A big data-based photovoltaic power generation management method

Abstract

This invention relates to the field of photovoltaic power generation management technology, specifically disclosing a photovoltaic power generation management method based on big data, comprising the following steps: S1: Obtain housing area, classify housing types based on housing area, obtain the number of households and daily electricity consumption of a certain type of household, and calculate the average electricity consumption; S2: Obtain the household mobility situation in the community, calculate the total electricity consumption increment based on household mobility situation, calculate the ideal electricity load, obtain the standard deviation of daily electricity consumption and set the ideal electricity consumption range; S3: Obtain the function of the daily photovoltaic power generation changing over time, calculate the total photovoltaic power generation, and manage photovoltaic power generation based on the ideal electricity consumption range and the total photovoltaic power generation. By segmenting users to achieve load forecasting, comparing the total photovoltaic power generation with the demand range, energy storage scheduling is optimized, photovoltaic self-consumption rate is improved, curtailment is reduced, and equipment lifespan is extended, achieving a dual improvement in the economy and reliability of the photovoltaic-storage system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation management technology, and more specifically to a photovoltaic power generation management method based on big data. Background Technology

[0002] As a renewable energy source that relies on natural sunlight, photovoltaic power generation is constantly affected by the dynamic weather conditions, exhibiting significant intermittent and fluctuating characteristics. Changes in sunshine duration and sunlight intensity can cause drastic fluctuations in power, thus requiring the storage of electricity generated by photovoltaic power generation.

[0003] Traditional energy storage charging and discharging strategies lack refined management capabilities, further amplifying the economic shortcomings of energy storage systems. Traditional strategies are mostly based on fixed time nodes or simple power threshold triggers, neglecting both grid load conditions and the health status of lithium batteries. Some projects adopt a fixed "sunrise charging, sunset discharging" pattern. If the grid load is low in the evening, energy storage discharging will further exacerbate the grid oversupply, forcing curtailment of solar power. Furthermore, blindly charging during off-peak hours and failing to discharge in a timely manner during peak hours means that the revenue from charging and discharging cannot cover the energy loss and equipment depreciation costs. Summary of the Invention

[0004] The purpose of this invention is to provide a photovoltaic power generation management method based on big data to solve the above-mentioned technical problems.

[0005] The objective of this invention can be achieved through the following technical solutions: A photovoltaic power generation management method based on big data includes the following steps; S1: Obtain the housing area S of each household in the community, divide the housing area S into small, medium and large types of housing, obtain the number of households I in small housing and the daily electricity consumption W, and calculate the average electricity consumption of small housing. Among them, W i This represents the electricity consumption of the i-th small apartment. The average electricity consumption W of the medium-sized apartment is calculated using the method described above. ave_M Average electricity consumption (W) of large apartments ave_L ; S2: Obtain the daily household mobility data within the community, including the number of households moving in (N) and the number of households moving out (O). Calculate the total electricity consumption increment W based on the household mobility data. z ; Calculate the ideal electrical load for the day: F=W last +W z Among them, W last The community's total electricity consumption for the previous day is represented by the standard deviation s of the community's daily electricity consumption, and the ideal electricity consumption range [Fs, F+s] is set. S3: Obtain the function G(t) that changes the photovoltaic power generation G on that day with time t, and calculate the total photovoltaic power generation. , where t start t represents the moment when photovoltaic power generation begins. end This represents the moment when photovoltaic power generation stops; If the total photovoltaic power generation G all Located outside the ideal electricity consumption range [Fs, F+s], based on the ideal electricity consumption range [Fs, F+s] and the total photovoltaic power generation G all To manage photovoltaic power generation.

[0006] As a further aspect of the present invention: In step S3, based on the ideal power consumption range [Fs, F+s] and the total photovoltaic power generation G... all The specific methods for managing photovoltaic power generation include: When G all When the power consumption is greater than F+s, the community's electricity consumption and photovoltaic power generation are in an unbalanced state. Therefore, G... all Store the amount of electricity in -(F+s); When G all When the value is less than Fs, there is an energy gap between the community's electricity consumption and photovoltaic power generation. The stored electricity is used to supply power to fill the energy gap.

[0007] As a further aspect of the present invention: calculate the energy gap Q = (Fs) - G all If the stored electrical energy D is obtained, and Q > D, then the stored electrical energy is insufficient to fill the energy gap Q. The staff is notified that the stored electrical energy is insufficient and the remaining energy gap is QD.

[0008] As a further aspect of the present invention: in step S1, the method for classifying apartments into small, medium, and large types based on the housing area S includes: Preset area gradient TD=λ×S t Where λ represents the preset gradient coefficient, λ=1,2,...,S t Representing a preset unit area, obtain the gradient coefficient range [λ] corresponding to the housing area S. s , λ s +1], if λ s If +1≤λ1, it is denoted as a small apartment; if λ1≤λ s And λ s If +1≤λ2, it is denoted as a medium-sized unit; if λ2≤λ s This is denoted as a large apartment, where λ1 and λ2 represent the preset first and second judgment coefficients.

[0009] As a further aspect of the present invention: in step S1, the total number of moves-in and moves-out of the same housing unit within a preset time period, H, is obtained. When the total number of moves-in and moves-out, H > H s When this occurs, the housing unit will be recorded as an abnormal housing unit, and the electricity consumption of abnormal housing units will not be included in subsequent calculations. s This represents the preset entry and exit threshold.

[0010] As a further aspect of the present invention: In step S2, if both the number of households moving in (N) and the number of households moving out (O) in the community on that day are 0, then the calculation of the total electricity consumption increment W is stopped. z .

[0011] As a further aspect of the present invention: in step S3, if Fs≤G all ≤F+s, at this time the community's electricity consumption and photovoltaic power generation are in a balanced state, so that photovoltaic power generation directly supplies electricity to the community.

[0012] As a further aspect of the present invention: in step S2, the total electricity consumption increment W is calculated based on the household flow situation. z The methods include: Calculate the increase in electricity consumption (W) for small apartments z_S =(N S -O S )×W ave_S , where N S O represents the number of small-sized families moving in. S The number of small-sized households moving out is used to calculate the increase in electricity consumption (W) for medium-sized and large-sized households using the method described above. z_M W z_L Calculate the total electricity consumption increment W z =W z_S +W z_M +W z_L .

[0013] The beneficial effects of this invention are as follows: First, to achieve management and optimization of community photovoltaic energy storage systems and overcome the shortcomings of traditional extensive strategies, the electricity consumption behavior characteristics of community users are profiled and quantitatively analyzed. The core is the construction of user grouping and benchmark electricity consumption profiles. This is achieved by acquiring the housing area data of each household in the community and dividing all households into three categories: small, medium, and large apartments. Based on this, the system further counts the total number of small apartment households and summarizes the daily electricity consumption data of this group. By calculating the average, the average electricity consumption of small apartments can be obtained. Similarly, the system can calculate the average electricity consumption of medium and large apartments in parallel. This calculation process lays a solid data foundation for subsequent strategy optimization. By classifying apartment types and calculating average electricity consumption, a community load prediction model can be constructed. Compared to treating the entire community as a homogeneous whole, apartment-type modeling can more accurately grasp the electricity consumption patterns and characteristics of different family groups.

[0014] After classifying apartment types, dynamic social data needs to be incorporated for real-time load forecasting to address the uncertainties brought about by community population movement. Specifically, the system obtains daily household movement data in real time, including the number of households moving in (N) and moving out (O), through a data interface connected to the community management platform. Based on this movement data, the system calculates the total electricity consumption increment W caused by population movement using a pre-set model. z This model comprehensively considers the average electricity consumption level of mobile households. Subsequently, the system performs dynamic load forecasting, calculating the community's total electricity consumption W from the previous day. last Compared with the daily increase in electricity consumption W z Adding them together, we get the basic load forecast value F for that day, i.e., F = W. last +W z To transform this point prediction into a more robust and safe operating range, the system uses historical data to calculate the standard deviation s of the community's daily total electricity consumption, and finally sets the ideal electricity consumption range for the day as [Fs, F+s]. The establishment of this range provides a dynamically adjusted benchmark for the subsequent adaptive control of the energy storage system, enabling the system to distinguish between normal fluctuations and abnormal conditions.

[0015] The core value of this approach lies in upgrading the management strategy of energy storage systems from a lagging response based on static historical patterns to a forward-looking intelligent control capable of sensing real-time social dynamics. This significantly improves the prediction accuracy and adaptive capabilities of community energy systems. By quantifying the key variable of household flow, the system can capture sudden changes in electricity demand caused by social activities, which are easily overlooked by traditional time series models. This makes the prediction of the total daily electricity load more accurate and reliable, providing a more scientific basis for the charging and discharging plans of energy storage systems and effectively avoiding energy mismatch caused by information lag or lack. More importantly, by establishing a safe operating range based on standard deviation rather than relying on a single predicted value, the system provides a control objective for energy storage dispatch that is both flexible and stable, balancing economy and reliability. The system does not need to make frequent charging and discharging adjustments that damage battery life in order to track a potentially fluctuating precise value; it only needs to maintain the community's net load within this range. This ensures the stable operation of the power grid and extends equipment life and optimizes operating costs through a gentler and smarter charging and discharging strategy.

[0016] After establishing the ideal electricity consumption range, the system needs to simultaneously monitor and coordinate the photovoltaic power generation side. The system obtains a function G(t) representing the daily photovoltaic power generation G as a function of time t. This function can be generated by fitting real-time data from the photovoltaic inverter with an irradiance prediction model. Based on this function, the system calculates the power generation from the start time t using integration. start until the end of power generation t end Total photovoltaic power generation G all Then, the system will G all Compare with the ideal power consumption range [Fs, F+s], if G all If the value is within this range, it indicates that photovoltaic power generation is basically matched with the community's electricity demand, and the system can operate according to conventional strategies; if G all If the situation exceeds this range, it means that there will be a significant power surplus or shortage, and the system must activate advanced management strategies.

[0017] A closed-loop intelligent decision-making system based on two-way perception was constructed, placing fluctuating photovoltaic (PV) power generation and dynamic community load within the same framework for collaborative optimization. This directly addresses the core pain points of "intermittency and volatility" and the "blindness" of traditional strategies mentioned in the background. By comparing the total PV power generation with dynamically predicted electricity demand ranges in real time, the system, for the first time, possesses the ability to proactively assess and regulate the global energy balance at the start of each day. When it is predicted that power generation will far exceed the upper limit of the demand range, the system can plan the charging strategy of the energy storage system in advance, or orderly connect excess electricity to the grid, rather than passively waiting for power to exceed the limit and forcibly "curtailing" PV, thereby maximizing the self-consumption rate of renewable energy and reducing energy waste. Conversely, when it is predicted that power generation will be insufficient, the system can reserve energy storage capacity in advance or adjust the discharge plan to make up for the expected power gap and ensure the stability of community power supply. This changes the passive response mode of traditional strategies, effectively smoothing out power fluctuations through feedforward control, not only improving grid friendliness, but also significantly enhancing the economy and operational reliability of the entire PV-storage system by optimizing energy flow. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart illustrating a photovoltaic power generation management method based on big data according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 As shown, this invention is a photovoltaic power generation management method based on big data, comprising the following steps: S1: Obtain the housing area S of each household in the community, divide the housing area S into small, medium and large types of housing, obtain the number of households I in small housing and the daily electricity consumption W, and calculate the average electricity consumption of small housing. Among them, W i This represents the electricity consumption of the i-th small apartment. The average electricity consumption W of the medium-sized apartment is calculated using the method described above. ave_M Average electricity consumption (W) of large apartments ave_L ; S2: Obtain the daily household mobility data within the community, including the number of households moving in (N) and the number of households moving out (O). Calculate the total electricity consumption increment W based on the household mobility data.z ; Calculate the ideal electrical load for the day: F=W last +W z Among them, W last The community's total electricity consumption for the previous day is represented by the standard deviation s of the community's daily electricity consumption, and the ideal electricity consumption range [Fs, F+s] is set. S3: Obtain the function G(t) that changes the photovoltaic power generation G on that day with time t, and calculate the total photovoltaic power generation. , where t start t represents the moment when photovoltaic power generation begins. end This represents the moment when photovoltaic power generation stops; If the total photovoltaic power generation G all Located outside the ideal electricity consumption range [Fs, F+s], based on the ideal electricity consumption range [Fs, F+s] and the total photovoltaic power generation G all To manage photovoltaic power generation.

[0022] It should be noted that the housing area data for each household in the community was obtained. Based on this housing area information, all households were divided into three categories: small, medium, and large apartments. After completing the apartment type classification, subsequent processing was carried out using the small apartment group as an example. The total number of residents in this group was counted, and the daily electricity consumption data of this group over a period of time was summarized. The daily electricity consumption data records the electricity consumption trajectory of each household. By analyzing this data, the average electricity consumption of small apartment residents can be calculated. This comprehensively reflects the overall electricity consumption level and characteristics of this group and has important reference value.

[0023] Similarly, the system employs the same parallel processing flow for both medium and large-sized apartments. This parallel computing approach not only improves data processing efficiency but also ensures that data from different apartment types are independent yet comparable, providing strong support for a comprehensive understanding of the community's electricity consumption structure.

[0024] On the one hand, by dividing the community into different apartment types and calculating the average electricity consumption of each type, a community load forecasting model is constructed. Compared with the traditional simple model that treats the entire community as a homogeneous whole, this apartment-type modeling method has the following advantages. First, different types of households, due to differences in living space size, family composition, and daily routines, will inevitably exhibit unique electricity consumption patterns and characteristics. Apartment-type modeling can keenly capture these subtle differences, thereby more accurately predicting the community's electricity demand at different times.

[0025] Secondly, based on this data and models, managers can formulate more personalized and refined scheduling strategies for photovoltaic energy storage systems. Simultaneously, they can provide targeted energy-saving suggestions and services for different types of households, guiding them to rationally adjust their electricity consumption behavior and further reduce energy consumption.

[0026] After classifying the apartment types, to effectively address the uncertainty in electricity consumption caused by community population movement and to achieve precise control and optimized scheduling of the photovoltaic energy storage system, the system is integrated with the community management platform. Utilizing its open data interface, it captures real-time household movement information, including the specific number of newly moved-in households (N) and the corresponding number of moved-out households (O). This constantly updated population change data provides clues for analyzing changes in electricity consumption trends.

[0027] Based on a specially designed algorithm model, the system comprehensively considers multiple factors, such as the average electricity consumption level of mobile households, to calculate the total increase in electricity consumption W caused by population migration. z This model can reflect the impact mechanism of different family sizes, lifestyles, and other factors on electricity consumption, ensuring a high degree of accuracy and reliability in the calculation results.

[0028] Subsequently, the system initiates a dynamic load forecasting process, using the total electricity consumption (W) of the community recorded the previous day. last As a base value, the incremental electricity consumption W due to population movement on that day is added. z This allows us to derive the predicted base load value F for the day. However, a single point prediction cannot fully encompass all fluctuations in actual electricity consumption. To enhance the robustness and practicality of the prediction results, the system further utilizes historical datasets accumulated over a long period and employs statistical methods to calculate the standard deviation s of the community's daily total electricity consumption. This indicator quantifies the dispersion of electricity consumption and intuitively reflects the range of fluctuations in daily electricity consumption behavior.

[0029] Based on this, an ideal daily electricity consumption range [Fs, F+s] is constructed. This range accommodates normal electricity consumption fluctuations while effectively identifying abnormal situations that exceed the normal range. It provides a dynamically adjustable reference benchmark for the subsequent adaptive control strategy of the energy storage system, enabling the system to distinguish between normal and abnormal fluctuations and respond appropriately in a timely manner. This ensures the stable and efficient operation of the entire community's photovoltaic energy storage system, achieving dynamic balance and optimized allocation of energy supply and demand.

[0030] This represents a comprehensive upgrade from the traditional passive and delayed response mechanism based on static historical patterns to a forward-looking intelligent control system capable of keenly sensing real-time social dynamics. It signifies a strategic leap in community energy management, from experience-driven to data-driven, and from extensive management to refined operation. By deeply integrating dynamic social data and electricity consumption behavior characteristics, the system's predictive accuracy and adaptive capabilities have been qualitatively improved, enabling it to respond to complex and ever-changing electricity consumption scenarios with greater timeliness and accuracy.

[0031] The significance of this approach is that it establishes a safe operating range based on standard deviation, rather than relying solely on a single predicted value as the control target. On one hand, range-based management inherently possesses a fault-tolerance mechanism, capable of accommodating normal electricity fluctuations and avoiding frequent adjustments caused by excessive pursuit of precise values. On the other hand, the reasonable fluctuation range determined through statistical principles ensures both the stability of the power grid operation and provides a clear decision boundary for energy storage dispatch. This "flexible control" strategy brings multiple benefits: it reduces mechanical wear and chemical stress on equipment, significantly extending the lifespan of energy storage devices; it also reduces peak-to-valley differences through smoothed charge-discharge curves, optimizing transaction costs in the electricity market; simultaneously, stable system output provides high-quality ancillary services to the power grid, improving the reliability and economy of the entire distribution system.

[0032] After establishing the ideal electricity consumption range, the system immediately transitions to a phase of in-depth monitoring and collaborative management of the photovoltaic power generation side. Through real-time connection with the millisecond-level data streams from the photovoltaic inverters and a high-precision irradiance prediction model, the system dynamically generates a continuous function G(t) representing the daily photovoltaic power generation over time. This function not only captures the natural fluctuations in solar irradiance but also incorporates microscopic characteristics formed by actual engineering factors such as component degradation and shading. Based on this, the system uses a numerical integration algorithm to calculate the total power generation G over the entire cycle from sunrise to sunset. all .

[0033] When G all When the power consumption falls within the preset ideal power consumption range [Fs, F+s], it signifies that the photovoltaic supply and community demand have reached a natural equilibrium. At this point, the system adopts a steady-state operation strategy, prioritizing the consumption of locally produced clean energy to meet the base load. Surplus electricity is then systematically injected into the energy storage system for tiered utilization, forming a virtuous cycle of "self-generation and self-consumption, surplus electricity storage." However, once G is detected... all Breaking through the interval boundary, whether it is due to excess power generation caused by extreme sunny weather or power shortage caused by continuous rain, the system immediately responds according to the specific situation.

[0034] In summary, by segmenting users to achieve accurate load forecasting, combining population flow data to establish dynamic electricity consumption zones, and intelligently comparing total photovoltaic power generation with demand zones, the system gains forward-looking control capabilities. This allows it to proactively smooth out fluctuations, optimize energy storage scheduling, significantly improve photovoltaic self-consumption rate, reduce curtailment, and extend equipment lifespan, ultimately achieving a dual improvement in the economics and reliability of the photovoltaic-energy storage system.

[0035] In another preferred embodiment of the present invention, based on the ideal power consumption range [Fs, F+s] and the total photovoltaic power generation G all The specific methods for managing photovoltaic power generation include: When G all When the power consumption is greater than F+s, the community's electricity consumption and photovoltaic power generation are in an unbalanced state. Therefore, G... all Store the amount of electricity in -(F+s); When G all When the value is less than Fs, there is an energy gap between the community's electricity consumption and photovoltaic power generation. The stored electricity is used to supply power to fill the energy gap.

[0036] It is worth noting that intelligent energy storage scheduling is achieved by comparing the difference between the total photovoltaic power generation and the ideal power consumption range. When there is excess power generation, the surplus power is calculated and storage is activated; when there is insufficient power generation, the energy storage is automatically released to make up for the shortfall. This closed-loop control mechanism effectively smooths out fluctuations, improves energy self-sufficiency, reduces curtailment of solar power, ensures stable grid operation, and significantly enhances the economic efficiency and reliability of the photovoltaic-energy storage system.

[0037] In a preferred embodiment, the energy gap Q is calculated as Q = (Fs) - G. all If the stored electrical energy D is obtained, and Q > D, then the stored electrical energy is insufficient to fill the energy gap Q. The staff is notified that the stored electrical energy is insufficient and the remaining energy gap is QD.

[0038] Understandably, when insufficient power generation is detected, the energy gap Q is calculated first. The energy storage capacity D is compared with the gap Q. When the energy storage is sufficient, the gap is automatically filled; when the energy storage is insufficient, an early warning is proactively sent to the staff and the remaining gap QD is quantified, thereby improving the reliability of energy management and emergency response capabilities.

[0039] In another preferred embodiment of the present invention, the method for classifying small, medium, and large apartments based on housing area S includes: Preset area gradient TD=λ×S t Where λ represents the preset gradient coefficient, λ=1,2,...,S t Representing a preset unit area, obtain the gradient coefficient range [λ] corresponding to the housing area S. s , λ s +1], if λs If +1≤λ1, it is denoted as a small apartment; if λ1≤λ s And λ s If +1≤λ2, it is denoted as a medium-sized unit; if λ2≤λ s This is denoted as a large apartment, where λ1 and λ2 represent the preset first and second judgment coefficients.

[0040] It is important to note that this classification method effectively avoids regional biases caused by fixed standards. It can adapt to the actual situation of different communities and ensure the logical consistency of the classification, laying a solid foundation for subsequent differentiated electricity consumption analysis.

[0041] In another preferred embodiment of the present invention, the total number of moves-in and moves-out times for the same housing unit within a preset time period, H, is obtained. When the total number of moves-in and moves-out times H > H s When this occurs, the housing unit will be recorded as an abnormal housing unit, and the electricity consumption of abnormal housing units will not be included in subsequent calculations. s This represents the preset entry and exit threshold.

[0042] It should be noted that abnormal households are identified by monitoring the total number of moves-in and moves-out transactions (H) of a single housing unit within a preset time period. When H exceeds a preset threshold... s When this occurs, the housing unit is marked as an abnormal housing unit, and its electricity consumption data is excluded from subsequent electricity consumption analysis and load forecasting calculations. This mechanism can effectively filter out interference data caused by abnormal occupancy behaviors such as short-term rentals and frequent changes of ownership, ensuring that the electricity consumption data used for analysis and modeling truly reflects the electricity consumption patterns of stable residents, thereby significantly improving the accuracy of load forecasting and the reliability of community energy management strategies.

[0043] In another preferred embodiment of the present invention, if both the number of households moving in (N) and the number of households moving out (O) in the community on that day are 0, then the calculation of the total electricity consumption increment W is stopped. z .

[0044] Understandably, when both the number of households moving in and out of the community on a given day are zero, the calculation of incremental electricity consumption is skipped. This avoids redundant calculations, improves system efficiency, and ensures the accuracy of load forecasting results when the community population is stable.

[0045] In another preferred embodiment of the present invention, if Fs≤G all ≤F+s, at this time the community's electricity consumption and photovoltaic power generation are in a balanced state, so that photovoltaic power generation directly supplies electricity to the community.

[0046] It is worth noting that when the total photovoltaic power generation G allWhen the system operates within the ideal power consumption range [Fs, F+s], it determines that the solar-load balance is achieved. At this point, the control system prioritizes direct photovoltaic power generation, utilizing all generated electricity for the community's load. This avoids energy and equipment losses in the energy storage process and achieves local consumption of clean energy through the simplest path, effectively improving system operating efficiency and economy.

[0047] In another preferred embodiment of the invention, the total electricity consumption increment W is calculated based on household mobility. z The methods include: Calculate the increase in electricity consumption (W) for small apartments z_S =(N S -O S )×W ave_S , where N S O represents the number of small-sized families moving in. S The number of small-sized households moving out is used to calculate the increase in electricity consumption (W) for medium-sized and large-sized households using the method described above. z_M W z_L Calculate the total electricity consumption increment W z =W z_S +W z_M +W z_L .

[0048] It is worth noting that the electricity consumption increments for small, medium, and large apartments are calculated separately, with the net number of mobile households for each apartment type multiplied by the corresponding average electricity consumption. Then, the electricity consumption increments for the three apartment types are summed to obtain the total electricity consumption increment W for the community. z It takes into account the differences in electricity consumption characteristics among different household structures, making it more accurate than simply using the overall average value, and providing a reliable data foundation for load forecasting.

[0049] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A photovoltaic power generation management method based on big data, characterized in that, Includes the following steps: S1: Obtain the housing area S of each household in the community, divide the housing area S into small, medium and large types of housing, obtain the number of households I in small housing and the daily electricity consumption W, and calculate the average electricity consumption of small housing. Among them, W i This represents the electricity consumption of the i-th small apartment. The average electricity consumption W of the medium-sized apartment is calculated using the method described above. ave_M Average electricity consumption (W) of large apartments ave_L ; S2: Obtain the daily household mobility data within the community, including the number of households moving in (N) and the number of households moving out (O). Calculate the total electricity consumption increment W based on the household mobility data. z ; Calculate the ideal electrical load for the day: F=W last +W z Among them, W last The community's total electricity consumption for the previous day is represented by the standard deviation s of the community's daily electricity consumption, and the ideal electricity consumption range [Fs, F+s] is set. S3: Obtain the function G(t) that changes the photovoltaic power generation G on that day with time t, and calculate the total photovoltaic power generation. , where t start t represents the moment when photovoltaic power generation begins. end This represents the moment when photovoltaic power generation stops; If the total photovoltaic power generation G all Located outside the ideal electricity consumption range [Fs, F+s], based on the ideal electricity consumption range [Fs, F+s] and the total photovoltaic power generation G all To manage photovoltaic power generation.

2. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S3, based on the ideal electricity consumption range [Fs, F+s] and the total photovoltaic power generation G all The specific methods for managing photovoltaic power generation include: When G all When the power consumption is greater than F+s, the community's electricity consumption and photovoltaic power generation are in an unbalanced state. Therefore, G... all Store the amount of electricity in -(F+s); When G all When the value is less than Fs, there is an energy gap between the community's electricity consumption and photovoltaic power generation. The stored electricity is used to supply power to fill the energy gap.

3. The photovoltaic power generation management method based on big data according to claim 2, characterized in that, Calculate the energy gap Q = (Fs) - G all If the stored electrical energy D is obtained, and Q > D, then the stored electrical energy is insufficient to fill the energy gap Q. The staff is notified that the stored electrical energy is insufficient and the remaining energy gap is QD.

4. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S1, the method for classifying apartments into small, medium, and large types based on the housing area S includes: Preset area gradient TD=λ×S t Where λ represents the preset gradient coefficient, λ=1,2,...,S t Representing a preset unit area, obtain the gradient coefficient range [λ] corresponding to the housing area S. s , λ s +1], if λ s If +1≤λ1, it is denoted as a small apartment; if λ1≤λ s And λ s If +1≤λ2, it is denoted as a medium-sized unit; if λ2≤λ s This is denoted as a large apartment, where λ1 and λ2 represent the preset first and second judgment coefficients.

5. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S1, the total number of moves-in and moves-out of the same housing unit within a preset time period, H, is obtained. When the total number of moves-in and moves-out, H > H s When this occurs, the housing unit will be recorded as an abnormal housing unit, and the electricity consumption of abnormal housing units will not be included in subsequent calculations. s This represents the preset entry and exit threshold.

6. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S2, if both the number of households moving in (N) and the number of households moving out (O) in the community on that day are 0, then the calculation of the total electricity consumption increment W is stopped. z .

7. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S3, if Fs≤G all ≤F+s, at this time the community's electricity consumption and photovoltaic power generation are in a balanced state, so that photovoltaic power generation directly supplies electricity to the community.

8. The photovoltaic power generation management method based on big data according to claim 1, characterized in that, In step S2, the total electricity consumption increment W is calculated based on household flow patterns. z The methods include: Calculate the increase in electricity consumption (W) for small apartments z_S =(N S -O S )×W ave_S , where N S O represents the number of small-sized families moving in. S The number of small-sized households moving out is used to calculate the increase in electricity consumption (W) for medium-sized and large-sized households using the method described above. z_M W z_L Calculate the total electricity consumption increment W z =W z_S +W z_M +W z_L .

Images