Frequency modulation sequence prediction method based on frequency domain resonance enhancement and random differential driving

Abstract

The invention relates to the technical field of power systems, and particularly provides a frequency modulation sequence prediction method based on frequency domain resonance enhancement and random differential driving. The method comprises the following steps: constructing a frequency domain mapping and resonance enhancement filter; performing frequency domain enhancement and inverse transformation; constructing a stochastic differential equation SDE for dynamic smoothing; discretizing the SDE to obtain a final optimized sequence; according to the method, the accurate prediction of the frequency modulation sequence of the power grid is improved, and the stability of the frequency of the power grid is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, and particularly relates to a frequency modulation sequence prediction method based on frequency domain resonance enhancement and random differential driving. BACKGROUND

[0002] Under the background of accelerating the construction of new power systems, the grid frequency fluctuation characteristics become increasingly complex due to the high proportion of renewable energy connected to the grid, and higher requirements are put forward for frequency regulation capability. Hybrid energy storage systems, due to their millisecond response speed and flexible power support capability, have become the key equipment for improving the frequency stability of the power grid. The core control logic of the hybrid energy storage system depends on the accurate prediction of the power grid frequency modulation instruction sequence, and the prediction accuracy is directly related to the optimization of the charge and discharge strategy of the energy storage unit, the life management and the overall frequency modulation effect. However, the power grid frequency modulation sequence is essentially a time domain mapping of the unbalanced power of the power grid, and is coupled by multiple factors such as unit start-stop, load mutation, intermittent renewable energy output, etc., showing strong non-stationary, nonlinear and high noise characteristics. Such sequences often contain multiple time scale characteristics: both the slow trend component determined by the system inertia and the instantaneous sharp pulse caused by random disturbances, and traditional time series prediction models cannot effectively extract generalizable regularity features from the original data.

[0003] The existing prediction technology usually relies on direct processing of the original sequence, which faces fundamental bottlenecks. First, although conventional preprocessing methods such as Kalman filtering or moving average can smooth noise, they have significant drawbacks: using fixed parameters and linear assumptions, they inevitably weaken the burst component and short time domain features in the sequence that carry key dynamic information while filtering high-frequency noise, resulting in degradation of the input signal features of the prediction model. Second, although methods based on frequency domain such as wavelet transform can extract multi-scale features, the selection of decomposition modes is highly dependent on prior knowledge, and the reconstruction prediction of different mode components has phase lag and energy leakage problems, making it difficult to ensure time consistency. The deeper problem is that existing methods generally treat noise reduction and feature extraction as two independent stages, and fail to construct a preprocessing process from the perspective of system dynamic evolution, resulting in a sequence that has improved statistical characteristics but unclear physical meaning, and the prediction model is difficult to establish a reliable causal relationship between the input and the output. SUMMARY

[0004] Therefore, the present application provides a frequency modulation sequence prediction method based on frequency domain resonance enhancement and random differential driving, to improve the accurate prediction of the power grid frequency modulation sequence and improve the stability of the power grid frequency.

[0005] In a first aspect, the present application provides a frequency modulation sequence prediction method based on frequency domain resonance enhancement and random differential driving, which comprises: Step 1, constructing a frequency domain mapping and resonance enhancement filter; Step 2, according to step 1, frequency domain enhancement and inverse transform are performed; Step 3, based on step 2, a stochastic differential equation (SDE) is constructed for dynamic smoothing; Step 4, using step 3, the SDE is discretized to obtain the final optimized sequence; Step 5, according to step 4, the frequency modulation sequence is predicted to obtain a predicted sequence.

[0006] Optionally, the step 1 comprises: First, the original frequency modulation sequence is mapped to the frequency domain by fast Fourier transform (FFT), and the frequency domain signal expression of the frequency domain transform is: ; wherein, ; is a complex component in the frequency domain, i is the imaginary unit.

[0007] Optionally, the step 1 further comprises: a resonance enhancement filter is constructed, the bandwidth and peak position of which are related to the characteristic frequency of the sequence, and the expression of the resonance enhancement filter is: ; wherein, is the angular frequency corresponding to the k th frequency point, ; ; the bandwidth parameter ; =2 and =0.5 are gain and offset constants, respectively, for controlling the amplitude of enhancement.

[0008] Optionally, the step 2 comprises: the frequency domain signal is multiplied by the resonance enhancement filter to obtain an enhanced frequency domain signal , which is then mapped back to the time domain by inverse fast Fourier transform (IFFT), and the expression is: ; ; wherein, is the intermediate sequence after frequency domain resonance enhancement.

[0009] Optionally, the step 3 comprises: the enhanced intermediate sequence The external force considered as a virtual Brownian motion particle is a variant of the Ornstein-Uhlenbeck (OU) process, and its steady-state solution is taken as the final output; a stochastic differential equation (SDE) is defined, and its expression is as follows: ; wherein, is the optimized sequence; is the time-varying long-term mean, which is set as , i.e. the enhanced sequence; is the mean reversion rate parameter, = 0.5, which is used to control the speed of towards ; is the volatility parameter, ; is the virtual Wiener process increment.

[0010] Optionally, the step 4 comprises: The SDE is discretized and solved by using the Euler-Maruyama method to obtain the final optimized sequence, and its expression is as follows: .

[0011] Optionally, the step 5 comprises: The original frequency modulation sequence is converted into , and is predicted, and the prediction formula is as follows: ; wherein, = 1, is an independent and identically distributed standard normal random variable, i.e. ~ N (0, 1); , avg represents the mean; .

[0012] Optionally, comprises: predicting , and the prediction formula is as follows: ; wherein, ; ; and so on, to predict ; Q is the length of the predicted sequence, Q is less than , represents rounding.

[0013] In a second aspect, an embodiment of the present application provides a computer readable storage medium, the computer readable storage medium comprising a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform the method for predicting frequency modulation sequence based on frequency domain resonance enhancement and stochastic differential drive.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs comprise instructions that, when executed by the device, cause the device to perform the method for predicting frequency modulation sequence based on frequency domain resonance enhancement and stochastic differential drive.

[0015] The technical solution provided by the present application comprises the following steps: constructing a frequency domain mapping and resonance enhancement filter; performing frequency domain enhancement and inverse transformation; constructing a stochastic differential equation (SDE) for dynamic smoothing; discretizing the SDE to obtain a final optimized sequence; and predicting the frequency modulation sequence to obtain a predicted sequence. The method improves the accurate prediction of the power grid frequency modulation sequence and improves the stability of the power grid frequency. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 A flowchart of the method for predicting frequency modulation sequence based on frequency domain resonance enhancement and stochastic differential drive provided by an embodiment of the present application is shown in the following figure: Figure 2 A schematic diagram of an electronic device provided by an embodiment of the present application is shown in the following figure: DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] It should be noted that the described embodiments are merely some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] The terms used in the embodiments of the present application are merely for the purpose of describing the specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0021] It should be understood that the term "and / or" used herein is merely a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are in an "or" relationship.

[0022] Depending on the context, the word "if" as used herein can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if it is determined" or "if (a stated condition or event) is detected" can be interpreted as "when it is determined" or "in response to determining" or "when (a stated condition or event) is detected" or "in response to detecting (a stated condition or event)".

[0023] Figure 1 The flowchart of the frequency domain resonance enhancement and random differential driving based frequency modulation sequence prediction method provided by the embodiments of the present application is shown in Figure 1 The method comprises: Step 1, constructing a frequency domain mapping and resonance enhancement filter.

[0024] In the embodiments of the present application, step 1 comprises: First, the original frequency modulation sequence is mapped to the frequency domain through fast Fourier transform (FFT), and the frequency domain signal expression of the frequency domain transformation is: ; Wherein, ; is a complex component in the frequency domain, i is an imaginary unit.

[0025] In the embodiments of the present application, the original frequency modulation sequence is mapped from the time domain to the frequency domain, and the frequency domain components in the sequence are identified, which provides a basis for subsequent frequency domain enhancement.

[0026] In the embodiments of the present application, step 1 further comprises: Constructing resonance enhancement filter , the gain of which reaches a peak value near a specific frequency, the bandwidth and the peak position of which are related to the characteristic frequency (system core frequency) of the sequence , the expression of the resonance enhancement filter is: ; Wherein, is the angular frequency corresponding to the k th frequency point, ; The frequency point with the strongest spectral energy (take the frequency point with the strongest spectral energy); the bandwidth parameter (Narrow bandwidth to enhance selectivity); =2 and =0.5 are gain and offset constants, respectively, for controlling the amplitude of enhancement.

[0027] In the embodiment of the present application, the resonance enhancement filter forms a fourth power steep drop characteristic; =2 can amplify the target frequency band, and the offset constant can avoid the gain of the low frequency band to be zero, =0.5 retains the basic signal. The fourth power design makes the gain in the target frequency band close to + , and the outside quickly decays to , which significantly improves the signal-to-noise ratio of the key frequency band.

[0028] Step 2, according to step 1, frequency domain enhancement and inverse transformation are performed.

[0029] In the embodiment of the present application, step 2 includes: The frequency domain signal is multiplied by the resonance enhancement filter to obtain an enhanced frequency domain signal , which is mapped back to the time domain through inverse Fourier transform IFFT, and the expression is: ; ; Wherein, is the intermediate sequence after frequency domain resonance enhancement.

[0030] In the embodiment of the present application, the enhanced frequency domain signal is reconstructed into a time domain sequence , which retains the physical nature while amplifying the key features.

[0031] Step 3, based on step 2, a stochastic differential equation SDE is constructed for dynamic smoothing.

[0032] In the embodiment of the present application, step 3 includes: The enhanced intermediate sequence The external force is regarded as a virtual Brownian motion particle, a variant of the Ornstein-Uhlenbeck (OU) process is constructed, and the steady-state solution is used as the final output; a stochastic differential equation (SDE) is defined, and the expression is as follows: Wherein, is the optimized sequence (continuous form); is the time-varying long-term mean, which is set to , that is, the enhanced sequence; is the mean reversion rate parameter, = 0.5, which is used to control the speed of reversion (the larger , , the faster tracks ); is the volatility parameter, , the volatility is associated with the standard deviation of the original sequence;

[0033] is the virtual Wiener process (Brownian motion) increment. In the embodiment of the application, the physical meaning of the SDE is: As a virtual smooth frequency modulation power, it is subjected to a force that reverts to the current enhanced signal , and is subjected to a random disturbance, and the solution will be a sequence that fluctuates around the enhanced signal but is smoother, that is, fluctuates around the enhanced signal , balances the tracking ability and smoothness,

[0034] and retains reasonable random disturbance.

[0035] In the embodiment of the application, step 4 comprises: The Euler-Maruyama method is used to discretize and solve the SDE to obtain the final optimized sequence, and the expression is as follows: .

[0036] Step 5, according to step 4, the frequency modulation sequence is predicted to obtain a predicted sequence.

[0037] In the embodiment of the application, step 5 comprises: The original frequency modulation sequence is converted to , and Prediction is made, and the prediction formula is: ; Wherein, =1, is a standard normal random variable, that is ~N(0,1); (dynamic updated historical mean), avg denotes the mean value; .

[0038] In the embodiment of the application, the prediction is made on , and the prediction formula is: ; Wherein, , and the mean value is ensured to be corrected in real time; ; By analogy, the prediction of is made; Q is the length of the prediction sequence, Q is less than , denotes rounding.

[0039] In the embodiment of the application, the dynamic makes the prediction continuously adapt to the sequence change, the historical value weight of the term is attenuated, and the recent trend is highlighted.

[0040] In the frequency modulation system, the original frequency modulation sequence , the subscript indicates the sampling number, sampling once per second, N ≥300; the core of the method is to regard the original frequency modulation sequence as an observation value of a virtual physical system (such as a frequency modulation energy field), map it to the frequency domain, design a resonance enhancement filter to amplify a specific frequency band, and the specific frequency band contains key prediction information. Subsequently, a stochastic differential equation (SDE) is introduced to simulate the system dynamics, and finally the SDE is discretized and solved, as an optimized sequence, to obtain a smoother and more regular sequence.

[0041] In the embodiment of the application, the simulation data verification is as follows: Input data: let the original frequency modulation sequence =[0.1,0.5,-0.2,0.7,-0.4], N =5, and the target prediction X 6; First, frequency domain enhancement is performed: is obtained by fast Fourier transform (FFT): : =[0.7,-0.25+0.4i,-0.1,-0.25-0.4i]; Computing ||=argmax ||=[0.7,0.47,0.1,0.47]argmax ||=1, = / 5; Constructing resonance-enhanced filter ( =2, =0.5): =[2.5,1.8,0.5,1.8]argmax at Inverse transform gives =[0.12,0.52,-0.18,0.68,-0.37].

[0042] Second, SDE dynamic smoothing: Parameters: =0.5, =0.1×std )=0.05; Initialization , =0.12; Recursion : , take =0.3, then ≈0.14; Final optimized sequence =[0.12,0.14,-0.05,0.41,-0.12].

[0043] Finally, prediction : = avg ( )=0.1; ; Predicted value .

[0044] The technical scheme provided by the application comprises the following steps: constructing a frequency domain mapping and a resonance enhancement filter; performing frequency domain enhancement and inverse transformation; constructing a stochastic differential equation (SDE) for dynamic smoothing; discretizing the SDE to obtain a final optimized sequence; and predicting the frequency modulation sequence to obtain a predicted sequence.

[0045] The various steps of the embodiment of the application can be executed by an electronic device. The electronic device includes, but is not limited to, a tablet computer, a portable PC, a desktop computer, and the like.

[0046] The embodiment of the application provides a computer readable storage medium, which comprises a stored program, wherein the program controls an electronic device in which the computer readable storage medium is located to execute the embodiment of the frequency modulation sequence prediction method based on frequency domain resonance enhancement and stochastic differential driving when the program is running.

[0047] Figure 2 An electronic device provided by the embodiment of the application is shown in FIG. 1. Figure 2 As shown in FIG. 1, the electronic device 21 comprises a processor 211, a memory 212, and a computer program 213 stored in the memory 212 and executable on the processor 211, wherein the computer program 213 is executed by the processor 211 to implement the frequency modulation sequence prediction method based on frequency domain resonance enhancement and stochastic differential driving in the embodiment. To avoid repetition, details are not described herein.

[0048] The electronic device 21 comprises, but is not limited to, the processor 211 and the memory 212. Those skilled in the art can understand that, Figure 2 The electronic device 21 is only an example and does not limit the electronic device 21, and can comprise more or fewer components than those shown in the figure, or combine certain components or different components, for example, the electronic device can further comprise an input / output device, a network access device, a bus, and the like.

[0049] The processor 211 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and the like. The general-purpose processor can be a microprocessor or any conventional processor.

[0050] The memory 212 can be an internal storage unit of the electronic device 21, for example, a hard disk or a memory of the electronic device 21. The memory 212 can also be an external storage device of the electronic device 21, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 21. Further, the memory 212 can also include both the internal storage unit and the external storage device of the electronic device 21. The memory 212 is used to store computer programs and other programs and data required by the network device. The memory 212 can also be used to temporarily store data that has been output or will be output.

[0051] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described herein.

[0052] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A frequency-modulated sequence prediction method based on frequency domain resonance enhancement and stochastic differential driving, characterized in that, The method includes: Step 1: Construct a frequency domain mapping and resonance enhancement filter; Step 2: Based on Step 1, perform frequency domain enhancement and inverse transform; Step 3: Based on Step 2, construct the stochastic differential equation SDE for dynamic smoothing; Step 4: Using the method from step 3, discretize the SDE to obtain the final optimized sequence; Step 5: Based on step 4, predict the frequency modulation sequence to obtain the predicted sequence.

2. The method according to claim 1, characterized in that, Step 1 includes: First, the original frequency modulation sequence The frequency domain signal expression after mapping to the frequency domain via Fast Fourier Transform (FFT) is as follows: ； in, ; For the complex components in the frequency domain i It is the imaginary unit.

3. The method according to claim 2, characterized in that, Step 1 further includes: Constructing a resonant enhancement filter Its bandwidth and peak position are related to the characteristic frequencies of the sequence. The relevant expression for the resonant enhancement filter is: ； in, For the first k The angular frequency corresponding to each frequency point ; Bandwidth parameters ; =2 and =0.5 are the gain and offset constants, respectively, used to control the magnitude of the enhancement.

4. The method according to claim 3, characterized in that, Step 2 includes: frequency domain signal With resonant enhancement filter Multiply to obtain the enhanced frequency domain signal. Then, it is mapped back to the time domain using the inverse Fourier transform (IFFT), and its expression is: ； ； in, This is the intermediate sequence after frequency domain resonance enhancement.

5. The method according to claim 4, characterized in that, Step 3 includes: The enhanced intermediate sequence Treating the external force acting on a virtual Brownian particle, we construct a variant of the Ornstein-Uhlenbeck (OU) process for it, and use its steady-state solution as the final output; we define the stochastic differential equation SDE, whose expression is: ； in, The optimized sequence; The long-term mean of the time-varying value is set as follows: That is, the enhanced sequence; The mean regression rate parameter, =0.5, used for control Towards The speed of regression; For volatility parameters, ; For the virtual Wiener process increment.

6. The method according to claim 5, characterized in that, Step 4 includes: The SDE is discretized and solved using the Euler-Maruyama method to obtain the final optimized sequence, which is expressed as follows: 。 7. The method according to claim 6, characterized in that, Step 5 includes: The original frequency modulation sequence Transform into ,right The prediction formula is as follows: ； in, =1, Let them be independent and identically distributed standard normal random variables, i.e. ~N(0,1); , avg This represents the mean; 。 8. The method according to claim 7, characterized in that, include: right The prediction formula is as follows: ； in, ; ； By analogy, the prediction can be made. ; Q To predict the length of the sequence, Q Less than , Indicates rounding down.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the frequency modulation sequence prediction method based on frequency domain resonance enhancement and stochastic differential driving as described in any one of claims 1 to 8.

10. An electronic device, characterized in that, include: One or more processors; Memory; And one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs including instructions that, when executed by the device, cause the device to perform the frequency modulation sequence prediction method based on frequency domain resonance enhancement and stochastic differential driving as described in any one of claims 1 to 8.

Images