Distributed photovoltaic power station intelligent safety operation and maintenance management system

Abstract

The invention discloses a distributed photovoltaic power station intelligent safety operation and maintenance management system, and relates to the technical field of photovoltaic power station operation and maintenance, the system comprises a data acquisition module, a data processing module, a calculation and analysis module and a management execution module, and the calculation and analysis module comprises a component health sub-module, a potential risk sub-module and an operation and maintenance analysis sub-module. According to the invention, a comprehensive attenuation coefficient is output through the assembly health sub-module, a dust coverage rate and other multi-dimensional parameters are integrated to quantify an attenuation state, a fault risk index is output based on the potential risk sub-module through the potential risk sub-module, and parameters such as current fluctuation are fused to construct a multi-precursor early warning system. The operation and maintenance analysis sub-module integrates parameters such as a risk index and outputs an operation and maintenance priority index, and the three components cooperate to realize full-period quantitative management from attenuation evaluation, fault early warning to operation and maintenance decision making, so that the traditional limitation is broken through, and the evaluation accuracy and the operation and maintenance efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant operation and maintenance technology, specifically to an intelligent safety operation and maintenance management system for distributed photovoltaic power plants. Background Technology

[0002] As an important form of clean energy utilization, distributed photovoltaic power stations are characterized by wide distribution, flexible access, and proximity to load centers, and have become a key component of global energy transformation. With the rapid growth of the installed capacity of distributed photovoltaic power stations, efficient operation and maintenance management is required.

[0003] The existing operation and maintenance management methods may rely heavily on manual inspections and on-site repairs, but this method is inefficient, costly, and lacks timeliness, and cannot meet users' needs for efficient, safe, and intelligent operation and maintenance of photovoltaic power plants. Furthermore, existing operation and maintenance management systems often focus on environmental factors (such as light and temperature) or single physical fields (such as photoelectric performance) when assessing the degradation of photovoltaic modules. They may not include implicit degradation factors such as mechanical structural stability (such as bracket tilt offset) and electrical connection status (such as the increase in contact resistance of wiring terminals) in the assessment. This results in a one-sided degradation assessment, which may not reflect the true condition of photovoltaic modules and affect the accuracy of subsequent risk assessment. Furthermore, existing fault warning methods in operation and maintenance management systems mostly analyze electrical parameters (such as current fluctuations) or environmental parameters (such as humidity) independently, without taking the overall degradation status of components as the basis for risk assessment. They ignore the core logic that degradation is the underlying cause of faults. At the same time, they lack integration of latent fault precursors such as insulation degradation (such as moisture penetration in the backplane) and safety grounding (such as increased grounding resistance), resulting in fragmented warning signals, making it difficult to distinguish between minor anomalies and serious risks, limiting the accuracy of warnings and resulting in poor intelligence. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent and safe operation and maintenance management system for distributed photovoltaic power plants, which solves the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an intelligent safety operation and maintenance management system for distributed photovoltaic power plants, comprising: Data acquisition module: used to collect component health correlation data, risk and fault correlation data, and component operation and maintenance correlation data of distributed photovoltaic power stations; Data processing module: Used to input component health correlation data, risk and fault correlation data and component operation and maintenance correlation data obtained by the data acquisition module, clean the input data, and then input the cleaned data into the analysis component; Calculation and Analysis Module: Based on the dynamic dust coverage, temperature stress coefficient, dynamic aging factor, operating years factor, tilt angle dynamic offset index, and terminal contact resistance increment factor in the component health correlation data, the comprehensive degradation coefficient of the component is output. The comprehensive degradation coefficient of the component is used to quantify the degree of performance degradation of the photovoltaic module in long-term operation. Based on the current fluctuation coefficient, infrared anomaly ratio, backsheet water vapor transmittance and grounding resistance dynamic increment factor in the risk fault correlation data, and combined with the module comprehensive attenuation coefficient, a potential fault risk index is output. The potential fault risk index is used to quantify the potential risk of photovoltaic module failure, and to provide a priority basis for fault early warning and operation and maintenance resource allocation. Based on the power generation contribution, safety hazard level, grid voltage fluctuation coupling coefficient, operation and maintenance resource accessibility coefficient, and historical average fault repair time in the component operation and maintenance related data, and combined with the component comprehensive attenuation coefficient and potential fault risk index, a safe operation and maintenance priority index is output. The safe operation and maintenance priority index is used to quantify the urgency of photovoltaic module operation and maintenance, and provide decision support for operation and maintenance resource scheduling and task prioritization. Management Execution Module: Used to input the component comprehensive attenuation coefficient, potential failure risk index and security operation and maintenance priority index output by the calculation and analysis module, and to perform operation and maintenance priority classification and operation and maintenance resource allocation based on the input data.

[0006] Optionally, the calculation and analysis module includes a component health submodule, a potential risk submodule, and an operation and maintenance analysis submodule.

[0007] Optionally, the processing procedure of the component health submodule is as follows: A1. By analyzing the real-time area ratio of dust coverage on the surface of photovoltaic modules, the degree of dust obstruction of the module's light absorption efficiency can be reflected, so as to calculate the dynamic dust coverage rate. A2. By collecting real-time temperature data of the photovoltaic module backsheet, we can analyze the fatigue damage of temperature fluctuations to the photovoltaic module materials and calculate the temperature stress coefficient. A3. By combining the daily average irradiance, daily average relative humidity and daily average wind speed of the environment in which the photovoltaic module is located, the aging of the photovoltaic module under the combined influence of light, humidity and mechanical vibration factors in actual operation is considered, so as to calculate the dynamic aging factor. A4. By analyzing the cumulative operating time of photovoltaic modules from commissioning to the present, the impact of time accumulation on module aging is reflected, so as to calculate the operating life factor; A5. By analyzing the comparison between the current tilt angle of the photovoltaic module and the initial installation tilt angle, the stability of the mechanical structure of the photovoltaic module support can be reflected, and the dynamic tilt angle offset index can be calculated. A6. By measuring the change in terminal resistance from its initial value within the combiner box, the degree of electrical connection degradation is reflected, and the terminal contact resistance increment factor is calculated, thereby ultimately outputting the overall component attenuation coefficient.

[0008] Optionally, the processing procedure for the potential risk submodule is as follows: B1. By analyzing the degree of fluctuation of the output current of photovoltaic modules within a unit time, the electrical characteristics are used as the basis for early fault warning, and the current fluctuation coefficient is calculated. B2. By analyzing the area ratio of abnormally high-temperature regions in the infrared thermal imaging of the photovoltaic module surface, the risk of local overheating can be reflected, and the proportion of infrared anomalies can be calculated. B3. By analyzing the real-time relative humidity of the environment where the photovoltaic modules are located, the water vapor content in the atmosphere can be reflected to calculate the ambient humidity; B4. By comparing the humidity inside and outside the photovoltaic module, the dynamic permeability of water vapor to the backsheet of the photovoltaic module in actual operation is analyzed to reflect the degree of deterioration of the insulation performance of the backsheet, so as to calculate the water vapor transmission rate of the backsheet. B5. By comparing the current grounding resistance of the photovoltaic module with the initial grounding resistance, the change in grounding resistance is calculated to reflect the safe current discharge capacity of the grounding system. This allows for the calculation of the dynamic increment factor of grounding resistance, which, combined with the module's comprehensive attenuation coefficient, outputs a potential fault risk index.

[0009] Optionally, the processing procedure of the operation and maintenance analysis submodule is as follows: C1. By analyzing the proportion of photovoltaic modules in the total power generation of the power plant, we can reflect their importance to the power generation of the power plant and calculate their contribution to power generation. C2. By analyzing the severity of the safety consequences caused by photovoltaic module failures, the safety risk level is reflected, and values ​​are assigned for various scenarios to calculate the safety hazard level. C3. By calculating the difference between the normal operating voltage and the fault simulation voltage of the photovoltaic module, the voltage fluctuation is obtained, and combined with the maximum allowable fluctuation of the distribution network, the potential impact of photovoltaic module faults on voltage stability is analyzed, so as to calculate the grid voltage fluctuation coupling coefficient. C4. Calculate the accessibility coefficient of maintenance resources by analyzing the standardized time cost for maintenance personnel to reach the photovoltaic module location from the nearest maintenance station; C5. By analyzing the normalized value of the average repair time of historical faults of photovoltaic modules of the same model, the average repair time of historical faults is calculated. Combined with the comprehensive degradation coefficient of the module and the potential fault risk index, the safety operation and maintenance priority index is calculated.

[0010] Optionally, the operation and maintenance priority classification in the management execution module is specifically as follows: A security operation and maintenance priority index of ≥0.7 indicates high priority and requires immediate intervention; A security operation and maintenance priority index of 0.3 ≤ security operation and maintenance priority index < 0.7 indicates medium priority, and a weekly maintenance plan should be developed. A security operation and maintenance priority index of <0.3 indicates low priority and should be included in the monthly routine inspection.

[0011] Optionally, the allocation of operation and maintenance resources in the management execution module is specifically as follows: When it is in a high priority state; Resource allocation: dispatch personnel from the nearest maintenance station and equip them with specialized tools; Intervention measures: Immediately stop the machine for inspection, focusing on investigating areas of infrared anomalies, and inspect and repair the grounding electrode and cables; Closed-loop verification: Retest the potential failure risk index within 24 hours after the repair to ensure it drops below 0.3, and update the overall component attenuation coefficient to the database simultaneously; When in medium priority; The plan is developed and incorporated into the weekly maintenance plan, with the task list arranged in descending order of security operation and maintenance priority index; Targeted measures include checking for damage to the backsheet and repairing it if the backsheet is damaged, and adjusting the bracket fastening bolts and re-measuring the tilt angle for components with excessive tilt angle deviation. Track the effects and retest relevant parameters weekly until the overall attenuation coefficient of the component stabilizes and decreases. When at low priority: Routine monitoring, included in monthly inspections, with a focus on dynamic dust coverage and temperature stress coefficient; Data accumulation: Continuously collect parameters and update the overall component attenuation coefficient. When the security operation and maintenance priority index increases and exceeds 0.3, it will be automatically upgraded to medium priority.

[0012] Optionally, the data processing module cleans the input data, specifically by: Outlier cleaning: Abnormal data caused by sensor malfunctions are removed using the 3σ principle, and the removed data is replaced with the average of the previous and next 10 minutes. Normalization and unification: Normalize all parameters to the 0-1 range according to the formula requirements to ensure that parameters with different dimensions can be calculated in a weighted manner; Time synchronization: unify the timestamps of all sensor data.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: I. This invention outputs a comprehensive degradation coefficient for photovoltaic modules through a module health submodule. By integrating multi-dimensional parameters such as dynamic dust coverage, temperature stress coefficient, dynamic aging factor, service life, dynamic tilt offset, and terminal contact resistance increment, it achieves a comprehensive quantitative assessment of the degradation status of photovoltaic modules. Among them, dynamic dust coverage and temperature stress coefficient reflect the impact of environmental erosion on the module's light absorption efficiency and material thermal aging. The dynamic aging factor combines dynamic environmental factors such as light, humidity, and wind speed to quantify the aging effect of multiple physical fields superimposed. Service life reflects the cumulative effect of time on the natural degradation of the module. The dynamic tilt offset and terminal contact resistance increment are respectively related to the implicit degradation of mechanical structure stability and electrical connection status. These parameters work together to break through the limitations of traditional single-dimensional assessment, extending module degradation from material aging to a comprehensive state of coordinated degradation of multiple systems including environment, machinery, and electrical systems. This provides a quantitative basis that truly reflects the module's health baseline for subsequent fault risk assessment, ensuring that the judgment of the degree of module degradation is closer to the complex physical processes in actual operation.

[0014] II. This invention outputs a potential fault risk index through a potential risk submodule. Based on the component's comprehensive attenuation coefficient, it integrates parameters such as current fluctuation coefficient, infrared anomaly ratio, ambient humidity, backsheet moisture transmittance, and dynamic increment of grounding resistance to construct a multi-precursor fusion fault risk early warning system based on attenuation status. The component's comprehensive attenuation coefficient serves as the underlying foundation for risk assessment, relating the degree of attenuation to the intrinsic correlation between fault occurrence. The current fluctuation coefficient and infrared anomaly ratio reflect obvious fault precursors such as abnormal electrical performance and local overheating. Ambient humidity exacerbates material corrosion, backsheet moisture transmittance quantifies the implicit process of internal insulation degradation in the component, and dynamic increment of grounding resistance directly relates to the safe discharge capacity of the grounding system. These parameters work synergistically to upgrade fault risk early warning from single-parameter anomaly identification to a multi-dimensional coupled assessment of attenuation status, obvious precursors, and implicit degradation. This achieves full-cycle risk quantification of components from the accumulation of implicit risks to the occurrence of obvious faults, avoiding the lag and one-sidedness of traditional early warning systems and improving the foresight and accuracy of fault early warning.

[0015] Third, this invention outputs a safety operation and maintenance priority index through an operation and maintenance analysis submodule. By integrating parameters such as potential fault risk index, component comprehensive attenuation coefficient, power generation contribution, safety hazard level, grid voltage fluctuation coupling coefficient, operation and maintenance resource accessibility coefficient, and historical average fault repair time, a multi-objective quantitative operation and maintenance decision priority system is constructed. Among them, the potential fault risk index and component comprehensive attenuation coefficient determine the basic priority of fault risk and attenuation; power generation contribution distinguishes the differences in power generation value of components; safety hazard level reflects the severity of the safety consequences that faults may cause; grid voltage fluctuation coupling coefficient relates to the impact of component faults on grid stability; and operation and maintenance resource accessibility coefficient and historical average fault repair time consider the actual operation and maintenance time cost and technical feasibility. These parameters work together to transform abstract operation and maintenance needs into comparable quantitative priorities, clarify the priority order of high-risk, high-value, and high-consequence components, avoid resource misallocation caused by traditional experience-based decision-making, ensure that operation and maintenance resources are tilted towards the most needed components, achieve synergistic optimization of safety assurance, power generation benefits, and grid stability, and improve the overall efficiency and scientific nature of distributed photovoltaic power station operation and maintenance management. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram of the structure of the calculation and analysis module of the present invention; Figure 3 This is a schematic diagram of the management execution module of the present invention; Figure 4 This is a schematic diagram of the operation flow of the calculation and analysis module of the present invention. Detailed Implementation

[0017] 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.

[0018] Please see Figures 1 to 4 This embodiment provides an intelligent safety operation and maintenance management system for distributed photovoltaic power plants, including: Data acquisition module: used to collect component health correlation data, risk and fault correlation data, and component operation and maintenance correlation data of distributed photovoltaic power stations; Data processing module: Used to input component health correlation data, risk and fault correlation data and component operation and maintenance correlation data obtained by the data acquisition module, clean the input data, and then input the cleaned data into the analysis component; The data cleaning process described above is as follows: Outlier cleaning: Abnormal data caused by sensor malfunctions are removed using the 3σ principle, and the removed data is replaced with the average of the previous and next 10 minutes. Normalization and unification: Normalize all parameters to the 0-1 range according to the formula requirements to ensure that parameters with different dimensions can be calculated in a weighted manner; Time synchronization: unifying the timestamps of all sensor data; Calculation and Analysis Module: Based on the dynamic dust coverage, temperature stress coefficient, dynamic aging factor, operating years factor, tilt angle dynamic offset index, and terminal contact resistance increment factor in the component health correlation data, the comprehensive degradation coefficient of the component is output. The comprehensive degradation coefficient of the component is used to quantify the degree of performance degradation of the photovoltaic module in long-term operation. Based on the current fluctuation coefficient, infrared anomaly ratio, backsheet water vapor transmittance and grounding resistance dynamic increment factor in the risk fault correlation data, and combined with the module comprehensive attenuation coefficient, a potential fault risk index is output. The potential fault risk index is used to quantify the potential risk of photovoltaic module failure, and to provide a priority basis for fault early warning and operation and maintenance resource allocation. Based on the power generation contribution, safety hazard level, grid voltage fluctuation coupling coefficient, operation and maintenance resource accessibility coefficient, and historical average fault repair time in the component operation and maintenance related data, and combined with the component comprehensive attenuation coefficient and potential fault risk index, a safe operation and maintenance priority index is output. The safe operation and maintenance priority index is used to quantify the urgency of photovoltaic module operation and maintenance, and provide decision support for operation and maintenance resource scheduling and task prioritization. The calculation and analysis module includes a component health submodule, a potential risk submodule, and an operation and maintenance analysis submodule. Management and execution module: Used to input the component comprehensive attenuation coefficient, potential failure risk index and security operation and maintenance priority index output by the calculation and analysis module, and to perform operation and maintenance priority classification and operation and maintenance resource allocation based on the input data; The above-mentioned operation and maintenance priority levels are as follows: A security operation and maintenance priority index of ≥0.7 indicates high priority and requires immediate intervention; A security operation and maintenance priority index of 0.3 ≤ security operation and maintenance priority index < 0.7 indicates medium priority, and a weekly maintenance plan should be developed. A security operation and maintenance priority index of <0.3 indicates low priority and should be included in the monthly routine inspection. The allocation of the aforementioned operation and maintenance resources is as follows: When it is in a high priority state; Resource allocation involves dispatching personnel from the nearest maintenance station and equipping them with specialized tools (such as infrared thermal imagers and grounding resistance testers). Intervention measures: Immediately stop the machine for inspection, focusing on checking abnormal infrared areas (locating hot spots), increased grounding resistance (inspecting and repairing the grounding electrode and cable), and contact resistance of the wiring terminals (re-crimping or replacing the terminals). Closed-loop verification: Retest the potential failure risk index within 24 hours after the repair to ensure it drops below 0.3, and update the overall component attenuation coefficient to the database simultaneously; When in medium priority; The plan is developed and incorporated into the weekly maintenance plan, with the task list arranged in descending order of security operation and maintenance priority index; Targeted measures include checking for damage to the backsheet and repairing it if the backsheet is damaged, and adjusting the bracket fastening bolts and re-measuring the tilt angle for components with excessive tilt angle deviation. Track the effects and retest relevant parameters weekly until the overall attenuation coefficient of the component stabilizes and decreases. When at low priority: Routine monitoring is included in monthly inspections, with a focus on dynamic dust coverage (scheduled for regular cleaning) and temperature stress coefficient (to avoid prolonged high-temperature operation). Data accumulation: Continuously collect parameters and update the overall component attenuation coefficient. When the security operation and maintenance priority index increases and exceeds 0.3, it will be automatically upgraded to medium priority.

[0019] Based on the above, the component health submodule (component comprehensive degradation coefficient KD) of this system serves as the starting point, accurately depicting the component health baseline and assessing the current health status of the components. The potential risk submodule (potential failure risk index RF) serves as the intermediate layer, identifying potential future problems based on the health baseline and assessing the level of component failure risk. The operation and maintenance analysis submodule (safe operation and maintenance priority index PT) serves as the endpoint, assessing the priority of operation and maintenance resources. This closed loop enables seamless integration of the entire process from component status monitoring to fault risk warning and operation and maintenance action decision-making, avoiding the problems of disconnect between monitoring data and decision-making, and separation between risk assessment and actual action in traditional operation and maintenance, thus enabling operation and maintenance management to shift from experience-driven to data-driven. Traditional operation and maintenance often judges the status of components based on macro indicators such as years of operation and overall power generation, ignoring individual differences (for example, components in the same batch may have significant differences in degradation due to different installation locations and environmental erosion). The three sets of sub-modules achieve accurate assessment of individual components through multi-dimensional micro parameters (such as tilt angle offset and contact resistance increment of a single component), avoiding the extensive management of over-maintaining healthy components and ignoring problematic components. Many failures in distributed photovoltaic power stations are caused by the accumulation of hidden risks (such as the slow increase in contact resistance of wiring terminals and the gradual infiltration of moisture into the back sheet). Traditional methods are difficult to detect in the early stages. The three sets of sub-modules integrate hidden parameters of multiple physical fields such as mechanical, electrical and material, and transform the invisible degradation process into quantifiable indicators, enabling operation and maintenance personnel to intervene in hidden risks in advance and fundamentally reduce the occurrence of obvious failures. The potential risk submodule (potential failure risk index RF) identifies potential safety hazards such as increased grounding resistance and moisture penetration in the backplane in advance, while the operation and maintenance analysis submodule (safe operation and maintenance priority index PT) prioritizes resource scheduling to handle high safety risk components, thereby reducing the probability of safety accidents such as electric shock and fire from the source. The component health submodule (component comprehensive degradation coefficient KD) identifies severely degraded components, while the operation and maintenance analysis submodule (safe operation and maintenance priority index PT) prioritizes the maintenance of high-value components based on their power generation contribution, reducing power generation losses caused by component failures and ensuring the long-term power generation benefits of the power plant. The Operation and Maintenance Analysis Submodule (Security Operation and Maintenance Priority Index PT) optimizes the operation and maintenance ranking based on parameters such as repair time and resource accessibility, avoids ineffective operation and maintenance (such as frequent inspections of low-risk components), reduces the manpower and material costs of operation and maintenance, and achieves the highest safety and power generation efficiency with the lowest cost. The three sub-modules transform complex operation and maintenance issues (component status, failure risks, and resource scheduling) into calculable and comparable quantitative indicators, providing standardized decision-making basis for intelligent operation and maintenance systems. Whether it is algorithm model iteration and operation and maintenance strategy optimization or cross-power station data comparison, it can all be carried out based on a unified quantitative framework, avoiding inconsistencies in decision-making caused by experience reliance and subjective judgment in traditional operation and maintenance, and promoting the standardization, intelligence, and sustainability of distributed photovoltaic power station operation and maintenance.

[0020] Please see Figure 1 , Figure 2 , Figure 3 and Figure 4 The specific components of the health submodule are as follows: A1. By analyzing the real-time area ratio of dust coverage on the surface of photovoltaic modules, the degree of dust obstruction of the module's light absorption efficiency can be reflected, so as to calculate the dynamic dust coverage rate. A2. By collecting real-time temperature data of the photovoltaic module backsheet, we can analyze the fatigue damage of temperature fluctuations to the photovoltaic module materials and calculate the temperature stress coefficient. A3. By combining the daily average irradiance, daily average relative humidity and daily average wind speed of the environment in which the photovoltaic module is located, the aging of the photovoltaic module under the combined influence of light, humidity and mechanical vibration factors in actual operation is considered, so as to calculate the dynamic aging factor. A4. By analyzing the cumulative operating time of photovoltaic modules from commissioning to the present, the impact of time accumulation on module aging is reflected, so as to calculate the operating life factor; A5. By analyzing the comparison between the current tilt angle of the photovoltaic module and the initial installation tilt angle, the stability of the mechanical structure of the photovoltaic module support can be reflected, and the dynamic tilt angle offset index can be calculated. A6. By measuring the change in terminal resistance from its initial value within the combiner box, the degree of electrical connection degradation can be reflected, and the terminal contact resistance increment factor can be calculated to ultimately output the overall component attenuation coefficient. The calculation formula for the component health submodule is as follows: ; in: KD refers to the overall degradation coefficient of a photovoltaic module, which is used to quantify the degree of performance degradation of a photovoltaic module due to multiple factors such as environment, mechanical and electrical factors during long-term operation. It is a core indicator for assessing the health status of the module and provides basic data support for subsequent fault risk assessment and operation and maintenance decisions. KDA stands for Dynamic Dust Coverage Ratio, which is the real-time area ratio of dust coverage on the surface of photovoltaic modules. It reflects the degree to which dust blocks the light absorption efficiency of the modules. The value ranges from 0 to 1 (0 means no dust, 1 means complete coverage). It can be obtained by collecting images of the module surface by a high-definition industrial camera deployed above the photovoltaic module array, and using image recognition algorithms (such as threshold segmentation) to identify the dust-covered area and calculate its ratio to the total area of ​​the module. The introduction of dynamic dust coverage (KDA) directly reflects the hindering effect of dust on light absorption by the modules. Dust coverage reduces the intensity of incident light, leading to a decrease in module output power. Simultaneously, dust accumulation increases thermal resistance, exacerbating localized overheating and indirectly accelerating material aging. This parameter provides a direct basis for cleaning and maintenance, avoiding hidden degradation caused by dust accumulation. KDB refers to the temperature stress coefficient, which is the cumulative stress level when the module's operating temperature deviates from standard test conditions (e.g., 25℃). Its value ranges from 0 to 1 (0 represents no stress, 1 represents extreme temperature stress). The module temperature KDBA can be collected in real time using a platinum resistance temperature sensor attached to the module's backsheet and calculated using the following formula: ; In the above formula, 50℃ is the limit temperature difference that the component can withstand for a long time. When it exceeds 50℃, it is calculated as 1. The introduction of the temperature stress coefficient KDB reflects the fatigue damage to component materials caused by temperature fluctuations. High temperatures accelerate the cross-linking and aging of EVA film and backsheet oxidation, while low temperatures lead to embrittlement of the glass and frame sealant. This parameter quantifies the latent damage of temperature to components and provides a key basis for assessing material aging rates. KDC refers to the dynamic aging factor, which is the aging coefficient that takes into account the combined effects of dynamic factors such as light, humidity, and mechanical vibration on the module during actual operation. Its value ranges from 0 to 1 (0 indicates no dynamic aging, and 1 indicates severe dynamic aging). It can be calculated based on the module's operating data (irradiance, humidity, and wind speed) using the following formula: ; In the above formula, KDCA is the daily average irradiance, and the unit is W / m² (watts per square meter). The benchmark irradiance under the standard test conditions of the photovoltaic industry can be 1000 W / m² (i.e., standard light intensity). Dividing by 1000 is to normalize the irradiance to the dimensionless range of 0-1, so as to ensure that it can be directly added to other dimensionless parameters in the formula (such as humidity and wind speed normalized values) and avoid weight imbalance due to dimensional differences. In the above formula, KDCB is the daily average relative humidity (%, normalized to 0-1), which is the average relative humidity over a single day of 24 hours. It needs to be normalized to 0-1 after averaging the humidity data throughout the day (for example, 50% humidity corresponds to KDCB=0.5). If the humidity fluctuates between 40% and 80% on a certain day, KDCB=60%, and after normalization, KDCB is 0.6. The daily average relative humidity KDCB is used for long-term aging trend assessment, reflecting the chronic damage of humidity to materials. In the above formula, KDCC is the daily average wind speed, and the unit is m / s (meters per second). The maximum tolerable wind speed usually considered in the design of distributed photovoltaic power stations is 10 m / s (corresponding to level 6 wind. Exceeding this wind speed may cause vibration fatigue of the support structure). Dividing by 10 is to normalize the wind speed to the dimensionless range of 0-1, which is consistent with the normalization logic of irradiance and humidity, and ensures that the contribution weight of wind speed to aging is within a reasonable range. The introduction of dynamic aging factor KDC breaks through the limitations of traditional static aging assessment. It comprehensively considers the dynamic damage of components caused by light intensity (accelerating light-induced degradation), humidity (accelerating hydrolysis reaction), and wind speed (mechanical vibration fatigue), which is closer to the aging law under actual operating environment and improves the authenticity of degradation assessment. KDD refers to the operating life factor, which is the cumulative operating time of photovoltaic modules from commissioning to the present, in years. It can be directly read from the equipment ledger of the power plant operation and maintenance management system. It is calculated based on the difference between the module commissioning date and the current date, and then normalized, as shown in the following formula: ; In the above formula, YY represents the operating years; In the above formula, 25 represents 25 years, because 25 years is the standard design life of photovoltaic modules; anything exceeding 25 years is calculated as 1. The introduction of the service life factor KDD reflects the impact of time accumulation on the aging of modules. The performance of module materials (such as silicon wafers and encapsulation films) will naturally degrade with the service life. This parameter provides a basic time dimension for assessing natural aging and is a necessary reference for degradation assessment. KDE refers to the Tilt Dynamic Offset Index, which is a normalized value of the dynamic offset angle of the component installation tilt angle relative to the initial design value. It reflects the stability of the support mechanical structure and has a value range of 0-1 (0 indicates no offset, 1 indicates severe offset). The offset angle can be calculated by comparing the current tilt angle KDEA with the initial installation tilt angle KDEB in real time using a dual-axis tilt sensor installed at the bottom of the component support, as shown in the following formula: KDES = |KDEA - KDEB|; Then, normalization is performed, as shown in the following formula: ; In the above formula, 3° is the safety offset threshold for the stent; if it exceeds this threshold, it is calculated as 1. The introduction of the tilt dynamic offset index KDE reflects the mechanical deformation of the support caused by foundation settlement, soil loosening, and strong wind load. Tilt offset will reduce the light receiving efficiency of the module, while aggravating the stress concentration at the connection between the frame and the back sheet, accelerating the aging of the sealant and cracking of the back sheet. This parameter links the mechanical structure stability with the module attenuation, filling the blind spot of traditional focus only on electrical performance. KDF refers to the terminal contact resistance increment factor, which is the dynamic increment of the component output terminal contact resistance relative to the initial value. It reflects the degree of deterioration of the electrical connection and has a value range of 0-1 (0 indicates no increment, 1 indicates severe deterioration). The increment can be calculated monthly by measuring the terminal resistance KDFA and the initial value KDFB using a DC micro-resistance tester integrated in the combiner box, as shown in the following formula: KDFS = KDFA - KDFB; The normalization process is then performed, as shown in the following formula: ; In the above formula, 50mΩ is the safe incremental threshold for contact resistance; if it exceeds this threshold, it is calculated as 1. The introduction of the terminal contact resistance increment factor KDF reflects poor contact caused by oxidation, insertion and removal wear, and rainwater erosion. Increased contact resistance can lead to localized heating, accelerate yellowing of the EVA film around the terminal and aging of the junction box plastic, and even cause the terminal to melt. This parameter quantifies the latent degradation of electrical connections and provides early warning for preventing DC side short circuit faults. K1, K2, K3, K4, and K5 respectively represent the weighting coefficients for dynamic dust coverage, temperature stress coefficient, dynamic aging factor, tilt angle dynamic offset index, and terminal contact resistance.

[0021] Based on the above, this module's health submodule quantifies the performance degradation of photovoltaic modules during long-term operation from multiple dimensions, including environmental erosion, material aging, mechanical stress, and electrical connections. It constructs a quantitative assessment benchmark for the health status of the module. Traditional assessments often focus on environmental factors such as light and temperature. However, this module's health submodule integrates implicit parameters such as dynamic tilt offset (mechanical stability) and terminal contact resistance increment (electrical connection degradation) to directly link the health of the mechanical structure and the state of electrical connections with the degradation of photovoltaic performance, thereby achieving a comprehensive assessment of the module's degradation throughout its entire life cycle and across multiple physical fields. Photovoltaic module degradation is the fundamental cause of failure. The quantitative result of the module's comprehensive degradation coefficient KD provides the core input for the potential risk sub-module, ensuring that the risk assessment is based on the actual health status of the module and avoiding unrealistic risk discussions that are divorced from actual degradation.

[0022] Please see Figure 1 , Figure 2 , Figure 3 and Figure 4 The potential risk submodule is as follows: B1. By analyzing the degree of fluctuation of the output current of photovoltaic modules within a unit time, the electrical characteristics are used as the basis for early fault warning, and the current fluctuation coefficient is calculated. B2. By analyzing the area ratio of abnormally high-temperature regions in the infrared thermal imaging of the photovoltaic module surface, the risk of local overheating can be reflected, and the proportion of infrared anomalies can be calculated. B3. By analyzing the real-time relative humidity of the environment where the photovoltaic modules are located, the water vapor content in the atmosphere can be reflected to calculate the ambient humidity; B4. By comparing the humidity inside and outside the photovoltaic module, the dynamic permeability of water vapor to the backsheet of the photovoltaic module in actual operation is analyzed to reflect the degree of deterioration of the insulation performance of the backsheet, so as to calculate the water vapor transmission rate of the backsheet. B5. By comparing the current grounding resistance of the photovoltaic module with the initial grounding resistance, the change in grounding resistance is calculated to reflect the safe discharge capacity of the grounding system. This allows for the calculation of the dynamic increment factor of grounding resistance, which, combined with the module's comprehensive attenuation coefficient, outputs a potential fault risk index. The calculation formula for the potential risk submodule is as follows: ; in: RF stands for Potential Failure Risk Index, which is used to quantify the potential risk of component failure (such as hot spots, short circuits and insulation failure) by comprehensively considering factors such as component degradation status, electrical anomalies and environmental corrosion. It provides a priority basis for fault early warning and operation and maintenance resource allocation. The introduction of the component comprehensive degradation coefficient KD serves as the basis for fault risk assessment. Components with severe degradation are already in a sub-healthy state and are more susceptible to failure triggered by external factors. It is the core weight item for risk assessment. RFA stands for Current Fluctuation Factor, which is the degree of fluctuation of the component's output current per unit time, reflecting current stability. Its value ranges from 0 to 1 (0 represents no fluctuation, 1 represents severe fluctuation). It can be calculated by acquiring the real-time current It using the Hall effect current sensor built into the string inverter, and then calculating the ratio of the current standard deviation to the average value over one minute, as shown in the following formula: RFA = σIt ÷ μIt; In the above formula, σIt refers to the standard deviation of the current within 1 minute; In the above formula, μIt refers to the average current over 1 minute; The introduction of the current fluctuation coefficient RFA reflects the abnormal current fluctuations caused by factors such as shading, microcracks, and loose wiring in the components. Severe current fluctuations can cause local overheating and inverter MPPT tracking failure, which are direct precursors to faults (such as hot spots and broken grids), providing electrical characteristic basis for early fault warning. RFB stands for Infrared Anomaly Ratio, which is the proportion of abnormally high-temperature areas in infrared thermal imaging of the component surface. It reflects the risk of local overheating and has a value range of 0-1 (0 indicates no anomaly and 1 indicates complete anomaly). The component surface temperature field image is captured regularly (once a month) by an infrared thermal imager mounted on a drone. The threshold segmentation method (the abnormal temperature threshold can be set to ambient temperature + 20℃) is used to identify abnormally high-temperature areas and calculate the ratio of the abnormal temperature area to the total area of ​​the component. The introduction of infrared anomaly ratio (RFB) directly reflects the local overheating caused by internal component faults (such as microcracks, loose solder joints, and diode failures). Infrared anomalies are the intuitive physical manifestations of faults, providing key information for locating specific fault locations and types. RFC stands for ambient humidity, which is the real-time relative humidity of the environment in which the component is located. It reflects the water vapor content in the atmosphere and has a value range of 0-1 (0 represents dryness and 1 represents saturated humidity). It is the real-time relative humidity at the current moment. It can be directly collected by the sensor and normalized to 0-1. Then, the obtained percentage value is divided by 100 to obtain a dimensionless value between 0 and 1. Ambient humidity RFC serves as an immediate fault risk warning and reflects the acute triggering effect of humidity on the current fault. The impact of humidity on component failure is immediate; sudden high humidity (such as H=0.95 after a heavy rain) may cause: A water film condenses on the surface of the terminals, accelerating oxidation and increasing contact resistance. A sudden increase in humidity inside the component in a short period of time may trigger an acute onset of potential-induced degradation. The electrochemical corrosion rate of metal components (such as frames and grounding terminals) increases instantaneously under high humidity; the ambient humidity RFC captures the current humidity status through real-time values, directly reflecting the triggering effect on the immediate risk of component failure, and providing a key basis for assessing whether a failure may occur at present; Temperature and humidity data can be collected directly from the temperature and humidity sensors deployed at the power station's meteorological station. The sensors are installed at the middle height of the component array (1.5m above the ground) to avoid obstruction and direct sunlight. RFD refers to the backsheet moisture vapor transmission rate, which is the dynamic ability of the module's backsheet to allow moisture to permeate during actual operation. It reflects the degree of degradation of the backsheet's insulation performance and has a value range of 0-1 (0 indicates no permeation, and 1 indicates severe permeation). A miniature humidity sensor can be pre-installed on the inside of the backsheet during module encapsulation to monitor the internal humidity HIN in real time and compare it with the ambient humidity, as shown in the following formula: RFD = (HIN - HBASE) ÷ (Ambient humidity - HBASE); In the above formula, HBASE is the internal reference humidity of the component when it leaves the factory, which is usually less than 5%RH. If it exceeds 1, it is calculated as 1. The introduction of backsheet moisture permeability (RFD) reflects the degree of failure of the backsheet as a waterproof barrier for the module. Moisture penetration can cause internal circuits to become damp and aluminum frame to corrode. It is a core parameter for assessing the internal insulation degradation of the module and fills the blind spot that traditional visual inspection cannot detect internal moisture. RFE refers to the dynamic increment factor of grounding resistance, which is the normalized value of the dynamic increment of the grounding resistance of the photovoltaic module / equipment grounding circuit relative to the initial value. It reflects the safe current discharge capacity of the grounding system and has a value range of 0-1 (0 indicates no increment, 1 indicates severe failure). It can be calculated by real-time acquisition of the current grounding resistance RFEA through an online grounding resistance monitoring module installed at the combiner box or inverter grounding terminal, comparing it with the initial grounding resistance RFEB, as shown in the following formula: RFES = RFEA - RFEB; The normalization process is shown in the following formula: ; In the above formula, 6Ω is the safe increment threshold for grounding resistance; if it exceeds this threshold, it is calculated as 1. The introduction of the dynamic increment factor RFE of grounding resistance reflects the decrease in the leakage capacity of the grounding system due to oxidation of the grounding electrode, damage to cable insulation and dry soil. Increased grounding resistance can lead to protection failure (risk of electric shock) during leakage faults and ground potential backflash during lightning strikes (risk of equipment damage). It is a core safety indicator to ensure the safe operation of the power station. R1, R2, R3, R4, and R5 represent the weighting factors for the current fluctuation coefficient, infrared anomaly ratio, ambient humidity, backplane water vapor transmittance, and grounding resistance dynamic increment factor, respectively.

[0023] Based on the above, this potential risk submodule uses the component's comprehensive attenuation coefficient KD as a basis, and integrates fault precursors such as electrical anomalies (current fluctuations and infrared hot spots), insulation degradation (backsheet moisture penetration and grounding resistance increase), and environmental erosion (humidity) to quantify the potential risk level of the component to cause obvious faults (such as hot spots, short circuits and insulation breakdown). Traditional operation and maintenance relies on post-fault repair or regular inspections, while this potential risk submodule upgrades the operation and maintenance mode from passive response to proactive early warning by integrating latent fault precursors (such as backplane moisture permeability reflecting internal insulation degradation and grounding resistance increment reflecting a decrease in safe leakage capacity), thus identifying latent risks that have not yet manifested as obvious faults in advance. A single parameter anomaly (such as current fluctuation) may be caused by a variety of reasons. This potential risk submodule distinguishes between risks caused by attenuation, risks caused by the environment, and risks caused by electrical connections by coupling the component comprehensive attenuation coefficient KD with each fault precursor through calculation, thereby avoiding misjudgment of a single parameter and improving the accuracy of early warning.

[0024] Please see Figure 1 , Figure 2 , Figure 3 and Figure 4 The operation and maintenance analysis submodule is as follows: C1. By analyzing the proportion of photovoltaic modules in the total power generation of the power plant, we can reflect their importance to the power generation of the power plant and calculate their contribution to power generation. C2. By analyzing the severity of the safety consequences caused by photovoltaic module failures, the safety risk level is reflected, and values ​​are assigned for various scenarios to calculate the safety hazard level. C3. By calculating the difference between the normal operating voltage and the fault simulation voltage of the photovoltaic module, the voltage fluctuation is obtained, and combined with the maximum allowable fluctuation of the distribution network, the potential impact of photovoltaic module faults on voltage stability is analyzed, so as to calculate the grid voltage fluctuation coupling coefficient. C4. Calculate the accessibility coefficient of maintenance resources by analyzing the standardized time cost for maintenance personnel to reach the photovoltaic module location from the nearest maintenance station; C5. By analyzing the normalized value of the average repair time of historical faults of the same type of photovoltaic modules, the average repair time of historical faults is calculated, and combined with the comprehensive degradation coefficient of the module and the potential fault risk index, the safety operation and maintenance priority index is calculated. The calculation formula for the operation and maintenance analysis submodule is as follows:

[0025] in: PT stands for Safety Operation and Maintenance Priority Index, which quantifies the urgency of component operation and maintenance by taking into account factors such as fault risk, power generation loss, safety consequences, grid impact and operation and maintenance feasibility. It provides decision support for operation and maintenance resource scheduling and task sequencing, ensuring that high-priority tasks are processed first. The introduction of the Potential Failure Risk Index (RF) serves as a core basis for operational prioritization. Components with high failure risk should be addressed first to prevent the failure from escalating and causing downtime or safety incidents. The introduction of the component's overall degradation coefficient KD, coupled with its power generation contribution, reflects the actual impact of degradation on power generation loss. Components with severe degradation and high power generation contribution have higher operation and maintenance value and should be prioritized. PTA refers to the power generation contribution, which is the proportion of photovoltaic modules in the total power generation of a power plant. It reflects the importance of the modules to the power generation of the power plant and has a value range of 0-1 (0 represents no contribution, and 1 represents the maximum contribution). It can be calculated based on string-level power monitoring data to determine the proportion of power generation of a single module to the total power generation of the array, as shown in the following formula: PTA = PTAA ÷ PTAB; In the above formula, PTAA represents the daily power generation of a single module (kWh). In the above formula, PTAB represents the total daily power generation of the array (kWh). The introduction of the Power Generation Contribution Amount (PTA) distinguishes the power generation importance of modules. If a module located in a high-irradiance area and with good orientation (high PTA value) fails, it will result in greater power generation loss. This parameter ensures that operation and maintenance resources are tilted towards high-value modules. PTB refers to the safety hazard level, which is the severity of the safety consequences that a component failure may cause. It reflects the level of safety risk and ranges from 0 to 1 (0 indicates no safety hazard, and 1 indicates an extremely serious safety hazard). The value can be based on the component type (e.g., whether it contains a flammable backsheet), installation location (e.g., whether it is near residential areas or flammable and explosive locations), and failure type (e.g., whether it involves high voltage or fire risk). The value can be manually assigned by maintenance personnel (e.g., 1.0 for rooftop power station components and 0.5 for remote ground-mounted power stations). This example provides several assignment scenarios: For rooftop power stations (residential / commercial buildings), with components containing flammable backsheets (such as TPT), PTB = 1.0 when near distribution boxes / gas pipelines; When dry grass / straw is piled under the components of an agricultural-solar hybrid power station, or when the greenhouse is covered with plastic film (PE / PVC), the PTB is 0.9. In industrial and commercial power plants, where components are located near production lines (such as chemical fiber or electronics workshops), a fault could cause the production line to shut down, resulting in a PTB of 0.8. For ground-mounted power stations, when the component array is close to a high-voltage line (10kV and above), and a fault may cause a short circuit in the line, PTB = 0.7. For ground-mounted power plants (in remote areas), where there are no flammable materials and they are far from residential areas, but the modules contain glass / metal frames, PTB = 0.5. When the solar-aquaculture hybrid power station's components float on the water surface, with no personnel activity below and no flammable materials nearby, PTB = 0.3. For off-grid power plants (such as remote communication base stations), with small component power (≤1kW), and operating independently without being connected to the public power grid, PTB = 0.2; When a decommissioned component is to be removed, has been de-energized and disconnected from the system, and has no power output, PTB = 0. The introduction of the Safety Hazard Level (PTB) reflects the differences in the safety consequences of faults. For example, a fault in a component near a residential area may cause personal injury or death and should be dealt with first to avoid secondary disasters. This is a direct manifestation of the safety priority principle in operation and maintenance decisions. PTC refers to the grid voltage fluctuation coupling coefficient, which is the potential impact of photovoltaic module failures on the voltage stability of the grid connection point. It reflects the risk of source-grid interaction and has a value range of 0-1 (0 indicates no impact, 1 indicates severe impact). It is calculated by collecting voltage data through voltage sensors installed at the grid connection point and combining it with module failure simulations (such as string open circuits). The voltage fluctuation PTCA is the ratio of this PTCA to the maximum allowable fluctuation PTCB (±7% of rated voltage) of the distribution network. Voltage fluctuation amount PTCA; Normal operating voltage: The grid connection point voltage under normal operating conditions is collected by a grid connection point voltage sensor; Fault simulation voltage: The grid connection point voltage when component faults (such as string open circuit and short circuit) are simulated by the power plant energy management system or simulation software, and then the fluctuation is calculated, which is the absolute value of the difference between the normal operating voltage and the fault simulation voltage. The calculation formula is as follows: PTC = PTCA / PTCB; The introduction of the grid voltage fluctuation coupling coefficient PTC in high-penetration distributed power stations is crucial because component failures may lead to voltage overruns (such as three-phase imbalance), resulting in penalties from the grid dispatching department or user complaints. This parameter binds component safety to grid stability, ensuring that operation and maintenance decisions take grid-side constraints into account. PTD stands for Accessibility Factor, which is the standardized time cost for maintenance personnel to reach the component location from the nearest maintenance station. It reflects the time feasibility of maintenance and ranges from 0 to 1 (0 indicates excellent accessibility, and 1 indicates extremely poor accessibility). Based on GIS system route planning, the ratio of actual arrival time PTDA (minutes) to standard response time PTDB (30 minutes) can be obtained, as shown in the following formula:

[0026] PTD = PTDA ÷ PTDB; When the result of the above calculation exceeds 2, it is calculated as 2, reflecting the extremely difficult situation to reach; In the above formula, the actual arrival time (PTDA) refers to the actual travel time (in minutes) required for maintenance personnel to travel from the nearest maintenance station (or warehouse) to the location of the component (such as the roof or field), including road travel, site access (such as factory registration and roof climbing), etc., reflecting geographical distance and site accessibility. In the above formula, the standard response time PTDB is a preset reasonable arrival time benchmark (such as 30 minutes). Combined with the average distribution density of power plants, its function is to convert the arrival time into a dimensionless difficulty coefficient, so as to avoid the distortion of priority assessment due to the difference in the location of power plants. The introduction of the Operation and Maintenance Resource Accessibility Factor (PTD) takes into account the differences in operation and maintenance difficulty caused by the dispersion of distributed power stations. For example, if the component failure of a power station in a remote mountainous area is not prioritized, the failure may be amplified due to response delay. This parameter ensures that high-risk and hard-to-reach components get resources first, avoiding operation and maintenance blind spots. PTE refers to the historical average fault repair time, which is the normalized value of the average repair time of historical faults of components of the same model or in the same region. It reflects the technical difficulty of operation and maintenance, and the value ranges from 0 to 1 (0 indicates that the repair is very simple and 1 indicates that the repair is very difficult). The repair time of the component / region faults in the past two years (from the time of reporting the fault to the time of restoration of operation) can be extracted from the operation and maintenance work order system, and the average value PTEA is calculated and normalized as follows: ; In the above formula, 4 hours is the standard repair time; anything exceeding that is calculated as 1 hour. The introduction of the historical average repair time (PTE) reflects the complexity of component failure repair. For example, imported components with special parts (high average repair time) require advance resource allocation. This parameter ensures that maintenance decisions take into account the actual repair difficulty and avoid secondary delays due to insufficient resources. P1, P2, P3, P4, P5, and P6 respectively refer to the weighting factors of the potential fault risk index, the comprehensive component attenuation coefficient, the power generation contribution, the safety hazard level, the grid voltage fluctuation coupling coefficient, the operation and maintenance resource accessibility coefficient, and the average historical fault repair time.

[0027] Based on the above, this operation and maintenance analysis submodule comprehensively considers factors such as potential fault risks, power generation value (power generation contribution), safety consequences (safety hazard level), grid impact (voltage fluctuation coupling), and operation and maintenance feasibility (repair time and resource accessibility) to quantify the urgency and priority of component operation and maintenance tasks. Distributed photovoltaic power stations have a large number of modules and are scattered. Traditional operation and maintenance (O&M) often leads to the problem of over-concentrating resources on low-risk modules or neglecting high-risk modules. This O&M analysis submodule uses multi-dimensional weighting (such as prioritizing modules with high safety hazard levels and high power generation contributions) to tilt O&M resources toward high-risk, high-value, and high-consequence tasks, thereby improving resource utilization efficiency. Operation and maintenance decisions need to balance safety (avoiding accidents), efficiency (reducing downtime), and economy (reducing costs). This operation and maintenance analysis submodule integrates the level of safety hazards (safety dimension), power generation contribution (economic dimension), and repair time (efficiency dimension) to ensure that decisions not only focus on the fault itself, but also take into account the overall benefits of operation and maintenance, avoiding the one-sidedness of repairing for the sake of repairing.

[0028] It is worth noting that this embodiment provides an iterative approach, further calculating the potential failure risk index RF of the potential risk submodule to iterate the terminal contact resistance weighting coefficient K5 in the component health submodule, thereby optimizing and adjusting the influence of the terminal contact resistance increment KDF, so that the overall system achieves the goal of cyclic optimization. The specific iterative process is as follows: K5 n =min(K5) n－1 +α×RF n－1 K5 max ); in: K5 n This refers to the weighting coefficient of the terminal contact resistance after the nth iteration. K5 n－1 This refers to the weighting coefficient of the terminal contact resistance after the (n-1)th iteration. RF n－1 Refers to the potential failure risk index after the (n-1)th iteration; α refers to the weight adjustment coefficient, which is set to 0.05 in this embodiment. This value is based on domain experience to ensure that the weight changes are stable. It is worth noting that an iteration termination condition is also set, and the iteration convergence is based on two termination conditions. The iteration terminates when either of the following two conditions is met. Condition 1: The number of iterations reaches the upper limit. In this embodiment, the maximum number of iterations is set to 5 to avoid the iteration from getting stuck in an infinite loop and to ensure the real-time performance of the system. Condition 2: ∣K5 n－1 -K5 n | <0.01 indicates that the weight adjustment has stabilized; Based on the above, through iteration, the component comprehensive attenuation coefficient KD of the component health submodule is no longer independent of the fault risk, but forms a closed-loop feedback with the potential fault risk index RF of the potential risk submodule. For components with higher fault risk, the contribution weight of the terminal contact resistance increment to attenuation is dynamically increased, making the attenuation assessment closer to the actual physical process of the attenuation accelerated by the fault precursor factors, and avoiding the problem that the impact of electrical connection degradation is underestimated in the traditional fixed weight model. For components in the early stages of failure (with a moderate but continuously rising potential failure risk index RF), the iterative process will gradually increase the weight of the contact resistance increment, so that the component's overall attenuation coefficient KD reflects the cumulative effect of electrical connection degradation earlier. This will further increase the potential failure risk index RF in the potential risk sub-module, forming an enhanced mechanism of implicit degradation, weight enhancement, attenuation prominence, and risk warning, thus solving the problem of the traditional model's lagging response to early failure precursors. The iterative component comprehensive attenuation coefficient KD incorporates fault risk feedback, making the severity of attenuation more safety-oriented. The attenuation of high-fault-risk components is given higher weight, and thus the coupling effect of attenuation and risk of such components is more significant in the operation and maintenance analysis submodule. This enables operation and maintenance decisions to prioritize key components with both high attenuation and fault risk, avoiding resource misallocation.

[0029] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A distributed photovoltaic power station intelligent safety operation and maintenance management system, characterized in that, include: Data acquisition module: used to collect component health correlation data, risk and fault correlation data, and component operation and maintenance correlation data of distributed photovoltaic power stations; Data processing module: Used to input component health correlation data, risk and fault correlation data and component operation and maintenance correlation data obtained by the data acquisition module, clean the input data, and then input the cleaned data into the analysis component; Calculation and Analysis Module: Based on the dynamic dust coverage, temperature stress coefficient, dynamic aging factor, operating years factor, tilt angle dynamic offset index, and terminal contact resistance increment factor in the component health correlation data, the comprehensive degradation coefficient of the component is output. The comprehensive degradation coefficient of the component is used to quantify the degree of performance degradation of the photovoltaic module in long-term operation. Based on the current fluctuation coefficient, infrared anomaly ratio, backsheet water vapor transmittance and grounding resistance dynamic increment factor in the risk fault correlation data, and combined with the module comprehensive attenuation coefficient, a potential fault risk index is output. The potential fault risk index is used to quantify the potential risk of photovoltaic module failure, and to provide a priority basis for fault early warning and operation and maintenance resource allocation. Based on the power generation contribution, safety hazard level, grid voltage fluctuation coupling coefficient, operation and maintenance resource accessibility coefficient, and historical average fault repair time in the component operation and maintenance related data, and combined with the component comprehensive attenuation coefficient and potential fault risk index, a safe operation and maintenance priority index is output. The safe operation and maintenance priority index is used to quantify the urgency of photovoltaic module operation and maintenance, and provide decision support for operation and maintenance resource scheduling and task sequencing. Management Execution Module: Used to input the component comprehensive attenuation coefficient, potential failure risk index and security operation and maintenance priority index output by the calculation and analysis module, and to perform operation and maintenance priority classification and operation and maintenance resource allocation based on the input data.

2. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 1, characterized in that: The calculation and analysis module includes a component health submodule, a potential risk submodule, and an operation and maintenance analysis submodule.

3. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 2, characterized in that: The processing procedure of the component health submodule is as follows: A1. By analyzing the real-time area ratio of dust coverage on the surface of photovoltaic modules, the degree of dust obstruction of the module's light absorption efficiency can be reflected, so as to calculate the dynamic dust coverage rate. A2. By collecting real-time temperature data of the photovoltaic module backsheet, we can analyze the fatigue damage of temperature fluctuations to the photovoltaic module materials and calculate the temperature stress coefficient. A3. By combining the daily average irradiance, daily average relative humidity and daily average wind speed of the environment in which the photovoltaic module is located, the aging of the photovoltaic module under the combined influence of light, humidity and mechanical vibration factors in actual operation is considered, so as to calculate the dynamic aging factor. A4. By analyzing the cumulative operating time of photovoltaic modules from commissioning to the present, the impact of time accumulation on module aging is reflected, so as to calculate the operating life factor; A5. By analyzing the comparison between the current tilt angle of the photovoltaic module and the initial installation tilt angle, the stability of the mechanical structure of the photovoltaic module support can be reflected, and the dynamic tilt angle offset index can be calculated. A6. By measuring the change in terminal resistance from its initial value within the combiner box, the degree of electrical connection degradation is reflected, and the terminal contact resistance increment factor is calculated, thereby ultimately outputting the overall component attenuation coefficient.

4. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 3, characterized in that: The processing procedure for the potential risk submodule is as follows: B1. By analyzing the degree of fluctuation of the output current of photovoltaic modules within a unit time, the electrical characteristics are used as the basis for early fault warning, and the current fluctuation coefficient is calculated. B2. By analyzing the area ratio of abnormally high-temperature regions in the infrared thermal imaging of the photovoltaic module surface, the risk of local overheating can be reflected, and the proportion of infrared anomalies can be calculated. B3. By analyzing the real-time relative humidity of the environment where the photovoltaic modules are located, the water vapor content in the atmosphere can be reflected to calculate the ambient humidity; B4. By comparing the humidity inside and outside the photovoltaic module, the dynamic permeability of water vapor to the backsheet of the photovoltaic module in actual operation is analyzed to reflect the degree of deterioration of the insulation performance of the backsheet, so as to calculate the water vapor transmission rate of the backsheet. B5. By comparing the current grounding resistance of the photovoltaic module with the initial grounding resistance, the change in grounding resistance is calculated to reflect the safe current discharge capacity of the grounding system. This allows for the calculation of the dynamic increment factor of grounding resistance, which, combined with the module's comprehensive attenuation coefficient, outputs a potential fault risk index.

5. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 4, characterized in that: The processing procedure of the operation and maintenance analysis submodule is as follows: C1. By analyzing the proportion of photovoltaic modules in the total power generation of the power plant, we can reflect their importance to the power generation of the power plant and calculate their contribution to power generation. C2. By analyzing the severity of the safety consequences caused by photovoltaic module failures, the safety risk level is reflected, and values ​​are assigned for various scenarios to calculate the safety hazard level. C3. By calculating the difference between the normal operating voltage and the simulated fault voltage of the photovoltaic module, the voltage fluctuation is obtained, and combined with the maximum allowable fluctuation of the distribution network, the potential impact of photovoltaic module faults on voltage stability is analyzed, so as to calculate the grid voltage fluctuation coupling coefficient. C4. Calculate the accessibility coefficient of maintenance resources by analyzing the standardized time cost for maintenance personnel to reach the photovoltaic module location from the nearest maintenance station; C5. By analyzing the normalized value of the average repair time of historical faults of photovoltaic modules of the same model, the average repair time of historical faults is calculated. Combined with the comprehensive degradation coefficient of the module and the potential fault risk index, the safety operation and maintenance priority index is calculated.

6. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 1, characterized in that: The specific operation and maintenance priority levels in the management execution module are as follows: A security operation and maintenance priority index of ≥0.7 indicates high priority and requires immediate intervention; A security operation and maintenance priority index of 0.3 ≤ security operation and maintenance priority index < 0.7 indicates medium priority, and a weekly maintenance plan should be developed. A security operation and maintenance priority index of <0.3 indicates low priority and should be included in the monthly routine inspection.

7. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 6, characterized in that: The allocation of operation and maintenance resources in the management execution module is specifically as follows: When it is in a high priority state; Resource allocation: dispatch personnel from the nearest maintenance station and equip them with specialized tools; Intervention measures: Immediately stop the machine for inspection, focusing on investigating areas of infrared anomalies, and inspect and repair the grounding electrode and cables; Closed-loop verification: Retest the potential failure risk index within 24 hours after the repair to ensure it drops below 0.3, and update the overall component attenuation coefficient to the database simultaneously; When in medium priority; The plan is developed and incorporated into the weekly maintenance plan, with the task list arranged in descending order of security operation and maintenance priority index; Targeted measures include checking for damage to the backsheet and repairing it if the backsheet is damaged, and adjusting the bracket fastening bolts and re-measuring the tilt angle for components with excessive tilt angle deviation. Track the effects and retest relevant parameters weekly until the overall attenuation coefficient of the component stabilizes and decreases. When at low priority: Routine monitoring, included in monthly inspections, with a focus on dynamic dust coverage and temperature stress coefficient; Data accumulation: Continuously collect parameters and update the overall component attenuation coefficient. When the security operation and maintenance priority index increases and exceeds 0.3, it will be automatically upgraded to medium priority.

8. The intelligent safety operation and maintenance management system for distributed photovoltaic power stations according to claim 1, characterized in that: The data processing module cleans the input data, specifically by: Outlier cleaning: Abnormal data caused by sensor malfunctions are removed using the 3σ principle, and the removed data is replaced with the average of the previous and next 10 minutes. Normalization and unification: Normalize all parameters to the 0-1 range according to the formula requirements to ensure that parameters with different dimensions can be calculated in a weighted manner; Time synchronization: unify the timestamps of all sensor data.

Images