An intelligent analysis system and method applied to a photovoltaic power station tool

Abstract

This invention discloses an intelligent analysis system and method for photovoltaic power plant tools, relating to the field of photovoltaic power plant operation and maintenance technology. The method includes: collecting operational datasets of photovoltaic equipment in a target photovoltaic power plant; analyzing dynamic health baselines and theoretical performance potential lines based on the operational datasets to obtain absolute health and real-time potential difference; dividing the photovoltaic equipment into multiple analysis regions according to a preset topological relationship and analyzing environmental risk entropy; constructing a three-dimensional dynamic evaluation coordinate system based on the absolute health, real-time potential difference, and environmental risk entropy of the photovoltaic equipment's region; generating a first operation and maintenance instruction when the photovoltaic equipment is mapped to a first quadrant; generating a second operation and maintenance instruction for the associated photovoltaic equipment when the photovoltaic equipment is mapped to a second quadrant; and executing a third operation and maintenance instruction to reduce the monitoring frequency when the photovoltaic equipment is mapped to a third quadrant.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant operation and maintenance technology, specifically an intelligent analysis system and method for photovoltaic power plant tools. Background Technology

[0002] As the scale of photovoltaic power plants continues to expand, traditional operation and maintenance methods are no longer sufficient to meet the demands for efficient and precise management. Current mainstream analysis methods often focus on a single dimension, relying on fixed thresholds for single-point alarms. This fails to distinguish between absolute performance degradation and temporary fluctuations caused by environmental factors or shading, resulting in a high false alarm rate. Simple horizontal comparisons struggle to identify "relative degradation" issues caused by different equipment aging rates—meaning equipment may still perform well but significantly lag behind similar healthy equipment. Furthermore, there is a lack of quantitative assessment of the operational correlations among equipment groups within the power plant. When an anomaly in one piece of equipment causes hidden interference to surrounding equipment (such as hot spots leading to localized temperature increases or mismatch losses), it is difficult to accurately pinpoint the source of the fault. These problems lead to passive and inefficient operation and maintenance work, making it impossible to classify and predict fault risks or optimize the allocation of monitoring resources for a large number of normally functioning devices. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent analysis system and method for photovoltaic power plant tools, in order to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent analysis system and method for photovoltaic power plant tools, the method comprising:

[0005] S100. Collect the operation dataset of photovoltaic equipment in the target photovoltaic power station. The operation dataset includes the output power, operating temperature and associated environmental irradiance data of the photovoltaic equipment. The output power is calculated by collecting the DC side voltage and current of the photovoltaic string. The association refers to matching the irradiance data collected by the meteorological monitoring point with the photovoltaic equipment operation data of the corresponding area through timestamps and geographical location.

[0006] S200. Based on the running dataset, analyze the dynamic health baseline and theoretical performance potential line of the photovoltaic equipment under standard environmental conditions to obtain the absolute health of the photovoltaic equipment and the real-time potential difference between the dynamic health baseline and the theoretical performance potential line.

[0007] S300. Divide the photovoltaic equipment into multiple analysis areas according to the preset topology relationship, and analyze the power generation consistency entropy value of the photovoltaic equipment in the analysis area at the same collection time, which is used as the environmental risk entropy of the analysis area.

[0008] S400. Based on the absolute health of the photovoltaic equipment, the real-time potential difference, and the environmental risk entropy of the area, a three-dimensional dynamic evaluation coordinate system is constructed; and the photovoltaic equipment is mapped to a preset operation and maintenance decision quadrant; the operation and maintenance decision quadrant includes a first quadrant, a second quadrant, and a third quadrant;

[0009] S500: When a photovoltaic device is mapped to the first quadrant, it is determined to be a high-priority fault risk target, and a first maintenance instruction is generated; when a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden fault risk in its analysis area, and a second maintenance instruction is generated for the associated photovoltaic device; when a photovoltaic device is mapped to the third quadrant, it is determined to be in a stable state, and a third maintenance instruction to reduce the monitoring frequency is executed.

[0010] According to the above scheme, step S200 includes:

[0011] S210. Based on the operating dataset, calculate the highest historical performance value of the photovoltaic equipment under preset standard environmental conditions within the current operating cycle, and use it as the dynamic health baseline of the photovoltaic equipment; the standard environmental conditions are standard test conditions; the highest historical performance value is obtained by correcting the actual performance data to the standard environmental conditions and selecting the maximum value within the current operating cycle.

[0012] S220. Based on the operational dataset, identify similar devices with the same model and orientation as the photovoltaic device and in a healthy operating state, and calculate the group optimal performance value of the similar devices under the standard environmental conditions as the theoretical performance potential line; the group optimal performance value is obtained by calculating the optimal quantile of the performance data of similar healthy devices.

[0013] S230. Determine the absolute health level as a percentage of the dynamic health baseline to the factory rated performance value of the photovoltaic equipment;

[0014] S240. Calculate the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference.

[0015] According to the above scheme, step S300 includes:

[0016] S310. The preset topology is the electrical connection hierarchy of photovoltaic equipment in the target photovoltaic power station. Photovoltaic equipment belonging to the same maximum power point tracker or the same combiner box is divided into the same analysis area.

[0017] S320. Select any synchronous sampling time, obtain the real-time output power of each photovoltaic device in the analysis area, and calculate the statistical entropy value, which characterizes the dispersion of the output power of all photovoltaic devices, based on the real-time output power of all photovoltaic devices at that time, as the power generation consistency entropy value; the statistical entropy value is calculated using the information entropy formula, and the input is the proportion of the output power of each device to the total output power of the area;

[0018] S330. The power generation consistency entropy value of the analysis area is used as the environmental risk entropy of all photovoltaic devices in the analysis area.

[0019] According to the above scheme, step S400 includes:

[0020] S410. Construct a three-dimensional dynamic evaluation coordinate system. The X-axis of the three-dimensional dynamic evaluation coordinate system represents the absolute health of the photovoltaic equipment, the Y-axis represents the real-time potential difference of the photovoltaic equipment, and the Z-axis represents the environmental risk entropy of the area where the photovoltaic equipment is located. The coordinate parameters are dynamically adjusted in real time as the photovoltaic equipment operation data is updated.

[0021] S420. The pre-defined quadrant division criteria for operation and maintenance decisions include: the first quadrant corresponds to the parameter combination of low absolute health, high real-time potential difference, and high environmental risk entropy; the second quadrant corresponds to the parameter combination of medium absolute health, low real-time potential difference, and high environmental risk entropy; and the third quadrant corresponds to the parameter combination of high absolute health, low real-time potential difference, and low environmental risk entropy.

[0022] S430. Extract the absolute health status, real-time potential difference, and environmental risk entropy data of the photovoltaic equipment and its region, and use them as coordinate points in the three-dimensional dynamic evaluation coordinates. Determine the operation and maintenance decision quadrant to which the coordinate point belongs according to the preset division conditions, and complete the mapping of the photovoltaic equipment to the operation and maintenance decision quadrant.

[0023] According to the above scheme, in the parameter combination corresponding to the first quadrant, the low absolute health is a value lower than the first health threshold, the high real-time potential difference is a value higher than the first potential difference threshold, and the high environmental risk entropy is a value higher than the first risk entropy threshold.

[0024] In the parameter combination corresponding to the second quadrant, the medium absolute health is a value between the lower limit and the upper limit of the second health threshold, the low real-time potential difference is a value lower than the second potential difference threshold, and the high environmental risk entropy is a value higher than the second risk entropy threshold.

[0025] In the parameter combination corresponding to the third quadrant, the high absolute health is a value higher than the third health threshold, the low real-time potential difference is a value lower than the third potential difference threshold, and the low environmental risk entropy is a value lower than the third risk entropy threshold.

[0026] According to the above scheme, step S500 includes:

[0027] S510. When a photovoltaic device is mapped to the first quadrant, it is determined to be a fault risk target with the highest maintenance urgency. Based on the operating data characteristics of the photovoltaic device, a matching analysis is performed through a pre-set fault mode library to generate the first maintenance instruction containing specific fault type predictions and priority maintenance suggestions.

[0028] S520. When a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden operational interference caused by other devices in the analysis area where it is located; based on the correlation between electrical connection relationship and operation data, at least one adjacent photovoltaic device with the highest correlation with the photovoltaic device is determined from the analysis area as the associated photovoltaic device, and a second operation and maintenance instruction for special inspection of the associated photovoltaic device is generated.

[0029] S530. When the photovoltaic equipment is mapped to the third quadrant, it is determined that it is in a healthy and stable operating state; according to the preset resource optimization strategy, the third operation and maintenance instruction, including extending the data acquisition interval and reducing the analysis and calculation frequency, is executed on the photovoltaic equipment.

[0030] An intelligent analysis system for photovoltaic power plant tools, comprising: a data acquisition module, an indicator analysis module, a situation assessment module, and an operation and maintenance decision-making module;

[0031] The data acquisition module is used to collect the operational dataset of photovoltaic equipment in the target photovoltaic power station. The operational dataset includes at least the output power and ambient irradiance data of the photovoltaic equipment. The index analysis module is used to analyze absolute health, real-time potential difference, and environmental risk entropy. The situation assessment module is used to construct a three-dimensional dynamic assessment coordinate system based on the absolute health, real-time potential difference, and environmental risk entropy output by the index analysis module, and to map the photovoltaic equipment to the corresponding operation and maintenance decision quadrant according to the preset operation and maintenance decision quadrant division conditions. The operation and maintenance decision module generates and outputs differentiated operation and maintenance instructions based on the mapping results of the situation assessment module.

[0032] According to the above scheme, the indicator analysis module includes a health assessment unit, a potential comparison unit, and an environmental perception unit. The health assessment unit is used to calculate the dynamic health baseline of the photovoltaic equipment based on the operating dataset, and determine the absolute health based on the percentage of the dynamic health baseline to the factory rated performance value of the photovoltaic equipment. The potential comparison unit is used to identify healthy similar equipment of the same model and orientation as the photovoltaic equipment based on the operating dataset, calculate the theoretical performance potential line of this type of equipment, and calculate the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference. The environmental perception unit is used to divide the analysis area according to the electrical connection relationship, calculate the statistical entropy value of the output power of the photovoltaic equipment in the analysis area at the synchronous sampling time as the power generation consistency entropy value, and use this entropy value as the environmental risk entropy.

[0033] According to the above scheme, the situation assessment module includes a coordinate construction unit and a quadrant mapping unit. The coordinate construction unit generates three-dimensional dynamic assessment coordinates of the photovoltaic equipment based on the absolute health, real-time potential difference, and environmental risk entropy. The quadrant mapping unit is used to determine the region to which the three-dimensional dynamic assessment coordinates belong in three-dimensional space according to preset division conditions, and map the photovoltaic equipment to the operation and maintenance decision quadrant corresponding to that region. The preset division conditions include: the first quadrant corresponds to a combination of low absolute health, high real-time potential difference, and high environmental risk entropy; the second quadrant corresponds to a combination of medium absolute health, low real-time potential difference, and high environmental risk entropy; and the third quadrant corresponds to a combination of high absolute health, low real-time potential difference, and low environmental risk entropy.

[0034] According to the above scheme, the differentiated operation and maintenance instructions include: when the photovoltaic equipment is mapped to the first quadrant, generating a first operation and maintenance instruction containing predictive fault types and priority maintenance suggestions; when mapped to the second quadrant, generating a second operation and maintenance instruction for special inspection of associated photovoltaic equipment determined by the correlation of operating data; and when mapped to the third quadrant, generating a third operation and maintenance instruction for reducing the monitoring frequency of the photovoltaic equipment.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. This invention constructs a three-dimensional dynamic evaluation coordinate system to comprehensively judge the equipment's own performance, relative degradation rate, and environmental stability, accurately distinguishing high-risk self-faulting equipment, equipment with hidden problems affected by related equipment, and healthy and stable equipment, thereby improving the accuracy of fault diagnosis and the pertinence of operation and maintenance decisions.

[0037] 2. This invention uses the discreteness of equipment operation within a region as an environmental risk entropy through quantitative analysis; it effectively perceives the hidden interference and systemic risks caused by local shadows, connection failures, or severe degradation of individual equipment to surrounding equipment.

[0038] 3. Based on the three-dimensional evaluation results, this invention maps the equipment to different operation and maintenance decision quadrants and automatically generates differentiated operation and maintenance instructions; enabling operation and maintenance resources to be prioritized for high-risk targets, while reducing unnecessary monitoring load on stable equipment. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating the steps of an intelligent analysis method for photovoltaic power plant tools according to the present invention.

[0040] Figure 2 This is a schematic diagram of the structure of an intelligent analysis system for photovoltaic power plant tools according to the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example: Figures 1-2 As shown, this invention provides a technical solution: an intelligent analysis system and method for photovoltaic power plant tools. The method includes the following steps:

[0043] S100. Collect the operation dataset of photovoltaic equipment in the target photovoltaic power station. The operation dataset includes the output power, operating temperature and associated environmental irradiance data of the photovoltaic equipment. The output power is calculated by collecting the DC side voltage and current of the photovoltaic string. The association refers to matching the irradiance data collected by the meteorological monitoring point with the photovoltaic equipment operation data of the corresponding area through timestamps and geographical location.

[0044] For example: Using a data acquisition device, the operating data of photovoltaic string A is synchronously collected over 30 consecutive days at 5-minute intervals. The operating data includes: Output power: The DC side voltage U and current I of photovoltaic string A are read through the DC smart meter, and the real-time output power P = U × I is calculated; for example, at noon on a sunny day, if U = 600V and I = 8A, then P = 4.8kW; Operating temperature: The operating temperature T is measured by a temperature sensor installed on the back panel of photovoltaic string A; for example, if T = 45°C is measured at the same time; Corresponding environmental irradiance data: The total irradiance G_h on the horizontal plane is collected at the same frequency through the total radiation meter of the power station's meteorological station. Based on the geographical location of the power station and the installation tilt and azimuth angle of photovoltaic string A, G_h is converted into the effective irradiance G_poa irradiated on the tilted surface of photovoltaic string A; for example, if G_poa = 850W / m² at the same time. 2 The effective irradiance G_poa data is matched with the power and temperature data of photovoltaic string A using precise timestamps; this is just an example and is not a limitation.

[0045] S200. Based on the operational dataset, analyze the dynamic health baseline and theoretical performance potential line of photovoltaic equipment under standard environmental conditions to obtain the absolute health of photovoltaic equipment and the real-time potential difference between the dynamic health baseline and the theoretical performance potential line.

[0046] Specifically, step S200 includes:

[0047] S210. Based on the operating dataset, calculate the highest historical performance value of the photovoltaic equipment under preset standard environmental conditions within the current operating cycle, and use it as the dynamic health baseline of the photovoltaic equipment; the standard environmental conditions are the standard test conditions; the highest historical performance value is obtained by correcting the actual performance data to the standard environmental conditions and selecting the maximum value within the current operating cycle.

[0048] Standard environmental conditions typically employ industry-standard testing conditions, such as: standard irradiance: 1000 watts per square meter, the benchmark light intensity for measuring solar cell performance; standard cell temperature: 25 degrees Celsius, the reference operating temperature of photovoltaic cells under standard testing conditions; and atmospheric quality: typically AM1.5, representing the standard spectral distribution of sunlight passing through the atmosphere to reach the ground.

[0049] For example: The actual output power of string A at all sampling times is corrected to a preset standard environmental condition using a standard performance model, such as a single diode model or its simplified form. The formula is: P_stc≈P_meas×(G_stc / G_poa)×[1+γ×(T_cell_stc-T_cell)]; where P_stc represents the estimated power value corrected to STC; P_meas represents the actual measured power; G_stc represents the standard irradiance; G_poa represents the actual measured effective irradiance of the tilted surface; T_cell_ stc represents the standard cell temperature; T_cell represents the estimated or measured cell operating temperature; γ represents the power temperature coefficient of the photovoltaic module, obtained from the module datasheet; the standard power value P_stc is obtained; within a 30-day operating cycle, the maximum value is selected from the sequence of standard power values ​​P_stc; this maximum value is P_stc_max = 5.2kW; P_stc_max = 5.2kW is used as the current dynamic health baseline of photovoltaic string A, representing the best performance that photovoltaic string A can achieve under ideal weather conditions in the current state; this is only an example and is not a limitation.

[0050] S220. Based on the operational dataset, identify similar equipment with the same model and orientation as the photovoltaic equipment and in a healthy operating state, and calculate the group optimal performance value of similar equipment under standard environmental conditions as the theoretical performance potential line; the group optimal performance value is obtained by calculating the optimal quantile of the performance data of similar healthy equipment.

[0051] For example: Within the power plant, other photovoltaic strings B, C, and D with the exact same model and installation orientation as photovoltaic string A, and which have not reported any alarms and have stable performance in the past week, are identified as healthy similar devices; the maximum standard power values ​​of photovoltaic strings B, C, and D under STC conditions are calculated for each of them within the same 30-day period; the P_stc_max of photovoltaic strings B, C, and D are 5.3kW, 5.35kW, and 5.25kW, respectively;

[0052] The upper quartile of the P_stc_max value of similar equipment is taken as the optimal performance value of the group; the calculated values ​​are: 5.3kW (B), 5.35kW (C), and 5.25kW (D); after sorting, the upper quartile value is 5.325kW; this value is the theoretical performance potential line of photovoltaic string A; this is only an example and is not a limitation.

[0053] S230. Determine the absolute health as a percentage of the dynamic health baseline to the factory-rated performance value of the photovoltaic equipment. For example: consult the datasheet of the photovoltaic module model used in photovoltaic string A to obtain its factory-rated power, i.e., the nominal power under STC, P_rated = 300 × 20 = 6.0 kW; calculate the absolute health H_absolute = (H_base / P_rated) × 100% = (5.2 / 6.0) × 100% = 86.7%; where H_base represents the dynamic health baseline, and P_rated represents the factory-rated power of the photovoltaic equipment, i.e., the nominal maximum output power under standard test conditions; the absolute health reflects the current best performance of photovoltaic string A relative to its factory design level; this is only an example and is not a limitation.

[0054] S240. Calculate the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference;

[0055] For example: Calculate the real-time potential difference D_pot = ((P_pot - H_base) / P_pot) × 100% ≈ 1.2%; where D_pot represents the real-time potential difference, that is, the gap between the current best performance of the equipment and its theoretical potential; P_pot represents the theoretical performance potential line, that is, the group's best performance value of similar healthy equipment under standard environmental conditions; a positive real-time potential difference indicates that the best performance of photovoltaic string A is different from the best level of similar equipment, and there is relative degradation; this is only an example and is not a limitation.

[0056] S300. Based on the preset topology, the photovoltaic equipment is divided into multiple analysis areas. The power generation consistency entropy value of the photovoltaic equipment in the analysis area at the same data collection time is analyzed and used as the environmental risk entropy of the analysis area.

[0057] Specifically, step S300 includes:

[0058] S310. The preset topology is the electrical connection hierarchy of photovoltaic equipment in the target photovoltaic power station. Photovoltaic equipment belonging to the same maximum power point tracker or the same combiner box is divided into the same analysis area.

[0059] For example: According to the power plant electrical drawings, photovoltaic string A and 7 other strings (a total of 8 strings) are connected to the same maximum power point tracker of the same inverter; therefore, these 8 strings are divided into the same analysis area; this is only an example and is not a limitation.

[0060] S320. Select any synchronous sampling time, obtain the real-time output power of each photovoltaic device in the analysis area, and calculate the statistical entropy value, which characterizes the dispersion of the output power of all photovoltaic devices, based on the real-time output power of all photovoltaic devices at that time, as the power generation consistency entropy value; the statistical entropy value is calculated using the information entropy formula, and the input is the proportion of the output power of each device to the total output power of the area;

[0061] For example: Select a clear, cloudless, and stable power generation period, and simultaneously acquire the real-time output power of all 8 strings within the analysis area at that moment; assume the power value list is: [P1, P2, P3, ..., P8], where P1 is the power of photovoltaic string A, 4.8kW; calculate the total power of the analysis area at that moment, P_total=Σ(P_i), and calculate the power proportion of each string: p_i=P_i / P_total; form a probability distribution {p1, p2, ..., p8}. The distribution satisfies Σ(p_i)=1. The entropy value of this distribution is calculated using the Shannon entropy formula, which is used as the power generation consistency entropy value. The formula is: E_con=-Σ(p_i×log2(p_i)); where Σ represents the summation of i from 1 to 8; p_i represents the power proportion of the i-th string; the entropy value is the largest when all strings have completely consistent power, indicating the worst consistency, i.e., the largest dispersion; the entropy value is the smallest, which is 0, when only one string generates power and the rest are zero; this is only an example and no restrictions are imposed.

[0062] S330. Use the power generation consistency entropy value of the analysis area as the environmental risk entropy of all photovoltaic devices in the analysis area.

[0063] S400 constructs a three-dimensional dynamic evaluation coordinate system based on the absolute health of photovoltaic equipment, real-time potential difference, and environmental risk entropy of the region; and maps the photovoltaic equipment to a preset operation and maintenance decision quadrant; the operation and maintenance decision quadrant includes the first quadrant, the second quadrant, and the third quadrant;

[0064] Specifically, step S400 includes:

[0065] S410. Construct a three-dimensional dynamic evaluation coordinate system. The X-axis of the three-dimensional dynamic evaluation coordinate system represents the absolute health of the photovoltaic equipment, the Y-axis represents the real-time potential difference of the photovoltaic equipment, and the Z-axis represents the environmental risk entropy of the area where the photovoltaic equipment is located. The coordinate parameters are dynamically adjusted in real time as the photovoltaic equipment operation data is updated.

[0066] For example: Construct a three-dimensional dynamic evaluation coordinate point for photovoltaic string A: the X-axis is the absolute health H_absolute=86.7%; the Y-axis is the real-time potential difference D_pot=1.2%; the Z-axis is the environmental risk entropy R_environment=2.95; the coordinate point is (86.7,1.2,2.95). This is only an example and is not a limitation.

[0067] S420. The pre-defined quadrant division criteria for operation and maintenance decisions include: the first quadrant corresponds to the parameter combination of low absolute health, high real-time potential difference, and high environmental risk entropy; the second quadrant corresponds to the parameter combination of medium absolute health, low real-time potential difference, and high environmental risk entropy; and the third quadrant corresponds to the parameter combination of high absolute health, low real-time potential difference, and low environmental risk entropy.

[0068] Furthermore, in the parameter combinations corresponding to the first quadrant, low absolute health is a value below the first health threshold, high real-time potential difference is a value above the first potential difference threshold, and high environmental risk entropy is a value above the first risk entropy threshold; in the parameter combinations corresponding to the second quadrant, medium absolute health is a value between the lower and upper limits of the second health threshold, low real-time potential difference is a value below the second potential difference threshold, and high environmental risk entropy is a value above the second risk entropy threshold; in the parameter combinations corresponding to the third quadrant, high absolute health is a value above the third health threshold, low real-time potential difference is a value below the third potential difference threshold, and low environmental risk entropy is a value below the third risk entropy threshold.

[0069] S430. Extract the absolute health status, real-time potential difference, and environmental risk entropy data of the photovoltaic equipment and its region, and use them as coordinate points in the three-dimensional dynamic evaluation coordinates. Determine the operation and maintenance decision quadrant to which the coordinate point belongs based on the preset division conditions, and complete the mapping of the photovoltaic equipment to the operation and maintenance decision quadrant.

[0070] The pre-defined quadrant division conditions for operation and maintenance decisions are achieved by setting specific numerical threshold ranges for each evaluation dimension. The threshold ranges are personalized and periodically optimized based on the statistical characteristics of the historical operation data of the target photovoltaic power plant, the type and aging stage of the photovoltaic equipment, and specific operation and maintenance strategies. For example, in the initial deployment phase, based on historical data during a stable period after the power plant is put into operation, the statistical distribution of indicators for each dimension is calculated, and the initial threshold boundaries are set accordingly to ensure that most of the equipment in a healthy state can be correctly classified. In subsequent operation, the above thresholds are adaptively adjusted according to the overall aging trend of the equipment, the configuration requirements of operation and maintenance resources, and the long-term impact of environmental climate, so that the entire evaluation system dynamically adapts to the actual operating status of the power plant and the operation and maintenance management objectives.

[0071] For example, the system's preset quadrant division thresholds are as follows:

[0072] First quadrant (high risk): H_absolute < 85%, D_pot > 1.5%, R_environment > 2.8;

[0073] Second quadrant (associated latent risks): 85%≤H_absolute≤95%, D_pot≤1.5%, R_environment>2.8;

[0074] Third quadrant (stable state): H_absolute>95%, D_pot≤1.5%, R_environment≤2.8.

[0075] The determination is made based on the coordinates (86.7, 1.2, 2.95) of photovoltaic string A:

[0076] H_absolute = 86.7% (≥85% and ≤95%), which meets the health criteria of the second quadrant;

[0077] D_pot=1.2% (≤1.5%), which meets the potential difference condition of the second quadrant;

[0078] R_environment=2.95 (>2.8), which meets the risk entropy condition of the second quadrant;

[0079] Therefore, the coordinates of photovoltaic string A fully satisfy all the preset conditions of the second quadrant.

[0080] Mapping results: Photovoltaic string A is clearly mapped to the second quadrant, indicating that there is a hidden operational disturbance in its analysis area. This is only an example and is not a limitation.

[0081] S500: When a photovoltaic device is mapped to the first quadrant, it is determined to be a high-priority fault risk target, and a first maintenance instruction is generated; when a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden fault risk in its analysis area, and a second maintenance instruction is generated for the associated photovoltaic device; when a photovoltaic device is mapped to the third quadrant, it is determined to be in a stable state, and a third maintenance instruction to reduce the monitoring frequency is executed.

[0082] Specifically, step S500 includes:

[0083] S510. When a photovoltaic device is mapped to the first quadrant, it is determined to be a fault risk target with the highest urgency of operation and maintenance. Based on the operating data characteristics of the photovoltaic device, a matching analysis is performed through a pre-set fault mode library to generate a first operation and maintenance instruction containing specific fault type predictions and priority maintenance suggestions.

[0084] For example: If photovoltaic string A is determined to be a high-priority fault risk target; the system calls the pre-set fault mode library; the fault mode library stores typical features; the recent detailed operating data of photovoltaic string A is matched with the fault library; the matching result shows that its features have the highest confidence level with series mismatch; the fault mode library is a structured database that stores feature vectors of various typical photovoltaic equipment faults; each fault entry includes at least: fault type, data feature vector, diagnostic confidence parameter, and cause and maintenance suggestions; the fault mode library collects detailed operating data of equipment with confirmed faults in the power plant's history for a period of time before the fault occurred, and based on photovoltaic modules and physical models, simulates the output characteristics under different fault types to generate theoretical feature vectors; it transforms industry-recognized fault diagnosis experience into quantitative feature descriptions; and as new fault cases accumulate, its feature entries are continuously expanded and optimized;

[0085] The first maintenance instruction is automatically generated: Photovoltaic string A has a high risk, and the predicted fault type is series mismatch. It is recommended to prioritize drone infrared inspection or on-site IV curve testing, and focus on checking the consistency of each component in the string. This is only an example and is not a limitation.

[0086] S520. When a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden operational interference caused by other devices in the analysis area where it is located; based on the correlation between electrical connection relationship and operation data, at least one adjacent photovoltaic device with the highest correlation with the photovoltaic device is identified as the associated photovoltaic device from the analysis area, and a second operation and maintenance instruction is generated for a special inspection of the associated photovoltaic device.

[0087] For example: If it is determined that there is latent operational interference within the MPPT where photovoltaic string A is located; the analysis area is locked; based on the correlation analysis of operational data, the Pearson correlation coefficient of the power time series of photovoltaic string A with the power time series of the other 7 photovoltaic strings is calculated; the calculation shows that the power curves of photovoltaic string A and photovoltaic string F are highly negatively correlated in multiple fluctuation periods, that is, when A decreases, F often increases; photovoltaic string F is identified as the adjacent photovoltaic device with the highest correlation; a second maintenance instruction is automatically generated: latent interference is detected in the area, photovoltaic string A may be affected by photovoltaic string F; it is recommended to conduct a special inspection of photovoltaic string F, focusing on whether its branch circuit breakers and connectors have poor contact, or whether there is shading; this is only an example and is not a limitation.

[0088] S530. When the photovoltaic equipment is mapped to the third quadrant, it is determined that it is in a healthy and stable operating state. Based on the preset resource optimization strategy, the third operation and maintenance instruction is executed on the photovoltaic equipment, including extending the data acquisition interval and reducing the frequency of analysis and calculation.

[0089] For example, if photovoltaic string A is determined to be in a stable state, the monitoring strategy for photovoltaic string X is automatically adjusted according to the preset resource optimization strategy; the data acquisition interval of photovoltaic string X is extended from 5 minutes to 30 minutes; in the backend analysis task, its health calculation frequency is reduced from once a day to once a week, thereby reducing the overall monitoring load of the system.

[0090] This invention provides another technical solution: an intelligent analysis system for photovoltaic power plant tools, which includes: a data acquisition module, an indicator analysis module, a situation assessment module, and an operation and maintenance decision-making module.

[0091] The data acquisition module is used to collect the operational dataset of photovoltaic equipment in the target photovoltaic power station. The operational dataset includes at least the output power and ambient irradiance data of the photovoltaic equipment. The index analysis module is used to analyze absolute health, real-time potential difference, and environmental risk entropy. The situation assessment module is used to construct a three-dimensional dynamic assessment coordinate based on the absolute health, real-time potential difference, and environmental risk entropy output by the index analysis module, and to map the photovoltaic equipment to the corresponding operation and maintenance decision quadrant according to the preset operation and maintenance decision quadrant division conditions. The operation and maintenance decision module generates and outputs differentiated operation and maintenance instructions based on the mapping results of the situation assessment module.

[0092] The indicator analysis module includes a health assessment unit, a potential comparison unit, and an environmental perception unit. The health assessment unit calculates the dynamic health baseline of the photovoltaic equipment based on the operational dataset and determines the absolute health level based on the percentage of the dynamic health baseline to the factory rated performance value of the photovoltaic equipment. The potential comparison unit identifies healthy similar photovoltaic equipment of the same model and orientation based on the operational dataset, calculates the theoretical performance potential line of this type of equipment, and calculates the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference. The environmental perception unit divides the analysis area according to the electrical connection relationship, calculates the statistical entropy value of the output power of the photovoltaic equipment in the analysis area at the synchronous sampling time as the power generation consistency entropy value, and uses this entropy value as the environmental risk entropy.

[0093] The situation assessment module includes a coordinate construction unit and a quadrant mapping unit. The coordinate construction unit generates three-dimensional dynamic assessment coordinates for photovoltaic equipment based on absolute health, real-time potential difference, and environmental risk entropy. The quadrant mapping unit is used to determine the region in three-dimensional space to which the three-dimensional dynamic assessment coordinates belong, and to map the photovoltaic equipment to the corresponding operation and maintenance decision quadrant according to preset division conditions. The preset division conditions include: the first quadrant corresponds to a combination of low absolute health, high real-time potential difference, and high environmental risk entropy; the second quadrant corresponds to a combination of medium absolute health, low real-time potential difference, and high environmental risk entropy; and the third quadrant corresponds to a combination of high absolute health, low real-time potential difference, and low environmental risk entropy.

[0094] The differentiated operation and maintenance instructions include: when the photovoltaic equipment is mapped to the first quadrant, a first operation and maintenance instruction containing predictive fault types and priority maintenance suggestions is generated; when mapped to the second quadrant, a second operation and maintenance instruction is generated for special inspection of the associated photovoltaic equipment determined by the correlation of operating data is generated; and when mapped to the third quadrant, a third operation and maintenance instruction is generated to reduce the monitoring frequency of the photovoltaic equipment.

[0095] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An intelligent analysis system and method for photovoltaic power plant tools, characterized in that: The method includes: S100. Collect the operation dataset of the photovoltaic equipment in the target photovoltaic power station. The operation dataset includes the output power, operating temperature and associated ambient irradiance data of the photovoltaic equipment. S200. Based on the running dataset, analyze the dynamic health baseline and theoretical performance potential line of the photovoltaic equipment under standard environmental conditions to obtain the absolute health of the photovoltaic equipment and the real-time potential difference between the dynamic health baseline and the theoretical performance potential line. S300. Divide the photovoltaic equipment into multiple analysis areas according to the preset topology relationship, and analyze the power generation consistency entropy value of the photovoltaic equipment in the analysis area at the same collection time, which is used as the environmental risk entropy of the analysis area. S400. Based on the absolute health of the photovoltaic equipment, the real-time potential difference, and the environmental risk entropy of the area, a three-dimensional dynamic evaluation coordinate system is constructed; and the photovoltaic equipment is mapped to a preset operation and maintenance decision quadrant; the operation and maintenance decision quadrant includes a first quadrant, a second quadrant, and a third quadrant; S500: When a photovoltaic device is mapped to the first quadrant, it is determined to be a high-priority fault risk target, and a first maintenance instruction is generated; when a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden fault risk in its analysis area, and a second maintenance instruction is generated for the associated photovoltaic device; when a photovoltaic device is mapped to the third quadrant, it is determined to be in a stable state, and a third maintenance instruction to reduce the monitoring frequency is executed.

2. The intelligent analysis method for photovoltaic power plant tools according to claim 1, characterized in that: Step S200 includes: S210. Based on the running dataset, calculate the highest historical performance value of the photovoltaic equipment under preset standard environmental conditions within the current operating cycle, and use it as the dynamic health baseline of the photovoltaic equipment. S220. Based on the operational dataset, identify similar devices with the same model and orientation as the photovoltaic device and in a healthy operating state, and calculate the group optimal performance value of similar devices under the standard environmental conditions as the theoretical performance potential line; S230. Determine the absolute health level as a percentage of the dynamic health baseline to the factory rated performance value of the photovoltaic equipment; S240. Calculate the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference.

3. The intelligent analysis method for photovoltaic power plant tools according to claim 1, characterized in that: Step S300 includes: S310. The preset topology is the electrical connection hierarchy of photovoltaic equipment in the target photovoltaic power station. Photovoltaic equipment belonging to the same maximum power point tracker or the same combiner box is divided into the same analysis area. S320. Select any synchronous sampling time, obtain the real-time output power of each photovoltaic device in the analysis area, and calculate the statistical entropy value that characterizes the dispersion of the output power of all photovoltaic devices based on the real-time output power of all photovoltaic devices at that time, as the power generation consistency entropy value. S330. The power generation consistency entropy value of the analysis area is used as the environmental risk entropy of all photovoltaic devices in the analysis area.

4. The intelligent analysis method for photovoltaic power plant tools according to claim 1, characterized in that: Step S400 includes: S410. Construct a three-dimensional dynamic evaluation coordinate system. The X-axis of the three-dimensional dynamic evaluation coordinate system represents the absolute health of the photovoltaic equipment, the Y-axis represents the real-time potential difference of the photovoltaic equipment, and the Z-axis represents the environmental risk entropy of the area where the photovoltaic equipment is located. The coordinate parameters are dynamically adjusted in real time as the photovoltaic equipment operation data is updated. S420. The pre-defined quadrant division criteria for operation and maintenance decisions include: the first quadrant corresponds to the parameter combination of low absolute health, high real-time potential difference, and high environmental risk entropy; the second quadrant corresponds to the parameter combination of medium absolute health, low real-time potential difference, and high environmental risk entropy; and the third quadrant corresponds to the parameter combination of high absolute health, low real-time potential difference, and low environmental risk entropy. S430. Extract the absolute health status, real-time potential difference, and environmental risk entropy data of the photovoltaic equipment and its region, and use them as coordinate points in the three-dimensional dynamic evaluation coordinates. Determine the operation and maintenance decision quadrant to which the coordinate point belongs according to the preset division conditions, and complete the mapping of the photovoltaic equipment to the operation and maintenance decision quadrant.

5. The intelligent analysis method for photovoltaic power station tools according to claim 4, characterized in that: In the parameter combination corresponding to the first quadrant, the low absolute health is a value lower than the first health threshold, the high real-time potential difference is a value higher than the first potential difference threshold, and the high environmental risk entropy is a value higher than the first risk entropy threshold. In the parameter combination corresponding to the second quadrant, the medium absolute health is a value between the lower limit and the upper limit of the second health threshold, the low real-time potential difference is a value lower than the second potential difference threshold, and the high environmental risk entropy is a value higher than the second risk entropy threshold. In the parameter combination corresponding to the third quadrant, the high absolute health is a value higher than the third health threshold, the low real-time potential difference is a value lower than the third potential difference threshold, and the low environmental risk entropy is a value lower than the third risk entropy threshold.

6. The intelligent analysis method for photovoltaic power plant tools according to claim 1, characterized in that: Step S500 includes: S510. When a photovoltaic device is mapped to the first quadrant, it is determined to be a fault risk target with the highest maintenance urgency. Based on the operating data characteristics of the photovoltaic device, a matching analysis is performed through a pre-set fault mode library to generate the first maintenance instruction containing specific fault type predictions and priority maintenance suggestions. S520. When a photovoltaic device is mapped to the second quadrant, it is determined that there is a hidden operational interference caused by other devices in the analysis area where it is located; based on the correlation between electrical connection relationship and operation data, at least one adjacent photovoltaic device with the highest correlation with the photovoltaic device is determined from the analysis area as the associated photovoltaic device, and a second operation and maintenance instruction for special inspection of the associated photovoltaic device is generated. S530. When the photovoltaic equipment is mapped to the third quadrant, it is determined that it is in a healthy and stable operating state; according to the preset resource optimization strategy, the third operation and maintenance instruction, including extending the data acquisition interval and reducing the analysis and calculation frequency, is executed on the photovoltaic equipment.

7. An intelligent analysis system for photovoltaic power plant tools, characterized in that: The system includes: a data acquisition module, an indicator analysis module, a situation assessment module, and an operation and maintenance decision-making module; The data acquisition module is used to collect the operating dataset of the photovoltaic equipment in the target photovoltaic power station. The operating dataset includes at least the output power and ambient irradiance data of the photovoltaic equipment. The indicator analysis module is used to analyze absolute health, real-time potential difference, and environmental risk entropy. The situation assessment module is used to construct a three-dimensional dynamic assessment coordinate based on the absolute health, real-time potential difference and environmental risk entropy output by the indicator analysis module, and to map the photovoltaic equipment to the corresponding operation and maintenance decision quadrant according to the preset operation and maintenance decision quadrant division conditions. The operation and maintenance decision module generates and outputs differentiated operation and maintenance instructions based on the mapping results of the situation assessment module.

8. The intelligent analysis system for photovoltaic power plant tools according to claim 7, characterized in that: The indicator analysis module includes a health assessment unit, a potential comparison unit, and an environmental perception unit. The health assessment unit is used to calculate the dynamic health baseline of the photovoltaic equipment based on the operating dataset, and to determine the absolute health based on the percentage of the dynamic health baseline to the factory rated performance value of the photovoltaic equipment. The potential comparison unit is used to identify healthy similar devices of the same model and orientation as photovoltaic devices based on the running dataset, calculate the theoretical performance potential line of such devices, and calculate the relative difference between the dynamic health baseline and the theoretical performance potential line as the real-time potential difference. The environmental sensing unit is used to divide the analysis area according to the electrical connection relationship, calculate the statistical entropy value of the output power of the photovoltaic equipment in the analysis area at the synchronous sampling time as the power generation consistency entropy value, and use the entropy value as the environmental risk entropy.

9. The intelligent analysis system for photovoltaic power plant tools according to claim 7, characterized in that: The situation assessment module includes a coordinate construction unit and a quadrant mapping unit; The coordinate construction unit generates three-dimensional dynamic evaluation coordinates for the photovoltaic equipment based on the absolute health, real-time potential difference, and environmental risk entropy. The quadrant mapping unit is used to determine the region in three-dimensional space to which the three-dimensional dynamic evaluation coordinates belong according to preset division conditions, and to map the photovoltaic equipment to the operation and maintenance decision quadrant corresponding to that region; the preset division conditions include: the first quadrant corresponds to the combination of low absolute health, high real-time potential difference and high environmental risk entropy; the second quadrant corresponds to the combination of medium absolute health, low real-time potential difference and high environmental risk entropy; and the third quadrant corresponds to the combination of high absolute health, low real-time potential difference and low environmental risk entropy.

10. The intelligent analysis system for photovoltaic power station tools according to claim 7, characterized in that: The differentiated operation and maintenance instructions include: when the photovoltaic equipment is mapped to the first quadrant, generating a first operation and maintenance instruction containing predictive fault types and priority maintenance suggestions; when mapped to the second quadrant, generating a second operation and maintenance instruction for special inspection of associated photovoltaic equipment determined by the correlation of operating data; and when mapped to the third quadrant, generating a third operation and maintenance instruction for reducing the monitoring frequency of the photovoltaic equipment.

Images