Photovoltaic power generation system load prediction method based on big data

Abstract

The invention discloses a photovoltaic power generation system load prediction method based on big data, and relates to the technical field of photovoltaic power generation, and the method comprises the steps: determining a plurality of power generation sub-units and a connection relation network; constructing constraint conditions, and constructing a plurality of power generation databases; collecting first weather prediction information of a preset power supply area in a first prediction time zone, and inputting the first weather prediction information into a plurality of power generation databases for analysis to generate initial power generation data of a plurality of subunits; analyzing the fusion loss of the initial power generation data of the plurality of subunits after the electric energy is transmitted from the power generation sub-unit to the inverter to be processed, and constructing the prediction load of the photovoltaic power generation system which can be input into the user side power grid in the first prediction time zone. The technical problem that the load prediction precision is insufficient due to the lack of processing on equipment aging, connection relation and multi-source data fusion analysis in the prior art is solved, and the technical effect of improving the adaptability and accuracy of the prediction result to the user side power grid is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and particularly relates to a method for predicting the load of a photovoltaic power generation system based on big data. Background Art

[0002] The existing methods for predicting the load of a photovoltaic power generation system mainly rely on static correlation modeling between meteorological parameters and power generation. Usually, single-dimensional meteorological data is used as input, ignoring the performance degradation of photovoltaic modules during long-term operation, as well as the influence of the connection structure between each generating sub-unit within the system on the energy transmission efficiency. In addition, traditional methods lack in-depth utilization of historical operation data and fail to construct a multi-dimensional data model reflecting equipment models, service time, and meteorological cycle changes, making it difficult to adapt to the complex and changeable operation environment in practical applications. As a result, the accuracy of the prediction results is not high, and it cannot meet the requirements of refined scheduling and power grid load management. Summary of the Invention

[0003] This application provides a method for predicting the load of a photovoltaic power generation system based on big data, which is used to solve the technical problem that the existing technology lacks the processing of equipment aging, connection relationship, and multi-source data fusion analysis, resulting in insufficient load prediction accuracy.

[0004] In view of the above problems, this application provides a method for predicting the load of a photovoltaic power generation system based on big data.

[0005] This application provides a method for predicting the load of a photovoltaic power generation system based on big data, and the method includes: Determine multiple generating sub-units within the photovoltaic power generation system in a preset power supply area and the connection relationship network of the multiple generating sub-units; collect the photovoltaic panel models and service times of the multiple generating sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and construct multiple power generation databases through historical data retrieval; collect the first meteorological prediction information in the first prediction time zone of the preset power supply area and input it into the multiple power generation databases for analysis to generate initial power generation data of multiple sub-units; based on the connection relationship network, analyze the fusion loss after the electric energy is processed by the generating sub-units and transmitted to the inverter for the initial power generation data of the multiple sub-units, and construct the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone.

[0006] In a possible implementation manner, determining a plurality of power generation sub-units in a photovoltaic power generation system within a preset power supply area, and a connection relationship network of the plurality of power generation sub-units includes: collecting each inverter processing node in the photovoltaic power generation system; using each inverter processing node as a partition node to partition the plurality of power generation sub-units to construct each power generation sub-unit array; constructing the connection relationship network with each power generation sub-unit array and each inverter processing node.

[0007] In a possible implementation manner, each power generation sub-unit array includes a plurality of power generation sub-units connected in series or in parallel, and the initial electric energy of the plurality of power generation sub-units flows into the same inverter processing node for processing.

[0008] In a possible implementation manner, collecting the photovoltaic panel models and service times of the plurality of power generation sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and constructing a plurality of power generation databases through historical data retrieval, including: constructing a first data constraint with the photovoltaic panel models and service times of the plurality of power generation sub-units, and retrieving a plurality of historical power generation data sets based on big data technology, where the historical power generation data includes historical meteorological information and corresponding historical power generation data at any historical moment; constructing a second data constraint with the photovoltaic panel models of the plurality of power generation sub-units and the meteorological cycle characteristics of the preset power supply area, and retrieving a plurality of historical power generation data time series that change in sequence with the service time; performing power generation quality degradation analysis on the plurality of power generation sub-units with the plurality of historical power generation data time series to construct a plurality of power generation quality degradation curves; establishing a mapping connection between the plurality of power generation quality degradation curves and the plurality of historical power generation data sets to obtain the plurality of power generation databases.

[0009] In a possible implementation manner, performing power generation quality degradation analysis on the plurality of power generation sub-units with the plurality of historical power generation data time series to construct a plurality of power generation quality degradation curves, including: decomposing the plurality of historical power generation data time series according to the types of power generation parameters to generate a plurality of historical power generation parameter time series combinations; based on the plurality of historical power generation parameter time series combinations, fitting the relationship between the degradation degree of power generation parameters and the service duration to generate the plurality of power generation quality degradation curves.

[0010] In a possible implementation, based on the connection relationship network, the fusion loss of the electric energy after the initial power generation data of the multiple sub-units is transmitted to the inverter for processing is analyzed, and the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone is constructed, including: based on the connection relationship network, the superposition and electric energy loss analysis of the initial power generation data of the multiple sub-units in the series and parallel states of each sub-unit power generation array are performed to generate the respective first output power generation data corresponding to each sub-unit power generation array; at each inverter processing node, based on the transmission parameter requirements of the user-side power grid, electric energy processing simulation is performed on the respective first output power generation data to generate respective second output power generation data; electric energy loss analysis is performed on the access circuits from each inverter processing node to the user-side power grid, and loss compensation is performed on the respective second output power generation data to generate the predicted load.

[0011] In a possible implementation, based on the connection relationship network, the superposition and electric energy loss analysis of the initial power generation data of the multiple sub-units in the series and parallel states of each sub-unit power generation array are performed to generate the respective first output power generation data corresponding to each sub-unit power generation array, including: constructing a series-parallel electric energy parameter superposition mechanism; collecting the connection circuit structure information of each sub-unit power generation array; retrieving historical electric energy transmission loss data based on the respective connection circuit structure information, analyzing the loss relationship between the electric energy loss and the circuit blocks with the historical electric energy transmission loss data, and constructing an electric energy loss compensation mechanism; combining the series-parallel electric energy parameter superposition mechanism and the electric energy loss compensation mechanism to perform superposition and loss analysis on the initial power generation data of the multiple sub-units to generate the respective first output power generation data.

[0012] In a possible implementation, at each inverter processing node, based on the transmission parameter requirements of the user-side power grid, electric energy processing simulation is performed on the respective first output power generation data to generate respective second output power generation data, including: performing twin simulation on each inverter processing node to establish respective twin inverters, where the twin simulation includes at least twin simulation of the DC input end, current control circuit, filter circuit, and AC output end; using the transmission parameter requirements as the simulation output constraint, and respectively using the respective first output power generation data as the simulation input through the respective twin inverters to perform simulation operation to generate the respective second output power generation data.

[0013] In a possible implementation, the transmission parameter requirements at least include the input voltage, frequency, power factor, and harmonic constraints of the user-side power grid.

[0014] In a possible implementation manner, after constructing the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone, the method further includes: reading the power supply planning information of the preset power supply area in the first prediction time zone; determining whether the predicted load exceeds the power supply planning information, if so, obtaining the excess load and storing it through an energy storage device, and if not, determining the power supply amount of the main distribution network connected to the user-side power grid based on the excess load.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application determines multiple power generation sub-units in a photovoltaic power generation system within a preset power supply area and the connection relationship network of the multiple power generation sub-units; collects the photovoltaic panel models and service times of the multiple power generation sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and constructs multiple power generation databases through historical data retrieval; collects the first meteorological prediction information of the preset power supply area in the first prediction time zone and inputs it into the multiple power generation databases for analysis to generate initial power generation data of multiple sub-units; based on the connection relationship network, analyzes the fusion loss of electric energy after being processed by the power generation sub-units and transmitted to the inverter for the initial power generation data analysis of the multiple sub-units, and constructs the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone. The present invention solves the technical problem that the prior art lacks the processing of equipment aging, connection relationship and multi-source data fusion analysis, resulting in insufficient load prediction accuracy. By constructing a power generation database containing multi-dimensional constraint conditions such as photovoltaic panel models, service times, and meteorological cycles, and combining the connection relationship network to analyze the electric energy fusion loss, the technical effect of improving the adaptability and accuracy of the prediction result to the user-side power grid is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings without creative efforts based on these drawings.

[0017] Figure 1 It is a schematic flowchart of a method for predicting the load of a photovoltaic power generation system based on big data provided by an embodiment of this application; Figure 2 It is a schematic flowchart of constructing multiple power generation databases in a method for predicting the load of a photovoltaic power generation system based on big data provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The present application provides a method for predicting the load of a photovoltaic power generation system based on big data, which is used to solve the technical problem that the existing technology lacks the processing of equipment aging, connection relationships, and multi-source data fusion analysis, resulting in insufficient accuracy of load prediction. By constructing a power generation database containing multi-dimensional constraint conditions such as photovoltaic panel models, service time, and meteorological cycles, and combining connection relationship network analysis to analyze the power fusion loss, the technical effect of improving the adaptability and accuracy of the prediction result to the user-side power grid is achieved.

[0019] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0020] It should be noted that any variations of the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0021] Example, as Figure 1 shown, the present application provides a method for predicting the load of a photovoltaic power generation system based on big data, and the method includes: Step S100: Determine multiple power generation sub-units in the photovoltaic power generation system within a preset power supply area, and the connection relationship network of the multiple power generation sub-units.

[0022] In the embodiments of the present application, first, the photovoltaic power generation system within a preset power supply area is structurally analyzed to determine multiple power generation sub-units. A power generation sub-unit refers to the basic power generation module that constitutes a photovoltaic power generation array, usually composed of several photovoltaic panels, and has independent power generation capabilities.

[0023] Next, a connection relationship network between the power generation sub-units is constructed. Specifically, first, each inverter processing node in the photovoltaic power generation system is collected; then, taking the inverter processing node as a partition node, the multiple power generation sub-units connected to it are divided into independent power generation sub-unit arrays, and each array represents a group of power generation modules with the same electrical attribution; finally, based on each power generation sub-unit array and the inverter processing node, a connection relationship network between the multiple power generation sub-units is constructed.

[0024] Further, in the method provided by the application embodiment, determining multiple power generation sub-units in the photovoltaic power generation system within the preset power supply area and the connection relationship network of the multiple power generation sub-units further includes: Collecting each inverter processing node in the photovoltaic power generation system; using each inverter processing node as a partition node to partition the multiple power generation sub-units and construct each power generation sub-unit array; constructing the connection relationship network with each power generation sub-unit array and each inverter processing node.

[0025] Further, the method provided by the application embodiment further includes: Each power generation sub-unit array includes a plurality of power generation sub-units connected in series or in parallel, and the initial electric energy of the plurality of power generation sub-units flows into the same inverter processing node for processing.

[0026] In the embodiment of the present application, first, multiple inverter processing nodes in the photovoltaic power generation system within the preset power supply area are determined. This process uses the method of system structure recognition and equipment information parsing, combines the electrical wiring diagram, deployment drawing, and operation configuration data in the SCADA platform to identify the positions of inverters, connection channels, and their electrical relationships with the power generation sub-units in the photovoltaic system, and determines each inverter processing node.

[0027] After identifying the inverter processing nodes, further use each inverter processing node as a partition node to partition the multiple power generation sub-units and construct each power generation sub-unit array. Specifically, by reading the wiring numbers, busbar connection methods, cable routing relationships, etc., identify multiple power generation sub-units connected to the same inverter and classify them into a power generation sub-unit array. Each array is composed of a plurality of power generation sub-units connected in series or in parallel. Among them, series connection can increase the output voltage, and parallel connection can enhance the output current capacity. Through the structure recognition method and actual operation parameter analysis, further clarify the internal connection method of multiple power generation sub-unit arrays and confirm that the initial electric energy of a plurality of power generation sub-units flows into the same inverter processing node for centralized processing.

[0028] After completing the above structure recognition and array division, finally construct the connection relationship network with each power generation sub-unit array and each inverter processing node, that is, through the graph structure modeling method, use the power generation sub-unit array and the inverter processing node as network nodes, and use their physical connection relationship and electrical connection path as the modeling edges to establish a complete structure model of the electric energy in the photovoltaic power generation system starting from the power generation sub-unit, converging through the array to the inverter processing node. What is finally obtained is the connection relationship network of multiple power generation sub-units.

[0029] Step S200: Collect the photovoltaic panel models and service times of the multiple power generation units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and construct multiple power generation databases through historical data retrieval.

[0030] In the embodiment of the present application, first, by referring to the equipment configuration files and installation records of the power station, clarify the photovoltaic panel types corresponding to each power generation unit and the installation and online time, and then calculate the actual service life of each photovoltaic panel. Subsequently, a periodic meteorological analysis method is used to collect the meteorological cycle characteristics of the preset power supply area. By calling the historical meteorological database (such as the national meteorological station or regional environmental monitoring station), key meteorological variables such as light intensity, ambient temperature, and wind speed in a long period are obtained, and a cycle modeling tool is used to perform trend extraction and periodic pattern analysis on the data. This process uses methods such as Fourier transform to identify typical cycle structures such as daily, weekly, and annual to characterize the environmental cycle background in which the photovoltaic system operates, so as to obtain the meteorological cycle characteristics of the preset power supply area.

[0031] After completing the multi-source information collection, first, construct the first data constraint based on the photovoltaic panel models and service times of the multiple power generation units, and use the big data retrieval method to screen out data records that meet the same component model and close service time from the system historical database. This data set contains meteorological information and corresponding power generation data at any historical time point. Then, construct the second data constraint based on the photovoltaic panel model and meteorological cycle characteristics, and screen out data record sequences that cover different meteorological stages and change with service time to form a historical power generation data time series with a time dimension.

[0032] After obtaining the historical time series data, use the power generation degradation analysis method to model the change trend of the power generation capacity of each power generation unit with service time based on multiple historical power generation data time series. Construct multiple power generation quality degradation curves through data fitting methods. Finally, map and connect the multiple power generation quality degradation curves with the multiple historical power generation data sets extracted in the first stage, that is, associate the time series degradation model with the actual power generation sample data, establish the relationship between the complete component characteristics - degradation model - environmental variables - power generation performance, and construct multiple power generation databases.

[0033] Further, as Figure 2 shown, in the method provided by the embodiment of the application, collecting the photovoltaic panel models and service times of the multiple power generation units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and constructing multiple power generation databases through historical data retrieval further includes: Construct a first data constraint based on the photovoltaic panel models and service times of the multiple power generation units. Based on big data technology, retrieve multiple historical power generation data sets, where the historical power generation data includes historical meteorological information and corresponding historical power generation data at any historical moment; construct a second data constraint based on the photovoltaic panel models of the multiple power generation units and the meteorological cycle characteristics of the preset power supply area, and retrieve multiple historical power generation data time series that change in sequence with service time; perform power generation quality degradation analysis on the multiple power generation units using the multiple historical power generation data time series, and construct multiple power generation quality degradation curves; establish a mapping connection between the multiple power generation quality degradation curves and the multiple historical power generation data sets to obtain the multiple power generation databases.

[0034] In the embodiment of the present application, first, an attribute screening method is adopted to construct a first data constraint based on the photovoltaic panel models and service times of multiple power generation units. This method is based on the component information table and operation record table in the big data platform. By setting the photovoltaic panel model (such as PERC) and service life interval as screening conditions, data entries that match the characteristics of the current power generation unit are screened out from the historical database. Through this operation, multiple historical power generation data sets are retrieved, and each data set contains the power generation output data of the photovoltaic module and the corresponding historical meteorological information at a certain historical moment, such as the light intensity, temperature, wind speed, etc. at that moment.

[0035] Subsequently, a meteorological cycle matching method is adopted to construct a second data constraint based on the photovoltaic panel models of multiple power generation units and the meteorological cycle characteristics of the preset power supply area. This method is based on time series analysis technology. Using STL decomposition (seasonal-trend-residual decomposition) to perform periodic modeling on the historical meteorological data of the target area, the stable daily cycle and annual cycle components are extracted, and combined with the photovoltaic panel model to limit the retrieval conditions, so that the data is not only consistent at the component level but also has periodic similarity in the meteorological pattern. In this way, multiple historical power generation data time series that change in sequence with service time are retrieved.

[0036] Next, a multi-time series degradation analysis method is adopted to perform power generation quality degradation analysis on the multiple power generation units using the multiple historical power generation data time series. This analysis first decomposes the multiple historical power generation data time series according to the type of power generation parameters, splits the power generation output according to key parameters (such as output power, voltage, current, temperature coefficient), and forms independent parameter change time series combinations respectively. Then, based on the multiple historical power generation parameter time series combinations, the relationship between the degradation degree of the power generation parameters and the service duration is fitted. Using common fitting methods such as linear regression, the degradation trend models of each parameter are obtained, and finally multiple power generation quality degradation curves are generated.

[0037] Finally, the data model mapping method is adopted to establish a mapping connection between multiple power generation quality degradation curves and multiple historical power generation data sets. Specifically, in each historical power generation data record, the degradation curve function corresponding to the component model and service time is embedded to form a data structure with triple information binding of time dimension, environmental background and degradation model.

[0038] Further, in the method provided by the application embodiment, the power generation quality degradation analysis is performed on the multiple power generation sub-units according to the time series of the multiple historical power generation data, and multiple power generation quality degradation curves are constructed, and it further includes: Decompose the time series of the multiple historical power generation data according to the power generation parameter type to generate multiple historical power generation parameter time series combinations; based on the multiple historical power generation parameter time series combinations, fit the relationship between the degradation degree of the power generation parameter and the service duration to generate the multiple power generation quality degradation curves.

[0039] In the embodiment of the present application, first, the parameter classification extraction method is adopted to decompose the time series of multiple historical power generation data according to the power generation parameter type. This method extracts key power generation parameters such as output power, voltage, current, temperature, etc. in the historical time series data one by one by parsing the parameter fields in the historical data records, and reorganizes them into independent time series data according to the time stamps. Each parameter forms a set of continuous numerical change sequences, so as to obtain multiple historical power generation parameter time series combinations.

[0040] Subsequently, the least squares regression fitting method is adopted to fit the relationship between the degradation degree of the power generation parameter and the service duration based on the multiple historical power generation parameter time series combinations. The service duration is obtained by subtracting the installation time of the component from the time of each data record and used as the independent variable input; the time series values corresponding to each parameter are used as the dependent variable input to establish a functional relationship between the parameter value and the service time. The least squares method is used to fit the data points, a suitable mathematical model is selected for expression (such as a linear model, an exponential model, etc.), and the fitting effect is evaluated through residual tests to obtain the degradation trend function of each power generation parameter, which can accurately depict the change trend of the parameter during the component aging process.

[0041] Finally, the trend curve construction method is adopted to generate multiple power generation quality degradation curves according to the parameter degradation functions obtained by fitting. Specifically, according to the mathematical expression of the degradation function, the parameter values corresponding to each time point are calculated within the set service time interval (such as 0 to 25 years), and a curve is plotted with time as the horizontal axis and the fitted value as the vertical axis. Each curve clearly shows the trajectory of a certain power generation parameter gradually degrading with the service duration, and has good visualization and quantification characteristics. Through this process, multiple power generation quality degradation curves are finally generated.

[0042] Step S300: Collect the first meteorological prediction information of the preset power supply area in the first prediction time zone, and input it into the multiple power generation databases for analysis to generate initial power generation data for multiple subunits.

[0043] In the embodiment of the present application, first, a numerical weather prediction interface calling method is adopted to collect the first meteorological prediction information of the preset power supply area in the first prediction time zone. By accessing the meteorological data service platform (such as ECMWF, GFS), the API interface is called to obtain the hourly meteorological data covering the target area. The data content includes key meteorological variables such as light intensity, ambient temperature, humidity, and wind speed, and the first meteorological prediction information of the first prediction time zone is obtained.

[0044] Next, a similar meteorological retrieval method is adopted to input the first meteorological prediction information of the first prediction time zone into the power generation database for matching analysis. Using indicators such as Euclidean distance, the meteorological parameters of the prediction time zone are compared with the typical meteorological scenarios recorded in the historical database, and the historical data set closest to the prediction conditions is retrieved. Subsequently, a degradation model calibration method is adopted to correct the retrieved historical power generation data combined with the corresponding power generation quality degradation curve. The degradation curve is fitted according to the model and service time of the photovoltaic module, reflecting the performance degradation trend of the module due to aging during actual operation. By substituting the historical power generation output data into the corresponding degradation function, the actual power generation capacity of the module at the current moment under the predicted meteorological conditions is calibrated, including key parameters such as power, voltage, and current.

[0045] Finally, a structured output generation method is adopted to generate initial power generation data for multiple subunits according to the calibrated results. The initial power generation data for subunits includes electrical energy parameters such as output power, voltage, and current.

[0046] Step S400: Based on the connection relationship network, analyze the fusion loss of the electrical energy after the initial power generation data of the multiple subunits is processed by the inverter, and construct the predicted load that the photovoltaic power generation system can input to the user-side power grid in the first prediction time zone.

[0047] In the embodiment of the present application, based on the connection relationship network, the initial power generation data of multiple subunits is fused and analyzed. First, the series and parallel connection methods of each power generation subunit array are identified, and the loss calculation during the electrical energy superposition and transmission of its initial power generation data is performed to obtain the first output power generation data corresponding to each power generation subunit array; then, at the inverter processing node, according to the power transmission parameter requirements of the user-side power grid, the electrical energy processing simulation of the first output power generation data is performed to obtain each second output power generation data; finally, combined with the access circuit from the inverter processing node to the user-side power grid, the electrical energy loss compensation is performed on the second output power generation data, and the predicted load that the photovoltaic power generation system can input to the user-side power grid in the first prediction time zone is constructed.

[0048] Further, in the method provided by the application embodiment, based on the connection relationship network, the fusion loss of the electric energy after the initial power generation data of the multiple sub-units is transmitted into the inverter for processing is analyzed, and the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone is constructed, further including: Based on the connection relationship network, the superposition and power loss analysis of the initial power generation data of multiple sub-units in the series and parallel states of each sub-unit power generation array are performed to generate each first output power generation data corresponding to each sub-unit power generation array; at each inverter processing node, based on the transmission parameter requirements of the user-side power grid, power processing simulations are respectively performed on each first output power generation data to generate each second output power generation data; for the access circuits from each inverter processing node to the user-side power grid, power loss analysis is performed, and loss compensation is performed on each second output power generation data to generate the predicted load.

[0049] In the embodiment of the present application, first, a series-parallel electrical energy parameter superposition method is adopted to analyze the structure of each sub-unit power generation array based on the connection relationship network. Specifically, the series or parallel state of each sub-unit in the array is identified, and based on the principles of voltage superposition (series) and current superposition (parallel), parameters such as power, voltage, and current in the initial power generation data of multiple sub-units are combined. To further correct the energy attenuation during transmission, a circuit structure extraction and matching method is adopted to collect connection circuit information such as the cable path, resistance structure, and wiring method of each array, and based on this information, historical power transmission loss data is retrieved to construct a power loss compensation mechanism to correct the superposed parameters and generate each first output power generation data corresponding to each sub-unit power generation array.

[0050] Subsequently, an inverter twin simulation method is adopted to simulate the actual power conversion process at each inverter processing node. Specifically, a twin inverter model including a DC input, current control, filtering, and AC output structure is established to simulate its response behavior to the received electric energy. Using the first output power generation data as the simulation input and combining the transmission parameter requirements of the user-side power grid (such as voltage level, power factor, frequency range, etc.) as the simulation output constraint, the simulation process is run to calculate the actual output capacity of the inverter under the current conditions, and each second output power generation data corresponding to each inverter processing node is obtained.

[0051] Afterwards, the access path exact impedance modeling method is used to analyze the losses of the access circuit from the inverter output to the user grid access point. By constructing the physical equivalent circuit model of the inverter output path, this method clarifies the equivalent series resistance and inductance of the cable, and combines the physical length, wiring method, and transformer parameters of each path to establish equivalent impedance models for single-phase and three-phase circuits respectively. Combining the current amplitude information in the second output power generation data, the active power loss calculation method is used to calculate the node-level losses of each path, and the power attenuation and voltage drop values of the entire access path are accumulated. Through deterministic physical calculations, this step fully reflects the electrical performance attenuation process from the inverter to the grid side.

[0052] Finally, the power fusion method based on path correction is used to compensate and integrate the output results. The path loss data obtained from the previous step is used to correct the power and voltage parameters in the second output power generation data. By directly subtracting the loss value from the AC output power, the net available power of each inverter node before grid connection is obtained. Subsequently, according to the multi-path power fusion rule of the system access point, the corrected output values of multiple inverters are summed up, and combined with the grid connection point power limit, phase voltage balance condition, and distribution network capacity constraint, the final predicted load that the photovoltaic power generation system can supply to the user-side grid in the first prediction time zone is generated.

[0053] Furthermore, in the method provided by the application embodiment, for the superposition of the initial power generation data of multiple sub-units in the series and parallel states of each sub-array of power generation units and the analysis of power loss, to generate each first output power generation data corresponding to each sub-array of power generation units, it further includes: Construct a series-parallel electrical energy parameter superposition mechanism; collect the connection circuit structure information of each sub-array of power generation units; retrieve historical electrical energy transmission loss data based on the connection circuit structure information of each sub-array, analyze the loss relationship between power loss and circuit blocks with the historical electrical energy transmission loss data, and construct a power loss compensation mechanism; combine the series-parallel electrical energy parameter superposition mechanism and the power loss compensation mechanism to perform superposition and loss analysis on the initial power generation data of multiple sub-units, and generate each first output power generation data.

[0054] In the embodiment of the present application, first, a series-parallel electrical energy parameter superposition mechanism is constructed. Based on the circuit characteristics of the existing photovoltaic system, this mechanism identifies the connection structure between each sub-array of power generation units, determines whether it is in series, parallel, or series-parallel state, and integrates the parameter of the initial power generation data of the sub-units according to electrical principles. Specifically, in the series structure, the current is the same and the voltage is accumulated; in the parallel structure, the voltage is the same and the current is accumulated. Through this mechanism, the voltage, current, and power parameters in the initial power generation data of multiple sub-units are logically superimposed according to the array topology structure to form the overall electrical energy output parameters of each sub-array of power generation units in an ideal state.

[0055] Subsequently, the connection circuit structure information of each power generation sub-unit array is collected by referring to the electrical installation drawings and the system wiring list, including specific physical parameters such as the wiring path between the array and the inverter, the cable material (such as copper, aluminum), the cable cross-sectional area, the resistance of a single section of the cable, the total length, the connection method (such as overhead, buried), and the number of joints. Through this process, the connection circuit structure information of each is obtained.

[0056] Then, based on the collected connection circuit structure information, historical power transmission loss data is further retrieved. By accessing the operation logs in the historical operation and maintenance platform of the photovoltaic power station, historical records with cable path parameters similar to the current structure are filtered out, and the current, voltage, actual power, and theoretical output data contained therein are extracted. Combining the difference between the ideal output and the measured output, the line loss level under specific operating conditions is compared, and the loss relationship between the power loss and the circuit block is analyzed accordingly. This analysis clarifies the corresponding relationship between different wiring structures, current intensities, and energy attenuation, providing quantitative support for the subsequent correction process.

[0057] Subsequently, based on the above analysis results, a power loss compensation mechanism is constructed. This mechanism is based on the commonly used physical calculation methods in engineering. By multiplying the array output current by the path resistance, the active power loss of the entire cable path is calculated; at the same time, based on the voltage drop formula, the voltage drops in the path are accumulated segment by segment to form a set of power and voltage correction rules for each path. This compensation mechanism real-time feedbacks the energy attenuation caused by the path to the array output parameters, ensuring that the subsequent calculation results are closer to the physical real state.

[0058] Finally, by combining the series-parallel power parameter superposition mechanism and the power loss compensation mechanism, the initial power generation data of each sub-unit is uniformly processed and loss analyzed. Through the integration of steps such as parameter superposition, structure identification, and path compensation, the comprehensive ability assessment of the power generation sub-unit array under the actual electrical configuration is completed, and finally the first output power generation data of each is generated.

[0059] Furthermore, in the method provided by the application embodiment, at each inverter processing node, based on the power transmission parameter requirements of the user-side power grid, power processing simulations are respectively performed on the first output power generation data of each to generate second output power generation data of each, and it further includes: Performing twin simulations on each inverter processing node to establish twin inverters, where the twin simulations at least include twin simulations of the DC input end, the current control circuit, the filter circuit, and the AC output end; using the power transmission parameter requirements as the simulation output constraints, and respectively using the first output power generation data of each as the simulation input through each twin inverter to perform simulation operations to generate the second output power generation data of each.

[0060] Further, the method provided by the application embodiment further includes: The power transmission parameter requirements at least include the input voltage, frequency, power factor, and harmonic constraints of the user-side power grid.

[0061] In the embodiment of the present application, in order to restore the power conversion process of the inverter in the photovoltaic power generation system, twin simulation is performed on each inverter processing node, that is, a simulation entity with the same functional structure as the actual inverter is established through digital modeling. This simulation uses a structural equivalent restoration method to ensure that the behavior of each inverter processing node in the system can be completely mapped in the simulation environment, so as to support the accurate prediction of the subsequent power output capacity. During the simulation process, each twin inverter is established, and each twin inverter includes key structural modules to achieve the electrical consistency restoration of the input-output behavior. This structure is divided into multiple functional sub-modules, covering the whole process from input acquisition, control calculation, filtering and shaping to output feedback, ensuring that the simulation data has response characteristics highly consistent with the operating state of the real inverter. Specifically, the twin simulation at least includes the twin simulation of the DC input terminal, the current control circuit, the filter circuit, and the AC output terminal. Among them, the DC input terminal is used to receive the first output power generation data on the photovoltaic array side, including parameters such as voltage, current, and power; the current control circuit uses a double-loop PI control method for voltage-current decoupling regulation to achieve stable tracking of the grid synchronization command; the filter circuit is constructed based on the LCL topology and is used to suppress the high-frequency harmonics generated during the inversion process and improve the output power quality; the AC output terminal simulates the electrical interface of the inverter grid connection port and is used to output the three-phase AC power after modulation and shaping to meet the grid connection standard.

[0062] After establishing the twin inverter structure, using the power transmission parameter requirements as the simulation output constraint, set the operating boundary conditions of each simulation inverter. The power transmission parameter requirements include the input voltage, frequency, power factor, and harmonic constraints of the user-side power grid, where the voltage and frequency are set to the national grid standard values, the power factor is used to constrain the reactive power characteristics of the inverter output, and the harmonic constraints ensure that the output power quality does not affect the grid stability. These parameters are implanted into the control module to dynamically adjust the PWM modulation strategy and output waveform characteristics, so that the simulation process strictly follows the grid connection rules.

[0063] Subsequently, through each twin inverter, using each first output power generation data as the simulation input, start the dynamic simulation process. After each inverter simulation unit receives the output power, current, and voltage parameters of the corresponding array, it starts the internal control calculation and power conversion process, and responds step by step from input to output. During the simulation operation, the controller dynamically adjusts the output amplitude and phase, the power device simulates the actual switching process, and the filter circuit follows synchronously, gradually generating the AC signal at the output terminal.

[0064] Through the above process, various second output power generation data are generated, that is, index data such as the output AC power, voltage, current, etc. that each inverter processing node can output under the set constraints and input operating conditions.

[0065] Further, in the method provided by the application embodiment, after constructing the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone, it further includes: Reading the power supply planning information of the preset power supply area in the first prediction time zone; judging whether the predicted load exceeds the power supply planning information, if so, obtaining the excess load and storing it through an energy storage device, if not, determining the power supply amount of the main distribution network connected to the user-side power grid based on the excess load.

[0066] In the embodiment of the present application, first, the time period supply and demand scheduling analysis method is used to read the power supply planning information of the preset power supply area in the first prediction time zone. This information comes from the regional energy management system (EMS) or the distribution automation platform, and its content includes the maximum power upper limit allowed to be connected in the target prediction time period of this area, the operating status of the energy storage device, the main distribution network power supply strategy, and the load level boundary, etc. Access the power grid scheduling system through the standard communication protocol, retrieve and analyze the power supply planning table, and extract the access capacity and load control strategy of the current time period as the basis for subsequent judgment and control.

[0067] Subsequently, the power boundary comparison and judgment method is used to judge whether the predicted load calculated by the photovoltaic system exceeds the allowable range in the power supply planning information. This method performs boundary condition judgment by comparing the predicted load obtained in the previous stage (that is, the available power supply capacity of the photovoltaic system under specific environmental and conversion conditions) with the access power upper limit in the power supply planning. If the predicted load is greater than the upper limit, it is determined as an excess load; otherwise, it is considered that the current power supply capacity can be completely absorbed by the power grid. This step provides the control branch logic for subsequent energy storage call or main network power supply replenishment strategy selection.

[0068] In the case where it is determined that the predicted load exceeds the planning upper limit, the dynamic energy absorption and regulation method is used to guide the excess electric energy to the local energy storage device for storage. This method is based on the bidirectional converter control mechanism, converts the excess power into the DC input required by the energy storage unit (such as a lithium battery or a super capacitor), and realizes the dynamic control of the energy storage process through setting the power upper limit, SOC state threshold, and charging curve control method. Before performing energy storage, verify the real-time SOC (state of charge) and available capacity of the energy storage device to ensure that the redundant power can be completely received. Finally, complete the temporary storage operation of the electric energy to avoid the waste of electricity caused by the photovoltaic power generation capacity exceeding the grid connection capacity.

[0069] When the judgment result is that the predicted load does not exceed the power supply planning limit, the main and distribution network power sharing method is used to determine that the effective part of the predicted load is jointly powered by the connection channels between the user-side power grid and the main and distribution network. This method performs power distribution calculations based on the electrical parameters of the user access point (such as grid connection switch status, voltage level, distribution transformer load rate, etc.) and combines the load margin information on the main grid side. Specific strategies can adopt the voltage adjustment method or the load transfer strategy to clarify the part of the current photovoltaic output that can be supported or accepted by the main grid and use it as the grid-connected power for output control to ensure the safe operation of the system.

[0070] In the embodiments of the present application, in summary, the embodiments of the present application at least have the following technical effects: The present application determines multiple power generation sub-units in the photovoltaic power generation system within the preset power supply area and the connection relationship network of the multiple power generation sub-units; collects the photovoltaic panel models and service times of the multiple power generation sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and constructs multiple power generation databases through historical data retrieval; collects the first meteorological prediction information in the first prediction time zone of the preset power supply area and inputs it into the multiple power generation databases for analysis to generate initial power generation data of multiple sub-units; based on the connection relationship network, analyzes the fusion loss of the electric energy after being processed by the power generation sub-units and transmitted to the inverter for the initial power generation data analysis of the multiple sub-units, and constructs the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone. The present invention solves the technical problem that the prior art lacks the processing of equipment aging, connection relationship and multi-source data fusion analysis, resulting in insufficient load prediction accuracy. By constructing a power generation database containing multi-dimensional constraint conditions such as photovoltaic panel models, service times, and meteorological cycles, and combining the connection relationship network to analyze the electric energy fusion loss, the technical effect of improving the adaptability and accuracy of the prediction result to the user-side power grid is achieved.

[0071] It should be noted that the above sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above describes specific embodiments of this specification. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

[0072] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0073] This specification and the drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A load forecasting method for a photovoltaic power generation system based on big data, characterized in that, Including: Determine multiple power generation sub-units within a photovoltaic power generation system in a preset power supply area, and the connection relationship network of the multiple power generation sub-units; Collect the photovoltaic panel models and service times of the multiple power generation sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and construct multiple power generation databases through historical data retrieval; Collect the first meteorological prediction information of the preset power supply area in the first prediction time zone, and input it into the multiple power generation databases for analysis to generate initial power generation data of multiple sub-units; Based on the connection relationship network, analyze the fusion loss of the electric energy after being processed by the inverter when the initial power generation data of the multiple sub-units is transmitted into the inverter, and construct the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone.

2. The load prediction method for a photovoltaic power generation system based on big data according to claim 1, wherein, Determine multiple power generation sub-units within a photovoltaic power generation system in a preset power supply area, and the connection relationship network of the multiple power generation sub-units, including: Collect each inverter processing node within the photovoltaic power generation system; Use each inverter processing node as a partition node to partition the multiple power generation sub-units and construct arrays of each power generation sub-unit; Construct the connection relationship network with the arrays of each power generation sub-unit and each inverter processing node.

3. The load prediction method for a photovoltaic power generation system based on big data according to claim 2, characterized in that Each array of power generation sub-units includes a number of power generation sub-units connected in series or in parallel, and the initial electric energy of the number of power generation sub-units flows into the same inverter processing node for processing.

4. A method for predicting the load of a photovoltaic power generation system based on big data according to claim 1, characterized in that, Collect the photovoltaic panel models and service times of the multiple power generation sub-units and the meteorological cycle characteristics of the preset power supply area to construct constraint conditions, and construct multiple power generation databases through historical data retrieval, including: Construct the first data constraint with the photovoltaic panel models and service times of the multiple power generation sub-units, and retrieve multiple historical power generation data sets based on big data technology, where the historical power generation data includes historical meteorological information and corresponding historical power generation data at any historical moment; Construct the second data constraint with the photovoltaic panel models of the multiple power generation sub-units and the meteorological cycle characteristics of the preset power supply area, and retrieve multiple historical power generation data time series that change in sequence with the service time; Conduct power generation quality degradation analysis on the multiple power generation sub-units with the multiple historical power generation data time series, and construct multiple power generation quality degradation curves; Establish a mapping connection between the multiple power generation quality degradation curves and the multiple historical power generation data sets to obtain the multiple power generation databases.

5. The load forecasting method for a photovoltaic power generation system based on big data according to claim 4, characterized in that, Conduct power generation quality degradation analysis on the multiple power generation sub-units with the multiple historical power generation data time series, and construct multiple power generation quality degradation curves, including: Decompose the multiple historical power generation data time series according to the type of power generation parameters to generate multiple combinations of historical power generation parameter time series; Based on the multiple combinations of historical power generation parameter time series, fit the relationship between the degradation degree of power generation parameters and the service duration to generate the multiple power generation quality degradation curves.

6. The load prediction method for a photovoltaic power generation system based on big data according to claim 2, wherein, Based on the connection relationship network, analyze the fusion loss of electric energy after the initial power generation data of the multiple sub-units is transmitted into the inverter for processing, and construct the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone, including: Based on the connection relationship network, perform superposition and power loss analysis on the initial power generation data of multiple sub-units in the series and parallel states of each sub-unit power generation array, and generate each first output power generation data corresponding to each sub-unit power generation array; At each inverter processing node, based on the transmission parameter requirements of the user-side power grid, perform power processing simulation on each of the first output power generation data respectively to generate each second output power generation data; Perform power loss analysis on the access circuit from each inverter processing node to the user-side power grid, and perform loss compensation on each of the second output power generation data to generate the predicted load.

7. The load prediction method for a photovoltaic power generation system based on big data according to claim 6, characterized in that Based on the connection relationship network, perform superposition and power loss analysis on the initial power generation data of multiple sub-units in the series and parallel states of each sub-unit power generation array, and generate each first output power generation data corresponding to each sub-unit power generation array, including: Construct a series-parallel power parameter superposition mechanism; Collect the connection circuit structure information of each sub-unit power generation array; Retrieve historical power transmission loss data based on the connection circuit structure information of each, analyze the loss relationship between power loss and circuit blocks with the historical power transmission loss data, and construct a power loss compensation mechanism; Combine the series-parallel power parameter superposition mechanism and the power loss compensation mechanism to perform superposition and loss analysis on the initial power generation data of the multiple sub-units to generate each first output power generation data.

8. The method for predicting the load of a photovoltaic power generation system based on big data according to claim 6, wherein, At each inverter processing node, based on the transmission parameter requirements of the user-side power grid, perform power processing simulation on each of the first output power generation data respectively to generate each second output power generation data, including: Perform twin simulation on each inverter processing node to establish each twin inverter, where the twin simulation includes at least twin simulation of the DC input terminal, current control circuit, filter circuit, and AC output terminal; Using the transmission parameter requirements as the simulation output constraint, perform simulation operation through each twin inverter with each of the first output power generation data as the simulation input to generate each second output power generation data.

9. A method for predicting the load of a photovoltaic power generation system based on big data according to claim 8, characterized in that, The transmission parameter requirements at least include the input voltage, frequency, power factor, and harmonic constraint of the user-side power grid.

10. A method for predicting the load of a photovoltaic power generation system based on big data according to claim 1, characterized in that, After constructing the predicted load that the photovoltaic power generation system can input into the user-side power grid in the first prediction time zone, it further includes: Read the power supply planning information of the preset power supply area in the first prediction time zone; Judge whether the predicted load exceeds the power supply planning information. If so, obtain the excess load and store it through the energy storage device. If not, determine the power supply amount of the main distribution network connected to the user-side power grid based on the excess load.

Images