Hydrogen production and storage unit load distribution method, system and device and storage medium

Abstract

The invention discloses a hydrogen production and storage unit load distribution method, system and device and a storage medium, and relates to the technical field of new energy. The method comprises the following steps: establishing a new energy prediction model, predicting the first generation power of a new energy power generation system at a target moment, adjusting the load of a hydrogen production and storage unit according to the first generation power, then obtaining the unit energy consumption of each hydrogen production unit, comparing the unit energy consumption of each hydrogen production unit to determine the distribution priority of each hydrogen production unit, and determining the distribution priority of each hydrogen production unit according to the distribution priority. The lower the unit energy consumption of the hydrogen production unit is, the higher the distribution priority of the hydrogen production unit is, and when the load of the hydrogen production and storage unit is smaller than a first threshold value, load distribution is carried out on each hydrogen production unit according to the distribution priority of each hydrogen production unit to obtain the load of each hydrogen production unit. According to the scheme, the dependence of the hydrogen production unit on the power grid can be reduced, the operation cost is reduced, and the economic benefit of hydrogen production is improved.

Description

Technical Field

[0001] The present application relates to the field of new energy technology, and in particular to a method, system, device and storage medium for load distribution of a hydrogen production and storage unit. Background Art

[0002] Hydrogen energy, with its abundant sources, high mass energy density, environmentally friendly use and zero carbon emissions, is touted as the ideal energy source of the 21st century.

[0003] In traditional hydrogen production and storage unit systems, the traditional load regulation module adopts a simple power average distribution principle, so its economic and environmental performance is often low. In terms of wind power forecasting, the relevant technology has a certain lag in load regulation of hydrogen production stations, increasing the dependence on power grid support. Summary of the invention

[0004] The main purpose of the embodiments of the present disclosure is to propose a load distribution method, system, device and storage medium for a hydrogen production and storage unit, which can reduce the dependence of the hydrogen production unit on the power grid, reduce its operating costs, and improve the economic benefits of hydrogen production.

[0005] On the one hand, an embodiment of the present invention provides a method for distributing load of a hydrogen production and storage unit, comprising the following steps:

[0006] Establish new energy forecasting models;

[0007] Predicting the first power generation power of the new energy power generation system at a target time by using the new energy prediction model;

[0008] adjusting the load of the hydrogen production and storage unit according to the first power generation, wherein the hydrogen production and storage unit comprises a plurality of hydrogen production units;

[0009] Obtaining the unit energy consumption of each of the hydrogen production units;

[0010] Comparing the unit energy consumption of each of the hydrogen production units to determine the allocation priority of each of the hydrogen production units, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority;

[0011] When the load of the hydrogen production and storage unit is less than a first threshold, the load of each hydrogen production unit is distributed according to the distribution priority of each hydrogen production unit to obtain the load of each hydrogen production unit.

[0012] In some embodiments, the step of establishing a new energy prediction model comprises the following steps:

[0013] Get the training dataset;

[0014] Preprocessing the training data set;

[0015] Initializing the new energy prediction model based on a long short-term memory network framework;

[0016] The initialized new energy prediction model is trained using the training data set.

[0017] In some embodiments, the method of using the new energy prediction model to predict the first power generation power of the new energy power generation system at a target time includes the following steps:

[0018] Collecting current environmental data and the current power generation of the new energy power generation system; wherein the environmental data includes light intensity, wind speed, wind direction, temperature, and humidity;

[0019] Inputting the plurality of environmental data and the current power generation into the new energy prediction model for multiple times to perform prediction, and obtaining the second power generation of the plurality of new energy power generation systems at target times;

[0020] An average value of a plurality of the second generated powers is calculated, and the average value is used as the first generated power.

[0021] In some embodiments, adjusting the load of the hydrogen production and storage unit according to the first power generation power comprises the following steps:

[0022] Determining whether the first power generation power is greater than the penetration power of the hydrogen production and storage unit;

[0023] When the first power generation is greater than the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be less than or equal to the penetration power;

[0024] When the first power generation is less than or equal to the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be equal to the first power generation;

[0025] The step of controlling the load of the hydrogen production and storage unit to be equal to the first power generation power specifically includes:

[0026] Determining whether the first generated power is less than the current load of the hydrogen production and storage unit of the hydrogen production and storage unit;

[0027] When the first power generation power is less than the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is reduced to the first power generation power in advance;

[0028] When the first power generation power is greater than or equal to the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is increased to the first power generation power in advance.

[0029] In some embodiments, obtaining the unit energy consumption of each of the hydrogen production units comprises the following steps:

[0030] Monitoring the operating data of each electrolyzer; wherein the operating data includes the energy consumption and hydrogen production of the electrolyzer;

[0031] The unit energy consumption of the hydrogen production unit is calculated according to the energy consumption of the electrolyzer and the hydrogen production amount.

[0032] In some embodiments, the calculating the unit energy consumption of the hydrogen production unit according to the energy consumption of the electrolyzer and the hydrogen production amount comprises the following steps:

[0033] Adding the electrolytic cell energy consumption of all the electrolytic cells in the hydrogen production unit to obtain a total energy consumption;

[0034] Adding the hydrogen production of all the electrolyzers in the hydrogen production unit to obtain a total hydrogen production;

[0035] The unit energy consumption of the hydrogen production unit is obtained according to the ratio of the total energy consumption to the total hydrogen production amount.

[0036] In some embodiments, the hydrogen production and storage unit load distribution method further includes the following steps:

[0037] Obtaining the unit energy consumption of each of the electrolyzers in the hydrogen production unit;

[0038] Comparing the unit energy consumption of each electrolyzer in the hydrogen production unit to determine the allocation priority of each electrolyzer; wherein the electrolyzer with lower unit energy consumption has higher allocation priority;

[0039] When the load of the hydrogen production unit is less than a second threshold, the load of the electrolyzers is distributed according to the distribution priority of each electrolyzer to obtain the load of each electrolyzer in the hydrogen production unit.

[0040] On the other hand, an embodiment of the present invention provides a load distribution system for a hydrogen production and storage unit, comprising:

[0041] The first module is used to establish a new energy prediction model;

[0042] The second module is used for the new energy prediction model to predict the first power generation power of the new energy power generation system at a target time;

[0043] A third module is used to adjust the load of the hydrogen production and storage unit according to the first power generation, wherein the hydrogen production and storage unit includes a plurality of hydrogen production units;

[0044] The fourth module is used to obtain the unit energy consumption of each of the hydrogen production units;

[0045] A fifth module is used to compare the unit energy consumption of each of the hydrogen production units to determine the allocation priority of each of the hydrogen production units, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority;

[0046] The sixth module is used to distribute the load to each hydrogen production unit according to the allocation priority of each hydrogen production unit when the load of the hydrogen production and storage unit is less than the first threshold value, so as to obtain the load of each hydrogen production unit.

[0047] On the other hand, an embodiment of the present invention provides an electronic device, including:

[0048] at least one processor;

[0049] at least one memory for storing at least one program;

[0050] When the at least one program is executed by the at least one processor, the at least one processor implements the load distribution method for the hydrogen production and storage unit as described above.

[0051] On the other hand, an embodiment of the present invention proposes a computer-readable storage medium, which stores a program executable by a processor, characterized in that the program executable by the processor is used to implement the load distribution method of the hydrogen production and storage unit as described above when executed by the processor.

[0052] The embodiments of the present application include at least the following beneficial effects: a method, system, device and storage medium for distributing the load of a hydrogen production and storage unit. The scheme can accurately predict the change of the power generation power of the new energy power generation system through the new energy prediction model, and timely adjust the load of the hydrogen production and storage unit, thereby achieving the purpose of reducing the dependence of the hydrogen production unit on the power grid, reducing power loss, and reducing its operating costs. By detecting the parameters of each hydrogen production unit, the load distribution strategy is optimized, thereby improving the economic benefits of hydrogen production. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of a method for distributing load of a hydrogen production and storage unit provided in an embodiment of the present invention;

[0054] Figure 2 It is a schematic diagram of the overall architecture of the system provided by an embodiment of the present invention;

[0055] Figure 3 It is a long short-term memory network workflow diagram provided by an embodiment of the present invention;

[0056] Figure 4 is a new energy prediction model training flow chart provided by an embodiment of the present invention;

[0057] Figure 5 It is a fitting diagram of a power generation prediction curve and an actual curve provided by an embodiment of the present invention;

[0058] Figure 6 is a training set loss and validation set loss curve diagram provided by an embodiment of the present invention;

[0059] Figure 7 It is a fitting diagram of a wind power data prediction curve and an actual curve of a certain wind farm provided by an embodiment of the present invention;

[0060] Figure 8 is a comparison diagram of energy consumption before and after load adjustment provided by an embodiment of the present invention;

[0061] Fig. 9 is a schematic diagram of the structure of a hydrogen production station provided by an embodiment of the present invention;

[0062] Fig.10 It is a schematic diagram of the relationship between the first threshold scheduling load and the hydrogen production unit load provided in an embodiment of the present invention;

[0063] Fig.11 is a load comparison diagram of each hydrogen production unit provided in an embodiment of the present invention;

[0064] Fig.12 is a load distribution flow chart of a hydrogen production and storage unit provided in an embodiment of the present invention;

[0065] Fig.13 It is a schematic diagram of the hardware structure of the electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0067] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0069] The embodiment of the present application provides a method for distributing the load of a hydrogen production and storage unit. The method for distributing the load of a hydrogen production and storage unit in the embodiment of the present application can be applied to a terminal, can be applied to a terminal, can be applied to a server, and can also be software running in a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0070] Reference Figure 1 As shown, the load distribution method of the hydrogen production and storage unit in the embodiment of the present invention includes but is not limited to steps S100 to S600.

[0071] Step S100, establishing a new energy prediction model;

[0072] Step S200, using a new energy prediction model to predict a first power generation power of a new energy power generation system at a target time;

[0073] Step S300, adjusting the load of the hydrogen production and storage unit according to the first power generation power, wherein the hydrogen production and storage unit includes a plurality of hydrogen production units;

[0074] Step S400, obtaining the unit energy consumption of each hydrogen production unit; wherein the hydrogen production and storage unit includes a plurality of hydrogen production units;

[0075] Step S500, comparing the unit energy consumption of each hydrogen production unit to determine the allocation priority of each hydrogen production unit, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority;

[0076] Step S600: When the load of the hydrogen production and storage unit is less than a first threshold, the load of each hydrogen production unit is distributed according to the distribution priority of each hydrogen production unit to obtain the load of each hydrogen production unit.

[0077] In some embodiments of step S100, the new energy prediction model may include but is not limited to a wind power generation prediction model and a photovoltaic power generation prediction model. Wind power can be described by two important indicators, namely wind speed and wind direction, while one of the most important influencing factors of photovoltaic power generation is light intensity.

[0078] Reference Figure 2 , Figure 2It is a schematic diagram of the overall system architecture of this embodiment. The new energy power generation system includes a wind power generation system, a photovoltaic power generation system, etc. to provide power for the hydrogen production unit. The prediction module predicts the power generation power based on the environment of the new energy power generation system and the current power generation situation. The prediction module transmits the prediction data to the voltage stabilization control center. The voltage stabilization control center controls the load of the hydrogen production station based on the prediction data. The load distribution module further distributes the load according to the load of the hydrogen production station and the unit energy consumption of each hydrogen production unit. Finally, the hydrogen production unit produces hydrogen.

[0079] In some embodiments, the new energy prediction model can use a radial basis function neural network. A neural network that uses a radial basis function as an activation mechanism is a radial basis function neural network form. This type of network regards the output as a weighted value of the input variable processed by the radial basis function and summed with the neuron parameters.

[0080] The output of the radial basis function neural network is obtained by summing up all the inputs with the corresponding radial basis functions and neural unit parameters according to the weights. The core idea of ​​the radial basis function neural network is to introduce radial basis functions as hidden layer nodes, and use this method to carefully divide different categories of data. The hidden layer constructed in this way suggests a "basis" space, whose function is to map the vector of the input pattern from the original low-dimensional space to a higher-dimensional space, in order to transform the situation that cannot be distinguished linearly in the low-dimensional space into a problem that can be linearly distinguished in the high-dimensional space.

[0081] In some embodiments, the new energy prediction model can use a recurrent neural network, which is designed to overcome the limitation of feedforward neural networks that ignore the correlation of data sequences. The feedforward network only generates output based on the current input and lacks reference to historical data information. In real applications, many problems involve sequence data, such as the periodic changes in outdoor temperature with the change of seasons, and language understanding also depends on context. This type of information usually has temporal continuity, that is, the network output at a certain moment is not only related to the current input, but also affected by the previous output. Feedforward neural networks cannot effectively handle this continuity because they lack the ability to transfer past information to future moments, and the core of recurrent neural networks is to have memory functions and use these memories to make judgments.

[0082] In some embodiments, the new energy prediction model can also use a long short-term memory neural network, which is based on a recurrent neural network and adds a processor to determine whether specific data is still valuable. The long short-term memory network is an extremely effective algorithm for solving wind forecasting, and this algorithm has very good compatibility.

[0083] In some embodiments, wind force prediction is mainly the prediction of wind speed changes. Usually, wind speed presents different data under different environmental conditions and different geographical locations. Its influencing factors include terrain, humidity, temperature, etc. Therefore, wind speed has a strong randomness, but the wind speed can still show certain regularity and correlation in a short time.

[0084] Furthermore, the expression of the mathematical model used to describe the wind speed is shown in formula (1), and the wind speed can be fitted.

[0085]

[0086] Among them, k is the shape coefficient, c is the scale coefficient, reflecting the annual average wind speed in the described area, and x represents the wind speed.

[0087] Furthermore, since there are many factors related to wind speed, and each factor has a different weight for the impact on wind speed, the weight of the same factor on wind speed at different times is also different. If each factor is treated equally, it will inevitably lead to distortion of the calculation results. In order to measure the correlation between different attributes and the changes in the wind speed data set, the embodiment of the present application analyzes the impact of each factor on wind speed.

[0088] The mutual information of two discrete random variables X and Y can be defined as:

[0089]

[0090] As shown in formula (2), p(x·y) represents the joint probability density of discrete random variables; p(x) and p(y) represent the marginal probability densities of two variables X and Y respectively; χ and γ are the value spaces of X and Y respectively. I(X;Y) represents the amount of information contained in the random variable X about another random variable Y, and can also measure the uncertainty reduction of one random variable under the condition of the existence of another random variable. The mutual information value can be used to represent the influence of each related factor on the wind speed.

[0091] In some embodiments, step S100 may include but is not limited to the following steps:

[0092] Step S110, obtaining a training data set;

[0093] Step S120, preprocessing the training data set;

[0094] Step S130, initializing a new energy prediction model based on a long short-term memory network framework;

[0095] Step S140: training the initialized new energy prediction model using the training data set.

[0096] In this embodiment, a long short-term memory network is selected as the basis of the new energy prediction model. The long short-term memory network (LSTM) is a time recurrent neural network, which is specially designed to solve the long-term dependency problem of general recurrent neural networks. On the basis of the recurrent neural network, the long short-term memory network adds a processor, which is used to determine whether specific data is still valuable. The processor consists of an input gate, a forget gate, and an output gate. When a piece of information is input, the long short-term memory network will determine whether it is useful according to the rules.

[0097] like Figure 3 As shown, Figure 3 It is the workflow diagram of the long short-term memory network. The definition expressions of each parameter in the diagram are shown in equations (3) to (5).

[0098] I t =σ(X t W xi +H t-1 W hi +b i ); (3)

[0099] F t =σ(X t W xf +H t-1 W hf +b f ); (4)

[0100] O t =σ(X t W xo +H t-1 W ho +b o ); (5)

[0101] Among them I t is the input gate, F t For the forget gate, O t is the output gate; W xi , W xf , W xo is the weight parameter; b i , b f , b o is the bias parameter. In order to alleviate the gradient vanishing problem and better capture the long-distance dependencies in the sequence, the same element-wise multiplication technique as before is used to obtain formula (6).

[0102]

[0103] Among them C t For memory unit; is the new data of the candidate hidden state; C t-1 is the old memory unit; it is overwritten after the calculation is completed. In this system, the values ​​of the input gate, forget gate and output gate are included, and the values ​​of these three gates are all in the range of (0, 1). If it is, it is permanently memorized, and if it is, it is forgotten, which meets the algorithm requirements of the embodiment of this application.

[0104] In one example, see Figure 4 The overall process of training a new energy prediction model based on a long short-term memory network in the embodiment of the present application is as follows.

[0105] First, select the location of the created algorithm folder and enter the command prompt panel.

[0106] Enter the command jupyter notebook, select the created code file, and enter PyCharm with the Anaconda environment configured.

[0107] More than 100,000 sets of data on actual wind speed in a certain place are used as training sets. The input function reads the file data and deletes the missing values ​​in the data set to facilitate the reference of the data set in subsequent operations.

[0108] Then the data set is translated into Chinese, and histograms, box plots, and scatter plot matrices are drawn for the relationship between time and power generation, and the icons are optimized. By observing the table, the distribution of data features is determined to prepare for the subsequent selection of feature values.

[0109] Create a table object about power.

[0110] Use pd.to_datetime() to merge the year, month, day, hour and minute into a timestamp. According to the above data characteristics, this data feature selection is determined by the function features = data[[′windspeedat100m(m / s)′]].values.astype(float).

[0111] Normalize the data to facilitate learning of the neural convolutional network. Prepare the training data and build a training data list. Based on the above data features, it can be deduced that the appropriate step size in this simulation is 3, so the step size is set to 3. Use the len() function to divide the data set into a training set and a test set. After the division, start building the LSTM model. First build the first layer model, the first layer LSTM, return the sequence for the next layer, and add Dropout for regularization. Then build the second layer LSTM and add Dropout for regularization again. Then build the output layer, compile the model, train the model, and test the model in sequence. Then train the LSTM based on the training set.

[0112] Use the np.sqr function and the mean_absolute_error function to calculate the root mean square error (RMSE) and the mean absolute error (MAE), respectively.

[0113] After the data learning is completed, in order to facilitate the visualization of the data, the predicted value and the true value are reverse normalized to facilitate the visualization of the results, such as Figure 5 shown.

[0114] The prediction effect of the neural convolutional network is evaluated. First, the model is trained through model.fit(), and then a line graph of the training set loss curve and the validation set loss curve is drawn based on the training history, such as Figure 6 shown.

[0115] The loss rates corresponding to the training set loss curve and the validation set loss curve are both small, which means that the predicted value in the algorithm has a good fitting effect with the actual value, indicating that the new energy prediction model has strong prediction capabilities and meets the requirements of the design of the embodiment of this application.

[0116] In some embodiments, the prediction verification is performed again based on the wind power data of a wind farm in another time period, and the data set of the time period is used as the verification set to perform wind power prediction, and the result is as follows: Figure 7 As shown in the figure, the final predicted wind speed and actual wind speed also meet the requirements.

[0117] In this embodiment, when the wind speed data is actually used to train the neural network, the data needs to be normalized to address the problems of the model's slow data processing speed and different feature dimensions. The idea is to use the data corresponding to the training set that has not been expanded in time steps as the standard and normalize other data. The specific formula is as shown in formula (7).

[0118]

[0119] Among them, X is the actual value of the sample feature, X′ is the normalized value of the sample feature, and X min is the minimum value of the sample feature, X max is the maximum value of the sample feature. Data normalization can eliminate the impact of dimension through normalization and also speed up the model to process data.

[0120] In some embodiments of step S200, new energy power generation includes hydroelectric power generation, solar power generation, wind power generation, etc. Hydroelectric power generation uses water with potential energy located in high places such as rivers and lakes to flow to low places, converts the potential energy contained therein into kinetic energy of a turbine, and then uses the turbine as the motive force to drive the generator to generate electricity. Wind power generation uses wind power to drive the windmill blades to rotate, and then increases the speed of rotation through a speed increaser to drive the generator to generate electricity. Solar power generation includes solar photovoltaic power generation and solar thermal power generation. Solar photovoltaic power generation is a power generation method that directly converts light energy into electrical energy without going through a thermal process, including photovoltaic power generation, photochemical power generation, photoinduction power generation, and photobiological power generation. Photovoltaic power generation is a direct power generation method that uses solar-grade semiconductor electronic devices to effectively absorb solar radiation energy and convert it into electrical energy. It is the mainstream of solar power generation today.

[0121] Furthermore, the new energy power generation system can be a wind power generation system or a photovoltaic power generation system, or a system combining wind power generation and photovoltaic power generation. Accordingly, if the new energy power generation system is a wind power generation system, the new energy prediction model is a wind power generation prediction model; if the new energy power generation system is a photovoltaic power generation system, the new energy prediction model is a photovoltaic power generation prediction model; if the new energy power generation system is a system combining wind power generation and photovoltaic power generation, the new energy prediction model is a model that can predict the sum of wind power generation and photovoltaic power generation. Specifically, the wind power generation prediction model and the photovoltaic power generation prediction model can be used to predict the wind power generation and photovoltaic power generation, and then the total can be calculated.

[0122] In some embodiments, step S200 may include but is not limited to the following steps:

[0123] Step S210, obtaining current environment data;

[0124] Step S220, obtaining the current power generation power of the new energy power generation system;

[0125] Step S230, using the new energy prediction model to predict according to the environmental data and the current power generation, to obtain the second power generation of the new energy power generation system at the target time;

[0126] Step S240, using the new energy prediction model to predict the second power generation at the target time multiple times; wherein each prediction is performed at intervals of a certain period of time;

[0127] Step S250, calculating an average value of the second generated power at multiple target moments as the first generated power.

[0128] In some embodiments, a new energy prediction model can be used to predict the second power generation power of the new energy power generation system at a target time, and the second power generation power can be directly used as the first power generation power.

[0129] In other embodiments, since the neural network has a certain error in each prediction, in order to avoid the result deviation caused by the randomness of a single prediction result, the prediction is repeated 5 times and the average value is taken as the final prediction result, which is then output to the voltage stabilization control center. At the same time, in order to perform better data quantitative analysis on the prediction results, this paper uses the prediction error rate err as an analysis and evaluation index to describe the accuracy of the prediction results. The calculation method of the prediction error rate err is shown in formula (8).

[0130]

[0131] Among them, Pr(t) is the actual value at time t, Pp(t) is the predicted value at time t, and Cap is the installed capacity in kW.

[0132] In this embodiment, wind power generation is predicted, and more than 100,000 sets of data on actual wind speed in a certain place in 2019 are used as training sets for neural network learning. The installed capacity is set to 400KW, and the step length is 5min to predict the output results. If the followability of the prediction results is very good, when the weather changes extremely and the wind power output changes greatly, the prediction error percentage can always be kept within 1%, and the err error rate is less than 0.2%, then this prediction method can be considered feasible and effective. Some neural network training sets are shown in Table 1:

[0133] Table 1 Part of the neural network training set

[0134]

[0135] In some embodiments of step S300, data such as the current wind speed and the current light intensity are fed back to the voltage stabilization control center to generate corresponding data. At the same time, in the system, the new energy power prediction system rolls up every 5 minutes, and uploads a 2-hour power prediction curve each time. The system intends to set the feedforward control time to 45 minutes. If the predicted power of renewable energy at any point in the next 45 minutes exceeds the maximum load regulation capacity of the hydrogen production station, data comparison will be performed after 5 minutes. If the predicted power of renewable energy at this time point is still measured to exceed the maximum load regulation capacity of the hydrogen production station, the dispatching system will issue a load reduction instruction to the hydrogen production station. The maximum load regulation capacity of the hydrogen production station is set to 1% / min.

[0136] In some embodiments, step S300 may include but is not limited to the following steps:

[0137] Step S310, determining whether the first power generation power is greater than the penetration power of the hydrogen production and storage unit;

[0138] Step S320, when the first power generation is greater than the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be less than or equal to the penetration power;

[0139] Step S330, when the first power generation power is less than or equal to the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be equal to the first power generation power;

[0140] The above-mentioned control of the load of the hydrogen storage unit to be equal to the first power generation power specifically includes:

[0141] Step S340, determining whether the first power generation power is less than the current load of the hydrogen production and storage unit;

[0142] Step S350, when the first power generation power is less than the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is reduced to the first power generation power in advance;

[0143] Step S360: When the first power generation power is greater than or equal to the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is increased to the first power generation power in advance.

[0144] In some embodiments of steps S310 to S320, a long short-term memory network framework is first built, and then trained using a training set, and the feasibility of the algorithm is determined after simulation. After being applied to the hydrogen production and storage unit, the wind energy prediction algorithm located in the electric energy prediction unit will transmit the predicted wind power data to the voltage stabilization control unit, and the load of each electrolyzer is actively adjusted through the voltage stabilization control unit, so as to maintain the power of the hydrogen production and storage unit below the penetration power while improving the economic benefits of the power grid, so as to prevent the occurrence of penetration phenomenon that seriously endangers the health of the equipment.

[0145] In some embodiments of step S330, refer to Figure 8 Since the load regulation capacity of hydrogen electrolysis and other downstream devices is limited and usually lower than the output change rate of renewable energy power generation systems such as wind power and photovoltaic power generation, the load regulation speed of the hydrogen production station usually lags behind the change rate of renewable energy, which leads to the problem that the difference between the load of the hydrogen production device and the output of renewable energy can only be supplemented by the power grid. Through the wind prediction model, the model uses a neural network to learn the wind power patterns in the current area and predict the changes in the wind power in the area in advance. If the wind power change value exceeds the load regulation capacity, the system will reduce the load of the hydrogen production unit in advance to reduce its dependence on the power grid and improve its economy.

[0146] In some embodiments of step S400, the hydrogen production and storage unit in the hydrogen production station is divided into several independent hydrogen production units, each of which contains multiple (e.g., 5) electrolyzers. Different hydrogen production units can be controlled independently, thereby improving the scheduling flexibility of the hydrogen production station. The hydrogen production station is divided into several hydrogen production units, each of which can be controlled independently, thereby improving the flexibility of load regulation of the hydrogen production station. The schematic diagram of the hydrogen production station structure is shown in FIG. Fig. 9This embodiment obtains the unit energy consumption of each hydrogen production unit, which facilitates the subsequent reasonable allocation of the hydrogen production unit load based on the unit energy consumption.

[0147] In some embodiments, step S400 may include but is not limited to the following steps:

[0148] Step S410, monitoring the operating data of each electrolytic cell;

[0149] Step S420, calculating the unit energy consumption of the hydrogen production unit according to the energy consumption of the electrolyzer and the hydrogen production amount.

[0150] In this embodiment, an equipment monitoring module is integrated into the hydrogen production station optimization scheduling system to monitor the voltage, current, energy consumption, output and other data of each electrolyzer, and the energy consumption indicators of the electrolyzer and hydrogen production unit are obtained and sorted through real-time data.

[0151] In some embodiments, step S420 may include but is not limited to the following steps:

[0152] Step S421, adding the electrolytic cell energy consumption of all electrolytic cells in the hydrogen production unit to obtain the total energy consumption;

[0153] Step S422, adding the hydrogen production of all electrolyzers in the hydrogen production unit to obtain the total hydrogen production;

[0154] Step S423, obtaining the unit energy consumption of the hydrogen production unit according to the ratio of the total energy consumption to the total hydrogen production amount.

[0155] In the embodiment of step S420, the unit energy consumption calculation expression of each electrolytic cell is shown in formula (9). The unit energy consumption calculation expression of each electrolytic cell is shown in formula (10).

[0156]

[0157]

[0158] Among them, ∑ energy consumption is the sum of energy consumption of all electrolyzers in the hydrogen production unit, and ∑ hydrogen production is the sum of hydrogen production of all electrolyzers in the hydrogen production unit.

[0159] In some embodiments of step S500, due to manufacturing errors, equipment aging and other reasons, the performance of each electrolyzer is not the same. Under the same energy efficiency, the hydrogen production of each electrolyzer is also different, resulting in a certain amount of waste. An equipment monitoring module is integrated into the hydrogen production station optimization and scheduling system to monitor the voltage, current, energy consumption, output and other data of each electrolyzer. The energy consumption indicators of the electrolyzer and the hydrogen production unit are obtained through real-time data. The lower the unit energy consumption, the higher the allocation priority of the hydrogen production unit. The higher the allocation priority, the higher the ranking of the hydrogen production unit. The lower the load of the hydrogen production unit, the load will be allocated to the hydrogen production unit with the higher ranking. This load distribution strategy can increase the hydrogen production of the hydrogen production station when it is running under low load and obtain more benefits.

[0160] In some embodiments of step S600, the unit energy consumption of each hydrogen production unit is determined by the load distribution module and the unit energy consumption of each hydrogen production unit is determined based on the weighted sum of the unit energy consumption of each electrolyzer. When the load of the water electrolysis hydrogen production device is greater than or equal to the first threshold, the load is evenly distributed to each hydrogen production unit; and when the load of the water electrolysis hydrogen production device is less than the first threshold, the load is preferentially distributed to the hydrogen production unit with lower unit energy consumption.

[0161] In some embodiments of step S600, refer to Fig.10 , set a limit load (i.e., the first threshold value). When the overall load of the hydrogen production station is greater than, the dispatch instruction is evenly distributed to each hydrogen production unit; when the overall load of the hydrogen production station is less than or equal to, the load is preferentially allocated to the hydrogen production unit with a high energy consumption index ranking. The relationship between the load of the hydrogen production unit and the overall load of the hydrogen production station according to an embodiment of the present invention. The horizontal axis expression corresponding to point A is shown in formula (11), and the horizontal axis expression corresponding to point B is shown in formula (12).

[0162] load 平均负荷 ×(i-1);(11)

[0163] ld 平均负荷 ×i; (12)

[0164] Where load1 is the first threshold, and the value is the ranking of hydrogen production units according to energy consumption indicators. It can be obtained from experiments that the overall energy consumption is reduced by about 5% to 7% compared with the traditional system.

[0165] For example, the energy consumption indicators of the four hydrogen production units are ranked, and the load scheduling limit is set to 0.8. When the load of the hydrogen production station is greater than this limit, the load is evenly distributed among each hydrogen production unit. When the load of the hydrogen production station is less than this limit, in the case of the overall scheduling instruction of the hydrogen production station, the hydrogen production unit with better energy consumption performance is prioritized according to the energy consumption ranking of the four hydrogen production units, and the load scheduling limit is set to 0.8. Fig.11, shows the load curve of each hydrogen production unit in the hydrogen production station according to the example of the present invention, and only the load change curve for the first 50 hours is drawn here. The energy consumption indicators of hydrogen production units 1 to 4 decrease in sequence. It can be seen from the curve that unit 1 has the best performance and the longest operating time, followed by unit 2, and so on. The average loads of units 1 to 4 are 0.49, 0.30, 0.20, and 0.13, respectively.

[0166] In some embodiments, the hydrogen production station is equipped with 4 hydrogen production units, each hydrogen production unit includes 5 electrolyzers, and the hydrogen production station is equipped with a total of 20 electrolyzers. The minimum stable operating load of each electrolyzer is 40%. When a traditional control system is used, the minimum load of the hydrogen production station is also 40%. After adopting the hydrogen production control system of the embodiment of the present application, the minimum load of a single module is 40%, and the minimum load of the hydrogen production station can be reduced to one quarter of the total load, that is, 10%.

[0167] In some embodiments, after step S600, the load distribution method of the hydrogen production and storage unit in the embodiment of the present application further includes but is not limited to the following steps:

[0168] Step S610, obtaining the unit energy consumption of each electrolyzer in the hydrogen production unit;

[0169] Step S620, comparing the unit energy consumption of each electrolyzer in the hydrogen production unit to determine the allocation priority of each electrolyzer;

[0170] Step S630: When the load of the hydrogen production unit is less than the second threshold, the load of the electrolyzers is distributed according to the distribution priority of each electrolyzer to obtain the load of each electrolyzer in the hydrogen production unit.

[0171] In some embodiments, the load distribution method of the electrolyzer is to distribute the load of each electrolyzer in the hydrogen production unit based on the unit energy consumption of each electrolyzer. The lower the unit energy consumption of the electrolyzer, the higher the allocation priority of the electrolyzer. When the load of the hydrogen production unit is greater than or equal to the second threshold, the load is evenly distributed to each electrolyzer; and when the load of the hydrogen production unit is less than the second threshold, the load is preferentially distributed to the electrolyzer with lower unit energy consumption.

[0172] Reference Fig.12 , the load distribution flow chart is as follows Fig.12 The unit energy consumption of each hydrogen production unit is determined by the load distribution module and the unit energy consumption of each hydrogen production unit is determined based on the weighted sum of the unit energy consumption of each electrolyzer.

[0173] When the load of the water electrolysis hydrogen production device is greater than or equal to the first threshold, the load is evenly distributed to each hydrogen production unit; and when the load of the water electrolysis hydrogen production device is less than the first threshold, the load is preferentially distributed to the hydrogen production unit with lower unit energy consumption.

[0174] Then, the load of each electrolyzer in the hydrogen production unit is allocated based on the unit energy consumption of each electrolyzer. That is, when the load of the hydrogen production unit is greater than or equal to the second threshold, the load is evenly allocated to each electrolyzer; and when the load of the hydrogen production unit is less than the second threshold, the load is preferentially allocated to the electrolyzer with lower unit energy consumption.

[0175] In this embodiment, in order to have better wind energy capture performance, this paper adopts MPPT technology in the wind turbine. The principle of MPPT is: when the wind speed is constant, the wind turbine rotor speed is maintained at the optimal speed corresponding to the optimal blade tip speed ratio to achieve maximum wind energy capture. This operating point is called the maximum power point. By connecting the maximum power points corresponding to different wind speeds, the optimal torque curve of the wind turbine can be obtained through the relationship between power and torque.

[0176] It is known that the resolution of existing turbulence frequency characterization indicators is insufficient. Therefore, a large amount of sample data is needed to show the statistical relationship between them and the optimal control parameters, which is very time-consuming and has a negative impact on the optimization results of the control parameters. Therefore, the concept of equivalent turbulent wind to sinusoidal wind realizes the extraction and index description of the main turbulent frequency components closely related to the optimal control parameters. Only the turbulent components whose frequencies match the sensitive frequency band of the wind turbine dynamic response (the frequency of turbulence matches the sensitive stage of the wind turbine dynamic response) can significantly change the mechanism of wind energy capture results. According to the criterion of wind energy equality, this paper proposes the theory of equivalent turbulent wind to sinusoidal wind, which realizes the extraction and single-value index description of turbulent frequency components sensitive to wind turbine dynamics. Experiments have shown that due to the focus on matching the main turbulent frequency components in the sensitive frequency band of the wind turbine dynamic response, the equivalent sinusoidal frequency is closely related to the optimal control parameters. The optimal parameter model constructed based on it estimates the optimal set value more accurately, so that the method in this paper can obtain higher wind energy capture efficiency.

[0177] In actual operation, the wind turbine rotor with large inertia moment cannot track the rapidly changing wind speed and is in the tracking process in most cases instead of working at the steady-state operating point. Therefore, the design perspective of wind turbine MPPT is extended from the steady-state state to the dynamic state.

[0178] Furthermore, the offline modeling and online application components are combined. First, the optimal control parameters corresponding to the wind speed sequence with different turbulence characteristics are searched offline to form samples. Then, based on the sample data, alternative models such as neural networks and polynomial response surfaces are used to describe the functional relationship between the optimal control parameters and the turbulence characteristics, and the optimal parameter model is constructed. In the online application, when the turbulent wind speed information is obtained, the control parameters can be optimized according to the optimal parameter model. The OPM method avoids the iterative search process and has strong adaptability to changes in turbulence characteristics. Under the premise of a more accurate model, a higher wind energy capture efficiency can be maintained.

[0179] Regarding equivalent sinusoidal wind, the characterization of turbulence frequency should focus on the main turbulence frequency components that match the sensitive frequency bands of wind turbine tracking dynamic response. Therefore, a method for extracting the main turbulence frequency components based on equivalent sinusoidal wind is proposed. The equivalent mechanism is that the total energy contained in a wind speed series depends on its mean wind speed and standard deviation. The energy that a wind turbine can capture from a wind speed series is the comprehensive result of the response of the wind turbine MPPT to different turbulence frequency components. Considering that wind turbines can only respond to turbulence frequency components that sensitively affect tracking dynamics and contribute to the vast majority of wind energy capture, the frequency equivalence process is based on the extraction and characterization of the main frequency components contained in the wind turbine MPPT for equal energy capture. The optimal control parameters are formed by applying the parameters of equivalent sinusoidal wind to the offline search, while predicting the future wind characteristics, and continuously iteratively upgrading the model through online application, so as to achieve the purpose of improving the efficiency of wind energy capture.

[0180] In some embodiments, hydrogen storage technology refers to methods and processes for safely storing hydrogen in various forms. There are three main hydrogen storage technologies, including gas storage, liquid storage, and solid state storage.

[0181] Gaseous hydrogen storage can be divided into two categories: compressed hydrogen storage and underground hydrogen storage. In compressed hydrogen storage, hydrogen is stored in a gaseous state under high pressure. Compressed hydrogen is stored in high-pressure cylinders or large storage containers designed to withstand high pressures. Underground gas storage refers to the storage of hydrogen in underground geological formations or structures, such as salt caverns, depleted oil and gas reservoirs, or aquifers.

[0182] Liquid hydrogen storage is the practice of storing substances in a liquid state, usually in specialized tanks designed to maintain the temperature and pressure conditions required. Storage and transportation of liquid hydrogen requires that it be kept cold to maintain its liquid state. Liquid hydrogen storage requires the use of specialized, insulated containers or tanks to store and transport liquid hydrogen. These containers are designed to minimize heat transfer and maintain the low temperatures required for liquid hydrogen.

[0183] Solid-state storage refers to the storage of hydrogen in solid materials, usually through chemical or physical bonding. This method provides high volumetric energy density and improved safety compared to gas or liquid storage methods. Solid-state storage includes a variety of materials, including metal hydrides, complex hydrides, and adsorbent materials. Metal hydrides are compounds formed when hydrogen interacts with metals or metal alloys to form stable reversible bonds. Complex hydrides are also a class of hydrogen storage materials, consisting of metals or metallic elements with hydrogen and other elements, such as aluminum, boron, or nitrogen. These materials can absorb hydride ions (H-) and generally have higher gravimetric storage capacities than metal hydrides. Adsorbent materials store hydrogen through a process called physical adsorption, where hydrogen molecules are physically attracted to the surface of a solid material.

[0184] A comparison of various hydrogen solid-state storage technologies is shown in Table 2.

[0185] In this embodiment, solid-state storage technology is used to store hydrogen energy, among which composite hydrides have the best storage performance.

[0186] The embodiment of the present invention further provides a load distribution system for a hydrogen production and storage unit, comprising:

[0187] The first module is used to establish a new energy prediction model;

[0188] The second module is used for the new energy prediction model to predict the first power generation power of the new energy power generation system at the target time;

[0189] A third module is used to adjust the load of the hydrogen production and storage unit according to the first power generation power, wherein the hydrogen production and storage unit includes a plurality of hydrogen production units;

[0190] The fourth module is used to obtain the unit energy consumption of each hydrogen production unit;

[0191] The fifth module is used to compare the unit energy consumption of each hydrogen production unit to determine the allocation priority of each hydrogen production unit, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority;

[0192] The sixth module is used to distribute the load to each hydrogen production unit according to the allocation priority of each hydrogen production unit when the load of the hydrogen production and storage unit is less than the first threshold value, so as to obtain the load of each hydrogen production unit.

[0193] The load distribution system for the hydrogen production and storage unit of the embodiment of the present application establishes a new energy prediction model to predict the first power generation power of the new energy power generation system at a target time, adjusts the load of the hydrogen production and storage unit according to the first power generation power, and then determines the allocation priority of each hydrogen production unit. According to the load distribution strategy, the load of each hydrogen production unit is distributed, and the load of each hydrogen production unit is obtained, which can reduce the dependence of the hydrogen production unit on the power grid, reduce its operating cost, and improve the economic benefit of hydrogen production.

[0194] It should be noted that the load distribution system for the hydrogen production and storage unit in the embodiment of the present application corresponds to the aforementioned load distribution method for the hydrogen production and storage unit. For specific steps of load distribution of the hydrogen production and storage unit, please refer to the aforementioned load distribution method for the hydrogen production and storage unit, which will not be described in detail here.

[0195] Table 2 Comparison of various hydrogen solid-state storage technologies

[0196]

[0197] Combine the following Fig.13 The electronic device according to the embodiment of the present application is introduced in detail.

[0198] like Fig.13 , Fig.13 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0199] The processor 1600 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present disclosure;

[0200] The memory 1700 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1700 can store an operating system and other application programs. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1700, and the processor 1600 is used to call and execute the load distribution method of the hydrogen production and storage unit in the embodiment of the present disclosure;

[0201] Input / output interface 1800, used to implement information input and output;

[0202] The communication interface 1900 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0203] Bus 2000 , which transmits information between various components of the device (e.g., processor 1600 , memory 1700 , input / output interface 1800 , and communication interface 1900 );

[0204] The processor 1600 , the memory 1700 , the input / output interface 1800 , and the communication interface 1900 are connected to each other in communication within the device via the bus 2000 .

[0205] The embodiment of the present disclosure also provides a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned hydrogen production and storage unit load distribution method.

[0206] The storage medium of the embodiment of the present application is used to execute the above-mentioned load allocation method for the hydrogen production and storage unit. By establishing a new energy prediction model, the first power generation power of the new energy power generation system at the target time is predicted, the load of the hydrogen production and storage unit is adjusted according to the first power generation power, and then the allocation priority of each hydrogen production unit is determined. The load of each hydrogen production unit is allocated according to the load allocation strategy to obtain the load of each hydrogen production unit, which can reduce the dependence of the hydrogen production unit on the power grid, reduce its operating cost, and improve the economic benefit of hydrogen production.

[0207] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0208] The embodiments described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0209] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present disclosure, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0210] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0211] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0212] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0213] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0214] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0215] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0216] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0217] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store programs.

[0218] The preferred embodiments of the present disclosure are described above with reference to the accompanying drawings, but the scope of the rights of the present disclosure is not limited thereto. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the present disclosure should be within the scope of the rights of the present disclosure.

Claims

1. A method for distributing load of a hydrogen production and storage unit, characterized in that: The method comprises: Establish new energy forecasting models; Predicting the first power generation power of the new energy power generation system at a target time by using the new energy prediction model; adjusting the load of the hydrogen production and storage unit according to the first power generation, wherein the hydrogen production and storage unit comprises a plurality of hydrogen production units; Obtaining the unit energy consumption of each of the hydrogen production units; Comparing the unit energy consumption of each of the hydrogen production units to determine the allocation priority of each of the hydrogen production units, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority; When the load of the hydrogen production and storage unit is less than a first threshold, the load of each hydrogen production unit is distributed according to the distribution priority of each hydrogen production unit to obtain the load of each hydrogen production unit.

2. The method for distributing load of a hydrogen production and storage unit according to claim 1, characterized in that: The step of establishing a new energy prediction model comprises the following steps: Get the training dataset; Preprocessing the training data set; Initializing the new energy prediction model based on a long short-term memory network framework; The initialized new energy prediction model is trained using the training data set.

3. The method for distributing load of a hydrogen production and storage unit according to claim 1, characterized in that: The method of using the new energy prediction model to predict the first power generation power of the new energy power generation system at a target time comprises the following steps: Collecting current environmental data and the current power generation of the new energy power generation system; wherein the environmental data includes light intensity, wind speed, wind direction, temperature, and humidity; Inputting the plurality of environmental data and the current power generation into the new energy prediction model for multiple times to perform prediction, and obtaining the second power generation of the plurality of new energy power generation systems at target times; An average value of a plurality of the second generated powers is calculated, and the average value is used as the first generated power.

4. The method for distributing load of a hydrogen production and storage unit according to claim 1, characterized in that: The step of adjusting the load of the hydrogen production and storage unit according to the first power generation power comprises the following steps: Determining whether the first power generation power is greater than the penetration power of the hydrogen production and storage unit; When the first power generation is greater than the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be less than or equal to the penetration power; When the first power generation is less than or equal to the penetration power of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is controlled to be equal to the first power generation; The step of controlling the load of the hydrogen production and storage unit to be equal to the first power generation power specifically includes: Determining whether the first generated power is less than the current load of the hydrogen production and storage unit of the hydrogen production and storage unit; When the first power generation power is less than the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is reduced to the first power generation power in advance; When the first power generation power is greater than or equal to the current load of the hydrogen production and storage unit, the load of the hydrogen production and storage unit is increased to the first power generation power in advance.

5. The method for distributing load of a hydrogen production and storage unit according to claim 1, characterized in that: The obtaining of the unit energy consumption of each hydrogen production unit comprises the following steps: Monitoring the operating data of each electrolyzer; wherein the operating data includes the energy consumption and hydrogen production of the electrolyzer; The unit energy consumption of the hydrogen production unit is calculated according to the energy consumption of the electrolyzer and the hydrogen production amount.

6. The method for distributing load of a hydrogen production and storage unit according to claim 5, characterized in that: The step of calculating the unit energy consumption of the hydrogen production unit according to the energy consumption of the electrolyzer and the hydrogen production amount comprises the following steps: Adding the electrolytic cell energy consumption of all the electrolytic cells in the hydrogen production unit to obtain a total energy consumption; Adding the hydrogen production of all the electrolyzers in the hydrogen production unit to obtain a total hydrogen production; The unit energy consumption of the hydrogen production unit is obtained according to the ratio of the total energy consumption to the total hydrogen production amount.

7. The method for distributing load of a hydrogen production and storage unit according to claim 1, characterized in that: Each of the hydrogen production units includes a plurality of electrolytic cells. After the step of distributing the loads of the hydrogen production units according to the allocation priorities of the hydrogen production units to obtain the loads of the hydrogen production units, the method for distributing the loads of the hydrogen production and storage units further includes the following steps: Obtaining the unit energy consumption of each of the electrolyzers in the hydrogen production unit; Comparing the unit energy consumption of each electrolyzer in the hydrogen production unit to determine the allocation priority of each electrolyzer; wherein the electrolyzer with lower unit energy consumption has higher allocation priority; When the load of the hydrogen production unit is less than a second threshold, the load of the electrolyzers is distributed according to the distribution priority of each electrolyzer to obtain the load of each electrolyzer in the hydrogen production unit.

8. A load distribution system for a hydrogen production and storage unit, characterized in that: include: The first module is used to establish a new energy prediction model; The second module is used for the new energy prediction model to predict the first power generation power of the new energy power generation system at a target time; A third module is used to adjust the load of the hydrogen production and storage unit according to the first power generation, wherein the hydrogen production and storage unit includes a plurality of hydrogen production units; The fourth module is used to obtain the unit energy consumption of each of the hydrogen production units; A fifth module is used to compare the unit energy consumption of each of the hydrogen production units to determine the allocation priority of each of the hydrogen production units, wherein the hydrogen production unit with lower unit energy consumption has higher allocation priority; The sixth module is used to distribute the load to each hydrogen production unit according to the allocation priority of each hydrogen production unit when the load of the hydrogen production and storage unit is less than the first threshold value, so as to obtain the load of each hydrogen production unit.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the load distribution method for the hydrogen production and storage unit as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the load distribution method of the hydrogen production and storage unit as described in any one of claims 1 to 7 when executed by the processor.

Images