A peak shaving and valley filling system for hydrogen fuel cells

Abstract

The present invention discloses a peak shaving and valley filling system for hydrogen fuel cells, and relates to the relevant field of electric energy storage systems. The system comprises: by receiving the peak shaving and valley filling instruction of the power grid, based on the embedded power grid load alignment rule, combined with the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution are retrieved. Based on the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution, the peak shaving and valley filling characteristic prediction results of the power grid and the power grid valley filling characteristic prediction results are obtained. The power grid is peak shaving regulated based on the hydrogen fuel cell module and the power grid peak shaving characteristic prediction results. The power grid is valley filling regulated based on the hydrogen fuel cell module and the power grid valley filling characteristic prediction results. The fixed parameter prediction and scheduling method used in the peak shaving and valley filling system in the prior art is solved, and the poor regulation and prediction accuracy affects the peak shaving and valley filling effect, resulting in waste of electric energy.

Description

Technical Field

[0001] The present application relates to the field of electric energy storage systems, and in particular to a peak shaving and valley filling system for hydrogen fuel cells. Background Art

[0002] With the continuous growth of energy demand and the increasing attention to environmental protection, the stability and efficiency of the power grid have become increasingly important. Traditional power systems mainly rely on fossil fuels for power generation, which is not only seriously polluting, but also has the problem of low energy utilization. In order to improve the operating efficiency of the power grid, peak shaving and valley filling technology has gradually become a hot topic in the power industry. Peak shaving and valley filling refers to transferring part of the load during the peak period of electricity demand to the valley period, thereby balancing the load of the power grid, reducing the waste of electricity resources, and improving the stability of the power grid operation. Traditional peak shaving and valley filling systems mostly use fixed parameter prediction and fixed parameter scheduling methods, which are difficult to cope with the dynamic changes of the power grid load, resulting in poor prediction accuracy in actual operation and affecting the control accuracy.

[0003] Therefore, the fixed parameter prediction and scheduling method used in the peak shaving and valley filling system in the prior art has poor regulation and prediction accuracy, which affects the effect of peak shaving and valley filling, leading to the technical problem of energy waste. Summary of the invention

[0004] This application provides a peak-shaving and valley-filling system for hydrogen fuel cells, which solves the technical problem that the fixed parameter prediction and scheduling method used in the peak-shaving and valley-filling system in the prior art has poor regulation and prediction accuracy, thus affecting the peak-shaving and valley-filling effect and causing waste of electric energy. It achieves the technical effect of accurately predicting and efficiently regulating the power grid load, improving the stability and efficiency of the power grid operation, and reducing the operating cost.

[0005] The present application provides a peak shaving and valley filling system for hydrogen fuel cells, the system comprising: a power grid interaction module, the power grid interaction module is used to receive a peak shaving and valley filling instruction of the power grid, wherein the peak shaving and valley filling instruction includes a predetermined future time zone. A power grid load information management module, the power grid load information management module retrieves a first power grid load characteristic distribution and a second power grid load characteristic distribution based on an embedded power grid load registration rule and in combination with the predetermined future time zone. A peak shaving and valley filling characteristic prediction module, the peak shaving and valley filling characteristic prediction module is used to predict the peak shaving and valley filling characteristics of the power grid based on the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution, and obtain a power grid peak shaving characteristic prediction result and a power grid valley filling characteristic prediction result. A hydrogen fuel cell module, the hydrogen fuel cell module comprising a plurality of hydrogen fuel cells. A power grid peak shaving control module, the power grid peak shaving control module is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module and the power grid peak shaving characteristic prediction result. A power grid valley filling control module, the power grid valley filling control module is used to perform valley filling control on the power grid based on the hydrogen fuel cell module and the power grid valley filling characteristic prediction result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The present application provides a hydrogen fuel cell peak shaving and valley filling system, comprising: a power grid interaction module, the power grid interaction module is used to receive a power grid peak shaving and valley filling instruction, wherein the peak shaving and valley filling instruction includes a predetermined future time zone. A power grid load information management module, the power grid load information management module retrieves a first power grid load characteristic distribution and a second power grid load characteristic distribution based on an embedded power grid load registration rule and in combination with the predetermined future time zone. A peak shaving and valley filling characteristic prediction module, the peak shaving and valley filling characteristic prediction module is used to predict the peak shaving and valley filling characteristics of the power grid based on the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution, and obtain a power grid peak shaving characteristic prediction result and a power grid valley filling characteristic prediction result. A hydrogen fuel cell module, the hydrogen fuel cell module includes a plurality of hydrogen fuel cells. A power grid peak shaving control module, the power grid peak shaving control module is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module and the power grid peak shaving characteristic prediction result. A power grid valley filling control module, the power grid valley filling control module is used to perform valley filling control on the power grid based on the hydrogen fuel cell module and the power grid valley filling characteristic prediction result. The fixed parameter prediction and dispatching method used in the existing peak shaving and valley filling system has been solved, and the poor regulation and prediction accuracy has affected the peak shaving and valley filling effect, resulting in the waste of electric energy. The technical effect of accurately predicting and efficiently regulating the grid load has been achieved, improving the stability and efficiency of the grid operation, and reducing the operating cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A schematic diagram of the structure of a peak shaving and valley filling system for a hydrogen fuel cell provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a process flow of peak shaving regulation of a power grid by a power grid peak shaving regulation module in a peak shaving and valley filling system of a hydrogen fuel cell in the present application.

[0011] Explanation of the reference numerals: power grid interaction module 11 , power grid load information management module 12 , peak shaving and valley filling characteristic prediction module 13 , hydrogen fuel cell module 14 , power grid peak shaving control module 15 , power grid valley filling control module 16 . DETAILED DESCRIPTION

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0013] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0014] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, systems, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0015] The present application embodiment provides a peak shaving and valley filling system for a hydrogen fuel cell, such as Figure 1 As shown, the system comprises:

[0016] The power grid interaction module 11 is used to receive a peak-shaving and valley-filling instruction of the power grid, wherein the peak-shaving and valley-filling instruction includes a predetermined future time zone.

[0017] The power grid load information management module 12 retrieves the first power grid load characteristic distribution and the second power grid load characteristic distribution based on the embedded power grid load registration rule and in combination with the predetermined future time zone.

[0018] The peak shaving and valley filling feature prediction module 13 is used to predict the peak shaving and valley filling features of the power grid based on the predetermined future time zone, the first power grid load feature distribution and the second power grid load feature distribution, and obtain power grid peak shaving feature prediction results and power grid valley filling feature prediction results.

[0019] The peak shaving and valley filling instruction of the power grid is received through the power grid interaction module 11, wherein the peak shaving and valley filling instruction includes a predetermined future time zone, and the predetermined future time zone is the execution time interval corresponding to the peak shaving and valley filling instruction. Subsequently, the first power grid load characteristic distribution and the second power grid load characteristic distribution are retrieved through the power grid load information management module 12 based on the embedded power grid load alignment rule and in combination with the predetermined future time zone. After obtaining the power grid load characteristic distribution, the peak shaving and valley filling characteristic prediction module 13 predicts the peak shaving and valley filling characteristics of the power grid based on the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution, and obtains the power grid peak shaving characteristic prediction results and the power grid valley filling characteristic prediction results, thereby obtaining the power grid load peak scale characteristics and the power grid load peak time characteristics, the power grid load valley value scale characteristics and the power grid load valley value time characteristics.

[0020] Furthermore, the power grid load information management module 12 is also used to: construct a first load-associated time zone based on the predetermined future time zone and the time zone neighborhood characteristic constraints.

[0021] Based on the first load-associated time zone, according to the power grid load information management module 12, the first power grid load peak and valley sample set is loaded, and data cleaning is performed according to the first power grid load peak and valley sample set to obtain the first power grid load characteristic distribution.

[0022] Based on the predetermined future time zone and the time zone attribute characteristic constraint, a second load-associated time zone is established.

[0023] Based on the second load associated time zone, according to the power grid load information management module 12, the second power grid load peak and valley sample set is loaded, and data cleaning is performed according to the second power grid load peak and valley sample set to obtain the second power grid load characteristic distribution.

[0024] The power grid load information management module 12 retrieves the first power grid load characteristic distribution and the second power grid load characteristic distribution based on the embedded power grid load matching rules and in combination with the predetermined future time zone, including: the power grid load matching rules include time zone neighborhood characteristic constraints and time zone attribute characteristic constraints, wherein the time zone neighborhood characteristic constraints are pre-set data time extraction lengths adjacent to the predetermined future time zone. If the predetermined future time zone is July-August and the time zone neighborhood characteristic constraints are 3 months, then the corresponding data obtained under the time zone neighborhood characteristic constraints are data from March to June. The time zone attribute characteristic constraints are historical time interval constraints that are the same as the predetermined future time zone. Based on the predetermined future time zone and the time zone neighborhood characteristic constraints, a first load-associated time zone is constructed, that is, according to the predetermined future time zone, the first load-associated time zone obtained under the time zone neighborhood characteristic constraints is obtained. The first load-associated time zone is a time interval adjacent to the predetermined future time zone. Based on the first load-associated time zone, according to the power grid load information management module 12, extract the first power grid load peak and valley sample set recorded in the historical data in the power grid load information management module 12, the first power grid load peak and valley sample set is a sample consisting of the peak parameters and valley parameters of the power grid load in each day, the corresponding occurrence time, and the generated power shortage and power surplus data, and perform data cleaning according to the first power grid load peak and valley sample set, remove duplicate data and abnormal deviation data therein, and obtain the first power grid load characteristic distribution. Further, based on the predetermined future time zone and the time zone attribute feature constraint, establish a second load-associated time zone, the second load-associated time zone is the same time interval as the predetermined future time zone in history, if the predetermined future time zone is the whole month of January, then the second load-associated time zone can be the whole month of January of the same period last year. Based on the second load-associated time zone, according to the power grid load information management module 12, load the second power grid load peak and valley sample set, and perform data cleaning according to the second power grid load peak and valley sample set, and obtain the second power grid load characteristic distribution as a sample consisting of the peak parameters and valley parameters of the power grid load in each day, the corresponding occurrence time, and the generated power shortage and power surplus data.

[0025] Furthermore, the peak shaving and valley filling characteristic prediction module 13 is also used to construct a first power grid load peak characteristic curve and a first power grid load valley characteristic curve according to the first power grid load characteristic distribution.

[0026] According to the second power grid load characteristic distribution, a second power grid load peak characteristic curve and a second power grid load valley characteristic curve are constructed.

[0027] Based on the peak shaving characteristic prediction component, the power grid peak shaving characteristic prediction result is generated according to the predetermined future time zone, the first power grid load peak characteristic curve and the second power grid load peak characteristic curve.

[0028] Based on the valley filling characteristic prediction component, the power grid valley filling characteristic prediction result is generated according to the predetermined future time zone, the first power grid load valley value characteristic curve and the second power grid load valley value characteristic curve.

[0029] The peak shaving and valley filling characteristic prediction module 13 includes: a peak shaving characteristic prediction component and a valley filling characteristic prediction component. Based on the first power grid load characteristic distribution, a first power grid load peak characteristic curve and a first power grid load valley characteristic curve are constructed, wherein the coordinates of the first power grid load peak characteristic curve and the first power grid load valley characteristic curve are time, power gap / surplus, peak parameter / load valley value, respectively. According to the second power grid load characteristic distribution, a second power grid load peak characteristic curve and a second power grid load valley characteristic curve are constructed, and the construction method of the second power grid load peak characteristic curve and the second power grid load valley characteristic curve is consistent with the construction method of the first power grid load peak characteristic curve and the first power grid load valley characteristic curve, and there is only a difference in construction data. Based on the peak shaving characteristic prediction component, according to the predetermined future time zone, the first power grid load peak characteristic curve and the second power grid load peak characteristic curve, the power grid peak shaving characteristic prediction result is generated, and the power grid peak shaving characteristic prediction result includes the specific parameters of the power grid load valley value and peak value of each day in the predetermined future time zone, as well as the time of occurrence and the corresponding power surplus and gap. Based on the valley filling characteristic prediction component, the power grid valley filling characteristic prediction result is generated according to the predetermined future time zone, the first power grid load valley value characteristic curve and the second power grid load valley value characteristic curve.

[0030] Furthermore, the peak shaving and valley filling feature prediction module 13 is also used for: the peak shaving feature prediction component includes a pre-constructed first grid load peak feature prediction model and a second grid load peak feature prediction model, wherein the first grid load peak feature prediction model and the second grid load peak feature prediction model both include multi-dimensional grid load peak feature prediction indicators, and the multi-dimensional grid load peak feature prediction indicators include grid load peak scale characteristics and grid load peak time characteristics.

[0031] The predetermined future time zone and the first power grid load peak characteristic curve are input into the first power grid load peak characteristic prediction model to obtain a first power grid load peak characteristic prediction result.

[0032] The predetermined future time zone and the second power grid load peak characteristic curve are input into the second power grid load peak characteristic prediction model to obtain a second power grid load peak characteristic prediction result.

[0033] The first power grid load peak characteristic prediction result and the second power grid load peak characteristic prediction result are integrated to obtain the power grid peak shaving characteristic prediction result.

[0034] Based on the peak shaving feature prediction component, the power grid peak shaving feature prediction result is generated according to the predetermined future time zone, the first power grid load peak feature curve and the second power grid load peak feature curve, including: the peak shaving feature prediction component includes a pre-built first power grid load peak feature prediction model and a second power grid load peak feature prediction model, wherein the first power grid load peak feature prediction model is used to predict the first power grid load peak feature curve and obtain the prediction result, and the second power grid load peak feature prediction model is used to predict the second power grid load peak feature curve and obtain the prediction result. In addition, the first power grid load peak feature prediction model and the second power grid load peak feature prediction model both include multi-dimensional power grid load peak feature prediction indicators, and the multi-dimensional power grid load peak feature prediction indicators include power grid load peak scale characteristics and power grid load peak time characteristics, the power grid load peak scale characteristics include specific parameters of the power grid load peak and the corresponding power shortage amount, and the power grid load peak time characteristics are the specific time when the peak occurs.

[0035] Among them, the first model for predicting the peak characteristics of power grid load is constructed based on the neural network model, and the historical peak shaving record data is obtained to obtain the recorded sample first power grid peak valley characteristic curve, the sample scheduled future time zone, and the marked sample power grid load peak scale characteristics and sample power grid load peak time characteristics. The above sample data is used as training data to supervise the neural network model until the power grid load peak scale characteristics and power grid load peak time characteristics output by the model meet the preset accuracy rate. The model training is completed to obtain the first model for predicting the peak characteristics of power grid load. The construction method of the second model for predicting the peak characteristics of power grid load is the same as the construction method of the first model for predicting the peak characteristics of power grid load, and there is only a difference in training data. The training data of the second model for predicting the peak characteristics of power grid load is the second power grid load peak characteristic curve and the corresponding training parameters.

[0036] Based on the predetermined future time zone and the first power grid load peak characteristic curve, a first model is predicted according to the power grid load peak characteristic, and a first power grid load peak characteristic prediction result is output. Based on the predetermined future time zone and the second power grid load peak characteristic curve, a second model is predicted according to the power grid load peak characteristic, and a second power grid load peak characteristic prediction result is output. The first power grid load peak characteristic prediction result and the second power grid load peak characteristic prediction result are integrated to obtain the power grid peak shaving characteristic prediction result, that is, the power grid load peak scale characteristics in the first power grid load peak characteristic prediction result and the second power grid load peak characteristic prediction result are averaged to obtain the average calculation result. The earliest prediction time of the predicted power grid load peak time characteristic is obtained, and the earliest prediction time is obtained as the prediction time in the power grid peak shaving characteristic prediction result.

[0037] Furthermore, the peak shaving and valley filling feature prediction module 13 is also used for: the valley filling feature prediction component includes a pre-constructed first grid load valley feature prediction model and a second grid load valley feature prediction model, wherein the first grid load valley feature prediction model and the second grid load valley feature prediction model both include multi-dimensional grid load valley feature prediction indicators, and the multi-dimensional grid load valley feature prediction indicators include grid load valley scale characteristics and grid load valley time characteristics.

[0038] Based on the predetermined future time zone and the first power grid load valley characteristic curve, a first model is predicted according to the power grid load valley characteristic, and a first power grid load valley characteristic prediction result is output.

[0039] Based on the predetermined future time zone and the second power grid load valley characteristic curve, a second model is predicted according to the power grid load valley characteristic, and a second power grid load valley characteristic prediction result is output.

[0040] The first power grid load valley characteristic prediction result and the second power grid load valley characteristic prediction result are integrated to obtain the power grid valley filling characteristic prediction result.

[0041] Based on the valley filling feature prediction component, according to the predetermined future time zone, the first grid load valley characteristic curve and the second grid load valley characteristic curve, the grid valley filling feature prediction result is generated, including: the valley filling feature prediction component includes a pre-constructed grid load valley characteristic prediction first model and a grid load valley characteristic prediction second model, wherein the grid load valley characteristic prediction first model is used to predict the first grid load valley characteristic curve and obtain the prediction result, and the grid load valley characteristic prediction second model is used to predict the second grid load valley characteristic curve and obtain the prediction result. In addition, the grid load valley characteristic prediction first model and the grid load valley characteristic prediction second model both include multi-dimensional grid load valley characteristic prediction indicators, and the multi-dimensional grid load valley characteristic prediction indicators include grid load valley scale characteristics and grid load valley time characteristics, the grid load valley scale characteristics include specific parameters of the grid load valley and the corresponding power surplus, and the grid load valley time characteristics are the specific time when the valley occurs.

[0042] Based on the predetermined future time zone and the first grid load valley characteristic curve, a first model is predicted according to the grid load valley characteristic, and a prediction result of the first grid load valley characteristic is output. Based on the predetermined future time zone and the second grid load valley characteristic curve, a second model is predicted according to the grid load valley characteristic, and a prediction result of the second grid load valley characteristic is output. The first grid load valley characteristic prediction result and the second grid load valley characteristic prediction result are integrated to obtain the grid valley filling characteristic prediction result, that is, the grid load valley scale characteristics in the first grid load valley characteristic prediction result and the second grid load valley characteristic prediction result are averaged to obtain the average calculation result. The predicted grid load valley time characteristics are obtained by interval acquisition, and the grid load valley time characteristics output by the two are constructed as a time interval. If the grid load valley time characteristics of the two are the same, the same grid load valley time characteristics are directly output.

[0043] The hydrogen fuel cell module 14 includes a plurality of hydrogen fuel cells.

[0044] A power grid peak shaving control module 15 is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module 14 and the power grid peak shaving characteristic prediction result.

[0045] A grid valley-filling control module 16 is used to control the grid valley-filling based on the hydrogen fuel cell module 14 and the grid valley-filling characteristic prediction result.

[0046] The system also includes: a hydrogen fuel cell module 14, wherein the hydrogen fuel cell module 14 includes a plurality of hydrogen fuel cells for storing and releasing electric energy. A power grid peak shaving control module 15, wherein the power grid peak shaving control module 15 is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module 14 and the prediction result of the power grid peak shaving characteristics, that is, to discharge the hydrogen fuel cell according to the prediction result of the power grid peak shaving characteristics, to supplement the power grid with electric energy, and to complete the power grid peak shaving control. A power grid valley filling control module 16, wherein the power grid valley filling control module 16 is used to perform valley filling control on the power grid based on the hydrogen fuel cell module 14 and the prediction result of the power grid valley filling characteristics, that is, to charge the hydrogen fuel cell according to the prediction result of the power grid valley filling characteristics, to supplement the hydrogen fuel cell with electric energy, and to complete the power grid valley filling control. This ensures that the power distribution is maintained in a relatively stable range to optimize the energy utilization efficiency of the power grid. The hydrogen fuel cell module 14 also includes a battery monitoring and early warning component, which is used to monitor and warn the real-time temperature and charge and discharge capacity of the multiple hydrogen fuel cells. When the temperature exceeds the safety value and / or the charge and discharge capacity exceeds the corresponding charge and discharge capacity of the battery energy storage auxiliary decision and the hydrogen power supply auxiliary decision, an early warning is issued for the corresponding hydrogen battery to ensure the safety of charging and discharging during the battery peak shaving and valley filling process.

[0047] Further, such as Figure 2 As shown, the power grid peak shaving control module 15 is also used to collect real-time energy storage information of the multiple hydrogen fuel cells to obtain a battery energy storage characteristic data set.

[0048] A hydrogen energy power supply auxiliary decision model is constructed, and the hydrogen energy power supply auxiliary decision model is embedded into the power grid peak shaving control module 15.

[0049] The battery energy storage characteristic data set and the power grid peak shaving characteristic prediction result are input into the hydrogen energy power supply auxiliary decision model to obtain the hydrogen energy power supply auxiliary decision.

[0050] The power grid is peak-cut and regulated according to the hydrogen energy power supply auxiliary decision and the hydrogen fuel cell module 14 .

[0051] Collect the real-time energy storage information of the multiple hydrogen fuel cells, the real-time energy storage information is the electric energy storage capacity of the battery, and obtain the battery energy storage feature data set. Subsequently, build a hydrogen power supply auxiliary decision model, and embed the hydrogen power supply auxiliary decision model into the power grid peak shaving control module 15. The hydrogen power supply auxiliary decision model is used to obtain the power supply difference according to the battery energy storage feature data set and the power grid peak shaving feature prediction results. The power supply difference is the difference between the battery energy storage characteristics and the power grid peak shaving feature power, and the final remaining power of the hydrogen fuel cell is obtained based on the difference and the number of hydrogen fuel cells. The battery discharge capacity is distributed based on the final remaining power and electric energy storage capacity of each hydrogen fuel cell. The discharge target is balanced discharge to ensure the balance of the power of each battery after the discharge is completed. Finally, the battery energy storage feature data set and the power grid peak shaving feature prediction results are input into the hydrogen power supply auxiliary decision model to obtain a hydrogen power supply auxiliary decision, which includes the specific discharge capacity parameters of each battery. The power grid is peak shaving controlled based on the hydrogen power supply auxiliary decision and the hydrogen fuel cell module 14.

[0052] Furthermore, the grid valley filling control module 16 is also used to collect energy storage gap information of the multiple hydrogen fuel cells to obtain a battery energy storage gap feature data set.

[0053] A battery energy storage auxiliary decision-making model is constructed, and the battery energy storage auxiliary decision-making model is embedded into the grid valley filling control module 16.

[0054] Based on the grid valley filling feature prediction result and the battery energy storage gap feature data set, a battery energy storage auxiliary decision is output according to the battery energy storage auxiliary decision model.

[0055] The grid is regulated to fill valleys according to the battery energy storage auxiliary decision and the hydrogen fuel cell module 14 .

[0056] Collect the real-time energy storage gap information of the multiple hydrogen fuel cells, the real-time energy storage information is the remaining electrical energy storage capacity of the battery, and obtain the battery gap feature data set. Subsequently, build a hydrogen energy storage auxiliary decision model, and embed the hydrogen energy storage auxiliary decision model into the power grid valley filling regulation module 16. The hydrogen energy storage auxiliary decision model is used to distribute the charge amount of each battery according to the battery gap feature data set and the power grid valley filling feature prediction results. The charging target is balanced charging to ensure the balance of the charge of each battery after charging is completed. Finally, the battery valley filling feature data set and the power grid valley filling feature prediction results are input into the hydrogen energy storage auxiliary decision model to obtain a hydrogen energy storage auxiliary decision, and the specific charge amount parameters of each battery are included in the hydrogen power supply storage decision. The power grid is valley-filled and regulated based on the hydrogen energy storage auxiliary decision and the hydrogen fuel cell module 14.

[0057] The embodiment of the present application receives a peak shaving and valley filling instruction of the power grid, wherein the peak shaving and valley filling instruction includes a predetermined future time zone. Based on the embedded power grid load alignment rule, in combination with the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution are retrieved. Based on the predetermined future time zone, the first power grid load characteristic distribution and the second power grid load characteristic distribution, the power grid is predicted to have peak shaving and valley filling characteristics, and the prediction results of the power grid peak shaving characteristics and the prediction results of the power grid valley filling characteristics are obtained. The power grid is peak shaving regulated based on the hydrogen fuel cell module and the prediction results of the power grid peak shaving characteristics. The power grid is valley filling regulated based on the hydrogen fuel cell module and the prediction results of the power grid valley filling characteristics. The fixed parameter prediction and scheduling method used in the peak shaving and valley filling system in the prior art is solved, and the poor regulation and prediction accuracy affects the effect of peak shaving and valley filling, resulting in waste of electric energy. The technical effect of accurately predicting and efficiently regulating the power grid load, improving the stability and efficiency of the power grid operation, and reducing the operating cost is achieved.

[0058] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does 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.

Claims

1. A peak shaving and valley filling system for a hydrogen fuel cell, characterized in that: The system comprises: A power grid interaction module, the power grid interaction module is used to receive a peak shaving and valley filling instruction of the power grid, wherein the peak shaving and valley filling instruction includes a predetermined future time zone; A power grid load information management module, wherein the power grid load information management module retrieves the first power grid load characteristic distribution and the second power grid load characteristic distribution based on an embedded power grid load registration rule and in combination with the predetermined future time zone; A peak shaving and valley filling feature prediction module, the peak shaving and valley filling feature prediction module is used to predict the peak shaving and valley filling features of the power grid based on the predetermined future time zone, the first power grid load feature distribution and the second power grid load feature distribution, and obtain a power grid peak shaving feature prediction result and a power grid valley filling feature prediction result; A hydrogen fuel cell module, wherein the hydrogen fuel cell module comprises a plurality of hydrogen fuel cells; A power grid peak shaving control module, the power grid peak shaving control module is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module and the power grid peak shaving characteristic prediction result; A power grid valley-filling control module is used to perform valley-filling control on the power grid based on the hydrogen fuel cell module and the power grid valley-filling characteristic prediction result.

2. The system according to claim 1, characterized in that The power grid load information management module retrieves the first power grid load characteristic distribution and the second power grid load characteristic distribution based on the embedded power grid load registration rule and in combination with the predetermined future time zone, including: The grid load matching rules include time zone neighborhood feature constraints and time zone attribute feature constraints; constructing a first load-associated time zone based on the predetermined future time zone and the time zone neighborhood characteristic constraint; Based on the first load-associated time zone, according to the power grid load information management module, loading the first power grid load peak and valley sample set, and performing data cleaning according to the first power grid load peak and valley sample set to obtain a first power grid load characteristic distribution; Establishing a second load-associated time zone based on the predetermined future time zone and the time zone attribute characteristic constraint; Based on the second load associated time zone and according to the power grid load information management module, the second power grid load peak and valley sample set is loaded, and data cleaning is performed according to the second power grid load peak and valley sample set to obtain the second power grid load characteristic distribution.

3. The system according to claim 1, characterized in that The peak shaving and valley filling feature prediction module is used to predict the peak shaving and valley filling features of the power grid based on the predetermined future time zone, the first power grid load feature distribution, and the second power grid load feature distribution, and obtain a power grid peak shaving feature prediction result and a power grid valley filling feature prediction result, including: The peak shaving and valley filling feature prediction module includes a peak shaving feature prediction component and a valley filling feature prediction component; Constructing a first power grid load peak characteristic curve and a first power grid load valley characteristic curve according to the first power grid load characteristic distribution; Constructing a second power grid load peak characteristic curve and a second power grid load valley characteristic curve according to the second power grid load characteristic distribution; Based on the peak shaving characteristic prediction component, generating the power grid peak shaving characteristic prediction result according to the predetermined future time zone, the first power grid load peak characteristic curve and the second power grid load peak characteristic curve; Based on the valley filling characteristic prediction component, the power grid valley filling characteristic prediction result is generated according to the predetermined future time zone, the first power grid load valley value characteristic curve and the second power grid load valley value characteristic curve.

4. The system according to claim 3, characterized in that Based on the peak shaving characteristic prediction component, generating the power grid peak shaving characteristic prediction result according to the predetermined future time zone, the first power grid load peak characteristic curve and the second power grid load peak characteristic curve, including: The peak shaving feature prediction component includes a pre-constructed first grid load peak feature prediction model and a second grid load peak feature prediction model, wherein the first grid load peak feature prediction model and the second grid load peak feature prediction model both include multi-dimensional grid load peak feature prediction indicators, and the multi-dimensional grid load peak feature prediction indicators include grid load peak scale characteristics and grid load peak time characteristics; Inputting the predetermined future time zone and the first power grid load peak characteristic curve into the first power grid load peak characteristic prediction model to obtain a first power grid load peak characteristic prediction result; Inputting the predetermined future time zone and the second power grid load peak characteristic curve into the second power grid load peak characteristic prediction model to obtain a second power grid load peak characteristic prediction result; The first power grid load peak characteristic prediction result and the second power grid load peak characteristic prediction result are integrated to obtain the power grid peak shaving characteristic prediction result.

5. The system according to claim 3, characterized in that Based on the valley filling characteristic prediction component, generating the grid valley filling characteristic prediction result according to the predetermined future time zone, the first grid load valley value characteristic curve and the second grid load valley value characteristic curve, including: The valley filling feature prediction component includes a pre-constructed first grid load valley feature prediction model and a second grid load valley feature prediction model, wherein the first grid load valley feature prediction model and the second grid load valley feature prediction model both include multi-dimensional grid load valley feature prediction indicators, and the multi-dimensional grid load valley feature prediction indicators include grid load valley scale characteristics and grid load valley time characteristics; Based on the predetermined future time zone and the first grid load valley characteristic curve, predicting a first model according to the grid load valley characteristic, and outputting a first grid load valley characteristic prediction result; Based on the predetermined future time zone and the second power grid load valley characteristic curve, predicting a second model according to the power grid load valley characteristic, and outputting a second power grid load valley characteristic prediction result; The first power grid load valley characteristic prediction result and the second power grid load valley characteristic prediction result are integrated to obtain the power grid valley filling characteristic prediction result.

6. The system according to claim 1, wherein: The power grid peak shaving control module is used to perform peak shaving control on the power grid based on the hydrogen fuel cell module and the power grid peak shaving characteristic prediction result, including: Collecting real-time energy storage information of the plurality of hydrogen fuel cells to obtain a battery energy storage characteristic data set; Building a hydrogen energy power supply auxiliary decision-making model, and embedding the hydrogen energy power supply auxiliary decision-making model into the power grid peak shaving control module; Inputting the battery energy storage characteristic data set and the power grid peak shaving characteristic prediction result into the hydrogen energy power supply auxiliary decision model to obtain a hydrogen energy power supply auxiliary decision; The power grid is peak-cut and regulated according to the hydrogen energy power supply auxiliary decision and the hydrogen fuel cell module.

7. The system according to claim 1, characterized in that The grid valley filling control module is used to perform valley filling control on the grid based on the hydrogen fuel cell module and the grid valley filling characteristic prediction result, including: Collecting energy storage gap information of the plurality of hydrogen fuel cells to obtain a battery energy storage gap feature data set; Building a battery energy storage auxiliary decision-making model, and embedding the battery energy storage auxiliary decision-making model into the grid valley filling control module; Based on the grid valley filling feature prediction result and the battery energy storage gap feature data set, outputting a battery energy storage auxiliary decision according to the battery energy storage auxiliary decision model; The grid is valley-filled and regulated according to the battery energy storage auxiliary decision and the hydrogen fuel cell module.

8. The system of claim 1, wherein: The hydrogen fuel cell module includes a battery monitoring and early warning component, and the battery monitoring and early warning component is used to monitor and warn the multiple hydrogen fuel cells.

Images