

[0119] Extraction unit: extracts the real-time operation sub-data and abnormality scores of all devices or components whose abnormality scores are not 0 in each first sub-category of the real-time operation report, and determines the real-time fault data;

[0120] Alarm unit: determines the alarm level of each device or component in the real-time fault data based on the alarm strategy and the abnormal score of each device or component in the real-time fault data;

[0121] Maintenance unit: issues warnings based on the alarm levels of all devices in real-time fault data, and performs maintenance on faulty devices in the PV power station based on the alarms;

[0122] Real-time alarm data unit: determines real-time alarm data based on maintenance results and real-time fault data.

[0123] In this embodiment, the real-time operation sub-data and corresponding abnormality scores of all devices or components in the first subcategory whose abnormality scores are not 0 are extracted from the real-time operation report. These data are called real-time fault data, which represent devices or components with failure risks. The extracted fault data will be used in subsequent alarm and maintenance processes.

[0124] In this embodiment, based on the set alarm strategy, the abnormal score of each device or component in the real-time fault data is analyzed to determine its alarm level. The alarm level is usually divided into multiple levels, such as normal, attention, warning, danger, etc., and the specific division is based on the size of the abnormal score.

[0125] In this embodiment, according to the alarm levels of all devices in the real-time fault data, the maintenance unit is responsible for issuing corresponding warnings and repairing the faulty equipment in the photovoltaic power station. This process may include arranging technicians to conduct on-site inspections, repairs or replacement of equipment to ensure the normal operation of the power station.

[0126] In this embodiment, real-time alarm data is generated based on the inspection results and real-time fault data. This data records the specific information and processing status of the fault, and can provide a basis for subsequent analysis and decision-making.

[0127] The beneficial effects of the above technical solution are: determining the real-time fault data of the photovoltaic power station based on the real-time operation report, and issuing an alarm based on the real-time fault data and alarm strategy, which can timely identify faults, ensure rapid communication of fault information, and improve the operation and maintenance efficiency of the photovoltaic power station.

[0128] Embodiment 8:

[0129] The embodiment of the present invention provides a photovoltaic power station alarm management system based on a drone, wherein the display module comprises:

[0130] Statistical display unit: counts the number of alarms of each alarm level in the historical alarm data, and visualizes the statistical results of all alarm levels within a specified time period.

[0131] In this embodiment, historical alarm data is extracted from the system and the number of alarms at each alarm level is counted. The unit can set a specific time period (such as one day, one week or one month) according to management needs, and present the statistical results of all alarm levels within the period in a visual form, usually using charts (such as bar charts, line charts, etc.) to show the statistical data of each alarm level.

[0132] The beneficial effects of the above technical solution are: extracting the historical alarm data of the photovoltaic power station within a specified time period stored in the system center and displaying it visually can help managers identify potential problems and trends, optimize operation and maintenance strategies, and enhance the overall safety and reliability of the system.

[0133] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0134] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic power station alarm management system based on drones, characterized in that: include: Monitoring module: The drone inspects the photovoltaic power station based on the inspection route and collects real-time operation data of the photovoltaic power station; Analysis module: transmits real-time operation data to the system center, analyzes the real-time operation data, and generates a real-time operation report of the photovoltaic power station; Alarm module: determines the real-time fault data of the photovoltaic power station based on the real-time operation report, and issues an alarm based on the real-time fault data and alarm strategy; Display module: extracts the historical alarm data of the photovoltaic power station within a specified time period stored in the system center and displays it visually.

2. The photovoltaic power station alarm management system based on drone according to claim 1 is characterized in that: The monitoring module comprises: First patrol route unit: The system center determines the first patrol route of the drone based on the geographic information of the PV power plant, the layout of all components, and the location of all equipment; Second patrol route unit: the system center obtains patrol environment data in real time, optimizes the first patrol route based on the patrol environment data, and determines the second patrol route; Real-time patrol parameter unit: extracts all component requirements and all equipment requirements of the photovoltaic power station, determines patrol demand data, and configures the real-time patrol parameters of the drone based on patrol environment data and patrol demand data; Inspection unit: uploads the second inspection route to the navigation system of the UAV, and the UAV inspects the photovoltaic power station based on the second inspection route and real-time inspection parameters.

3. The photovoltaic power station alarm management system based on drone according to claim 2 is characterized in that: The monitoring module further includes: Flight route unit: locates the drone in real time based on GPS and ground station signals to determine the flight route of the drone; Deviation unit: if the flight route deviates from the second patrol route, the adjustment patrol parameters of the drone are determined based on the deviation route that the drone has deviated from and the second patrol route, and the drone continues to patrol the photovoltaic power station based on the second patrol route and the adjusted patrol parameters; Real-time operation data unit: The UAV collects the real-time operation sub-data of each component or device of the photovoltaic power station along the second patrol route, and determines the real-time operation data based on the real-time operation sub-data of all components and all devices.

4. The photovoltaic power station alarm management system based on drone according to claim 1 is characterized in that: The analysis module comprises: Transmission unit: The drone transmits the collected real-time operation data to the system center, and the system center pre-processes the received real-time operation data; A first classification unit: classifies the real-time operation data based on the device characteristics or component characteristics of each real-time operation sub-data in the preprocessed real-time operation data, determines that all the real-time operation sub-data with the same device characteristics or component characteristics are first sub-categories, and determines the first category based on all the first sub-categories; Key operation sub-data unit: extracting key operation sub-data of each real-time operation sub-data in the first sub-category based on the device characteristics or component characteristics of each first sub-category in the first category; An operation feature vector unit: extracting features of key operation sub-data of each real-time operation sub-data in the first sub-category, and determining an operation feature vector of a device or component corresponding to each real-time operation sub-data; An operation characteristic matrix unit: determining an operation characteristic matrix of each first subcategory based on the operation characteristic vectors of all real-time operation sub-data in each first subcategory respectively; A feature range unit: determining a feature range of each feature in the first subcategory based on the running feature matrix of each first subcategory; Anomaly scoring unit: calculating anomaly scores for each device or component in the first subcategory based on an operation feature vector of the real-time operation subdata of each device or component in the first subcategory and a feature range of all features in the first subcategory; Real-time operation reporting unit: generates a real-time operation report based on the real-time operation sub-data, operation feature vectors, anomaly scores of all devices or components in all first sub-categories in the first category, and feature ranges of all features in all first sub-categories in the first category.

5. The photovoltaic power station alarm management system based on drone according to claim 4 is characterized in that: The feature range unit comprises: The lower range limit and upper range limit subunits are configured to determine the lower range limit and the upper range limit of each feature in the first subcategory based on the operating feature matrix of each first subcategory; MD j =Median(|fe ij -Med(M j )|); Among them, FL j represents the lower limit of the range of the jth feature in the first subcategory, FU j represents the upper limit of the range of the jth feature in the first subcategory, represents the α quantile of all devices or components in the first subcategory on the jth feature, represents the 1-α quantile of all devices or components in the first subcategory on the jth feature, MD j represents the absolute median deviation of the jth feature in the first subcategory, δ j represents the offset parameter of the jth feature in the first subcategory, γ j represents the concentration parameter of the jth feature in the first subcategory, Med(M j ) represents the median of the jth feature in the first subcategory, ε1 j represents the first adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, ε2 j represents the second adjustment coefficient of the jth feature in the first subcategory based on the absolute median deviation, fe ij represents the characteristic value of the i-th device or component in the first subcategory on the j-th characteristic, μ j represents the average value of the jth feature in the first subcategory, σ j represents the variance of the jth feature in the first subcategory, and N1 represents the number of devices or components in the first subcategory; Feature range subunit: determines a feature range of each feature in the first subcategory based on a lower range limit and an upper range limit of each feature in the first subcategory.

6. The photovoltaic power station alarm management system based on drone according to claim 4 is characterized in that: The abnormal score unit comprises: Among them, A ij represents the outlier value of the i-th device or component in the first subcategory on the j-th feature, represents the third adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, represents the fourth adjustment coefficient of the i-th device or component in the first subcategory on the j-th feature, S i represents the anomaly score of the i-th device or component in the first subcategory.

7. The photovoltaic power station alarm management system based on drone according to claim 1 is characterized in that: The alarm module comprises: Extraction unit: extracts the real-time operation sub-data and abnormality scores of all devices or components whose abnormality scores are not 0 in each first sub-category of the real-time operation report, and determines the real-time fault data; Alarm unit: determines the alarm level of each device or component in the real-time fault data based on the alarm strategy and the abnormal score of each device or component in the real-time fault data; Maintenance unit: issues warnings based on the alarm levels of all devices in real-time fault data, and performs maintenance on faulty devices in the PV power station based on the alarms; Real-time alarm data unit: determines real-time alarm data based on maintenance results and real-time fault data.

8. The photovoltaic power station alarm management system based on drone according to claim 1 is characterized in that: The display module comprises: Statistical display unit: counts the number of alarms of each alarm level in the historical alarm data, and visualizes the statistical results of all alarm levels within a specified time period.

Images