Energy storage control method based on improved variational mode decomposition and Hilbert transform

Abstract

The invention discloses an energy storage control method based on improved variational mode decomposition and Hilbert transform, and relates to the technical field of energy storage optimization control, and the energy storage control method comprises the steps: calculating the reference power of an energy storage system composed of an energy storage battery and a super capacitor based on the preprocessed new energy power generation measurement data; decomposing the energy storage system reference power into a plurality of energy storage system power sub-mode components by using an improved variational mode decomposition technology; generating an energy spectrum of a plurality of energy storage system power sub-modal components by using a Hilbert transform technology, and selecting a boundary frequency when the modal aliasing energy is minimum based on the energy spectrum; and reconstructing the plurality of energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency. Compared with empirical mode decomposition and local mean decomposition, the improved variational mode decomposition technology can restrain continuous propagation of envelope line errors, and the problem of mode aliasing of a traditional method is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage optimization control, and in particular to an energy storage control method based on improved variational mode decomposition and Hilbert transform. Background Art

[0002] At present, wind power and photovoltaic power generation technologies are relatively mature and have great development potential, and have been widely used at home and abroad. However, since the output power of photovoltaic and wind power is greatly affected by environmental factors, and is intermittent and uncertain, the output of power stations that use wind or solar energy as a single energy source fluctuates greatly, which is easy to cause impact on the power system. With the increase in the penetration rate of distributed power sources, it will seriously affect the safe, stable and reliable operation of the power grid.

[0003] Energy storage systems can reasonably solve a series of problems due to their bidirectional power regulation capabilities and rapid response characteristics. Through effective energy storage system optimization configuration and coordinated control methods, they can absorb or release power in a timely manner, smooth out the volatility of new energy injection power, and improve the utilization rate of intermittent renewable energy. Therefore, if the charging and discharging power of the energy storage system can be reasonably controlled to smooth out the fluctuations caused by its grid connection, through the coordinated control between the energy storage system and renewable power sources such as wind power and photovoltaics, it can not only reduce the impact on the power system and ensure the reliability of power supply, but also reduce the backup capacity of the power system, improve the operating economy of the power system and the ability to accept renewable energy. Summary of the invention

[0004] In view of the problems in the related art, the present invention proposes an energy storage control method based on improved variational mode decomposition and Hilbert transform to overcome the above technical problems existing in the existing related art.

[0005] To this end, the specific technical solution adopted by the present invention is as follows:

[0006] An energy storage control method based on improved variational mode decomposition and Hilbert transform, the energy storage control method comprising:

[0007] S1. Collect and pre-process the renewable energy power generation measurement data, and calculate the reference power of the energy storage system composed of energy storage batteries and supercapacitors based on the pre-processed renewable energy power generation measurement data;

[0008] S2. Using improved variational modal decomposition technology, the reference power of the energy storage system is decomposed into several energy storage system power sub-modal components;

[0009] S3. Generate energy spectra of several power sub-modal components of the energy storage system using Hilbert transform technology, and select the boundary frequency when the modal aliasing energy is minimum based on the energy spectrum;

[0010] S4. Reconstruct several energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency to achieve energy storage distribution between the energy storage battery and the supercapacitor power in the energy storage system.

[0011] Preferably, collecting and preprocessing the renewable energy power generation measurement data, and calculating the reference power of the energy storage system composed of energy storage batteries and supercapacitors based on the preprocessed renewable energy power generation measurement data includes:

[0012] S11, collecting renewable energy power generation measurement data, and using a sliding average filter algorithm to separate noise in the renewable energy power generation measurement data and renewable energy grid-connected reference power;

[0013] S12. The difference between the new energy grid-connected reference power and the predefined new energy synthetic output power is used as the reference power of the energy storage system composed of energy storage batteries and supercapacitors.

[0014] Preferably, the expression of the new energy grid-connected reference power is:

[0015] P G (t)=(P G (t-(s-1))+P G (t-(s-2))+...+P G (t)+...+P G (t+s)) / N;

[0016] in:

[0017] P G (t) = P PV (t)+P WG (t);

[0018] Where P G (t) represents the reference power of new energy grid connection; P PV (t) represents the photovoltaic power output power; P WG (t) represents wind power output power; s represents the number of renewable energy power generation measurement data; N represents the length of the sliding time window; t represents the time vector.

[0019] Preferably, the improved variational modal decomposition technique is used to decompose the energy storage system reference power into several energy storage system power sub-modal components including:

[0020] S21. Obtaining a number of energy storage system power sub-modes under preset constraint conditions, and forming constrained optimization variational conditions based on the energy storage system power sub-modes;

[0021] S22, introducing a quadratic penalty factor, a primary penalty factor and a Lagrange multiplication operator to convert the constrained optimization variational condition into an unconstrained form;

[0022] S23, using the alternating direction multiplier method to solve the saddle point of the augmented Lagrangian function in unconstrained form;

[0023] S24, transforming the unconstrained form into the frequency domain, iteratively updating the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency in the frequency domain to obtain power sub-modal components of the energy storage system at several different frequencies.

[0024] Preferably, the unconstrained form is transformed into the frequency domain, and the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency are iteratively updated in the frequency domain to obtain several power sub-modal components of the energy storage system at different frequencies, including:

[0025] S241, transforming the unconstrained form into the frequency domain using Fourier isometric transform technology, and converting the frequency domain form into a non-negative frequency interval integral form;

[0026] S242, generating an optimized updated solution based on a non-negative frequency interval integral form, and obtaining an iteratively updated and variationally modally decomposed reference power component of the energy storage system;

[0027] S243, converting the solution function of the center frequency corresponding to the power component of the energy storage system after variational modal decomposition into the frequency domain to obtain the iteratively updated center frequency;

[0028] S244, repeating the iterative process of step S241-step S243 until the iteration stop condition is met, and obtaining the energy storage system power sub-modal components at different frequencies based on the iteratively updated and variationally modally decomposed energy storage system reference power components and their corresponding center frequencies.

[0029] Preferably, the Hilbert transform technique is used to generate energy spectra of several power sub-modal components of the energy storage system, and the boundary frequencies at which the modal aliasing energy is minimized are selected based on the energy spectra, including:

[0030] S31, performing Hilbert transform on each power sub-modal component of the energy storage system to obtain the instantaneous amplitude and frequency of the power sub-modal of the energy storage system;

[0031] S32, calculating and summarizing the energy spectrum of the power sub-modal component of the energy storage system based on the instantaneous amplitude and frequency to obtain the target signal energy time-frequency spectrum;

[0032] S33, integrating the time-frequency spectrum of the target signal energy to obtain the time-frequency spectrum energy of the power sub-modal component of the energy storage system, and formulating an instantaneous frequency-time curve based on the time-frequency spectrum energy;

[0033] S34, analyzing the instantaneous frequency-time curve and extracting the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system.

[0034] Preferably, analyzing the instantaneous frequency-time curve and extracting the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system includes:

[0035] S341, extracting two adjacent curves from the instantaneous frequency-time curve, and defining a temporary frequency division frequency between the adjacent curves to distinguish the power distribution of the energy storage battery and the supercapacitor;

[0036] S342, comparing the frequency of each energy storage system power sub-modal component with the temporary frequency division frequency, obtaining the time when the frequency of each energy storage system power sub-modal component is higher or lower than the temporary frequency division frequency and the energy storage system sub-modal power;

[0037] S343. Calculate the sub-modal aliasing energy of the energy storage system based on time and the sub-modal power of the energy storage system, and select the boundary frequency corresponding to the minimum sub-modal aliasing of the energy storage system.

[0038] Preferably, the expression of sub-modal aliasing energy of the energy storage system is:

[0039]

[0040] In the formula, E overlap represents the sub-modal aliasing energy of the energy storage system; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is higher than the temporary division frequency; k represents the number of center frequencies.

[0041] Preferably, reconstructing a plurality of energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency to achieve energy storage distribution between energy storage batteries and supercapacitor powers in the energy storage system includes:

[0042] S41, reconstructing the power sub-modal component of the energy storage system into a low-frequency component and a high-frequency component based on the demarcation frequency corresponding to the minimum sub-modal aliasing of the energy storage system;

[0043] S42, dividing the energy storage system reference power into a high-frequency part and a low-frequency part based on the reconstructed low-frequency component and high-frequency component, and using them as reference powers of the supercapacitor and the energy storage battery respectively;

[0044] S43. Calculate the actual power of the supercapacitor and the energy storage battery in each time period in combination with the constraints during the actual operation of the energy storage system, and realize the power distribution between the energy storage battery and the supercapacitor in the energy storage system based on the actual power.

[0045] Preferably, the expression of the reference power of the energy storage battery is:

[0046]

[0047] The expression of the reference power of the supercapacitor is:

[0048]

[0049] Where P B0 (t) represents the reference power of the energy storage battery; P C0 (t) represents the reference power of the supercapacitor; y s+i Indicates the measured s+i-th renewable energy power generation data; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the submodal power of the energy storage system when the frequency of the submodal component of the power of the energy storage system is higher than the temporary frequency division frequency; N represents the length of the sliding time window; p and q both represent positive integers within a preset range; ω i Represents the weight coefficient.

[0050] The beneficial effects of the present invention are:

[0051] 1. The present invention adopts variational mode decomposition technology to decompose the reference power of the energy storage system. By improving the variational mode decomposition technology compared with empirical mode decomposition and local mean decomposition, it can suppress the continuous propagation of envelope errors, thereby effectively solving the problem of modal aliasing in traditional methods, which is more advantageous than traditional mode decomposition technology.

[0052] 2. The present invention uses Hilbert transform to perform spectral analysis on the modal components, finds the instantaneous frequency-time function of each modal component, and distinguishes the original signal according to different degrees of fluctuation, thereby overcoming the limitation of traditional Fourier transform that the instantaneous frequency value of the signal must have at least one fully oscillating sine or cosine wave, and further solves the problem of spectral analysis of non-stationary waveforms.

[0053] 3. The present invention is based on the modal functions from low to high frequencies obtained by improving the variational modal decomposition. By establishing a modal aliasing energy minimum model, a reasonable demarcation frequency is selected, and the modal components are reconstructed into low-frequency components and high-frequency components. The energy storage output is reasonably controlled based on the high and low-frequency components, thereby improving the energy storage output performance while meeting the requirements of the energy storage system to smooth wind and light power fluctuations, thereby increasing the service life of the energy storage and reducing the cost of energy storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 is a flow chart of an energy storage control method based on improved variational mode decomposition and Hilbert transform according to an embodiment of the present invention;

[0056] Figure 2 is a flow chart of an improved variational mode decomposition algorithm in an energy storage control method based on improved variational mode decomposition and Hilbert transform according to an embodiment of the present invention;

[0057] Figure 3 It is a control flow chart of an energy storage system in an energy storage control method based on improved variational mode decomposition and Hilbert transform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0059] According to an embodiment of the present invention, a method for energy storage control based on improved variational mode decomposition and Hilbert transform is provided.

[0060] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to an energy storage control method based on improved variational mode decomposition and Hilbert transform according to an embodiment of the present invention, the energy storage control method includes:

[0061] S1. Collect and pre-process the renewable energy power generation measurement data (wind and solar power generation measurement data). Based on the pre-processed renewable energy power generation measurement data, calculate the reference power of the energy storage system composed of energy storage batteries and supercapacitors.

[0062] Among them, the new energy power generation measurement data is collected and preprocessed, and based on the preprocessed new energy power generation measurement data, the reference power of the energy storage system composed of energy storage batteries and supercapacitors is calculated, including:

[0063] S11, collecting renewable energy power generation measurement data, and using a sliding average filter algorithm to separate noise in the renewable energy power generation measurement data and renewable energy grid-connected reference power;

[0064] S12. The difference between the new energy grid-connected reference power and the predefined new energy synthetic output power is used as the reference power of the energy storage system composed of energy storage batteries and supercapacitors.

[0065] It should be noted that the noise filtering of renewable energy power generation measurement data is performed based on the sliding average filtering algorithm. The sliding average is to take m adjacent data one by one along the full length of N data as the direct arithmetic average, that is, the m adjacent data y s-n ,y s-n+1 , ..., y s , ..., y s+n The smoothed data f s =y s is equivalent to, and its expression is:

[0066]

[0067] In the formula, y s+i is the measured s+ith renewable energy power generation data, ω i is the weight coefficient, and p and q are any positive integers less than m, m is an arbitrarily selected number of adjacent measurement data, p is the p-th power generation measurement data in N, q is the q-th power generation measurement data in N, and p+q+1=m. Different methods of taking these parameters form different sliding average methods.

[0068] The present invention selects a sliding time window of length N to perform sliding average filtering on the wind and light combined output power. The expression of the wind and light combined output power at time t is:

[0069] P G (t)=(P G (t-(s-1))+P G (t-(s-2))+...+P G (t)+...+P G (t+s)) / N;

[0070] in:

[0071]

[0072] PG (t) = P PV (t)+P WG (t);

[0073] The grid-connected reference power P of the wind-solar-storage combined power generation system out (t) Combined output power with wind and light (reference power for new energy grid connection) P G The difference between (t) and (t) is used as the reference power of the energy storage system, and its expression is:

[0074] P HESS (t) = P out (t)-P G (t);

[0075] Where P HESS (t) represents the reference power of the energy storage system; P G (t) represents the reference power of new energy grid connection; P PV (t) represents the photovoltaic power output power; P WG (t) represents wind power output power; k represents the number of measurement data of renewable energy power generation; N represents the length of the sliding time window; t represents the time vector.

[0076] S2. Using improved variational modal decomposition technology, the energy storage system reference power is decomposed into several energy storage system power sub-modal components.

[0077] Among them, the improved variational modal decomposition technology is used to decompose the energy storage system reference power into several energy storage system power sub-modal components including:

[0078] S21. Obtaining a number of energy storage system power sub-modes under preset constraint conditions, and forming constrained optimization variational conditions based on the energy storage system power sub-modes;

[0079] S22, introducing a quadratic penalty factor, a primary penalty factor and a Lagrange multiplication operator to convert the constrained optimization variational condition into an unconstrained form;

[0080] S23, using the alternating direction multiplier method to solve the saddle point of the augmented Lagrangian function in unconstrained form;

[0081] S24, transforming the unconstrained form into the frequency domain, iteratively updating the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency in the frequency domain to obtain power sub-modal components of the energy storage system at several different frequencies.

[0082] Among them, the unconstrained form is transformed into the frequency domain, and the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency are iteratively updated in the frequency domain to obtain several power sub-modal components of the energy storage system at different frequencies, including:

[0083] S241, transforming the unconstrained form into the frequency domain using Fourier isometric transform technology, and converting the frequency domain form into a non-negative frequency interval integral form;

[0084] S242, generating an optimized updated solution based on a non-negative frequency interval integral form, and obtaining an iteratively updated and variationally modally decomposed reference power component of the energy storage system;

[0085] S243, converting the solution function of the center frequency corresponding to the power component of the energy storage system after variational modal decomposition into the frequency domain to obtain the iteratively updated center frequency;

[0086] S244, repeating the iterative process of step S241-step S243 until the iteration stop condition is met, and obtaining the energy storage system power sub-modal components at different frequencies based on the iteratively updated and variationally modally decomposed energy storage system reference power components and their corresponding center frequencies.

[0087] It should be noted that the reference power of the energy storage system is decomposed into k quasi-orthogonal modal components (power sub-modal components of the energy storage system) with different center frequencies based on the improved variational modal decomposition technology, including:

[0088] Under the constraint that the sum of the power sub-modal functions of the energy storage system is equal to the original power signal, k power sub-modes of the energy storage system with finite bandwidth and center frequency are sought so that the sum of the power sub-modal bandwidths of the energy storage system is minimized. The constrained optimization variational problem can be expressed as:

[0089]

[0090] Where x(t) represents the reference power signal of the energy storage system, u k (t) and ω k represents the reference power component of the energy storage system obtained by variational mode decomposition and its corresponding center frequency, δ(t) represents the Dirac distribution function, (δ(t)+j / πt)*u k (t) represents the Hilbert transform of each mode u k (t) The convolution operation obtains the analytical signal, the exponential term e -jωt It means modulating the analytical signal to the baseband, t and j represent time vectors, and k represents the number of center frequencies.

[0091] The expression of the constrained optimization variational problem is converted into an unconstrained form by introducing a quadratic penalty factor α, a linear penalty factor β, and a Lagrange multiplication operator λ:

[0092]

[0093] The alternating direction multiplier method is used to solve the saddle point of the augmented Lagrangian function, and then iteratively updates in the frequency domain and λ n+1 , that is u k ,ω k and λ, which specifically include:

[0094] The Fourier isometry transform (combined with Parseval's theorem or Plancher's theorem) is used to transform the converted unconstrained form into the frequency domain, and its expression is:

[0095]

[0096] Transform the converted unconstrained form to the frequency domain expression and replace the ω in the first and second terms with ω-ω k , whose expression is:

[0097]

[0098] Replace ω with ω-ω k The expression is converted into the form of non-negative frequency interval integral:

[0099]

[0100] The updated solution to the quadratic optimization problem is generated as:

[0101]

[0102] Similarly, The value problem is converted to the frequency domain:

[0103]

[0104] generate The updated solution is:

[0105]

[0106] Update λ:

[0107]

[0108] In the formula, It means the value obtained in the n+1th iteration τ represents the time step of the dual ascent method; the discrimination accuracy e is given in advance, and the iteration stops when the preset conditions are met. The k energy storage power submodal functions u can be obtained through the inverse Fourier transform k (t), where the precondition is:

[0109]

[0110] It should be noted that the specific implementation process of improved variational mode decomposition is as follows: Figure 2As shown, according to the above method, the reference power of the energy storage system is improved by variational mode decomposition, and the k center frequencies are ω k The inherent energy storage system power submodal component u k (t), its expression is:

[0111]

[0112] In the above formula, u k represents the kth power component of the energy storage system obtained through variational mode decomposition, represents the Fourier transformed u k , It means the value obtained in the n+1th iteration ω k represents the center frequency corresponding to the kth power component of the energy storage system obtained through variational mode decomposition, represents the original input signal after Fourier transform, represents the Lagrange multiplication operator after Fourier transformation, sgn represents the sign function, Represents the i-th power component u of the energy storage system after Fourier transformation k , represents the ω obtained in the n+1th iteration k , α represents the quadratic penalty factor, β represents the primary penalty factor, and λ represents the Lagrange multiplication operator.

[0113] S3. Use Hilbert transform technology to generate energy spectra of several power sub-modal components of the energy storage system, and select the boundary frequency when the modal aliasing energy is minimum based on the energy spectrum.

[0114] Among them, the Hilbert transform technology is used to generate the energy spectra of several power sub-modal components of the energy storage system, and the boundary frequencies with the minimum modal aliasing energy are selected based on the energy spectrum, including:

[0115] S31, performing Hilbert transform on each power sub-modal component of the energy storage system to obtain the instantaneous amplitude and frequency of the power sub-modal of the energy storage system;

[0116] S32, calculating and summarizing the energy spectrum of the power sub-modal component of the energy storage system based on the instantaneous amplitude and frequency to obtain the target signal energy time-frequency spectrum;

[0117] S33, integrating the time-frequency spectrum of the target signal energy to obtain the time-frequency spectrum energy of the power sub-modal component of the energy storage system, and formulating an instantaneous frequency-time curve based on the time-frequency spectrum energy;

[0118] S34, analyzing the instantaneous frequency-time curve and extracting the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system.

[0119] Among them, the analysis of the instantaneous frequency-time curve and the extraction of the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system include:

[0120] S341, extracting two adjacent curves from the instantaneous frequency-time curve, and defining a temporary frequency division frequency between the adjacent curves to distinguish the power distribution of the energy storage battery and the supercapacitor;

[0121] S342, comparing the frequency of each energy storage system power sub-modal component with the temporary frequency division frequency, obtaining the time when the frequency of each energy storage system power sub-modal component is higher or lower than the temporary frequency division frequency and the energy storage system sub-modal power;

[0122] S343. Calculate the sub-modal aliasing energy of the energy storage system based on time and the sub-modal power of the energy storage system, and select the boundary frequency corresponding to the minimum sub-modal aliasing of the energy storage system.

[0123] It should be noted that the instantaneous amplitude, frequency and phase of k energy storage power sub-modal components are obtained based on Hilbert transform, the Hilbert spectrum of each energy storage power sub-modal component is obtained, and the boundary frequency is determined based on the minimum aliasing energy of the energy storage power sub-modal:

[0124] For each energy storage power submodal component u k (t) Perform Hilbert transform to obtain the information of the instantaneous amplitude and frequency phase of the energy storage power sub-mode, and obtain the Hilbert spectrum of each energy storage power sub-mode component according to the obtained instantaneous spectrum (the Hilbert spectrum here is the energy spectrum in the above content), and its expression is:

[0125]

[0126] Construct the analytical signal of each energy storage power sub-modal component, and its expression is:

[0127]

[0128] The expression of the instantaneous frequency of each energy storage power sub-modal component is:

[0129]

[0130] Energy storage power sub-mode instantaneous amplitude a k The time-frequency distribution of (t) is defined as the energy storage power submodal component u k The Hilbert time spectrum of (t) is expressed as:

[0131] H k (ω,t)=H k (ω k (t),t)=a k(t);

[0132] The Hilbert time-frequency spectrum of all energy storage power sub-modal components is summarized to obtain the Hilbert time-frequency spectrum of the target signal, which is expressed as:

[0133]

[0134] The Hilbert time spectrum energy is obtained by integrating the square of the Hilbert time spectrum H(ω, t) over time and frequency. Its expression is:

[0135]

[0136] In the formula, a k (t) represents the energy storage power sub-modal component u k (t) is the instantaneous amplitude of the energy storage power submode, Represents the energy storage power submodal component u k (t), H(ω, t) represents the Hilbert time-frequency spectrum, [0, T] represents the time range of the defined data, [ω1, ω2] represents the frequency band of the Hilbert time-frequency spectrum, that is, the frequency range of the energy storage system that needs to be integrated, H k (ω,t) represents the energy storage power submodal component u k (t) is the Hilbert time-frequency spectrum, E represents the Hilbert time-frequency spectrum energy, and τ represents the time step of the dual rise method; z k (t) represents the analytical signal of the kth modal component of the energy storage power. This analytical signal contains two important parts, amplitude and phase.

[0137] It should be noted that the Hilbert transform is performed on each natural modal component of the energy storage system power to obtain several corresponding instantaneous frequency-time curves f k (t), and the curve f in the improved variational mode decomposition i+1 The frequency of (t) is generally higher than that of curve f i (t). The energy storage battery is responsible for smoothing low-frequency fluctuations, and the supercapacitor is responsible for smoothing high-frequency fluctuations. i (t) and f i+1 (t) Take the crossover frequency f m , then the curve f i+1 (t) The frequency is lower than f m The power corresponding to part of the curve is allocated to the battery (equivalent to the energy storage battery), which increases the number of charge and discharge times of the battery and reduces its service life; curve f i (t) The frequency is higher than f mThe power corresponding to the part is allocated to the supercapacitor, increasing the supercapacitor capacity. Since the supercapacitor energy storage density is not high, this part of energy cannot be too high, otherwise it cannot meet the requirements. Therefore, find a suitable crossover frequency f m , so that with f m The adjacent curve f i (t) and f i+1 (t) Minimum modal aliasing is crucial for developing appropriate power control strategies for energy storage systems.

[0138] When the frequency of each modal component of the energy storage system power is lower than the given frequency f m The corresponding time is t l (l=1,…,j), the corresponding submodal power of the energy storage system is u l (t l ); When the frequency of each modal component of the energy storage system power is higher than the given frequency f m The corresponding time is t h (h=1,…,k), the corresponding submodal power of the energy storage system is u h (t h ). The product of the absolute value of the power corresponding to the aliasing part of the energy storage system power sub-modal and the corresponding time is defined as the energy storage system power sub-modal aliasing energy, and its expression is:

[0139]

[0140] In the formula, E overlap represents the sub-modal aliasing energy of the energy storage system; f m represents the temporary decomposition frequency; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the energy storage system power is higher than the temporary frequency division frequency; when the frequency of each modal component of the energy storage system power is lower than the given frequency f m The corresponding time is t l (l=1,…,j), so it can be considered that at t j When the frequency of each modal component is exactly equal to f m .

[0141] It should be noted that the frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system is the boundary frequency f between high frequency and low frequency.m . Note f i+1 The maximum value of (t) is f max , f i The minimum value of (t) is f min From f min Start taking values, starting with f max As the end point, each time 0.0001Hz is increased as the temporary frequency division frequency, and the corresponding sub-modal aliasing energy of the energy storage system is calculated. The frequency corresponding to the minimum modal aliasing energy is the dividing frequency f between high frequency and low frequency. m .

[0142] S4. Reconstruct several energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency to achieve energy storage distribution between the energy storage battery and the supercapacitor power in the energy storage system.

[0143] Among them, according to the boundary frequency, several energy storage system power sub-modal components are reconstructed into low-frequency components and high-frequency components to realize the energy storage distribution between the energy storage battery and the supercapacitor power in the energy storage system, including:

[0144] S41, reconstructing the power sub-modal component of the energy storage system into a low-frequency component and a high-frequency component based on the demarcation frequency corresponding to the minimum sub-modal aliasing of the energy storage system;

[0145] S42, dividing the energy storage system reference power into a high-frequency part and a low-frequency part based on the reconstructed low-frequency component and high-frequency component, and using them as reference powers of the supercapacitor and the energy storage battery respectively;

[0146] S43. Calculate the actual power of the supercapacitor and the energy storage battery in each time period in combination with the constraints during the actual operation of the energy storage system, and realize the power distribution between the energy storage battery and the supercapacitor in the energy storage system based on the actual power.

[0147] It should be noted that the k sub-modal components of the energy storage system are reconstructed into low-frequency components and high-frequency components according to the boundary frequency to achieve a reasonable distribution of the power of the energy storage battery and supercapacitor inside the energy storage system, including:

[0148] For the wind and solar combined output power that does not meet the grid connection requirements, the sliding average method is used to obtain the grid connection power, and the difference between it and the actual wind and solar output power is the reference power of the hybrid energy storage system. In order to make full use of the complementary characteristics of energy storage and power storage, the sub-modal functions u1(t)~u2(t) of the energy storage system from low to high frequencies are obtained based on the improved variational mode decomposition. k (t), select a reasonable demarcation frequency f mThe above modal components are reconstructed into low-frequency components and high-frequency components, and the reference power of the hybrid energy storage system is divided into high-frequency and low-frequency parts as the reference power of the supercapacitor and the battery respectively. Finally, considering the constraints such as the state of charge during the actual operation of the energy storage, the actual power of the supercapacitor and the battery at each moment is obtained. The power allocation method is as follows: Figure 3 As shown. Among them, the frequency is less than or equal to f m The modal component of is smoothed by the battery as a low-frequency component, which is greater than f m The modal components are smoothed as high-frequency components by supercapacitors.

[0149] The charging and discharging power (reference power) of energy storage batteries and supercapacitors can be expressed as:

[0150]

[0151] Where P B0 (t) represents the charging and discharging power of the energy storage battery; P C0 (t) represents the charging and discharging power of the supercapacitor; y s+i Indicates the measured s+i-th renewable energy power generation data; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is higher than the temporary frequency division frequency; N represents the length of the sliding time window; p and q both represent positive integers within a preset range, p is the p-th power generation measurement data in N, and q is the q-th power generation measurement data in N; ω i Represents the weight coefficient.

[0152] The reference output power of the battery and supercapacitor is calculated through sliding average filtering and improved variational mode decomposition. However, in practice, constraints such as the state of charge of the energy storage system should also be considered.

[0153] The rated power of the battery is P B，e ; The rated power of the supercapacitor is P C，e ; E B，max , E C，max , E B，min , E C，min are the upper and lower limits of the remaining power of the battery and supercapacitor respectively; P B (t) and P C(t) are the actual charging and discharging power of the battery and supercapacitor at time t, respectively.

[0154] The output power of the energy storage system should meet the following constraints:

[0155] -P B,e ≤P B (t)≤P B,e ;

[0156] -P C,e ≤P C (t)≤P C,e ;

[0157] The remaining power of the energy storage device should meet the following constraints:

[0158] E B,min ≤E B (t)≤E B,max ;

[0159] E C,min ≤E C (t)≤E C,max ;

[0160] in:

[0161]

[0162] In the formula, SOC B，max , SOC C，max , SOC B，min , SOC C，min E are the upper and lower limits of the state of charge allowed for batteries and supercapacitors respectively; B , E C are the rated capacities of the battery and supercapacitor respectively.

[0163] The relationship between the state of charge of batteries and supercapacitors and the charge and discharge power is expressed as:

[0164]

[0165] Where Δt is the sampling time interval; SOC B (t), SOC C (t), SOC B (t+Δt), SOC C (t+Δt) are the charge states of the battery and supercapacitor at time t and time t+Δt respectively; E B 、E C are the capacities of the battery and supercapacitor respectively; η Bc , η Cc , η Bd , η Cdare the charging and discharging efficiencies of batteries and supercapacitors, respectively.

[0166] The actual charging and discharging power of the energy storage system is calculated as follows:

[0167]

[0168] P B0 ≥0;

[0169]

[0170] P B0 <0;

[0171]

[0172] P C0 ≥0;

[0173]

[0174] P C0 <0;

[0175] At this time, the actual output power of the wind-solar-storage combined power generation system at time t is:

[0176] P out,r (t) = P G (t)+P B (t)+P C (t);

[0177] Where P out,r (t) represents the actual output power of the wind-solar-storage combined power generation system at time t; P B (t) and P C (t) are the actual charging and discharging power of the battery and supercapacitor at time t; P G (t) represents the reference power of new energy grid-connected.

[0178] In summary, with the help of the above technical solution of the present invention, the present invention uses variational modal decomposition technology to decompose the reference power of the energy storage system. By improving the variational modal decomposition technology, compared with empirical mode decomposition and local mean decomposition, it can suppress the continuous propagation of envelope errors, thereby effectively solving the problem of modal aliasing in traditional methods, and has more advantages than traditional modal decomposition technology. The present invention uses Hilbert transform to perform spectral analysis on the modal components, find the instantaneous frequency-time function of each modal component, and distinguish the original signal according to different fluctuation degrees, thereby overcoming the limitation of the traditional Fourier transform that the instantaneous frequency value of the signal must have at least one fully oscillating sine or cosine wave, thereby solving the problem of spectrum analysis of non-stationary waveforms. The present invention is based on the modal functions from low to high frequencies obtained by improving variational modal decomposition. By establishing a modal aliasing energy minimum model, a reasonable demarcation frequency is selected, and the modal components are reconstructed into low-frequency components and high-frequency components. The energy storage output is reasonably controlled based on the high and low-frequency components, thereby improving the energy storage output performance while meeting the requirements of the energy storage system to smooth wind and light power fluctuations, thereby increasing the service life of the energy storage and reducing the cost of energy storage.

[0179] 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 spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An energy storage control method based on improved variational mode decomposition and Hilbert transform, characterized in that: The energy storage control method comprises: S1. Collect and pre-process the renewable energy power generation measurement data, and calculate the reference power of the energy storage system composed of energy storage batteries and supercapacitors based on the pre-processed renewable energy power generation measurement data; S2. Using improved variational modal decomposition technology, the reference power of the energy storage system is decomposed into several energy storage system power sub-modal components; S3. Generate energy spectra of several power sub-modal components of the energy storage system using Hilbert transform technology, and select the boundary frequency when the modal aliasing energy is minimum based on the energy spectrum; S4. Reconstruct several energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency to achieve energy storage distribution between the energy storage battery and the supercapacitor power in the energy storage system.

2. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 1, characterized in that: The collecting and preprocessing of the renewable energy power generation measurement data, and the calculation of the reference power of the energy storage system composed of the energy storage battery and the supercapacitor based on the preprocessed renewable energy power generation measurement data, comprises: S11, collecting renewable energy power generation measurement data, and using a sliding average filter algorithm to separate noise in the renewable energy power generation measurement data and renewable energy grid-connected reference power; S12. The difference between the new energy grid-connected reference power and the predefined new energy synthetic output power is used as the reference power of the energy storage system composed of energy storage batteries and supercapacitors.

3. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 2, characterized in that: The expression of the new energy grid-connected reference power is: P G (t)=(P G (t-(s-1))+P G (t-(s-2))+...+P G (t)+...+P G (t+s)) / N; in: P G (t)=P PV (t)+P WG (t); Where P G (t) represents the reference power of new energy grid connection; P PV (t) represents the photovoltaic power output power; P WG (t) represents wind power output power; s represents the number of renewable energy power generation measurement data; N represents the length of the sliding time window; t represents the time vector.

4. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 3 is characterized in that: The improved variational modal decomposition technique is used to decompose the energy storage system reference power into several energy storage system power sub-modal components, including: S21. Obtaining a number of energy storage system power sub-modes under preset constraint conditions, and forming constrained optimization variational conditions based on the energy storage system power sub-modes; S22, introducing a quadratic penalty factor, a primary penalty factor and a Lagrange multiplication operator to convert the constrained optimization variational condition into an unconstrained form; S23, using the alternating direction multiplier method to solve the saddle point of the augmented Lagrangian function in unconstrained form; S24, transforming the unconstrained form into the frequency domain, iteratively updating the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency in the frequency domain to obtain power sub-modal components of the energy storage system at several different frequencies.

5. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 4, characterized in that: The unconstrained form is transformed into the frequency domain, and the reference power component of the energy storage system after variational modal decomposition and its corresponding center frequency are iteratively updated in the frequency domain to obtain several power sub-modal components of the energy storage system at different frequencies, including: S241, transforming the unconstrained form into the frequency domain using Fourier isometric transform technology, and converting the frequency domain form into a non-negative frequency interval integral form; S242, generating an optimized updated solution based on a non-negative frequency interval integral form, and obtaining an iteratively updated and variationally modally decomposed reference power component of the energy storage system; S243, converting the solution function of the center frequency corresponding to the power component of the energy storage system after variational modal decomposition into the frequency domain to obtain the iteratively updated center frequency; S244, repeating the iterative process of step S241-step S243 until the iteration stop condition is met, and obtaining the energy storage system power sub-modal components at different frequencies based on the iteratively updated and variationally modally decomposed energy storage system reference power components and their corresponding center frequencies.

6. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 5, characterized in that: The Hilbert transform technique is used to generate energy spectra of several power sub-modal components of the energy storage system, and the boundary frequency when the modal aliasing energy is the smallest is selected based on the energy spectrum, including: S31, performing Hilbert transform on each power sub-modal component of the energy storage system to obtain the instantaneous amplitude and frequency of the power sub-modal of the energy storage system; S32, calculating and summarizing the energy spectrum of the power sub-modal component of the energy storage system based on the instantaneous amplitude and frequency to obtain the target signal energy time-frequency spectrum; S33, integrating the time-frequency spectrum of the target signal energy to obtain the time-frequency spectrum energy of the power sub-modal component of the energy storage system, and formulating an instantaneous frequency-time curve based on the time-frequency spectrum energy; S34, analyzing the instantaneous frequency-time curve and extracting the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system.

7. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 6, characterized in that: The analysis of the instantaneous frequency-time curve and the extraction of the boundary frequency corresponding to the minimum sub-modal aliasing energy of the energy storage system include: S341, extracting two adjacent curves from the instantaneous frequency-time curve, and defining a temporary frequency division frequency between the adjacent curves to distinguish the power distribution of the energy storage battery and the supercapacitor; S342, comparing the frequency of each energy storage system power sub-modal component with the temporary frequency division frequency, obtaining the time when the frequency of each energy storage system power sub-modal component is higher or lower than the temporary frequency division frequency and the energy storage system sub-modal power; S343. Calculate the sub-modal aliasing energy of the energy storage system based on time and the sub-modal power of the energy storage system, and select the boundary frequency corresponding to the minimum sub-modal aliasing of the energy storage system.

8. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 7, characterized in that: The expression of submodal aliasing energy of the energy storage system is: In the formula, E overlap represents the sub-modal aliasing energy of the energy storage system; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is higher than the temporary division frequency; k represents the number of center frequencies.

9. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 8, characterized in that: The reconstructing of a plurality of energy storage system power sub-modal components into low-frequency components and high-frequency components according to the boundary frequency to achieve energy storage distribution between energy storage batteries and supercapacitor powers in the energy storage system includes: S41, reconstructing the power sub-modal component of the energy storage system into a low-frequency component and a high-frequency component based on the demarcation frequency corresponding to the minimum sub-modal aliasing of the energy storage system; S42, dividing the energy storage system reference power into a high-frequency part and a low-frequency part based on the reconstructed low-frequency component and high-frequency component, and using them as reference powers of the supercapacitor and the energy storage battery respectively; S43. Calculate the actual power of the supercapacitor and the energy storage battery in each time period in combination with the constraints during the actual operation of the energy storage system, and realize the power distribution between the energy storage battery and the supercapacitor in the energy storage system based on the actual power.

10. The energy storage control method based on improved variational mode decomposition and Hilbert transform according to claim 9, characterized in that: The expression of the reference power of the energy storage battery is: The expression of the reference power of the supercapacitor is: Where P B0 (t) represents the reference power of the energy storage battery; P C0 (t) represents the reference power of the supercapacitor; y s+i Indicates the measured s+i-th renewable energy power generation data; t l Indicates the time when the frequency of the power sub-modal component of the energy storage system is lower than the temporary division frequency; u l (t l ) represents the sub-modal power of the energy storage system when the frequency of the sub-modal component of the power of the energy storage system is lower than the temporary frequency division frequency; t h Indicates the time when the frequency of the power sub-modal component of the energy storage system is higher than the temporary division frequency; u h (t h ) represents the submodal power of the energy storage system when the frequency of the submodal component of the power of the energy storage system is higher than the temporary frequency division frequency; N represents the length of the sliding time window; p and q both represent positive integers within a preset range; ω i Represents the weight coefficient.

Images