Micro-grid power supply coordination control method, device and equipment

Abstract

The invention provides a micro-grid power supply coordination control method, device and equipment, and relates to the technical field of micro-grid power supply control. The method comprises the following steps: acquiring historical power consumption data of a micro-grid before a current moment, and predicting a power consumption trend curve of a future set time period based on the data and a preset power consumption prediction model; determining short-term fluctuation characteristics and long-term trend characteristics of electricity consumption through multiple ensemble empirical mode decomposition; based on the two types of characteristics, electric quantity gathering calculation and redistribution are carried out, and a coordination control scheme is determined. By means of the excellent decomposition capability of ensemble empirical mode decomposition on non-linear and non-stationary signals, refined distinguishing of loads is achieved, and a stable electricity utilization component and a fluctuating electricity utilization component are obtained; the operation stability and the power supply continuity of the main power grid are improved through a differential power supply mode that the main power grid specially supplies the stable component and the new energy and stored energy cooperatively undertakes the fluctuation component.

Description

Technical Field

[0001] This invention relates to the field of microgrid power supply control technology, and in particular to a method, apparatus and equipment for coordinated control of microgrid power supply. Background Technology

[0002] With the development of microgrids, such as the upgrading of industrial structure and diversification of functional business formats in industrial parks, the composition of electricity load is becoming increasingly complex, covering high-power production equipment, precision instruments, and intermittently operating supporting facilities. Moreover, the load exhibits typical characteristics of violent fluctuations and strong randomness. For example, the start and stop of production equipment, the start and stop of air conditioning systems depending on the weather, and the concentrated power consumption of temporary activities can all cause the load to fluctuate by several times in a short period of time, and the start and end time and intensity of the fluctuations are difficult to predict accurately.

[0003] Traditional microgrid power supply systems have long relied on the main grid for full load, lacking both accurate analysis and understanding of complex load characteristics and the construction of a diversified power supply system adapted to fluctuating loads. This results in various load impacts acting directly on the main grid without any buffer, requiring the main grid to continuously cope with frequent switching between overload and underload, leading to a sharp increase in power supply pressure. Ultimately, this causes a series of power supply instability problems such as voltage fluctuations, frequency instability, and frequent equipment failures in the main grid. This not only affects the production continuity of microgrid enterprises but also increases the operation and maintenance costs and emergency repair pressure of the main grid. Summary of the Invention

[0004] This invention provides a microgrid power supply coordination control method, apparatus, and equipment to solve the problem of instability in the main power grid caused by random load impacts.

[0005] In a first aspect, embodiments of the present invention provide a microgrid power supply coordination control method, comprising: acquiring historical electricity consumption data of the microgrid to be coordinated up to the current moment; predicting the electricity consumption trend curve of the microgrid to be coordinated within a future set period after the current moment based on the historical electricity consumption data and a preset electricity consumption prediction model; performing multiple ensemble empirical mode decompositions based on the electricity consumption trend curve to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within the future set period; and performing electricity aggregation calculation and electricity redistribution based on the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption to determine a coordination control scheme for the microgrid to be coordinated, wherein the coordination control scheme includes a stable power supply from the main grid and a coordinated power supply from a combination of new energy sources and energy storage systems.

[0006] In one possible implementation, based on the electricity consumption trend curve, multiple ensemble empirical mode decompositions are performed to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set period. This includes: performing multiple ensemble empirical mode decompositions on the electricity consumption trend curve, with each decomposition yielding multiple-order original intrinsic mode functions and one original residual component; calculating the average value of the original intrinsic mode functions of the same order from the multiple decompositions to obtain stable multiple-order intrinsic mode functions; calculating the average value of all original residual components to obtain stable residual components; wherein, stable intrinsic mode functions with frequencies higher than a preset high-frequency threshold are used to aggregate to obtain short-term fluctuation characteristics, and stable intrinsic mode functions with frequencies lower than a preset low-frequency threshold are used to merge with stable residual components to obtain long-term trend characteristics.

[0007] In one possible implementation, based on the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption, electricity consumption is aggregated and redistributed to determine the coordinated control scheme of the microgrid to be coordinated. This includes: based on the short-term fluctuation characteristics, electricity consumption is aggregated and calculated to determine the fluctuating electricity consumption component; based on the long-term trend characteristics, electricity consumption is aggregated and calculated to determine the stable electricity consumption component; and based on the fluctuating and stable electricity consumption components, electricity is redistributed to determine the coordinated control scheme of the microgrid to be coordinated.

[0008] In one possible implementation, based on short-term fluctuation characteristics, electricity consumption is aggregated and calculated to determine the fluctuating electricity consumption component. This includes: extracting the original intrinsic mode functions corresponding to the short-term fluctuation characteristics from the results of multiple ensemble empirical mode decompositions, and sorting them from high to low frequency to obtain the sorted results; calculating the average value of the original intrinsic mode functions with the same index in the sorted results to obtain stable high-frequency intrinsic mode functions; optimizing the stable high-frequency intrinsic mode functions to obtain optimized stable high-frequency intrinsic mode functions; and superimposing the optimized stable high-frequency intrinsic mode functions to obtain the fluctuating electricity consumption component.

[0009] In one possible implementation, the stable high-frequency intrinsic mode functions (HIMFs) are optimized to obtain optimized stable high-frequency HIMFs. This includes: calculating the Pearson correlation coefficient between each stable high-frequency HIMF and the electricity consumption trend curve to obtain a correlation score for each stable high-frequency HIMF; classifying the stable high-frequency HIMFs into high-frequency, mid-frequency, and low-frequency groups based on a preset period threshold; performing initial threshold screening for the stable high-frequency HIMFs in the high-frequency, mid-frequency, and low-frequency groups based on a preset threshold to obtain initial screening results for the stable high-frequency HIMFs in the high-frequency, mid-frequency, and low-frequency groups; and performing refined screening for the stable high-frequency HIMFs in the high-frequency, mid-frequency, and low-frequency groups based on the initial screening results and ranking them according to a preset percentage to obtain refined screening results for the stable high-frequency HIMFs in the high-frequency, mid-frequency, and low-frequency groups.

[0010] In one possible implementation, before performing multiple ensemble empirical mode decompositions based on the electricity consumption trend curve to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set period, the method further includes: calculating the standard deviation and average value of the electricity consumption trend curve; calculating the fluctuation coefficient of the electricity consumption trend curve based on the standard deviation and average value of the electricity consumption trend curve; determining the amplitude of Gaussian white noise to be added to the electricity consumption trend curve based on the fluctuation coefficient of the electricity consumption trend curve; determining the Gaussian white noise to be added to the electricity consumption trend curve based on a preset multiple and the amplitude; and adding the preset multiple of Gaussian white noise to the electricity consumption trend curve.

[0011] In one possible implementation, based on long-term trend characteristics, electricity consumption is aggregated and calculated to determine the stable electricity consumption component. This includes: calculating the trend slope of the long-term trend characteristics using the sliding window method; calculating the absolute value of the trend slope of all windows within a preset time period; if the maximum absolute value is less than a preset trend slope threshold, the long-term trend characteristics are determined as the stable electricity consumption component; if the maximum absolute value is greater than or equal to the preset trend slope threshold, the amplitude and preset multiple of the Gaussian white noise are readjusted, and ensemble empirical mode decomposition is performed again.

[0012] In one possible implementation, power redistribution is performed based on fluctuating and stable power consumption components to determine the coordinated control scheme for the microgrid to be coordinated. This includes: for high-frequency fluctuating power consumption components, priority is given to distributed photovoltaic and distributed wind power, with lithium battery energy storage systems providing supplementary power; for medium-frequency fluctuating power consumption components, priority is given to distributed photovoltaic and distributed wind power, with gas turbines and supercapacitors providing supplementary power; and for low-frequency fluctuating power consumption components, power is supplied through expanded distributed photovoltaic and distributed wind power, and long-term energy storage systems working together.

[0013] Secondly, embodiments of the present invention provide a microgrid power supply coordination and control device, comprising: a communication module for acquiring historical electricity consumption data of the microgrid to be coordinated up to the current moment; a processing module for predicting, based on the historical electricity consumption data and a preset electricity consumption prediction model, the electricity consumption trend curve of the microgrid to be coordinated within a future set period after the current moment; performing multiple ensemble empirical mode decompositions based on the electricity consumption trend curve to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within the future set period; and performing electricity aggregation calculation and electricity redistribution based on the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption to determine a coordination and control scheme for the microgrid to be coordinated, wherein the coordination and control scheme includes a stable power supply from the main grid and a coordinated power supply from a combination of new energy sources and energy storage systems.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0015] In this embodiment of the invention, based on historical electricity consumption data and a preset electricity consumption prediction model, an electricity consumption trend curve for a preset future period after the current moment of the microgrid to be coordinated is generated. Then, through multiple ensemble empirical mode decompositions, the algorithm's excellent decomposition capability for nonlinear and non-stationary signals is utilized to separate the long-term trend characteristics and short-term fluctuation characteristics of electricity consumption. Furthermore, based on the two types of characteristics, the stable electricity consumption component corresponding to the stable load and the fluctuating electricity consumption component corresponding to the short-term random and periodic load are determined respectively, achieving refined differentiation of the load. Finally, a differentiated power supply mode is adopted, in which the main grid provides the stable electricity consumption component and distributed new energy and energy storage work together to undertake the fluctuating component, thereby cutting off the direct impact of random loads on the main grid from the root and improving the operational stability and power supply continuity of the main grid. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the implementation of the microgrid power supply coordination control method provided in this embodiment of the invention. Figure 2 This is a schematic diagram of the microgrid power supply coordination and control device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] See Figure 1 The flowchart illustrating the implementation of the microgrid power supply coordination control method provided in this embodiment of the invention is described in detail below: Step 101: Obtain the historical electricity consumption data of the microgrid to be coordinated up to the current moment.

[0019] In some embodiments, historical electricity consumption data refers to various records related to electricity consumption of a microgrid to be coordinated, such as the park's electricity consumption data over a period of time prior to the current moment. These data are an objective reflection of the microgrid's past electricity consumption behavior and can reflect the historical patterns and characteristics of the microgrid's electricity consumption.

[0020] In this embodiment, historical electricity consumption data typically includes time-based data, load type data, associated environmental data, and power interaction data.

[0021] For example, time-dimensional data includes electricity consumption records at different time granularities, such as minute-level, hour-level, daily-level, weekly-level, and monthly-level electricity consumption data, which are used to reflect the changing patterns of electricity consumption over time.

[0022] For example, load type data is electricity consumption data divided according to the main electricity users in the microgrid, such as the electricity consumption of different types of loads such as industrial production equipment, commercial office facilities, public lighting, and residential life, which can reflect the electricity consumption characteristics of different scenarios.

[0023] For example, the associated environmental data are external environmental parameters that affect electricity consumption, such as historical temperature, humidity, weather conditions (sunny, rainy, snowy, etc.), seasonal characteristics, holiday information, etc. These factors are usually strongly correlated with electricity demand, such as the increase in air conditioning load as the temperature rises in summer.

[0024] For example, power interaction data includes historical power interaction between the microgrid and the main grid (such as power purchase and sales), historical power generation of new energy sources (photovoltaics, wind power, etc.), and historical charging and discharging of energy storage systems, which are used to reflect the matching relationship between historical power supply structure and power consumption.

[0025] As one possible implementation, embodiments of the present invention obtain historical electricity consumption data to provide raw data for analysis and processing in subsequent steps.

[0026] Step 102: Based on historical electricity consumption data and a preset electricity consumption prediction model, predict the electricity consumption trend curve of the microgrid to be coordinated in the future within a set period after the current moment.

[0027] In some embodiments, the electricity consumption prediction model is a data-driven time series prediction model that learns the patterns contained in historical electricity consumption data to quantitatively predict electricity consumption for a future set period (such as the next 24 hours, 7 days, etc.) and outputs a continuous electricity consumption trend curve.

[0028] In some embodiments, the electricity consumption forecasting model is used to provide a predictive basis for subsequent coordinated control: by predicting future electricity consumption trends, the power supply strategies of the main power grid, new energy sources, and energy storage systems can be planned in advance to avoid power surplus or shortage and improve energy utilization efficiency. It also assists in identifying peak / valley periods of electricity consumption: the forecast results can help determine the timing and magnitude of future peak and valley electricity loads, providing support for power dispatching (such as selecting the timing of energy storage charging and discharging, and matching the output of new energy sources).

[0029] In some embodiments, the electricity consumption prediction model is obtained by training a neural network based on historical electricity consumption data and historical electricity consumption trend curves.

[0030] In some embodiments, the electricity consumption trend curve is the output of the electricity consumption prediction model. It is a continuous curve with time as the horizontal axis and predicted electricity consumption as the vertical axis, which intuitively reflects the overall trend and fluctuation characteristics of electricity consumption over a set period of time in the future for the microgrid to be coordinated. The electricity consumption trend curve includes time axis information, electricity consumption values, and trend characteristics.

[0031] For example, timeline information clearly indicates the future time period covered by the curve (such as the next 24 hours or the next 7 days), and the horizontal axis scale is usually consistent with the time granularity of historical data (such as one data point per hour for hourly curves).

[0032] For example, the vertical axis of electricity consumption represents the predicted electricity consumption (the unit is the same as the historical data), and each point in time corresponds to a predicted value, reflecting the electricity demand at that moment.

[0033] For example, trend features include the overall direction of the curve (such as an upward trend, a downward trend, or a stable trend), periodic fluctuations (such as differences in electricity consumption between weekdays and weekends, or intraday peak-valley patterns), and sudden fluctuation points (such as possible temporary high-load events). These features are the direct basis for subsequent steps (such as ensemble empirical mode decomposition) to extract short-term fluctuations and long-term trends.

[0034] As one possible implementation, step 102 can be specifically implemented as steps 1021-1024.

[0035] Step 1021: Preprocess the historical electricity consumption data.

[0036] For example, preprocessing includes cleaning historical electricity consumption data to remove outliers and missing values; standardizing or normalizing the data to eliminate interference from data of different magnitudes on the model; and extracting time features and related influencing factors to construct the model input feature set.

[0037] Step 1022: Determine the forecast period and model parameters.

[0038] For example, the prediction period is the specific duration of a future set period. The model parameters are the input and output dimensions of a preset electricity consumption prediction model, such as the hourly / minute electricity consumption for the future set period.

[0039] Step 1023: Input the processed historical data into the prediction model.

[0040] For example, select the latest historical data segment before the current moment. For instance, when predicting electricity consumption for the next 24 hours, input the electricity consumption data and related characteristics for the same period over the past 7 days, organizing the input according to the format required by the model. Run the prediction model, and through forward propagation calculations using a neural network, output the predicted electricity consumption value for each time point within the set future time period.

[0041] Step 1024: Generate electricity consumption trend curve.

[0042] For example, by plotting time (e.g., hours, minutes) on the horizontal axis and the predicted electricity consumption value on the vertical axis, the discrete predicted values ​​output by the model are concatenated into a continuous curve. The curve is then smoothed (e.g., by using a moving average) to eliminate minor oscillations in the model predictions, resulting in an electricity consumption trend curve that reflects the overall trend of future electricity consumption changes.

[0043] As one possible implementation, embodiments of the present invention can purify and standardize historical data, extract key features, and provide accurate and standardized input for the prediction model. By clearly defining the prediction period and model parameters, the prediction target is adapted to the needs of coordinated control, ensuring model controllability. Through a neural network model, future electricity consumption data is extrapolated based on historical electricity consumption patterns, achieving the core calculation from history to the future. Discrete predicted values ​​are transformed into smooth trend curves, highlighting electricity consumption patterns and facilitating subsequent feature extraction and analysis.

[0044] In this embodiment, past electricity consumption patterns are transformed into future electricity consumption trends through data modeling. This process requires combining the electricity consumption characteristics of the microgrid to complete data preprocessing, model training, prediction correction, and curve output, providing a basis for subsequent steps.

[0045] Step 103: Based on the electricity consumption trend curve, perform multiple ensemble empirical mode decompositions to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set time period.

[0046] In some embodiments, ensemble empirical mode decomposition is an adaptive data decomposition method for processing nonlinear and non-stationary signals. By adding Gaussian white noise multiple times to the original signal and then performing multiple empirical mode decompositions (EMDs), the noise interference is finally suppressed by averaging, and the complex signal is decomposed into a series of physically meaningful intrinsic mode functions (IMFs) and a residual component (reflecting the overall trend of the signal).

[0047] In some embodiments, ensemble empirical mode decomposition includes multiple empirical mode decompositions, Gaussian white noise, and averaging.

[0048] For example, in multiple empirical mode decompositions, each decomposition separates the signal layer by layer into intrinsic mode functions (IMFs) and residual components of different frequencies. The IMFs must satisfy the conditions that the mean of the upper and lower envelopes is zero, and the number of extrema and zero-crossings is equal or differs by at most 1.

[0049] For example, Gaussian white noise solves the modal aliasing problem that may occur in traditional EMD (i.e., a single IMF contains features of different scales) by adding different Gaussian white noise to the original signal.

[0050] For example, averaging is used to average the IMFs and residual components of the same order obtained from multiple decompositions to obtain stable decomposition results (the final multi-order IMFs and residual components).

[0051] As one possible implementation, embodiments of the present invention can decompose the electricity consumption trend curve for a future set period into IMFs and residual components of different frequencies, achieving precise separation of short-term fluctuation characteristics (high-frequency IMF aggregation) and long-term trend characteristics (low-frequency IMF and residual component merging). By clarifying the fluctuation patterns of electricity consumption (such as short-term sudden fluctuations) and the overall trend (such as long-term stable demand), characteristic support is provided for subsequent electricity consumption allocation (the main grid undertakes the stable component, and new energy and energy storage work together to cope with the fluctuation component), ensuring that the power supply scheme is more in line with actual electricity consumption characteristics.

[0052] In some embodiments, short-term fluctuation characteristics refer to the rapid and high-frequency changes in electricity consumption within a short period of time (such as minutes or hours) in the future, reflecting the instantaneous fluctuation pattern of electricity load.

[0053] In some embodiments, long-term trend characteristics refer to the overall trend of electricity consumption over a longer time scale (such as daily, weekly, or monthly) within a future set period, reflecting the basic demand and macroeconomic patterns of electricity load.

[0054] As one possible implementation, embodiments of the present invention can determine the basic load and macroscopic patterns of microgrid electricity consumption, providing a basis for the stable power supply planning of the main grid. Such trend-based demands need to be borne by the main grid with high power supply stability to ensure the continuous satisfaction of core electricity demand.

[0055] In one possible implementation, step 103 can be specifically implemented as steps 1031-1033.

[0056] Step 1031: Perform multiple ensemble empirical mode decompositions on the electricity consumption trend curve. Each decomposition yields multiple original intrinsic mode functions and one original residual component.

[0057] Step 1032: Calculate the average value of the original intrinsic mode functions of the same order after multiple decompositions to obtain stable multi-order intrinsic mode functions.

[0058] Step 1033: Calculate the average value of all original residual components to obtain stable residual components; Among them, stable intrinsic mode functions with frequencies higher than a preset high-frequency threshold are used to aggregate and obtain short-term fluctuation characteristics, while stable intrinsic mode functions with frequencies lower than a preset low-frequency threshold are used to merge with stable residual components to obtain long-term trend characteristics.

[0059] In some embodiments, the original intrinsic mode function (IMF) is an oscillating component with specific frequency characteristics that is separated layer by layer from the signal during a single empirical mode decomposition (EMD) of the electricity consumption trend curve, and it is the basic unit constituting the original signal. The original IMF includes high-frequency, mid-frequency, and low-frequency IMFs. The high-frequency IMF corresponds to rapid, short-term fluctuations in the signal, such as minute-level electricity consumption jumps; the mid-frequency IMF corresponds to medium-period fluctuations, such as hourly electricity consumption rhythm changes; and the low-frequency IMF corresponds to slow, long-period fluctuations, such as intraday electricity consumption peak-valley trends.

[0060] As one possible implementation, the embodiments of the present invention serve as the basic component for signal decomposition. The original IMF reflects the fluctuation characteristics of different time scales in the electricity consumption trend curve. After averaging the original IMFs of the same order obtained from multiple decompositions, the randomness of a single decomposition can be eliminated, forming a stable intrinsic mode function, providing raw data for subsequent differentiation between short-term fluctuation characteristics (high-frequency aggregation) and long-term trend characteristics (low-frequency merging).

[0061] In some embodiments, the original residual component is the remaining part that cannot be further decomposed into the IMF by a single empirical mode decomposition (EMD), reflecting the overall baseline or macro trend of the electricity consumption trend curve. The original residual component mainly reflects the long-term trend of the signal, such as: the base load level of microgrid electricity consumption (e.g., the lowest electricity consumption baseline within 24 hours), the overall increasing / decreasing trend (e.g., electricity consumption growth with the expansion of production scale), and long-period patterns (e.g., seasonal differences in electricity consumption baseline). Its characteristics are slow change and no obvious oscillation, making it the most stable component in the signal.

[0062] As one possible implementation, the embodiment of the present invention uses the original residual component as the core component reflecting the overall trend of electricity consumption. After multiple decompositions and averaging, the original residual component is merged with the low-frequency intrinsic mode function to form a long-term trend feature, which provides a macro-based basis for determining the stable power supply of the main grid and ensures that the power supply scheme conforms to the basic needs and overall laws of microgrid electricity consumption.

[0063] Step 104: Based on the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption, perform electricity aggregation calculation and electricity redistribution to determine the coordination control scheme of the microgrid to be coordinated. The coordination control scheme includes the stable power supply from the main grid and the collaborative power supply from the joint power supply of new energy and energy storage systems.

[0064] In one possible implementation, step 104 can be specifically implemented as steps 1041-1043.

[0065] Step 1041: Based on the short-term fluctuation characteristics, perform a summary calculation of electricity consumption to determine the fluctuating electricity consumption component; Step 1042: Based on long-term trend characteristics, perform electricity consumption summary calculations to determine the stable electricity consumption component; Step 1043: Based on the fluctuating power consumption component and the stable power consumption component, perform power redistribution to determine the coordination control scheme for the microgrid to be coordinated.

[0066] In some embodiments, the fluctuating electricity consumption component is a quantitative value that reflects the scale of short-term electricity consumption fluctuations in a microgrid, obtained by summarizing and calculating electricity consumption based on short-term fluctuation characteristics.

[0067] In some embodiments, the stable electricity consumption component is a quantitative value that reflects the long-term basic electricity consumption scale of the microgrid after the electricity consumption is aggregated and calculated based on long-term trend characteristics.

[0068] In some embodiments, power redistribution refers to the process of scientifically allocating the total power consumption of a microgrid to different types of power supply resources based on the specific values ​​of fluctuating power consumption components and stable power consumption components.

[0069] In some embodiments, the coordination control scheme, after power redistribution, ultimately forms a specific strategy that guides the coordinated operation of various power supply resources, which is the final output of the entire microgrid power optimization.

[0070] As one possible implementation, embodiments of the present invention can bridge the gap between electricity consumption characteristics and power supply strategies by transforming short-term fluctuation characteristics and long-term trend characteristics into fluctuating electricity consumption components and stable electricity consumption components, respectively. This allows abstract electricity consumption patterns to be directly applied to power allocation. It clarifies that different types of power sources will handle different electricity consumption components (e.g., stable components are guaranteed by the main grid, while fluctuating components are addressed collaboratively by new energy sources and energy storage), ensuring a match between power supply resources and electricity demand characteristics, and improving power supply efficiency and stability. By calculating specific component values ​​through electricity aggregation, the power supply scheme shifts from qualitative analysis to quantitative decision-making, ensuring that the output planning of the main grid, new energy sources, and energy storage systems is operable and executable, ultimately achieving optimized coordination of microgrid power supply.

[0071] As one possible implementation, step 1041 can be specifically implemented as steps A11-A14.

[0072] A11: Extract the original intrinsic mode functions corresponding to the short-term fluctuation characteristics from the results of multiple ensemble empirical mode decompositions, and sort them from high to low frequency to obtain the sorted results.

[0073] A12: Calculate the average value of the original intrinsic mode functions with the same index in the sorting results to obtain stable high-frequency intrinsic mode functions.

[0074] A13: Optimize the stable high-frequency intrinsic mode function to obtain the optimized stable high-frequency intrinsic mode function.

[0075] A14: The optimized and stable high-frequency intrinsic mode functions are superimposed to obtain the fluctuating power consumption component.

[0076] In some embodiments, the original intrinsic mode function corresponding to the short-term fluctuation characteristics refers to the original intrinsic mode function that represents the short-term fluctuation pattern of electricity consumption, selected from all decomposition results after multiple EEMD decompositions.

[0077] In some embodiments, averaging the original intrinsic mode functions (IMFs) with the same index is a key step in improving data stability. Specifically, since EEMD involves multiple decompositions (e.g., 50 decompositions), each decomposition yields a set of original IMFs sorted by frequency from high to low. For example, the first decomposition yields sorted IMF1-1, IMF1-2, ...; the second decomposition yields IMF2-1, IMF2-2, ... . The original IMFs with the same index from all decomposition results (e.g., IMF1-1 from the first decomposition, IMF2-1 from the second decomposition, ..., IMF50-1 from the 50th decomposition, all being the highest-frequency IMFs in each decomposition) are taken, and their arithmetic mean is calculated. Because of slight differences in the noise introduced in each EEMD decomposition, the original IMFs may exhibit small fluctuations (i.e., instability). By averaging the original IMFs with the same index, noise interference can be offset, resulting in more reliable fluctuation characteristics.

[0078] In some embodiments, a stable high-frequency intrinsic mode function is obtained by averaging the original IMFs with the same index, eliminating random noise interference. Compared to the original IMFs, its fluctuation pattern is more stable and more accurately reflects the inherent short-term fluctuations of microgrid electricity consumption (rather than spurious fluctuations caused by noise).

[0079] In some embodiments, stable high-frequency intrinsic mode function (IMF) optimization involves further processing of the stable high-frequency IMF to remove invalid information and enhance effective fluctuation characteristics. This further refines the stable high-frequency IMF, retaining only components that are meaningful for describing short-term electricity consumption fluctuations, thus reducing the interference of invalid data on the final result.

[0080] In some embodiments, the optimized and stable high-frequency intrinsic mode function is the high-frequency intrinsic mode function that is ultimately retained after optimization steps and accurately reflects the core characteristics of short-term power consumption fluctuations in the microgrid.

[0081] As one possible implementation, embodiments of the present invention can eliminate the randomness of a single decomposition by sorting the original intrinsic mode functions (IMFs) from multiple ensemble empirical mode decompositions by frequency and taking the average value, thus obtaining stable high-frequency IMFs and ensuring more reliable extracted fluctuation characteristics. By screening and optimizing the high-frequency IMFs, irrelevant or interfering fluctuations are eliminated, retaining the core components that truly reflect dynamic changes in electricity consumption, improving the accuracy of fluctuating electricity consumption components. The final superimposed fluctuating electricity consumption components clearly quantify short-term electricity consumption fluctuations at different frequencies (such as high-frequency, medium-frequency, and low-frequency fluctuations), providing a specific quantitative target for the coordinated power supply strategy of new energy and energy storage systems, ensuring a more accurate and efficient response to short-term electricity consumption fluctuations.

[0082] As one possible implementation, step A13 can be specifically implemented as steps B11-B14.

[0083] B11: Calculate the Pearson correlation coefficient between each stable high-frequency intrinsic mode function and the electricity consumption trend curve to obtain the correlation score of each stable high-frequency intrinsic mode function.

[0084] B12: Based on a preset period threshold, stable high-frequency intrinsic mode functions are divided into high-frequency band group, mid-frequency band group and low-frequency band group.

[0085] B13: For stable high-frequency intrinsic mode functions of the high-frequency band, mid-frequency band, and low-frequency band, a threshold screening is performed based on a preset threshold to obtain the preliminary screening results of stable high-frequency intrinsic mode functions of the high-frequency band, mid-frequency band, and low-frequency band.

[0086] B14: Based on the initial screening results, the stable high-frequency intrinsic mode functions of the high-frequency band, mid-frequency band, and low-frequency band are ranked and finely screened according to a preset percentage to obtain the fine screening results of the stable high-frequency intrinsic mode functions of the high-frequency band, mid-frequency band, and low-frequency band.

[0087] In some embodiments, a stable high-frequency intrinsic mode function refers to a stable fluctuation component obtained by averaging the original high-frequency intrinsic mode functions of the same order after multiple ensemble empirical mode decompositions (EEMD). It reflects the high-frequency fluctuation characteristics in electricity consumption trends and is the basic unit constituting short-term electricity consumption fluctuations.

[0088] In some embodiments, the preset period threshold refers to a manually set critical value (such as time length) used to divide the fluctuation period, and is used to distinguish between high-frequency, medium-frequency, and low-frequency fluctuations. For example, a period of <1 hour can be set as the high-frequency band, 1-6 hours as the medium-frequency band, and 6-24 hours as the low-frequency band (the specific threshold is determined according to the power consumption characteristics of the microgrid).

[0089] In some embodiments, the Pearson correlation coefficient is a statistical indicator that measures the degree of linear correlation between two variables (in this case, a stable high-frequency intrinsic mode function and an electricity consumption trend curve), and its value ranges from [-1, 1]. The closer the absolute value is to 1, the stronger the correlation between the two variables; the closer it is to 0, the weaker the correlation.

[0090] In some embodiments, the correlation score, i.e. the calculation result of the Pearson correlation coefficient, is used to quantify the degree of correlation between a single stable high-frequency intrinsic mode function and the original electricity consumption trend curve. The higher the score, the more significant the impact of the fluctuation component on the actual electricity consumption change.

[0091] In some embodiments, preset thresholds, such as a first benchmark threshold, a second benchmark threshold, and a third benchmark threshold, are correlation score thresholds set for different frequency band groups (high frequency, mid frequency, and low frequency). For example, a higher threshold (e.g., 0.6) is set for the high frequency band group, and a medium threshold (e.g., 0.4) is set for the mid frequency band group, to initially filter out invalid fluctuation components with excessively low correlation.

[0092] In some embodiments, the preset percentage is a threshold for selecting the top N% (e.g., top 80%) of the retained fluctuation components after initial screening, sorted by relevance score. This ranking-based fine screening further eliminates components with relatively low relevance, retaining the core fluctuation characteristics.

[0093] In some embodiments, the initial screening results and the refined screening results refer to the sets of stable high-frequency intrinsic mode functions retained after threshold screening and percentage ranking screening, respectively. The refined screening results are the final purification of the fluctuation components, ensuring that they can accurately reflect the key power consumption fluctuation characteristics of different frequency bands and provide a reliable basis for subsequent power dispatching.

[0094] As one possible implementation, embodiments of the present invention can divide stable high-frequency intrinsic mode functions into high-frequency, mid-frequency, and low-frequency groups by setting a preset periodic threshold. This clearly distinguishes short-term fluctuations at different time scales (such as minute-level instantaneous jumps and hourly-level rhythm changes), laying a classification foundation for matching power sources with different response speeds (such as high-frequency fluctuations corresponding to supercapacitors and mid-frequency fluctuations corresponding to gas turbines). By calculating the Pearson correlation coefficient between each high-frequency intrinsic mode function and the original electricity consumption trend curve, fluctuation components with strong correlation to actual electricity consumption changes are screened out, while irrelevant fluctuations caused by noise, decomposition errors, etc., are eliminated to avoid invalid fluctuation data interfering with subsequent power supply strategy formulation. Through a dual screening logic of threshold initial screening and percentage ranking fine screening, the fluctuation components of different frequency band groups are further purified to ensure that the final retained high-frequency intrinsic mode functions not only meet the fluctuation characteristic classification criteria but also truly reflect the dynamic changes in electricity demand. This makes the optimized fluctuation electricity consumption components more realistic, providing high-quality data support for the precise coordinated scheduling of new energy and energy storage systems (such as charging and discharging timing and output control).

[0095] As one possible implementation, step B13 can be specifically implemented as steps C11-C13.

[0096] C11: For the high-frequency band group, remove stable high-frequency intrinsic mode functions whose correlation scores are below the first benchmark threshold.

[0097] C12: For the mid-frequency band group, remove stable high-frequency intrinsic mode functions with correlation scores below the second benchmark threshold.

[0098] C13: For the low-frequency group, remove stable high-frequency intrinsic mode functions with correlation scores below the third benchmark threshold.

[0099] As one possible implementation, embodiments of the present invention can set a first benchmark threshold, a second benchmark threshold, and a third benchmark threshold for high-frequency, mid-frequency, and low-frequency bands respectively, and delete components with correlation scores lower than the corresponding thresholds. This can filter out noise or decomposition error components that are weakly correlated with the original electricity consumption trend, avoiding interference from these invalid data in the judgment of real electricity consumption fluctuations. Different frequency band fluctuations have different requirements for power supply response (e.g., high-frequency fluctuations require rapid response, while low-frequency fluctuations require continuous adjustment). Setting thresholds for each frequency band can better reflect the physical meaning of each band. For example, the high-frequency band threshold can be set higher to strictly retain components strongly correlated with instantaneous electricity consumption changes; the low-frequency band threshold can be appropriately relaxed to ensure that all meaningful slow fluctuations are captured, making the screening results more consistent with the actual functional requirements of each frequency band. After screening by frequency band thresholds, the inherent mode functions retained in each frequency band group are all highly correlated with the electricity consumption trend, and the superimposed fluctuating electricity consumption components can more realistically reflect the electricity consumption fluctuation patterns at different time scales.

[0100] As one possible implementation, step B14 can be specifically implemented as steps D11-D14.

[0101] D11: The stable high-frequency intrinsic mode functions of the high-frequency, mid-frequency, and low-frequency groups are ranked from high to low according to their correlation scores.

[0102] D12: For the high-frequency band group, delete the stable high-frequency intrinsic mode functions that rank last in the correlation score and are at the first preset percentage.

[0103] D13: For the mid-frequency band group, delete the stable high-frequency intrinsic mode functions that are ranked second to last in the correlation score by the preset percentage.

[0104] D14: For the low-frequency band group, the stable high-frequency intrinsic mode functions that rank third from the bottom of the correlation score are deleted to obtain the optimized stable high-frequency intrinsic mode functions.

[0105] As one possible implementation, embodiments of the present invention can further improve the quality of high-frequency IMFs by ranking and eliminating components based on frequency bands, on the basis of initial threshold screening. This ensures that the retained components are the most core and relevant fluctuation features in each frequency band, providing a more reliable foundation for the accurate calculation of subsequent fluctuation power consumption components and optimization of power supply strategies.

[0106] As one possible implementation, step 1042 can be specifically implemented as steps E11-E14.

[0107] E11: Calculate the trend slope of long-term trend characteristics using the sliding window method.

[0108] E12: The absolute value of the trend slope of all windows within a preset time period.

[0109] E13: If the maximum absolute value is less than the preset trend slope threshold, the long-term trend feature will be determined as the stable electricity consumption component.

[0110] E14: If the maximum absolute value is greater than or equal to the preset trend slope threshold, readjust the amplitude and preset multiple of the Gaussian white noise, and re-perform ensemble empirical mode decomposition.

[0111] In some embodiments, the sliding window method is a common approach for local analysis of time series data. By setting a fixed-length window and letting it slide sequentially along the time axis, indicators are calculated separately for the data within each window.

[0112] In some embodiments, long-term trend characteristics refer to the overall direction of change of time series data (such as electricity consumption) over a longer time scale (such as several months or several years), and are not affected by short-term random fluctuations (such as sudden power outages on a certain day or changes in electricity consumption caused by abnormal weather in a certain week).

[0113] In some embodiments, the trend slope is a key indicator for quantifying long-term trend characteristics. Essentially, it is the rate of change of time series data within a window. By linearly fitting the electricity consumption data within the window (drawing a straight line that best represents the change of data in that window), the slope of this straight line is the trend slope.

[0114] In some embodiments, the trend slope threshold is a pre-set slope value standard used to determine whether the long-term trend characteristics are stable. When the absolute value of the trend slope does not exceed the threshold, the trend is considered to be stable; if it exceeds the threshold, the trend is considered to be too volatile and does not meet the stability requirements.

[0115] In some embodiments, Gaussian white noise is a type of random interference signal commonly used in signal processing. It has two key characteristics: Gaussian distribution, in which the probability distribution of its values ​​(interference intensity) conforms to a normal distribution (the interference is weak most of the time and strong only a very few times); and white noise, in which the interference at different time points is independent of each other and has no discernible pattern (similar to the hissing sound of a radio when there is no signal).

[0116] As one possible implementation, embodiments of the present invention can calculate the trend slope of the long-term trend feature using the sliding window method and calculate its absolute value to determine whether the long-term trend is in a stable state (the absolute value of the slope is less than a preset threshold). If the condition is met, the feature is confirmed as a stable power consumption component, ensuring the reliability of the main grid's power supply planning. If the absolute value of the trend slope exceeds the threshold, it indicates that the long-term trend feature obtained by the current decomposition still has significant fluctuations, and the amplitude and preset multiple of the Gaussian white noise need to be readjusted, and ensemble empirical mode decomposition needs to be performed again. This mechanism can correct deviations in the decomposition process, avoid including unstable components in the long-term trend, and ensure the accuracy of the stable power consumption component. By rigorously verifying the stability of the long-term trend feature, it ensures that the stable power supply undertaken by the main grid is accurately matched with the actual basic power demand of the microgrid, avoiding power shortages or resource waste caused by misjudgments of the long-term trend, and providing a fundamental guarantee for the effectiveness of the entire coordinated control scheme.

[0117] As one possible implementation, step 1043 can be specifically implemented as steps F11-F13.

[0118] F11: For high-frequency fluctuating power consumption components, priority is given to power supply from distributed photovoltaic and distributed wind power, with lithium battery energy storage systems providing supplementary power.

[0119] F12: For the fluctuating power consumption component of medium frequency, distributed photovoltaic and distributed wind power are given priority for power supply, with gas turbines and supercapacitors supplementing the power supply.

[0120] F13: For low-frequency fluctuating power consumption components, utilize the coordinated power supply of expanded distributed photovoltaic, expanded distributed wind power, and long-term energy storage systems.

[0121] As one possible implementation, the embodiments of the present invention can solve the contradiction between different fluctuation characteristics and power supply equipment through a combination strategy of high-frequency fast response, medium-frequency stable and continuous operation, and low-frequency large capacity equipment, thereby maximizing the utilization of new energy sources, reducing costs, and ensuring power supply stability.

[0122] In this embodiment of the invention, based on historical electricity consumption data and a preset electricity consumption prediction model, an electricity consumption trend curve for a preset future period after the current moment of the microgrid to be coordinated is generated. Then, through multiple ensemble empirical mode decompositions, the algorithm's excellent decomposition capability for nonlinear and non-stationary signals is utilized to separate the long-term trend characteristics and short-term fluctuation characteristics of electricity consumption. Furthermore, based on the two types of characteristics, the stable electricity consumption component corresponding to the stable load and the fluctuating electricity consumption component corresponding to the short-term random and periodic load are determined respectively, achieving refined differentiation of the load. Finally, a differentiated power supply mode is adopted, in which the main grid provides the stable electricity consumption component and distributed new energy and energy storage work together to undertake the fluctuating component, thereby cutting off the direct impact of random loads on the main grid from the root and improving the operational stability and power supply continuity of the main grid.

[0123] In one possible implementation, G11-G15 are also included before step 103.

[0124] G11: Calculate the standard deviation and mean of the electricity consumption trend curve.

[0125] G12: Calculate the fluctuation coefficient of the electricity consumption trend curve based on the standard deviation and average value of the electricity consumption trend curve.

[0126] G13: Determine the amplitude of the Gaussian white noise to be added to the electricity consumption trend curve based on the fluctuation coefficient of the electricity consumption trend curve.

[0127] G14: Determine the amount of Gaussian white noise to be added to the electricity consumption trend curve based on the preset multiple and amplitude.

[0128] G15: Adds Gaussian white noise of a preset multiple to the electricity consumption trend curve.

[0129] As one possible implementation, embodiments of the present invention can dynamically determine the amplitude of Gaussian white noise by calculating the fluctuation coefficient (based on standard deviation and average value) of the electricity consumption trend curve, and then add noise by a preset multiple. The introduction of noise makes the signal characteristics clearer at different scales, avoiding the mixing of different frequency components into the same intrinsic mode function (IMF) in a single decomposition, thus improving the stability of the decomposition results. Noise parameters are customized according to the degree of fluctuation (fluctuation coefficient) of the electricity consumption trend curve, so that the noise intensity matches the fluctuation characteristics of the signal itself; signals with large fluctuations correspond to stronger noise, and signals with small fluctuations correspond to weaker noise, ensuring that the noise both assists in decomposition and does not mask the true electricity consumption characteristics. The signal after noise optimization processing can obtain purer intrinsic mode functions and residual components in subsequent EEMD decompositions, making the separation of short- and medium-term fluctuation characteristics from long-term trend characteristics more accurate, ultimately providing a more reliable characteristic basis for coordinated control schemes.

[0130] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0131] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0132] Figure 2 A schematic diagram of the microgrid power supply coordination control device provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the microgrid power supply coordination control device 2 includes: Communication module 21 is used to acquire historical electricity consumption data of the microgrid to be coordinated up to the current moment; The processing module 22 is used to predict the electricity consumption trend curve of the microgrid to be coordinated within a set future period after the current moment based on historical electricity consumption data and a preset electricity consumption prediction model; based on the electricity consumption trend curve, it performs multiple ensemble empirical mode decompositions to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within the set future period; based on the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption, it performs electricity aggregation calculation and electricity redistribution to determine the coordination control scheme of the microgrid to be coordinated. The coordination control scheme includes a stable power supply from the main grid and a collaborative power supply from new energy sources and energy storage systems.

[0133] In this embodiment of the invention, based on historical electricity consumption data and a preset electricity consumption prediction model, an electricity consumption trend curve for a preset future period after the current moment of the microgrid to be coordinated is generated. Then, through multiple ensemble empirical mode decompositions, the algorithm's excellent decomposition capability for nonlinear and non-stationary signals is utilized to separate the long-term trend characteristics and short-term fluctuation characteristics of electricity consumption. Furthermore, based on the two types of characteristics, the stable electricity consumption component corresponding to the stable load and the fluctuating electricity consumption component corresponding to the short-term random and periodic load are determined respectively, achieving refined differentiation of the load. Finally, a differentiated power supply mode is adopted, in which the main grid provides the stable electricity consumption component and distributed new energy and energy storage work together to undertake the fluctuating component, thereby cutting off the direct impact of random loads on the main grid from the root and improving the operational stability and power supply continuity of the main grid.

[0134] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 3 As shown, the electronic device 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0135] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0136] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0137] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0138] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0139] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for coordinated control of power supply sources in a microgrid, characterized in that, include: Obtain historical electricity consumption data of the microgrid to be coordinated up to the current moment; Based on the historical electricity consumption data and the preset electricity consumption prediction model, the electricity consumption trend curve of the microgrid to be coordinated in the future within a set period after the current moment is predicted. Based on the electricity consumption trend curve, multiple ensemble empirical mode decompositions are performed to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set time period. Based on the short-term fluctuation characteristics and long-term trend characteristics of the electricity consumption, the electricity consumption is aggregated and redistributed to determine the coordination control scheme of the microgrid to be coordinated. The coordination control scheme includes the stable power supply from the main grid and the collaborative power supply from the joint power supply of new energy sources and energy storage systems.

2. The microgrid power supply coordination control method according to claim 1, characterized in that, The process of performing multiple ensemble empirical mode decompositions based on the electricity consumption trend curve to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set time period includes: The electricity consumption trend curve is subjected to multiple ensemble empirical mode decompositions, and each decomposition yields multiple original intrinsic mode functions and an original residual component. The average value of the original intrinsic mode functions of the same order obtained from multiple decompositions is calculated to obtain stable multi-order intrinsic mode functions; The average value of all original residual components is calculated to obtain the stable residual components; Among them, stable intrinsic mode functions with frequencies higher than a preset high-frequency threshold are used to aggregate and obtain short-term fluctuation characteristics, while stable intrinsic mode functions with frequencies lower than a preset low-frequency threshold are used to merge with stable residual components to obtain long-term trend characteristics.

3. The microgrid power supply coordination control method according to claim 1, characterized in that, The process of calculating and redistributing electricity consumption based on its short-term fluctuations and long-term trends to determine the coordinated control scheme for the microgrid to be coordinated includes: Based on short-term fluctuation characteristics, electricity consumption is aggregated and calculated to determine the fluctuating electricity consumption component. Based on long-term trend characteristics, electricity consumption is aggregated and calculated to determine the stable electricity consumption component. Based on fluctuating and stable power consumption components, power redistribution is performed to determine the coordination control scheme for the microgrid to be coordinated.

4. The microgrid power supply coordination control method according to claim 3, characterized in that, The process of summarizing and calculating electricity consumption based on short-term fluctuation characteristics to determine the fluctuating electricity consumption component includes: For the results of multiple ensemble empirical mode decompositions, the original intrinsic mode functions corresponding to short-term fluctuation characteristics are extracted and sorted from high to low frequency to obtain the sorted results; The average value of the original intrinsic mode functions with the same index in the sorting results is calculated to obtain the stable high-frequency intrinsic mode functions; The stable high-frequency intrinsic mode function is optimized to obtain the optimized stable high-frequency intrinsic mode function; The optimized and stable high-frequency intrinsic mode functions are superimposed to obtain the fluctuating power consumption component.

5. The microgrid power supply coordination control method according to claim 4, characterized in that, The optimization of the stable high-frequency intrinsic mode function to obtain the optimized stable high-frequency intrinsic mode function includes: Calculate the Pearson correlation coefficient between each stable high-frequency intrinsic mode function and the electricity consumption trend curve to obtain the correlation score of each stable high-frequency intrinsic mode function; Based on a preset period threshold, stable high-frequency intrinsic mode functions are divided into high-frequency band group, mid-frequency band group and low-frequency band group. For stable high-frequency intrinsic mode functions of the high-frequency, mid-frequency, and low-frequency groups, a threshold screening is performed based on a preset threshold to obtain the preliminary screening results of stable high-frequency intrinsic mode functions of the high-frequency, mid-frequency, and low-frequency groups. Based on the initial screening results, the stable high-frequency intrinsic mode functions of the high-frequency, mid-frequency, and low-frequency groups are ranked and refined according to a preset percentage to obtain the refined screening results of the stable high-frequency intrinsic mode functions of the high-frequency, mid-frequency, and low-frequency groups.

6. The microgrid power supply coordination control method according to claim 2, characterized in that, Before determining the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set period by performing multiple ensemble empirical mode decompositions based on the electricity consumption trend curve, the method further includes: Calculate the standard deviation and average value of the electricity consumption trend curve; Calculate the fluctuation coefficient of the electricity consumption trend curve based on the standard deviation and average value of the electricity consumption trend curve; Based on the fluctuation coefficient of the electricity consumption trend curve, determine the amplitude of the Gaussian white noise to be added to the electricity consumption trend curve; Based on the preset multiple and amplitude, determine the preset multiple of Gaussian white noise to be added to the electricity consumption trend curve; Add a preset multiple of Gaussian white noise to the power consumption trend curve.

7. The microgrid power supply coordination control method according to claim 3, characterized in that, The process of summarizing and calculating electricity consumption based on long-term trend characteristics to determine the stable electricity consumption component includes: The trend slope of long-term trend characteristics is calculated using the sliding window method; Calculate the absolute value of the trend slope of all windows within a preset time period; If the maximum absolute value is less than the preset trend slope threshold, the long-term trend feature is determined as a stable electricity consumption component. If the maximum absolute value is greater than or equal to the preset trend slope threshold, readjust the amplitude and preset multiple of the Gaussian white noise, and re-perform ensemble empirical mode decomposition.

8. The microgrid power supply coordination control method according to claim 3, characterized in that, The process of redistributing electricity based on fluctuating and stable electricity consumption components to determine the coordinated control scheme for the microgrid to be coordinated includes: For high-frequency fluctuating electricity consumption components, distributed photovoltaic and distributed wind power are given priority for power supply, and lithium battery energy storage systems are used to supplement the power supply. For the fluctuating power consumption components of medium frequency, distributed photovoltaic and distributed wind power are given priority for power supply, while gas turbines and supercapacitors supplement the power supply. For low-frequency fluctuating electricity consumption components, power supply is provided in a coordinated manner by expanding distributed photovoltaic capacity, expanding distributed wind power capacity, and long-term energy storage systems.

9. A microgrid power supply coordination control device, characterized in that, include: The communication module is used to acquire historical electricity consumption data of the microgrid to be coordinated up to the current moment; The processing module is used to predict the electricity consumption trend curve of the microgrid to be coordinated in the future set period after the current moment, based on the historical electricity consumption data and the preset electricity consumption prediction model. Based on the electricity consumption trend curve, multiple ensemble empirical mode decompositions are performed to determine the short-term fluctuation characteristics and long-term trend characteristics of electricity consumption within a future set time period. Based on the short-term fluctuation characteristics and long-term trend characteristics of the electricity consumption, the electricity consumption is aggregated and redistributed to determine the coordination control scheme of the microgrid to be coordinated. The coordination control scheme includes the stable power supply from the main grid and the collaborative power supply from the joint power supply of new energy sources and energy storage systems.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images