Collaborative design integrated optimization method and system based on optical storage power station

Abstract

The invention discloses a collaborative design integrated optimization method and system based on an optical storage power station, and relates to the technical field of industrial collaborative design optimization. The collaborative design integrated optimization method and system based on the optical storage power station comprises the steps that S1, field structure data, environment and layout data and state performance data in the construction and operation process of the optical storage power station are collected and preprocessed, and a standardized component state data set is constructed; s2, performing performance evaluation on the real-time power generation state of the assembly, identifying whether the assembly enters a low-efficiency operation section or not, and touching corresponding operation and maintenance regulation and control operation; s3, performing operation analysis on the operation consistency, and dynamically adjusting the cleaning scheduling priority; and S4, comprehensively evaluating the task propulsion state of the component, and dynamically deducing the current optimal task adjustment intensity. The problem that the efficiency of the whole period is difficult to guarantee due to the disjunction phenomenon that the construction design tends to the static working condition and the operation, maintenance, adjustment and optimization lacks the design context in the construction of the current optical storage power station is solved.

Description

Technical Field

[0001] This invention relates to the field of industrial collaborative design optimization technology, specifically to an integrated optimization method and system for collaborative design of photovoltaic and energy storage power stations. Background Technology

[0002] With the continuous growth in demand for intensive construction and intelligent operation and maintenance of new energy power plants, the collaborative design capability among photovoltaic modules, energy storage devices, and power dispatching equipment is increasingly becoming a key factor affecting the performance of power plants throughout their entire life cycle. Currently, during the design phase, photovoltaic-storage power plants typically use offline simulation to quantitatively calculate module layout, support angles, cable routes, etc., and evaluate energy efficiency output and load distribution effects based on typical operating conditions. During the operation and maintenance phase, they rely heavily on data collected from photovoltaic monitoring terminals, environmental sensing devices, and energy storage management systems, using rule triggering and anomaly alarms to guide the scheduling of operation and maintenance tasks. Some high-level photovoltaic-storage power plants have begun to introduce BIM models, electrical topology modeling, and energy flow tracing analysis to attempt to achieve model mapping from design parameters to operational data, thereby improving the accuracy and response efficiency of collaborative management and control.

[0003] For example, invention patent CN104077728B discloses a method for selecting energy storage configurations for photovoltaic power plants that considers economic factors. This method includes: establishing an energy storage cost function model that does not consider electricity price costs; determining the relationship between the amount of curtailed solar power and the annual power generation of the energy storage system based on the aforementioned energy storage cost function model according to the power curtailment situation; inputting this relationship into a MATLAB tool for optimization calculation; and obtaining the optimal photovoltaic energy storage ratio based on the calculation results. By comprehensively considering both technical and economic factors, the feasibility of combined photovoltaic and energy storage operation is analyzed, and a practical configuration scheme is provided, thereby achieving the goal of improving the economic efficiency of use.

[0004] For example, invention patent CN118587038B discloses a method and device for real-time verification and reporting of power generation data from an integrated wind-solar-storage power station. The method includes: acquiring the maximum available power and actual power output fed back by the energy management platform; determining the reporting dispatch step size based on the data frequency of the actual power output fed back by the energy management platform; collecting the maximum available power output fed back by the energy management platform according to the reporting dispatch step size; comparing the collected maximum available power output with the actual power output at the same time; updating the current maximum available power output to the actual power output at the same time when the collected maximum available power output is less than the actual power output at the same time; and uploading the updated maximum available power output and actual power output to the power grid dispatch center according to the reporting dispatch step size. This invention overcomes the problem that the maximum available power output of the data transmitted to the power grid is less than the actual power output due to differences in communication links and update time scales, as well as the inertia of the generating units during normal operation.

[0005] However, current design and operation still suffer from data disconnect, model fragmentation, and lack of feedback: On the one hand, static structural parameters generated during the design phase are difficult to dynamically compare during operation, making it impossible to trace the origins of issues such as component shading, cable hotspots, and voltage fluctuations back to the initial layout; on the other hand, a large amount of operational data generated during operation fails to be used in reverse for structural optimization and task reconfiguration, leading to a mismatch between cleaning, maintenance, and reconfiguration plans and actual deviations, thus affecting energy efficiency and the rationality of task resource allocation. Furthermore, the lack of integrated modeling mechanisms and dynamic reconfiguration strategies during task optimization makes it difficult to consider diverse factors such as on-site disturbances, equipment degradation, and grid demands, resulting in delayed scheduling strategies and lengthy optimization cycles, failing to meet the requirements for efficient, closed-loop, and adaptive operation and management of large-scale photovoltaic-storage power plants.

[0006] Therefore, there is an urgent need for an integrated optimization method and system for the collaborative design of photovoltaic and energy storage power plants. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides an integrated optimization method and system for collaborative design of photovoltaic and energy storage power plants. This solves the problem that current photovoltaic and energy storage power plant construction often suffers from a disconnect between construction design and operation and maintenance optimization, which is biased towards static conditions and lacks design context, making it difficult to guarantee efficiency throughout the entire lifecycle.

[0009] Technical solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: an integrated optimization method and system for collaborative design of photovoltaic-storage power plants, comprising: S1, collecting on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic-storage power plant; preprocessing the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset; S2, based on the standardized component status dataset, performing performance evaluation on the real-time power generation status of the components, and identifying whether the components have entered an inefficient operating segment based on the performance evaluation results and triggering corresponding operation and maintenance control operations; S3, based on the standardized component status dataset, performing operational analysis on the operational consistency from the component energy output, and dynamically adjusting the cleaning scheduling priority based on the operational analysis results; S4, using the performance evaluation results and operational analysis results as input, comprehensively evaluating the component task progress status, and dynamically deriving the current optimal task adjustment intensity based on the comprehensive evaluation results.

[0011] Furthermore, the specific steps for collecting on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic-storage power station are as follows: Collect on-site structural data of the photovoltaic array area, including: the static installation tilt angle of each photovoltaic module, module surface image, hot spot recurrence frequency, cable path length, surface temperature of the thermally sensitive area on the back of the module, and the connection path length between each connection point; Collect environmental and layout data during construction, including: the real-time irradiance value received by each photovoltaic module, the change in cable contact resistance of the module connection loop, the ambient temperature at the location of the photovoltaic array, the return path length for operation and maintenance scheduling, and the instantaneous wind speed disturbance value, while simultaneously calculating and recording the standard deviation of the wind speed disturbance; Collect performance data of the operating status, including: the DC output power, output energy, real-time output voltage of each photovoltaic module, the absorbed energy of the energy storage channel, and the real-time connection voltage and instantaneous temperature rise value at each connection point, while simultaneously calculating and recording the average output energy, average absorbed energy, and average output voltage.

[0012] Furthermore, the specific steps for preprocessing the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset are as follows: For the on-site structural data, the static installation tilt angle of each photovoltaic module is uniformly converted to angle units, and outliers exceeding the physical range are corrected. Broken path records in the cable path length and connection point path length are interpolated and completed. The hot spot recurrence frequency is aligned to the specific component number according to the construction time axis, and median filtering is used to eliminate sudden noise in the surface temperature of the heat-sensitive area. For the environmental and layout data, the real-time received light intensity value is processed using a sliding mean filter, and the cable contact resistance change is combined with the group... The component numbering system is unified and mapped to a numbering system. Missing data in ambient temperature is filled using linear interpolation, and physical anomalies are eliminated. Operation and maintenance scheduling return paths are matched to geographic location numbers, and disconnected and erroneous path records are removed. In the standardization of operational status performance data, DC output power and output energy are timestamped, format errors and duplicate records in energy storage channel absorption are removed, and the numbering mapping between components and energy storage channels is verified. Extreme values ​​in output voltage and instantaneous temperature rise are removed and replaced with the mean value within a window. The standardized field structure data, environmental and layout data, and status performance data are normalized to construct a standardized component status dataset.

[0013] Furthermore, the specific steps for evaluating the real-time power generation status of the photovoltaic module based on the standardized module state dataset are as follows: Obtain the real-time connection voltage and connection path length at the normalized connection point positions; calculate the voltage difference between adjacent time periods using a sliding window method; divide the voltage difference between adjacent time periods by the corresponding connection path length to obtain the cable voltage drop rate per unit length; multiply the cosine of the module's static installation tilt angle by the difference between one and the light intensity value, and then add this difference to the cable contact resistance change to obtain the tilt angle shading correction value; divide the surface temperature of the thermistor area by the ambient temperature and add one to obtain the temperature rise correction ratio; take the cable path length as the power of the cable voltage drop rate per unit length to obtain the path length loss ratio; use the DC output power of the photovoltaic module as the numerator, and the product of the tilt angle shading correction value, the temperature rise correction ratio, and the path length loss ratio as the denominator to obtain the structural performance expression value.

[0014] Furthermore, the specific steps for identifying whether a component has entered an inefficient operating zone and triggering corresponding operation and maintenance control operations based on performance evaluation results are as follows: Real-time comparison of the current photovoltaic module's structural performance expression value with the expression threshold: When the structural performance expression value is less than or equal to the expression threshold, the module is determined to have entered an inefficient operating zone. The DC combiner switch of the branch where the module is located is immediately shut off, the module surface cleaning is initiated and the pollution level is simultaneously marked, the priority of the corresponding energy storage discharge channel of the module is adjusted to the lowest, and a high-frequency relocation image acquisition task for the shading source is pushed to the operation and maintenance terminal; When the structural performance expression value is greater than the expression threshold, the current circuit on / off state and power allocation strategy remain unchanged, the current module is included in the performance stable section monitoring list, and the heat load change of the cable path corresponding to the current module is updated and recorded.

[0015] Furthermore, the specific steps for analyzing operational consistency based on component energy output using the standardized component state dataset are as follows: The difference between the component's output power and the average output power is divided by the sum of the average output power and 1 to obtain the power generation fluctuation offset rate; the absorbed power of the energy storage channel is divided by the average absorbed power to obtain the energy storage response matching rate; the real-time output voltage of the component is divided by 1 plus the sum of the average output voltage, then 1 is added, the natural logarithm is taken, and 1 is added again to obtain the voltage deviation correction value; the instantaneous wind speed disturbance value at the component's location is squared, divided by the sum of the square of the wind speed disturbance standard deviation and 1, and then 1 is added to obtain the disturbance intensity normalization value; the reciprocal of the voltage deviation correction value is multiplied by the energy storage response matching rate to obtain the voltage-energy storage comprehensive value; the absolute value of the difference between the power generation fluctuation offset rate and the voltage-energy storage comprehensive value is multiplied by the disturbance intensity normalization value to obtain the component operational deviation value.

[0016] Furthermore, the specific steps for dynamically adjusting the cleaning scheduling priority based on the operational analysis results are as follows:

[0017] The component operating deviation values ​​of all components are sorted from low to high. Components with operating deviation values ​​in the last q quantile are identified as components with disordered power generation behavior and their task scheduling list is cleaned. The relative increase of the operating deviation value of the same component in two consecutive calculations is calculated. The Holt-Winters trend model is input to perform multi-step analysis of the deviation trend. Components with a significant positive increase trend are marked as components with increasing deviation trend. The structure of these components is immediately reviewed by UAVs to check for surface obstruction, cable connection points, and backplane contamination. The two-dimensional geographic coordinates of all components are used as input. A density-based spatial clustering algorithm is used to spatially cluster the components. The mean operating deviation value of the components in each group is calculated. The group with the highest mean operating deviation value is obtained, and a redistribution suggestion for the energy storage channel is generated. At the same time, a component operating deviation heatmap is constructed based on the component two-dimensional coordinates and the distribution of component operating deviation values. The area with the highest mean operating deviation value is marked as a dynamic hotspot segment, and the sampling frequency of the area is increased.

[0018] Furthermore, the specific steps for comprehensively evaluating the component task progress status using performance evaluation results and operational analysis results as input are as follows: Obtain a component surface image; using image recognition and dust occlusion density recognition algorithms, segment and evaluate the component surface image based on texture, color, and edge blurring to obtain the component's area occlusion coverage rate; multiply the structural performance expression value, component operation deviation value, and maintenance scheduling return path length to obtain the structural maintenance evaluation value; add the area occlusion coverage rate to the instantaneous temperature rise value of the cable connection node to obtain the wiring displacement correction value; add the instantaneous wind speed disturbance value and the hot spot recurrence frequency, add one, take the natural logarithm, and add another one to obtain the comprehensive environmental disturbance value; divide the structural maintenance evaluation value by the product of the wiring displacement correction value and the comprehensive environmental disturbance value, add one, and take the natural logarithm to obtain the task optimization feedback value.

[0019] Furthermore, the specific steps for dynamically deriving the current optimal task adjustment intensity based on the comprehensive evaluation results are as follows: Real-time comparison of the current task optimization feedback value with the optimization feedback threshold, which includes a first optimization threshold and a second optimization threshold: When the task optimization feedback value is less than or equal to the second optimization threshold, stop the current area's operation and maintenance dispatch, and simultaneously extract the bracket tilt angle, cable laying length, and energy storage interface configuration within the section, archiving them as standard samples; When the task optimization feedback value is greater than the second optimization threshold but less than or equal to the first optimization threshold, rearrange the operation and maintenance inspection rhythm within the area, increase the cleaning frequency and sampling frequency, and simultaneously generate a comparison list of component construction constraints and operational deviations, marking specific structural information in the comparison list and pushing out structural feedback suggestions; When the task optimization feedback value is greater than the first optimization threshold, immediately initiate on-site intervention, adjust the component cleaning task priority within the area to the highest level, generate an on-site task order on the operation and maintenance terminal, synchronously retrieve and upload the current area's construction information, and continuously record supplementary information including the number of bracket adjustments, cable rewiring distance, and manual dust removal time during the on-site intervention process.

[0020] The second aspect of this invention provides an integrated optimization method and system for collaborative design of photovoltaic-storage power plants, comprising: a multi-source data acquisition module for acquiring on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic-storage power plant; preprocessing the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset; a structural performance mapping module for evaluating the real-time power generation status of components based on the standardized component status dataset, and identifying whether components have entered an inefficient operating segment and triggering corresponding operation and maintenance control operations based on the performance evaluation results; an operation deviation identification module for performing operation analysis on the operational consistency of components based on the standardized component status dataset and dynamically adjusting the cleaning scheduling priority based on the operation analysis results; and an operation and maintenance task reconstructing module for comprehensively evaluating the component task progress status using the performance evaluation results and operation analysis results as input, and dynamically deriving the current optimal task adjustment intensity based on the comprehensive evaluation results.

[0021] Beneficial effects

[0022] The present invention has the following beneficial effects:

[0023] (1) The integrated optimization method and system for collaborative design of photovoltaic and energy storage power stations achieves a data closed loop from static design to dynamic operation and maintenance by integrating configuration parameters in the construction phase and deviation behavior data in the operation phase, thereby improving the consistency of structural response and the ability to trace the source of problems throughout the entire life cycle.

[0024] (2) The integrated optimization method and system based on the collaborative design of photovoltaic and energy storage power stations automatically generates suggestions for energy storage scheduling, component replacement and shading path adjustment by establishing an optimization recommendation logic based on structural constraints and historical deviation trends, thereby reducing reliance on manual judgment and improving response efficiency.

[0025] (3) The integrated optimization method and system based on the collaborative design of photovoltaic and energy storage power stations can dynamically identify areas with disordered power generation behavior by clustering analysis of the two-dimensional coordinates and physical layout information of components, automatically generate hot spot segment markers and high-frequency sampling strategies, and improve the coverage and frequency control capability of regional response.

[0026] (4) The integrated optimization method and system based on the collaborative design of photovoltaic and energy storage power stations, through standardized interaction with the operation and maintenance terminal, ensures the synchronous controllability of each link of task generation, dispatch, execution and feedback, and realizes closed-loop scheduling and remote linkage of on-site tasks.

[0027] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0028] Figure 1 This is a flowchart of the integrated optimization method for collaborative design of photovoltaic and energy storage power stations based on the present invention.

[0029] Figure 2 This is a structural diagram of the integrated optimization system for collaborative design of photovoltaic and energy storage power stations based on the present invention.

[0030] Figure 3 This is a heat map showing the operational deviation of the components involved in this invention;

[0031] Figure 4 This is a line graph of the task optimization feedback value involved in this invention. Detailed Implementation

[0032] 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.

[0033] Please see Figures 1-4This invention provides a technical solution: a collaborative design integrated optimization method and system for photovoltaic-storage power plants, comprising: S1, collecting on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic-storage power plant; preprocessing the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset; S2, based on the standardized component status dataset, performing performance evaluation on the real-time power generation status of the components, and identifying whether the components have entered an inefficient operating section based on the performance evaluation results and triggering corresponding operation and maintenance control operations; S3, based on the standardized component status dataset, performing operational analysis on the operational consistency from the component energy output, and dynamically adjusting the cleaning scheduling priority based on the operational analysis results; S4, using the performance evaluation results and operational analysis results as input, comprehensively evaluating the component task progress status, and dynamically deriving the current optimal task adjustment intensity based on the comprehensive evaluation results.

[0034] Specifically, the steps for collecting on-site structural data, environmental and layout data, and status performance data during the construction and operation of photovoltaic and energy storage power stations are as follows:

[0035] The on-site structural data of the photovoltaic array area was collected, including: the static installation tilt angle of each photovoltaic module, module surface images, hot spot recurrence frequency, cable path length, surface temperature of the thermistor area on the back of the module, and connection path length between each connection point. Among them, the static installation tilt angle reflects the pitch state of the module during actual installation, which directly affects the light reception efficiency. The module surface images can be used to identify apparent anomalies such as contamination, shading, and microcracks. The hot spot recurrence frequency serves as a time-dependent indicator of internal heating anomalies in the module, which helps to analyze structural stress and local aging. The cable path length and connection path length reflect the internal wiring complexity and impedance accumulation characteristics. The surface temperature of the thermistor area characterizes the local heat dissipation performance of the backsheet area and is an important basis for judging the consistency of module heat dissipation and load response distribution.

[0036] Environmental and layout data were collected during the construction process. This data included: real-time irradiance values ​​received by each photovoltaic module, changes in cable contact resistance of the module connection loops, ambient temperature at the location of the photovoltaic array, return path length for operation and maintenance scheduling, and instantaneous wind speed disturbance values. The standard deviation of wind speed disturbances was also calculated and recorded. Among these, irradiance values ​​reflect the incident energy level of the modules and are a direct parameter driving power generation efficiency. Changes in cable contact resistance characterize the reliability of connection points and the aging trend of the lines. Ambient temperature, as a macroscopic thermal field distribution factor, affects the offset of the module's operating point and the heat dissipation boundary conditions. Return path length measures operation and maintenance response efficiency and regional accessibility. Instantaneous wind speed disturbance values ​​and their standard deviation reflect the frequency and amplitude of external disturbances impacting the module's thermal field and the stability of the supporting structure.

[0037] The system collects performance data on operational status, including: DC output power, output energy, real-time output voltage of each photovoltaic module, absorbed energy of the energy storage channel, and real-time connection voltage and instantaneous temperature rise at each connection point. Simultaneously, it calculates and records average output energy, average absorbed energy, and average output voltage. DC output power reflects the instantaneous power generation capacity of the module under current environmental and structural conditions; output energy is used to track medium- and long-term operating efficiency trends; real-time output voltage helps identify module operating condition fluctuations and local mismatch risks; absorbed energy of the energy storage channel is used to assess the dynamic coupling degree of the energy storage load to the photovoltaic power generation response; and real-time connection voltage and instantaneous temperature rise serve as feedback indicators of connection stability and localized heating conditions, and are key parameters for assessing electrical connection safety.

[0038] This implementation plan provides a comprehensive and structured data foundation for the integrated optimization design of photovoltaic and energy storage power stations, supporting the collaborative optimization of subsequent task scheduling, performance evaluation, and control strategy formulation. By collecting three types of core data covering the structural, construction, and operational layers, a multi-dimensional characterization of the photovoltaic module installation status, construction layout integrity, and dynamic changes in operational performance can be achieved, bridging the information gap between the construction and operation and maintenance phases. The coordinated collection of these three types of data not only enhances the comprehensiveness and continuity of status awareness but also lays a data support foundation for constructing dynamic task control logic, thereby effectively mitigating scheduling deviations and resource waste caused by information inconsistencies throughout the entire lifecycle.

[0039] Specifically, the preprocessing of on-site structural data, environmental and layout data, and performance data to construct a standardized component status dataset involves the following steps: For on-site structural data, the static installation tilt angles of each photovoltaic module must first be uniformly converted into angle units to ensure unit consistency. Then, outliers are eliminated and upper and lower limits are corrected based on physical boundary conditions to avoid misleading subsequent shading judgments due to incorrect tilt angle entries. If broken path records are found in the cable path length and connection point path length, linear interpolation should be used to complete them. The interpolation strategy is based on the physical distance between the path start and end points and the assumption of layout continuity, and is limited to the maximum interpolation path length threshold to prevent overfitting. The frequency of hot spot recurrence is aligned on the construction time axis, and a dual mapping table of component number and construction node is introduced to ensure positioning accuracy. The surface temperature data of the heat-sensitive area is processed using a three-point sliding median filter method, with the window size set to 5 consecutive time steps, combined with the physical upper limit of temperature fluctuations to filter out abrupt abnormal data.

[0040] When processing environmental and layout data, the light intensity value is introduced into a 5-minute moving average processing window to smooth sampling fluctuations and retain periodic trends. Before uniformly mapping the change in cable contact resistance to the component numbering system, the conductor loop topology mapping must be completed first. Missing ambient temperature values ​​are filled in using linear interpolation, which is based on the temperature change rate constraint of adjacent time nodes. At the same time, physical outlier criteria are established for removal. The return path length of operation and maintenance scheduling is mapped to the geographic location number through the binding relationship between GIS coordinates and component numbers, and disconnected paths with a path length of 0 and erroneous path records that exceed the site boundary are cleared.

[0041] During the processing of operational performance data, the DC output power and output energy are first aligned with a 5-second precision timestamp to ensure consistent sampling time. In the energy storage channel's energy absorption, records with incorrect formats are removed using regular expression validation, and deduplication is performed to prevent data accumulation and offset. A one-to-one mapping table is constructed between the modules and the energy storage channel for number verification, clearing missing and duplicate numbers. For extreme values ​​identified in the output voltage and instantaneous temperature rise, such as those exceeding 15% of the module's rated voltage and those with instantaneous temperature rises greater than 40°C, the mean value within a window is used for substitution. The window size is two data points before and after the current time to reduce the impact of interference on the dynamic trend.

[0042] Finally, the standardized on-site structural data, environmental and layout data, and state performance data are uniformly normalized. A 0-to-1 distributed standardized component state dataset is constructed using a minimum-maximum normalization strategy, providing unified dimensional support for subsequent correlation analysis and optimized scheduling.

[0043] This implementation scheme eliminates heterogeneity in acquisition accuracy, recording format, time granularity, and physical units among photovoltaic array on-site structural data, construction process layout data, and operational performance data through standardization and normalization of multi-source raw data. This ensures the comparability of different data dimensions in the numerical space and the accuracy of fusion calculations. By removing and correcting outliers, interpolating and completing missing data, constructing consistent structural mappings, and smoothing out extreme values, data integrity and stability are improved. Furthermore, a unified-dimensional component status dataset is constructed after normalization, providing highly reliable and consistent foundational data support for subsequent key tasks such as component operation deviation analysis, anomaly identification, and dynamic scheduling optimization. This avoids analytical biases and misjudgments caused by inconsistent raw data.

[0044] Specifically, based on a standardized component state dataset, the performance evaluation of the component's real-time power generation status involves the following steps: obtaining the normalized real-time connection voltage and connection path length at the positions before and after the connection point; calculating the voltage difference between adjacent time periods using a sliding window method; dividing the voltage difference between adjacent time periods by the corresponding connection path length to obtain the cable's voltage drop per unit length rate of change; multiplying the cosine of the component's static installation tilt angle by the difference between one and the light intensity value, and then adding it to the change in cable contact resistance to obtain the tilt angle shading correction value; dividing the surface temperature of the thermistor zone by the ambient temperature and then adding one to obtain the temperature rise correction ratio; taking the cable path length as the power of the cable's voltage drop per unit length rate of change to obtain the path length loss ratio; and using the DC output power of the photovoltaic module as the numerator, and the product of the tilt angle shading correction value, the temperature rise correction ratio, and the path length loss ratio as the denominator to obtain the structural performance expression value.

[0045] The formula for calculating the structural performance expression value is:

[0046] ;

[0047] In the formula: This represents the DC output power of the photovoltaic module, reflecting its energy output capability under current operating conditions. It is a direct indicator of the module's power generation efficiency and originates from the module-level voltage and current acquisition module. The static installation tilt angle of the component is used to describe its fixed angle relative to the horizontal plane. It is an important structural parameter for calculating the relationship between light incidence efficiency and shading. It is derived from the structural design drawings and digital modeling results of the support posture during the construction phase. This represents the real-time light intensity value received by the module, which is used to measure the light input per unit area. It is an important external condition affecting power generation efficiency and originates from the light intensity sensor on the module panel. This represents the change in cable contact resistance in the component connection loop, used to reflect changes in the conduction state at the line connection, and is derived from the real-time acquisition and calculation results of the cable voltage drop to current ratio; It represents the surface temperature of the thermally sensitive area on the back of the component, which reflects the heat load level during operation. It is an important physical quantity that causes efficiency changes due to temperature, and is obtained from the thermocouple and infrared thermal sensor acquisition module. The ambient temperature at the location of the photovoltaic array is used to measure the current external cooling capacity and is derived from a local ambient temperature sensor. It represents the cable path length between the component and the nearest combiner box, and is used to reflect the cable impedance effect in the signal and power transmission process. It is an important factor in assessing long-distance cabling loss and is derived from cabling design drawings and GIS path reconstruction data. It represents the rate of change of voltage drop per unit length of cable, used to reflect the degree of nonlinear loss generated in different line sections during actual operation. It is a calculation parameter for quantifying the transmission attenuation characteristics of the path, derived from multi-point voltage measurement data collected synchronously under different path lengths, and calculated by fitting the relationship between the voltage at the beginning and end of the cable and its corresponding length.

[0048] This implementation plan comprehensively evaluates the overall structural performance of photovoltaic modules under current operating conditions. Its core lies in coupling the module's actual power generation capacity with its static installation structure, solar environment, cable connection quality, and thermal load in heat-sensitive areas. By dividing the output power value by multiple correction factors based on structure and operating status, the performance deviation of the module under the current arrangement and environmental constraints can be quantified. This reveals the true state of how module performance is affected by structural parameters, providing a quantitative reference for subsequent anomaly diagnosis, tuning, and structural optimization. In practical applications, this structural performance value can serve as input for key control logic such as photovoltaic-energy storage coordinated scheduling, module status identification, and maintenance priority ranking.

[0049] Specifically, the steps for identifying whether a module has entered an inefficient operating zone based on performance evaluation results and triggering corresponding operation and maintenance control operations are as follows: Real-time comparison of the current structural performance expression value of the photovoltaic module with the expression threshold: When the structural performance expression value is less than or equal to the expression threshold, the module is determined to have entered an inefficient operating zone. This state is usually accompanied by physical structural degradation problems such as obstructed light incidence, abnormal load accumulation in the backsheet thermal sensitive area, and abnormal voltage drop at the cable connection end. If operation continues, it will cause uneven power output, thermal mismatch accumulation, and energy transmission blockage in the series channel, further leading to increased string loss and decreased stability. To avoid the aforementioned adverse consequences, the DC combiner switch of the branch where the component is located was immediately shut down to prevent the degraded component from causing power interference to the upstream and downstream channels. At the same time, the surface cleaning of the component was initiated and the pollution level was marked simultaneously. The output capacity of the component was restored by improving the incident light efficiency. The priority of the energy storage discharge channel corresponding to the component was adjusted to the lowest level to reduce its control influence on the power output path and prevent inefficient components from occupying the discharge bandwidth. A high-frequency relocation image acquisition task for the shading source was pushed to the operation and maintenance terminal. Through multi-temporal shading source tracing and positioning, possible environmental shading objects and structural shading coupling problems were identified and eliminated to fundamentally restore the unobstructed power generation channel.

[0050] When the structural performance value exceeds the threshold, it indicates that the component is in a good structural condition and operating stably. Currently, there are no abnormal disturbances in illumination, thermal state, cable connections, or structural attitude. Therefore, the current circuit on / off state and power distribution strategy should be maintained to avoid unnecessary switching interference. Simultaneously, this component will be included in the performance stability monitoring list as a source of subsequent performance benchmark samples. Furthermore, the thermal load changes along the corresponding cable path will be continuously updated and recorded to establish a path load curve model under typical operating conditions, providing comparative data support for identifying potential structural changes and layout thermal coupling risks in the future.

[0051] This implementation plan achieves dynamic control of partitioned operation strategies based on component structural performance status and a linkage mechanism for structural degradation early warning, thereby improving the operational efficiency of photovoltaic-storage synergy, delaying component performance degradation, and enhancing disturbance rejection stability. By comparing structural performance values ​​with threshold values ​​in real time, multi-dimensional joint control measures can be triggered promptly when components enter an inefficient operating state: on the one hand, disconnecting inefficient component branches from energy storage discharge channels prevents crosstalk and energy storage scheduling conflicts caused by power instability, reducing string losses and power management difficulty; on the other hand, initiating surface cleaning and pollution level labeling directly improves the component's power generation environment and enhances its subsequent recovery potential; simultaneously, based on the push of high-frequency shading source relocation tasks, rapid source tracing and remediation of environmental shading are achieved, further improving the environmental adaptability of the site operation and the efficiency of image recognition data accumulation. When components are in a high structural performance state, the on / off and power strategies remain stable, reducing unnecessary adjustments, improving control response efficiency, and using the updated accumulation of cable path heat load data to build a long-term baseline for operational safety monitoring, providing data support for subsequent trend prediction and degradation anomaly identification. Overall, this step constructs a closed-loop control chain from anomaly identification to dynamic intervention, performance recovery, and state accumulation, providing key technical support for building a highly reliable, low-loss, and evolvable operating system for photovoltaic-storage power stations.

[0052] Specifically, based on a standardized component state dataset, the operational analysis of operational consistency from the perspective of component energy output involves the following steps: First, the difference between the component's output energy and the average output energy is divided by the sum of the average output energy and 1 to obtain the power generation fluctuation deviation rate. This ratio measures the deviation of a single component's power generation state relative to the stable mean of its array, reflecting the degree of local instability in its instantaneous power generation behavior and serving as a prerequisite for identifying abnormal power generation behavior. Second, the absorbed energy of the energy storage channel is divided by the average absorbed energy to obtain the energy storage response matching rate, characterizing the dynamic synchronization degree of the current energy storage channel with the component's output response and reflecting the adaptive absorption capacity of the energy storage side to power generation behavior. Third, the real-time output voltage of the component is divided by the sum of 1 and the average output voltage, then 1 is added, the natural logarithm is taken, and 1 is added again to obtain the voltage deviation correction value. A logarithmic transformation is introduced to enhance the identification capability of low-amplitude fluctuations, avoid the dominant influence of abnormal voltage values ​​on deviation indicators, and improve the response sensitivity to boundary disturbances. Sensitivity; The square of the instantaneous wind speed disturbance value at the component location is divided by the sum of the square of the standard deviation of the wind speed disturbance and one, and then one is added to obtain the normalized value of the disturbance intensity. This normalization expression is used to eliminate the influence of fluctuations in local abnormal wind speed data and ensure that the disturbance index has a consistent dimensional standard and physical interpretability at the full array scale. The reciprocal of the voltage deviation correction value is multiplied by the energy storage response matching rate to obtain the voltage-energy storage comprehensive value. This combined structure is used to reflect the degree of coordination between voltage fluctuations and energy storage absorption. When the voltage deviation is large but the energy storage response is rapid, this value can still remain stable, thus reflecting the compensation capability of coupling mismatch. The absolute value of the difference between the power generation fluctuation offset rate and the voltage-energy storage comprehensive value is multiplied by the normalized value of the disturbance intensity to obtain the component operation deviation value. This structure, by jointly expressing the bidirectional deviation between the power generation side and the energy storage side and performing scale unification under the disturbance weight, realizes the integrated evaluation mechanism of the consistency of operating status, response matching and local disturbance robustness.

[0053] The formula for calculating the component's operational deviation is:

[0054] ;

[0055] In the formula: It represents the electrical energy output of the component and is used to quantify the actual energy output level of the component within this cycle. It is a basic parameter for measuring operational stability and energy output integrity, and is derived from the energy accumulation data of the inverter and the power metering module. It represents the average output energy of the component within the sliding time window, and is used to reflect the statistical baseline of recent energy output. It is a comparative parameter for identifying energy fluctuation trends and is derived from the sliding average calculation of historical energy collection data. It represents the absorbed electrical energy of the energy storage channel and is used to evaluate the real-time matching degree between the photovoltaic output and the energy storage response. It is a direct parameter for judging the consistency of energy coupling and comes from the charging metering module of the energy storage converter. It represents the average electrical energy absorbed by the energy storage channel within the sliding time window, used to establish a historical energy storage response baseline, and is a reference parameter for judging the current rate of change in energy absorption. It is derived from the time series data collected and recorded by the BMS. It represents the real-time output voltage of the component, which reflects the electrical output stability under the current operating state. It is a key signal for identifying local impedance rise and poor contact, and it originates from the DC combiner box and the component-level voltage sampling module. It represents the average output voltage of components within the same string in the current cycle, used to construct a comparative reference for voltage distribution, and is a group indicator for judging the degree of deviation of the output of a single component. It is derived from the aggregation results of the multi-node voltage sampling module. It represents the instantaneous wind speed disturbance value at the location of the component, used to characterize the immediate impact of external environmental fluctuations on the operation of the component, and is a disturbance correction parameter for identifying operational anomalies, which is derived from the micro-meteorological wind speed sensor in the array area; It represents the standard deviation of wind speed disturbance within the sliding window where the component is located. It is used to quantify the stability of wind speed fluctuations and serves as a reference for normalizing the impact of environmental disturbances. It is derived from time series statistical analysis of wind speed data.

[0056] This implementation plan is used to assess the deviation of energy output of each photovoltaic module within a specific period and to identify modules with poor energy storage coordination capabilities and unstable electrical fluctuations. This supports the execution of high-precision module operation status diagnosis and energy storage channel adjustment strategies. The module operation deviation value output by the formula serves as the criterion for subsequent deviation level classification, energy storage priority ranking, module operation and maintenance early warning, and local thermal barrier identification, which can significantly improve the automation and accuracy of photovoltaic module fault location and coordinated energy storage control.

[0057] Specifically, the steps for dynamically adjusting the cleaning scheduling priority based on the operational analysis results are as follows: In order to improve the targeting of photovoltaic module operation status identification and the ability of group dynamic control, the operation deviation values ​​of all modules are first sorted from low to high. Then, based on the inflection point fitting of the overall deviation distribution, piecewise gradient change and Knee-point learning, the quantile threshold q is automatically selected to accurately determine the boundary of high-risk modules. Then, modules with deviation values ​​in the last q quantiles are identified as modules with disordered power generation behavior and included in the scheduling list of cleaning tasks.

[0058] Subsequently, the relative increase in deviation values ​​of the aforementioned components over two consecutive measurement periods was calculated to capture the time dynamic trend of their changes. This increase sequence was fed into a Holt-Winters trend model for multi-step trend prediction analysis to comprehensively determine whether it has a significant positive growth trend. Once the trend is significant, it is marked as a component with an increasing deviation trend, triggering a structural inspection process. A drone platform is then used to conduct joint infrared and visible light inspections, focusing on checking the component surface shading, the physical integrity of cable connections, and the backplane contamination level to prevent hot spot accumulation and output loss caused by continuous deviation growth.

[0059] At the spatial identification level, the two-dimensional geographic coordinates of all components are further input into the density clustering model DBSCAN to form local spatial clustering units. The average operational deviation value of the components within each cluster is taken, and the spatial group with the highest average deviation value is identified. Based on this, suggestions for the redistribution of energy storage channels are generated to improve the overall energy management balance. At the same time, an operational deviation heatmap is constructed based on the coordinates and deviation values ​​of each component, and the area with the highest average deviation value is highlighted in the map and marked as a dynamic hotspot segment. Its sampling frequency is automatically increased to enhance data timeliness and response speed to local operational fluctuations, enabling early identification and key monitoring of high-risk areas.

[0060] like Figure 3 The diagram shown is a heatmap of component operation deviation provided in this application example, used to reflect the current operating behavior and state fluctuations of the components. The horizontal axis represents component column numbers C1 to C10, and the vertical axis represents component row numbers R1 to R10, forming a 10×10 spatial layout matrix. The color uses a gradient from dark blue to dark red to represent the degree of deviation; the closer the color is to red, the greater the component operation deviation value and the more unstable the operating state. The component operation deviation value is calculated comprehensively based on factors including the dynamic inconsistency between actual power generation output and energy storage response, voltage deviation, and the intensity of local disturbance fluctuations, possessing good data acquisition and real-time performance. It is clearly visible in the diagram that a high deviation area, ranging from red to orange, forms slightly to the right of the center of the component array, indicating that the components in this area exhibit significant performance degradation, abnormal energy output, and environmental adaptation imbalances within the current cycle. This concentrated deviation area is usually caused by dust accumulation, obstruction, poor connection contact, and enhanced wind disturbance. The outer blue-green area shows low component operation deviation values, representing good operating conditions and stable output. This deviation heatmap not only serves as a visual representation of operational health status assessment but also supports operational and maintenance decision support. It can automatically prioritize high-deviation areas for cleaning and structural inspection, increasing sampling frequency and enhancing trend tracking; simultaneously, it uses low-deviation fluctuation areas as a stable operational benchmark to support adaptive scheduling optimization and local reference.

[0061] This implementation scheme achieves the distribution identification, trend tracking, spatial clustering, and regional hotspot labeling of photovoltaic module operating deviations, thereby accurately locating potentially risky modules and mismatched areas, and dynamically optimizing energy storage paths and sampling strategies to enhance overall operational stability and control accuracy. A deviation threshold q is selected through a self-learning method, avoiding rigid misjudgments caused by manual settings and better reflecting the actual evolution and distribution of operating deviations, thus improving the adaptability of identifying power generation behavior disturbances. The Holt-Winters model is used to identify modules with positive deviation growth trends, enabling proactive screening and intervention before significant power losses and latent faults occur, significantly reducing energy losses caused by surface shading, loose cables, and thermal mismatch. Furthermore, spatial clustering enables local area deviation mean analysis, improving the resolution and focus of regional diagnostics. The construction of dynamic deviation heatmaps and the adjustment of sampling frequencies enhance the tracking intensity of high-risk hotspots in both spatiotemporal dimensions, providing valuable references for regional key cleaning, energy storage scheduling reconfiguration, and priority allocation for UAV re-inspections. Overall, this step expands operational deviations from static identification to dynamic evolution and spatial coordination, effectively supporting the leap from passive response to proactive prediction in photovoltaic module operation and maintenance management.

[0062] Specifically, using performance evaluation results and operational analysis results as input, the comprehensive evaluation of the component task progress status involves the following steps: acquiring component surface images; using image recognition and dust occlusion density recognition algorithms, segmenting and evaluating the component surface images based on texture, color, and edge blurring to obtain the component area occlusion coverage rate; multiplying the structural performance expression value, component operation deviation value, and maintenance scheduling return path length to obtain the structural maintenance evaluation value; adding the area occlusion coverage rate to the instantaneous temperature rise value of the cable connection node to obtain the wiring displacement correction value; adding the instantaneous wind speed disturbance value to the hot spot recurrence frequency, adding one, taking the natural logarithm, and adding one again to obtain the comprehensive environmental disturbance value; dividing the structural maintenance evaluation value by the product of the wiring displacement correction value and the comprehensive environmental disturbance value, adding one, and taking the natural logarithm to obtain the task optimization feedback value.

[0063] The formula for calculating the task optimization feedback value is:

[0064] ;

[0065] In the formula: It represents the structural performance value, used to evaluate the spatial binding relationship of the component's power generation performance based on the support posture and cable laying status; This represents the component's operational deviation value, used to reflect the consistency of the component's current energy output with its optimal output. This indicates the length of the operation and maintenance scheduling return path, which is used to measure the migration distance of operation and maintenance personnel and robots required when the current work order is executed. It is derived from the GIS operation and maintenance map. The area occlusion coverage of the component is calculated by image recognition and dust occlusion density recognition algorithms, reflecting the local cleanliness level; This indicates the instantaneous temperature rise of the cable connection node, which is obtained by the laying point temperature sensor and used to identify areas with high contact resistance. It represents the instantaneous wind speed disturbance value at the location of the component, used to characterize the immediate impact of external environmental fluctuations on the operation of the component, and is a disturbance correction parameter for identifying operational anomalies, which is derived from the micro-meteorological wind speed sensor in the array area; This indicates the frequency of hot spot recurrence in the area where the component unit is located. It is statistically analyzed by the infrared imaging module and reflects the activity level of local abnormal heat distribution.

[0066] In this implementation example, Example 1 sets the structural performance expression value to 1.2, the component operation deviation value to 0.80, the operation and maintenance scheduling return path length to 5.2, the area shading coverage rate to 0.50, the instantaneous temperature rise value to 0.60, the instantaneous wind speed disturbance value to 0.70, and the hot spot recurrence frequency to 1.10.

[0067] In Example 2, the structural performance expression value is set to 1.4, the component operation deviation value is 0.90, the operation and maintenance scheduling return path length is 4.8, the area shading coverage rate is 0.40, the instantaneous temperature rise value is 0.65, the instantaneous wind speed disturbance value is 0.60, and the hot spot recurrence frequency is 1.05.

[0068] In Example 3, the structural performance expression value is set to 1.1, the component operation deviation value is 0.85, the operation and maintenance scheduling return path length is 6.0, the area shading coverage rate is 0.60, the instantaneous temperature rise value is 0.58, the instantaneous wind speed disturbance value is 0.75, and the hot spot recurrence frequency is 1.15.

[0069] In Example 4, the structural performance expression value is set to 1.6, the component operation deviation value is 0.95, the operation and maintenance scheduling return path length is 5.5, the area shading coverage rate is 0.55, the instantaneous temperature rise value is 0.70, the instantaneous wind speed disturbance value is 0.65, and the hot spot recurrence frequency is 1.08.

[0070] In Example 5, the structural performance expression value is set to 1.3, the component operation deviation value is 0.88, the operation and maintenance scheduling return path length is 5.0, the area shading coverage rate is 0.45, the instantaneous temperature rise value is 0.62, the instantaneous wind speed disturbance value is 0.68, and the hot spot recurrence frequency is 1.12.

[0071] In Example 6, the structural performance expression value is set to 1.7, the component operation deviation value is 1.00, the operation and maintenance scheduling return path length is 4.9, the area shading coverage rate is 0.43, the instantaneous temperature rise value is 0.67, the instantaneous wind speed disturbance value is 0.70, and the hot spot recurrence frequency is 1.07.

[0072] In Example 7, the structural performance expression value is set to 1.5, the component operation deviation value is 0.92, the operation and maintenance scheduling return path length is 5.4, the area shading coverage rate is 0.52, the instantaneous temperature rise value is 0.66, the instantaneous wind speed disturbance value is 0.72, and the hot spot recurrence frequency is 1.09. The task optimization feedback value for each example is calculated, as shown in Table 1.

[0073] Table 1 Task Optimization Feedback Value Data Table

[0074]

[0075] like Figure 4 As shown in Table 1, this is a line graph of the task optimization feedback values ​​provided in this application example. Figure 4 As can be seen, Example 3 has the lowest task optimization feedback value, reflecting that this component has a highly consistent optimization response potential in terms of structural performance expression value, component operation deviation, temperature rise state, and hot spot reproduction. This indicates that its current control path has good thermal stability characteristics and dynamic compensation plasticity, making it suitable as a key optimization component to implement a refined scheduling strategy to achieve shading avoidance and heat load balance. Conversely, Example 6 has the highest task optimization feedback value. Although its hot spot reproduction frequency is relatively high, the component operation deviation and temperature rise disturbance fluctuation amplitude are large, resulting in poor overall feedback response stability. It is suitable to maintain the current state monitoring level and delay intervention to avoid resource waste and false triggering. The horizontal comparison of task optimization feedback values ​​shows the response consistency and control adaptability of each example under multi-parameter disturbances. The higher the feedback value, the more significant the task effect response and the more timely the adaptive adjustment. It can be used as a priority processing unit in the distributed scheduling strategy to support the dynamic improvement of the overall thermal coupling dissipation efficiency and shading robustness of the photovoltaic module network. The line graph of task optimization feedback values ​​further reveals the discrete characteristics of the responses of different components in the optimization path, which facilitates the formulation of target selection strategies for subsequent UAV review, cleaning scheduling and energy storage redistribution.

[0076] Specifically, the steps for dynamically deriving the current optimal task adjustment intensity based on the comprehensive evaluation results are as follows: The current task optimization feedback value is compared with the optimization feedback threshold in real time, and the feedback results are dynamically graded according to the first optimization threshold and the second optimization threshold. The optimization feedback value is constructed by fusing multi-dimensional information such as structural performance expression, component operation deviation, instantaneous disturbance, occlusion rate, and hotspot recurrence frequency, and has a strong ability to characterize coupling mismatch response.

[0077] When the task optimization feedback value is less than or equal to the second optimization threshold, it is determined that the component operation status in this area is stable and the intervention benefit is low. The operation and maintenance dispatch in the current area is immediately stopped to avoid resource waste. At the same time, the bracket tilt angle, cable laying length and energy storage interface configuration parameters corresponding to the components in this area are extracted and archived as standard samples with good operating performance for subsequent configuration reference and planning optimization calls in low feedback areas, so as to support the optimization transmission at the structural level and component selection recommendation.

[0078] When the task optimization feedback value is greater than the second optimization threshold and less than or equal to the first optimization threshold, the region is identified as being in a performance transition zone, with a slight tendency for operational disorder. The maintenance and inspection rhythm within the region is automatically rearranged, increasing the cleaning frequency and data sampling frequency to improve local response sensitivity. Based on the current component configuration information, a comparison list of construction constraints and operational deviation values ​​is generated. The comparison list details the structural information of the component's bracket type, connection method, wiring mode, and installation location. Structural feedback suggestions are pushed through the maintenance terminal for maintenance personnel to review and make decisions.

[0079] When the task optimization feedback value exceeds the first optimization threshold, it is determined that there is a significant structural-operational coupling mismatch in the area, requiring immediate on-site intervention. The cleaning tasks for all components in this area are prioritized to the highest level. A high-priority on-site task order is immediately generated on the operations and maintenance terminal, and historical construction records, drawings, and configuration parameters for the current area are simultaneously uploaded to the on-site dispatch platform. During the on-site intervention, the operations and maintenance terminal will record supplementary information in real time, including the number of support adjustments, cable rewiring distance, and manual dust removal time. This information is automatically merged into the operations and maintenance record database for this area, providing rich samples of intervention behaviors and operational feedback for the subsequent adaptive adjustment of the task optimization feedback mechanism.

[0080] This implementation plan utilizes dynamic hierarchical control based on task optimization feedback values ​​to achieve refined scheduling of operation and maintenance resources and rapid identification and intervention of structural mismatch issues. By introducing a first optimization threshold and a second optimization threshold, the operational status of components in the current region can be judged at multiple levels, corresponding to three response levels: "maintaining stable sample archiving," "generating moderate adjustment strategies," and "on-site intervention for severe problems." When the feedback value is low, redundant dispatching is stopped, and high-quality samples are extracted for standard configuration archiving, reducing ineffective operation and maintenance costs and accumulating excellent structural templates. When the feedback value is in the middle range, potential degradation trends are identified in advance through frequency adjustment and the push of structure-performance comparison lists, enhancing the stability of component response. When the feedback value exceeds the highest threshold, high-priority on-site intervention can be initiated immediately to quickly address critical issues such as severe shading, wiring errors, and hot spot recurrence, ensuring a closed loop in regional performance recovery and control rhythm. This mechanism realizes closed-loop linkage and adaptive intervention strategy generation between operation and maintenance tasks and structural configurations, improving the operational resilience and response efficiency of photovoltaic-storage power stations in complex environments.

[0081] The second aspect of this invention provides an integrated optimization method and system for collaborative design of photovoltaic and energy storage power stations, including: a multi-source data acquisition module, used to collect on-site structural data, environmental and layout data and state performance data during the construction and operation of photovoltaic and energy storage power stations, and to perform feature standardization and spatial matching correction on the collected data, and to construct a standardized component state dataset with location labels, structural attributes and energy efficiency index correlation features, to provide a unified analysis basis for subsequent performance mapping and deviation identification, and to ensure that the analysis process has cross-component consistency and historical trend comparability;

[0082] The structural performance mapping module is used to perform multi-index performance evaluation of the real-time power generation status of the components based on a standardized component status dataset. It comprehensively considers the component output power, temperature rise response and shading characteristics. By constructing a mapping relationship between structural information and energy efficiency output, it identifies whether the components have entered the inefficient operation section. After the identification result is triggered, it links the downstream operation deviation analysis and cleaning scheduling process to achieve early identification and closed-loop response to shading mismatch and structural mismatch problems.

[0083] The operation deviation identification module is used to perform deviation modeling and analysis on the operation consistency based on the standardized component status dataset and the component energy output characteristics. It calculates the operation deviation value by combining wind speed disturbance, voltage response and energy storage absorption synergy behavior, and dynamically adjusts the cleaning scheduling priority of the components based on the ranking results of the operation deviation value, so as to realize the real-time allocation optimization of resource deployment and the reconstruction of control path.

[0084] The operation and maintenance task refactoring module is used to construct a comprehensive quantitative indicator system for task progress status based on performance evaluation results and operation analysis results. It integrates the task completion progress, structural intervention history and component recovery trend in the current area, conducts multi-dimensional iterative analysis of control intervention and resource allocation strategies, and dynamically derives the current optimal task adjustment intensity based on the comprehensive evaluation results. This realizes an adaptive control closed loop driven by component state evolution, thereby improving the task coordination efficiency and problem response time during the operation of the photovoltaic and energy storage power station.

[0085] In this implementation plan, the multi-source data acquisition module is responsible for collecting on-site structural data, environmental and layout data, and status and performance data during the construction and operation of the photovoltaic-storage power station. After collection, the various heterogeneous data are standardized and preprocessed to construct a standardized component status dataset with a unified format, spatial matching attributes, and timestamp alignment. The key role of this module is to provide high-quality, full-volume input data for subsequent structural evaluation and scheduling optimization, ensuring the consistency and accuracy of the analysis logic.

[0086] The structural performance mapping module, based on a standardized component status dataset, constructs a correlation model between structural parameters such as support type, component orientation, and arrangement density, and component power output and voltage deviation. Through real-time performance evaluation, it determines whether components are operating inefficiently, such as experiencing continuous power degradation or rapid temperature rise. Upon identifying inefficient operating sections, it triggers relevant operation and maintenance control operations, such as local cleaning and component replacement recommendations. The module's function is to establish a structure-performance coupling mapping mechanism, quickly identifying potential structural problems affecting power generation efficiency.

[0087] The operational deviation identification module further analyzes the consistency and coordination of component operation. Based on signals from component energy output, energy storage response, and voltage offset, it constructs an operational deviation value index to quantify whether component behavior deviates from the normal range. Components with large operational deviation values ​​typically indicate contamination, hot spots, shading, and abnormal energy storage connections. This module sorts the operational deviation values ​​and dynamically adjusts the priority of cleaning scheduling, thereby achieving precise resource allocation and improving operation and maintenance efficiency.

[0088] The operation and maintenance task reconfiguration module takes structural performance evaluation results and operational deviation analysis results as input to comprehensively assess the current task progress status of components, considering multiple factors including: task completion rate, frequency of on-site intervention, and response to structural changes. By dynamically deriving the intensity of task adjustments, it rationally plans the inspection rhythm, resource allocation, and construction priority within the area, ensuring stronger intervention for high-deviation and high-sensitivity components and reducing resource scheduling in low-risk areas, thereby achieving dynamic reconfiguration and real-time optimal scheduling of photovoltaic and energy storage power station operation and maintenance tasks.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0090] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A collaborative design and integrated optimization method based on photovoltaic-storage power stations, characterized by: include: S1 collects on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic and energy storage power station. It preprocesses the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset. S2, based on a standardized component status dataset, performs performance evaluation on the real-time power generation status of the components, and identifies whether the components have entered an inefficient operating section based on the performance evaluation results and triggers corresponding operation and maintenance control operations. S3, based on a standardized component state dataset, performs operational consistency analysis on component energy output and dynamically adjusts the cleaning scheduling priority based on the operational analysis results; S4 takes the performance evaluation results and operation analysis results as input, performs a comprehensive evaluation of the component task progress status, and dynamically derives the current optimal task adjustment intensity based on the comprehensive evaluation results.

2. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for collecting on-site structural data, environmental and layout data, and status performance data during the construction and operation of the photovoltaic and energy storage power station are as follows: Collect on-site structural data of the photovoltaic array area. The on-site structural data includes: static installation tilt angle of each photovoltaic module, module surface image, hot spot recurrence frequency, cable path length, surface temperature of the thermal sensitive area on the back of the module, and connection path length between each wiring point. Collect environmental and layout data during the construction process. The environmental and layout data include: real-time light intensity value received by each photovoltaic module, change in cable contact resistance of module connection circuit, ambient temperature at the location of the photovoltaic array, return path length of operation and maintenance scheduling, instantaneous wind speed disturbance value, and calculate and record the standard deviation of wind speed disturbance. The system collects performance data on the operating status, including: DC output power, output energy, real-time output voltage of each photovoltaic module, absorbed energy of the energy storage channel, and real-time connection voltage and instantaneous temperature rise at each connection point. It also calculates and records the average output energy, average absorbed energy, and average output voltage.

3. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for preprocessing on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset are as follows: For the on-site structural data, the static installation tilt angle of each photovoltaic module is uniformly converted into angle units, and abnormal values ​​that exceed the physical range are corrected. The broken path records in the cable path length and the connection path length of the connection point are interpolated and completed. The hot spot recurrence frequency is aligned to the specific module number according to the construction time axis. The median filtering method is used to eliminate the sudden noise of surface temperature change in the heat-sensitive area. For environmental and layout data, the real-time received light intensity values ​​are processed by moving average filtering. The change in cable contact resistance is mapped to a unified numbering system in conjunction with the component number. The missing data in the ambient temperature is filled by linear interpolation, and physical anomalies are removed. The return path length of operation and maintenance scheduling is matched to the geographical location number, and disconnected and erroneous path records are cleared. In the standardization of operating status performance data, the DC output power and output energy are timestamped, format errors and duplicate records in the energy absorbed by the energy storage channel are removed, and the number mapping verification between the components and the energy storage channel is completed. Extreme values ​​of output voltage and instantaneous temperature rise are removed and replaced with the mean value within the window. The standardized on-site structural data, environmental and layout data, and status performance data are normalized to construct a standardized component status dataset.

4. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for evaluating the real-time power generation status of components based on the standardized component status dataset are as follows: Obtain the real-time wiring voltage and connection path length before and after the normalized wiring point, calculate the voltage difference between adjacent time periods using a sliding window method, and divide the voltage difference between adjacent time periods by the corresponding connection path length to obtain the voltage drop change rate per unit length of the cable. Multiply the cosine of the static installation tilt angle of the component by the difference between the light intensity value and the value of the light intensity, and then add it to the change in cable contact resistance to obtain the tilt angle shading correction value. Divide the surface temperature of the heat-sensitive area by the ambient temperature and then add one to obtain the temperature rise correction ratio. The path length loss ratio is obtained by raising the cable path length to the power of the rate of change of voltage drop per unit length of the cable. The structural performance value is obtained by taking the DC output power of the photovoltaic module as the numerator and the product of the tilt angle shading correction value, the temperature rise correction ratio and the path length loss ratio as the denominator.

5. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for identifying whether a component has entered an inefficient operating segment based on performance evaluation results and triggering corresponding operation and maintenance control operations are as follows: Real-time comparison of the current structural performance expression value and expression threshold of the photovoltaic module: when the structural performance expression value is less than or equal to the expression threshold, the module is determined to have entered the inefficient operation zone, the DC combiner switch of the branch where the module is located is immediately turned off, the module surface cleaning is started and the pollution level is marked simultaneously, the priority of the corresponding energy storage discharge channel of the module is adjusted to the lowest, and the high-frequency repositioning image acquisition task of the shading source is pushed to the operation and maintenance terminal. When the structural performance value exceeds the threshold, the current circuit on / off state and power distribution strategy remain unchanged. The current component is then included in the performance stable section monitoring list, and the thermal load change of the corresponding cable path of the current component is updated and recorded.

6. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for performing operational consistency analysis based on component energy output using a standardized component state dataset are as follows: The difference between the component's output power and the average output power is divided by the sum of the average output power and the average output power to obtain the power generation fluctuation offset rate. The energy storage response matching rate is obtained by dividing the absorbed energy of the energy storage channel by the average absorbed energy. Divide the real-time output voltage of the component by the sum of the average output voltage, add one, take the natural logarithm, and add one again to obtain the voltage deviation correction value. The square of the instantaneous wind speed disturbance value at the component location is divided by the sum of the square of the standard deviation of the wind speed disturbance and one, and then one is added to obtain the normalized value of the disturbance intensity. Multiply the reciprocal of the voltage deviation correction value by the energy storage response matching rate to obtain the voltage-energy storage comprehensive value. Multiply the absolute value of the difference between the power generation fluctuation deviation rate and the voltage-energy storage comprehensive value by the disturbance intensity normalization value to obtain the component operation deviation value.

7. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for dynamically adjusting the cleaning scheduling priority based on the operational analysis results are as follows: The component operating deviation values ​​of all components are sorted from low to high. Components with operating deviation values ​​in the last q quantile are identified as components with disordered power generation behavior and the task scheduling list is cleaned. The relative increase of the component operating deviation value of the same component in disordered power generation behavior is calculated from two consecutive calculations. The Holt-Winters trend model is input to perform multi-step analysis on the deviation trend. Components with a significant positive increase trend are marked as components with increasing deviation trend. The UAV structure is immediately reviewed to check the surface obstruction, cable connection points and backplate contamination status of components with increasing deviation trend. Using the two-dimensional geographic coordinates of all components as input, a density-based spatial clustering algorithm is used to spatially cluster the components. The mean value of component operation deviation within each group is calculated, and the group with the highest mean value of component operation deviation is obtained to generate a redistribution suggestion for energy storage channels. At the same time, a component operation deviation heatmap is constructed based on the component two-dimensional coordinates and the distribution of component operation deviation values. The area with the highest mean value of component operation deviation is marked as a dynamic hot spot segment, and the sampling frequency of the area is increased.

8. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for comprehensively evaluating the component task progress status using performance evaluation results and operational analysis results as input are as follows: The component surface image is acquired. Based on the texture, color and edge blur of the component surface image, the component surface image is segmented and evaluated at the pixel level to obtain the component area occlusion coverage. The structural performance value, component operation deviation value, and operation and maintenance scheduling return path length are multiplied to obtain the structural operation and maintenance evaluation value. The wiring displacement correction value is obtained by adding the area shading coverage rate to the instantaneous temperature rise value of the cable connection node; Add the instantaneous wind speed disturbance value to the hot spot recurrence frequency, add one, take the natural logarithm, and add one again to obtain the comprehensive environmental disturbance value. Divide the structural operation and maintenance assessment value by the product of the wiring displacement correction value and the comprehensive environmental disturbance value, add one and take the natural logarithm to obtain the task optimization feedback value.

9. The integrated optimization method for collaborative design of photovoltaic-storage power stations according to claim 1, characterized in that: The specific steps for dynamically deriving the current optimal task adjustment intensity based on comprehensive evaluation results are as follows: Real-time comparison of the current task optimization feedback value with the optimization feedback threshold, which includes a first optimization threshold and a second optimization threshold: When the task optimization feedback value is less than or equal to the second optimization threshold, stop the operation and maintenance dispatch in the current area, and extract the bracket tilt angle, cable laying length and energy storage interface configuration in the section and archive them as standard samples. When the task optimization feedback value is greater than the second optimization threshold and less than or equal to the first optimization threshold, the operation and maintenance inspection rhythm in the area is rearranged, the cleaning frequency and sampling frequency are increased, and a comparison list of component construction constraints and operation deviations is generated. Specific structural information is marked in the comparison list, and structural feedback suggestions are pushed out. When the task optimization feedback value exceeds the first optimization threshold, on-site intervention is immediately initiated. The priority of component cleaning tasks in the area is adjusted to the highest level, and an on-site task order is generated on the operation and maintenance terminal. Simultaneously, the construction information of the current area is retrieved and uploaded. During the on-site intervention process, supplementary information including the number of brackets adjusted, the distance of cable rewiring, and the time taken for manual dust removal is continuously recorded.

10. A collaborative design integrated optimization system based on photovoltaic-storage power stations, employing the collaborative design integrated optimization method based on photovoltaic-storage power stations as described in any one of claims 1-9, characterized in that: include: The multi-source data acquisition module is used to collect on-site structural data, environmental and layout data, and status performance data during the construction and operation of photovoltaic and energy storage power stations. It preprocesses the on-site structural data, environmental and layout data, and status performance data to construct a standardized component status dataset. The structural performance mapping module is used to evaluate the real-time power generation status of components based on a standardized component status dataset, and to identify whether the components have entered an inefficient operating section based on the performance evaluation results and trigger corresponding operation and maintenance control operations. The operation deviation identification module is used to perform operation consistency analysis based on the component energy output from the standardized component status dataset, and dynamically adjust the cleaning scheduling priority based on the operation analysis results; The operation and maintenance task refactoring module is used to comprehensively evaluate the progress status of component tasks based on performance evaluation results and operation analysis results, and dynamically deduce the current optimal task adjustment intensity based on the comprehensive evaluation results.

Images