A new energy photovoltaic module health assessment method and system based on intelligent monitoring

Abstract

The application discloses a new energy photovoltaic module health evaluation method and system based on intelligent monitoring, relates to the photovoltaic equipment evaluation technical field, and comprises the following steps: constructing a three-dimensional photovoltaic module structure model according to photovoltaic structure information and photovoltaic electrical information of the photovoltaic module, and dividing the three-dimensional photovoltaic module structure model into a photovoltaic structure model and a photovoltaic electrical model according to structure and electricity; linkage of the three-dimensional photovoltaic module structure model; establishment of photovoltaic structure defect grading and photovoltaic electrical diagnosis grading; linkage analysis of multiple models; and construction of a photovoltaic linkage evaluation model to evaluate the health of the photovoltaic module. In the application, the real-time changing conversion efficiency and the state of the photovoltaic module are displayed through linkage of the three-dimensional photovoltaic module structure model, the photovoltaic structure model and the photovoltaic electrical model, and the health of the photovoltaic module is monitored in real time according to the photovoltaic linkage evaluation model, so that traditional discrete and lagging fault troubleshooting is changed into continuous and predictive comprehensive health state evaluation.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic equipment evaluation technology, specifically to a health evaluation method and system for new energy photovoltaic modules based on intelligent monitoring. Background Technology

[0002] Photovoltaic power generation is a technology that uses the photovoltaic effect to directly convert sunlight into electrical energy. Its core equipment is photovoltaic modules. Multiple modules form an array, and the generated direct current is converted into alternating current by an inverter and fed into the power grid or used by loads.

[0003] Traditional photovoltaic module health assessments often rely on manual equipment inspections or diagnosing equipment status based on changes in power generation. This single-dimensional assessment method has limitations and may be unable to address changes in photovoltaic modules under different conditions, potentially leading to inaccurate assessments of the module's health status. Summary of the Invention

[0004] To address the aforementioned technical issues, this paper provides a health assessment method and system for new energy photovoltaic modules based on intelligent monitoring. This technical solution solves the problem mentioned in the background that traditional photovoltaic module health assessments often rely on manual equipment inspection or rely on changes in power generation to diagnose equipment status. The single-dimensional assessment method has certain limitations and may be unable to cope with changes in photovoltaic modules under different conditions, which may lead to incorrect assessments of the health status of photovoltaic modules.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A health assessment method and system for new energy photovoltaic modules based on intelligent monitoring, comprising: Acquire photovoltaic structure and electrical information of photovoltaic modules, as well as photovoltaic usage and installation environment information; Based on photovoltaic usage information and installation environment information, a historical photovoltaic module health matrix is ​​constructed; A three-dimensional photovoltaic module structural model is constructed using the photovoltaic structural information and photovoltaic electrical information of the photovoltaic module. The three-dimensional photovoltaic module structural model is then divided to obtain the photovoltaic structural model and the photovoltaic electrical model. By combining historical photovoltaic module health matrix and photovoltaic structure model, photovoltaic modules are inspected in stages, and a photovoltaic structure defect classification is established; By combining photovoltaic usage information, installation environment information, and photovoltaic electrical models, a hierarchical inspection of photovoltaic modules is conducted to establish a photovoltaic electrical diagnostic classification. By integrating photovoltaic structural defect classification with photovoltaic electrical diagnosis classification, a photovoltaic linkage assessment model is established. The photovoltaic linkage assessment model is then corrected based on photovoltaic usage information and real-time photovoltaic information, and the health of photovoltaic modules is then assessed in real time through the photovoltaic linkage assessment model.

[0006] Preferably, the acquisition of photovoltaic structure information and photovoltaic electrical information of the photovoltaic module, as well as photovoltaic usage information and installation environment information, specifically includes: The photovoltaic usage information includes the historical usage time and historical failure event information of the photovoltaic structure; the installation environment information includes environmental temperature and humidity information and environmental suspended particulate matter concentration. The photovoltaic structural information includes the structural performance report of the photovoltaic module and the real-time structural information of the photovoltaic module; the photovoltaic electrical information includes the original performance report and the real-time electrical information of the photovoltaic module. The structural performance report includes manufacturing data and initial inspection reports of the photovoltaic modules at the time of manufacture; the real-time structural information includes periodic inspection data on surface stains and damage of the photovoltaic modules; the original performance report includes photovoltaic module performance data sheets and grid-connected test and commissioning records; the real-time electrical information includes SCADA system data and real-time IV curve test data; the historical usage time includes the grid-connected operation time and total operation time of the photovoltaic modules; the historical fault event information includes maintenance records, fault records, and periodic cleaning records of the photovoltaic modules; the environmental temperature and humidity information includes annual average temperature, annual average temperature difference, and annual extreme temperature; and the environmental suspended particulate matter concentration includes annual average PM10 concentration and quarterly average PM10 concentration.

[0007] Preferably, the step of constructing a historical photovoltaic module health matrix based on photovoltaic usage information and installation environment information specifically includes: The installation environment information and photovoltaic usage information are cleaned and time-series aligned according to time sequence, and feature factors are extracted from them; The photovoltaic modules are divided according to their structure based on the photovoltaic structure information, and a weight is defined for each photovoltaic structure region as a spatial weight. The characteristic factors are quantified and standardized. A two-dimensional matrix is ​​constructed based on the time-characteristic factors. Spatial weight mapping is introduced, and a historical photovoltaic module health matrix is ​​constructed by combining the spatial weights with the two-dimensional matrix.

[0008] Preferably, the step of constructing a three-dimensional photovoltaic module structural model using the photovoltaic structural information and photovoltaic electrical information of the photovoltaic module, and dividing the three-dimensional photovoltaic module structural model to obtain the photovoltaic structural model and the photovoltaic electrical model, specifically includes: Based on the photovoltaic structure information, a geometric model of the photovoltaic module is constructed using modeling software, which is called a three-dimensional photovoltaic module structure model. The photovoltaic structure information and photovoltaic electrical information are mapped to the three-dimensional photovoltaic module structure model, parameters are assigned to the corresponding geometry, and the power flow direction is defined. Based on photovoltaic structural information and photovoltaic electrical information as the core of the division, the three-dimensional photovoltaic module structural model is segmented according to the material mechanical characteristics and circuit topology data of the photovoltaic module. All geometric bodies are extracted and mapped with material mechanical characteristics, and all circuit topologies are extracted and assigned equivalent electrical parameters to obtain sub-models, which are respectively labeled as photovoltaic structural model and photovoltaic electrical model. A spatial coordinate system is introduced to link the sub-model with the three-dimensional photovoltaic module structure model.

[0009] Preferably, the step of combining historical photovoltaic module health matrices and photovoltaic structure models to conduct phased inspections of photovoltaic modules and establish a photovoltaic structural defect classification specifically includes: Based on the degradation trend of photovoltaic modules in the historical photovoltaic module health matrix, different photovoltaic modules in the photovoltaic structure model are classified and described, and the structural performance report of photovoltaic modules in the photovoltaic structure information is used to check for possible defects in photovoltaic modules; A four-level structural defect impact standard is established, namely L1, L2, L3 and L4. Among them, L1: affects appearance but does not affect normal use; L2: has a significant impact on appearance and may cause power loss or reliability risk; L3: has a significant power loss or safety hazard; L4: has functional failure or causes safety accident. The defects identified during the inspection were integrated and classified according to different photovoltaic modules. At the same time, the defects were graded according to their impact on the function of the photovoltaic modules. A photovoltaic structural defect classification was established based on the photovoltaic module location and the four-level structural defect impact standard. The photovoltaic structural defect classification was then mapped onto a three-dimensional photovoltaic module structural model.

[0010] Preferably, the step of combining photovoltaic usage information, installation environment information, and photovoltaic electrical models to perform hierarchical checks on photovoltaic modules and establish a photovoltaic electrical diagnostic grading system specifically includes: Based on the original performance report and installation environment information of the photovoltaic modules in the photovoltaic electrical information, the electrical defective modules in the photovoltaic modules are marked, and the real-time operating status of the electrical defective modules is monitored based on the photovoltaic usage information; Data generated during real-time operation is collected, cleaned and standardized, and the IV characteristic curve is reconstructed based on the processed data. The single diode model is then fitted to obtain the electrical parameters of the electrically defective component. Based on the comparison between the health benchmark value and electrical parameters in the photovoltaic electrical information, a four-level electrical defect classification is established, namely T1, T2, T3 and T4. Among them, T1: does not affect normal use; T2: may have early degradation or slight contamination; T3: power or parameters have changed significantly or there is a clear electrical fault; T4: there is a serious safety hazard or the module has failed. A photovoltaic electrical diagnostic classification is established based on the four-level electrical defect classification and the electrical parameters of the electrical defect components, and then mapped to a three-dimensional photovoltaic module structural model.

[0011] Preferably, the process of integrating photovoltaic structural defect classification with photovoltaic electrical diagnostic classification to establish a photovoltaic linkage assessment model, and then correcting the photovoltaic linkage assessment model based on photovoltaic usage information and real-time photovoltaic information, thereby enabling real-time assessment of the health of photovoltaic modules through the photovoltaic linkage assessment model, specifically includes: Based on spatial consistency and causal relationship, a fusion rule base is established. By using spatial positioning technology, the specific location information of photovoltaic structural defects is matched with the electrical parameter collection points in the corresponding photovoltaic structural area. The causal relationship mechanism between structural defects and electrical performance abnormalities is analyzed, and fusion judgment rules are constructed. The photovoltaic structural defect classification and photovoltaic electrical diagnosis classification are input into the rule base, and the initial comprehensive health level is obtained by cross-comparing the preset weight coefficients and correlation strength. The weights of the fusion rule base are adjusted based on photovoltaic usage information and real-time structural and electrical information of photovoltaic modules. The adjustment scope includes the association rules, influencing factors and initial weights between different data. Based on the comparative analysis of real-time data and historical data, the weight values ​​of each parameter are dynamically adjusted according to the percentage change in the conversion power of photovoltaic modules. The effect of weight adjustment is verified by combining the actual power generation fluctuations in photovoltaic usage information, and the photovoltaic linkage evaluation model is obtained by iteratively optimizing the fusion rule base. The health of photovoltaic modules is monitored and evaluated in real time based on the obtained photovoltaic linkage assessment model.

[0012] The next step is a health assessment system for new energy photovoltaic modules based on intelligent monitoring, used to implement the above assessment method, characterized by comprising: The information acquisition and processing module is used to read the structural performance report of the photovoltaic module at the time of its manufacture and all information during the use of the photovoltaic module. It cleans and sorts all the information in chronological order to facilitate subsequent reading. At the same time, it establishes an interrelated historical photovoltaic module health matrix based on the photovoltaic structural information and photovoltaic electrical information.

[0013] Furthermore, it also includes: The model building and splitting module is used to build a three-dimensional photovoltaic module structure model. It maps photovoltaic structural information and photovoltaic electrical information to the three-dimensional photovoltaic module structure model to form a structural-electrical linkage model. Based on the structural-electrical division, the three-dimensional photovoltaic module structure model is divided into two corresponding sub-models. At the same time, the spatial coordinate system is added to the three models to obtain the linkage between the three-dimensional photovoltaic module structure model and the photovoltaic structural model and the photovoltaic electrical model.

[0014] Furthermore, it also includes: The dual-level defect classification module is used to inspect and monitor the structural and electrical performance of photovoltaic modules, and to establish photovoltaic structural defect classification and photovoltaic electrical diagnosis classification based on real-time operating data. The detection and classification results are mapped to the corresponding sub-models and linked to the three-dimensional photovoltaic module structural model. The linkage assessment and correction module is used to establish a fusion rule base. Based on the fusion rule base, photovoltaic structural defects are classified and photovoltaic electrical diagnoses are classified to establish an initial comprehensive health level. The rule base is then adjusted according to photovoltaic usage information and real-time information of photovoltaic modules to construct a photovoltaic linkage assessment model.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: In this invention, a historical photovoltaic module health matrix is ​​constructed based on photovoltaic structural information, photovoltaic electrical information, photovoltaic usage information, and installation environment information. A three-dimensional photovoltaic module structural model is constructed using the photovoltaic structural and photovoltaic electrical information, and the photovoltaic electrical model and photovoltaic structural model are divided. The photovoltaic modules are then inspected in stages using the historical photovoltaic module health matrix and the photovoltaic structural model, and a four-level photovoltaic structural defect classification is established. The photovoltaic modules are then inspected hierarchically using the photovoltaic usage information, installation environment information, and photovoltaic electrical model, and a four-level photovoltaic electrical diagnosis classification is established. The photovoltaic structural defect classification and the photovoltaic electrical diagnosis classification are fused into an initial comprehensive health level by establishing a fusion rule base through spatial consistency and causal correlation. The weights of the fusion rule base are adjusted according to the photovoltaic usage information and real-time photovoltaic information to establish a photovoltaic linkage evaluation model, thereby achieving accurate assessment of the health of photovoltaic modules. Attached Figure Description

[0016] Figure 1 This is a flowchart of a health assessment method for new energy photovoltaic modules based on intelligent monitoring, as described in this invention. Figure 2 This is a flowchart illustrating the process of obtaining the historical photovoltaic module health matrix in this invention; Figure 3 This is a flowchart illustrating the construction and partitioning of the three-dimensional photovoltaic module structural model in this invention; Figure 4 This is a flowchart illustrating the process of constructing a photovoltaic structure defect classification in this invention; Figure 5 This is a flowchart for constructing a photovoltaic electrical diagnostic grading system in this invention; Figure 6 This is a flowchart of the photovoltaic linkage evaluation model established in this invention; Figure 7 This is a system framework diagram of a new energy photovoltaic module health assessment system based on intelligent monitoring, as described in this invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Reference Figure 1 As shown, a health assessment method and system for new energy photovoltaic modules based on intelligent monitoring includes: Acquire photovoltaic structure and electrical information of photovoltaic modules, as well as photovoltaic usage and installation environment information; Based on photovoltaic usage information and installation environment information, a historical photovoltaic module health matrix is ​​constructed; A three-dimensional photovoltaic module structural model is constructed using the photovoltaic structural information and photovoltaic electrical information of the photovoltaic module. The three-dimensional photovoltaic module structural model is then divided to obtain the photovoltaic structural model and the photovoltaic electrical model. By combining historical photovoltaic module health matrix and photovoltaic structure model, photovoltaic modules are inspected in stages, and a photovoltaic structure defect classification is established; By combining photovoltaic usage information, installation environment information, and photovoltaic electrical models, a hierarchical inspection of photovoltaic modules is conducted to establish a photovoltaic electrical diagnostic classification. By integrating photovoltaic structural defect classification with photovoltaic electrical diagnosis classification, a photovoltaic linkage assessment model is established. The photovoltaic linkage assessment model is then corrected based on photovoltaic usage information and real-time photovoltaic information, and the health of photovoltaic modules is then assessed in real time through the photovoltaic linkage assessment model.

[0019] Reference Figure 2 As shown, the acquisition of photovoltaic structure information and photovoltaic electrical information of photovoltaic modules, as well as photovoltaic usage information and installation environment information, specifically includes: The photovoltaic usage information includes the historical usage time and historical failure event information of the photovoltaic structure; the installation environment information includes environmental temperature and humidity information and environmental suspended particulate matter concentration. The photovoltaic structural information includes the structural performance report of the photovoltaic module and the real-time structural information of the photovoltaic module; the photovoltaic electrical information includes the original performance report and the real-time electrical information of the photovoltaic module. The structural performance report includes manufacturing data and initial inspection reports of the photovoltaic modules at the time of manufacture; the real-time structural information includes periodic inspection data on surface stains and damage of the photovoltaic modules; the original performance report includes photovoltaic module performance data sheets and grid-connected test and commissioning records; the real-time electrical information includes SCADA system data and real-time IV curve test data; the historical usage time includes the grid-connected operation time and total operation time of the photovoltaic modules; the historical fault event information includes maintenance records, fault records, and periodic cleaning records of the photovoltaic modules; the environmental temperature and humidity information includes annual average temperature, annual average temperature difference, and annual extreme temperature; and the environmental suspended particulate matter concentration includes annual average PM10 concentration and quarterly average PM10 concentration.

[0020] The structural performance report of the photovoltaic (PV) modules in this plan includes manufacturing data and initial inspection reports at the time of PV module delivery, providing data support for subsequent use. Real-time structural information includes regular inspections of PV module surface stains and damage. PV electrical information includes PV module performance data tables from the original performance report and grid-connected test and commissioning records. Real-time electrical information includes SCADA system data and real-time N-curve tests. The historical usage time of the PV modules specifically includes the grid-connected operation start time, cumulative grid-connected operation time, and off-grid standby time. The cumulative grid-connected operation time is accurate to the hour, and the off-grid standby time needs to be distinguished for maintenance. The time periods corresponding to shutdowns, fault shutdowns, and grid-limited power generation shutdowns, historical fault event information covers maintenance records, fault records, and regular cleaning records throughout the entire life cycle of photovoltaic modules. The environmental temperature and humidity information specifically includes the annual average temperature, annual average diurnal temperature range, annual extreme maximum temperature, annual extreme minimum temperature, monthly average relative humidity, and daily average relative humidity during the rainy season. All temperature and humidity data must correspond to the location data of the photovoltaic module installation area. The environmental suspended particulate matter concentration specifically includes the annual average PM10 concentration, quarterly average PM10 concentration, and peak instantaneous PM10 concentration during key pollution periods. The concentration data must be synchronized with the monitoring cycle of dust pollution levels of photovoltaic modules.

[0021] Furthermore, the construction of a historical photovoltaic module health matrix based on photovoltaic usage information and installation environment information specifically includes: The installation environment information and photovoltaic usage information are cleaned and time-series aligned according to time sequence, and feature factors are extracted from them; The photovoltaic modules are divided according to their structure based on the photovoltaic structure information, and a weight is defined for each photovoltaic structure region as a spatial weight. The characteristic factors are quantified and standardized, and a two-dimensional matrix is ​​constructed based on time-characteristic factors. Spatial weight mapping is introduced, and a historical photovoltaic module health matrix is ​​constructed by combining spatial weights with the two-dimensional matrix. In this scheme, all data are unified with a timestamp according to time sequence, sorted according to the same time interval, and characteristic factors contained in various photovoltaic module data are calculated, such as damp heat stress index, particulate matter accumulation, and maintenance events. Then, the photovoltaic modules are disassembled according to components and structure, and the pure structural part and electrical part of the photovoltaic module are separated into multiple photovoltaic structural regions. The spatial weight of each photovoltaic structural region is defined by experts based on materials science, failure physics, environmental engineering, and statistical data. For example, based on materials science and failure physics, the temperature cycle weight of different photovoltaic structural regions of the photovoltaic module is defined, where the base temperature cycle weight is set to 1, and the weight of different regions is set based on the base temperature cycle weight. The weight of the edge photovoltaic structural region is set to 1.3 times the base temperature cycle weight. Different materials have the greatest difference in thermal expansion coefficient in the edge photovoltaic structural region, and shear stress is concentrated. The central photovoltaic structure area is set to have a base temperature cycle weight of 1.5 times: the heat dissipation of the central photovoltaic structure area is relatively poor, and the fatigue damage caused by the diurnal temperature difference is more significant. The cells and main grid lines are set to have a base temperature cycle weight of 1.8 times: the solder ribbons and grid lines directly bear repeated tension and shear caused by temperature differences, and are the photovoltaic structure areas most sensitive to fatigue fracture. The characteristic factors are cleaned in chronological order. For characteristic factors with clear physical mechanisms, physical failure models are used to quantify the characteristic factors. Then, statistical tools are used to perform regression analysis on the historical data of different regions to obtain empirical formulas. For characteristic factors with strong statistical correlation, empirical formulas are used to quantify them, and the time span is unified. Then, the units are standardized and unified. Finally, a two-dimensional matrix is ​​constructed according to time and characteristic factors. Each value in the two-dimensional matrix is ​​multiplied by the corresponding spatial weight to obtain the corresponding time-characteristic sequence, thus completing the construction of the historical photovoltaic module health matrix.

[0022] Reference Figure 3 As shown, further, the step of constructing a three-dimensional photovoltaic module structural model using the photovoltaic structural information and photovoltaic electrical information of the photovoltaic module, and dividing the three-dimensional photovoltaic module structural model to obtain the photovoltaic structural model and photovoltaic electrical model, specifically includes: Based on the photovoltaic structure information, a geometric model of the photovoltaic module is constructed using modeling software, which is called a three-dimensional photovoltaic module structure model. The photovoltaic structure information and photovoltaic electrical information are mapped to the three-dimensional photovoltaic module structure model, parameters are assigned to the corresponding geometry, and the power flow direction is defined. Based on photovoltaic structural information and photovoltaic electrical information as the core of the division, the three-dimensional photovoltaic module structural model is segmented according to the material mechanical characteristics and circuit topology data of the photovoltaic module. All geometric bodies are extracted and mapped with material mechanical characteristics, and all circuit topologies are extracted and assigned equivalent electrical parameters to obtain sub-models, which are respectively labeled as photovoltaic structural model and photovoltaic electrical model. A spatial coordinate system is introduced to link the sub-model with the three-dimensional photovoltaic module structural model; In this solution, based on the structural information of the photovoltaic (PV) module, existing 3D modeling software is used to transform the geometric configuration and material information of the PV module into a 3D PV module structural model. Simultaneously, relevant information about each structure or component of the PV module is mapped onto the 3D PV module structural model, and the electrical information of the PV module is also mapped onto the 3D PV module structural model. Corresponding parameters are assigned to the corresponding geometric bodies, and the power flow direction is defined. Subsequently, the 3D PV module structural model is divided according to structural and electrical aspects, segmented based on the material mechanical characteristics and circuit topology data of the PV module, and all geometric entities in the 3D PV module structural model are extracted. The corresponding material mechanical characteristics are then mapped onto the corresponding geometric entities. In the 3D photovoltaic module structural model, all circuit topologies are assigned equivalent electrical parameters to each component. The geometric solid model containing material mechanical characteristics is treated as an independent sub-model, and the circuit topology containing equivalent electrical parameters is treated as an independent sub-model, resulting in three 3D models. These include an undivided main model and two divided sub-models. A spatial coordinate system is introduced to link the three models together, allowing modifications to the three models to be made synchronously. This results in the main model being the 3D photovoltaic module structural model, and the two sub-models being the photovoltaic structure model and the photovoltaic electrical model. The photovoltaic structure model retains only the material mechanical information of the photovoltaic module, while the photovoltaic electrical model simplifies the material structure information of the photovoltaic module and retains the circuit topology data of the photovoltaic module.

[0023] Reference Figure 4 As shown, furthermore, the process of combining historical photovoltaic module health matrices and photovoltaic structure models to conduct phased inspections of photovoltaic modules and establish a photovoltaic structural defect classification includes: Based on the degradation trend of photovoltaic modules in the historical photovoltaic module health matrix, different photovoltaic modules in the photovoltaic structure model are classified and described, and the structural performance report of photovoltaic modules in the photovoltaic structure information is used to check for possible defects in photovoltaic modules; A four-level structural defect impact standard is established, namely L1, L2, L3 and L4. Among them, L1: affects appearance but does not affect normal use; L2: has a significant impact on appearance and may cause power loss or reliability risk; L3: has a significant power loss or safety hazard; L4: has functional failure or causes safety accident. The defects identified during the inspection were integrated and classified according to different photovoltaic modules. At the same time, the defects were graded according to their impact on the function of the photovoltaic modules. A photovoltaic structural defect classification was established based on the photovoltaic module location and the four-level structural defect impact standard. The photovoltaic structural defect classification was then mapped to a three-dimensional photovoltaic module structural model. This solution involves statistically analyzing the current performance of each structure within a photovoltaic (PV) module according to its degradation trend. This statistically analyzed performance is then linked to its corresponding structure. Different current performance conditions are listed and reported. Based on these reports, potential defects are inspected and statistically analyzed. These defects are categorized according to the PV module's structure. A four-level structural defect impact standard is established, classifying defects based on their degree of impact on the PV module's structure. Thresholds for judging the four levels of structural defect impact are set based on PV module safety requirements. The L1 threshold is set to reflect the impact of current and future performance on the PV module's structure. Safety is not affected. The L2 threshold is set as the existence of a potential damage mode or the power loss is less than 5%. The L3 threshold is set as the existence of a clear safety hazard or the power loss is greater than 5% but less than 20%. The L4 threshold is set as the complete or most of the module's function is lost or the power loss is more than 20%, with the risk of fire, electric shock, or structural collapse. Then, the classified defects are associated with the corresponding photovoltaic module structure, and the defects are mapped to the photovoltaic structure model and associated with the three-dimensional photovoltaic module structure model. The photovoltaic structure defect classification is based on the photovoltaic module structure as the primary classification, and the four-level structural defect impact standard is a secondary classification.

[0024] Reference Figure 5 As shown, furthermore, the process of combining photovoltaic usage information, installation environment information, and photovoltaic electrical models to perform hierarchical checks on photovoltaic modules and establish a photovoltaic electrical diagnostic grading system specifically includes: Based on the original performance report and installation environment information of the photovoltaic modules in the photovoltaic electrical information, the electrical defective modules in the photovoltaic modules are marked, and the real-time operating status of the electrical defective modules is monitored based on the photovoltaic usage information; Data generated during real-time operation is collected, cleaned and standardized, and the IV characteristic curve is reconstructed based on the processed data. The single diode model is then fitted to obtain the electrical parameters of the electrically defective component. Based on the comparison between the health benchmark value and electrical parameters in the photovoltaic electrical information, a four-level electrical defect classification is established, namely T1, T2, T3 and T4. Among them, T1: does not affect normal use; T2: may have early degradation or slight contamination; T3: power or parameters have changed significantly or there is a clear electrical fault; T4: there is a serious safety hazard or the module has failed. A photovoltaic electrical diagnostic classification is established based on the four-level electrical defect classification and the electrical parameters of the electrical defect components, and then mapped to a three-dimensional photovoltaic module structural model. In this solution, the original performance report includes the initial STC power, initial IV curve, and manufacturing date of the photovoltaic module at the time of manufacture. Combined with installation environment information, potential electrical defects that the photovoltaic module may encounter are statistically analyzed, and these defects are confirmed based on photovoltaic usage information. Confirmed electrical defects are then statistically analyzed and ranked according to their impact on the photovoltaic module. Simultaneously, real-time operating data from photovoltaic modules containing electrical defects is collected, processed, and timestamped. The IV characteristic curve is reconstructed from the processed data, fitted, and its electrical parameters are obtained. These parameters are then compared with the electrical parameters of the initial IV curve. Different data points are extracted and analyzed, correlated with electrical defects, and a four-level electrical defect classification is established. Based on a combination of photovoltaic module safety assessment and expert evaluation, four levels of electrical defects are categorized. The system sets thresholds for electrical defects. For T1, the threshold is set as power attenuation less than or equal to 3% or an undistorted IV curve. For T2, the threshold is set as power attenuation greater than 3% but less than or equal to 5% or a slightly distorted but smooth IV curve. For T3, the threshold is set as power attenuation greater than 5% but less than or equal to 20% or the appearance of localized hot spots (i.e., temperature changes exceeding 20 degrees Celsius). For T4, the threshold is set as power attenuation greater than 20% or insulation failure or no output from the module. The electrical parameters of the defective module are mapped to the four levels of electrical defects to obtain the corresponding classifications for different electrical defects. This allows for the construction of a photovoltaic electrical diagnostic classification system. All obtained electrical defects and their corresponding classifications are mapped onto the 3D photovoltaic module structure model and the photovoltaic electrical model. Simultaneously, the 3D photovoltaic module structure model and the photovoltaic electrical model are linked to ensure that the electrical information of the two models is updated synchronously.

[0025] Reference Figure 6 As shown, furthermore, the integration of photovoltaic structural defect classification and photovoltaic electrical diagnostic classification to establish a photovoltaic linkage assessment model, the correction of the photovoltaic linkage assessment model based on photovoltaic usage information and real-time photovoltaic information, and the real-time assessment of the health of photovoltaic modules through the photovoltaic linkage assessment model, specifically includes: Based on spatial consistency and causal relationship, a fusion rule base is established. By using spatial positioning technology, the specific location information of photovoltaic structural defects is matched with the electrical parameter collection points in the corresponding photovoltaic structural area. The causal relationship mechanism between structural defects and electrical performance abnormalities is analyzed, and fusion judgment rules are constructed. The photovoltaic structural defect classification and photovoltaic electrical diagnosis classification are input into the rule base, and the initial comprehensive health level is obtained by cross-comparing the preset weight coefficients and correlation strength. The weights of the fusion rule base are adjusted based on photovoltaic usage information and real-time structural and electrical information of photovoltaic modules. The adjustment scope includes the association rules, influencing factors and initial weights between different data. Based on the comparative analysis of real-time data and historical data, the weight values ​​of each parameter are dynamically adjusted according to the percentage change in the conversion power of photovoltaic modules. The effect of weight adjustment is verified by combining the actual power generation fluctuations in photovoltaic usage information, and the photovoltaic linkage evaluation model is obtained by iteratively optimizing the fusion rule base. The health of photovoltaic modules is monitored and evaluated in real time based on the obtained photovoltaic linkage assessment model; In this scheme, the rule base first uses spatial positioning technology to match the specific location information of photovoltaic structural defects with the electrical parameter collection points within the corresponding photovoltaic structure area, ensuring a clear spatial correspondence between structural defects and electrical anomalies. Next, it analyzes the causal relationship between different types of structural defects and electrical performance anomalies. For example, microcracks in modules may cause local current shading, leading to a decrease in string power; tilted supports may change the optimal tilt angle of modules, resulting in reduced power generation. Based on this correlation, fusion judgment rules are constructed. Then, the photovoltaic structural defect classification and photovoltaic electrical diagnosis classification are input into the rule base. Through cross-comparison, the results of structural defects and electrical diagnoses are comprehensively evaluated. When the structural defect level is L2 and the corresponding photovoltaic structure area's electrical diagnosis level is T2, the initial comprehensive health level of the photovoltaic structure area is determined to be (L2, T2+) based on the preset weight coefficients and correlation strength in the rule base. This ultimately yields an initial comprehensive health level that reflects the health status of the photovoltaic modules. Subsequently, photovoltaic usage information, real-time structural information of the photovoltaic modules, and real-time electrical information are integrated. Based on this data, the weights in the fusion rule base are adjusted, including correlation rules, influencing factors, and initial weights between different data sets. Then, a gradient descent adaptive algorithm is introduced, and based on a comparative analysis of real-time structural and electrical information with historical data, the photovoltaic module's conversion power variation is considered. The system dynamically adjusts the weights of parameters in the rule base based on percentages. The sign of the percentage determines the adjustment level, and the magnitude of the percentage determines the adjustment ratio. For example, if shading in a photovoltaic structure area suddenly increases, the weight of the shading coefficient in the real-time structural information will be increased. Conversely, if abnormal temperature increases in the modules lead to a decrease in conversion efficiency, the weight of the temperature coefficient in the real-time electrical information will be adjusted based on the percentage fluctuation in conversion efficiency. Simultaneously, the system cleanses the actual power generation fluctuations in the photovoltaic usage information by incorporating environmental temperature and humidity information and the concentration of suspended particulate matter, eliminating extreme operating conditions of the photovoltaic modules caused by extreme environmental conditions. The effectiveness of the weight adjustment is then verified through feedback. The new weight is obtained by multiplying the difference between the power generation after weight adjustment and the expected power generation after weight adjustment by the percentage of the expected power generation after weight adjustment and the adjusted weight. At the same time, according to the photovoltaic power plant operation and maintenance technical specifications, the feedback verification upper limit of the T3 level four electrical defect classification is set. When the power change is greater than 20%, the weight adjustment is stopped, an alarm is issued and manual intervention is required. Through continuous iteration, the fusion rule base can reflect the degree of impact of different factors on the health of photovoltaic modules. Finally, based on the optimized weight configuration, a photovoltaic linkage assessment model that can analyze the structural status, electrical performance and usage requirements in real time is constructed, and the photovoltaic linkage assessment model can be used to achieve accurate assessment of the health of photovoltaic modules.

[0026] Reference Figure 7As shown, the present invention also provides a health assessment system for new energy photovoltaic modules based on intelligent monitoring, used to implement the above-mentioned assessment method. The system includes: The information acquisition and processing module is used to read the structural performance report of the photovoltaic module at the time of its manufacture and all information during the use of the photovoltaic module. It cleans and sorts all the information in chronological order to facilitate subsequent reading. At the same time, it establishes an interrelated historical photovoltaic module health matrix based on the photovoltaic structural information and photovoltaic electrical information.

[0027] Furthermore, it also includes: The model building and splitting module is used to build a three-dimensional photovoltaic module structure model. It maps photovoltaic structural information and photovoltaic electrical information to the three-dimensional photovoltaic module structure model to form a structural-electrical linkage model. Based on the structural-electrical division, the three-dimensional photovoltaic module structure model is divided into two corresponding sub-models. At the same time, the spatial coordinate system is added to the three models to obtain the linkage between the three-dimensional photovoltaic module structure model and the photovoltaic structural model and the photovoltaic electrical model.

[0028] Furthermore, it also includes: The dual-level defect classification module is used to inspect and monitor the structural and electrical performance of photovoltaic modules, and to establish photovoltaic structural defect classification and photovoltaic electrical diagnosis classification based on real-time operating data. The detection and classification results are mapped to the corresponding sub-models and linked to the three-dimensional photovoltaic module structural model. The linkage assessment and correction module is used to establish a fusion rule base. Based on the fusion rule base, photovoltaic structural defects are classified and photovoltaic electrical diagnoses are classified to establish an initial comprehensive health level. The rule base is then adjusted according to photovoltaic usage information and real-time information of photovoltaic modules to construct a photovoltaic linkage assessment model.

[0029] The advantages of this invention lie in its construction of a historical photovoltaic module health matrix. By combining a two-dimensional matrix of time and feature factors with spatial weights, it reflects the changes in the health status of different photovoltaic structural regions of the photovoltaic module over time. It utilizes photovoltaic structural and electrical information to construct and decompose a three-dimensional model into a photovoltaic structural model and a photovoltaic electrical model, introducing a spatial coordinate system to achieve model linkage, enabling precise analysis of the structural and electrical components separately. It performs phased structural checks on the photovoltaic module and establishes a four-level structural defect impact standard, clearly defining the degree of impact of structural defects on the photovoltaic module, and mapping the defect classification to the three-dimensional model for intuitive viewing. It combines photovoltaic usage and installation environment information with the photovoltaic electrical model to perform hierarchical electrical checks, reconstructing the IV characteristic curve and fitting electrical parameters to establish a four-level electrical defect classification, accurately assessing electrical performance and defect impact, which is also mapped to the three-dimensional model for visualization. Finally, it establishes a fusion rule base to integrate the photovoltaic structural defect classification with the photovoltaic electrical diagnostic classification to obtain an initial comprehensive health level. Based on photovoltaic usage information and real-time information, it corrects the weights of the fusion rule base to construct a photovoltaic linkage assessment model, achieving a precise comprehensive assessment of the photovoltaic module's health.

[0030] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A health assessment method for new energy photovoltaic modules based on intelligent monitoring, characterized in that, include: Acquire photovoltaic structure and electrical information of photovoltaic modules, as well as photovoltaic usage and installation environment information; Based on photovoltaic usage information and installation environment information, a historical photovoltaic module health matrix is ​​constructed; A three-dimensional photovoltaic module structural model is constructed using the photovoltaic structural information and photovoltaic electrical information of the photovoltaic module. The three-dimensional photovoltaic module structural model is then divided to obtain the photovoltaic structural model and the photovoltaic electrical model. By combining historical photovoltaic module health matrix and photovoltaic structure model, photovoltaic modules are inspected in stages, and a photovoltaic structure defect classification is established; By combining photovoltaic usage information, installation environment information, and photovoltaic electrical models, a hierarchical inspection of photovoltaic modules is conducted to establish a photovoltaic electrical diagnostic classification. By integrating photovoltaic structural defect classification with photovoltaic electrical diagnosis classification, a photovoltaic linkage assessment model is established. The photovoltaic linkage assessment model is then corrected based on photovoltaic usage information and real-time photovoltaic information, and the health of photovoltaic modules is then assessed in real time through the photovoltaic linkage assessment model.

2. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 1, characterized in that, The acquisition of photovoltaic structural information and photovoltaic electrical information of photovoltaic modules, as well as photovoltaic usage information and installation environment information, specifically includes: The photovoltaic usage information includes the historical usage time and historical failure event information of the photovoltaic structure; the installation environment information includes environmental temperature and humidity information and environmental suspended particulate matter concentration. The photovoltaic structural information includes the structural performance report of the photovoltaic module and the real-time structural information of the photovoltaic module; the photovoltaic electrical information includes the original performance report and the real-time electrical information of the photovoltaic module. The structural performance report includes manufacturing data and initial inspection reports of the photovoltaic modules at the time of manufacture; the real-time structural information includes periodic inspection data on surface stains and damage of the photovoltaic modules; the original performance report includes photovoltaic module performance data sheets and grid-connected test and commissioning records; the real-time electrical information includes SCADA system data and real-time IV curve test data; the historical usage time includes the grid-connected operation time and total operation time of the photovoltaic modules; the historical fault event information includes maintenance records, fault records, and periodic cleaning records of the photovoltaic modules; the environmental temperature and humidity information includes annual average temperature, annual average temperature difference, and annual extreme temperature; and the environmental suspended particulate matter concentration includes annual average PM10 concentration and quarterly average PM10 concentration.

3. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 2, characterized in that, The construction of a historical photovoltaic module health matrix based on photovoltaic usage information and installation environment information specifically includes: The installation environment information and photovoltaic usage information are cleaned and time-series aligned according to time sequence, and feature factors are extracted from them; The photovoltaic modules are divided according to their structure based on the photovoltaic structure information, and a weight is defined for each photovoltaic structure region as a spatial weight. The characteristic factors are quantified and standardized. A two-dimensional matrix is ​​constructed based on the time-characteristic factors. Spatial weight mapping is introduced, and a historical photovoltaic module health matrix is ​​constructed by combining the spatial weights with the two-dimensional matrix.

4. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 3, characterized in that, The process involves constructing a three-dimensional photovoltaic module structural model using the photovoltaic structural and electrical information of the photovoltaic module, and then dividing the three-dimensional photovoltaic module structural model to obtain the photovoltaic structural model and the photovoltaic electrical model. Specifically, this includes: Based on the photovoltaic structure information, a geometric model of the photovoltaic module is constructed using modeling software, which is called a three-dimensional photovoltaic module structure model. The photovoltaic structure information and photovoltaic electrical information are mapped to the three-dimensional photovoltaic module structure model, parameters are assigned to the corresponding geometry, and the power flow direction is defined. Based on photovoltaic structural information and photovoltaic electrical information as the core of the division, the three-dimensional photovoltaic module structural model is segmented according to the material mechanical characteristics and circuit topology data of the photovoltaic module. All geometric bodies are extracted and mapped with material mechanical characteristics, and all circuit topologies are extracted and assigned equivalent electrical parameters to obtain sub-models, which are respectively labeled as photovoltaic structural model and photovoltaic electrical model. A spatial coordinate system is introduced to link the sub-model with the three-dimensional photovoltaic module structure model.

5. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 4, characterized in that, The process of combining historical photovoltaic module health matrices and photovoltaic structure models to conduct phased inspections of photovoltaic modules and establish a photovoltaic structural defect classification includes: Based on the degradation trend of photovoltaic modules in the historical photovoltaic module health matrix, different photovoltaic modules in the photovoltaic structure model are classified and described, and the structural performance report of photovoltaic modules in the photovoltaic structure information is used to check for possible defects in photovoltaic modules; A four-level structural defect impact standard is established, namely L1, L2, L3 and L4. Among them, L1: affects appearance but does not affect normal use; L2: has a significant impact on appearance and may cause power loss or reliability risk; L3: has a significant power loss or safety hazard; L4: has functional failure or causes safety accident. The defects identified during the inspection were integrated and classified according to different photovoltaic modules. At the same time, the defects were graded according to their impact on the function of the photovoltaic modules. A photovoltaic structural defect classification was established based on the photovoltaic module location and the four-level structural defect impact standard. The photovoltaic structural defect classification was then mapped onto a three-dimensional photovoltaic module structural model.

6. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 5, characterized in that, The method combines photovoltaic usage information, installation environment information, and photovoltaic electrical models to conduct hierarchical inspections of photovoltaic modules and establish a photovoltaic electrical diagnostic grading system, specifically including: Based on the original performance report and installation environment information of the photovoltaic modules in the photovoltaic electrical information, the electrical defective modules in the photovoltaic modules are marked, and the real-time operating status of the electrical defective modules is monitored based on the photovoltaic usage information; Data generated during real-time operation is collected, cleaned and standardized, and the IV characteristic curve is reconstructed based on the processed data. The single diode model is then fitted to obtain the electrical parameters of the electrically defective component. Based on the comparison between the health benchmark value and electrical parameters in the photovoltaic electrical information, a four-level electrical defect classification is established, namely T1, T2, T3 and T4. Among them, T1: does not affect normal use; T2: may have early degradation or slight contamination; T3: power or parameters have changed significantly or there is a clear electrical fault; T4: there is a serious safety hazard or the module has failed. A photovoltaic electrical diagnostic classification is established based on the four-level electrical defect classification and the electrical parameters of the electrical defect components, and then mapped to a three-dimensional photovoltaic module structural model.

7. The health assessment method for new energy photovoltaic modules based on intelligent monitoring according to claim 6, characterized in that, The process involves integrating photovoltaic structural defect classification with photovoltaic electrical diagnostic classification to establish a photovoltaic linkage assessment model. This model is then revised based on photovoltaic usage information and real-time photovoltaic data. Finally, the photovoltaic linkage assessment model is used to conduct real-time assessments of the health of photovoltaic modules. Specifically, this includes: Based on spatial consistency and causal relationship, a fusion rule base is established. By using spatial positioning technology, the specific location information of photovoltaic structural defects is matched with the electrical parameter collection points in the corresponding photovoltaic structural area. The causal relationship mechanism between structural defects and electrical performance abnormalities is analyzed, and fusion judgment rules are constructed. The photovoltaic structural defect classification and photovoltaic electrical diagnosis classification are input into the rule base, and the initial comprehensive health level is obtained by cross-comparing the preset weight coefficients and correlation strength. The weights of the fusion rule base are adjusted based on photovoltaic usage information and real-time structural and electrical information of photovoltaic modules. The adjustment scope includes the association rules, influencing factors and initial weights between different data. Based on the comparative analysis of real-time data and historical data, the weight values ​​of each parameter are dynamically adjusted according to the percentage change in the conversion power of photovoltaic modules. The effect of weight adjustment is verified by combining the actual power generation fluctuations in photovoltaic usage information, and the photovoltaic linkage evaluation model is obtained by iteratively optimizing the fusion rule base. The health of photovoltaic modules is monitored and evaluated in real time based on the obtained photovoltaic linkage assessment model.

8. A health assessment system for new energy photovoltaic modules based on intelligent monitoring, used to implement the assessment method described in claims 1-7, characterized in that, include: The information acquisition and processing module is used to read the structural performance report of the photovoltaic module at the time of its manufacture and all information during the use of the photovoltaic module. It cleans and sorts all the information in chronological order to facilitate subsequent reading. At the same time, it establishes an interrelated historical photovoltaic module health matrix based on the photovoltaic structural information and photovoltaic electrical information.

9. A health assessment system for new energy photovoltaic modules based on intelligent monitoring according to claim 8, characterized in that, Also includes: The model building and splitting module is used to build a three-dimensional photovoltaic module structure model. It maps photovoltaic structural information and photovoltaic electrical information to the three-dimensional photovoltaic module structure model to form a structural-electrical linkage model. Based on the structural-electrical division, the three-dimensional photovoltaic module structure model is divided into two corresponding sub-models. At the same time, the spatial coordinate system is added to the three models to obtain the linkage between the three-dimensional photovoltaic module structure model and the photovoltaic structural model and the photovoltaic electrical model.

10. A health assessment system for new energy photovoltaic modules based on intelligent monitoring according to claim 9, characterized in that, Also includes: The dual-level defect classification module is used to inspect and monitor the structural and electrical performance of photovoltaic modules, and to establish photovoltaic structural defect classification and photovoltaic electrical diagnosis classification based on real-time operating data. The detection and classification results are mapped to the corresponding sub-models and linked to the three-dimensional photovoltaic module structural model. The linkage assessment and correction module is used to establish a fusion rule base. Based on the fusion rule base, photovoltaic structural defects are classified and photovoltaic electrical diagnoses are classified to establish an initial comprehensive health level. The rule base is then adjusted according to photovoltaic usage information and real-time information of photovoltaic modules to construct a photovoltaic linkage assessment model.

Images