Virtual power plant control method and device, equipment, storage medium and program product

Abstract

The invention discloses a control method and device of a virtual power plant, equipment, a storage medium and a program product, and the method comprises the steps: predicting a predicted power error of a next moment of a current moment through a power deviation sequence of a previous time window of the current moment; and adjusting a control influence factor of the accumulated power error at the next moment according to the magnitude relationship between the accumulated power error of the current time window to which the current moment belongs and an error threshold value, and determining the control power of the virtual power plant according to the control influence factor, the accumulated power error and the predicted power error. And controlling the operation of the virtual power plant at the next moment based on the control power. According to the method, external disturbance is predicted and feedforward compensation is carried out, and feedback correction is carried out in combination with a real-time response result of a system, so that collaborative optimization regulation and control of various distributed energy resources in a virtual power plant are realized.

Description

Technical Field

[0001] The present disclosure generally relates to the field of control, and more particularly to a control method for a virtual power plant, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] A virtual power plant is not a physical power plant in the traditional sense. Instead, it is a "virtualized" power system that aggregates dispersed distributed energy resources (such as photovoltaic, wind power, energy storage, electric vehicles, adjustable loads, etc.) through advanced information and communication technology (ICT) and intelligent algorithms, forming a power system that can be uniformly dispatched and operate collaboratively. It does not rely on a physical power plant but can participate in power market transactions and grid dispatching like a traditional power plant. The efficient operation of a virtual power plant depends on its three core components: distributed energy, energy storage system, and intelligent control system. They cooperate with each other to jointly form the "intelligent brain" of the virtual power plant.

[0003] Currently, open-loop control methods are mostly used in the actual control of virtual power plants at home and abroad, that is, the target power is allocated according to the preset target power curve and the adjustable ability evaluated for each aggregated adjustable resource. However, for distributed resources with weak control rights such as electric vehicle charging loads, the errors generated by open-loop control will be uncontrollable. Summary of the Invention

[0004] In view of the above defects or deficiencies in the prior art, it is desired to provide a control method for a virtual power plant, a computer device, a computer-readable storage medium, and a computer program product that can improve the control accuracy of the virtual power plant.

[0005] In a first aspect, a control method for a virtual power plant is provided, the method including: Obtaining a power deviation sequence of a previous time window at the current moment; Predicting a predicted power error at the next moment of the current moment according to the power deviation sequence of the previous time window; Obtaining the cumulative power error of the current time window to which the current moment belongs, and adjusting the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and the error threshold; Determining the control power of the virtual power plant according to the control influence factor, the cumulative power error, and the predicted power error, and controlling the operation of the virtual power plant at the next moment based on the control power.

[0006] The control method of the virtual power plant provided by this application takes into account that the existing open-loop control methods for the virtual power plant will produce uncontrollable errors. This application predicts the predicted power error at the next moment of the current moment through the power deviation sequence in the previous time window of the current moment, and adjusts the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error in the current time window to which the current moment belongs and the error threshold. Then, based on the control influence factor, the cumulative power error, and the predicted power error, it determines the control power of the virtual power plant, and controls the operation of the virtual power plant at the next moment based on the control power. This method predicts external disturbances and performs feedforward compensation, and at the same time combines the real-time response results of the system for feedback correction to achieve the coordinated and optimized control of various distributed energy resources in the virtual power plant. This method has both fast response ability and robustness, can improve the response speed and regulation accuracy of the virtual power plant, and is applicable to real-time regulation scenarios under high uncertainty environments.

[0007] In a second aspect, a control device for a virtual power plant is provided. The device includes: An acquisition module, configured to acquire the power deviation sequence in the previous time window of the current moment; A prediction module, configured to predict the predicted power error at the next moment of the current moment according to the power deviation sequence in the previous time window; An acquisition and adjustment module, configured to acquire the cumulative power error in the current time window to which the current moment belongs, and adjust the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and the error threshold; A determination module, configured to determine the control power of the virtual power plant according to the control influence factor, the cumulative power error, and the predicted power error, and control the operation of the virtual power plant at the next moment based on the control power.

[0008] In a third aspect, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the control method of the virtual power plant in the first aspect.

[0009] In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the control method of the virtual power plant in the first aspect.

[0010] In a fourth aspect, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, it implements the control method of the virtual power plant in the first aspect. Description of the Drawings

[0011] Other features, objectives, and advantages of the present application will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings: Figure 1 It is a flowchart of the steps of a control method for a virtual power plant provided by the present application; Figure 2 It is a flowchart of the steps of another control method for a virtual power plant provided by the present application; Figure 3 It is a flowchart of the steps of another control method for a virtual power plant provided by the present application; Figure 4 It is a flowchart of the steps of another control method for a virtual power plant provided by the present application; Figure 5 It is a flowchart of the steps of another control method for a virtual power plant provided by the present application; Figure 6 It is a schematic structural diagram of a control device for a virtual power plant provided by the present application. Detailed implementation manners

[0012] The present application will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention and are not intended to limit the invention. Additionally, it should be noted that for the convenience of description, only the parts related to the invention are shown in the drawings.

[0013] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0014] Currently, in the aspect of the closed-loop control method of a virtual power plant, it can be divided into an energy management strategy based on the Markov decision process, a regulation strategy based on robust optimization, and a regulation strategy based on model predictive control; among them, the Markov decision process method relies on a large amount of historical data for training, while the regulation data of the actual virtual power plant is very little and has characteristics such as privacy, complexity, and high confidentiality, so it is obviously not suitable for actual popularization and application; the robust optimization method tends to provide conservative solutions, which puts very high requirements on the accuracy of the evaluation of the regulation ability of the virtual power plant, and this is obviously not suitable in actual applications either; model predictive control can perform rolling optimization according to the real-time information of the virtual power plant and is more suitable for the real-time optimization control of the virtual power plant. Currently, the conventional model predictive control method needs to model the internal resources of the virtual power plant and then perform online solution according to the set optimization objectives, which will obviously cause a control lag in the actual application of the method.

[0015] In view of the above problems, the present application online corrects the control power to be issued according to the predicted future possible error and the current real-time cumulative error, so that the output power of the virtual power plant fits the target power, improving the control accuracy.

[0016] The following combines Figure 1 to explain the control method of the virtual power plant provided by the present application. Please refer to Figure 1 , Figure 1 which is the step flowchart of a control method for a virtual power plant provided by an exemplary embodiment of the present application. The step flowchart includes the following steps: Step S20, obtaining the power deviation sequence of the previous time window at the current moment; Among them, the current moment refers to the real-time time point when this "obtaining" operation occurs, which can be any moment during the control of the virtual power plant; the time window is a time interval with a fixed length, and this length is preset in advance, for example, it can be 30 minutes, 1 hour, 50 minutes, etc. The previous time window refers to the fixed-length time interval immediately before the current moment.

[0017] The power deviation sequence refers to a set of data points arranged in chronological order, and this set of data points represents the value of the control power deviation. In the present application, all "control power deviation" values obtained within the time period of the "previous time window" can be extracted from the stored historical data, and these values are arranged in chronological order to obtain the power deviation sequence of the previous time window at the current moment.

[0018] Exemplarily, assume that the time window length is 1 second. Then at the current moment t, the previous time window refers to the interval from t - 1 second to t seconds. What needs to be obtained is the 10 control power deviations recorded within the interval from t - 1 second to t seconds, and they are sorted in time series to obtain the power deviation sequence.

[0019] Optionally, the present application can obtain the above power deviation sequence through the following method: Collect the first operating power at each moment within the previous time window; Subtract the issued power from each first operating power respectively to obtain a plurality of first power deviations; Arrange the plurality of first power deviations in chronological order to obtain the power deviation sequence.

[0020] Here, it needs to be explained that the issued power is a set value that allows the virtual power plant to operate at this power value in an ideal state. However, often due to the influence of other factors, there is a deviation between the actual operating power of the virtual power plant and the issued power. Therefore, it is necessary to obtain this deviation and optimize the subsequent control of the virtual power plant according to this deviation.

[0021] Step S30: Predict the predicted power error at the next moment of the current moment based on the power deviation sequence of the previous time window. After obtaining the power deviation sequence of the previous time window according to the above method, the control power deviation of the virtual power plant at the next moment can be predicted in advance based on this sequence. That is, the future control power deviation is predicted based on the historical power deviation, converting passive response into active pre-control, achieving the technical effect of compensating for the predicted deviation in advance and reducing the fluctuation of the actual operating power. In addition, adjusting the control strategy before the disturbance in this application can improve the resilience of the control system.

[0022] This application can use the power deviation sequence of the previous time window as the model input, organize the data through a sliding time window structure to capture the temporal dependence and fluctuation law of the power deviation sequence, and then use a time series prediction algorithm to analyze the evolution trend of the power deviation sequence and output the predicted power error at the next moment. Exemplarily, this application can use an LSTM model or the like to implement the above operations.

[0023] Step S40: Obtain the cumulative power error of the current time window to which the current moment belongs, and adjust the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and the error threshold. Based on the above explanation of the previous time window, the current time window has the same time length as the previous time window. The current time window is a fixed-length time interval adjacent to the previous time window, and the current moment belongs to this time interval.

[0024] The cumulative power error refers to the sum of the control power errors at all moments in the current time window. The error threshold is a preset critical value used to determine whether the cumulative power error exceeds the allowable range. The control influence factor is a proportional coefficient when the cumulative power error exerts a control effect on the virtual power plant, and is used to adjust the control influence of the cumulative power error on the virtual power plant.

[0025] The magnitude relationship between the cumulative power error and the error threshold includes two cases: the cumulative power error is greater than or equal to the error threshold and the cumulative power error is less than the error threshold. In this application, when the cumulative power error is greater than or equal to the error threshold, the control influence factor can be increased to more aggressively eliminate the cumulative deviation of the power; in addition, when the cumulative power error is less than the error threshold, the control influence factor can be decreased to avoid over-regulation causing oscillations. Of course, in this application, when the cumulative power error is less than the error threshold, the control influence factor can also not be adjusted to maintain the current control intensity.

[0026] Exemplarily, the present application includes a relationship information table, which includes the correspondence between the magnitude relationship between the cumulative power error and the error threshold and the control influence factor, for example: Magnitude relationship between cumulative power error and error threshold Control influence factor Less than a Greater than or equal to b When the magnitude relationship between the cumulative power error and the error threshold is determined, the control influence factor can be directly read from the relationship information table, and then the control influence factor of the cumulative power error at the next moment can be adjusted to the corresponding value.

[0027] Exemplarily, the present application can also adjust the control influence factor of the cumulative power error at the next moment through the following formula: , where a next is the control influence factor, k1 is the first adjustment coefficient (such as 1.2 - 1.5), k2 is the second adjustment coefficient (such as 0.6 - 0.8), a curret is the reference influence factor, is a preset value, for example 1; E cum is the error threshold, E th is the cumulative power error.

[0028] Optionally, the present application can obtain the cumulative power error according to the following steps: Collect the second operating power at each moment within the current time window; Subtract the issued power from each second operating power to obtain multiple second power deviations; Sum up the multiple second power deviations to obtain the cumulative power error.

[0029] Among them, the issued power is as explained above, and the calculation method of the control power deviation is the same and will not be elaborated here.

[0030] Step S50, determine the control power of the virtual power plant according to the control influence factor, the cumulative power error and the predicted power error, and control the operation of the virtual power plant at the next moment based on the control power.

[0031] Among them, the control influence factor determines the influence degree of the cumulative power error and the predicted power error on the control instruction of the issued target power at the next moment. The cumulative power error reflects the control deviation direction of the system, while the predicted power error quantifies an expected deviation of uncertainty. The combination of the three, through the combination of the prediction of external disturbances introduced by forward-looking input processing (predicted power error) and the closed-loop feedback correction (cumulative power error), realizes the dynamic optimization and coordinated control of multi-source heterogeneous resources, effectively enhances the stability of the system, and improves the response speed and control accuracy of the virtual power plant.

[0032] Exemplarily, the control power of the virtual power plant can be determined according to the following formula: , where P is the control power of the virtual power plant, and P1 is the preset reference power; a next is the control influence factor, E cum is the error threshold, and E th is the cumulative power error.

[0033] In an alternative embodiment, as Figure 2 shown, Figure 2 is an alternative method embodiment for determining the predicted power error provided by an exemplary embodiment of the present application. The method embodiment includes the following steps: Step S201: Decompose the high-frequency part and the low-frequency part in the power deviation sequence to obtain a high-frequency component and a low-frequency component; Among them, the high-frequency part in the power deviation sequence reflects the instantaneous fluctuation of the control power deviation and requires rapid response adjustment. The low-frequency part in the power deviation sequence controls the long-term trend of the control power deviation and corresponds to the steady-state regulation demand of the control system.

[0034] The present application can, for example, decompose the high-frequency part and the low-frequency part in the power deviation sequence to obtain a high-frequency component and a low-frequency component according to discrete wavelet transform, wavelet inverse transform, Fourier filtering, VMD algorithm, etc.

[0035] Step S202: Synthesize the predicted power error based on the high-frequency component and the low-frequency component.

[0036] Among them, after separating the high-frequency component and the low-frequency component according to the above steps, the present application can then synthesize the predicted power error according to the following formula.

[0037] , where E pred is the predicted power error, E L is the low-frequency component, and E H is the high-frequency component.

[0038] By predicting the power error through frequency division, the present application can improve the prediction accuracy and reduce the control error.

[0039] In another alternative embodiment, the present application can synthesize the predicted power error according to the method steps as Figure 3 shown: Step S301: Perform filtering estimation processing on the high-frequency component to obtain a corrected high-frequency component; Among them, the core purpose of filtering and estimating the high-frequency components is to preserve the effective fluctuation characteristics and suppress the noise interference, so as to improve the prediction and control accuracy.

[0040] In this application, for example, after obtaining the high-frequency components through wavelet decomposition technology, the effective fluctuations and interference noises in the high-frequency components can be distinguished by Kalman filtering, threshold filtering or frequency-domain band-pass filtering technology, and then inverse transformation and reconstruction can be carried out through technologies such as wavelet reconstruction and IFFT to obtain the pure and corrected high-frequency components.

[0041] By processing the high-frequency components, high-frequency noise can be eliminated and the true fluctuation characteristics can be retained, so as to achieve the technical effects of suppressing high-frequency noise, preventing the control system from oscillating, and reducing harmonic resonance.

[0042] Step S302: Capture the evolution law of the low-frequency components, predict the low-frequency components according to the evolution law, and obtain the corrected low-frequency components; Among them, the core logic of low-frequency component correction is: by mining the regular patterns in historical data, a prediction model is constructed to actively compensate for system deviations.

[0043] In this application, after separating the low-frequency components through technologies such as STL decomposition and Hodrick-Prescott filtering, the deterministic evolution laws such as periodicity and trend hidden in the low-frequency components are identified, and then the low-frequency components are predicted according to the evolution laws to obtain the corrected low-frequency components.

[0044] Exemplarily, in this application, the periodic evolution law hidden in the low-frequency components can be obtained through Fourier series fitting technology. The trend evolution law hidden in the low-frequency components can be obtained through polynomial regression / Prophet model. The corrected low-frequency components are output by inputting the periodic evolution law and the trend evolution law into the LSTM neural network.

[0045] Optionally, in this application, the low-frequency components can also be input into the LightGBM model, and the corrected low-frequency components are obtained according to the prediction of the model.

[0046] Step S303: Synthesize the predicted power error according to the corrected high-frequency components and the corrected low-frequency components.

[0047] Among them, exemplarily, this application can be synthesized according to the following formula: , ΔP low-corrected is the corrected low-frequency component; ΔPhigh-filtered is the corrected high-frequency component; Rresidual is the preset residual compensation term; α, β, γ are dynamic weight coefficients, for example, α = 0.7, β = 0.25, γ = 0.05.

[0048] In yet another alternative embodiment, the present application may synthesize the predicted power error according to the method steps as Figure 4 shown: Step S401: Sort the corrected high-frequency components and the corrected low-frequency components in a preset arrangement order to obtain a component sequence. The preset arrangement order is that the corrected low-frequency components are in the front, and the corrected high-frequency components are arranged in reverse order according to the decomposition levels; Among them, in the present application, placing the corrected low-frequency components at the starting position of the sequence and arranging the corrected high-frequency components in reverse order of the original decomposition levels to obtain the component sequence can ensure strict correspondence between the high-frequency components and the low-frequency components. In addition, this can preferentially retain the main trend features, and the reverse arrangement of the high-frequency components conforms to the physical logic of signal reconstruction.

[0049] Step S402: Through the linear combination of wavelet basis functions, fuse the time-frequency domain information of the high-frequency part and the low-frequency part in the component sequence into a complete time-domain signal to obtain the predicted power error.

[0050] Among them, the core process of using wavelet transform to realize signal reconstruction is essentially to fuse multi-scale components into a time-domain signal through orthogonal projection.

[0051] Exemplarily, the present application may be through the following formula: , △Perror(t) is the predicted power error, ϕ j,k (t) is the corrected low-frequency component; ψ j,k (t) is the corrected high-frequency component; c L,k is the preset low-frequency component coefficient; d j,k is the preset high-frequency component coefficient.

[0052] In an alternative embodiment, the present application may adjust the control influence factor through the following steps. This method embodiment includes the following steps: If the cumulative power error is less than the error threshold, then adjust the control influence factor to be smaller according to the preset adjustment integral gain; If the cumulative power error is greater than or equal to the error threshold, then use the preset adjustment integral gain as the control influence factor.

[0053] Among them, the above description can be expressed by the following formula: , is the control influence factor, is the error threshold, is the preset adjustment integral gain, and e(t - 1) is the cumulative power error.

[0054] In an alternative embodiment, as Figure 5 shown, Figure 5 This is an alternative method embodiment for determining the control power of a virtual power plant provided by an exemplary embodiment of the present application. The method embodiment includes the following steps: Step S501: Update the cumulative power error according to the control influence factor to obtain the target cumulative power error; Step S502: Determine the control power according to the target cumulative power error and the predicted power error.

[0055] Among them, updating the cumulative power error according to the control influence factor is, for example, taking the product of the control influence factor and the cumulative power error as the target cumulative power error.

[0056] Determining the control power according to the target cumulative power error and the predicted power error can be, for example, taking the sum of the target cumulative power error and the predicted power error as the control power, and controlling the operation of the virtual power plant at the next moment through this control power.

[0057] It is represented by the formula as follows: , where is the target cumulative power error, is the control influence factor, is the predicted power error.

[0058] It should be noted here that the above control method can be executed cyclically within a control time window. That is to say, after determining the control power of the virtual power plant at the next moment and controlling the operation of the virtual power plant according to this control power, before the control time window ends, it is still necessary to continue to obtain the power deviation sequence of the previous time window at the current moment, but at this time the current moment is replaced by the next moment, that is, the next moment is used as the current moment, and the operation of determining the control power of the virtual power plant at the next moment is executed cyclically until the control time window ends.

[0059] It should be noted that although the operations of the method of the present invention are described in a specific order in the drawings, this does not require or imply that these operations must be executed in that specific order, or that all the operations shown must be executed to achieve the desired result. On the contrary, the steps depicted in the flowchart can change the execution order.

[0060] Further referring to Figure 6 , as Figure 6 shown, Figure 6 The present application provides a control device for a virtual power plant. The control device includes an acquisition module 601, a prediction module 602, an acquisition and adjustment module 603, and a determination module 604.

[0061] An acquisition module 601, configured to acquire a power deviation sequence of a previous time window at the current moment; A prediction module 602, configured to predict a predicted power error at the next moment of the current moment according to the power deviation sequence of the previous time window; An acquisition and adjustment module 603, configured to acquire an accumulated power error of the current time window to which the current moment belongs, and adjust a control influence factor of the accumulated power error at the next moment according to a magnitude relationship between the accumulated power error and an error threshold; A determination module 604, configured to determine a control power of the virtual power plant according to the control influence factor, the accumulated power error, and the predicted power error, and control the operation of the virtual power plant at the next moment based on the control power.

[0062] In an optional embodiment, the prediction module 602 is specifically configured to decompose a high-frequency part and a low-frequency part in the power deviation sequence to obtain a high-frequency component and a low-frequency component; Synthesize a predicted power error based on the high-frequency component and the low-frequency component.

[0063] In an optional embodiment, the prediction module 602 is specifically further configured to perform a filtering estimation process on the high-frequency component to obtain a corrected high-frequency component; Capture an evolution law of the low-frequency component, predict the low-frequency component according to the evolution law, and obtain a corrected low-frequency component; Synthesize a predicted power error according to the corrected high-frequency component and the corrected low-frequency component.

[0064] In an optional embodiment, the prediction module 602 is specifically further configured to sort the corrected high-frequency component and the corrected low-frequency component according to a preset sorting order to obtain a component sequence, and the preset sorting order is that the corrected low-frequency component is in the front, and the corrected high-frequency component is arranged in reverse order according to the decomposition level; Fuse time-frequency domain information of the high-frequency part and the low-frequency part in the component sequence into a complete time domain signal through a linear combination of wavelet basis functions to obtain a predicted power error.

[0065] In an optional embodiment, the acquisition and adjustment module 603 is specifically configured to, if the magnitude relationship is that the accumulated power error is less than the error threshold, reduce the control influence factor according to a preset adjustment integral gain; If the magnitude relationship is that the accumulated power error is greater than or equal to the error threshold, use the preset adjustment integral gain as the control influence factor.

[0066] In an optional embodiment, the determination module 604 is specifically configured to update the accumulated power error according to the control influence factor to obtain a target accumulated power error; Determine a control power according to the target accumulated power error and the predicted power error.

[0067] Each module in the above control device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0068] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0069] The units or modules described in the embodiments of the present application can be implemented in software or in hardware. The described units or modules can also be provided in a processor. For example, it can be described as: a processor includes XX unit, YY unit, and ZZ unit. Among them, the names of these units or modules do not constitute a limitation on the units or modules themselves in some cases. For example, the XX unit can also be described as "a unit for XX".

[0070] As another aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium can be the computer-readable storage medium included in the device described in the above embodiments; or it can exist separately and be unassembled into the device. The computer-readable storage medium stores one or more programs, and the one or more programs are used by one or more processors to execute the control method of the virtual power plant described in the present application.

[0071] The above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) disclosed in the present application that have similar functions.

Claims

1. A control method for a virtual power plant, characterized in that The method includes: Obtaining a power deviation sequence of the previous time window at the current moment; Predicting a predicted power error at the next moment of the current moment according to the power deviation sequence of the previous time window; Obtaining the cumulative power error of the current time window to which the current moment belongs, and adjusting the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and an error threshold; Determining the control power of the virtual power plant according to the control influence factor, the cumulative power error, and the predicted power error, and controlling the operation of the virtual power plant at the next moment based on the control power.

2. The control method according to claim 1, wherein The predicting the predicted power error at the next moment of the current moment according to the power deviation sequence of the previous time window includes: Decomposing the high-frequency part and the low-frequency part in the power deviation sequence to obtain a high-frequency component and a low-frequency component; Synthesizing the predicted power error based on the high-frequency component and the low-frequency component.

3. The control method according to claim 2, characterized in that The synthesizing the predicted power error based on the high-frequency component and the low-frequency component includes: Performing a filtering estimation process on the high-frequency component to obtain a corrected high-frequency component; Capturing the evolution law of the low-frequency component, predicting the low-frequency component according to the evolution law, and obtaining a corrected low-frequency component; Synthesizing the predicted power error according to the corrected high-frequency component and the corrected low-frequency component.

4. The control method according to claim 3, characterized in that, The synthesizing the predicted power error according to the corrected high-frequency component and the corrected low-frequency component includes: Sorting the corrected high-frequency component and the corrected low-frequency component according to a preset arrangement order to obtain a component sequence, where the preset arrangement order is that the corrected low-frequency component is in the front, and the corrected high-frequency component is arranged in reverse order according to the decomposition level; Fusing the time-frequency domain information of the high-frequency part and the low-frequency part in the component sequence into a complete time domain signal through a linear combination of wavelet basis functions to obtain the predicted power error.

5. The control method according to claim 1, characterized in that, The adjusting the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and the error threshold includes: If the magnitude relationship is that the cumulative power error is less than the error threshold, then adjusting the control influence factor to be smaller according to a preset adjustment integral gain; If the magnitude relationship is that the cumulative power error is greater than or equal to the error threshold, then using the preset adjustment integral gain as the control influence factor.

6. The control method according to claim 1, characterized in that, The determining the control power of the virtual power plant according to the control influence factor, the cumulative power error, and the predicted power error includes: Updating the cumulative power error according to the control influence factor to obtain a target cumulative power error; Determining the control power according to the target cumulative power error and the predicted power error.

7. A control device for a virtual power plant, characterized in that, The device includes: An acquisition module for acquiring a power deviation sequence of the previous time window at the current moment; A prediction module for predicting a predicted power error at the next moment of the current moment according to the power deviation sequence of the previous time window; An acquisition and adjustment module, configured to acquire the cumulative power error of the current time window to which the current moment belongs, and adjust the control influence factor of the cumulative power error at the next moment according to the magnitude relationship between the cumulative power error and the error threshold; A determination module, configured to determine the control power of the virtual power plant according to the control influence factor, the cumulative power error, and the predicted power error, and control the operation of the virtual power plant at the next moment based on the control power.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images