A microgrid control system and method

Abstract

The present application relates to the field of microgrid control technology, and specifically to a control system and method for a microgrid, the method comprising: obtaining the actual power sequence and various power influence sequences of each electrolytic cell; obtaining each steady-state submatrix of each electrolytic cell; constructing the steady-state power of each preset time period of each electrolytic cell; obtaining the power matching cluster and historical power matching cluster of each electrolytic cell; obtaining the best matching power of each preset time period of each electrolytic cell; constructing a first difference and a second difference of each electrolytic cell; determining the microgrid power priority index of each electrolytic cell to control the operating power of the microgrid. The present application equips each electrolytic cell with a DC power supply, provides the control adaptability of the electrolytic cell, improves the compatibility of the microgrid control method and the electrolytic cell, and effectively increases the service life of the electrolytic cell.

Description

Technical Field

[0001] The present application relates to the field of microgrid control technology, and in particular to a control system and method for a microgrid. Background Art

[0002] As a clean energy, hydrogen energy has the characteristics of high energy density, renewability, storability, and pollution-free. It is hailed as the best way to solve the energy resource crisis and environmental crisis. The water electrolysis hydrogen production technology with electrolyzer as the key equipment has broad application prospects and market scale. Microgrid is a micro power supply network composed of distributed power sources, loads and energy storage equipment in a certain area. Electrolyzers have unique operating characteristics. Electrolysis hydrogen production microgrids are prone to uneven power distribution, resulting in frequent shutdown of electrolyzers, which directly affects the equivalent service life of the electrolyzer system. Appropriate microgrid control strategies must be adopted to maintain the operating stability of the entire system and obtain higher electrolysis hydrogen production efficiency.

[0003] In the prior art, the DC power supply used for hydrogen production by electrolysis is a single power supply or several power supplies connected in parallel. However, due to the fluctuation of the external load during the operation of the electrolyzer, there is a problem that the operating characteristics of a specific electrolyzer cannot be controlled, so that the electrolyzer cannot operate at the optimal power, thereby greatly reducing the operating life of the electrolyzer and the hydrogen production efficiency. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a control system and method for a microgrid, and the technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a microgrid control method, the method comprising the following steps:

[0006] Obtain the actual power data of each electrolytic cell at all acquisition times, and the data of various power influencing factors of the electrolytic cell at all acquisition times, and form various power influencing sequences of each electrolytic cell;

[0007] Various power impact sequences in each preset time period are arranged in rows to form a steady-state submatrix;

[0008] Analyze the average distribution of actual power in each preset time period in each electrolytic cell to construct the steady-state power of each preset time period in each electrolytic cell;

[0009] The electrolytic cell whose actual power data is equal to the preset optimal power is used as the electrolytic cell to be analyzed;

[0010] The various power influence sequences of the electrolytic cells to be analyzed are combined into a historical state matrix of each electrolytic cell to be analyzed;

[0011] Clustering each steady-state submatrix of each electrolytic cell and all the historical state matrices to obtain clusters, and recording the clusters that do not contain the steady-state submatrix as historical power matching clusters;

[0012] Obtain the historical power data of each electrolytic cell, and in each preset time period of each electrolytic cell, obtain the best matching power of each preset time period of each electrolytic cell according to the distribution of the historical power data corresponding to all historical state matrices in the cluster where the steady-state submatrix is ​​located;

[0013] Constructing a first difference of the cluster where each steady-state submatrix of each electrolytic cell is located according to the difference between the steady-state submatrix in the cluster where each steady-state submatrix is ​​located and all the historical state matrices, and the difference between all the historical state matrices in the historical power matching cluster;

[0014] Constructing a second difference for each electrolytic cell according to the difference between the steady-state power and the best matching power in each preset time period of each electrolytic cell and the average distribution of the first differences of the clusters where all the steady-state sub-matrices of each electrolytic cell are located;

[0015] The pause power urgency of each electrolyzer is constructed according to the change of the actual power data of each electrolyzer, and the microgrid power priority index of each electrolyzer is determined in combination with the second difference to control the power of the DC power supply of the microgrid.

[0016] Preferably, the steady-state power of each electrolytic cell in each preset time period is the average of the actual power data of each electrolytic cell in each preset time period.

[0017] Preferably, the process of constructing the optimal matching power for each preset time period of each electrolytic cell is as follows:

[0018] In each preset time period of each electrolytic cell, determine the mean of all historical actual powers of each historical state matrix in the cluster where the steady-state submatrix is ​​located;

[0019] In each preset time period of each electrolytic cell, the average value of the means of all historical state matrices in the cluster where the steady-state submatrix is ​​located is determined, and recorded as the best matching power in each preset time period of each electrolytic cell.

[0020] Preferably, the process of constructing the first difference of the clusters where each steady-state submatrix of each electrolytic cell is located is:

[0021] Determine the mean of the distances between each steady-state submatrix of each electrolytic cell and all historical state matrices in the cluster to which it belongs, and record it as the first mean of the cluster to which each steady-state submatrix of each electrolytic cell belongs;

[0022] Analyze the mean of the distances between all historical state matrices in each historical power matching cluster of each electrolytic cell, and record it as the second mean of each historical power matching cluster of each electrolytic cell;

[0023] The first difference of the cluster where each steady-state sub-matrix of each electrolytic cell is located is the average of the differences between the first mean of the cluster where each steady-state sub-matrix of each electrolytic cell is located and all the second means.

[0024] Preferably, the construction process of the second difference of each electrolytic cell is:

[0025] Constructing the long-term deviation increasing significance of each electrolytic cell according to the difference between the steady-state power and the best matching power of each electrolytic cell in each preset time period;

[0026] Analyze the mean of the first difference of the clusters of all steady-state sub-matrices of each electrolyzer, and record it as the power adaptation excellence of each electrolyzer;

[0027] The second difference of each electrolytic cell is determined based on the significant degree of long-term deviation increase and the excellent degree of power adaptation, and the second difference of each electrolytic cell is the result of the fusion of the significant degree of long-term deviation increase and the excellent degree of power adaptation of each electrolytic cell.

[0028] Preferably, the process of constructing the significant degree of long-term deviation increase of each electrolytic cell is as follows:

[0029] Determine the difference between the steady-state power and the optimal matching power of each electrolytic cell in each preset time period, and record it as the power modulation deviation of each electrolytic cell in each preset time period;

[0030] The power modulation deviations of all preset time periods of each electrolytic cell are arranged in ascending time order to form a power modulation sequence of each electrolytic cell;

[0031] A trend test algorithm is used to obtain the trend test statistics of the power modulation sequence of each electrolytic cell;

[0032] The degree of significant increase in long-term deviation trend of each electrolyzer is determined based on all the power modulation deviations and trend test statistics. The power adaptation excellence of each electrolyzer is the result of the fusion of the mean value of the power modulation deviations of all steady-state sub-matrices of each electrolyzer and the trend test statistics.

[0033] Preferably, the construction process of the pause power urgency of each electrolyzer is:

[0034] The actual power data of all acquisition moments in each electrolytic cell constitute the actual power sequence of each electrolytic cell;

[0035] Taking the last data element with a value of 0 in the actual power sequence of each electrolytic cell as the center, the number of values ​​0 to the right of the center is counted as the last pause duration of each electrolytic cell;

[0036] The time interval from the center to the last actual power data in the actual power sequence is taken as the last startup operation duration of each electrolytic cell;

[0037] The pause power urgency of each electrolytic cell is the ratio of the last pause duration to the last start-up operation duration of each electrolytic cell.

[0038] Preferably, the microgrid power priority index of each electrolyzer is a result of a fusion of the second difference and the pause power urgency of each electrolyzer.

[0039] Preferably, the method of controlling the power of the DC power supply of the microgrid includes:

[0040] The power modulation sequence of each electrolyzer is used as the input of the autoregressive integrated moving average model, and the power adjustment prediction value of each electrolyzer is output;

[0041] When the power adjustment prediction value is negative, the power of the DC power supply corresponding to each electrolyzer is subtracted from the corresponding microgrid power priority index power adjustment prediction value;

[0042] When the power adjustment prediction value is positive, the power of the DC power supply corresponding to the electrolyzer in the power absorption group plus all electrolyzers whose corresponding microgrid power priority index is negative are recorded as the power supply group.

[0043] In a second aspect, an embodiment of the present application further provides a control system for a microgrid, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0044] This application has at least the following beneficial effects:

[0045] By analyzing the closeness of the feature data matching in the clustering results corresponding to each steady-state submatrix of each electrolyzer, the first difference of each steady-state submatrix of each electrolyzer is obtained to avoid the data of the steady-state submatrix in the electrolyzer having a unique pattern, resulting in poor data matching and mining effect based on the historical operation data of the electrolyzer, and large error in the best matching power obtained, which affects the accuracy of subsequent microgrid control; based on the pause and restart characteristics in the actual power data sequence of each electrolyzer, the microgrid power priority index of the electrolyzer is obtained, and the start-stop characteristics of the electrolyzer are comprehensively considered to prompt the microgrid to give priority to the restarted electrolyzer to provide power, provide the speed at which the electrolyzer reaches the conditions for electrolytic hydrogen production, avoid frequent pauses and restarts of the electrolyzer, and improve the efficiency of electrolytic hydrogen production. This application is equipped with a DC power supply for each electrolyzer, provides the control adaptability of the electrolyzer, improves the fit between the microgrid control method and the electrolyzer, and effectively increases the service life of the electrolyzer. At the same time, modularization greatly reduces the cost of later maintenance, is easy to replace, and the faulty module does not affect the normal operation of other electrolyzers. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 A flow chart of steps of a control method for a microgrid;

[0048] Figure 2 Extract a flow chart for the second difference;

[0049] Figure 3 Extracting flow chart for microgrid power priority index;

[0050] Figure 4 Schematic diagram of the process of regulating the operating power of the microgrid. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the control system and method of a microgrid proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0053] The specific scheme of a microgrid control system and method provided by the present application is described in detail below with reference to the accompanying drawings.

[0054] See also Figure 1 , which shows a flow chart of a microgrid control method provided by an embodiment of the present application, the method comprising the following steps:

[0055] Step S1: collect data from each electrolytic cell and perform preprocessing.

[0056] The present application abandons the use of a single DC power supply or several power supplies in parallel, and each electrolyzer in the electrolyzer system is equipped with a separate DC power supply. A microgrid is used to control all DC power supplies. When the operating power required by the electrolyzer is lower than the rated output power of the DC power supply, the microgrid absorbs and consumes the power of the electrolyzer. When the operating power required by the electrolyzer is higher than the rated output power of the DC power supply, the microgrid provides additional power support to the electrolyzer.

[0057] The acquisition module is used to obtain all types of electrolyzer power influencing factor data and actual power data in each electrolyzer, wherein the electrolyzer power influencing factor data include: electrolyte concentration, tank temperature, ambient temperature, hydrogen production, working voltage, electrolysis current, and sampling is performed at a preset time interval T, and the total sampling time is preset t. The sampled electrolyzer power influencing factor data are arranged in chronological order to form a data sequence of various electrolyzer power influencing factors for each electrolyzer; the actual power data at all acquisition times are arranged in ascending time order to form an actual power sequence of each electrolyzer.

[0058] It should be noted that the values ​​of the preset time interval and the preset duration are both manually set. In this embodiment, the value of the preset time interval is 1s, and the value of the preset duration is 30min. The implementer can set them according to his own situation, and this embodiment does not impose any special restrictions.

[0059] In this embodiment, all kinds of power influence sequences of each electrolytic cell are normalized to eliminate dimension. As an embodiment of the present application, the Z-score normalization method is adopted in this embodiment. The implementer may also adopt other normalization methods in the prior art. This embodiment does not make any special restrictions. The Z-score normalization method is a well-known technology and the specific process will not be repeated.

[0060] Step S2: Analyze the average distribution of the actual power in each preset time period in each electrolytic cell to construct the steady-state power of each preset time period in each electrolytic cell.

[0061] The optimal power during the operation of the electrolyzer is affected by the electrolyzer equipment and working environment. For example, if the ambient temperature of the electrolyzer is low, part of the DC power supply's electrical energy is used to increase the temperature of the tank because electrolytic hydrogen production requires a high temperature environment. At this time, the rated power of the DC power supply is likely to be lower than the optimal power of the electrolyzer. The DC power supply cannot meet the power requirements of the electrolyzer, and additional power support is required for the electrolyzer based on the microgrid. In the process of controlling the electrolytic hydrogen production microgrid, it is usually necessary to perform macroscopic control based on the optimal power of the electrolyzer. However, the factors affecting the electrolyzer power have a large waveform, and the optimal power of the electrolyzer also changes accordingly, making it difficult to dynamically track the optimal power of the electrolyzer.

[0062] The average distribution of the actual power in each preset time period in each electrolytic cell is analyzed to construct the steady-state power of each electrolytic cell in each preset time period.

[0063] The data sequence of the power influencing factors of each electrolytic cell in each electrolytic cell is divided into various sub-sequences according to preset time periods, recorded as various power influence sequences in each preset time period, and the various power influence sequences in each preset time period form a steady-state sub-matrix, recorded as each steady-state sub-matrix of each electrolytic cell.

[0064] It should be noted that the value of the preset time period is manually set. In this embodiment, the value of the preset time period is 3 minutes. The implementer can set it according to the specific situation. This embodiment does not impose any special restrictions.

[0065] Furthermore, the average of the actual power data of each electrolytic cell in each preset time period is analyzed as the steady-state power of each electrolytic cell in each preset time period.

[0066] Step S3: Obtain the historical power data of each electrolytic cell, and in each preset time period of each electrolytic cell, obtain the best matching power of each preset time period of each electrolytic cell according to the distribution of historical power data corresponding to all historical state matrices in the cluster where the steady-state submatrix is ​​located.

[0067] In the process of controlling the electrolysis hydrogen production microgrid, macro control is usually required based on the optimal power of the electrolyzer. However, the factors affecting the electrolyzer power have a large waveform, and the optimal power of the electrolyzer also changes accordingly, making it difficult to dynamically track the optimal power of the electrolyzer.

[0068] The historical operation data of hydrogen production by electrolysis contains the operation characteristic information of the electrolyzer when the actual power of the electrolyzer reaches the optimal power. By analyzing the correlation between the historical operation data and the steady-state submatrix, the deep-level characteristics of the electrolyzer data can be mined to obtain the optimal matching power of the electrolyzer, specifically:

[0069] S301: Acquire all historical power influencing factor data of each electrolytic cell to be analyzed; cluster each steady-state sub-matrix of each electrolytic cell and all the historical state matrices to obtain a power matching cluster and a historical power matching cluster of each steady-state sub-matrix of each electrolytic cell.

[0070] The electrolytic cell with actual power data equal to the preset optimal power is taken as the electrolytic cell to be analyzed. All historical power influencing factor data of each electrolytic cell to be analyzed are obtained from the database, and according to the method for obtaining the electrolytic cell power influencing factor data sequence, all historical electrolytic cell power influencing factor data sequences of a preset number of electrolytic cells to be analyzed are obtained from the database; all the historical power influencing factor data sequences are arranged in rows to form a historical state matrix of each electrolytic cell to be analyzed.

[0071] It should be noted that the sequence length of the historical electrolyzer power influencing factor data sequence is L. The sequence length L, the value of the preset number, and the value of the preset optimal power are all set manually. In this embodiment, the value of the sequence length L is 5min, the value of the preset number is 50, and the value of the preset optimal power is 100kW. The implementer can set it according to the specific situation, and this embodiment does not impose any special restrictions.

[0072] Furthermore, all the steady-state submatrices and all the historical state matrices of each electrolyzer are used as inputs of the CURE clustering algorithm, and the cumulative sum of the distances of the same electrolyzer power influencing factor data sequences between the steady-state submatrices and all the historical state matrices is calculated. The size of the shrinkage factor in the CURE algorithm is set to a preset value, and the clustering result is output. Among them, the clustering cluster that does not contain the steady-state submatrix is ​​recorded as the historical power matching cluster, and there is only one steady-state submatrix in the clustering cluster where the steady-state submatrix is ​​located.

[0073] It should be noted that there are many commonly used methods for measuring distance. As an embodiment of the present application, DTW distance is used to calculate the metric distance in this embodiment. The implementer may also use Euclidean distance, Manhattan distance or Lagrange distance to measure the distance between matrices. This embodiment does not impose any special restrictions.

[0074] It should be noted that the value of the shrinkage factor in the CURE algorithm is artificially set. In this embodiment, the preset value is 0.5. The implementer can set it according to the specific situation. This embodiment does not impose any special restrictions.

[0075] The CURE clustering algorithm is a well-known technology, and the specific process will not be described in detail.

[0076] S302: In each preset time period of each electrolytic cell, the best matching power of each preset time period of each electrolytic cell is obtained according to the distribution of historical power data corresponding to all historical state matrices in the cluster where the steady-state submatrix is ​​located.

[0077] Further, in each preset time period of each electrolytic cell, the mean of all historical actual powers of each historical state matrix in the cluster where the steady-state submatrix is ​​located is determined;

[0078] In each preset time period of each electrolytic cell, the average value of the means of all historical state matrices in the cluster where the steady-state submatrix is ​​located is determined, and recorded as the best matching power in each preset time period of each electrolytic cell.

[0079] It should be noted that one preset time period corresponds to one steady-state sub-matrix.

[0080] The closer the metric distance between the data in the clusters, the better the data matching mining effect of the steady-state submatrix and the historical state matrix, and the more it can avoid the steady-state submatrix data in the electrolyzer having a unique pattern, resulting in poor data matching mining effect based on the historical operation data of the electrolyzer, and a large error in the obtained optimal matching power, which affects the accuracy of subsequent microgrid control.

[0081] Step S4: construct the first difference of the cluster where each steady-state submatrix of each electrolytic cell is located according to the difference between the steady-state submatrix in the cluster where each steady-state submatrix is ​​located and all the historical state matrices, and the difference between all the historical state matrices in the historical power matching cluster.

[0082] In order to avoid the unique pattern of the steady-state submatrix data in the electrolyzer, which leads to poor data matching and mining effect based on the historical operation data of the electrolyzer, the error of the best matching power obtained is large, which affects the accuracy of the subsequent microgrid control. Therefore, by analyzing the closeness of the matching of the characteristic data in the clustering results corresponding to each steady-state submatrix of each electrolyzer, the first difference of each steady-state submatrix of each electrolyzer is obtained, which is specifically:

[0083] Determine the mean of the distances between each steady-state submatrix of each electrolytic cell and all historical state matrices in the cluster to which it belongs, and record it as the first mean of the cluster to which each steady-state submatrix of each electrolytic cell belongs;

[0084] Analyze the mean of the distances between all historical state matrices in each historical power matching cluster of each electrolytic cell, and record it as the second mean of each historical power matching cluster of each electrolytic cell;

[0085] It should be noted that there are many commonly used methods for measuring distance. As an embodiment of the present application, Euclidean distance is used to calculate the metric distance in this embodiment. The implementer may also use Manhattan distance or Lagrange distance to measure the distance between matrices. This embodiment does not impose any special restrictions.

[0086] The first difference of each steady-state submatrix of each electrolytic cell is determined based on the first mean and all the second means, and the first difference of the cluster where each steady-state submatrix of each electrolytic cell is located is the mean of the difference between the first mean of the cluster where each steady-state submatrix of each electrolytic cell is located and all the second means.

[0087] It should be noted that there are many ways to measure differences. In this embodiment, the difference between the first mean and the second mean is measured by calculating the absolute value of the difference between them. The implementer may also use other methods that reflect the difference, such as ratio, to measure the difference between the values. This embodiment does not impose any special restrictions.

[0088] The larger the first difference is, the better the data matching and mining effect between the steady-state submatrix and the historical state matrix is, and the more likely the best matching power obtained is to be the optimal power of the electrolyzer; conversely, the smaller the first difference is, the worse the data matching and mining effect between the steady-state submatrix and the historical state matrix is, and the less likely the best matching power obtained is to be the optimal power of the electrolyzer.

[0089] Step S5: construct the second difference of each electrolytic cell according to the difference between all steady-state powers of each electrolytic cell and the best matching power of each steady-state sub-matrix, the variation trend of the difference and the overall distribution of the first differences of all steady-state sub-matrices of each electrolytic cell.

[0090] In the process of hydrogen production by electrolysis, the electrolyzer has many performance parameters, high nonlinearity, and complex coupling relationships. Moreover, after long-term operation, the characteristics of the electrolyzer components may shift, resulting in a deviation between the actual power of the electrolyzer and the optimal power of the electrolyzer. The greater the deviation, the lower the electrolysis efficiency of the electrolyzer, and the greater the need for microgrid control.

[0091] Therefore, the second difference of each electrolytic cell is determined by analyzing the difference between all steady-state powers of each electrolytic cell and the best matching power of each steady-state sub-matrix, the variation trend of the difference, and the overall distribution of the first differences of all steady-state sub-matrices of each electrolytic cell, specifically:

[0092] Determine the difference between the steady-state power and the optimal matching power of each electrolytic cell in each preset time period, and record it as the power modulation deviation of each electrolytic cell in each preset time period;

[0093] Arranging the power modulation deviations of all preset time periods of each electrolytic cell in ascending time order to form a power modulation sequence of each electrolytic cell;

[0094] The power modulation sequence of each electrolytic cell is used as the input of the trend test algorithm, and the product of the output trend test statistic and the mean value of all the power modulation deviations in the power modulation sequence is used as the significance degree of the long-term deviation increase of each electrolytic cell;

[0095] It should be noted that there are many commonly used trend test algorithms. In this embodiment, the Mann-Kendall trend test algorithm is used to obtain the trend test statistic. The implementer may also use other trend test algorithms, and this embodiment does not impose any special restrictions.

[0096] Further, the mean of the first difference of the clusters of all the steady-state sub-matrices of each electrolyzer is analyzed, and recorded as the power adaptation excellence of each electrolyzer;

[0097] The second difference of each electrolytic cell is determined based on the significant degree of long-term deviation increase and the excellent degree of power adaptation, and the second difference of each electrolytic cell is the result of the fusion of the significant degree of long-term deviation increase and the excellent degree of power adaptation of each electrolytic cell.

[0098] Preferably, as an embodiment of the present application, the second difference extraction flow chart is as follows: Figure 2 shown.

[0099] It should be understood that fusion refers to combining two or more indicators by adding or multiplying them together in order to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analysis methods. The implementer can choose according to the specific situation, and this embodiment does not impose any special restrictions.

[0100] Preferably, as an embodiment of the present application, the second difference of each electrolytic cell is the product of the significant degree of the long-term deviation increase and the excellent degree of power adaptation of each electrolytic cell.

[0101] Preferably, as another embodiment of the present application, the second difference of each electrolytic cell is an exponential function value with a natural constant as the base and the sum of the significant degree of long-term deviation increase and the degree of power adaptation excellence of each electrolytic cell as the independent variable.

[0102] The greater the significant degree of long-term deviation increase and the greater the degree of power adaptation excellence, it means that the electrolyzer is in a power underload state for a long time during the process of hydrogen production by electrolysis. The more accurate the power modulation deviation obtained by the electrolyzer and the smaller the error of the obtained long-term deviation significant degree increase, the greater the second difference. Conversely, the smaller the significant degree of long-term deviation increase and the smaller the degree of power adaptation excellence, it means that the electrolyzer is not in a power underload state during the process of hydrogen production by electrolysis. The greater the error of the obtained long-term deviation significant degree increase, the smaller the second difference.

[0103] Step S6: construct the pause power urgency of each electrolyzer according to the change of the actual power data in the actual power data sequence of each electrolyzer, and determine the microgrid power priority index of each electrolyzer in combination with the second difference.

[0104] During the electrolysis and hydrogen production process of the electrolyzer, the electrolyzer is affected by the power supply and the working conditions of the electrolyzer. For example, when the power supply is unstable and the operating parameters of the electrolyzer exceed the safety limit, the electrolyzer will be suspended and restarted. Due to the start-stop characteristics of the electrolyzer, when the electrolyzer starts, the temperature of the electrolyzer usually does not reach the temperature condition for producing hydrogen. At this time, the electrolyzer needs more power. In order to ensure the stable operation of the electrolyzer, it is urgent to give priority to additional power support from the microgrid.

[0105] Therefore, the pause power urgency of each electrolyzer is constructed by analyzing the change of the actual power data in the actual power data sequence of each electrolyzer, and the microgrid power priority index of each electrolyzer is determined in combination with the second difference, specifically:

[0106] All actual power data of each electrolytic cell are sorted in ascending time order to form an actual power data sequence of each electrolytic cell.

[0107] Taking the last data element with a value of 0 in the actual power sequence of each electrolytic cell as the center, the number of values ​​0 to the right of the center is used as the last pause duration of each electrolytic cell;

[0108] The time interval from the center to the last actual power data in the actual power sequence is taken as the last startup operation duration of each electrolytic cell;

[0109] The ratio of the last pause time to the last start-up time of each electrolyzer is calculated as the pause power urgency of each electrolyzer.

[0110] Preferably, as an embodiment of the present application, the microgrid power priority index extraction flow chart is as follows: Figure 3 shown.

[0111] Among them, the longer the last pause time, the longer the electrolyzer was last shut down, the more the electrolyzer temperature drops, resulting in more power required to restart the electrolyzer, and the greater the urgency of the pause power; the shorter the last startup operation time, the more the electrolyzer has just switched from a shutdown state to a restart state. In order to quickly meet the conditions for electrolytic hydrogen production, the electrolyzer needs more power support, and the greater the urgency of the pause power.

[0112] The microgrid power priority index of each electrolyzer is obtained based on the second difference and the suspension power urgency of each electrolyzer. The microgrid power priority index of each electrolyzer is a fusion result of the second difference and the suspension power urgency of each electrolyzer.

[0113] It should be understood that fusion refers to combining two or more indicators by adding or multiplying them together in order to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analysis methods. The implementer can choose according to the specific situation, and this embodiment does not impose any special restrictions.

[0114] Preferably, as an embodiment of the present application, the microgrid power priority index of each electrolyzer is the product of the second difference of each electrolyzer and the urgency of the suspension power.

[0115] Preferably, as another embodiment of the present application, the microgrid power priority index of each electrolyzer is an exponential function value with a natural constant as the base and the sum of the second difference and the pause power urgency of each electrolyzer as the independent variable.

[0116] Among them, the larger the second difference, the more the overall operating condition of the electrolyzer is in an underload state, the more it needs microgrid power support to achieve the best matching power, improve the maximum hydrogen production efficiency of the electrolyzer, and the larger the microgrid power priority index; the greater the urgency of the power suspension, the greater the impact of the last shutdown on the electrolyzer, the more it needs microgrid power support to achieve the conditions for electrolysis hydrogen production, and the larger the microgrid power priority index; conversely, the smaller the second difference, the more the overall operating condition of the electrolyzer is in an underload state, the more it needs microgrid power support to achieve the best matching power, reduce the minimum hydrogen production efficiency of the electrolyzer, and the smaller the microgrid power priority index; the smaller the urgency of the power suspension, the less the impact of the last shutdown on the electrolyzer, the less it needs microgrid power support to achieve the conditions for electrolysis hydrogen production, and the smaller the microgrid power priority index.

[0117] Step S7: Based on the second difference and the urgency of the pause power of the electrolyzer, the microgrid in the electrolyzer system is controlled in combination with the electrolyzer power modulation sequence of each electrolyzer.

[0118] The power modulation sequence of each electrolyzer is used as the input of the ARIMA autoregressive integrated moving average model algorithm, and the power adjustment prediction value of each electrolyzer is output. The negative power adjustment prediction value and the smaller the power adjustment prediction value, the more likely the electrolyzer is in an overload state, and the more power absorption of the microgrid is needed. All electrolyzers with negative power adjustment prediction values ​​are recorded as the power absorption group; all electrolyzers with positive power adjustment prediction values ​​are recorded as the power supply group.

[0119] Among them, the ARIMA autoregressive integrated moving average model algorithm is a well-known technology, and the specific process will not be repeated here.

[0120] When the electrolytic cell is in an overload state, the power of the DC power supply corresponding to each electrolytic cell is subtracted from the corresponding microgrid power priority index, thereby achieving regulation of the electrolytic cell overload state.

[0121] When the electrolyzer is in an underload state, the power of the DC power supply corresponding to the electrolyzer in the power absorption group is added to the corresponding microgrid power priority index to achieve regulation of the underload state of the electrolyzer.

[0122] At this point, the power regulation of the microgrid electrolyzer is completed based on the electrolyzer power modulation sequence of each electrolyzer and the microgrid power priority index.

[0123] Preferably, as an embodiment of the present application, the process diagram of regulating the operating power of the microgrid is as follows: Figure 4 shown.

[0124] Based on the same inventive concept as the above method, an embodiment of the present application also provides a control system for a microgrid, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned microgrid control methods are implemented.

[0125] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0126] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0127] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A microgrid control method, characterized in that: The method comprises the following steps: Obtain the actual power data of each electrolytic cell at all acquisition times, and the data of various power influencing factors of the electrolytic cell at all acquisition times, and form various power influencing sequences of each electrolytic cell; Various power impact sequences in each preset time period are arranged in rows to form a steady-state submatrix; Analyze the average distribution of actual power in each preset time period in each electrolytic cell to construct the steady-state power of each preset time period in each electrolytic cell; The electrolytic cell whose actual power data is equal to the preset optimal power is used as the electrolytic cell to be analyzed; The various power influence sequences of the electrolytic cells to be analyzed are combined into a historical state matrix of each electrolytic cell to be analyzed; Clustering each steady-state submatrix of each electrolytic cell and all the historical state matrices to obtain clusters, and recording the clusters that do not contain the steady-state submatrix as historical power matching clusters; Obtain the historical power data of each electrolytic cell, and in each preset time period of each electrolytic cell, obtain the best matching power of each preset time period of each electrolytic cell according to the distribution of the historical power data corresponding to all historical state matrices in the cluster where the steady-state submatrix is ​​located; Constructing a first difference of the cluster where each steady-state submatrix of each electrolytic cell is located according to the difference between the steady-state submatrix in the cluster where each steady-state submatrix is ​​located and all the historical state matrices, and the difference between all the historical state matrices in the historical power matching cluster; Constructing a second difference for each electrolytic cell according to the difference between the steady-state power and the best matching power in each preset time period of each electrolytic cell and the average distribution of the first differences of the clusters where all the steady-state sub-matrices of each electrolytic cell are located; The pause power urgency of each electrolyzer is constructed according to the change of the actual power data of each electrolyzer, and the microgrid power priority index of each electrolyzer is determined in combination with the second difference to control the power of the DC power supply of the microgrid.

2. A microgrid control method according to claim 1, characterized in that: The steady-state power of each electrolytic cell in each preset time period is the average value of the actual power data of each electrolytic cell in each preset time period.

3. A microgrid control method according to claim 1, characterized in that: The construction process of the optimal matching power of each preset time period of each electrolytic cell is as follows: In each preset time period of each electrolytic cell, determine the mean of all historical actual powers of each historical state matrix in the cluster where the steady-state submatrix is ​​located; In each preset time period of each electrolytic cell, the average value of the means of all historical state matrices in the cluster where the steady-state submatrix is ​​located is determined, and recorded as the best matching power in each preset time period of each electrolytic cell.

4. A microgrid control method according to claim 1, characterized in that: The construction process of the first difference of the clusters where each steady-state submatrix of each electrolytic cell is located is: Determine the mean of the distances between each steady-state submatrix of each electrolytic cell and all historical state matrices in the cluster to which it belongs, and record it as the first mean of the cluster to which each steady-state submatrix of each electrolytic cell belongs; Analyze the mean of the distances between all historical state matrices in each historical power matching cluster of each electrolytic cell, and record it as the second mean of each historical power matching cluster of each electrolytic cell; The first difference of the cluster where each steady-state sub-matrix of each electrolytic cell is located is the average of the differences between the first mean of the cluster where each steady-state sub-matrix of each electrolytic cell is located and all the second means.

5. A microgrid control method according to claim 1, characterized in that: The construction process of the second difference of each electrolytic cell is: Constructing the long-term deviation trend significance of each electrolytic cell according to the difference between the steady-state power and the best matching power of each electrolytic cell in each preset time period; Analyze the mean of the first difference of the clusters of all steady-state sub-matrices of each electrolyzer, and record it as the power adaptation excellence of each electrolyzer; The second difference of each electrolytic cell is determined based on the significant degree of long-term deviation increase and the excellent degree of power adaptation, and the second difference of each electrolytic cell is the result of the fusion of the significant degree of long-term deviation increase and the excellent degree of power adaptation of each electrolytic cell.

6. A microgrid control method as claimed in claim 5, characterized in that: The process of constructing the significant degree of long-term deviation increase of each electrolytic cell is as follows: Determine the difference between the steady-state power and the optimal matching power of each electrolytic cell in each preset time period, and record it as the power modulation deviation of each electrolytic cell in each preset time period; The power modulation deviations of all preset time periods of each electrolytic cell are arranged in ascending time order to form a power modulation sequence of each electrolytic cell; A trend test algorithm is used to obtain the trend test statistics of the power modulation sequence of each electrolytic cell; The degree of significant increase in long-term deviation trend of each electrolyzer is determined based on all the power modulation deviations and trend test statistics. The power adaptation excellence of each electrolyzer is the result of the fusion of the mean value of the power modulation deviations of all steady-state sub-matrices of each electrolyzer and the trend test statistics.

7. A microgrid control method as claimed in claim 1, characterized in that: The construction process of the pause power urgency of each electrolyzer is as follows: The actual power data of all acquisition moments in each electrolytic cell constitute the actual power sequence of each electrolytic cell; Taking the last data element with a value of 0 in the actual power sequence of each electrolytic cell as the center, the number of values ​​0 to the right of the center is counted as the last pause duration of each electrolytic cell; The time interval from the center to the last actual power data in the actual power sequence is taken as the last startup operation duration of each electrolytic cell; The pause power urgency of each electrolytic cell is the ratio of the last pause duration to the last start-up operation duration of each electrolytic cell.

8. A microgrid control method as claimed in claim 1, characterized in that: The microgrid power priority index of each electrolyzer is the result of the fusion of the second difference and the pause power urgency of each electrolyzer.

9. A microgrid control method as claimed in claim 6, characterized in that: The method for controlling the power of the DC power supply of the microgrid includes: The power modulation sequence of each electrolyzer is used as the input of the autoregressive integrated moving average model, and the power adjustment prediction value of each electrolyzer is output; When the power adjustment prediction value is negative, the power of the DC power supply corresponding to each electrolyzer is subtracted from the corresponding microgrid power priority index power adjustment prediction value; When the power adjustment prediction value is positive, the power of the DC power supply corresponding to the electrolyzer in the power absorption group plus all electrolyzers whose corresponding microgrid power priority index is negative are recorded as the power supply group.

10. A control system for a microgrid, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images