Hybrid energy storage system power distribution method based on energy-entropy ratio
By using the energy entropy ratio method in the hybrid energy storage system to identify the noise signal and allocate the power of the energy storage unit, the problem of unreasonable power distribution of the hybrid energy storage system in the prior art when processing non-stationary signals is solved, and the energy utilization efficiency and noise recognition ability of the system are improved.
Patent Information
- Application Number
- CN202411332009.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-11-10
- Filing Date
- 2024-09-24
- Publication Date
- 2025-03-21
AI Technical Summary
When existing hybrid energy storage systems deal with non-stationary power scheduling signals, there is a problem of unreasonable power allocation, and traditional methods fail to effectively consider the noise influence during signal transmission.
The power distribution method of hybrid energy storage system based on energy entropy ratio is adopted, and the dispatch signal is decomposed into a sub-frequency domain through fast Fourier transform, the energy spectrum entropy of the sub-frequency domain is calculated, and the energy entropy ratio is calculated to identify the noise signal, and the power of the energy storage unit is reasonably allocated.
The energy storage system has improved its ability to identify noise signals, achieved reasonable distribution of power of different energy storage units in hybrid energy storage systems, and optimized the energy conversion efficiency and working performance of the system.
Smart Images
Figure CN119675129A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power distribution of hybrid energy storage systems, and in particular relates to a power distribution method of a hybrid energy storage system based on energy entropy ratio. Background Art
[0002] With the development of new energy technologies, the problem of power imbalance in the power grid caused by the randomness and intermittency of new energy generation has attracted much attention. Therefore, in order to maintain the stability of the power grid, it is necessary to add an energy storage system to smooth out power fluctuations. The hybrid energy storage system consists of energy storage units and power storage units. The power storage unit absorbs / outputs peak power, and the energy storage unit absorbs / outputs stable power. The combination of the two can effectively simplify the capacity attenuation problem of energy storage units represented by lithium-ion batteries, and effectively increase the comprehensive energy utilization and service life of the energy storage system.
[0003] However, during the operation of the system, the imbalance of power supply and demand between the hybrid energy storage system, the new energy generation system and the load will cause bus voltage fluctuations. At the same time, the unreasonable coordination and distribution of energy between energy storage components will also lead to unnecessary energy loss and reduce the energy utilization efficiency of the hybrid energy storage.
[0004] Regarding the coordinated control method of the two energy storage units in hybrid energy storage, the current mainstream method is to obtain the high-frequency separation and low-frequency components of the hybrid energy storage scheduling signal through FFT decomposition. The high-frequency power part is absorbed / output by the power-type energy storage unit, and the low-frequency power part is absorbed / output by the energy-type energy storage unit, thereby realizing the complementary advantages of the two energy storage units and improving the utilization efficiency of the hybrid energy storage units.
[0005] Problems with existing technologies: 1. The traditional hybrid energy storage unit decomposes the energy storage power dispatch signal into high and low frequency parts through filters or Fourier transform. The low frequency power part is borne by the energy type energy storage unit, and the high frequency power part is borne by the power type energy storage unit. However, this method is more suitable for analyzing steady signals. For non-steady power dispatch signals, there is the problem of unreasonable power allocation.
[0006] 2. The energy storage power dispatching signal is decomposed into high-frequency, medium-frequency and low-frequency signals by Fourier transform method, and different energy storage units bear the power of corresponding frequencies. However, the high-frequency, medium-frequency and low-frequency signals decomposed by this method are aliased, resulting in that energy-type energy storage units and power-type energy storage units cannot be reasonably used according to their respective characteristics; 3. Apply the modal decomposition algorithm to decompose the hybrid energy storage system dispatching signal into control signals of different modes. Consider the characteristics of the energy storage unit and distribute the control signal to different energy storage units to realize charge / discharge control.
[0007] The above method only targets the hybrid energy storage system when the dispatching signal is pure, but does not consider the problem of unreasonable control signal distribution caused by the noise in the signal transmission process.
[0008] In the existing patent CN115663853 A "Unbalanced power distribution method of hybrid energy storage based on real-time empirical mode decomposition", the unbalanced power is distributed by collecting the last data section data of different intrinsic mode functions and comparing them with the capacity of each energy storage unit in the hybrid energy storage. However, the above patent does not involve the impact of information such as noise in the collected signal on power distribution.
[0009] At the same time, the paper "Energy Storage Control Method Based on Battery State of Charge and Variable Filter Time Constant" (Zhang Ye et al., Automation of Electric Power Systems, 2012) only separates the high and low frequencies of the energy storage power distribution signal through the filtering time constant, and does not consider the power distribution signal noise and power distribution signal aliasing. Summary of the invention
[0010] In order to overcome the above technical problems, the purpose of the present invention is to provide a power allocation method for a hybrid energy storage system based on energy entropy ratio, in which an energy storage system is constructed by energy-type energy storage units and power-type energy storage units, and the power of the energy-type energy storage units and the power-type energy storage units in the hybrid energy storage system are reasonably allocated according to their respective working charge and discharge rates and SOC information.
[0011] In order to achieve the above object, the technical solution adopted by the present invention is: A power allocation method for a hybrid energy storage system based on energy entropy ratio comprises the following steps; Step 1: Calculate the spectral entropy of the energy storage system scheduling information containing noise signals to obtain the energy spectral entropy in the sub-frequency domain; Step 2: sub-frequency domain energy entropy ratio; When the power system dispatch signal is transmitted to different energy storage units, the sub-frequency energy and sub-frequency energy spectrum entropy are combined to obtain the sub-frequency energy entropy ratio G, which is as follows: (5).
[0012] The step 1 is specifically as follows: Assume that the charge / discharge scheduling signal of the hybrid energy storage system is x(t), and divide this signal into n time periods: t i The scheduling signal is x i (m), through the fast Fourier transform (FFT), the scheduling signal x i (m) Convert the scheduled time domain signal into a frequency domain signal: (1) In formula (1), X i (k) Yes t i The spectrum amplitude value of the kth frequency domain point in the time period, N is the number of sampling points, ω (n) is the window function (Hamming window, Blackman window, etc. can be selected as needed). The window function can effectively reduce signal energy leakage; Each time period t i The frequency domain signal is divided into h sub-frequency domains, each sub-frequency domain consists of j frequency domain points, and the energy of the energy storage unit scheduling signal in each sub-frequency domain is calculated. t i The energy of the dth sub-frequency domain signal in a time period is: (2) The energy probability of the corresponding sub-frequency domain signal is: (3) Then the energy spectrum entropy H(d) in the sub-frequency domain can be obtained as: (4).
[0013] In step 2, the scheduling signal is sliced and divided. The signal energy of each sub-frequency domain is the energy signal that the hybrid energy storage at this frequency domain point needs to be charged or released. The energy storage unit is given a charging and discharging energy signal, and the energy storage unit will charge and discharge according to this signal to release energy (high power). The sub-frequency domain energy of the hybrid energy storage system can reflect the amount of energy required by the hybrid energy storage system during this period of time.
[0014] In step 2, the sub-frequency energy spectrum entropy reflects the degree of confusion of the hybrid energy storage system scheduling signal during this period of time. The larger the sub-frequency domain energy is, the smaller its energy spectrum entropy value is. Therefore, the value of the sub-frequency domain energy in the noise interval is small, and the corresponding energy spectrum entropy value is large.
[0015] In step 2, the energy entropy ratio is the frequency domain energy and energy spectrum entropy of the hybrid energy storage system divided by each other. The energy entropy ratio can better identify the dispatching signal and noise signal of the hybrid energy storage system; at the same time, for each time period, the energy entropy ratio can also be used to determine whether the energy storage unit needs to be charged / discharged or the power storage unit needs to be charged / discharged. The specific method is as follows: In a cycle, the energy entropy ratio is less than a certain threshold αG max (α is the noise threshold coefficient, which can be revised according to the actual project situation. According to Gaussian distribution, α can be preferably 0.0456). If G(i) <αG max This section is considered to be a noise signal. At this time, the signal is considered to be a noise signal, and the hybrid energy storage system maintains the operating state of the previous moment; If the energy-entropy ratio within a cycle is greater than αG max , and less than βG max ,(for β The power division coefficient can be revised according to the actual project situation, and can be preferably 0.25). At this time, the hybrid energy storage unit is charged / discharged, and the power type energy storage unit is preferentially charged / discharged; If the energy-entropy ratio within a cycle is greater than βG max At this time, the hybrid energy storage unit is charged / discharged, and the energy type energy storage unit is selected to charge / discharge.
[0016] Beneficial effects of the present invention.
[0017] The present invention proposes a noise identification algorithm in the scheduling signal of a hybrid energy storage system based on the energy entropy ratio. In the frequency domain, the spectrum energy is combined with the spectrum entropy, which improves the noise identification ability in the energy storage system scheduling process. At the same time, the energy entropy ratio can also be used to reasonably allocate power to different energy storage units in the hybrid energy storage system.
[0018] The present invention proposes a noise identification algorithm in the dispatching signal of a hybrid energy storage system based on the energy entropy ratio. In the field of energy, the reliability and efficiency of the energy storage system are crucial to the stable operation of the power system. However, the hybrid energy storage system is often faced with interference from high-frequency, low-frequency dispatching signals and noise signals during actual operation, which will affect its reliability. In order to solve this problem, the present invention proposes a hybrid energy storage power allocation method considering a noise identification algorithm to improve the ability to accurately identify noise during the dispatching process of the energy storage system.
[0019] The present invention is mainly based on the energy-entropy ratio principle, combining spectrum energy with spectrum entropy in the frequency domain. Spectral energy reflects the energy distribution of the signal on each frequency component, while spectral entropy reflects the complexity and uncertainty of the signal spectrum. By combining these two indicators, we can more comprehensively analyze the noise characteristics in the energy storage system dispatch signal.
[0020] Specifically, the algorithm first uses Fourier transform to convert the signal into the frequency domain to obtain the relevant data of spectrum energy and spectrum entropy. Then, the energy entropy ratio is used to evaluate the relative strength of noise in the signal. Compared with traditional methods, the energy entropy ratio can better capture the difference between noise and signal, thereby improving the ability to identify noise.
[0021] In addition, the present invention also has the ability to distribute power to different energy storage units in the hybrid energy storage system. By analyzing the energy entropy ratio of different energy storage units, we can reasonably allocate power according to their characteristics and requirements to optimize the energy conversion efficiency and working performance of the system.
[0022] In summary, the dispatch signal control algorithm containing noise in the hybrid energy storage system based on the energy entropy ratio proposed in this invention improves the energy storage system's ability to identify noise by combining spectrum energy and spectrum entropy, and can also effectively realize the reasonable power allocation of different energy storage units in the hybrid energy storage system. The application of this technology will help improve the performance and reliability of the energy storage system and further promote the sustainable development of the power system. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 This is a structural block diagram of the hybrid energy storage system of the present invention.
[0024] Figure 2 Energy allocation flow chart of hybrid energy storage system based on energy-entropy ratio. DETAILED DESCRIPTION
[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0026] The present invention aims at the hybrid energy storage unit power distribution process. When the signal-to-noise ratio of the hybrid energy storage system dispatch signal is low, the method based on energy entropy ratio is proposed to achieve the reasonable distribution of energy storage unit power under the condition of low signal-to-noise ratio. The power system structure based on new energy power generation system and hybrid energy storage system.
[0027] like Figure 1 shown.
[0028] Spectral entropy calculation Assume that the charge / discharge scheduling signal of the hybrid energy storage system is x(t), and divide this signal into n time periods: t i The scheduling signal is x i (m), through the fast Fourier transform (FFT), the scheduling signal x i (m) The transformation from time domain to frequency domain is: (1) In formula (1), X i(k) Yes t i The spectrum amplitude value of the kth frequency domain point in the time period, N is the number of sampling points, ω (n) is the window function (Hamming window, Blackman window, etc. can be selected as needed). The window function can effectively reduce signal energy leakage.
[0029] Each time period t i The frequency domain signal is divided into h sub-frequency domains, each sub-frequency domain consists of j frequency domain points, and the energy of each sub-frequency domain is calculated. t i The energy of the dth sub-frequency domain in a time period is: (2) The corresponding energy probability in the sub-frequency domain is: (3) Then the energy spectrum entropy H(d) in the sub-frequency domain can be obtained as: (4) Sub-frequency domain energy entropy ratio When the power system dispatch signal is transmitted to different energy storage units, it will inevitably be mixed with noise signals due to reasons such as line transmission distance. In order to better remove the noise information in the dispatch signal and enhance the charging / discharging accuracy of each energy storage unit, the present invention proposes to combine the sub-frequency energy and the sub-frequency energy spectrum entropy to obtain the sub-frequency energy entropy ratio G, which is as follows: (5) Algorithm description: The sub-frequency domain energy of the hybrid energy storage system can reflect the amount of energy required by the hybrid energy storage system during this period of time; The energy spectrum entropy reflects the degree of confusion of the hybrid energy storage system dispatching signal during this period of time. The larger the sub-frequency domain energy, the smaller the energy spectrum entropy value. Therefore, the value of the sub-frequency domain energy in the noise interval is small, and the corresponding energy spectrum entropy value is large.
[0030] The energy entropy ratio is the frequency domain energy and energy spectrum entropy of the hybrid energy storage system divided by each other. The energy entropy ratio can better identify the dispatching signal and noise signal of the hybrid energy storage system; at the same time, for each time period, the energy entropy ratio can also be used to determine whether the energy storage unit needs to be charged / discharged or the power storage unit needs to be charged / discharged. The specific method is as follows: In a cycle, the energy entropy ratio is less than a certain threshold αG max (α is the noise threshold coefficient, which can be revised according to the actual project situation. According to Gaussian distribution, α can be preferably 0.0456). If G(i) < αGmax This section is considered to be a noise signal. At this time, the signal is considered to be a noise signal, and the hybrid energy storage system maintains the operating state of the previous moment; If the energy-entropy ratio within a cycle is greater than αG max , and less than βG max ,(for β The power division coefficient can be revised according to the actual engineering situation, and 0.25 can be preferably selected. At this time, the hybrid energy storage unit is charged / discharged, and the power type energy storage unit is preferentially charged / discharged.
[0031] If the energy-entropy ratio within a cycle is greater than βG max At this time, the hybrid energy storage unit is charged / discharged, and the energy type energy storage unit is selected to charge / discharge.
[0032] Algorithm flow of the present invention The algorithm flow proposed by the invention is as follows Figure 2 As shown, the specific steps of the algorithm are as follows: 1. First, according to the renewable energy power generation and load power consumption in the power system, the charging / discharging dispatching signal required by the hybrid energy storage system is obtained according to the power system demand; 2. Divide the dispatch signal into i time lengths t i Sub-scheduling signals, multiply each sub-scheduling signal by a window function (windowing, the specific window function can be selected according to the actual situation, such as Hanning window, Hamming window, and Blackman window, etc., because Hanning window has better low-pass characteristics, it can be preferred); 3. Perform fast Fourier transform (FFT) on the windowed signal according to formula (1) to obtain the representation of the sub-scheduling signal in the frequency domain; 4. Divide the scheduling signal obtained in step 4 into h sub-frequency domains, each sub-frequency domain consists of j frequency domain points; 5. Calculate the energy of each sub-frequency domain using formula (2); 6. Calculate the energy probability density distribution function of each sub-frequency domain by formula (3); 7. Obtain the energy spectrum entropy value of each sub-frequency domain through formula (4); 8. Divide the sub-frequency domain energy and spectral entropy values obtained in step 5 and step 7, and obtain the energy entropy ratio of the sub-frequency domain by formula (5): G ( i ); 9. Calculate the maximum value of the energy entropy ratio in the entire frequency domain G max ; 10. Energy entropy ratio for each time period G( i ) and G max To compare, G ( i )<αG max When , the energy dispatching signal of the hybrid energy storage system is a noise signal, and the hybrid energy storage system maintains the state of the previous moment; 11. When αG max < G ( i )<βG max At this time, the power type energy storage unit is preferentially selected for charging / discharging response.
[0033] 12. When βG max < G ( i ), at this time, the energy storage unit is preferentially selected for charge / discharge response.
[0034] (1) The invention performs frequency domain decomposition of the hybrid energy storage dispatching signal, considers the influence of noise in the dispatching signal, and realizes the identification of the noise signal by calculating the energy entropy ratio; (2) The hybrid energy storage unit power allocation algorithm proposed in this invention based on energy entropy ratio combines frequency domain energy signals and spectral entropy to improve the recognition of noise signals. (3) The present invention can simultaneously realize noise signal recognition and the allocation of charging / discharging power of energy-type energy storage units and power-type energy storage units through the energy entropy ratio.
Claims
1. A power allocation method for a hybrid energy storage system based on energy entropy ratio, characterized in that: The steps include: Step 1: Calculate the spectral entropy of the energy storage system scheduling information containing noise signals to obtain the energy spectral entropy in the sub-frequency domain; Step 2: sub-frequency domain energy entropy ratio; When the power system dispatch signal is transmitted to different energy storage units, the sub-frequency energy and sub-frequency energy spectrum entropy are combined to obtain the sub-frequency energy entropy ratio G, which is as follows: (5)。 2. A hybrid energy storage system power allocation method based on energy entropy ratio according to claim 1, characterized in that: The step 1 is specifically as follows: Assume that the charge / discharge scheduling signal of the hybrid energy storage system is x(t), and divide this signal into n time periods: t i The scheduling signal is x i (m), through the fast Fourier transform (FFT), the scheduling signal x i (m) Convert the scheduled time domain signal into a frequency domain signal: (1) In formula (1), X i (k) Yes t i The spectrum amplitude value of the kth frequency domain point in the time period, N is the number of sampling points, ω (n) is the window function, which can effectively reduce signal energy leakage; Each time period t i The frequency domain signal is divided into h sub-frequency domains, each sub-frequency domain consists of j frequency domain points, and the energy of the energy storage unit scheduling signal in each sub-frequency domain is calculated. t i The energy of the dth sub-frequency domain signal in a time period is: (2) The energy probability of the corresponding sub-frequency domain signal is: (3) Then the energy spectrum entropy H(d) in the sub-frequency domain can be obtained as: (4)。 3. The power allocation method of a hybrid energy storage system based on energy entropy ratio according to claim 1 is characterized in that: In step 2, the scheduling signal is sliced and divided. The signal energy of each sub-frequency domain is the energy signal that the hybrid energy storage at this frequency domain point needs to be charged or released. The energy storage unit is given a charging and discharging energy signal, and the energy storage unit will charge and discharge according to this signal to release energy (high power). The sub-frequency domain energy of the hybrid energy storage system can reflect the amount of energy required by the hybrid energy storage system during this period of time.
4. The power allocation method of a hybrid energy storage system based on energy entropy ratio according to claim 3 is characterized in that: In step 2, the sub-frequency energy spectrum entropy reflects the degree of confusion of the hybrid energy storage system scheduling signal during this period of time. The larger the sub-frequency domain energy is, the smaller its energy spectrum entropy value is. Therefore, the value of the sub-frequency domain energy in the noise interval is small, and the corresponding energy spectrum entropy value is large.
5. A hybrid energy storage system power allocation method based on energy entropy ratio according to claim 4, characterized in that: In step 2, the energy entropy ratio is the frequency domain energy and energy spectrum entropy of the hybrid energy storage system divided by each other. The energy entropy ratio can better identify the dispatching signal and noise signal of the hybrid energy storage system; at the same time, for each time period, the energy entropy ratio can also be used to determine whether the energy storage unit needs to be charged / discharged or the power storage unit needs to be charged / discharged. The specific method is as follows: In a cycle, the energy entropy ratio is less than a certain threshold αG max (α is the noise threshold coefficient, which can be revised according to the actual project situation. According to Gaussian distribution, α can be preferably 0.0456). If G(i) < αG max This section is considered to be a noise signal. At this time, the signal is considered to be a noise signal, and the hybrid energy storage system maintains the operating state of the previous moment; If the energy-entropy ratio within a cycle is greater than αG max , and less than βG max ,(for β The power division coefficient can be revised according to the actual project situation, and can be preferably 0.25). At this time, the hybrid energy storage unit is charged / discharged, and the power type energy storage unit is preferentially charged / discharged; If the energy-entropy ratio within a cycle is greater than βG max At this time, the hybrid energy storage unit is charged / discharged, and the energy type energy storage unit is selected to charge / discharge.