Shale oil exploitation microgrid collaborative optimization scheduling method based on source-load interaction

Abstract

This application relates to the field of power grid dispatching technology, specifically to a collaborative optimization dispatching method for shale oil extraction microgrids based on source-load interaction. The method includes: acquiring the photovoltaic power generation, wind power generation, and load demand power in the shale oil extraction microgrid; analyzing the random variation characteristics and deviation degree of the data in each power residual component, and obtaining the random abrupt change degree of each power residual component; performing frequency domain analysis on the power residual components to obtain the random complexity of each power residual component, and then obtaining the random characteristic values ​​of each power residual component, which are used to predict the photovoltaic power generation, wind power generation, and load demand forecasts for all time periods within the current day; establishing an objective function; and using an optimization algorithm to perform collaborative optimization dispatching of the shale oil extraction microgrid. This application can improve the dispatching accuracy of microgrids.

Description

Technical Field

[0001] This application relates to the field of power grid dispatching technology, specifically to a collaborative optimization dispatching method for shale oil extraction microgrids based on source-load interaction. Background Technology

[0002] Currently, microgrids face significant challenges in shale oil extraction, particularly in the Gulong shale oil extraction, where high energy consumption and large load fluctuations are prevalent. Furthermore, the output of renewable energy sources on the source side is significantly affected by natural environmental conditions; for example, wind and photovoltaic power generation in microgrids are intermittent and fluctuating. Therefore, to respond to the load demands of shale oil extraction, accurate and coordinated optimization scheduling of microgrids for shale oil extraction is necessary.

[0003] To address the uncertainty of power generation and load in microgrids used in shale oil extraction, existing technologies for collaborative optimization scheduling of shale oil extraction microgrids can be divided into two stages. The first stage uses a deep learning model to predict photovoltaic power generation, wind power generation, and load demand. The second stage establishes a microgrid optimization scheduling model and its constraints based on the power prediction results, and solves the model using an optimization algorithm to achieve collaborative optimization scheduling of the shale oil extraction microgrid. Due to the randomness of wind and solar power generation and fracturing load fluctuations in microgrids, existing technologies generally use neural networks to predict power generation in shale oil extraction microgrids. However, these technologies primarily focus on the power data itself and do not fully consider the abnormal randomness of power generation and load in shale oil extraction microgrids. This can easily lead to poor accuracy in power prediction, resulting in low reliability of the microgrid optimization scheduling model and affecting the effectiveness of collaborative optimization scheduling of shale oil extraction microgrids. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a collaborative optimization scheduling method for shale oil extraction microgrids based on source-load interaction, thereby resolving the existing issues.

[0005] The collaborative optimization scheduling method for shale oil extraction microgrids based on source-load interaction proposed in this application adopts the following technical solution:

[0006] One embodiment of this application provides a collaborative optimization scheduling method for microgrids in shale oil extraction based on source-load interaction, including the following steps:

[0007] Obtain the photovoltaic power generation, wind power generation, and load demand in the microgrid for shale oil extraction;

[0008] Extract each power residual component of photovoltaic power generation in each time period, analyze the randomness variation characteristics and the degree of deviation of the distribution of the data in each power residual component, and obtain the random abrupt change degree of each power residual component.

[0009] Frequency domain analysis is performed on the power residual components. By analyzing the complexity and randomness of the frequency energy variation, the randomness complexity of each power residual component is obtained. Combined with the randomness abrupt change, the randomness characteristic value of each power residual component is obtained.

[0010] By analyzing the photovoltaic power generation, wind power generation, and load demand at different time periods, and combining the randomness characteristics of photovoltaic power generation, wind power generation, and load demand, the predicted photovoltaic power generation, wind power generation, and load demand for all time periods within the current day are forecasted. Then, an objective function is established, and an optimization algorithm is used to perform collaborative optimization scheduling of the shale oil extraction microgrid.

[0011] Preferably, a singular spectrum analysis algorithm is used to obtain each power residual component of the photovoltaic power generation in each time period, and the randomness variation rate of each element in each power residual component is calculated: ,in, Let be the rate of random change of the j-th element in the i-th power residual component. and These are the j-th and (j-1)-th elements in the i-th power residual component, respectively. For a preset minimum positive number; when When, the rate of change of randomness It is 0.

[0012] Preferably, the formula for obtaining the random abrupt change degree of each power residual component is: In the formula, Let be the random abrupt change degree of the i-th power residual component. Let represent the degree of dispersion of the rate of change of randomness for all elements in the i-th power residual component. Let be the number of elements in the i-th power residual component. Let be the rate of random change of the j-th element in the i-th power residual component. Let be the mean of the random rate of change of all elements in the i-th power residual component.

[0013] Preferably, before obtaining the randomness complexity, the amplitude of all frequency components of each power residual component is thresholded, and the frequency components corresponding to the amplitudes greater than or equal to the threshold are taken as the key frequency components of each power residual component.

[0014] Preferably, for each power residual component, the square of the amplitude corresponding to each key frequency component is divided by the sum of the squares of the amplitudes corresponding to all key frequency components to obtain the energy proportion of each key frequency component for each power residual component.

[0015] Preferably, the formula for obtaining the randomness complexity of each power residual component is:

[0016] In the formula, Let be the random complexity of the i-th power residual component. Let Q be the information entropy representing the energy proportion of all critical frequency components in the i-th power residual component, and let Q be the number of critical frequency components in the i-th power residual component. For the i-th power residual component, the s-th critical frequency component is... Let be the mean of the remaining frequency components in the i-th power residual component, excluding the critical frequency component. It is a preset minimum positive number.

[0017] Preferably, the formula for obtaining the randomness characteristic value of each power residual component is: In the formula, Let i be the random eigenvalue of the i-th power residual component. Let be the randomness abruptness and randomness complexity of the i-th power residual component, respectively.

[0018] Preferably, the feature vectors for each time period are formed by the photovoltaic power generation, wind power generation, load demand, the maximum value of the random characteristic value of photovoltaic power generation, the maximum value of the random characteristic value of wind power generation, and the maximum value of the random characteristic value of load demand power for each time period of the previous day. The predicted photovoltaic power generation, wind power generation, and load demand for all time periods of the current day are obtained through a predictive neural network model.

[0019] Preferably, the objective function is to minimize the cost function C, and the specific formula is as follows:

[0020]

[0021] In the formula, min represents the minimum value. To optimize the number of time periods within the coordinated scheduling time of microgrids, and These represent the prices at which the microgrid purchases and sells electricity to the main grid during the t-th time period. and These represent the electricity purchased and sold by the microgrid to the main grid during the t-th time period, respectively. The unit operation and maintenance cost of photovoltaic power generation. Let be the predicted photovoltaic power generation of the microgrid in the t-th time period. The unit operation and maintenance cost of wind power generation, Let be the predicted wind power generation of the microgrid in time period t. This refers to the unit depreciation cost of energy storage equipment. Let be the operating power of the energy storage device in the microgrid during the t-th time period.

[0022] Preferably, in the collaborative optimization scheduling process of shale oil extraction microgrids, the constraints of the optimization algorithm are energy storage device constraints, power balance constraints, and main grid power interaction constraints. The main grid power interaction constraint is that the electricity purchased and sold by the microgrid to the main grid shall not exceed the maximum value allowed by the main grid.

[0023] This application has at least the following beneficial effects:

[0024] This invention considers the abnormal mutation phenomenon of random changes in source load power in shale oil extraction microgrids. By analyzing the fluctuation characteristics of random changes in source load power and combining the average deviation characteristics and overall deviation characteristics of random changes in source load power, the degree of abnormal mutation of random changes in source load power in shale oil extraction microgrids is accurately measured, which more clearly reflects the abnormal randomness characteristics of source load power in shale oil extraction microgrids.

[0025] To more fully explore the anomalous randomness of source-load power in shale oil extraction microgrids, this invention analyzes the frequency and amplitude characteristics of random changes in source-load power in shale oil extraction microgrids, and more accurately measures the complexity of random changes in source-load power in shale oil extraction microgrids, which is then used to more accurately measure the anomalous randomness of source-load power in shale oil extraction microgrids.

[0026] This invention combines the degree of anomalous change in the randomness of source and load power to more accurately measure the anomalous randomness of source and load power in shale oil extraction microgrids. It also fully considers these anomalous randomness characteristics and employs a neural network model to more accurately predict the power consumption of shale oil extraction microgrids. Based on the power consumption prediction results, a more reliable microgrid optimization scheduling model and model constraints are established, enabling the collaborative optimization scheduling scheme for shale oil extraction microgrids to more accurately respond to fracturing load fluctuations during shale oil extraction. Attached Figure Description

[0027] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 The flowchart illustrates the steps of the collaborative optimization scheduling method for shale oil extraction microgrids based on source-load interaction provided in this application. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the source-load interaction-based microgrid collaborative optimization scheduling method for shale oil extraction proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0030] Unless otherwise defined, terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0031] The following description, in conjunction with the accompanying drawings, details the specific scheme of the source-load interaction-based microgrid collaborative optimization scheduling method for shale oil extraction provided in this application.

[0032] This application provides an embodiment of a collaborative optimization scheduling method for microgrids used in shale oil extraction based on source-load interaction. For details, please refer to [link to relevant documentation]. Figure 1 This includes the following steps:

[0033] Step 1: Obtain the photovoltaic power generation, wind power generation, and load demand power in the shale oil extraction microgrid.

[0034] To improve the effectiveness of collaborative optimization scheduling of microgrids for shale oil extraction, it is necessary to fully explore the anomalous randomness of source and load power in the microgrid and use a neural network model to more accurately predict the power consumption of the microgrid. Based on the power consumption prediction results of the microgrid, an optimal scheduling model and model constraints should be established to enable the collaborative optimization scheduling scheme of the microgrid to respond more accurately to the fluctuations in fracturing load during shale oil extraction.

[0035] Therefore, in order to more accurately coordinate and optimize the scheduling of shale oil extraction microgrids, photovoltaic power generation from the source-side photovoltaic generators in the shale oil extraction microgrid is collected through photovoltaic inverters, and wind power generation from the source-side wind turbines in the shale oil extraction microgrid is collected through a phasor measurement unit. At the same time, load demand power from the load side in the shale oil extraction microgrid is collected through smart meters. In this embodiment, the sampling rate of photovoltaic power generation, wind power generation, and load demand power is 10Hz, and the duration of each time period is 15 minutes, so that the source and load power time series data for each time period can be obtained, including photovoltaic power generation time series data, wind power generation time series data, and load demand power time series data.

[0036] Step 2: Extract each power residual component of photovoltaic power generation in each time period, analyze the randomness variation characteristics and the degree of deviation of the distribution of the data in each power residual component, and obtain the randomness abrupt change degree of each power residual component.

[0037] Due to the randomness of wind and solar power generation and fracturing load fluctuations in shale oil extraction microgrids, the collaborative optimization scheduling method for shale oil extraction microgrids needs to fully exploit the abnormal randomness of source and load power in the microgrids. This requires using a neural network prediction model to accurately predict the power consumption of the shale oil extraction microgrids, and then establishing a more reliable microgrid optimization scheduling model and model constraints based on the power consumption prediction results.

[0038] Because the fluctuations in wind and solar power generation on the source side and fracturing load on the load side of the microgrid both exhibit anomalous randomness, this analysis uses photovoltaic power generation time-series data as an example. Wind power generation time-series data and load demand power time-series data are extracted using the same method. Therefore, to fully exploit the anomalous randomness of source and load power in the shale oil extraction microgrid, the photovoltaic power generation time-series data for each time period is used as input to the Singular Spectrum Analysis algorithm, with a preset window length of [value missing]. ,in, It is a rounding function. To determine the number of photovoltaic power generation time series data for each time period, a weighted correlation graph is used for grouping. The singular spectrum analysis algorithm is used to obtain each power residual component for each time period, reflecting the random variation characteristics of source and load power time series data in the microgrid. This is beneficial for more accurately measuring the abnormal randomness characteristics of source and load power time series changes in the future.

[0039] Furthermore, in order to analyze the differences in the random variation of source-load power, the random variation rate of each element in each power residual component is calculated. The method for calculating the random variation rate is as follows: ,in, Let be the rate of random change of the j-th element in the i-th power residual component. and These are the j-th and (j-1)-th elements in the i-th power residual component, respectively. To predetermine a very small positive number, its purpose is to avoid the denominator being 0. The value range is 0.01-0.2; in this embodiment, it is set to 0.05. The rate of random variation reflects the magnitude of the difference in the random variation of source load power. The larger the rate of random variation, the greater the difference in the random variation of source load power in the shale oil extraction microgrid, and the more it reflects the random fluctuation characteristics of source load power in the microgrid. Specifically, when... When, the rate of change of randomness A value of 0 indicates that the difference in the random variation of source and load power in the microgrid is zero at this time.

[0040] In general, the random changes in source load power in shale oil extraction microgrids will exhibit anomalous abrupt changes. For example, rapid cloud movement, gust changes, and the impact of fracturing loads on high-power equipment during shale oil extraction will cause all random change rates in the power residual components to deviate abnormally from their distribution. The more significant the anomalous deviation of all random change rates, the more it reflects the anomalous abrupt changes in the random changes of source load power.

[0041] Based on the above analysis, for each time period, calculate the degree of random abrupt change for each power residual component:

[0042] In the formula, Let be the random abrupt change degree of the i-th power residual component. Let represent the degree of dispersion of the random rate of change of all elements in the i-th power residual component. The degree of dispersion can be variance, standard deviation, or coefficient of variation. Let be the number of elements in the i-th power residual component. Let be the rate of random change of the j-th element in the i-th power residual component. Let be the mean of the random rate of change of all elements in the i-th power residual component.

[0043] Based on existing feature measurement methods, the above formula uses the average difference between the random change rate of each element and its average level to characterize the average deviation feature of all random change rates in the power residual component. At the same time, it uses the dispersion of the random change rate of all elements to characterize the overall deviation feature of all random change rates in the power residual component. Then, it combines the average deviation feature and the overall deviation feature and performs feature fusion by multiplication to achieve the measurement of random change rate.

[0044] It can be understood that the randomness abruptness reflects the degree of abnormal abrupt change in the randomness of source load power in the shale oil extraction microgrid. The greater the randomness abruptness, the greater the degree of abnormal abrupt change in the randomness of source load power in the shale oil extraction microgrid. In this case, there is a greater possibility of the rapid movement of clouds, gust changes, and the impact of fracturing load on high-power equipment during the shale oil extraction process, thus more clearly reflecting the abnormal randomness of source load power in the shale oil extraction microgrid.

[0045] Step 3: Perform frequency domain analysis on the power residual components. By analyzing the complexity and randomness of the frequency energy changes, obtain the randomness complexity of each power residual component. Combine this with the randomness abrupt change to obtain the randomness characteristic value of each power residual component.

[0046] However, relying solely on the randomness of the power residual components in the time domain is insufficient to effectively measure the frequency and amplitude characteristics of the random variations in source load power. To more fully exploit the anomalous randomness of source load power in shale oil extraction microgrids, it is necessary to analyze the frequency domain characteristics of each power residual component. Therefore, each power residual component is used as the input to a Fourier transform, which can be either a Fast Fourier Transform (FFT) or a Discrete Fourier Transform (DFT). The FFT extracts the amplitude of all frequency components in each power residual component, reflecting the frequency characteristics of the random variations in source load power in shale oil extraction microgrids.

[0047] To further analyze the frequency and amplitude energy characteristics of the random variations in source-load power in shale oil extraction microgrids, the amplitudes of all frequency components of each power residual component are used as inputs to the Otsu's inter-class variance algorithm. The Otsu's algorithm is used to obtain a segmentation threshold. Each frequency component with an amplitude greater than or equal to the segmentation threshold is designated as a key frequency component of each power residual component. For each power residual component, the square of the amplitude corresponding to each key frequency component is divided by the sum of the squares of the amplitudes corresponding to all key frequency components to obtain the energy proportion of each key frequency component of each power residual component. This reflects the main frequency energy characteristics of the random variations in source-load power in the shale oil extraction microgrid. It should be noted that, to prevent the denominator from being zero and thus uncalculation, a very small positive number is added to the denominator to avoid a zero denominator during the ratio calculation. In this embodiment, the value is 0.0001.

[0048] Under normal circumstances, the rapid movement of clouds, changes in gusts, and the impact of fracturing loads will cause the random changes in source and load power in microgrids to have many rapid and drastic components. This will result in high complexity of energy changes in all key frequency components in each power residual component, and the key frequency components will be significantly higher than the average of the other frequency components.

[0049] Based on the above analysis, in order to more fully exploit the anomalous randomness of source-load power in the shale oil extraction microgrid for each time period, the randomness complexity of each power residual component is calculated:

[0050] In the formula, Let be the random complexity of the i-th power residual component. Let Q be the information entropy representing the energy proportion of all critical frequency components in the i-th power residual component, and let Q be the number of critical frequency components in the i-th power residual component. For the i-th power residual component, the s-th critical frequency component is... It is the mean of the remaining frequency components in the i-th power residual component, excluding the critical frequency component.

[0051] The above formula, based on existing feature measurement methods, uses a ratio to measure the magnitude between each key frequency component and the mean of the remaining frequency components, and employs information entropy to measure the energy variation complexity of all key frequency components in the power residual component. Furthermore, by combining the calculation results of information entropy and the ratio, the randomness complexity of the power residual component is measured from a frequency domain perspective. It should be noted that the specific calculation process of information entropy is prior art known to those skilled in the art and will not be elaborated upon in this embodiment.

[0052] It is understandable that the randomness complexity reflects the complexity of the random changes in source load power in the shale oil extraction microgrid. The greater the randomness complexity, the greater the likelihood that it will be affected by rapid cloud movement, gust changes, and fracturing load impacts. This makes the randomness of the source load power in the shale oil extraction microgrid more complex and better reflects the abnormal randomness of the source load power in the shale oil extraction microgrid.

[0053] Therefore, in order to fully consider the anomalous randomness of source and load power in shale oil extraction microgrids, a neural network model is used to more accurately predict the power consumption of shale oil extraction microgrids. Combining the randomness abruptness and randomness complexity of time-frequency domain analysis, the randomness characteristic value of each power residual component is calculated:

[0054] In the formula, Let be the randomness characteristic value of the i-th power residual component. To avoid the influence of different dimensions, this embodiment uses normalization to bring the randomness mutation degree and randomness complexity to the same order of magnitude. Many existing normalization methods exist, and no limitation is made here. This embodiment uses exponential normalization to normalize the range of randomness mutation degree and randomness complexity to 0 to 1. The specific process is existing technology and will not be described in this embodiment.

[0055] Therefore, by using a product approach to fuse and measure the randomness abruptness and randomness complexity of time-frequency domain analysis, the abnormal randomness characteristics of source-load power in shale oil extraction microgrids are more clearly reflected. This allows for the use of neural network prediction models to accurately predict the power consumption of shale oil extraction microgrids, which is beneficial for establishing more reliable microgrid optimization scheduling models and model constraints based on the power consumption prediction results of shale oil extraction microgrids.

[0056] Step 4: By analyzing the photovoltaic power generation, wind power generation, and load demand for each time period, and combining the randomness characteristics of photovoltaic power generation, wind power generation, and load demand, predict the photovoltaic power generation, wind power generation, and load demand for all time periods within the current day. Then, establish the objective function and use optimization algorithms to perform collaborative optimization scheduling of the shale oil extraction microgrid.

[0057] Therefore, in order to accurately predict photovoltaic power generation, wind power generation, and load demand, the time-series data of photovoltaic power generation, wind power generation, and load demand power for each time period are integrated along the time dimension of each time period to obtain the photovoltaic power generation, wind power generation, and load demand for each time period. At the same time, feature extraction is performed on the time-series data of photovoltaic power generation, wind power generation, and load demand power for each time period using the methods in steps 2 and 3, respectively. This allows us to obtain the maximum values ​​of the randomness characteristics of all power residual components of photovoltaic power generation, wind power generation, and load demand power for each time period.

[0058] To accurately predict the power consumption of shale oil extraction microgrids using a neural network prediction model, a vector composed of the maximum random eigenvalues ​​of photovoltaic power generation, wind power generation, load demand, and load demand power for each time period was used as the feature vector for each time period. The feature vectors of all time periods within 24 hours of each day over the past three months were used as input to train an LSTM (Long Short-Term Memory) neural network model. The model then predicted the photovoltaic power generation, wind power generation, and load demand for all time periods within 24 hours of the day following the current day. The Adam optimization algorithm was used as the optimizer with a learning rate of 0.001, and the mean squared error (MSE) was selected as the loss function to obtain the trained prediction neural network model.

[0059] Therefore, the feature vectors of all time periods within the previous 24 hours are used as input to the trained predictive neural network model, and the photovoltaic power generation forecast, wind power generation forecast, and load demand forecast for all time periods within the current day are obtained through the trained predictive neural network model.

[0060] Furthermore, based on the power forecast results, a microgrid optimal scheduling model and model constraints are established. The objective function of the current shale oil extraction microgrid optimal scheduling model is to minimize the cost function C, which can be expressed as:

[0061]

[0062] In the formula, min represents the minimum value. To optimize the number of time periods within the coordinated scheduling time of microgrids, and These represent the prices at which the microgrid purchases and sells electricity to the main grid during the t-th time period. and These represent the electricity purchased and sold by the microgrid to the main grid during the t-th time period, respectively. The unit operation and maintenance cost of photovoltaic power generation is 0.04-0.06 yuan / kWh. Let be the predicted photovoltaic power generation of the microgrid in the t-th time period. The unit operation and maintenance cost of wind power generation is 0.05-0.07 yuan / kWh. Let be the predicted wind power generation of the microgrid in time period t. The unit depreciation cost of energy storage equipment is calculated, ranging from 0.05 to 0.09 yuan / kW. Let be the operating power of the energy storage device in the microgrid during time period t. Where, when When the value is less than 0, the energy storage device is charged. When the value is greater than 0, the energy storage device discharges. When the value is 0, the energy storage device is not currently in operation.

[0063] The objective function of the collaborative optimization scheduling model for shale oil extraction microgrids is used to determine the model constraints of the collaborative optimization scheduling model for shale oil extraction microgrids. Preferably, in this embodiment, these constraints include energy storage device constraints, power balance constraints (determined based on load forecast demand), and main grid power interaction constraints. The main grid power interaction constraints refer to the requirement that the electricity purchased and sold by the microgrid to the main grid must not exceed the maximum value allowed by the main grid. This is a basic requirement under the conditions of microgrid grid-connected operation. Meanwhile, the establishment of energy storage device constraints and power balance constraints in the field of grid technology is existing technology. Implementers can set them themselves in actual application scenarios. This embodiment will not elaborate on them or impose any special limitations on them.

[0064] Furthermore, the objective function of the microgrid optimization scheduling model is solved using the particle swarm optimization algorithm. Under the model constraints of the collaborative optimization scheduling model for shale oil extraction microgrids, the purchased or sold electricity and the charging or discharging power of energy storage units for all time periods within the current day are randomly generated as 192-dimensional particles. The particle population size is 100, the initial inertia weight is 0.8, the inertia weight typically adopts a linear decreasing strategy, the final inertia weight is 0.4, the learning factor is 1.5, and the maximum number of iterations is 200. The objective function of the current day's shale oil extraction microgrid optimization scheduling model, which minimizes the cost function value, is used as the fitness function of each particle in the particle swarm optimization algorithm. Iterative optimization is performed using the particle swarm optimization algorithm to finally obtain the optimal scheme for collaborative optimization scheduling of the current day's shale oil extraction microgrid. Thus, energy scheduling is carried out through the optimal scheme for collaborative optimization scheduling of the current day's shale oil extraction microgrid, improving the effectiveness of collaborative optimization scheduling of the shale oil extraction microgrid.

[0065] It is understood that references to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include the specific features, structures, or characteristics described in connection with that embodiment. Therefore, the appearance of phrases such as "in one embodiment," "in some embodiments," "in other embodiments," or "in still other embodiments" in different parts of this specification does not necessarily refer to the same embodiment, but rather means "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0066] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous. Moreover, the sequence numbers of the steps in the embodiments do not imply a specific 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 in this specification.

[0067] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A shale oil exploitation microgrid collaborative optimization scheduling method based on source-load interaction, characterized in that, Includes the following steps: Obtain the photovoltaic power generation, wind power generation, and load demand in the microgrid for shale oil extraction; Extract each power residual component of photovoltaic power generation in each time period, analyze the randomness variation characteristics and the degree of deviation of the distribution of the data in each power residual component, and obtain the random abrupt change degree of each power residual component. Frequency domain analysis is performed on the power residual components. By analyzing the complexity and randomness of the frequency energy variation, the randomness complexity of each power residual component is obtained. Combined with the randomness abrupt change, the randomness characteristic value of each power residual component is obtained. By analyzing the photovoltaic power generation, wind power generation, and load demand at different time periods, and combining the randomness characteristics of photovoltaic power generation, wind power generation, and load demand, the predicted photovoltaic power generation, wind power generation, and load demand for all time periods within the current day are forecasted. Then, an objective function is established, and an optimization algorithm is used to perform collaborative optimization scheduling of the shale oil extraction microgrid.

2. The collaborative optimization scheduling method for shale oil extraction microgrids based on source-load interaction as described in claim 1, characterized in that, The singular spectrum analysis algorithm is used to obtain each power residual component of the photovoltaic power in each period, and the randomness change rate of each element in each power residual component is calculated: wherein, is the randomness change rate of the jth element in the ith power residual component, and are the jth and j-1th elements in the ith power residual component, respectively, is a preset minimum positive number; when the randomness change rate is 0.

3. The shale oil extraction microgrid collaborative optimization scheduling method based on source-load interaction as described in claim 2, characterized in that, The acquisition formula of the randomness mutation degree of each power residual component is: ; in the formula, is the randomness mutation degree of the i th power residual component, is the dispersion degree of the randomness change rate of all elements in the i th power residual component, is the number of elements in the i th power residual component, is the randomness change rate of the j th element in the i th power residual component, is the average of the randomness change rate of all elements in the i th power residual component.

4. The collaborative optimization scheduling method for shale oil extraction microgrids based on source-load interaction as described in claim 1, characterized in that, Before obtaining the randomness complexity, the amplitude of all frequency components of each power residual component is thresholded, and the frequency components corresponding to the amplitudes greater than or equal to the threshold are taken as the key frequency components of each power residual component.

5. The shale oil extraction microgrid collaborative optimization scheduling method based on source-load interaction as described in claim 4, characterized in that, For each power residual component, the square of the amplitude corresponding to each key frequency component is divided by the sum of the squares of the amplitudes corresponding to all key frequency components to obtain the energy proportion of each key frequency component for each power residual component.

6. The shale oil extraction microgrid collaborative optimization scheduling method based on source-load interaction as described in claim 5, characterized in that, The formula for obtaining the randomness complexity of each power residual component is: In the formula, Let be the random complexity of the i-th power residual component. Let Q be the information entropy representing the energy proportion of all critical frequency components in the i-th power residual component, and let Q be the number of critical frequency components in the i-th power residual component. For the i-th power residual component, the s-th critical frequency component is... Let be the mean of the remaining frequency components in the i-th power residual component, excluding the critical frequency component. It is a preset minimum positive number.

7. The method for collaborative optimization scheduling of shale oil extraction microgrids based on source-load interaction as described in claim 1, characterized in that, The formula for obtaining the randomness eigenvalues ​​of each power residual component is as follows: In the formula, Let i be the random eigenvalue of the i-th power residual component. Let be the randomness abruptness and randomness complexity of the i-th power residual component, respectively.

8. The method for collaborative optimization scheduling of shale oil extraction microgrids based on source-load interaction as described in claim 1, characterized in that, The feature vectors for each time period are formed by taking the photovoltaic power generation, wind power generation, load demand, the maximum random characteristic value of photovoltaic power generation, the maximum random characteristic value of wind power generation, and the maximum random characteristic value of load demand power from the previous day's time period. The predicted photovoltaic power generation, wind power generation, and load demand for all time periods within the current day are then obtained through a predictive neural network model.

9. The method for collaborative optimization scheduling of shale oil extraction microgrids based on source-load interaction as described in claim 1, characterized in that, The objective function is to minimize the cost function C, and the specific formula is as follows: In the formula, min represents the minimum value. To optimize the number of time periods within the coordinated scheduling time of microgrids, and These represent the prices at which the microgrid purchases and sells electricity to the main grid during the t-th time period. and These represent the electricity purchased and sold by the microgrid to the main grid during the t-th time period, respectively. The unit operation and maintenance cost of photovoltaic power generation. Let be the predicted photovoltaic power generation of the microgrid in the t-th time period. The unit operation and maintenance cost of wind power generation, Let be the predicted wind power generation of the microgrid in time period t. This refers to the unit depreciation cost of energy storage equipment. Let be the operating power of the energy storage device in the microgrid during the t-th time period.

10. The method for collaborative optimization scheduling of shale oil extraction microgrids based on source-load interaction as described in claim 9, characterized in that, In the process of collaborative optimization scheduling of microgrids for shale oil extraction, the constraints of the optimization algorithm are energy storage device constraints, power balance constraints, and main grid power interaction constraints. The main grid power interaction constraint is that the electricity purchased and sold by the microgrid to the main grid shall not exceed the maximum value allowed by the main grid.

Images