Source network load storage integrated configuration optimization method and system considering external influence

Abstract

The invention relates to the technical field of intelligent power grids, in particular to a source network load storage integrated configuration optimization method and system considering external influences, and the method comprises the steps: enabling the working states of energy storage equipment to comprise a network construction state and a network following state, controlling the virtual rotational inertia and the virtual damping coefficient of the energy storage equipment in the network construction state through employing an MPC algorithm, the first fluctuation change characteristic of the frequency in the power grid is used for correction; evaluating a second input capacity of the energy storage equipment in the network following state according to a second fluctuation change characteristic of the load power in the power grid, and evaluating a first input capacity of the energy storage equipment in the network construction state according to the first fluctuation change characteristic; and updating the correction coefficient according to the difference between the total capacity of all the energy storage devices and the first input capacity and the second input capacity. According to the method, the stability of the power grid frequency and the load is considered, and the problem of instability of a new energy highly-permeable power grid under the influence of an external environment is avoided.

Description

Technical Field

[0001] This invention relates to the field of smart grid technology, and more specifically to an integrated configuration optimization method and system for source-grid-load-storage that takes into account external influences. Background Technology

[0002] With the increasing proportion of renewable energy generation, integrated power grids with grid-load-storage systems have become an important technological path for achieving high-proportion renewable energy consumption. However, power grid systems with high renewable energy penetration face many challenges: the large-scale replacement of traditional synchronous generators by power electronic devices such as wind power and photovoltaics has led to a significant reduction in the total rotational inertia of the power grid. When power deficits or external disturbances occur (such as sudden weather changes, equipment failures, or severe load fluctuations), the frequency drops faster and more significantly, and it is difficult to automatically recover to the rated value. Existing energy storage devices typically operate according to fixed parameters (such as virtual rotational inertia and damping coefficients in grid-connected mode, or PQ control mode in grid-connected mode), and cannot be dynamically adjusted according to the real-time status of the power grid. When changes in the external environment cause frequency shifts and increased load fluctuations, the initial configuration of energy storage quickly fails, which is detrimental to the long-term stability of the power grid. Summary of the Invention

[0003] To address the aforementioned problems, this invention provides a method and system for optimizing the integrated configuration of power generation, grid, load, and storage, taking into account external influences.

[0004] The integrated source-grid-load-storage configuration optimization method and system of the present invention, taking into account external influences, adopts the following technical solution: One embodiment of the present invention provides an integrated source-grid-load-storage configuration optimization method that takes into account external influences. The method includes the following steps: The working states of energy storage devices in the integrated power grid include grid connection state and grid connection state. Grid connection state refers to the energy storage device being connected to the grid as a virtual synchronous generator. The virtual moment of inertia and virtual damping coefficient of the virtual synchronous generator are controlled by the MPC algorithm. The virtual moment of inertia and virtual damping coefficient are referred to as virtual parameters. In grid connection state, the energy storage device is connected to the grid using an active-reactive control mode. The first frequency fluctuation characteristic in the power grid is monitored in real time, and the virtual parameters under the control of the MPC algorithm are corrected according to the first fluctuation characteristic using a preset correction coefficient. The second fluctuation characteristic of the load power in the power grid is detected in real time. The second input capacity of the energy storage device in the grid-connected state is evaluated based on the second fluctuation characteristic. The first input capacity of the energy storage device in the grid-connected state is evaluated based on the first fluctuation characteristic F of the frequency obtained under the corrected virtual parameters. The first input capacity is positively correlated with the first fluctuation characteristic F, and the second input capacity is positively correlated with the second fluctuation characteristic. The correction coefficients are updated based on the differences between the total capacity of all energy storage devices and the first and second input capacities, respectively.

[0005] Preferably, the specific steps of correcting the virtual parameters under the MPC algorithm control using a preset correction coefficient based on the first fluctuation change characteristic are as follows: The first fluctuation change characteristics include frequency offset characteristics and frequency fluctuation characteristics, denoted as a and b respectively; the virtual parameters obtained at each moment under the control of the MPC algorithm include virtual moment of inertia and virtual damping coefficient, denoted as J0 and D0 respectively; Decrease the value of J0, with the magnitude of the decrease being positively correlated with a and m1 respectively; increase the value of D0, with the magnitude of the increase being positively correlated with b and m2 respectively; use the decreased J0 and increased D0 as the corrected virtual parameters; where m1 represents the preset first coefficient and m2 represents the preset second coefficient, and m1 and m2 are collectively referred to as the correction coefficients; After each moment, the energy storage device, acting as a virtual synchronous generator, operates with the corrected virtual parameters.

[0006] Preferably, the specific steps for assessing the second operational capacity of the energy storage device based on the second fluctuation change characteristics and its grid connection status are as follows: The sum of the capacities of all grid-connected energy storage devices at the current moment is denoted as C20; the sum of the capacities of all energy storage devices is denoted as C0; the second fluctuation characteristic is denoted as F2; ​​the second input capacity is equal to the capacity expanded based on C20, where the expanded capacity is determined by F2 times C0.

[0007] Preferably, the specific steps for evaluating the first operational capacity of the energy storage device based on the first frequency fluctuation characteristic F obtained under the corrected virtual parameters are as follows: The first fluctuation change feature F includes frequency offset feature and frequency fluctuation feature, and the mean of frequency offset feature and frequency fluctuation feature is denoted as F1; The sum of the capacities of the energy storage devices in the current grid state is denoted as C10, and the sum of the capacities of all energy storage devices is denoted as C0. The first input capacity is equal to the capacity expanded based on C10, where the expanded capacity is determined by F1 times C0.

[0008] Preferably, the specific steps for updating the correction coefficient based on the difference between the total capacity of all energy storage devices and the first and second input capacities are as follows: The sum of the first and second input capacities obtained at the current moment is denoted as the first comprehensive capacity; When the sum of the capacities of all energy storage devices is less than the first comprehensive capacity, the absolute value of the difference between the sum of the capacities of all energy storage devices and the first comprehensive capacity at the current moment is denoted as Cu. An adjustment coefficient k is defined, k = Cu / C0, where C0 represents the sum of the capacities of all energy storage devices. The correction coefficient is updated using an adjustment factor k, and the magnitude of the update is positively correlated with k. If the sum of the capacities of all energy storage devices is greater than or equal to the first combined capacity, the correction factor will not be updated.

[0009] Preferably, the specific steps for obtaining the first fluctuation change feature are as follows: At each time point, the frequency of the power grid collected within a first preset time period prior to each time point is acquired to form a first frequency sequence; Obtain the first mean of all frequencies in the first frequency sequence, calculate the absolute value of the difference between the first mean and the preset reference frequency, and obtain the frequency offset feature. The frequency offset feature is positively correlated with the absolute value. The first frequency sequence is filtered to obtain the filtered sequence. The difference sequence between the first frequency sequence and the filtered sequence is obtained, and the frequency fluctuation characteristics are obtained. The frequency fluctuation characteristics are positively correlated with the mean of the absolute values ​​of all elements in the difference sequence.

[0010] Preferably, the specific steps for obtaining the second fluctuation change characteristic are as follows: At the current moment, obtain the load power at each moment within the second preset time period before the current moment and form a load power sequence; obtain the power output of all energy storage devices at each moment within the second preset time period before the current moment and form an energy storage power sequence. All elements in the energy storage power series and the load power series are linearly normalized together. The average difference between the elements at the same position in the normalized load power series and the normalized energy storage power series is recorded as the relative power difference. The Pierce correlation coefficient between the normalized energy storage power series and the normalized load power series is used to assess the difference in the trend. The second fluctuation variation characteristic is positively correlated with the difference in relative power and the difference in variation trend, respectively.

[0011] Preferably, the step of updating the correction coefficient based on the difference between the total capacity of all energy storage devices and the first and second input capacities includes the following specific steps: When F1 is less than or equal to the preset warning threshold, several energy storage devices in the grid-connected state are randomly selected, and the energy storage devices in the grid-connected state are switched to the grid-connected state in turn, so that the capacity of all energy storage devices in the grid-connected state is greater than or equal to the second input capacity. When F1 exceeds the preset warning threshold, several energy storage devices in grid-connected state are randomly selected and switched to grid-connected state in sequence, so that the capacity of all energy storage devices in grid-connected state is greater than or equal to the first input capacity.

[0012] Preferably, the method of updating the correction coefficient using an adjustment coefficient k, wherein the update magnitude is positively correlated with k, includes the following specific steps: At the current moment, the first coefficient in the correction coefficients before the update is denoted as m10, and the first coefficient after the update is set as m10×(1+k); At the current moment, the second coefficient in the correction coefficients before the update is denoted as m20, and the second coefficient after the update is set as m20×(1-k).

[0013] Another embodiment of the present invention provides an integrated source-grid-load-storage configuration optimization system that takes into account external influences. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor runs the computer program, it implements all the steps of the above-described integrated source-grid-load-storage configuration optimization method that takes into account external influences.

[0014] The beneficial effects of the technical solution of the present invention are: By monitoring the fluctuation characteristics of the power grid frequency in real time and dynamically adjusting the virtual parameters (virtual moment of inertia and damping coefficient) output by the MPC algorithm using correction coefficients, energy storage devices in the grid configuration can more accurately suppress frequency fluctuations. This closed-loop correction mechanism effectively avoids the response lag problem caused by the fixed parameters of traditional MPC.

[0015] By analyzing the fluctuation characteristics of load power, the required capacity of energy storage devices in grid-connected state is dynamically assessed; by analyzing the fluctuation characteristics of frequency, the required capacity of energy storage devices in grid-connected state is also dynamically assessed. This method enables the energy storage system to track load changes in real time, reducing reliance on direct power output from new energy sources; and enables the energy storage system to support the stability of grid frequency.

[0016] By comparing the total energy storage capacity with the dynamically assessed input capacity (first input capacity in grid-connected state + second input capacity in grid-connected state), the control strategy for virtual parameters is corrected in reverse. This allows the energy storage devices to coordinate reasonably between grid-connected and grid-connected states, with mutual feedback and dependence, forming a parameter adaptive closed loop under capacity constraints. This approach aims to simultaneously ensure the stability of grid frequency and load, minimizing instability issues in a grid with high penetration of new energy sources due to external environmental influences. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the steps of an integrated source-grid-load-storage configuration optimization method that takes into account external influences, as provided in an embodiment of the present invention. Detailed Implementation

[0019] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the integrated source-grid-load-storage configuration optimization method and system considering external influences proposed according to the present invention. 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.

[0020] Unless otherwise defined, 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 invention pertains.

[0021] The following description, in conjunction with the accompanying drawings, details the specific scheme of the integrated source-grid-load-storage configuration optimization method and system that takes into account external influences provided by the present invention.

[0022] Example 1: Please see Figure 1 The diagram illustrates a flowchart of a source-grid-load-storage integrated configuration optimization method considering external influences, according to an embodiment of the present invention. The method includes the following steps: Step S101: The working state of energy storage devices in the integrated power grid includes grid connection state and grid connection state. Grid connection state refers to the energy storage devices being connected to the power grid as virtual synchronous generators. The virtual rotational inertia and virtual damping coefficient of the virtual synchronous generators are controlled by the MPC algorithm.

[0023] In this embodiment, it is considered that for power supply areas with high penetration of new energy sources, the power grid in such areas often has the following characteristics: the synchronous generators in the power station are replaced by power electronic equipment such as wind power and photovoltaics, the total moment of inertia is greatly reduced, when a power deficit occurs, the frequency drops faster and lower, and after a disturbance, it is difficult to establish a new active power balance, the frequency may deviate from the rated value for a long time and cannot be automatically restored.

[0024] Based on the above characteristics, the stability of the power grid can be maintained to a certain extent by constructing an integrated power grid of source, grid, load and storage.

[0025] However, when considering the impact and disturbances of external factors on the power grid, such as changes in weather, aging and failure of power electronic equipment, large changes in electricity demand in a short period of time, and the inability to match and balance electricity demand and supply, the power grid often experiences unpredictable disturbances. This can lead to the inability of the integrated power grid (source, grid, load, and storage) to operate stably for a long time as intended. For example, the access strategy or capacity of energy storage devices in the power grid may not be able to adapt to the dynamic and unpredictable state of the power grid.

[0026] Therefore, it is necessary to optimize the configuration of the integrated power grid of source, grid, load and storage. This embodiment takes the optimization of energy storage equipment as an example to describe how the optimized power grid can adapt to the state of the power grid and maintain the stability of the power grid as much as possible.

[0027] Specifically, for all energy storage devices in the integrated power grid, each energy storage device in this embodiment has two working states: grid-connected state and grid-connected state. Each energy storage device can switch between these two states.

[0028] In grid-connected mode, energy storage devices are integrated into the power grid as virtual synchronous generators. These virtual synchronous generators possess virtual rotational inertia (simulating the physical inertia of a synchronous generator). When grid frequency fluctuates, they absorb or release electrical energy using their simulated inertia, mitigating frequency changes and providing a stable foundation for the grid. The virtual synchronous generator also has a virtual damping coefficient (simulating the damping effect of a synchronous generator), effectively suppressing power oscillations. Furthermore, it can automatically adjust reactive power output through droop control to support grid voltage.

[0029] The energy storage device connected to the grid in grid-connected mode adopts active-reactive power control mode (PQ mode). In this mode, the energy storage device can track the grid voltage and frequency in real time and inject (or absorb) power into the grid, so that the grid is as stable as possible at the target power (set by the EMS module of the energy storage device), in order to suppress grid load fluctuations, such as avoiding grid overload and realizing peak shaving and valley filling. The target power mentioned in this embodiment is set as the rated power of the energy storage device connection point.

[0030] It should be noted that, due to the high penetration of new energy sources into the integrated power grid, grid, load, and storage system, the grid is highly unstable under external influences. This means that the virtual moment of inertia and virtual damping coefficient set by the energy storage equipment in the grid operation state cannot dynamically adapt to the complex and ever-changing power grid, and thus cannot effectively maintain the stability of the power grid for a long time.

[0031] Based on this, this embodiment employs a Model Predictive Control (MPC) algorithm to control the virtual moment of inertia and virtual damping coefficient. The general working principle of MPC in this embodiment is as follows: For any given moment, after the energy storage device acquires the grid frequency using phase-locked loop (PLL) technology, it first predicts the frequencies for N subsequent moments (excluding the given moment) using the MPC algorithm. The control quantity of the MPC algorithm consists of the virtual moment of inertia and virtual damping coefficient. Then, based on the predicted N frequencies, the MPC algorithm optimizes the control quantity for that moment through quadratic programming. The control quantity for that moment refers to the virtual moment of inertia and virtual damping coefficient of the virtual synchronous generator (i.e., the energy storage device in grid-connected state) operating at that moment and after that moment (i.e., the virtual moment of inertia and virtual damping coefficient obtained under MPC algorithm control). Furthermore, in this embodiment, the grid reference frequency (rated frequency) is 50Hz; therefore, the MPC algorithm sets the reference value for the N subsequent moments to 50Hz.

[0032] The MPC algorithm is a well-known technology, and will not be described in detail in this embodiment. Furthermore, this embodiment uses 1 second as a time interval; in other embodiments, considering the long execution time of the MPC algorithm (e.g., due to insufficient computing power of the hardware device running the MPC algorithm), a 2-second time interval can be used as a time interval.

[0033] For ease of description, the virtual moment of inertia and virtual damping coefficient are referred to as virtual parameters in this embodiment.

[0034] Step S102: Monitor the first fluctuation change characteristics of the frequency in the power grid in real time, and use a preset correction coefficient to correct the virtual parameters under the control of the MPC algorithm according to the first fluctuation change characteristics.

[0035] In this embodiment, a portion of the energy storage devices are randomly assigned to operate in a grid-connected state, while another portion operates in a grid-following state. For example, 30% of the energy storage devices operate in a grid-connected state, and 70% operate in a grid-following state. This embodiment does not impose any specific limitations.

[0036] As the power grid operates (note that energy storage devices in both grid-connected and grid-connected states operate according to the method in step S101), at each moment, the frequency of the power grid collected within a first preset time period prior to each moment is acquired, forming a first frequency sequence. In this embodiment, the first preset time period refers to one minute prior to each moment (including each moment). This embodiment uses phase-locked loop (PLL) technology to collect the power grid frequency, which collects the frequency once every 0.1 seconds.

[0037] Obtain the first fluctuation change characteristics of the first frequency sequence, and correct the virtual parameters under the control of the MPC algorithm based on the first fluctuation change characteristics.

[0038] In this process, the first fluctuation characteristic describes the frequency deviation and frequency fluctuation that still exist in the power grid under the frequency support of the energy storage devices in the grid-connected state. The more severe the frequency deviation and frequency fluctuation, the more difficult it is for the energy storage devices in the grid-connected state to further maintain the stability of the power grid, even under the control of the MPC algorithm. One reason for this is that the virtual parameters under the control of the MPC algorithm are inaccurate, making it difficult to effectively suppress abnormal situations such as frequency deviation and frequency oscillation. The less obvious the frequency deviation and frequency fluctuation, the more likely it is that the energy storage devices in the grid-connected state can maintain the stability of the power grid under the control of the MPC algorithm, and the virtual parameters effectively suppress abnormal situations such as frequency deviation and frequency oscillation.

[0039] In the above process, the virtual parameters under the control of the MPC algorithm are corrected according to the first fluctuation change characteristics, which helps to make the virtual parameters under the control of the MPC algorithm more accurate and further restore the grid frequency to stability.

[0040] As an example, the method for obtaining the first fluctuation change feature of the first frequency sequence includes: Obtain the first mean of all frequencies in the first frequency sequence, calculate the absolute value of the difference between the first mean and the reference frequency, and denote the ratio of this absolute value to a preset normalization coefficient as the frequency offset characteristic. The larger the frequency offset characteristic, the more significant the shift in the real-time frequency of the power grid relative to the reference frequency; the smaller the frequency offset characteristic, the less significant the shift in the real-time frequency of the power grid.

[0041] The first frequency sequence is filtered using a mean filter with a kernel length of 9, resulting in a filtered sequence. The difference sequence between the first frequency sequence and the filtered sequence is then obtained (i.e., the difference is calculated for elements at the same position). The ratio of the mean of the absolute values ​​of all elements in the difference sequence to a preset normalization coefficient is denoted as the frequency fluctuation characteristic. A larger frequency fluctuation characteristic indicates more drastic fluctuations in the elements of the first frequency sequence, suggesting significant unstable fluctuations in the power grid frequency.

[0042] It should be noted that in this embodiment, the step size of the mean filter is 1, and the length of the filtered sequence and the first frequency sequence are the same by padding the first frequency sequence with 0s.

[0043] Thus, frequency offset characteristics and frequency fluctuation characteristics were obtained at each time point. The frequency change characteristics of the historical period before each time point were described from the two aspects of the offset and fluctuation of the power grid frequency. The tuple composed of frequency offset characteristics and frequency fluctuation characteristics was used as the first fluctuation change characteristic obtained at each time point.

[0044] It should be noted that in this embodiment, whether the frequency fluctuation amplitude in the power grid exceeds 2Hz within a short period of time is a criterion for judging whether a major fault has occurred in the power grid. Based on this, the preset normalization coefficient is set to 2Hz in this embodiment to remove dimensions and orders of magnitude. Specifically, when the frequency offset feature and frequency fluctuation feature are both greater than 0.95, the frequency offset feature and frequency fluctuation feature are set to 0.95 respectively. Specifically, when the average value of the frequency offset feature or the average value of the frequency fluctuation feature obtained within 10 consecutive seconds is greater than 0.95, it indicates that the power grid may be in a major fault. At this time, this embodiment will no longer be run, and some loads need to be discarded, such as directly disconnecting all loads, or issuing an early warning so that relevant power grid management personnel can disconnect the loads and carry out manual maintenance.

[0045] As an example, the virtual parameters under the control of the MPC algorithm are corrected based on the characteristics of the first fluctuation, including the following methods: The larger the first fluctuation change characteristic, the greater the error of the virtual parameter under the control of the MPC algorithm. When correcting the virtual parameter, a larger correction is required. In this example, the correction magnitude is positively correlated with the first fluctuation change characteristic.

[0046] As an example, the formulas for modifying virtual parameters are as follows: Let the frequency offset feature and frequency fluctuation feature in the first wave change feature be denoted as a and b, respectively. Let the virtual moment of inertia and virtual damping coefficient (i.e., virtual parameters) obtained at each moment under the control of the MPC algorithm be denoted as J0 and D0, respectively.

[0047] The larger 'a' is, the greater the deviation of the grid frequency from the reference frequency. In this case, it is more necessary to significantly reduce J0. By reducing J0, the grid frequency can be made to quickly approach the original state (that is, to quickly approach the reference frequency) with the support of the virtual synchronous generator. This avoids the consequences of the virtual synchronous generator becoming "sluggish" in its frequency response due to an excessively large J0, which would lead to a continuous frequency deviation.

[0048] Based on this, the correction magnitude of J0 is positively correlated with a; as an example, the correction result of J0 is J=J0×(1-m1×a), where m1 represents the preset first coefficient, which is used to describe the degree of attention paid to frequency deviation during correction.

[0049] The larger b is, the more severe the oscillations are when the frequency of the power grid changes. In this case, D0 needs to be increased significantly. Increasing D0 increases the oscillation suppression capability of the virtual synchronous generator, thereby smoothing out frequency fluctuations.

[0050] Based on this, the correction magnitude of D0 is positively correlated with b; as an example, the correction result of D0 is D=D0×(1+m2×b), where m2 represents the preset second coefficient, which is used to describe the degree of attention paid to the frequency oscillation change during correction.

[0051] In this embodiment, m1 and m2 are collectively referred to as correction coefficients, which are used to describe the degree of attention paid to historical frequency changes when correcting virtual parameters. As an example, the initial values ​​of m1 and m2 are set to 0.5, and the preferred range of their initial values ​​is (0, 1).

[0052] Thus, at each time point, after executing the MPC algorithm and obtaining the virtual parameters under the control of the MPC algorithm, the virtual parameters are corrected, and the energy storage devices in the grid state at subsequent times operate using the corrected virtual parameters.

[0053] In special cases, if less than one minute has passed before the current time, the virtual parameters are not corrected (or it is determined that the virtual parameters before and after the correction are the same).

[0054] Step S103: Real-time detection of the second fluctuation characteristics of load power in the power grid, and assessment of the second operational capacity of the energy storage device in grid status based on the second fluctuation characteristics.

[0055] For all low-voltage substations (e.g., 220V box-type substations) that directly supply power to users in the power grid, power meters are installed on the low-voltage output bus of all substations. Power is collected every second. At each moment, the average of all power collected by all power meters in the most recent hour is obtained and recorded as the load power at each moment.

[0056] It should be noted that reactive power droop control exists in the grid-connected state, while the PQ mode is used in the grid-connected state. Moreover, the reactive power in the power grid is less affected by external factors. Therefore, reactive power is ignored in this embodiment. The power collected above refers to active power. In addition, in this embodiment, all energy storage device access points are located on the line between the new energy power station and the low-voltage substation.

[0057] At the current moment, all load power obtained within a second preset time period prior to the current moment is acquired to form a load power sequence. In this embodiment, the second preset time period is set to the most recent 24 hours, including the current moment.

[0058] This embodiment obtains the second fluctuation change characteristics of the load power sequence, and evaluates the second operational capacity of the energy storage device in grid-connected status based on the second fluctuation change characteristics.

[0059] The second fluctuation characteristic describes the load changes in the power grid over a historical period, supported by all energy storage devices. A large second fluctuation characteristic indicates that, even with the support of all energy storage devices, it is impossible to effectively suppress load fluctuations in the power grid. For example, there may be situations such as high electricity demand during peak hours, overload of power equipment (e.g., low-voltage substations), unstable electricity demand, and large differences in electricity demand between off-peak and peak hours. All energy storage devices in the power grid are unable to cope with or resolve these situations. This is because the capacity of the grid-connected energy storage devices is insufficient, or in other words, the capacity of the grid-connected energy storage devices at the current moment is insufficient to meet the dynamic electricity demand of users. In this case, it is necessary to optimize and reconfigure the capacity of the grid-connected energy storage devices.

[0060] It should be noted that the capacity mentioned in this embodiment refers to power capacity, which represents the rate at which energy is absorbed or released per unit time. When the energy storage device's power is equal to 100% or less than 15%, it no longer absorbs or releases energy, and the capacity at this time is considered to be 0.

[0061] As an example, obtaining the second fluctuation variation characteristics of the load power sequence includes the following steps: A power meter is also installed at the grid connection point of each energy storage device (including those in grid-connected and those in stand-alone grid states) to collect the power absorbed or released by each energy storage device (specifically, active power), with power collected every second. At each time moment, the average of all power collected by the power meters of all energy storage devices at that time moment and within the most recent hour is obtained and recorded as the energy storage power at that time moment (released power is recorded as a positive number, and absorbed power as a negative number).

[0062] The energy storage power obtained within the second preset time period prior to the current moment constitutes the energy storage power sequence. All elements in both the energy storage power sequence and the load power sequence are linearly normalized together. The purpose of this is to eliminate the order of magnitude of the elements and focus only on the relative differences between the energy storage power sequence and the load power sequence.

[0063] The second fluctuation characteristic is obtained from the relative power difference and trend difference between the normalized energy storage power sequence and the normalized load power sequence.

[0064] The relative power difference represents the difference between elements at the same position in the normalized energy storage power sequence and the normalized load power sequence. The trend difference describes the correlation between the normalized energy storage power sequence and the normalized load power sequence. A larger relative power difference and trend difference indicate, on the one hand, a significant difference between the power released by all energy storage devices and the power demanded by users. This suggests that energy storage devices are outputting less power to meet user electricity needs, relying more heavily on renewable energy sources. In this case, the power regulation function of energy storage devices is weak, which is detrimental to grid steady-state adjustment. For example, when renewable energy output fluctuates greatly, it cannot provide stable power supply to users. On the other hand, it indicates that the output power of energy storage devices cannot match or keep up with user electricity demand. For example, during peak electricity demand periods, it is expected that energy storage devices will release sufficient power, but in reality, the released power is insufficient or non-existent. Similarly, during off-peak electricity demand periods, it is expected that energy storage devices will absorb power for storage to be released later during peak demand periods, but in reality, the energy storage devices do not absorb enough power, resulting in low energy utilization. Based on the above two aspects, this embodiment needs to expand the capacity of energy storage devices in grid-connected mode, thereby alleviating the problems mentioned above.

[0065] As an example, the second fluctuation characteristic is obtained from the relative power difference and trend difference between the normalized energy storage power sequence and the normalized load power sequence, including the following methods: The difference between each element in the normalized load power sequence and each element in the normalized energy storage power sequence is calculated. The mean of all the differences is denoted as the relative power difference. Specifically, when the relative power difference is less than 0, the relative power difference is set to 0.

[0066] Calculate the Pielsen correlation coefficient between the normalized energy storage power series and the normalized load power series, denoted as r. Let exp(-r) represent the difference in trends. A smaller Pielsen correlation coefficient indicates that the normalized energy storage power series and the normalized load power series have significantly different trends, and the greater the difference in trends. Specifically, when the difference in trends is greater than or equal to 1, we set the difference in trends to 1.

[0067] The second fluctuation variation characteristic is positively correlated with the difference in relative power and the difference in variation trend, respectively.

[0068] As an example, the second fluctuation change characteristic is equal to the mean of the relative power difference and the difference in change trend.

[0069] As an example, the method for assessing the second operational capacity of energy storage devices based on the second fluctuation change characteristics and grid connection status includes: The sum of the capacities of all grid-connected energy storage devices at the current moment is denoted as C20; the second input capacity is denoted as C2, the sum of the capacities of all energy storage devices (including grid-connected and grid-connected devices) is denoted as C0, and the second fluctuation characteristic is denoted as F2. Let C2 = C20 + x2 × C0 × F2, which means that a portion of the capacity will be added on the basis of the total capacity of the current grid-connected energy storage devices. The added capacity is F2 × x2 times the total capacity of all energy storage devices, where x2 represents the preset second expansion coefficient, which is used to describe the attention paid to the grid load situation in historical periods when expanding the capacity of grid-connected energy storage devices. In this embodiment, x2 = 0.5 is used as an example for description.

[0070] Step S104: Evaluate the first operational capacity of the energy storage equipment in the grid state based on the first frequency fluctuation characteristic F obtained under the corrected virtual parameters.

[0071] The first fluctuation characteristic obtained at the current moment is denoted as F. The larger F is (that is, the larger the frequency offset and frequency fluctuation characteristics in F), the more frequency changes and instability still exist after the MPC algorithm is corrected in step S102. This indicates that simply correcting the virtual parameters cannot further guarantee the stability of the power grid. The reason is that the capacity of the energy storage devices in the grid configuration is insufficient and cannot provide enough power throughput for the operation of the energy storage devices. At this time, it is necessary to expand the capacity of the energy storage devices in the grid configuration. The larger the first fluctuation characteristic F1 is, the more capacity needs to be expanded.

[0072] Based on this, the initial input capacity is positively correlated with F.

[0073] As an example, methods for obtaining the initial input capacity include: The sum of the capacities of the energy storage devices in the current grid-connected state is denoted as C10. The first installed capacity is denoted as C1, and the sum of the capacities of all energy storage devices is denoted as C0. Let C1 = C10 + x1 × C0 × F1. This indicates that a portion of the capacity will be added based on the total capacity of the energy storage devices in the current grid-connected state. The added capacity is F1 × x1 times the total capacity of all energy storage devices, where x1 represents the preset first expansion coefficient, which is used to describe the degree of attention paid to the grid load situation in the historical period when expanding the capacity of the energy storage devices in the grid-connected state. In this embodiment, x1 = 0.5 is used as an example. F1 represents the average value of the frequency offset characteristic and frequency fluctuation characteristic in F.

[0074] Step S105: Update the correction coefficients based on the differences between the total capacity of all energy storage devices and the first and second input capacities, respectively.

[0075] The first and second input capacities were obtained at the current moment. The sum of the first and second input capacities is denoted as the first comprehensive capacity.

[0076] When the total capacity of all energy storage devices (i.e., the sum of the capacities of all energy storage devices) is less than the first comprehensive capacity, it indicates that the capacity of the existing energy storage devices in the power grid cannot meet the current capacity demand.

[0077] One approach is to modify the power grid to accommodate more energy storage devices. However, this method takes a long time and cannot meet current capacity demands in a short period.

[0078] Based on this, in this embodiment, the control and correction process of the virtual parameters in step S102 is fed back according to the difference between the total capacity of all energy storage devices and the first comprehensive capacity, thereby reducing the capacity demand of energy storage devices in the grid-connected state, and ultimately reducing the current capacity demand. This avoids as much as possible the problem that the capacity of existing energy storage devices is insufficient, which leads to the inability to simultaneously meet the capacity requirements of energy storage devices in the grid-connected and grid-connected states.

[0079] Specifically, since the output power of the energy storage devices in the grid state is mainly used to power the operation of the virtual synchronous generator, in this embodiment, when the difference between the total capacity of all energy storage devices and the first comprehensive capacity is less than 0, and the smaller the difference is, the more necessary it is to reduce the power consumption of the virtual synchronous generator during operation. This embodiment reduces the power consumption of the virtual synchronous generator during operation by avoiding setting the virtual moment of inertia and virtual damping coefficient of the virtual synchronous generator too large.

[0080] Based on this, after obtaining the difference between the total capacity of all energy storage devices and the first comprehensive capacity at the current moment, if the difference is less than 0, the absolute value of the difference is recorded as Cu, and an adjustment coefficient k is defined, k=Cu / C0, where C0 represents the total capacity of all energy storage devices. Using it as the denominator is for normalization and dimension removal.

[0081] This embodiment achieves the goal of reducing the power consumption of the virtual synchronous generator during operation by updating the correction coefficient in S102.

[0082] When the correction factor is updated, the virtual synchronous generator is run using the updated correction factor in subsequent times.

[0083] As an example, the methods for updating the correction coefficient include: In the correction coefficients before the update at the current moment, the first coefficient is represented as m10, and the updated first coefficient is set as m10×(1+k). At this time, the larger k is (that is, the less the capacity of the existing energy storage devices in the grid can meet the current capacity demand), the more necessary it is to significantly reduce the capacity demand of the grid-connected energy storage devices (reduce their power throughput). This means that the larger the updated first coefficient is, the smaller the virtual rotational inertia can be when the virtual synchronous generator is running. Based on step S102, this reduces the power consumption of the virtual synchronous generator during operation and reduces its power throughput.

[0084] In the correction coefficients before the update at the current moment, the second coefficient is denoted as m20, and the updated second coefficient is set to m20×(1-k). At this time, the larger k is, the smaller the updated second coefficient is. Therefore, when running the virtual synchronous generator in the future, it is possible to avoid a large virtual damping coefficient, thereby reducing the power consumption of the virtual synchronous generator during operation.

[0085] Therefore, the updated first and second coefficients are used as the updated correction coefficients.

[0086] If the difference between the total capacity of all energy storage devices and the first comprehensive capacity is greater than or equal to 0, the correction factor will not be updated.

[0087] At this point, the first input capacity, the second input capacity, and the correction coefficient have been obtained and updated at the current moment.

[0088] Next, randomly select some energy storage devices in the grid-connected state and switch them to the grid-connected state in turn. Stop switching when the capacity of all grid-connected energy storage devices is greater than or equal to the second input capacity.

[0089] Specifically, when F1 exceeds the preset warning threshold, it indicates that the grid frequency has severely lost its steady state. In this case, to ensure the stability of the grid frequency and avoid large-scale power outages, some energy storage devices in grid-connected mode are randomly selected and switched to grid-connected mode in sequence. Switching stops when the capacity of all grid-connected energy storage devices is greater than or equal to the first operational capacity. If F1 is less than or equal to the preset warning threshold, the grid-connected mode is switched to grid-connected mode according to the above method.

[0090] The preset warning threshold in this embodiment is 0.8, and the preferred value range is [0.6, 0.9].

[0091] This concludes the example.

[0092] When the next moment arrives, continue to execute the following methods included in this embodiment: (1) Monitor the first fluctuation change characteristics of the frequency in the power grid in real time, and use the updated correction coefficient to correct the virtual parameters under the control of the MPC algorithm again based on the first fluctuation change characteristics; (2) Calculate the second input capacity and the first input capacity; (3) Use the obtained first fluctuation change characteristics to update the correction coefficient again based on the second input capacity and the first input capacity; (4) Change the working state of the energy storage device.

[0093] The above method is followed at every subsequent time point.

[0094] In this process, energy storage devices in grid-connected mode support grid frequency stability, while those in grid-connected mode regulate and stabilize user electricity load, such as facilitating peak shaving and valley filling. This embodiment optimizes the capacity of energy storage devices in different operating states, enabling dynamic switching between grid-connected and grid-connected modes. The two modes can coordinate and provide feedback effectively, maximizing the balance between grid frequency and load stability and minimizing instability issues in grids with high renewable energy penetration due to external environmental influences.

[0095] Example 2: Step S102 in Embodiment 1 includes: correcting the virtual parameters; as another example, it includes the following formula: When a is small, for example, when a is less than the threshold th1, it means that the frequency of the power grid has not deviated significantly from the reference frequency. In this case, J0 can be increased to make the virtual synchronous generator more "slow" to respond to frequency changes, thereby suppressing further frequency deviation.

[0096] Based on this, when a is less than the threshold th1, the correction result J of J0 is J = J0 × (1 + th1 - a); when a is greater than or equal to the threshold th1, it is implemented according to Example 1.

[0097] Furthermore, considering that when b is small, for example, when b is less than the threshold th2, it indicates that there is no obvious oscillation in the frequency of the power grid. In this case, D0 can be reduced, which helps to reduce the damping when the virtual synchronous generator recovers to the reference frequency.

[0098] Based on this, when a is less than the threshold th2, the correction result of D0 is D=D0×[1-(th2-b)]; when b is greater than or equal to the threshold th2, it is implemented according to Example 1.

[0099] As an example, th1=th2=0.07, and the preferred value range is [0.005, 0.1].

[0100] Step S103 of Embodiment 1 includes: assessing the second operational capacity of the energy storage device based on the second fluctuation change characteristics; as another example, the included formula is: Considering that when F2 is small, for example, when F2 is less than or equal to th3, it indicates that the second fluctuation characteristic is small, which means that the grid load has sufficient stability with the support of all energy storage devices. In this case, the capacity of the energy storage devices connected to the grid is not expanded.

[0101] In some preferred embodiments, the capacity of the energy storage device in grid-connected state can also be reduced. In this case, the second input capacity C2 = C20 - C0 × (th3 - F2) means that the smaller F2 is, the greater the capacity reduction. The purpose is that the reduced capacity can be switched to grid-connected state, thereby providing reliable support for the stability of the grid frequency.

[0102] Specifically, in implementing step S105: some grid-connected energy storage devices are randomly selected and switched to grid-connected status in sequence, and the switching stops when the capacity of all grid-connected energy storage devices is less than or equal to the second input capacity.

[0103] When F2 is greater than th3, then the implementation shall be carried out in accordance with Example 1.

[0104] As an example, th3 = 0.3; its preferred value range is [0.1, 0.5].

[0105] Example 3: In steps S103 and S104 of Embodiment 1, a first expansion coefficient x1 and a second expansion coefficient x2 are set. This embodiment provides another method for setting x1 and x2: At the current moment, obtain C20, F2, C10, and F1 from each moment in history, and use these data as a sample.

[0106] At this point, multiple samples have been obtained.

[0107] Step S104 includes C1 = C10 + x1 × C0 × F1, where C1 is considered a parametric curve about the independent variable F1, and the parameter is x1; step S103 includes the formula C2 = C20 + x2 × C0 × F2, where C2 is considered a parametric curve about the independent variable F2, and the parameter is x2. This embodiment aims to achieve the following: after substituting all samples into the above parametric curves, C1 and C2 are equal to the sum of the capacities of all energy storage devices, C0.

[0108] The specific method is as follows: for each sample, after substituting the above parameter curves, we obtain C1 and C2. The mean of C1 obtained from all samples is denoted as c1, and the mean of C2 obtained from all samples is denoted as c2.

[0109] Let the loss function A pair of x1 and x2 is randomly initialized. The simulated annealing algorithm is used to obtain the values ​​of x1 and x2 when Q is minimized, and these values ​​are used as the x1 and x2 set at the current time.

[0110] This process ensures that the x1 and x2 set at the current moment are as close as possible to or equal to the sum of the first input capacity and the second input capacity, thus avoiding the error introduced when x1 and x2 are set inappropriately in step S105 when the correction coefficients are updated based on the difference between the total capacity of all energy storage devices and the first input capacity and the second input capacity.

[0111] In addition, this embodiment takes into account the need to avoid the impact on the power grid when switching energy storage devices from grid-connected to grid-connected status, or switching energy storage devices from grid-connected to grid-connected status. Therefore, this embodiment switches only one energy storage device at any given time, prioritizing the switching of the energy storage device with the smallest capacity.

[0112] In other embodiments, an integrated source-grid-load-storage configuration optimization system that takes into account external influences is provided. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor runs the computer program, it implements all the steps of all the embodiments described above.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the integrated configuration of power generation, grid, load, and storage considering external influences, characterized in that, The method includes the following steps: The working states of energy storage devices in the integrated power grid include grid connection state and grid connection state. Grid connection state refers to the energy storage device being connected to the grid as a virtual synchronous generator. The virtual moment of inertia and virtual damping coefficient of the virtual synchronous generator are controlled by the MPC algorithm. The virtual moment of inertia and virtual damping coefficient are referred to as virtual parameters. In grid connection state, the energy storage device is connected to the grid using an active-reactive control mode. The first frequency fluctuation characteristic in the power grid is monitored in real time, and the virtual parameters under the control of the MPC algorithm are corrected according to the first fluctuation characteristic using a preset correction coefficient. The second fluctuation characteristic of the load power in the power grid is detected in real time. The second input capacity of the energy storage device in the grid-connected state is evaluated based on the second fluctuation characteristic. The first input capacity of the energy storage device in the grid-connected state is evaluated based on the first fluctuation characteristic F of the frequency obtained under the corrected virtual parameters. The first input capacity is positively correlated with the first fluctuation characteristic F, and the second input capacity is positively correlated with the second fluctuation characteristic. The correction coefficients are updated based on the differences between the total capacity of all energy storage devices and the first and second input capacities, respectively.

2. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 1, characterized in that, The specific steps involved in correcting the virtual parameters under the MPC algorithm control using a preset correction coefficient based on the first fluctuation change characteristic are as follows: The first fluctuation change characteristics include frequency offset characteristics and frequency fluctuation characteristics, denoted as a and b respectively; the virtual parameters obtained at each moment under the control of the MPC algorithm include virtual moment of inertia and virtual damping coefficient, denoted as J0 and D0 respectively; Decrease the value of J0, with the magnitude of the decrease being positively correlated with a and m1 respectively; increase the value of D0, with the magnitude of the increase being positively correlated with b and m2 respectively; use the decreased J0 and increased D0 as the corrected virtual parameters; where m1 represents the preset first coefficient and m2 represents the preset second coefficient, and m1 and m2 are collectively referred to as the correction coefficients; After each moment, the energy storage device, acting as a virtual synchronous generator, operates with the corrected virtual parameters.

3. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 1, characterized in that, The specific steps involved in assessing the second operational capacity of the energy storage device based on the second fluctuation change characteristics and its grid connection status are as follows: The sum of the capacities of all grid-connected energy storage devices at the current moment is denoted as C20; the sum of the capacities of all energy storage devices is denoted as C0; and the second fluctuation characteristic is denoted as F2. The second input capacity is equal to the capacity expanded based on C20, where the expanded capacity is determined by F2 times C0.

4. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 1, characterized in that, The specific steps for evaluating the first operational capacity of energy storage devices based on the first frequency fluctuation characteristic F obtained under the corrected virtual parameters are as follows: The first fluctuation change feature F includes frequency offset feature and frequency fluctuation feature, and the mean of frequency offset feature and frequency fluctuation feature is denoted as F1; The sum of the capacities of the energy storage devices in the current grid state is denoted as C10, and the sum of the capacities of all energy storage devices is denoted as C0. The first input capacity is equal to the capacity expanded based on C10, where the expanded capacity is determined by F1 times C0.

5. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 2, characterized in that, The specific steps involved in updating the correction coefficient based on the difference between the total capacity of all energy storage devices and the first and second operational capacities are as follows: The sum of the first and second input capacities obtained at the current moment is denoted as the first comprehensive capacity; When the sum of the capacities of all energy storage devices is less than the first comprehensive capacity, the absolute value of the difference between the sum of the capacities of all energy storage devices and the first comprehensive capacity at the current moment is denoted as Cu. An adjustment coefficient k is defined, k = Cu / C0, where C0 represents the sum of the capacities of all energy storage devices. The correction coefficient is updated using an adjustment factor k, and the magnitude of the update is positively correlated with k. If the sum of the capacities of all energy storage devices is greater than or equal to the first combined capacity, the correction factor will not be updated.

6. The integrated source-grid-load-storage configuration optimization method taking into account external influences according to any one of claims 1, 2, or 4, characterized in that, The specific steps for obtaining the first fluctuation change feature are as follows: At each time point, the frequency of the power grid collected within a first preset time period prior to each time point is acquired to form a first frequency sequence; Obtain the first mean of all frequencies in the first frequency sequence, calculate the absolute value of the difference between the first mean and the preset reference frequency, and obtain the frequency offset feature. The frequency offset feature is positively correlated with the absolute value. The first frequency sequence is filtered to obtain the filtered sequence. The difference sequence between the first frequency sequence and the filtered sequence is obtained, and the frequency fluctuation characteristics are obtained. The frequency fluctuation characteristics are positively correlated with the mean of the absolute values ​​of all elements in the difference sequence.

7. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 1, characterized in that, The specific steps for obtaining the second fluctuation change characteristic are as follows: At the current moment, obtain the load power at each moment within the second preset time period before the current moment and form a load power sequence; obtain the power output of all energy storage devices at each moment within the second preset time period before the current moment and form an energy storage power sequence. All elements in the energy storage power series and the load power series are linearly normalized together. The average difference between the elements at the same position in the normalized load power series and the normalized energy storage power series is recorded as the relative power difference. The Pierce correlation coefficient between the normalized energy storage power series and the normalized load power series is used to assess the difference in the trend. The second fluctuation variation characteristic is positively correlated with the difference in relative power and the difference in variation trend, respectively.

8. The integrated source-grid-load-storage configuration optimization method considering external influences according to claim 4, characterized in that, The step of updating the correction coefficient based on the difference between the total capacity of all energy storage devices and the first and second input capacities, and then including the following specific steps: When F1 is less than or equal to the preset warning threshold, several energy storage devices in the grid-connected state are randomly selected, and the energy storage devices in the grid-connected state are switched to the grid-connected state in turn, so that the capacity of all energy storage devices in the grid-connected state is greater than or equal to the second input capacity. When F1 exceeds the preset warning threshold, several energy storage devices in grid-connected state are randomly selected and switched to grid-connected state in sequence, so that the capacity of all energy storage devices in grid-connected state is greater than or equal to the first input capacity.

9. The integrated source-grid-load-storage configuration optimization method considering external influences as described in claim 5, characterized in that, The method of updating the correction coefficient using an adjustment factor k, wherein the magnitude of the update is positively correlated with k, includes the following specific steps: At the current moment, the first coefficient in the correction coefficients before the update is denoted as m10, and the first coefficient after the update is set as m10×(1+k); At the current moment, the second coefficient in the correction coefficients before the update is denoted as m20, and the second coefficient after the update is set as m20×(1-k).

10. An integrated power generation, grid, load, and storage configuration optimization system taking into account external influences, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor runs the computer program, it implements all the steps of the source-grid-load-storage integrated configuration optimization method according to any one of claims 1 to 9, taking into account external influences.

Images