Virtual power plant day-ahead optimization scheduling method fusing multiple types of electric hydrogen production devices

Abstract

The invention relates to the technical field of energy storage and power dispatching, and discloses a virtual power plant day-ahead random optimization dispatching method fusing multiple types of electric hydrogen production devices. Comprising the following steps: S1, obtaining fan and photovoltaic output data in a virtual power plant; s2, performing multi-band decomposition on the output data of the fan and the photovoltaic power obtained in the step S1 by adopting empirical mode decomposition to obtain a low-frequency component, an intermediate-frequency component and a high-frequency component; s3, according to grading, an alkaline electrolytic cell, a proton exchange membrane electrolytic cell and an energy storage system are matched; s4, constructing a scheduling optimization model of the virtual power plant; and S5, solving the scheduling optimization model of the virtual power plant constructed in the step S4, and outputting a result. According to the method, the proposed strategy can solve the problems of wind and light power generation fluctuation stabilization and efficient utilization of renewable energy sources, collaborative optimization operation of energy storage and an electric hydrogen production device is achieved through frequency division matching and a collaborative regulation and control mechanism, the electricity abandoning rate can be effectively reduced, and the wind and light absorption efficiency and the system economy are remarkably improved.

Description

Technical Field

[0001] This invention relates to the fields of energy storage and power dispatching technology, and in particular to a virtual power plant day-ahead optimization dispatching method that integrates multiple types of electric hydrogen production devices. Background Technology

[0002] Virtual power plants (VPPs) are an important carrier for integrating distributed energy resources, energy storage devices and controllable loads. Their value in improving the flexibility of the power system through resource aggregation and coordinated regulation is becoming increasingly prominent.

[0003] Furthermore, hydrogen production by electricity, as an emerging energy storage method, is gradually attracting attention. However, the intermittency and volatility of wind and solar power output not only affect the stable operation of the power system but also significantly harm the operational reliability, lifespan, and efficiency of hydrogen production equipment. Therefore, how to effectively mitigate the fluctuations in wind and solar power output and improve the system's capacity to absorb renewable energy has become a hot research topic. It is necessary to decompose the original wind and solar power output and rationally determine the optimal operating depth of each dispatching device to achieve safe and economical operation of the VPP (Vehicle Power Plant).

[0004] Empirical mode decomposition (EMD) can decompose signals and identify different frequency components, helping to identify hidden patterns and features in the signal. Therefore, it has been widely used in the field of power system dispatching. However, different types of hydrogen production units have significant differences in response speed and regulation costs. Existing technologies that only decompose the new energy power signal into high-frequency and low-frequency components cannot effectively match the actual operating characteristics of hydrogen production units, which may cause the hydrogen production equipment to operate under heavy load or inefficiently, thus shortening its service life. Summary of the Invention

[0005] To address the problem that existing technologies, which only decompose renewable energy power signals into high-frequency and low-frequency components, cannot effectively match the actual operating characteristics of hydrogen production units, potentially leading to overloaded or inefficient operation of hydrogen production equipment and shortening its service life, this invention proposes a virtual power plant day-ahead optimization scheduling method that integrates multiple types of electric hydrogen production devices. Through frequency division matching and collaborative control mechanisms, this method achieves coordinated and optimized operation of energy storage and electric hydrogen production devices, enabling the proposed strategy to effectively reduce curtailment rates and significantly improve wind and solar power integration efficiency and system economy.

[0006] This invention is achieved through the following technical solution, including the following steps:

[0007] S1. Obtain the wind turbine WT and photovoltaic (PV) output data in the virtual power plant (VPP);

[0008] S2, adopting empirical mode decomposition (EMD) to decompose the output data of the fan WT and the photovoltaic PV obtained in step S1 into multiple frequency bands to obtain three components of low frequency, medium frequency and high frequency;

[0009] S3, according to the classification, respectively matching alkaline (alkaline, ALK) electrolytic tank, proton exchange membrane (proton exchange membrane, PEM) electrolytic tank and energy storage system (energy storage system, ESS);

[0010] S4, constructing a scheduling optimization model of a virtual power plant (VPP);

[0011] S5, solving the scheduling optimization model of the virtual power plant (VPP) constructed in step S4, and outputting the result.

[0012] Compared with the prior art, the beneficial effects of the present application are:

[0013] (1) In the prior art, the single-type electrolytic tank scheduling strategy (such as only using alkaline ALK electrolytic tank or proton exchange membrane PEM electrolytic tank) cannot adapt to the wide frequency fluctuation of wind and light output due to the single response characteristic of the equipment, and the power abandonment rate is generally high. The present application decomposes the wind and light output into high, medium and low frequency components by empirical mode decomposition (EMD), respectively matches the operating characteristics of the energy storage system (ESS), the proton exchange membrane PEM electrolytic tank and the alkaline ALK electrolytic tank, realizes frequency division collaborative consumption, and the power abandonment rate can be as low as 0.14%, close to the complete consumption effect.

[0014] (2) The prior art has the contradiction between cost and consumption efficiency: single alkaline ALK electrolytic tank has low cost but high power abandonment loss; single proton exchange membrane PEM electrolytic tank or combination of energy storage system ESS and proton exchange membrane PEM electrolytic tank has strong consumption capacity but high equipment investment cost. The present application realizes precise matching of "high frequency-energy storage system ESS, medium frequency-proton exchange membrane PEM electrolytic tank, low frequency-alkaline ALK electrolytic tank", reduces unnecessary high-cost equipment investment while ensuring low power abandonment rate. Compared with the optimal profit scheme (energy storage + alkaline electrolytic tank strategy profit 10117 yuan) in the prior art, the profit of the present application can reach 10234.7 yuan, increasing by 1.16%, and the power abandonment rate is reduced from 1.75% to 0.14%; the comprehensive cost is reduced by 597.5 yuan compared with the single proton exchange membrane PEM strategy, increasing by 6.2%, and the power abandonment rate is also reduced from 0.39% to 0.14%. The present application realizes the optimal balance between consumption efficiency and economy.

[0015] (3) Prior art does not decompose wind and light output into multiple frequency bands, resulting in devices being forced to bear non-adaptive fluctuations (such as alkaline ALK electrolytic cell being prone to damage due to frequent adjustment when dealing with high-frequency fluctuations, and proton exchange membrane PEM electrolytic cell being underutilized when dealing with low-frequency stable load for a long time). The present application is based on the frequency distribution mechanism of empirical mode decomposition EMD, so that the alkaline electrolytic cell only bears smooth low-frequency load (fluctuation amplitude less than 23kW / 5min), the proton exchange membrane PEM electrolytic cell matches the medium-frequency fluctuation (fluctuation amplitude between 23-94kW / 5min), and the energy storage system ESS deals with high-frequency impact (fluctuation amplitude higher than 94kW / 10min), which completely fits the response speed and adjustment capacity of various devices, and can effectively avoid device overload or inefficient operation. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The structure diagram of the virtual power plant VPP constructed by the present application.

[0017] Figure 2 The flowchart of the virtual power plant VPP optimization scheduling based on empirical mode decomposition EMD in the present application.

[0018] Figure 3 The high, medium and low frequency component screening diagram of wind and light signals in the simulation experiment of the present application.

[0019] Figure 4a The ALK electrolytic cell power distribution diagram in scenario 1 of the simulation experiment of the present application.

[0020] Figure 4b The PEM electrolytic cell power distribution diagram in scenario 1 of the simulation experiment of the present application.

[0021] Figure 4c The energy storage system ESS power distribution diagram in scenario 1 of the simulation experiment of the present application.

[0022] Figure 5 The energy storage system ESS power diagram before and after the simulation of the present application.

[0023] Figure 6 The cost, profit and power abandonment rate diagram under different scenarios in the simulation experiment of the present application. DETAILED DESCRIPTION

[0024] The advantages and characteristics of the present application will be illustrated and explained by the following non-limiting description of preferred embodiments, which are given by way of example only with reference to the accompanying drawings.

[0025] As Figure 1As shown in the structural diagram of the virtual power plant (VPP) constructed by the application, the virtual power plant VPP includes a wind turbine (WT), a photovoltaic (PV), a power-to-hydrogen (P2H), an energy storage system (ESS), and an electric load (EL). The power-to-hydrogen P2H includes alkaline (ALK) electrolyzer hydrogen production and proton exchange membrane (PEM) electrolyzer hydrogen production. The power-to-hydrogen P2H system can realize flexible consumption of electricity in a clean manner. When the electric load EL cannot consume the output of the wind turbine WT and the photovoltaic PV, that is, the output of the wind turbine WT and the photovoltaic PV is excessive, the power-to-hydrogen P2H system and the energy storage system ESS will actively participate in the consumption of the excessive electric energy. Conversely, when the wind turbine WT and the photovoltaic PV cannot meet the demand of the electric load EL, that is, the output of the wind turbine WT and the photovoltaic PV is insufficient, the system will enable the standby resource, that is, the electric energy stored in the energy storage system ESS. When the electric energy in the energy storage system ESS still cannot meet the demand of the electric load EL, electricity needs to be purchased from the electricity market to meet the demand.

[0026] As Figure 2 shown, the application provides a virtual power plant day-ahead optimization scheduling method fusing multiple types of power-to-hydrogen devices, including the following steps:

[0027] Step S1, obtaining the output data of the wind turbine WT and the photovoltaic PV in the virtual power plant VPP.

[0028] The output data of the wind turbine WT and the photovoltaic PV can be directly collected by the wind turbine factory and the photovoltaic factory, and the output of the wind turbine WT and the photovoltaic PV has intermittency and volatility. When the output data of the wind turbine WT and the photovoltaic PV cannot be directly obtained from the wind turbine factory and the photovoltaic factory, the K-means clustering algorithm can be used to perform clustering analysis on the annual wind turbine and photovoltaic output data, and the cumulative fluctuation median is used as an index to select a typical output scenario. Each scenario selects a representative typical day as the output data of the wind turbine WT and the photovoltaic PV as input for the next step.

[0029] Step S2, using empirical mode decomposition (EMD) to perform multi-frequency band decomposition on the output data of the wind turbine WT and the photovoltaic PV obtained in step S1 to obtain three components of low frequency, medium frequency, and high frequency.

[0030] Step S21, obtaining the remaining power according to the load balancing formula of the virtual power plant VPP.

[0031]

[0032] In the formula, P t This represents the sum of the output of the wind turbine WT and the photovoltaic power generation PV at time t, i.e. The remaining power at time t is the remaining power after the wind turbine WT and photovoltaic PV meet the electrical load EL demand at time t. and These represent the original wind turbine WT signal and the original photovoltaic PV signal at time t, respectively, i.e., the wind turbine WT and photovoltaic PV output data obtained in step S1; Let EL be the electrical load power at time t.

[0033] Step S22, if The virtual power plant (VPP) then needs to purchase electricity from the electricity market to meet the demand of the electrical load (EL), without needing to [address the issue of electricity]. Perform Empirical Mode Decomposition (EMD); otherwise, proceed to step S23.

[0034] Step S23: The remaining power obtained in step S21... The decomposition is as follows:

[0035]

[0036] In the formula, IMF i,t Represents the remaining power at time t The i-th intrinsic mode function (IMF); r n,t Indicates remaining power The nth-order residual signal or balance after decomposition;

[0037] Step S24: Use Empirical Mode Decomposition (EMD) to analyze the remaining power. The components are decomposed into low-frequency, mid-frequency, and high-frequency components.

[0038] Step S241, Define the signal sequence to be decomposed: Let The initial signal to be decomposed is the remaining power at time t, and the intrinsic mode function (IMF) counter is defined as: i = 1;

[0039] Step S242: For the current signal to be decomposed r i-1,t (Initial signal is r) 0,t ), traverse r i-1,t Determine the set of all local maxima and local minima, and construct the maximum envelope. and minimum envelope

[0040] Step S243: Introduce intermediate variables to represent the iterative process, as shown in the following formula:

[0041]

[0042]

[0043] wherein, is the mean value of the maximum envelope and the minimum envelope ; H i,t is an intermediate variable;

[0044] Step S244, check whether H i,t satisfies the definition of IMF: (1) the difference between the number of extreme points (maxima + minima) and the number of zero-crossing points in the signal is not more than 1, (2) the average value of the upper and lower envelopes at any time If the definition condition is not satisfied, H i,t is taken as the new signal to be decomposed, i.e. let r i-1,t = H i,t , repeat steps S242-S243 until H i,t satisfies the IMF condition. At this time, the component satisfying the IMF condition is defined as the i-th IMF, denoted as IMF i,t = H i,t ;

[0045] Step S245, remove the extracted IMF from the current signal to be decomposed to obtain a new residual signal: r i,t = r i-1,t - IMF i,t , check whether the residual signal satisfies the following termination condition: (1) the number of extreme points in r i,t is less than or equal to 1, (2) the energy of r i,t is less than 1% of the energy of the original signal. If the termination condition is satisfied, the decomposition process ends; otherwise, let i = i + 1 and return to step S242 to continue extracting the next IMF;

[0046] Step S246, the original signal is finally represented as the superposition of all IMFs and the residual signal: wherein IMF 1,t , IMF 2,t ,…, IMF n,t are IMFs arranged in descending order of frequency, and r n,t represents the final residual component.

[0047] Step S3, according to the classification, respectively match the alkaline ALK electrolyzer, the proton exchange membrane PEM electrolyzer and the energy storage system ESS.

[0048] Step S31, the low frequency, medium frequency and high frequency components obtained in step S246 are used for hydrogen production by the alkaline ALK electrolyzer, hydrogen production by the proton exchange membrane PEM electrolyzer and energy storage system ESS storage respectively. In the selection process of the low frequency, medium frequency and high frequency components, the modal components lower than the maximum fluctuation limit of the low frequency within 5 min are the low frequency, the modal components between the maximum fluctuation limits of the low frequency and the high frequency within 5 min are the medium frequency, and the remaining modal components are the high frequency. The maximum fluctuation limit of the low frequency is preferably 23 KW / 5 min, and the maximum fluctuation limit of the high frequency is preferably 94 KW / 5 min.

[0049] As shown in Figure 3 , the balance power relationship is as follows:

[0050]

[0051] In the formula: represents the electricity purchase amount of the virtual power plant VPP in the power market at time t (P represents that the VPP does not purchase electricity from the power market) ; and are the working powers of the alkaline ALK electrolyzer and the proton exchange membrane PEM electrolyzer at time t respectively; and represent the charging power and discharging power of the energy storage device ESS at time t respectively, is the power of the electrical load EL at time t.

[0052] The residual power is decomposed by empirical mode decomposition EMD, and the component frequency decreases with the increase of the decomposition layer, and the signal smoothness gradually improves. Based on this characteristic, the residual power can be decomposed into three types of low frequency, medium frequency and high frequency components. Under the premise of meeting the demand of the electrical load, the low frequency component is used for hydrogen production by the ALK electrolyzer, the medium frequency component is used for hydrogen production by the PEM electrolyzer, and the high frequency component is stored by the ESS.

[0053] Step S32, the residual power is reconstructed as follows:

[0054]

[0055] In the formula: is the high frequency component of the residual power at time t; is the medium frequency component; is the low frequency component.

[0056] Step S4, a scheduling optimization model of the virtual power plant VPP is constructed.

[0057] Step S41, a mathematical model is established with the configuration cost and the operation cost as the objective function to obtain a cost objective function minΓ, and the formula is as follows:

[0058]

[0059] In the formula, Γ P2H is the hydrogen production cost; Γ ESS is the energy storage cost; Γ buy is the electricity purchase cost; Γ cut is the electricity abandonment cost; is the hydrogen sales revenue.

[0060] The core target of the virtual power plant VPP constructed in the application is to realize efficient consumption of renewable energy by integrating the electricity-to-hydrogen P2H equipment and the energy storage system ESS, and meanwhile, the cost optimization and the economic benefit improvement are taken into account, so the mathematical model with the minimization of the configuration cost and the operation cost as the target, that is, the cost objective function, is constructed.

[0061] Step S411, the hydrogen production cost is constructed, and the formula is as follows:

[0062]

[0063]

[0064] In the formula, Γ inv is the investment cost of the hydrogen production equipment, is the investment cost of the unit hydrogen production equipment j, is the rated power of the electrolyzer j, φ represents the equipment discount rate, the value of the application is 0.05, y is the equipment service life, and the application is set to 20 years. Γ om is the operation and maintenance cost of the electrolyzer, ζ P2H is the ratio of the operation and maintenance cost to the present value of the investment cost of the hydrogen production equipment, and the value of the application is 0.04. Γ po is the electricity consumption cost of the hydrogen production link, is the electricity price of the hydrogen production link. Γ wa is the water consumption cost of the hydrogen production link, c wa is the unit water consumption cost, is the hydrogen production water consumption of the hydrogen production equipment j at the t time, η wa is the unit hydrogen production water consumption.

[0065] Step S412, the energy storage cost is constructed, and the formula is as follows:

[0066]

[0067]

[0068] In the formula, and respectively total investment cost and total operation and maintenance cost of the energy storage system ESS; and respectively unit power investment cost and unit capacity investment cost of the energy storage system ESS; ζ ESS is the ratio of operation and maintenance cost to investment cost present value of the energy storage system ESS, and the present application takes the value of 0.04.

[0069] Step S413, constructing the electricity purchase cost, the formula is as follows:

[0070]

[0071] In the formula: is the unit electricity price of purchasing electricity from the electricity market at time t, is the electricity purchase amount at time t.

[0072] Step S414, constructing the electricity abandonment cost, the formula is as follows:

[0073]

[0074]

[0075] In the formula: is the electricity abandonment amount at time t; c cut represents the unit electricity abandonment cost; and respectively actual power generation amount of the wind turbine WT and the photovoltaic PV at time t. The electricity abandonment cost is the abandonment cost of the wind power and the light power provided by the wind turbine WT and the photovoltaic PV.

[0076] Step S415, constructing the hydrogen sales revenue, the formula is as follows:

[0077]

[0078] In the formula: is the unit hydrogen sales price, and the present application takes the value of 40 yuan / kg; represents the hydrogen production amount of the electrolyzer j at time t. When the electrolyzer is running, the hydrogen production rate is directly related to the input power. The hydrogen production amount of the electrolyzer j at time t can be expressed as:

[0079]

[0080] In the formula: is the input power of the electrolyzer j at time t; represents the hydrogen production efficiency of the electrolyzer j, wherein the hydrogen production efficiencies of the alkaline electrolyzer ALK and the proton exchange membrane electrolyzer PEM are respectively 5.0 and 4.5 kWh / m 3The corresponding energy consumption levels are 56.0 and 50.4 kWh / kg, respectively; A represents an alkaline ALK electrolyzer; and P represents a proton exchange membrane PEM electrolyzer.

[0081] Step S42, constructing a constraint condition of a cost target function:

[0082] Step S421, constructing an electrolyzer hydrogen production power constraint, the formula being as follows:

[0083]

[0084] In the formula: is the input power of the electrolyzer j at time t, and j represents the electrolyzer type; A represents an alkaline ALK electrolyzer; and P represents a proton exchange membrane PEM electrolyzer.

[0085] Step S422, constructing an electrolyzer ramping constraint, the formula being as follows:

[0086]

[0087] In the formula: and are the lower limit and the upper limit of the power ramping of the electrolyzer j, respectively.

[0088] The power ramping constraint of the electrolyzer is realized by the difference between the powers at adjacent times, and the mathematical expression is shown in formula (3).

[0089] Step S423, constructing an energy storage system ESS constraint,

[0090] Step S4231, constructing a state of charge (SoC) constraint, the formula being as follows:

[0091]

[0092] In the formula: E t is the storage capacity of the energy storage system ESS at time t; E min and E max represent the lower limit and the upper limit of the storage capacity of the energy storage system ESS, respectively.

[0093] The state of charge SoC represents the percentage of the current remaining capacity of the battery to the rated capacity. The state of charge SoC constraint avoids the influence of overcharging or overdischarging on the performance and service life of the battery by limiting the operation range of the battery capacity, and the mathematical expression is shown in formula (4).

[0094] Step S4232, constructing an energy storage system ESS charging and discharging power constraint, the formula being as follows:

[0095] During the charging and discharging process of the energy storage system ESS, the charging power and the discharging power need to meet the range limit, and the mathematical expression is as follows:

[0096]

[0097]

[0098]

[0099] In the formula: and respectively, the charging power and discharging power at time t; and respectively, the maximum value of the charging and discharging power at time t; and respectively, the minimum value of the charging and discharging power at time t; It indicates that the energy storage device can only be in the charging or discharging state at a single time. In addition, the power update equation of the ESS at time t+Δt is:

[0100]

[0101] In the formula: parameter μ is the self-discharge rate of the energy storage device, and the value of the present application is 0.001; η ch and η dis respectively, the charging efficiency and discharging efficiency, both of which are 0.9 in the present application.

[0102] Step S424, construct the energy (power) balance constraint, the formula is as follows:

[0103]

[0104] Step S425, construct the actual output constraint of the wind turbine WT and the photovoltaic PV, the formula is as follows:

[0105]

[0106]

[0107] Step S43, integrate the constraint condition of the cost objective function minΓ constructed in step S41 and the cost objective function minΓ constructed in step S42, construct the scheduling optimization model of the virtual power plant VPP, the formula is as follows:

[0108]

[0109] The scheduling optimization model of the virtual power plant VPP, i.e. formula (12), is a linear programming model, which can be solved by calling the Yalmip toolbox and the CPLEX solver in Matlab.

[0110] Step S5, the scheduling optimization model of the virtual power plant VPP constructed in step S4 is solved based on the Matlab platform, Yalmip toolbox and CPLEX solver, the optimization model with the minimum cost as the target is solved while meeting the power constraint of the electrolytic cell; a comparison scene is set, the solving process is repeated and all scenes are covered; and the results are output, including the scheduling scheme, cost benefit index and power abandonment rate under each scene.

[0111] In the simulation environment, the effectiveness of the method is verified, the output data of the wind turbine WT and the photovoltaic PV are obtained, and the effectiveness of the collaborative optimization and scheduling strategy of the energy storage and hydrogen production equipment in the virtual power plant VPP is researched. The total capacity of wind power and photovoltaic power reaches 2.3MW. The K-means clustering algorithm is used to analyze the clustering of the annual wind and light output data, and the data is divided into 8 typical output scenes. However, in the case of extreme wind and light data, the K-means algorithm may cause a certain degree of deviation of the clustering center. Therefore, the cumulative fluctuation amount median is used as an index to select the typical day of each scene: after sorting the candidate days according to the cumulative fluctuation amount from small to large, the median corresponding date is taken as the representative typical day of the scene. This method can effectively reduce the interference of extreme weather conditions on the selection of typical days through the anti-extreme value characteristics of the median.

[0112] The parameters used in the subsequent experiments are given in Table 1.

[0113] Table 1: Optimization scheduling related parameters

[0114]

[0115]

[0116] In the present application, the maximum fluctuation limit of low frequency and high frequency is used as the dividing line to divide the power tasks of the energy storage system ESS, the proton exchange membrane PEM electrolyzer hydrogen production and the alkaline ALK electrolyte hydrogen production. Taking the typical day 6 as an example, the calculation results are shown in Figure 4a to Figure 4c Figure 4a to Figure 4c respectively, the power allocation of the alkaline ALK electrolyzer, the power allocation of the proton exchange membrane PEM electrolyzer and the power allocation of the energy storage system ESS in scene 1. The analysis results show that the fluctuation value of the high frequency component exceeds the set maximum fluctuation limit of high frequency, so the high frequency reconstruction component is determined as the power task of the energy storage system ESS. The fluctuation value of the low frequency component is lower than the maximum fluctuation limit of low frequency, so the low frequency reconstruction component is designated as the power task of the alkaline ALK electrolyzer hydrogen production. In addition, the fluctuation value of the medium frequency component is between the maximum fluctuation limit of low frequency and the maximum fluctuation limit of high frequency, so ​The power task of hydrogen production by the proton exchange membrane PEM electrolyzer. In the present application, the high-frequency maximum fluctuation limit value is set to 94 kW / 10 min, and the low-frequency maximum fluctuation limit value is set to 23 kW / 10 min.

[0117] In the simulation analysis process, comparative analysis is carried out for the following five configuration scenarios: 1) Scenario 1 is the comprehensive optimization scheduling strategy proposed in the present application; 2) Scenario 2 adopts the single optimization scheduling strategy for hydrogen production by alkaline ALK electrolyte; 3) Scenario 3 uses the optimization scheduling strategy for hydrogen production by proton exchange membrane PEM electrolyzer; 4) Scenario 4 is the cooperative optimization scheduling strategy for hydrogen production by alkaline ALK electrolyzer and energy storage system ESS; 5) Scenario 5 is the joint optimization scheduling strategy for hydrogen production by proton exchange membrane PEM electrolyzer and energy storage system ESS. The configuration results are shown in Table 2, wherein the scheduling period is set to 24 hours.

[0118] It should be noted that in Scenario 2 and Scenario 3, the original output signal does not need to be decomposed. In Scenario 4 and Scenario 5, the decomposed high-frequency signal is used as the power task of the energy storage system ESS, and the remaining signal is used as the power task of the alkaline ALK (Scenario 4) and proton exchange membrane PEM (Scenario 5) electrolyzer, respectively. The power task allocation in Scenario 1 is shown in Figure 4a to Figure 4c As can be seen from the figure, the power task of the alkaline ALK electrolyzer is relatively stable with small fluctuations. In contrast, the power task of the proton exchange membrane PEM electrolyzer fluctuates significantly with a larger fluctuation range than that of the alkaline ALK electrolyzer. The power task of the energy storage system ESS fluctuates the most, which is consistent with its device characteristics and can adapt to high-frequency output scenarios. The configuration results are shown in Table 2.

[0119] As can be seen from Table 2, the results of Scenario 1 are obtained based on the strategy proposed in the present application. Figure 5 As can be seen from the figure, most of the fluctuation components have been effectively absorbed and compensated, but some points with large amplitudes have not been completely compensated. Analysis shows that by using the strategy proposed in the present application, through reasonable decomposition and allocation of the initial signal, the consumption effect of wind and solar output can be significantly improved, and the curtailment rate is as low as 0.14%. This result shows that the method proposed in the present application can effectively suppress the fluctuations and uncertainties of wind and solar output, and is close to the effect of complete suppression.

[0120] Table 2 scheduling results

[0121]

[0122] Based on the configuration data of Scenario 2-5 in Table 2, combined with the cost-profit-curtailment rate comparative analysis shown in Figure 6

[0123] ​(1) Single alkaline ALK electrolyzer hydrogen production strategy (Scenario 2): The system comprehensive cost is 8380.7 yuan, which is reduced by 436.4 yuan compared with Scenario 1, but its abandoned electricity rate is as high as 3.14% (Scenario 1 abandoned electricity rate is only 0.14%), resulting in insufficient renewable energy consumption, and the final profit is reduced by 131.5 yuan compared with Scenario 1. The results show that the low dynamic response characteristics of the alkaline ALK electrolyzer cannot effectively track high-frequency power fluctuations, resulting in power loss.

[0124] (2) Single proton exchange membrane PEM electrolyzer hydrogen production strategy (Scenario 3): Due to the high investment cost of the proton exchange membrane PEM electrolyzer, its comprehensive cost is increased by 287.6 yuan compared with Scenario 1, the abandoned electricity rate is 1.87%, and the system profit is reduced by 214.3 yuan. The results verify that single device regulation cannot balance economic efficiency and fluctuation suppression effect.

[0125] (3) Energy storage system ESS and alkaline ALK electrolyzer hydrogen production combined regulation strategy (Scenario 4): The collaborative control of energy storage system ESS and alkaline ALK electrolyzer reduces the abandoned electricity rate to 1.75%, but due to the investment cost of energy storage system ESS, Scenario 4 is increased by 44.4 yuan compared with Scenario 2. Due to the high-frequency rapid response characteristics of the energy storage system ESS, it is complementary to the alkaline ALK electrolyzer hydrogen production, and the abandoned electricity rate is reduced to 1.75%, which is slightly lower than Scenario 2, but overall it is still not as good as Scenario 1.

[0126] (4) Energy storage system ESS and proton exchange membrane PEM electrolyzer hydrogen production combined regulation strategy (Scenario 5): The remaining signal is allocated to the proton exchange membrane PEM electrolyzer with faster response speed, which can theoretically improve the fluctuation suppression effect. But the actual results show that due to the high investment cost of the proton exchange membrane PEM electrolyzer, its comprehensive cost is increased by 950.5 yuan compared with Scenario 1, the abandoned electricity rate is 0.44%, and the profit is reduced by 1007.4 yuan. The results reveal that the selection of equipment needs to balance the response characteristics and economic cost.

[0127] After simulation verification, the hierarchical optimization scheduling strategy (Scenario 1) proposed in this paper realizes the optimal balance in abandoned electricity rate (0.14%), comprehensive cost (8817.1 yuan) and system profit (10234.7 yuan) through multi-device collaboration and frequency division regulation mechanism, which verifies the effectiveness of the proposed method in improving renewable energy consumption and system economy.

[0128] The present application aims at the wind and light power generation fluctuation suppression and renewable energy efficient utilization problem, and proposes a frequency division regulation strategy of energy storage system ESS and electrolytic cell cooperation. Multi-scene comparative simulation results show that: based on the experience mode decomposition EMD high, medium and low frequency fluctuation limit division, combined with the characteristic matching mechanism of alkali ALK, proton exchange membrane PEM and energy storage system ESS, the strategy can realize the significant improvement of wind and light output consumption efficiency, and the abandoned power rate is as low as 0.14%. Compared with single type alkali ALK or proton exchange membrane PEM strategy, it is reduced by 2.00%~3.00%, compared with energy storage system ESS-single type electrolytic cell cooperation strategy, it is reduced by 1.61%, which shows that the multi-device frequency division cooperation regulation has the advantages in improving the wind and light consumption capacity and system economy.

[0129] In addition to the above embodiments, the present application can have other implementation manners, and any technical solutions formed by equivalent replacement or equivalent transformation shall fall within the protection scope required by the present application.

Claims

1. A method for day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices, characterized in that: It comprises the following steps: S1, obtaining wind turbine WT and photovoltaic PV output data in a virtual power plant VPP; S2, performing multi-band decomposition on the wind turbine WT and photovoltaic PV output data obtained in step S1 by using empirical mode decomposition EMD, to obtain three components of low frequency, medium frequency and high frequency; S3, matching alkaline ALK electrolyzer, proton exchange membrane PEM electrolyzer and energy storage system ESS according to the classification; S4, constructing a scheduling optimization model of the virtual power plant VPP; S5, solving the scheduling optimization model of the virtual power plant VPP constructed in step S4, and outputting the result.

2. The method of claim 1, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The specific steps of step S2 are as follows: S21, obtain the residual power according to a load balancing formula of the virtual power plant VPP The formula is as follows: where P t represents the sum of the outputs of the wind turbine WT and the photovoltaic PV at time t; is the residual power at time t; and represent the original wind turbine WT signal and the original photovoltaic PV signal at time t, respectively; is the electrical load EL power at time t; S22, if The virtual power plant (VPP) then needs to purchase electricity from the electricity market to meet the demand of the electrical load (EL), without needing to [address the issue of electricity]. Perform Empirical Mode Decomposition (EMD); otherwise proceed to step S23. S23, the remaining power obtained in step S21 is Decomposition is performed, and the formula is as follows: In the formula, IMF i,t represents the residual power at time t the i-th intrinsic mode function IMF; r n,t represents the residual power the n-th order residual signal or the residual after decomposition S24, decomposing the residual power by using empirical mode decomposition (EMD) to obtain three components of low frequency, medium frequency and high frequency. decomposing the residual power by using empirical mode decomposition (EMD) to obtain three components of low frequency, medium frequency and high frequency.

3. The method of claim 2, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The specific steps of step S24 are as follows: S241、define the signal sequence to be decomposed: let That is, the initial signal to be decomposed is the residual power at time t, and the intrinsic mode function IMF counter is defined: i = 1; S242、to the current signal to be decomposed r i-1,t , the initial signal is r 0,t , traverse r i-1,t , determine all local maximum point set and local minimum point set, construct maximum envelope line and minimum envelope line S243, introducing an intermediate variable to represent the iteration process, and the formula is as follows: wherein is the mean value of the maximum envelope and the minimum envelope H i,t is an intermediate variable S244, check H i,t whether the definition of IMF is satisfied: (1) the difference between the number of extreme points and the number of zero-crossing points in the signal is not more than 1, (2) at any time, the average value of the upper and lower envelope lines if the definition condition is not satisfied, H i,t is taken as a new signal to be decomposed, i.e. let r i-1,t = H i,t , repeat steps S242-S243 until H i,t satisfies the IMF condition; at this time, the component satisfying the IMF condition is defined as the i-th IMF, denoted as IMF i,t = H i,t ; S245, eliminate the extracted IMF from the current signal to be decomposed, and obtain a new residual signal: r i,t = r i-1,t -IMF i,t , check whether the residual signal satisfies the following termination conditions: (1) the number of extreme points in r i,t is less than or equal to 1, (2) the energy of r i,t is less than 1% of the energy of the original signal; If the termination condition is met, the decomposition process ends; otherwise, let i=i+1, return to step S242, and continue to extract the next IMF; S246, The original signal is finally represented as the superposition of all IMFs and the residual signal: where IMFs 1,t , IMF 2,t ,..., IMF n,t are the IMFs arranged from high to low in frequency, and r n,t represents the final residual component.

4. The method of claim 3, wherein the method is characterized in that: The specific steps of step S3 are as follows: S31, using the low-frequency, medium-frequency and high-frequency components obtained in step S246 for hydrogen production by alkaline ALK electrolyzer, hydrogen production by proton exchange membrane PEM electrolyzer and energy storage by energy storage system ESS, and balancing the power relationship as follows: In the formula: represents the electricity purchase amount of the virtual power plant VPP in the electricity market at time t; and respectively represent the working power of the alkaline electrolyzer ALK and the proton exchange membrane electrolyzer PEM at time t; and respectively represent the charging power and the discharging power of the energy storage device ESS at time t; S32, the remaining electric power is restructured as follows: In the formula: is the high frequency component of the residual power at time t; is the medium frequency component; is the low frequency component.

5. The method of claim 4, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The specific steps of step S4 are as follows: S41, establishing a mathematical model with configuration cost and operation cost as objective functions to obtain a cost objective function minΓ, and the formula is as follows: where Γ P2H is the cost of hydrogen production; Γ ESS is the cost of energy storage; Γ buy is the cost of electricity purchase; Γ cut is the cost of electricity rejection; is the revenue of hydrogen sale; S42, constructing constraint conditions of the cost objective function; S43, integrating the cost objective function minΓ constructed in step S41 and the constraint conditions of the cost objective function minΓ constructed in step S42 to construct a scheduling optimization model of the virtual power plant VPP.

6. The method of claim 5, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The specific steps of step S41 are as follows: S411, constructing hydrogen production cost, and the formula is as follows: wherein: Γ inv is the investment cost of the hydrogen production equipment; is the investment cost of the unit hydrogen production equipment j; is the rated power of the electrolyzer j; φ represents the equipment discount rate; y is the equipment service life; Γ om is the operation and maintenance cost of the electrolyzer; ζ P2H is the ratio of the operation and maintenance cost to the present value of the investment cost of the hydrogen production equipment; Γ po is the electricity consumption cost of the hydrogen production link; is the electricity price of the hydrogen production link; Γ wa is the water consumption cost of the hydrogen production link; c wa is the unit water consumption cost; is the water consumption of the hydrogen production equipment j at time t; η wa is the unit hydrogen production water consumption; S412, constructing energy storage cost, and the formula is as follows: In the formula: and respectively represent the total investment cost and the total operation and maintenance cost of the energy storage system ESS; and respectively represent the unit power investment cost and the unit capacity investment cost of the energy storage system ESS; ζ ESS is the ratio of the operation and maintenance cost to the present value of the investment cost of the energy storage system ESS. S413, constructing electricity purchase cost, and the formula is as follows: In the formula: is the unit price of electricity purchased from the electricity market at time t, is the electricity purchase amount at time t; S414, constructing abandoned electricity cost, and the formula is as follows: In the formula: is the abandoned power at time t; c cut represents the unit abandoned power cost; and are the actual power generation of the wind turbine WT and the photovoltaic PV at time t, respectively; S415, constructing hydrogen sales revenue, and the formula is as follows: In the formula: is the unit price of hydrogen; represents the hydrogen production amount of electrolyzer j at time t. The hydrogen production amount of electrolyzer j at time t is represented by: In the formula: Pj(t) is the input power of the electrolyzer j at time t; represents the hydrogen production efficiency of the electrolyzer j.

7. The method of claim 6, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The specific steps of step S42 are as follows: S421, constructing electrolyzer hydrogen production power constraint, and the formula is as follows: wherein: Pj(t) is the input power of electrolyzer j at time t, j represents the electrolyzer type; A denotes an alkaline ALK electrolyzer; P denotes a proton exchange membrane PEM electrolyzer; S422, constructing electrolyzer climbing constraint, and the formula is as follows: In the formula: and are the lower and upper bounds of the power ramp of the electrolyzer j, respectively. S423, constructing energy storage system ESS constraint, S4231, constructing charge constraint, and the formula is as follows: E min ≤E t ≤E max (4) In the formula: E t E(t) is the energy storage amount of the energy storage system ESS at time t min E max respectively represent the lower limit and the upper limit of the energy storage amount of the energy storage system ESS S4232, constructing energy storage system ESS charging and discharging power constraint, and the formula is as follows: wherein: Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; Pcharge(t) and Pdischarge(t) are the charging and discharging power at time t, respectively; wherein: the parameter μ is the self-discharge rate of the energy storage device; η ch and η dis are the charging and discharging efficiencies, respectively. S424, constructing energy balance constraint, and the formula is as follows: S425, constructing actual output constraint of wind turbine WT and photovoltaic PV, and the formula is as follows:

8. The method of claim 7, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The scheduling optimization model formula of the virtual power plant VPP in step S43 is as follows:

9. The method of claim 1 to 8, wherein the method is characterized in that: The method for obtaining wind turbine WT and photovoltaic PV output data in the virtual power plant VPP in step S1 is as follows: using K-means clustering algorithm to perform clustering analysis on annual wind turbine and photovoltaic output data, selecting typical output scenarios with cumulative fluctuation amount median as an index, and selecting representative typical days in each scenario as wind turbine WT and photovoltaic PV output data.

10. The method of claim 8, wherein the method is a method of day-ahead optimal scheduling of a virtual power plant integrated with multi-type electrolytic hydrogen generation devices. The solving method of the scheduling optimization model of the virtual power plant VPP constructed in step S4 is that, based on a Matlab platform, a Yalmip toolbox and a CPLEX solver are called to solve.

Images