Photovoltaic power generation access capacity optimization method and device based on big data analysis

Abstract

The invention discloses a photovoltaic power generation access capacity optimization method and device based on big data analysis, and belongs to the technical field of new energy power generation prediction and power distribution network optimization operation, and the method comprises the steps: determining the power-on amount of each time period through analyzing the discrete degree of the historical power generation data of a photovoltaic power station; determining an actual consumption value according to the discrete degree of the historical power consumption data of the access point; screening a reference access point from the photovoltaic access points; the reference access point and the to-be-determined access points are combined to form a plurality of access point groups; calculating an access negative value of each access point group; calculating a total distance value of each access point group; weighting and adding the access negative value and the total distance value of each access point group to obtain an evaluation value of the group; and selecting the access point group with the optimal evaluation value as a target access point group for photovoltaic power generation access capacity optimization configuration. According to the invention, the dynamic optimal configuration of the photovoltaic power generation capacity is realized, the photovoltaic consumption rate is effectively improved, the pressure of the energy storage equipment is reduced, and the method is suitable for a multi-access-point power supply scene.

Description

TECHNICAL FIELD

[0001] The application relates to a photovoltaic power generation access capacity optimization method and device based on big data analysis, and belongs to the technical field of new energy power generation prediction and power distribution network optimal operation. BACKGROUND

[0002] With the large-scale promotion of distributed photovoltaics, a single centralized or large-scale distributed photovoltaic power station (or a park-level photovoltaic cluster) needs to provide power for multiple access points which are geographically dispersed, have different electrical characteristics, and have different load demands, and access the power grid. For example, a large industrial and commercial roof photovoltaic power station needs to provide power for multiple different workshops, office buildings and even adjacent buildings in the factory area and access the grid. Under the premise of meeting the safety and stability constraints of the power grid, how to maximize the economic benefits (such as self-generation and self-use rate, investment return rate) of photovoltaic power generation and improve the power grid access capacity becomes a key and extremely challenging problem.

[0003] Patent No. CN102684220A discloses a ship photovoltaic power generation grid-connected experimental platform, which comprises a solar photovoltaic array, a photovoltaic lightning protection combiner box, a direct-current lightning protection power distribution cabinet, a grid-connected inverter, an alternating-current lightning protection distribution box, a motor, a generator, a simulated ship load, an environment monitor, an oscilloscope and an electric energy quality tester. The direct-current electricity generated by the solar photovoltaic array is inverted into 380V, 50Hz three-phase alternating-current electricity by the grid-connected inverter, and is connected with the ship power grid. The ship power grid is powered by the motor from the power grid, drives the generator to generate electricity and is connected in series with the simulated ship load. The ship solar photovoltaic power generation system can simulate the grid-connected operation performance of the ship photovoltaic power generation system under different external environments, and can measure data such as voltage, current, frequency, power and power factor. The influence of the external environment and the ship load on the operation performance of the ship photovoltaic power generation system in actual use is restored, so as to provide a basis for the structure and optimal design of the ship photovoltaic power generation system.

[0004] In the prior art, each access point is regarded as an independent individual, and its access capacity is optimized separately. Simple load matching or local grid constraints are used, but multiple access points usually share the resources of the upper grid, and the photovoltaic injection of one access point will affect the voltage level and line flow of other access points. Independent optimization cannot capture these mutual influences, which may lead to local optimization but global suboptimization or even infeasibility, sacrificing the great potential brought by system-level coordinated optimization. For example, the load curve and photovoltaic output curve of a typical day (such as summer and winter) are used for optimization calculation, but photovoltaic output and load demand have significant uncertainty and randomness, such as rapid movement of clouds and equipment start-stop. Static models cannot capture this volatility, and the optimization results may fail when facing actual fluctuations. Limited model accuracy: simplified models may not accurately reflect the nonlinear characteristics of the power grid, leading to infeasible or degraded performance of the optimization results in the real power grid.

[0005] In actual use, especially when a photovoltaic power station supplies power to several access points, how to determine better use nodes for access capacity optimization is a difficult problem, based on this, the application provides a photovoltaic power access capacity optimization scheme based on big data analysis. SUMMARY

[0006] In order to solve the above problems, the application provides a photovoltaic power access capacity optimization method and device based on big data analysis, which can improve the photovoltaic consumption rate and reduce the energy storage pressure.

[0007] The technical scheme adopted by the application to solve the technical problems is: In a first aspect, the application provides a photovoltaic power access capacity optimization method based on big data analysis, including the following steps: Step S1, according to the historical power data generated by the photovoltaic power station in several time periods within several days, analyze the data dispersion degree of each time period power, and determine the corresponding upper power of each time period; Step S2, obtain the historical consumption value data of the multiple photovoltaic access points in the multiple time periods, analyze the dispersion degree of the consumption value of each access point in each time period, and determine the actual consumption value of each photovoltaic access point in each time period; Step S3, select the photovoltaic access point whose actual consumption value in each time period is less than the upper power of the corresponding time period and the sum of the difference is the smallest from all photovoltaic access points, and mark it as a reference access point; Step S4, combine the reference access point with the remaining undetermined access points to form multiple access point groups, wherein the sum of the actual consumption values of all photovoltaic access points in each access point group is less than the sum of the upper powers of all time periods of the photovoltaic power station; Step S5, calculate the access negative value of each access point group, the access negative value is the number of time periods in which the sum of the actual consumption values of any time period in the group is greater than the upper power of the corresponding time period; Step S6, calculate the distance total value of each access point group, the distance total value is determined by the real-time distance between each undetermined access point in the group and the photovoltaic power station; Step S7, add the weight of the access negative value and the distance total value of each access point group respectively to obtain the evaluation value of the group; Step S8, select the access point group with the optimal evaluation value as the target access point group for photovoltaic power access capacity optimization configuration.

[0008] As a possible implementation manner of the embodiment, the step S1 includes the following steps: Step S11, divide a day into several time periods with uniform time length, collect historical power data of each time period in each day in the past month by the power metering device, and the interval is not more than 1 hour; Step S12, for any target time period, record the historical power data of the time period as Di, wherein i=1, 2...n, n is the number of days in the past month, calculate the arithmetic mean P of Di, screen out the abnormal data satisfying “|Di-P|> preset abnormal threshold X1”, and calculate the ratio of the number of abnormal data to the total data n to obtain the screening ratio; Step S13, if the screening ratio is less than or equal to a preset proportion threshold B1, take the maximum value of all Di in the time period as the upper power; if the screening ratio is greater than B1, take the maximum value of Di in the last ten days in the time period as the upper power; Step S14, repeat step S12 and step S13 for all divided time periods to obtain the upper power corresponding to each time period and form the upper power basic data.

[0009] As a possible implementation manner of the embodiment, the preset abnormal threshold X1 is 5-15 kWh, the preset proportion threshold B1 is 0.1-0.2, the time length of the time period can be adjusted in the range of 1-4 hours, and the total time length is 24 hours.

[0010] As a possible implementation manner of the embodiment, the step S2 comprises the following steps: Step S21, collect historical consumption value data of each time period in each day in the past month of all access points in the power supply range of the photovoltaic power station, and record the consumption value data of any access point in any time period as Fi, i=1, 2...n, n is the number of days in the past month; Step S22, calculate the arithmetic mean U and the mean square deviation of Fi, obtain the deviation W by the ratio of the mean square deviation to U, and judge whether W exceeds a preset fluctuation threshold X2; Step S23, if W is less than or equal to X2, take the minimum value of Fi as the consumption real value of the access point corresponding to the time period; if W is greater than X2, sequentially remove Fi according to “|Fi-U| from large to small” and recalculate W until W is less than or equal to X2; calculate the deletion ratio of the ratio of the number of removed data to the total data n, and if the deletion ratio is less than or equal to B1, take the minimum value of the remaining Fi as the consumption real value, otherwise take the minimum value of the original Fi as the consumption real value; Step S24, repeat step S22 and step S23 for all access points and all time periods to obtain the consumption real value corresponding to each time period.

[0011] As a possible implementation manner of the embodiment, the preset fluctuation threshold X2 is 0.2-0.4, the power supply range is 3-10 km around the photovoltaic power station, and the sampling frequency of the historical consumption value data collecting device and the power metering device is consistent.

[0012] As a possible implementation manner of the embodiment, the step S3 comprises the following steps. Step S31, calling the power-on amount Sj of each period and the accommodation real value Aj corresponding to any access point, calculating the sum of the difference values C = Σ (Sj-Aj), j = 1, 2...m, m is the total number of periods; Step S32, eliminating the access points with any period Aj≥Sj, and selecting the access point with the minimum C value from the remaining access points as the reference access point.

[0013] As a possible implementation manner of the embodiment, the step S4 comprises the following steps. Step S41: marking the remaining valid access points other than the selected reference access point as pending access points; Step S42: combining the reference access point with one or more pending access points to form multiple access point groups, and the combination of the reference access point and the pending access points needs to meet the condition that the sum of the accommodation real values of all access points in the group in each period ≤ the power-on amount of the corresponding period, and the sum of the total accommodation real values in the group ≤ the sum of the total power-on amounts; Step S43: if a pending access point violates the above condition after being added, it is not included in the group, and finally multiple access point groups meeting the constraint are formed.

[0014] As a possible implementation manner of the embodiment, the access point groups are generated in the order of "the number of pending access points from small to large" during the combination process, and the remaining capacity of "total power-on amount-total accommodation real value" is calculated in real time. When the remaining capacity is insufficient to accommodate the minimum accommodation demand of the next pending access point, the combination operation of the group is stopped.

[0015] As a possible implementation manner of the embodiment, the step S5 comprises the following steps. Step S51, for any access point group generated, calculating the sum Tj of the accommodation real values of the group in each period, Tj = the sum of Aj corresponding to all access points in the group in the period; Step S52, comparing Tj with the power-on amount Sj of the corresponding period, and counting the number of periods that satisfy "Tj>Sj", which is the access negative value of the access point group.

[0016] As a possible implementation manner of the embodiment, the step S6 comprises the following steps. Step S61, measuring the actual distance between all pending access points in each access point group and the photovoltaic power station; Step S62, selecting a calculation method according to the application scenario, taking the sum of the distances of all pending access points in the group as the total distance value in the industrial and commercial scenario, and taking the average distance of all pending access points in the group as the total distance value in the residential community scenario.

[0017] As a possible implementation manner of the embodiment, the step S7 comprises the following steps: The step S71 sets the weight coefficient of the access negative value as 0.58 and the weight coefficient of the distance total value as 0.42. The step S72 calculates the evaluation value of each access point group according to the formula "evaluation value = 0.58*access negative value + 0.42*distance total value".

[0018] As a possible implementation manner of the embodiment, the step S8 comprises the following steps: The step S81 sorts the evaluation values of all the access point groups. The step S82 selects the access point group with the minimum evaluation value as the target access point group; if the evaluation values are the same, the group with the smaller access negative value is selected preferentially; if the access negative values are still the same, the group with the smaller distance total value is selected preferentially. The step S83 takes the target access point group as the configuration basis for the photovoltaic power generation access capacity optimization.

[0019] As a possible implementation manner of the embodiment, if the evaluation value of the target access point group exceeds a preset warning threshold, a warning signal is sent to prompt the checking of the completeness of the historical data and the running state of the acquisition device.

[0020] In a second aspect, the embodiment of the present application provides a photovoltaic power generation access capacity optimization device based on big data analysis, comprising: The power generation amount determination module is configured to analyze the data dispersion degree of the power generation amount of each time period according to the historical power generation amount data of the photovoltaic power station in multiple time periods within a plurality of days, and determine the corresponding power generation amount of each time period. The consumption real value determination module is configured to obtain historical consumption value data of a plurality of photovoltaic access points in multiple time periods, analyze the dispersion degree of the consumption value of each access point in each time period, and determine the consumption real value of each photovoltaic access point in each time period. The reference access point screening module is configured to screen a photovoltaic access point with the smallest sum of the difference between the consumption real value and the power generation amount of each time period from all photovoltaic access points, and mark the photovoltaic access point as a reference access point. The access point group construction module is configured to combine the reference access point with the remaining undetermined access points to form a plurality of access point groups, wherein the sum of the consumption real values of all photovoltaic access points in each access point group is less than the sum of the power generation amounts of all time periods of the photovoltaic power station. The access negative value calculation module is configured to calculate the access negative value of each access point group, wherein the access negative value is the number of time periods in which the sum of the consumption real values of any time period in the group is greater than the power generation amount of the corresponding time period. The total distance calculation module is used to calculate the total distance of each access point group. The total distance is determined by the real-time distance between each undetermined access point in the group and the photovoltaic power station. The evaluation value calculation module is used to assign weights to the negative access value and total distance value of each access point group, and then add them together to obtain the evaluation value of the group. The target access point group selection module is used to select the access point group with the best evaluation value as the target access point group for the optimized configuration of photovoltaic power generation access capacity.

[0021] The beneficial effects of the technical solutions of the embodiments of the present invention are as follows: This invention, utilizing big data analytics and combining the characteristics of photovoltaic (PV) power generation and consumption, provides a scientific, accurate, and practical method for optimizing PV power grid connection capacity. Through dynamic optimization, it eliminates non-technical defects, improves PV absorption rate, reduces energy storage pressure, and is applicable to multi-connection point power supply scenarios. This invention has significant beneficial effects in improving the economic efficiency of PV power generation, enhancing grid stability, and reducing investment costs, and is of great significance for promoting the large-scale application of distributed PV power generation and optimizing the energy structure.

[0022] This invention uses big data analysis of historical power generation and consumption data to accurately determine the power generation and actual consumption value at each time period, achieving precise matching between photovoltaic power generation and electricity demand. This significantly improves the self-consumption rate, reduces curtailment, and enhances the economic benefits and return on investment of photovoltaic power generation.

[0023] This invention employs discreteness analysis and anomaly data screening mechanisms to effectively identify and handle abnormal fluctuations during power generation and consumption, avoiding grid instability caused by random factors such as cloud movement and equipment failure, thereby improving power supply quality and reliability.

[0024] This invention achieves global optimization rather than local optima by selecting benchmark access points and optimizing access point groups, taking into account the mutual influence between multiple access points. It fully leverages the potential of system-level coordinated optimization and avoids suboptimal solutions caused by independent optimization.

[0025] This invention reduces reliance on energy storage devices through precise capacity optimization configuration, thereby lowering the investment cost and operation and maintenance pressure of energy storage systems. At the same time, it can still ensure reliable power supply to relevant access points while reducing the use of energy storage devices.

[0026] This invention employs a dynamic data update and processing mechanism, which can adapt to the randomness and uncertainty of photovoltaic power output and load demand, overcome the problem that traditional static models may fail when faced with actual fluctuations, and improve the practicality and adaptability of optimization results.

[0027] This invention calculates and optimizes the total distance value, prioritizing the selection of combinations of access points that are closer together, effectively reducing line losses, cable investment and operating costs, and improving the overall system economy.

[0028] This invention, while meeting the constraints of safe and stable grid operation, improves the grid's capacity to accept distributed photovoltaic power generation through scientific capacity optimization, thereby promoting the large-scale development and utilization of renewable energy.

[0029] This invention establishes a complete evaluation system and optimization algorithm, providing a scientific basis and decision support for the configuration of photovoltaic power generation access capacity, reducing the uncertainty caused by human factors, and improving the accuracy and reliability of decision-making.

[0030] This invention achieves intelligent operation and maintenance management of photovoltaic power generation systems through big data analysis and automatic optimization algorithms, reducing manual intervention, improving operation and maintenance efficiency, and lowering operation and maintenance costs.

[0031] This invention promotes the large-scale application of clean energy, optimizes the energy structure, and reduces dependence on traditional fossil fuels by improving the utilization efficiency of photovoltaic power generation and grid acceptance capacity, thus having significant environmental and social benefits. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating a photovoltaic power generation access capacity optimization method based on big data analysis, according to an exemplary embodiment. Figure 2 This is a schematic diagram of a photovoltaic power generation access capacity optimization device based on big data analysis, according to an exemplary embodiment. Detailed Implementation

[0033] To more clearly illustrate the technical features of the present invention, the present invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings.

[0034] like Figure 1 As shown in the figure, an embodiment of the present invention provides a photovoltaic power generation access capacity optimization method based on big data analysis, which includes the following steps: Step S1: Based on the historical power generation data of the photovoltaic power station in multiple time periods over several days, analyze the data dispersion of power generation in each time period and determine the power generation corresponding to each time period; Step S2: Obtain historical power consumption data of multiple photovoltaic access points in multiple time periods, analyze the dispersion of power consumption values ​​of each access point in each time period, and determine the actual power consumption value of each photovoltaic access point in each time period. Step S3: Select the photovoltaic access point from all photovoltaic access points where the actual value of the power consumption in each time period is less than the power generation in the corresponding time period and the sum of the differences between the two is the smallest, and mark it as the benchmark access point. Step S4: Combine the reference access point with the remaining undetermined access points to form multiple access point groups, wherein the sum of the actual absorption values ​​of all photovoltaic access points in each access point group is less than the sum of the electricity generated by the photovoltaic power station in all time periods. Step S5: Calculate the access negative value for each access point group. The access negative value is the number of time periods in the group where the sum of the actual consumption values ​​for any time period is greater than the electricity consumption for the corresponding time period. Step S6: Calculate the total distance for each access point group, whereby the total distance is determined by the real-time distance between each pending access point in the group and the photovoltaic power station. Step S7: Assign weights to the negative access value and total distance value of each access point group, then sum them up to obtain the evaluation value of the group; Step S8: Select the access point group with the best evaluation value as the target access point group for photovoltaic power generation access capacity optimization configuration.

[0035] As one possible implementation of this embodiment, step S1 includes the following steps: Step S11: Divide the day into several time periods of uniform duration, and collect historical electricity data generated by the photovoltaic power station in each of the aforementioned time periods within the past month through an electricity metering device, with a collection interval of no more than 1 hour. Step S12: For any target time period, record the historical electricity data of that time period as Di, where i = 1, 2...n, and n is the number of days in the past month. Calculate the arithmetic mean P of Di, filter out abnormal data that satisfy "|Di-P|>preset abnormal threshold X1", and calculate the ratio of the number of abnormal data to the total data volume n to obtain the filtering ratio. Step S13: If the screening ratio is less than or equal to the preset ratio threshold B1, take the maximum value of all Di in the time period as the power consumption; if the screening ratio is greater than B1, take the maximum value of Di in the most recent ten days in the time period as the power consumption. Step S14: Repeat steps S12 and S13 for all time periods to obtain the electricity consumption corresponding to each time period, forming the basic data of electricity consumption.

[0036] As one possible implementation of this embodiment, the preset anomaly threshold X1 is set to 5-15 kWh, the preset proportion threshold B1 is set to 0.1-0.2, the time period can be adjusted within the range of 1-4 hours, and the total time period is 24 hours. X1 is used to determine whether the power generation data of a certain period deviates too much from the average, thereby identifying abnormal power generation days (such as cloudy / rainy days, faults, etc.). The value range of X1 can usually be set to 5-15 kWh, and the specific value should be combined with the historical power generation fluctuation of the photovoltaic power station; if the power station has high power generation stability, a smaller value can be used (such as 5 kWh); if the fluctuation is large, it can be appropriately relaxed to 10-15 kWh. B1 is used to control the proportion of abnormal data removal to avoid insufficient representativeness due to the removal of too much data.

[0037] As one possible implementation of this embodiment, step S2 includes the following steps: Step S21: Collect historical absorption values ​​of all access points within the power supply range of the photovoltaic power station for each time period within the past month. Record the absorption value of any access point for any time period as Fi, i=1, 2...n, where n is the number of days in the past month. Step S22: Calculate the arithmetic mean U and the root mean square deviation of Fi, and obtain the deviation value W by the ratio of the root mean square deviation to U. Determine whether W exceeds the preset fluctuation threshold X2; Step S23: If W≤X2, take the minimum value of Fi as the actual value of the time period corresponding to the access point; if W>X2, remove Fi in descending order of "|Fi-U|" and recalculate W until W≤X2; calculate the deletion ratio by the ratio of the amount of data removed to the total amount of data n; if the deletion ratio≤B1, take the minimum value of the remaining Fi as the actual value of the time period; otherwise, take the minimum value of the original Fi as the actual value of the time period. Step S24: Repeat steps S22 and S23 for all access points and all time periods to obtain the actual absorption value corresponding to each time period.

[0038] As one possible implementation of this embodiment, the preset fluctuation threshold X2 is set to 0.2-0.4, the power supply range is 3-10km around the photovoltaic power station, and the sampling frequency of the historical absorption value data acquisition device and the power metering device are consistent. X2 is used to determine whether the absorption value fluctuation of a certain access point is within a reasonable range, so as to avoid the impact of abnormal power consumption behavior such as production stoppage or maintenance on the accuracy of the actual absorption value; it can usually be set to 0.2-0.4, that is, the coefficient of variation relative to the mean. If the absorption behavior is relatively stable, it can be set to 0.2-0.3; if the fluctuation is large, it can be relaxed to 0.3-0.4.

[0039] As one possible implementation of this embodiment, step S3 includes the following steps: Step S31: Call up the electricity consumption Sj for each time period and the actual consumption value Aj corresponding to any access point, and calculate the sum of the differences C=Σ(Sj-Aj), j=1, 2...m, where m is the total number of time periods; Step S32: Remove access points where Aj≥Sj exists for any time period, and select the access point with the smallest C value from the remaining access points and mark it as the baseline access point.

[0040] As one possible implementation of this embodiment, step S4 includes the following steps: Step S41: Mark the remaining valid access points other than the selected baseline access points as pending access points; Step S42: Combine the reference access point with one or more pending access points to form multiple access point groups. The combination of the reference access point and the pending access point must meet the following conditions: the sum of the actual consumption values ​​of all access points in the group in each time period is less than or equal to the electricity consumption in the corresponding time period, and the sum of the total actual consumption values ​​in the group is less than or equal to the total electricity consumption. Step S43: If a pending access point violates the above conditions after being added, it will not be included in the group, and finally multiple access point groups that meet the constraints will be formed.

[0041] As one possible implementation of this embodiment, during the combination process, access point groups are generated in the order of "number of pending access points from few to many", and the remaining capacity of "total electricity consumption - total actual consumption value" is calculated in real time. When the remaining capacity is insufficient to accommodate the minimum consumption requirement of the next pending access point, the combination operation of the group is stopped.

[0042] As one possible implementation of this embodiment, step S5 includes the following steps: Step S51: For any generated access point group, calculate the total real value of the group's absorption in each time period, Tj, where Tj = the sum of the corresponding time periods Aj of all access points in the group; Step S52: Compare Tj with the corresponding power consumption Sj during the same time period, and count the number of time periods that satisfy "Tj > Sj". This number is the negative access value of the access point group.

[0043] As one possible implementation of this embodiment, step S6 includes the following steps: Step S61: Measure the actual distance between all undetermined access points in each access point group and the photovoltaic power station; Step S62: Select the calculation method according to the application scenario. In the industrial and commercial scenario, the sum of the distances of all undetermined access points in the group is taken as the total distance value. In the residential community scenario, the average distance of all undetermined access points in the group is taken as the total distance value. The distance mentioned in this invention is the physical distance.

[0044] In industrial and commercial scenarios, the total distance is calculated by summing the distances of all undetermined access points within the group. Specifically, the real-time distances between all undetermined access points in the access point group and the photovoltaic power station are obtained, and these real-time distances are summed to obtain the total distance value. This total distance value is suitable for scenarios where total line loss and total investment cost are more sensitive. This method focuses more on the absolute value of the total distance and is suitable for scenarios where overall line loss or cable investment needs to be controlled.

[0045] In residential community scenarios, the average distance between all undetermined access points within a group is taken as the total distance. Specifically, the real-time distances between all undetermined access points in the access point group and the photovoltaic power station are obtained, and the average of these real-time distances is marked as the total distance. This total distance is suitable for scenarios where average power supply distance and voltage stability are of greater concern. This method better reflects the average distribution of access points and is suitable for scenarios with high power quality requirements.

[0046] As one possible implementation of this embodiment, step S7 includes the following steps: Step S71: Set the weight coefficient for negative values ​​to 0.58 and the weight coefficient for distance from the total value to 0.42; Step S72: For each access point group, calculate the evaluation value of the group according to the formula "evaluation value = 0.58 × access negative value + 0.42 × total distance value".

[0047] As one possible implementation of this embodiment, step S8 includes the following steps: Step S81: Sort the evaluation values ​​of all access point groups; Step S82: Select the access point group with the smallest evaluation value as the target access point group; if the evaluation values ​​are the same, prioritize the group with the smaller negative value; if the negative values ​​are still the same, prioritize the group with the smaller total distance value. Step S83: Use the target access point group as the configuration basis for optimizing photovoltaic power generation access capacity.

[0048] As one possible implementation of this embodiment, if the evaluation value of the target access point group exceeds the preset warning threshold, a reminder signal is issued to prompt the user to check the integrity of historical data and the operating status of the acquisition device.

[0049] like Figure 2 As shown in the figure, an embodiment of the present invention provides a photovoltaic power generation access capacity optimization device based on big data analysis, comprising: The power generation determination module is used to analyze the data dispersion of power generation in each time period based on the historical power generation data of the photovoltaic power station in multiple time periods over several days, and to determine the power generation corresponding to each time period. The actual consumption value determination module is used to acquire historical consumption value data of multiple photovoltaic access points in multiple time periods, analyze the dispersion of the consumption value of each access point in each time period, and determine the actual consumption value of each photovoltaic access point in each time period. The benchmark access point screening module is used to select the photovoltaic access points from all photovoltaic access points where the actual value of the power consumption in each time period is less than the power generation in the corresponding time period and the sum of the differences between the two is the smallest, and these points are marked as benchmark access points. The access point group construction module is used to combine the benchmark access point with the remaining undetermined access points to form multiple access point groups, wherein the sum of the actual absorption values ​​of all photovoltaic access points in each access point group is less than the sum of the electricity generated by the photovoltaic power station in all time periods. The access negative value calculation module is used to calculate the access negative value for each access point group. The access negative value is the number of time periods in the group where the sum of the actual consumption values ​​in any time period is greater than the electricity consumption in the corresponding time period. The total distance calculation module is used to calculate the total distance of each access point group. The total distance is determined by the real-time distance between each undetermined access point in the group and the photovoltaic power station. The evaluation value calculation module is used to assign weights to the negative access value and total distance value of each access point group, and then add them together to obtain the evaluation value of the group. The target access point group selection module is used to select the access point group with the best evaluation value as the target access point group for the optimized configuration of photovoltaic power generation access capacity.

[0050] Taking a small industrial and commercial park as an example, a photovoltaic power station (with an average daily power generation of 2000 kWh) needs to optimize the grid connection capacity for three factories (A / B / C) within a radius of 5 km. The specific process of optimizing the photovoltaic grid connection for this small industrial and commercial park using the method described in this invention is as follows.

[0051] First, divide the time periods into: peak hours (10:00-14:00), normal hours (8:00-10:00, 14:00-16:00), and off-peak hours (the rest of the time). Preset the parameters: X1=10, B1=0.15, X2=0.3.

[0052] Step 1: Calculate the "electricity" of the photovoltaic power station at different times.

[0053] Peak power generation data (unit: kWh): Data for the past 30 days: [220, 215, 230, 10 (heavy rain days), 218, ..., 225] (n=30); The mean P = (ΣDi - 10) / 29 ≈ 221.7; Filtering: |Di-221.7|≤10 → Days with Di=10 during heavy rain are filtered out; Screening ratio = 1 / 30 ≈ 0.033 < 0.15 → Power consumption = max(Di) = 230; The algorithm is the same for other time periods. Assuming the results are: peak time 230, normal time 180, and valley time 120.

[0054] Step 2: Calculate the actual consumption value of each factory.

[0055] Taking the peak consumption value Fi of factory A as an example (unit: kWh): [100,105,98,20 (shutdown day),102,...,101] (n=30); The mean U = (ΣFi - 20) / 29 ≈ 102.4; The deviation W = √[Σ(Fi-U)² / (n-1)] / U ≈ 0.35 > 0.3, therefore it needs to be filtered. Sorting deviation: |20-102.4|=82.4 is the largest, so it should be deleted first. Recalculate W'≈0.12<0.3 → Screening ratio = 1 / 30≈0.033<0.15; Actual value of absorption = min(remaining Fi) = 98; (The same applies to factories B and C, assuming the results: peak value of A is 98, peak value of B is 85, and peak value of C is 90).

[0056] Step 3: Calculate the absorption difference and select the benchmark access point.

[0057] If C>0, all points are reserved as pending access points; Baseline access point: Select factory A with the smallest C value and Sj≥Aj for all time periods (satisfying 230>98, 180>95,...).

[0058] Step 4: Network optimization.

[0059] Undetermined groups: [A+B], [A+C], [A+B+C]; Checking peak load absorption for group [A+B]: 98+85=183<230 (no over-generation); negative value at connection = 0; Total distance: A (1km) + B (3km) = 4km; Evaluation value: 0.58×0 + 0.42×4 = 1.68, here it is assumed that the evaluation value of [A+C] is 2.1, and the evaluation value of [A+B+C] due to the excess power generation during the valley is 5.8; target group: [A+B], with the smallest evaluation value; Conclusion: Plants A and B should be prioritized as the grid connection group, with a total peak load of 183 kWh (accounting for 80% of the power plant's peak power generation).

[0060] The data in the above formula are all calculated by removing the dimensions and taking the numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for optimizing photovoltaic power generation grid connection capacity based on big data analysis, characterized in that, Includes the following steps: Step S1: Based on the historical power generation data of the photovoltaic power station in multiple time periods over several days, analyze the data dispersion of power generation in each time period and determine the power generation corresponding to each time period; Step S2: Obtain historical power consumption data of multiple photovoltaic access points in multiple time periods, analyze the dispersion of power consumption values ​​of each access point in each time period, and determine the actual power consumption value of each photovoltaic access point in each time period. Step S3: Select the photovoltaic access point from all photovoltaic access points where the actual value of the power consumption in each time period is less than the power generation in the corresponding time period and the sum of the differences between the two is the smallest, and mark it as the benchmark access point. Step S4: Combine the reference access point with the remaining undetermined access points to form multiple access point groups, wherein the sum of the actual absorption values ​​of all photovoltaic access points in each access point group is less than the sum of the electricity generated by the photovoltaic power station in all time periods. Step S5: Calculate the access negative value for each access point group. The access negative value is the number of time periods in the group where the sum of the actual consumption values ​​for any time period is greater than the electricity consumption for the corresponding time period. Step S6: Calculate the total distance for each access point group, whereby the total distance is determined by the real-time distance between each pending access point in the group and the photovoltaic power station. Step S7: Assign weights to the negative access value and total distance value of each access point group, then sum them up to obtain the evaluation value of the group; Step S8: Select the access point group with the best evaluation value as the target access point group for photovoltaic power generation access capacity optimization configuration.

2. The photovoltaic power generation grid connection capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Divide the day into several time periods of uniform duration, and collect historical electricity data generated by the photovoltaic power station in each of the aforementioned time periods within the past month through an electricity metering device, with a collection interval of no more than 1 hour. Step S12: For any target time period, record the historical electricity data of that time period as Di, where i = 1, 2...n, and n is the number of days in the past month. Calculate the arithmetic mean P of Di, filter out abnormal data that satisfy "|Di-P|>preset abnormal threshold X1", and calculate the ratio of the number of abnormal data to the total data volume n to obtain the filtering ratio. Step S13: If the screening ratio is less than or equal to the preset ratio threshold B1, take the maximum value of all Di in the current period as the power consumption; if the screening ratio is greater than or equal to the preset ratio threshold B1, take the maximum value of Di in the most recent ten days in the current period as the power consumption. Step S14: Repeat steps S12 and S13 for all time periods to obtain the electricity consumption corresponding to each time period, forming the basic data of electricity consumption.

3. The photovoltaic power generation access capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Collect historical absorption values ​​of all access points within the power supply range of the photovoltaic power station for each time period within the past month. Record the absorption value of any access point for any time period as Fi, i=1, 2...n, where n is the number of days in the past month. Step S22: Calculate the arithmetic mean U and the root mean square deviation of Fi. Obtain the deviation value W by the ratio of the root mean square deviation to U, and determine whether W exceeds the preset fluctuation threshold X2. Step S23: If W≤X2, take the minimum value of Fi as the actual value of the time period corresponding to the access point; if W>X2, remove Fi in descending order of "|Fi-U|" and recalculate W until W≤X2; calculate the deletion ratio by the ratio of the amount of data removed to the total amount of data n; if the deletion ratio≤B1, take the minimum value of the remaining Fi as the actual value of the time period; otherwise, take the minimum value of the original Fi as the actual value of the time period. Step S24: Repeat steps S22 and S23 for all access points and all time periods to obtain the actual absorption value corresponding to each time period.

4. The photovoltaic power generation access capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Call up the electricity consumption Sj for each time period and the actual consumption value Aj corresponding to any access point, and calculate the sum of the differences C=Σ(Sj-Aj), j=1, 2...m, where m is the total number of time periods; Step S32: Remove access points where Aj≥Sj exists for any time period, and select the access point with the smallest C value from the remaining access points and mark it as the baseline access point.

5. The photovoltaic power generation grid connection capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Mark the remaining valid access points other than the selected baseline access points as pending access points; Step S42: Combine the reference access point with one or more pending access points to form multiple access point groups. The combination of the reference access point and the pending access point must meet the following conditions: the sum of the actual consumption values ​​of all access points in the group in each time period is less than or equal to the electricity consumption in the corresponding time period, and the sum of the total actual consumption values ​​in the group is less than or equal to the total electricity consumption. Step S43: If a pending access point violates the above conditions after being added, it will not be included in the group, and finally multiple access point groups that meet the constraints will be formed.

6. The photovoltaic power generation access capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: For any generated access point group, calculate the total real value of the group's absorption in each time period, Tj, where Tj = the sum of the corresponding time periods Aj of all access points in the group; Step S52: Compare Tj with the power consumption Sj of the corresponding time period, and count the number of time periods that satisfy "Tj>Sj". This number is the access negative value of the access point group.

7. The photovoltaic power generation grid connection capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S6 includes the following steps: Step S61: Measure the actual distance between all undetermined access points in each access point group and the photovoltaic power station; Step S62: Select the calculation method according to the application scenario. For industrial and commercial scenarios, take the sum of the distances of all undetermined access points in the group as the total distance value. For residential community scenarios, take the average distance of all undetermined access points in the group as the total distance value.

8. The photovoltaic power generation access capacity optimization method based on big data analysis according to claim 1, characterized in that, Step S7 includes the following steps: Step S71: Set the weight coefficient for negative values ​​to 0.58 and the weight coefficient for distance from the total value to 0.42; Step S72: For each access point group, calculate the evaluation value of the group according to the formula "evaluation value = 0.58 × access negative value + 0.42 × total distance value".

9. The photovoltaic power generation access capacity optimization method based on big data analysis according to any one of claims 1-8, characterized in that, Step S8 includes the following steps: Step S81: Sort the evaluation values ​​of all access point groups; Step S82: Select the access point group with the smallest evaluation value as the target access point group; if the evaluation values ​​are the same, prioritize the group with the smaller negative value; if the negative values ​​are still the same, prioritize the group with the smaller total distance value. Step S83: Use the target access point group as the configuration basis for optimizing photovoltaic power generation access capacity.

10. A photovoltaic power generation access capacity optimization device based on big data analysis, characterized in that, include: The power generation determination module is used to analyze the data dispersion of power generation in each time period based on the historical power generation data of the photovoltaic power station in multiple time periods over several days, and to determine the power generation corresponding to each time period. The actual consumption value determination module is used to acquire historical consumption value data of multiple photovoltaic access points in multiple time periods, analyze the dispersion of the consumption value of each access point in each time period, and determine the actual consumption value of each photovoltaic access point in each time period. The benchmark access point screening module is used to select the photovoltaic access points from all photovoltaic access points where the actual value of the power consumption in each time period is less than the power generation in the corresponding time period and the sum of the differences between the two is the smallest, and these points are marked as benchmark access points. The access point group construction module is used to combine the benchmark access point with the remaining undetermined access points to form multiple access point groups, wherein the sum of the actual absorption values ​​of all photovoltaic access points in each access point group is less than the sum of the electricity generated by the photovoltaic power station in all time periods. The access negative value calculation module is used to calculate the access negative value of each access point group. The access negative value is the number of time periods in the group where the sum of the actual consumption values ​​in any time period is greater than the electricity consumption in the corresponding time period. The total distance calculation module is used to calculate the total distance of each access point group. The total distance is determined by the real-time distance between each undetermined access point in the group and the photovoltaic power station. The evaluation value calculation module is used to assign weights to the negative access value and total distance value of each access point group, and then add them together to obtain the evaluation value of the group. The target access point group selection module is used to select the access point group with the best evaluation value as the target access point group for the optimized configuration of photovoltaic power generation access capacity.

Images