Micro-grid load fluctuation suppression method

Abstract

The invention provides a micro-grid load fluctuation suppression method, which belongs to the technical field of micro-grid operation, and comprises the following steps: S1, collecting data; s2, establishing a load prediction model; s3, establishing a micro-grid complementary system; and S4, the energy storage system suppresses load fluctuation. According to the micro-grid load fluctuation suppression method provided by the invention, by comprehensively collecting the data of the power supply end and the load end and generating the prediction curve, a reliable basis is provided for subsequent model establishment and system adjustment; prediction deviation is reduced through dynamic adjustment, and the accuracy of load prediction is improved; for the corrected predicted load, power supply ends in different power supply advantage periods are adopted at the power supply ends for complementary matching, energy storage sharing is achieved, the load is reasonably coordinated, and load fluctuation is remarkably reduced; charging and discharging of the energy storage system are adjusted according to the power balance state and the load prediction result, load fluctuation is further stabilized, power balance of the micro-grid is guaranteed, and the operation stability of the micro-grid is improved.

Description

Technical Field

[0001] This invention belongs to the field of microgrid operation technology, and more specifically, relates to a method for suppressing load fluctuations in microgrids. Background Technology

[0002] The rapid development of modern industry has led to unprecedented energy demands, and the traditional energy structure based on fossil fuels can no longer meet these growing needs. To address the challenges posed by energy pollution and environmental pollution, a green, environmentally friendly, and low-carbon lifestyle has become a primary goal. Renewable energy is clean, renewable, and widely distributed, but its utilization rate in daily life remains low due to its randomness and geographical limitations. To improve the utilization rate of renewable energy, microgrid operation has emerged. Microgrids can integrate renewable energy and employ energy storage systems to maximize the utilization rate of renewable energy. A typical smart microgrid system usually consists of distributed generation equipment, energy storage systems, energy management systems, and transmission and distribution systems.

[0003] In the actual operation of microgrids, the output of distributed generation equipment is highly random due to the influence of natural conditions, while the load carried by the microgrid (such as residential electricity consumption and industrial electricity consumption) also exhibits a certain degree of fluctuation. This dual uncertainty directly affects the power balance of the microgrid, thereby threatening the stability of its operation. Summary of the Invention

[0004] The purpose of this invention is to provide a method for suppressing load fluctuations in microgrids, which aims to improve the operational stability of microgrid systems.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is: to provide a microgrid load fluctuation suppression method, comprising the following steps: S1. Data collection; collect rated output parameters and historical output data of the microgrid power supply end to generate power supply fluctuation prediction curves; collect historical and real-time power consumption data of the load end to generate load fluctuation prediction curves. S2. Establish a load forecasting model; dynamically adjust the forecasting results using load fluctuation forecasting curves and real-time load to reduce forecasting deviations. S3. Establish a microgrid complementary system; by coordinating the loads by complementing the power supply terminals of multiple microgrids with different power supply advantages during different power supply periods, and by using a shared energy storage system to achieve energy storage sharing, the load fluctuation of the microgrid can be reduced. S4. Energy storage system suppresses load fluctuations; based on the power balance status of the microgrid and the load forecast results, the charging and discharging power and charging and discharging time of the energy storage system are adjusted to smooth out load fluctuations.

[0006] As another embodiment of this application, in step S1, the historical power output data of the microgrid power supply end covers the hourly power output value and corresponding meteorological condition records of each day in the past 12 months. Based on the above data, a power supply fluctuation prediction curve is generated by time series decomposition method. Historical and real-time electricity consumption data are collected from the load side. Historical electricity consumption data includes daily time-of-day electricity consumption records for different user types over the past 12 months. Real-time electricity consumption data is collected from smart meters at 10-30 minute intervals to collect the current electricity load value. Combined with user type tags and time period characteristics, a load fluctuation prediction curve is generated.

[0007] In another embodiment of this application, in step S2, the data from step S1 is analyzed and processed to establish a load forecasting model and predict the load changes of the microgrid over a future period of time; the prediction deviation threshold is 25% of the predicted load value. When the deviation between the real-time collected current power load value and the predicted load value exceeds the threshold, the dynamic adjustment mechanism is triggered; real-time power consumption data is re-collected, and the newly collected real-time power consumption data, meteorological data, and real-time power output data from the power supply end are input into the load prediction model to regenerate the load prediction curve.

[0008] In another embodiment of this application, a deviation correction factor is introduced when obtaining the prediction deviation threshold; When real-time influencing factors exist, the prediction deviation threshold of the predicted load is first corrected using the deviation correction factor, and then the current electricity load value is compared with the corrected prediction deviation threshold of the predicted load.

[0009] In another embodiment of this application, in step S3, based on the principle of complementary advantages during peak periods, at least two microgrid power supply terminals with off-peak output characteristics are connected in a network to build a collaborative power supply network, and real-time data interaction and power transmission between the power supply terminals are realized through the energy management bus.

[0010] In another embodiment of this application, a shared energy storage system is set up for the power supply network, and the capacity of the shared energy storage system is greater than 20% of the maximum load of the power supply network; Multiple shared energy storage systems can be set up, and each shared energy storage system is connected to at least one microgrid power supply terminal with peak-shifting output characteristics.

[0011] As another embodiment of this application, in step S4, a power supply data platform is set up: microgrid power supply information, energy storage information and electricity consumption information are collected, a data platform is established, and a unified data interface and interaction protocol are formulated. The power supply data platform monitors the distribution status of the microgrid in real time. If the predicted load is rising and the current power supply output is insufficient, the energy storage system will start discharging preparation 20 minutes in advance. If the predicted load is falling and the current power supply output is excessive, the energy storage system will start charging preparation 20 minutes in advance.

[0012] In another embodiment of this application, in step S4, the power supply data platform monitors the status of the energy storage system; When the charge level of a certain energy storage system in a discharge state drops below 20%, that energy storage system stops discharging. When the charge level of a certain energy storage system exceeds 80% while it is charging, that energy storage system stops charging.

[0013] As another embodiment of this application, the power supply data platform includes real-time monitoring and control of the voltage and frequency of the microgrid; When the voltage or frequency of a microgrid exceeds the normal range, the voltage and frequency can be restored to the normal range by adjusting the output of distributed power sources, the charging and discharging of energy storage systems, and load demand.

[0014] As another embodiment of this application, it also includes: S5. Establish a user-side electricity price adjustment mechanism; monitor users' electricity load in real time, and send electricity price adjustment signals to users when the microgrid load reaches peak or off-peak times to encourage users to reduce electricity consumption during peak hours and increase electricity consumption during off-peak hours.

[0015] The beneficial effects of the microgrid load fluctuation suppression method provided by this invention are as follows: Compared with the prior art, this invention's microgrid load fluctuation suppression method comprehensively collects data from both the power supply end and the load end and generates prediction curves, providing a reliable basis for subsequent model establishment and system adjustment, and helping to more accurately grasp the operating status of the microgrid; by dynamically adjusting to reduce prediction deviations, it improves the accuracy of load prediction and provides a more effective reference for the stable operation of the microgrid; for the corrected predicted load, by using complementary matching of power supply ends with different power supply advantage periods and realizing energy storage sharing at the power supply end, it can fully utilize the advantages of each power supply end, rationally coordinate the load, and significantly reduce load fluctuations; by adjusting the charging and discharging of the energy storage system according to the power balance state and load prediction results, it further smooths out load fluctuations, effectively copes with the dual uncertainties brought about by the randomness of distributed generation equipment output and load fluctuations, ensures the power balance of the microgrid, and improves the stability of microgrid operation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the microgrid load fluctuation suppression method provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0019] Please see Figure 1 The microgrid load fluctuation suppression method provided by the present invention will now be described. The microgrid load fluctuation suppression method includes the following steps: S1. Data collection; collect rated output parameters and historical output data of the microgrid power supply end to generate power supply fluctuation prediction curves; collect historical and real-time power consumption data of the load end to generate load fluctuation prediction curves. S2. Establish a load forecasting model; dynamically adjust the forecasting results using load fluctuation forecasting curves and real-time load to reduce forecasting deviations. S3. Establish a microgrid complementary system; by coordinating the loads by complementing the power supply terminals of multiple microgrids with different power supply advantages during different power supply periods, and by using a shared energy storage system to achieve energy storage sharing, the load fluctuation of the microgrid can be reduced. S4. Energy storage system suppresses load fluctuations; based on the power balance status of the microgrid and the load forecast results, the charging and discharging power and charging and discharging time of the energy storage system are adjusted to smooth out load fluctuations.

[0020] The microgrid load fluctuation suppression method provided by this invention, compared with existing technologies, comprehensively collects data from both the power supply and load sides and generates prediction curves, providing a reliable basis for subsequent model building and system adjustment, and helping to more accurately grasp the operating status of the microgrid; by dynamically adjusting to reduce prediction deviations, it improves the accuracy of load prediction and provides a more effective reference for the stable operation of the microgrid; for the corrected predicted load, it adopts complementary matching of power supply ends with different power supply advantage periods and realizes energy storage sharing, which can make full use of the advantages of each power supply end, reasonably coordinate the load, and significantly reduce load fluctuations; by adjusting the charging and discharging of the energy storage system according to the power balance status and load prediction results, it further smooths out load fluctuations, effectively copes with the dual uncertainties brought about by the randomness of distributed generation equipment output and load fluctuations, ensures the power balance of the microgrid, and improves the stability of microgrid operation.

[0021] Specifically, conventional microgrid systems include photovoltaic power generation systems, wind power generation systems, diesel power generation systems, and hybrid electric vehicle charging systems.

[0022] In step S1, the rated power of photovoltaic power generation equipment and historical output data under different seasons and weather conditions over the past year are collected, and combined with information such as local sunshine duration, to generate a power supply fluctuation prediction curve for photovoltaic power generation; the rated power of wind power generation equipment and output data corresponding to historical wind speeds are collected to generate a power supply fluctuation prediction curve for wind power generation; the rated output parameters and historical operating output data of diesel generators are collected to generate their power supply fluctuation prediction curves.

[0023] Historical electricity consumption data from the past year for residents and small industrial users within the coverage area of ​​the corresponding microgrid, as well as real-time electricity consumption data from the user end, are collected to generate load fluctuation prediction curves. This data will provide a basis for subsequent steps.

[0024] In step S2, the load fluctuation prediction curve generated in step S1 is used in conjunction with the real-time load data of residential air conditioners, lighting, and industrial equipment currently being monitored. This data is then compared with the initial predicted load result and dynamically adjusted. For example, the weight of peak residential electricity consumption periods is adjusted to reduce prediction bias and enable the model to more accurately predict the load situation in the next few hours or even days.

[0025] In step S3, considering the different advantages of various power generation devices, at least two devices with different advantages are interconnected to form a microgrid system. Taking a microgrid system comprising photovoltaic (PV), wind, and diesel power generation as an example, based on the advantages of PV power generation during strong sunlight, wind power generation at night or during periods of high wind speed, and diesel power generation as a stable supplement, these three power sources are complementary to form a microgrid complementary system. Simultaneously, a shared energy storage system is established. When PV power generation is excessive during the day, the excess energy is stored in the shared energy storage system; when wind power generation is insufficient at night, the energy in the shared energy storage system is released. This coordinated approach reduces load fluctuations.

[0026] In step S4, the power balance of the microgrid is monitored in real time. If a sudden increase in load is detected at a certain moment, causing the power output to be less than the load demand, the load trend for the next 2 hours is predicted based on the load forecast results obtained in S2. At the same time, the charging and discharging power of the shared energy storage system is adjusted, either by reducing the charging power or increasing the discharging power, until the load change is gradual, in order to supplement the power supply gap, smooth out load fluctuations, and ensure the stable operation of the microgrid.

[0027] In some possible embodiments, in step S1, the historical output data of the microgrid power supply end covers the hourly output values ​​and corresponding meteorological condition records for each day over the past 12 months. Based on the above data, a power supply fluctuation prediction curve is generated using the time series decomposition method. Historical and real-time electricity consumption data are collected from the load end. The historical electricity consumption data includes daily time-segmented electricity consumption records for different user types over the past 12 months. The real-time electricity consumption data is collected from smart meters at 10-30 minute intervals to collect the current electricity load value. The load fluctuation prediction curve is generated by combining user type tags and time period characteristics.

[0028] The aforementioned meteorological records include solar radiation intensity, wind speed, temperature, and weather conditions. Based on this data, a time series decomposition method is used to break down historical power output data into trend, periodic, and random components. Each component is analyzed and predicted separately before being integrated to generate a power supply fluctuation prediction curve.

[0029] The trend term primarily reflects the overall trend of power output over a longer period, unaffected by short-term fluctuations. For example, for photovoltaic power generation, if the average annual sunshine duration in the region has gradually increased over the past 12 months, the trend term might show a slow overall increase in photovoltaic power output over time; for wind power generation, if the average wind speed in the region has shown a decreasing trend year by year, the trend term might show a gradual overall decrease in wind power output.

[0030] The periodic term reflects the recurring fluctuation pattern of power output at the power supply end within a fixed time period, which is usually related to periodic time units such as seasons, months, and weeks.

[0031] The stochastic term refers to power output fluctuations caused by random factors that cannot be explained by trend and periodic terms. Examples include the impact of gusts of wind and short-duration heavy rainfall on power generation equipment.

[0032] The power supply fluctuation prediction curve is generated by using the time series decomposition method based on the historical power output data of the power supply end. This method can accurately separate the trend, periodic and random components in the data, improve the accuracy of the prediction curve, and make the understanding of power supply fluctuations more in line with the actual situation.

[0033] Based on historical electricity consumption data from the load end, recorded by user type and daily time period, the electricity consumption characteristics of different users at different times can be clearly presented. Real-time electricity consumption data is collected by smart meters at 10-30 minute intervals and associated with user type tags and time period characteristics, which can capture the dynamic changes of the current electricity load in a timely manner.

[0034] In some possible embodiments, in step S2, the data in step S1 is analyzed and processed to establish a load forecasting model and predict the load changes of the microgrid in the future; the prediction deviation threshold is 25% of the predicted load value; when the deviation between the real-time collected current power load value and the predicted load value exceeds the threshold, a dynamic adjustment mechanism is triggered; real-time power consumption data is re-collected, and the newly collected real-time power consumption data, meteorological data, and real-time power output data of the power supply end are input into the load forecasting model to regenerate the load forecasting curve.

[0035] The data in step S1 is preprocessed, such as by removing abrupt changes. Based on the preprocessed data, an LSTM neural network is used to build a load forecasting model, and the model is used to output the predicted load change curve for the next 72 hours.

[0036] The predicted load value is extracted from the predicted load change curve for a specific time period, and the prediction deviation threshold is determined by the average value of this predicted load. For example, if the predicted peak load for residential users in a certain area is 80KW and the peak load for commercial users is 50KW, the total predicted load is 130KW. The prediction deviation threshold is 130KW × 25% = 32.5KW. That is, when the deviation between the real-time collected current electricity load value and the predicted load value for that time period is greater than 32.5KW, the dynamic adjustment mechanism is triggered.

[0037] The dynamic adjustment mechanism requires re-collecting real-time electricity consumption data, shortening the time interval between real-time data collection, and re-collecting meteorological information. The new electricity consumption data, meteorological information, and real-time power output data from the power supply end are input into the load forecasting model, and a revised load forecast curve is output. This revised load forecast curve provides a basis for subsequent microgrid load adjustments.

[0038] In some possible embodiments, a deviation correction factor is introduced when obtaining the prediction deviation threshold; when there are real-time influencing factors, the prediction deviation threshold of the predicted load is first corrected using the deviation correction factor, and then the current electricity load value is compared with the corrected prediction deviation threshold of the predicted load.

[0039] Real-time influencing factors include large-scale events and extreme heat weather. For example, under the extreme heat weather factor, if the correction factor is set to 1.3, then the prediction deviation threshold is predicted load × 25% × 1.3. The deviation correction factor for large-scale residential events is 1.2, then the prediction deviation threshold is predicted load × 25% × 1.2.

[0040] When multiple real-time influencing factors exist, a weighted average method is considered to derive the deviation correction factor. For example, if extreme high temperatures and large-scale events coexist, the impact of extreme high temperatures on the load is slightly greater than that of large-scale events. Extreme high temperatures are assigned a weight of 0.6, and large-scale events are assigned a weight of 0.4. The weighted average calculation yields a comprehensive deviation correction factor of 1.3 × 0.6 + 1.2 × 0.4 = 1.26.

[0041] The real-time load value is compared with the predicted load deviation threshold after correction. If the deviation exceeds the threshold, dynamic adjustment is triggered in a timely manner.

[0042] In some possible embodiments, in step S3, based on the principle of complementary advantages during peak periods, at least two microgrid power supply terminals with off-peak output characteristics are connected in a network to build a collaborative power supply network, and real-time data interaction and power transmission between the power supply terminals are realized through the energy management bus.

[0043] When constructing a coordinated power supply network based on the principle of complementary advantages during peak periods, it is first necessary to conduct a detailed analysis of the output characteristics of each microgrid power supply end and select at least two power supply ends with obvious peak-shifting output characteristics. For example, photovoltaic power plants and wind power plants can be selected as grid-connected entities, forming a significant peak-shifting effect between them during peak output periods.

[0044] When it is in the advantageous period of photovoltaic power generation during the day, if the output of the photovoltaic power station is excessive, the energy management bus will automatically control the photovoltaic power station to transmit the excess electric energy to the shared energy storage system associated with the wind farm through the coordinated power supply network according to the load prediction results and the real-time output of the wind farm for energy storage reserve. And when it comes to the advantageous period of wind power generation at night, if the output of the wind farm is excessive, the excess electric energy will also be transmitted to the energy storage device supporting the photovoltaic power station through this network. When the output of a certain power supply end is insufficient, the energy management bus will quickly call the reserved electric energy or real-time output of the other power supply end for supplementation to ensure the stability of the overall power supply.

[0045] Optionally, a shared energy storage system is set for the power supply network, and the capacity of the shared energy storage system is greater than 20% of the maximum load of the power supply network; multiple groups of shared energy storage systems can be provided, and each group of shared energy storage systems is at least connected to a group of microgrid power supply ends with the characteristic of peak-shifting output.

[0046] When setting the shared energy storage system, it is first necessary to clarify the maximum load value of the power supply network. For example, through the statistical analysis of the historical operation data of this network, it is obtained that its maximum load is 1000 kW, then the total capacity of the shared energy storage system needs to be greater than 1000 kW × 120% = 1200 kW to meet the energy storage demand for coping with load fluctuations.

[0047] In some possible embodiments, in step S4, a power supply data platform is set: collect the microgrid power supply information, energy storage information and power consumption information, establish a data platform, and formulate a unified data interface and interaction protocol; the power supply data platform monitors the microgrid distribution status in real time. If the predicted load shows an upward trend and the current power supply end has insufficient output, start the discharge preparation of the energy storage system 20 minutes in advance; if the predicted load shows a downward trend and the current power supply end has excessive output, start the charge preparation of the energy storage system 20 minutes in advance.

[0048] When establishing the data platform, formulate a unified data interface standard, such as using the Modbus TCP / IP protocol as the hardware interface communication standard to ensure that the data of devices such as each power supply end, energy storage system and smart meter can be smoothly connected to the platform. In terms of the interaction protocol, stipulate that the data transmission format is JSON, clarify the definition of data fields, data update frequency and abnormal data processing mechanism.

[0049] In step S4, the power supply data platform monitors the status of the energy storage system; when the power in a certain group of energy storage systems in the discharge state is lower than 20%, this group of energy storage systems stops discharging; when the power in a certain group of energy storage systems in the charge state exceeds 80%, this group of energy storage systems stops charging.

[0050] When the state of charge (SBC) of a group of energy storage systems in a discharging state falls below 20%, the power supply data platform will immediately send a stop-discharge command to that group of energy storage systems. For example, if the SBC of the first shared energy storage system drops below 20% during discharge, the platform will detect this data and instantly trigger the stop-discharge mechanism. The inverter of that group of energy storage systems will immediately disconnect the discharge circuit and stop supplying power to the load side, thus avoiding the impact on battery life due to over-discharge. At the same time, it can connect to other groups of energy storage systems or other microgrids to output power in real time.

[0051] When the state of charge (SBC) of a group of energy storage systems in the charging state exceeds 80%, the power supply data platform will also promptly issue a stop charging command. Taking the second group of shared energy storage systems as an example, during the charging process, the SBC rises from 78% to 82%. After detecting this, the platform quickly instructs the charging circuit of this group of energy storage systems to disconnect, stop receiving excess power, and prevent overcharging from damaging the battery.

[0052] The power supply data platform includes real-time monitoring and control of the voltage and frequency of the microgrid; when the voltage or frequency of the microgrid exceeds the normal range, it restores the voltage and frequency of the microgrid to the normal range by adjusting the output of distributed power sources, the charging and discharging of energy storage systems, and load demand.

[0053] In some possible embodiments, the microgrid load fluctuation suppression method further includes: S5. Establish a user-side electricity price adjustment mechanism; monitor users' electricity load in real time, and send electricity price adjustment signals to users when the microgrid load reaches peak or off-peak times to encourage users to reduce electricity consumption during peak hours and increase electricity consumption during off-peak hours.

[0054] When a microgrid needs to be connected to the main grid, the electricity price is adjusted based on the main grid's demand and the microgrid's surplus capacity. For example, when the microgrid has a large surplus load, with a surplus ratio > 50%, the price is 90%-100% of the base price to encourage the main grid to purchase and absorb the redundancy; when the surplus ratio is < 20%, the price is 120%-150% of the base price to suppress the main grid's purchase, thereby reducing unnecessary purchases by the main grid and ensuring emergency needs are met through higher prices.

[0055] Establishing a user-side electricity price adjustment mechanism can fully leverage the regulatory role of price, guide electricity consumption behavior from the user side, effectively alleviate load pressure during peak hours, balance load fluctuations in the microgrid, and synergize with previous steps through energy storage systems and power supply coordination to further enhance the stability of the microgrid.

[0056] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for suppressing load fluctuations in microgrids, characterized in that, Includes the following steps: S1. Data collection; collect rated output parameters and historical output data of the microgrid power supply end to generate power supply fluctuation prediction curves; collect historical and real-time power consumption data of the load end to generate load fluctuation prediction curves. S2. Establish a load forecasting model; dynamically adjust the forecasting results using load fluctuation forecasting curves and real-time load to reduce forecasting deviations. S3. Establish a microgrid complementary system; by coordinating the loads by complementing the power supply terminals of multiple microgrids with different power supply advantages during different power supply periods, and by using a shared energy storage system to achieve energy storage sharing, the load fluctuation of the microgrid can be reduced. S4. Energy storage system suppresses load fluctuations; based on the power balance status of the microgrid and the load forecast results, the charging and discharging power and charging and discharging time of the energy storage system are adjusted to smooth out load fluctuations.

2. The microgrid load fluctuation suppression method as described in claim 1, characterized in that, In step S1, the historical power output data of the microgrid power supply end covers the hourly power output value and corresponding meteorological condition records for each day over the past 12 months. Based on the above data, a power supply fluctuation prediction curve is generated using the time series decomposition method. Historical and real-time electricity consumption data are collected from the load side. Historical electricity consumption data includes daily time-of-day electricity consumption records for different user types over the past 12 months. Real-time electricity consumption data is collected from smart meters at 10-30 minute intervals to collect the current electricity load value. Combined with user type tags and time period characteristics, a load fluctuation prediction curve is generated.

3. The microgrid load fluctuation suppression method as described in claim 1, characterized in that, In step S2, the data from step S1 is analyzed and processed to establish a load forecasting model and predict the load changes of the microgrid over a future period; the prediction deviation threshold is 25% of the predicted load value. When the deviation between the real-time collected current power load value and the predicted load value exceeds the threshold, the dynamic adjustment mechanism is triggered; real-time power consumption data is re-collected, and the newly collected real-time power consumption data, meteorological data, and real-time power output data from the power supply end are input into the load prediction model to regenerate the load prediction curve.

4. The microgrid load fluctuation suppression method as described in claim 3, characterized in that, A deviation correction factor is introduced when obtaining the prediction deviation threshold; When real-time influencing factors exist, the prediction deviation threshold of the predicted load is first corrected using the deviation correction factor, and then the current electricity load value is compared with the corrected prediction deviation threshold of the predicted load.

5. The microgrid load fluctuation suppression method as described in claim 1, characterized in that, In step S3, based on the principle of complementary advantages during peak periods, at least two microgrid power supply terminals with staggered peak output characteristics are connected in a network to build a collaborative power supply network. Real-time data interaction and power transmission between the power supply terminals are realized through the energy management bus.

6. The microgrid load fluctuation suppression method as described in claim 5, characterized in that, A shared energy storage system should be installed in the power supply network, with the capacity of the shared energy storage system exceeding 20% ​​of the maximum load of the power supply network. Multiple shared energy storage systems can be set up, and each shared energy storage system is connected to at least one microgrid power supply terminal with peak-shifting output characteristics.

7. The microgrid load fluctuation suppression method as described in claim 1, characterized in that, In step S4, a power supply data platform is set up: microgrid power supply information, energy storage information, and electricity consumption information are collected, a data platform is established, and a unified data interface and interaction protocol are formulated. The power supply data platform monitors the distribution status of the microgrid in real time. If the predicted load is rising and the current power supply output is insufficient, the energy storage system will start discharging preparation 20 minutes in advance. If the predicted load is falling and the current power supply output is excessive, the energy storage system will start charging preparation 20 minutes in advance.

8. The microgrid load fluctuation suppression method as described in claim 7, characterized in that, In step S4, the power supply data platform monitors the status of the energy storage system; When the charge level of a certain energy storage system in a discharge state drops below 20%, that energy storage system stops discharging. When the charge level of a certain energy storage system exceeds 80% while it is charging, that energy storage system stops charging.

9. The microgrid load fluctuation suppression method as described in claim 7, characterized in that, The power supply data platform includes real-time monitoring and control of the voltage and frequency of the microgrid; When the voltage or frequency of a microgrid exceeds the normal range, the voltage and frequency can be restored to the normal range by adjusting the output of distributed power sources, the charging and discharging of energy storage systems, and load demand.

10. The microgrid load fluctuation suppression method as described in claim 7, characterized in that, Also includes: S5. Establish a user-side electricity price adjustment mechanism; monitor users' electricity load in real time, and send electricity price adjustment signals to users when the microgrid load reaches peak or off-peak times to encourage users to reduce electricity consumption during peak hours and increase electricity consumption during off-peak hours.

Images