Hydrogen production performance prediction method and equipment of hydrogen production system and storage medium

Abstract

The invention provides a hydrogen production performance prediction method and device of a hydrogen production system and a storage medium. The method comprises the steps that a target sub-region and a target hydrogen production system in the target sub-region are determined in a target region; obtaining target data of each sub-region based on the hydrogen production performance prediction model; performing vectorization processing on the target data of each sub-region to obtain a plurality of vector matrixes; carrying out segmentation processing on the vector matrix, and determining a prediction vector according to a vector obtained after segmentation; and predicting the hydrogen production performance of the target hydrogen energy product prepared by the target hydrogen production system according to the prediction vector, and obtaining and outputting a hydrogen production performance prediction value of the target hydrogen production system. According to the method, multi-dimensional data input and a machine learning method are adopted, the accuracy of hydrogen production performance prediction is remarkably improved, and effective decision support is provided for planning and optimization of a hydrogen production system.

Description

Technical Field

[0001] The present application relates to the technical field of hydrogen production, and particularly to a method, device and storage medium for predicting the hydrogen production performance of a hydrogen production system. Background Art

[0002] The performance prediction of a hydrogen production system has important guiding significance for industrial layout planning and investment decision-making. However, the prediction methods in related technologies have the following deficiencies: the prediction model relies on single-dimensional data input, mainly based on the regional hydrogen production raw material reserve for prediction, and fails to fully consider multi-dimensional influencing factors such as the regional economic development level, technological innovation ability and industrial structure; and the existing prediction methods mostly adopt static analysis models, and it is difficult to effectively capture and respond to the impact of dynamic change factors such as technological progress and industrial structure adjustment on the hydrogen production system performance. These technical defects lead to problems such as insufficient accuracy and low prediction efficiency of the prediction results. Therefore, how to establish a multi-dimensional and dynamic hydrogen production system performance prediction method to improve the accuracy and efficiency of prediction has become an urgent problem to be solved. Summary of the Invention

[0003] The present application provides a method, device and storage medium for predicting the hydrogen production performance of a hydrogen production system, aiming to provide a multi-dimensional and dynamic hydrogen production system performance prediction method to solve the problems of low accuracy and efficiency in hydrogen production performance prediction in related technologies.

[0004] In a first aspect, the present application provides a method for predicting the hydrogen production performance of a hydrogen production system, and the method for predicting the hydrogen production performance of the hydrogen production system includes the following steps:

[0005] According to a plurality of hydrogen production systems included in a target region, divide the target region into a plurality of sub-regions, and each sub-region includes one of the hydrogen production systems;

[0006] Determine at least one target sub-region in the target region, and determine the hydrogen production system in the target sub-region as a target hydrogen production system;

[0007] Based on the data acquisition sub-model of the hydrogen production performance prediction model, obtain the target data of each sub-region, and the target data includes the per capita gross regional product and the export value of the regional export products of the corresponding sub-region;

[0008] Based on the vectorization sub-model of the hydrogen production performance prediction model, perform vectorization processing on the target data of each sub-region to obtain a plurality of vector matrices;

[0009] Based on the vector processing sub-model of the hydrogen production performance prediction model, perform segmentation processing on the vector matrix, and determine a prediction vector according to the vectors obtained after segmentation;

[0010] Based on the prediction sub-model of the hydrogen production performance prediction model, the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product is predicted according to the prediction vector, and the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product is obtained and output.

[0011] In a second aspect, the present application also provides a computer device, which includes a processor, a memory, and a computer program stored on the memory and executable by the processor. When the computer program is executed by the processor, the steps of the hydrogen production performance prediction method of the hydrogen production system as described above are implemented.

[0012] In a third aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the hydrogen production performance prediction method of the hydrogen production system as described above are implemented.

[0013] The present application provides a hydrogen production performance prediction method, device, and storage medium for a hydrogen production system. By inputting the target data of each region into the hydrogen production performance prediction model, the hydrogen production performance of the hydrogen production system corresponding to the target sub-region is predicted at least through the per capita gross domestic product of each region and the export value of the regional export products. Through multi-dimensional data input, the present application avoids single-data prediction, improves the comprehensiveness of prediction, and realizes dynamic prediction of the hydrogen production performance of the region through the obtained per capita gross domestic product of the region and the export value of the regional export products; and after vectorizing the data through the hydrogen production performance prediction model and segmenting the vector, prediction is performed based on the segmented prediction vector, which improves the data processing rate and prediction accuracy, and further enables the obtained prediction results to provide more scientific and accurate decision-making support for policymakers and investors. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is a schematic flowchart of a hydrogen production performance prediction method for a hydrogen production system provided by an embodiment of the present application;

[0016] Figure 2 It is a schematic block diagram of the structure of a computer device related to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0018] The flowcharts shown in the accompanying drawings are only illustrative examples, and do not necessarily include all the content and operations / steps, nor do they necessarily need to be executed in the described order. For example, some operations / steps can be decomposed, combined, or partially merged, so the actual execution order may change according to the actual situation.

[0019] The embodiments of the present application provide a method, device, and storage medium for predicting the hydrogen production performance of a hydrogen production system. Among them, the method for predicting the hydrogen production performance of the hydrogen production system can be applied to a terminal device, which can be an electronic device such as a tablet computer, a notebook computer, or a desktop computer, and can also be applied to a server, which can be a cloud server or a server cluster, etc.

[0020] Next, in conjunction with the accompanying drawings, some embodiments of the present application will be described in detail. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0021] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for predicting the hydrogen production performance of a hydrogen production system provided by an embodiment of the present application.

[0022] As Figure 1 shown, the method for predicting the hydrogen production performance of the hydrogen production system includes steps S101 to S106.

[0023] Step S101: Divide the target area into a plurality of sub-areas according to a plurality of hydrogen production systems included in the target area, and each of the sub-areas includes one of the hydrogen production systems.

[0024] In a specific implementation process, the target area includes, but is not limited to, areas divided geographically, such as countries, cities, etc., which are not limited in the present application. The target area is divided into a plurality of sub-areas according to a preset division rule, and each sub-area includes one hydrogen production system. It should be understood that the preset division rule includes, but is not limited to, administrative area division rules. For example, if the target area includes a country, the sub-areas include the cities of the country; if the target area includes a city, the sub-areas include the administrative regions of the city. The hydrogen production system includes, but is not limited to, equipment, factories, industries, etc. for producing hydrogen energy products such as hydrogen, which are also not limited in the present application.

[0025] Step S102: Determine at least one target sub-region in the target region, and determine the hydrogen production system within the target sub-region as the target hydrogen production system.

[0026] Exemplarily, the target hydrogen production system is used to indicate the hydrogen production system whose hydrogen production performance is to be predicted. During the prediction process, one or more target hydrogen production systems can be predicted simultaneously to improve the prediction efficiency of the hydrogen production performance of the hydrogen production system.

[0027] Step S103: Based on the data acquisition sub-model of the hydrogen production performance prediction model, obtain the target data of each sub-region. The target data includes the regional per capita GDP and the export value of the regional export products corresponding to the sub-region.

[0028] Exemplarily, after determining the target hydrogen production system corresponding to the target sub-region, input the information corresponding to the hydrogen production systems of each sub-region into the hydrogen production performance prediction model, so as to obtain information such as the regional per capita GDP and the export value of the regional export products of each sub-region through the data acquisition sub-model of the hydrogen production performance prediction model.

[0029] In some embodiments, obtaining the target data of each sub-region based on the data acquisition sub-model of the hydrogen production performance prediction model includes: obtaining target text information and / or target images, where the target text information and / or target images include at least one piece of information indicating the regional per capita GDP, the export value of the regional export products, and the total regional product export value of the sub-region; based on the data acquisition sub-model, perform text recognition processing on the target text information to extract the target data in the target text information; and / or perform image recognition processing on the target image to extract the target data in the target image.

[0030] Exemplarily, the data acquisition sub-model can obtain the above target data through data crawlers and other means. For example, obtain the regional export products and the corresponding export value in the commodity trade database, and obtain the regional per capita GDP in the economic data disclosure website. It can be understood that the above data can also be obtained from other databases or websites. The obtained information corresponds to historical time information, such as the regional export products, the export value of the regional export products, and the regional per capita GDP within the past three months. When the corresponding information in this region changes, the information recorded in each database or website will also change. Therefore, the authenticity of information acquisition can be ensured. The present application does not limit the specific acquisition method of the target data.

[0031] In a specific implementation process, texts and / or images containing target data may be obtained from a database or a data disclosure website. These texts and / or images are determined as target texts and / or target images to obtain the target texts and / or target images, and the target texts are subjected to text recognition processing to obtain the target data in the target texts, and / or the target images are subjected to image recognition processing to obtain the target data in the target images. Among them, the target data may also be recorded in other ways, such as audio or video, etc. The data acquisition sub-model can also obtain the target data from the target audio or target video.

[0032] Step S104: Based on the vectorization sub-model of the hydrogen production performance prediction model, perform vectorization processing on the target data of each sub-region to obtain a plurality of vector matrices.

[0033] Exemplarily, after obtaining the target data of each sub-region, perform vectorization processing on the target data of each sub-region to obtain a vector matrix corresponding to the target data of each sub-region.

[0034] In a specific implementation process, perform vector encoding processing on the target data to obtain a vector matrix. For example, use one-hot encoding to generate a vector matrix corresponding to the target data; for example, in the vectorization sub-model, perform vectorization processing on the recognition text obtained after image recognition based on a picture to obtain a vector matrix.

[0035] Step S105: Based on the vector processing sub-model of the hydrogen production performance prediction model, perform segmentation processing on the vector matrix, and determine a prediction vector according to the vectors obtained after segmentation.

[0036] Exemplarily, input the generated vector matrix into the vector processing sub-model to segment the vector matrix in the vector processing sub-model to obtain a plurality of vectors, and determine a prediction vector through the plurality of vectors.

[0037] In a specific implementation process, set in the vector processing sub-model including but not limited to a U-Net network or an LSTM network to segment the vector matrix to obtain a plurality of vectors.

[0038] In some embodiments, the vector processing sub-model based on the hydrogen production performance prediction model performs segmentation processing on the vector matrix and determines a prediction vector according to the vectors obtained after segmentation, including: based on the vector matrix segmentation network of the vector processing sub-model, perform segmentation processing on each vector matrix to obtain a plurality of vectors; based on the clustering network of the vector processing sub-model, perform classification aggregation processing on the plurality of vectors to obtain a plurality of vector clusters; based on the prediction vector generation network of the vector processing sub-model, generate the prediction vector according to the clustering vectors corresponding to the plurality of vector clusters.

[0039] Exemplarily, the segmented vectors include, but are not limited to, a first vector for indicating the per capita GDP of a region and a second vector for indicating the export value of the export products of the region. It should be understood that the segmented vectors may also include other vectors such as a vector for indicating export products, etc., which are not limited in this application.

[0040] After the vector matrix is segmented into multiple vectors, the multiple vectors are clustered in a clustering network to assign the multiple vectors to different vector clusters. It should be noted that the same vector can be assigned to different vector clusters, which is not limited in this application.

[0041] In a specific implementation process, the clustering network can implement vector clustering through the K-means clustering algorithm, or set four preset vector clusters, each of which corresponds to a clustering preset value. The Euclidean distance between each vector and the clustering preset value of the vector cluster is used to determine the vector cluster to which each vector is assigned, so as to implement vector clustering and obtain four vector clusters after clustering is completed. The generated four vector clusters are input into the prediction vector generation network to generate a prediction vector according to the clustering vectors corresponding to the multiple vector clusters. By classifying and aggregating the vectors through the clustering network, the data processing rate can be effectively improved, and by generating the prediction vector after classifying the vectors, the prediction accuracy can be improved.

[0042] In some embodiments, the classifying and aggregating the multiple vectors to obtain multiple vector clusters includes: classifying and aggregating the vectors to obtain a first vector cluster, a second vector cluster, a third vector cluster, and a fourth vector cluster; wherein, the first vector cluster includes a first vector for indicating the per capita GDP of the region and a second vector for indicating the export value of the export products of the region; the second vector cluster and the third cluster both include the second vector; the fourth vector cluster includes a third vector for indicating the product proximity between the export products of the target sub-region and the target hydrogen energy products; the generating the prediction vector according to the clustering vectors corresponding to the multiple vector clusters includes: generating the prediction vector according to the clustering vectors corresponding to the first vector cluster, the second vector cluster, the third vector cluster, and the fourth vector cluster respectively.

[0043] In a specific embodiment, through a preset vector allocation rule, the multiple vectors are allocated to the first vector cluster, the second vector cluster, the third vector cluster, and the fourth vector cluster, so that the obtained first vector cluster includes a first vector of the per capita GDP of the region and a second vector for indicating the export value of the export products of the region; the second vector cluster and the third vector cluster both include the second vector, and the fourth vector cluster includes a third vector for indicating the product proximity between the export products of the target sub-region and the target hydrogen energy products.

[0044] After completing the vector clustering of the first vector clustering, the second vector clustering, the third vector clustering, and the fourth vector clustering, calculate the clustering vectors corresponding to each vector clustering respectively, so as to generate a prediction vector according to the clustering vectors.

[0045] In some embodiments, generating a prediction vector according to the clustering vectors corresponding to the first vector clustering, the second vector clustering, the third vector clustering, and the fourth vector clustering respectively includes: performing a first preset vector operation process on the vectors in the first vector clustering to obtain a first clustering vector; determining a second clustering vector according to the first clustering vector and the vectors in the second vector clustering; performing a second preset vector operation process on the vectors in the third vector clustering to obtain a third clustering vector; determining a fourth clustering vector according to the third clustering vector and the vectors in the fourth vector clustering; generating the prediction vector according to the first clustering vector, the second clustering vector, the third clustering vector, and the fourth clustering vector.

[0046] Exemplarily, after calculating the clustering vectors of each vector clustering to determine a prediction vector, the hydrogen production performance prediction of the hydrogen production system is completed, and multiple index information related to the hydrogen production performance of the hydrogen production system can be determined. Furthermore, the hydrogen production performance prediction of the hydrogen production system is realized, and by clustering multiple vectors into different vector clusterings to calculate the clustering vectors of multiple vector clusterings, relevant index information can be obtained, which can effectively improve the calculation efficiency and accuracy of the index information, and further improve the accuracy and efficiency of the hydrogen production performance prediction.

[0047] In a specific implementation manner, based on the following formula, perform a first preset vector operation process on the vectors in the first vector clustering to obtain a first clustering vector:

[0048]

[0049] where PRODY p is the first clustering vector, used to characterize the technical complexity of the export product p in the target sub-region c, X c,p is the second vector, used to indicate the export value of the target export product p in the target sub-region c during the target time period; Y c is the first vector, used to indicate the per capita gross regional product in the target sub-region c during the target time period. It should be noted that the target time is the time corresponding to the target data.

[0050] In a specific implementation process, first determine the target export products corresponding to the target sub-region. That is, if the hydrogen production system in the target sub-region c is regarded as the target hydrogen production system, then determine the export product p exported from the target sub-region c, so that the technological complexity corresponding to each export product p in the target sub-region c can be determined through the above formula.

[0051] It should be noted that the PRODY value is actually used to indicate the technological complexity of a certain export product in a certain region, and is used to indicate the complexity of the preparation process or the scientific and technological level required for preparing this export product. That is, the higher the PRODY value of the export product, the more complex the export product is, and higher technical level and production capacity are required. If the PRODY values of the products containing hydrogen production raw materials exported from a certain region are higher and the number is larger, it indicates that this region has stronger technical capabilities and a more mature industrial foundation. That is, the hydrogen production performance of the hydrogen production system in this region is also correspondingly greater.

[0052] In a specific implementation manner, determine the second clustering vector based on the following formula:

[0053]

[0054] where EXPY c is the second clustering vector, used to characterize the complexity of the export product portfolio of the target sub-region c, X c,p is the second vector, used to indicate the export value of the target export product p in the target sub-region c during the target time period, and PRODY p is the first clustering vector.

[0055] It should be noted that the EXPY value, as a comprehensive index reflecting the overall export complexity of a region, is positively correlated with the overall complexity of the export product portfolio of the region. And the overall complexity of the export products of the region is related to the overall technical level, industrial structure and innovation ability of the region. Therefore, a high EXPY value often means a stronger technical foundation, a more complete industrial system and greater development potential, and further makes the ability of the hydrogen production system in this region to prepare target hydrogen energy products higher. Specifically, the EXPY value of the target sub-region is positively correlated with the average PRODY value of the target sub-region.

[0056] In a specific implementation manner, determine the third clustering vector based on the following formula:

[0057]

[0058] where RCA c,p is the third clustering vector, and X c,p is used to indicate the export value of the export product p in the target sub-region c during the target time period.

[0059] It should be noted that RCAc,p It is actually used to indicate the ratio of the export value of product p in the target sub-region c during the target time period to the total export value of the target sub-region c, to the ratio of the export value of the export product p in all sub-regions within the target region to the total export value of all export products within the target region, so as to be based on the RCA c,p The value can measure whether the export product p of the target sub-region c has a comparative advantage among all export products, as well as the magnitude of the comparative advantage, and can reflect the resource endowment and industrial characteristics of this sub-region. It should be understood that if the target sub-region c has an export product p with RCA c,p ≥ 1, and this export product p includes the raw materials for preparing the target hydrogen energy product, then it can be determined that the target sub-region c has an advantage in preparing this target hydrogen energy product.

[0060] In a specific implementation manner, the fourth clustering vector is determined based on the following formula:

[0061]

[0062] where Distance c,p is the fourth clustering vector, φ p,p′ is the third vector, and M c,p′ is determined according to the above-mentioned third clustering vector.

[0063] Specifically, the third vector is used to characterize the product proximity between the export product p and the target hydrogen energy product p'.

[0064] It should be noted that Distance c,p is actually used to indicate the distance between the existing product structure of the target sub-region c and the target hydrogen energy product. Among them, the distance between products is used to indicate the difference between the existing product structure and the new product. A smaller distance is used to indicate that this region is relatively easy to produce this new product. The performance of the hydrogen energy system in the region for preparing the target hydrogen energy product is negatively correlated with the distance between the current product structure in this region and the target hydrogen energy product. It can be seen that the fourth clustering vector can reflect the hydrogen production performance of the hydrogen production system in the target sub-region.

[0065] Specifically, φ p,p′ can be calculated based on the following formula:

[0066] φ p,p′ = min{p(RCA P ≥ 1|RCA P′ ≥ 1), p(RCA P′ ≥ 1|RCA P ≥ 1)}

[0067] It should be noted that if more sub-regions in the target region simultaneously export two products p and p' and have a comparative advantage, it can be determined that the product proximity of these two products is higher, that is, the preparation capabilities required for these two products are more similar. Among them, the preparation capabilities include but are not limited to preparation conditions and product raw materials, etc. Therefore, the product proximity can determine the possibility of the product preparation system in the current region jumping from preparing one product to preparing another product. Starting from a specific product, the higher the product proximity, the higher the possibility of achieving the expansion of the target product preparation; the lower the product proximity, the higher the probability of failure in expanding to the target product. Theoretically, the product proximity between specific products should be equal, but due to different denominators, the probability that p' has a comparative advantage under the condition that p has a comparative advantage and the probability that p has a comparative advantage under the condition that p' has a comparative advantage are not equal. Therefore, the smaller value of the proximity between specific product pairs is taken to meet the prudence requirement.

[0068] In the specific implementation process, if RCA c,p ≥ 1, then the corresponding M c,p′ = 1, otherwise M c,p′ = 0.

[0069] It should be noted that the above method of determining the prediction vector through each clustering vector and predicting the hydrogen production performance of the target hydrogen production system can realize the prediction of the hydrogen production performance of the target hydrogen production system based on multi-dimensional information, avoiding the limitations of single-data prediction. And when the target data changes, the above clustering vectors will also change accordingly, thus realizing the prediction of hydrogen production performance according to dynamic data, and further improving the practicality of the prediction result.

[0070] In some embodiments, the method further includes: obtaining the proportion data of the preparation raw materials of the target hydrogen energy product and converting the proportion data of the preparation raw materials into a fourth vector; determining an update vector according to the fourth vector and a fifth vector for indicating the export products of the target sub-region; the step of generating the prediction vector according to the first clustering vector, the second clustering vector, the third clustering vector and the fourth clustering vector includes: performing a vector value update process on the first clustering vector and the third clustering vector according to the update vector to obtain an updated first clustering vector and an updated third clustering vector; generating the prediction vector according to the updated first clustering vector, the second clustering vector, the updated third clustering vector and the fourth clustering vector.

[0071] Exemplarily, determine the hydrogen production process and preparation raw materials for preparing the target hydrogen energy product, and determine the usage proportion of each preparation raw material to obtain the proportion data of the preparation raw materials, so as to improve the prediction accuracy of the hydrogen production performance of the hydrogen production system for preparing the target hydrogen energy product.

[0072] For ease of understanding, taking hydrogen gas as an example of the target hydrogen energy product, the obtained hydrogen production processes and the proportion of their raw materials are as follows:

[0073] 1.1 Fossil fuel reforming (about 76%)

[0074] (1) Steam Methane Reforming (SMR): It has the highest proportion, about 48%, and the raw material is natural gas (the main component is methane). The reaction formula is as follows:

[0075]

[0076] Coal Gasification: The proportion is about 18%, and the main raw material is coal. The reaction formula is as follows:

[0077]

[0078] (3) Partial Oxidation of Petroleum: The proportion is about 10%, and the raw materials are mainly petroleum, heavy oil, etc.

[0079] 1.2 Industrial by - product hydrogen (about 20%)

[0080] It mainly includes ethylene cracking, chlor - alkali industry, and refineries, and the main raw materials are petroleum and water.

[0081] 1.3 Hydrogen production by electrolysis of water (about 2%)

[0082] The main raw materials are water and electricity (renewable energy or traditional electricity). The reaction formula is as follows:

[0083] 2H 2 O → 2H 2 + O 2

[0084] 1.4 Hydrogen production from biomass (about 1%)

[0085] The main raw material is biomass, which mainly includes grass, straw, wood residues, sugarcane residues, etc.

[0086] 1.5 Others (about 1%)

[0087] Mainly include photocatalytic hydrogen production, microbial electrolysis cell hydrogen production, etc., which are in the laboratory research and development stage.

[0088] To sum up, the hydrogen production raw material with the largest proportion is natural gas, accounting for 48%, petroleum accounting for 30%, coal accounting for 18%, electric energy accounting for 2%, and biomass accounting for 1%.

[0089] It should be noted that the above data is for illustrative purposes only. With the upgrading and updating of the hydrogen production process, both the raw materials for preparing the target hydrogen energy products and their usage ratios will change, thereby causing the hydrogen production performance of the hydrogen production system to change accordingly. This application does not limit the raw materials for preparing the target hydrogen energy products and their usage ratios.

[0090] After determining the raw materials for preparing the target hydrogen energy products and their usage ratios, a fourth vector is generated, and an update vector is determined based on the fourth vector and a fifth vector to update the first clustering vector and the third clustering vector based on the update vector. The fifth vector is used to represent the export products of the target sub-region.

[0091] In a specific implementation process, the update process of the first clustering vector is as follows:

[0092]

[0093] Among them, is used to indicate the updated first clustering vector, PRODY p1 、PRODY p2 … are used to indicate the unupdated first clustering vectors, W 1 、W 2 … are used to indicate the fourth vector. The fifth vector includes vectors used to indicate the export products p1, p2. A mapping relationship between the export product p and the preparation raw material weight W corresponding to the export product p is determined according to the fourth vector and the fifth vector, so as to update the unupdated first clustering vector according to the mapping relationship to obtain the updated first clustering vector.

[0094] Specifically, PRODY p1 is used to represent the technical complexity corresponding to the export product p1, and W 1 is used to represent the usage ratio of the preparation raw materials included in the export product p1. The usage ratio of the preparation raw materials can be the data provided above. That is, when the export product p1 is natural gas, W 1 can be 0.48, or the usage ratio data of the preparation raw materials obtained from other data sources, which is not limited herein. By updating the usage ratio of the preparation raw materials corresponding to each export product p to the first clustering vector corresponding to each export product p, the accuracy of hydrogen production performance prediction is improved.

[0095] It should be understood that the update process of the third clustering vector is as follows:

[0096]

[0097] Among them, is used to indicate the updated third clustering vector, RCA c,p1 is used to indicate the unupdated third clustering vector, W 1 、W2 … is the fourth vector, which is used to represent the usage proportion of hydrogen production raw materials contained in export products p1, p2, ….

[0098] Through the above process, the first clustering vector and the third clustering vector can be updated. Based on the updated first clustering vector and third clustering vector, as well as the second clustering vector and the fourth clustering vector, a prediction vector is determined, so as to realize the hydrogen production performance prediction of the target hydrogen production system according to the prediction vector, improving the accuracy of the prediction result.

[0099] Step S106: Based on the prediction sub-model of the hydrogen production performance prediction model, predict the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product according to the prediction vector, and obtain and output the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product.

[0100] Exemplarily, after determining the prediction vector according to the clustering vector, the prediction vector is input into the prediction sub-model to predict the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product in the prediction sub-model, so as to obtain the corresponding hydrogen production performance prediction value.

[0101] In some embodiments, the prediction sub-model based on the hydrogen production performance prediction model predicts the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product according to the prediction vector, and obtains and outputs the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product, including: based on the normalization network of the prediction sub-model, performing standardization processing on the first clustering vector, the second clustering vector, the updated third clustering vector and the fourth clustering vector to obtain a first prediction vector, a second prediction vector, a third prediction vector and a fourth prediction vector; based on the activation network of the prediction sub-model, predicting the hydrogen production performance of the target hydrogen production system according to the first prediction vector, the second prediction vector, the third prediction vector and the fourth prediction vector.

[0102] Exemplarily, based on the following process, perform standardization processing on the first clustering vector to obtain the first prediction vector:

[0103]

[0104] where NTP is the first prediction vector, is the first clustering vector, which is used to represent the total technical complexity of export products containing the preparation raw materials of the target hydrogen energy product, TPRODY p_max 、TPRODY p_min respectively represent the maximum value and the minimum value of the total technical complexity of export products containing the preparation raw materials of the target hydrogen energy product in all sub-regions of the target region.

[0105] Based on the following process, the second clustering vector is normalized to obtain the second prediction vector:

[0106]

[0107] where NTE is the second prediction vector, and EXPY c_i is the second clustering vector, which is used to characterize the economic complexity of the product export basket of the target sub-region. EXPY c_max and EXPY c_min respectively represent the maximum and minimum values of the economic complexity of the product export baskets of all sub-regions in the target region.

[0108] Exemplarily, based on the following process, the third clustering vector is normalized to obtain the third prediction vector. It should be understood that since the hydrogen production performance of the target hydrogen production system is negatively correlated with the vector value of the third clustering vector, the third prediction vector is determined by using inverse normalization:

[0109]

[0110] where ND is the third prediction vector, and Distance c,p_i is the third clustering vector, which is used to characterize the distance between the currently prepared product of the target sub-region and the target hydrogen energy product. Distance c,p_max and Distance c,p_min respectively represent the maximum and minimum values of the distances between the products prepared by all sub-regions in the target region and the target hydrogen energy product.

[0111] Exemplarily, based on the following process, the fourth clustering vector is normalized to obtain the fourth prediction vector:

[0112]

[0113] where NTR is the fourth prediction vector, is the fourth clustering vector, which is used to characterize the total comparative advantage of the target sub-region in the export products containing hydrogen production raw materials, respectively represent the maximum and minimum values of the total comparative advantages of the export products containing hydrogen production raw materials of all sub-regions in the target region.

[0114] Exemplarily, through the above processes, the normalization of each clustering vector is completed to obtain the prediction vector corresponding to each clustering vector, so as to predict the hydrogen production performance of the target hydrogen production system, thereby improving the accuracy and practicality of the prediction result.

[0115] In a specific implementation process, based on an activation network, the hydrogen production performance of a target hydrogen production system is predicted according to a first prediction vector, a second prediction vector, a third prediction vector, a fourth prediction vector, and a fifth prediction vector.

[0116] Exemplarily, the fifth prediction vector is determined by a fifth clustering vector obtained by clustering a fifth vector including a second vector, wherein the fifth clustering vector is determined based on the following formula:

[0117] TX c,p =X c,p1 +X c,p2 +X c,p3 ···

[0118] wherein, TX c,p is the fifth clustering vector, and X c,p is the second vector, which is used to represent the export value of the export product p in the sub-region c during the target time period, that is, the fifth clustering vector is used to represent the total export value of the sub-region during the target time period;

[0119] Then the fifth prediction vector is determined based on the following formula:

[0120]

[0121] wherein, NTX is the fifth prediction vector, and TX c,p_i is the fifth clustering vector, and TX c,p_max , TX c,p_min are respectively used to represent the maximum value and the minimum value in the total export values of the export products containing hydrogen production raw materials in all sub-regions in the target region.

[0122] The hydrogen production performance of the target hydrogen production system is predicted through the first prediction vector, the second prediction vector, the third prediction vector, the fourth prediction vector, and the fifth prediction vector.

[0123] In a specific implementation process, the activation network is further used to determine the weights corresponding to each prediction vector, so as to predict a hydrogen production performance prediction value based on an activation calculation function according to the weights corresponding to each prediction vector and the prediction vectors. The activation calculation function is shown in the following formula:

[0124] HPI = v 1 NTX + v 2 NTP + v 3 NE + v 4 ND + v 5 NTR

[0125] wherein, HPI is the hydrogen production performance prediction value, and v 1 , v 2 , v 3 , v 4 , v 5is the weight value of each prediction vector, and the sum is 1, such as v 1 = 0.2, v 2 = 0.1, v 3 = 0.3, v 4 = 0.3, v 5 = 0.1; It should be noted that the above values of the weight are for illustrative purposes only and do not limit the specific values of the weight.

[0126] It should be noted that since the value of the first prediction vector can represent the product manufacturing technology level of the target sub-region, the value of the second prediction vector can represent the overall complexity of the export products in the target sub-region, the value of the third prediction vector can represent the comparative advantage of the export products in the target sub-region, and the comparative advantage of the export products is positively correlated with the number of raw materials for preparing the target hydrogen energy product; the value of the fourth prediction vector can represent the difficulty of changing from preparing the current product to preparing the target hydrogen energy product in the target sub-region. Therefore, the hydrogen production performance of the target hydrogen energy system in the target sub-region is positively correlated with the values of the above-mentioned prediction vectors. Furthermore, by determining the sum of the vector values of the above-mentioned prediction vectors, the hydrogen production performance prediction value of the target hydrogen production system can be predicted.

[0127] In some embodiments, the method further includes: when it is determined that the target hydrogen production system meets the preparation conditions of the target hydrogen energy product according to the hydrogen production performance prediction value, generating the target operation parameters of the target hydrogen production system; based on the target operation parameters, adjusting the operation state of the target hydrogen production system.

[0128] Exemplarily, after obtaining the hydrogen production performance prediction value of the target hydrogen production system, according to the hydrogen production performance prediction value and the actual demand of the target sub-region, it is determined whether the target hydrogen production system meets the preparation conditions of the target hydrogen energy product. For example, if the hydrogen production performance prediction value is greater than the preset threshold and there is a preparation requirement for the target hydrogen energy product, it is determined that the target hydrogen production system meets the preparation conditions of the target hydrogen energy product. According to the current operation parameters of the target hydrogen production system, the target operation parameters are determined to adjust the operation state of the target hydrogen production system based on the target operation parameters. It should be understood that the target hydrogen production system includes, but is not limited to, factories in various industries and hydrogen energy product preparation equipment, etc. The current operation parameters include intermediate products prepared, product preparation directions, etc. By adjusting the production direction of the factory or equipment, the target hydrogen production system can be made to prepare the target hydrogen energy product.

[0129] The hydrogen production performance prediction method of the hydrogen production system provided by the above embodiment inputs the target data of each region into the hydrogen production performance prediction model to predict the hydrogen production performance of the hydrogen production system corresponding to the target sub-region at least through the per capita GDP corresponding to each region and the export value of the region's export products, avoiding the prediction of single data, improving the comprehensiveness of data prediction, and improving the prediction efficiency through the hydrogen production performance prediction model, so that the obtained prediction results can provide more scientific and accurate decision-making support for policymakers and investors.

[0130] Please refer to Figure 2 , Figure 2 which is a schematic block diagram of the structure of a computer device provided by an embodiment of the present application. The computer device can be a server or a terminal.

[0131] As Figure 2 shown, the computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the memory can include a non-volatile storage medium and an internal memory.

[0132] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any hydrogen production performance prediction method of the hydrogen production system.

[0133] The processor is used to provide computing and control capabilities to support the operation of the entire computer device.

[0134] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any hydrogen production performance prediction method of the hydrogen production system.

[0135] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that Figure 2 the structure shown in

[0136] It should be understood that the processor can be a Central Processing Unit (CPU), and the processor can also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0137] Among them, in one embodiment, the processor is used to run a computer program stored in a memory to implement the following steps:

[0138] According to multiple hydrogen production systems included in the target area, divide the target area into multiple sub-areas, and each sub-area contains one of the hydrogen production systems;

[0139] Determine at least one target sub-area in the target area, and determine the hydrogen production system in the target sub-area as the target hydrogen production system;

[0140] Based on the data acquisition sub-model of the hydrogen production performance prediction model, obtain the target data of each sub-area, and the target data includes the per capita gross regional product of the corresponding sub-area and the export value of the regional export products;

[0141] Based on the vectorization sub-model of the hydrogen production performance prediction model, perform vectorization processing on the target data of each sub-area to obtain multiple vector matrices;

[0142] Based on the vector processing sub-model of the hydrogen production performance prediction model, perform segmentation processing on the vector matrix, and determine the prediction vector according to the vectors obtained after segmentation;

[0143] Based on the prediction sub-model of the hydrogen production performance prediction model, predict the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product according to the prediction vector, and obtain and output the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product.

[0144] In one embodiment, when the processor implements the vector processing sub-model of the hydrogen production performance prediction model, performs segmentation processing on the vector matrix, and determines the prediction vector according to the vectors obtained after segmentation, it is used to implement:

[0145] The vector matrix splitting network based on the vector processing sub-model splits each of the vector matrices to obtain multiple vectors;

[0146] The clustering network based on the vector processing sub-model classifies and aggregates multiple vectors to obtain multiple vector clusters;

[0147] The prediction vector generation network based on the vector processing sub-model generates the prediction vector according to the clustering vectors corresponding to multiple vector clusters.

[0148] In one embodiment, when the processor realizes classifying and aggregating multiple vectors to obtain multiple vector clusters, it is used to realize:

[0149] Classify and aggregate the vectors to obtain a first vector cluster, a second vector cluster, a third vector cluster, and a fourth vector cluster;

[0150] Among them, the first vector cluster includes a first vector for indicating the per capita GDP of the region and a second vector for indicating the export value of the region's export products; the second vector cluster and the third cluster both include the second vector; the fourth vector cluster includes a third vector for indicating the product proximity between the export products of the target sub-region and the target hydrogen energy products;

[0151] When the processor realizes generating the prediction vector according to the clustering vectors corresponding to multiple vector clusters, it is also used to realize:

[0152] Generate the prediction vector according to the clustering vectors corresponding to the first vector cluster, the second vector cluster, the third vector cluster, and the fourth vector cluster respectively.

[0153] In one embodiment, when the processor realizes generating a prediction vector according to the clustering vectors corresponding to the first vector cluster, the second vector cluster, the third vector cluster, and the fourth vector cluster respectively, it is used to realize:

[0154] Perform a first preset vector operation on the vectors in the first vector cluster to obtain a first clustering vector;

[0155] Determine a second clustering vector according to the first clustering vector and the vectors in the second vector cluster;

[0156] Perform a second preset vector operation on the vectors in the third vector cluster to obtain a third clustering vector;

[0157] Determine a fourth clustering vector according to the third clustering vector and the vectors in the fourth vector cluster;

[0158] Generate the prediction vector according to the first clustering vector, the second clustering vector, the third clustering vector, and the fourth clustering vector.

[0159] In one embodiment, when implementing the hydrogen production performance prediction method of the hydrogen production system, the processor is further configured to implement:

[0160] Obtain the proportion data of the raw materials for preparing the target hydrogen energy product, and convert the proportion data of the raw materials for preparation into a fourth vector;

[0161] Determine an update vector according to the fourth vector and a fifth vector for indicating the export product of the target sub-region;

[0162] When implementing the generation of the prediction vector according to the first clustering vector, the second clustering vector, the third clustering vector, and the fourth clustering vector, the processor is further configured to implement:

[0163] Perform a vector value update process on the first clustering vector and the third clustering vector according to the update vector to obtain an updated first clustering vector and an updated third clustering vector;

[0164] Generate the prediction vector according to the updated first clustering vector, the second clustering vector, the updated third clustering vector, and the fourth clustering vector.

[0165] In one embodiment, when implementing the prediction sub-model based on the hydrogen production performance prediction model to predict the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product according to the prediction vector, and obtaining and outputting the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product, the processor is configured to implement:

[0166] Based on the normalization network of the prediction sub-model, perform a standardization process on the first clustering vector, the second clustering vector, the updated third clustering vector, and the fourth clustering vector to obtain a first prediction vector, a second prediction vector, a third prediction vector, and a fourth prediction vector;

[0167] Based on the activation network of the prediction sub-model, predict the hydrogen production performance prediction value of the target hydrogen production system for preparing the target hydrogen energy product according to the first prediction vector, the second prediction vector, the third prediction vector, and the fourth prediction vector.

[0168] In one embodiment, when implementing the hydrogen production performance prediction of the hydrogen production system, the processor is further configured to implement:

[0169] Generate the target operating parameters of the target hydrogen production system when it is determined according to the hydrogen production performance prediction value that the target hydrogen production system meets the preparation conditions of the target hydrogen production product;

[0170] Adjust the operating state of the target hydrogen production system based on the target operating parameters.

[0171] In one embodiment, when the processor implements the data acquisition sub-model of the hydrogen production performance prediction model to obtain the target data of each sub-region, it is used to implement:

[0172] Obtain target text information and / or target images, where the target text information and / or target images include at least one piece of information indicating the per capita GDP of the region, the export value of the region's export products, and the total regional product export value of the sub-region;

[0173] Based on the data acquisition sub-model, perform text recognition processing on the target text information to extract the target data in the target text information; and / or perform image recognition processing on the target image to extract the target data in the target image.

[0174] It should be noted that those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above computer device can refer to the corresponding process in the embodiment of the hydrogen production performance prediction method of the hydrogen production system described above, and will not be elaborated here.

[0175] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and the computer program includes program instructions. The method implemented when the program instructions are executed can refer to each embodiment of the hydrogen production performance prediction method of the hydrogen production system of the present application.

[0176] Among them, the computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc.

[0177] It should be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0178] It should also be understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. It should be noted that, in this text, the term "comprises", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article or system comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "comprising a..." does not preclude the presence of additional identical elements in the process, method, article or system comprising the element.

[0179] The serial numbers of the embodiments of the present application above are for description only and do not represent the superiority or inferiority of the embodiments. The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for predicting hydrogen production performance of a hydrogen production system, characterized in that: include: According to the multiple hydrogen production systems included in the target area, the target area is divided into multiple sub-areas, each of which includes one hydrogen production system; Determine at least one target sub-region in the target region, and determine the hydrogen production system in the target sub-region as a target hydrogen production system; Based on the data acquisition sub-model of the hydrogen production performance prediction model, target data of each sub-region is acquired, wherein the target data includes the regional per capita GDP and the export value of regional export products of the corresponding sub-region; Based on the vectorized sub-model of the hydrogen production performance prediction model, vectorized processing is performed on the target data of each of the sub-regions to obtain a plurality of vector matrices; Based on the vector processing sub-model of the hydrogen production performance prediction model, the vector matrix is ​​segmented and processed, and a prediction vector is determined according to the vectors obtained after segmentation; Based on the prediction sub-model of the hydrogen production performance prediction model, the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product is predicted according to the prediction vector, and a predicted value of the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product is obtained and output.

2. The method for predicting hydrogen production performance of a hydrogen production system according to claim 1, characterized in that: The vector processing sub-model based on the hydrogen production performance prediction model performs segmentation processing on the vector matrix and determines the prediction vector according to the vector obtained after segmentation, including: Based on the vector matrix segmentation network of the vector processing sub-model, segmentation processing is performed on each of the vector matrices to obtain multiple vectors; Based on the clustering network of the vector processing sub-model, a plurality of the vectors are classified and aggregated to obtain a plurality of vector clusters; A prediction vector generation network based on the vector processing sub-model generates the prediction vector according to cluster vectors corresponding to a plurality of the vector clusters.

3. The method for predicting hydrogen production performance of a hydrogen production system according to claim 2, characterized in that: The classifying and aggregating the plurality of vectors to obtain a plurality of vector clusters includes: Classifying and clustering the vectors to obtain a first vector cluster, a second vector cluster, a third vector cluster, and a fourth vector cluster; The first vector cluster includes a first vector for indicating the regional per capita GDP and a second vector for indicating the export value of regional export products; the second vector cluster and the third cluster both include the second vector; the fourth vector cluster includes a third vector for indicating the product proximity between the export products of the target sub-region and the target hydrogen energy product; The step of generating the prediction vector according to the cluster vectors corresponding to the plurality of vector clusters comprises: The prediction vector is generated according to the cluster vectors corresponding to the first vector cluster, the second vector cluster, the third vector cluster, and the fourth vector cluster.

4. The method for predicting hydrogen production performance of a hydrogen production system according to claim 3, characterized in that: The generating a prediction vector according to the cluster vectors respectively corresponding to the first vector cluster, the second vector cluster, the third vector cluster and the fourth vector cluster comprises: Performing a first preset vector operation process on the vectors in the first vector cluster to obtain a first cluster vector; determining a second clustering vector based on the first clustering vector and a vector in the second vector cluster; Performing a second preset vector operation process on the vectors in the third vector cluster to obtain a third cluster vector; Determining a fourth cluster vector based on the third cluster vector and a vector in the fourth vector cluster; The prediction vector is generated according to the first clustering vector, the second clustering vector, the third clustering vector and the fourth clustering vector.

5. The method for predicting hydrogen production performance of a hydrogen production system according to claim 4, characterized in that: The method further comprises: Obtaining the raw material proportion data for preparing the target hydrogen energy product, and converting the raw material proportion data for preparing the target hydrogen energy product into a fourth vector; determining an update vector according to the fourth vector and a fifth vector indicating export products of the target sub-region; The step of generating the prediction vector according to the first clustering vector, the second clustering vector, the third clustering vector and the fourth clustering vector comprises: performing vector value updating processing on the first clustering vector and the third clustering vector according to the update vector to obtain an updated first clustering vector and an updated third clustering vector; The prediction vector is generated according to the updated first clustering vector, the second clustering vector, the updated third clustering vector and the fourth clustering vector.

6. The method for predicting hydrogen production performance of a hydrogen production system according to claim 5, characterized in that: The prediction sub-model based on the hydrogen production performance prediction model predicts the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product according to the prediction vector, and obtains and outputs the predicted value of the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product, including: Based on the normalized network of the prediction sub-model, the first clustering vector, the second clustering vector, the updated third clustering vector and the fourth clustering vector are normalized to obtain a first prediction vector, a second prediction vector, a third prediction vector and a fourth prediction vector; Based on the activation network of the prediction sub-model, according to the first prediction vector, the second prediction vector, the third prediction vector and the fourth prediction vector, a predicted value of the hydrogen production performance of the target hydrogen production system for preparing the target hydrogen energy product is predicted.

7. The method for predicting hydrogen production performance of a hydrogen production system according to any one of claims 1 to 6, characterized in that: The method further comprises: generating target operating parameters when it is determined that the target hydrogen production system meets the preparation conditions of the target hydrogen production product according to the predicted value of hydrogen production performance; Based on the target operating parameters, the operating state of the target hydrogen production system is adjusted.

8. The method for predicting hydrogen production performance of a hydrogen production system according to any one of claims 1 to 6, characterized in that: The data acquisition sub-model based on the hydrogen production performance prediction model acquires target data of each of the sub-areas, including: Acquire target text information and / or target image, wherein the target text information and / or target image includes at least one of information indicating regional per capita gross domestic product, regional export products, export value of regional export products, and total export value of regional products of the sub-region; Based on the data acquisition sub-model, text recognition processing is performed on the target text information to extract target data in the target text information; and / or image recognition processing is performed on the target image to extract target data in the target image.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the method for predicting the hydrogen production performance of the hydrogen production system as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the method for predicting hydrogen production performance of a hydrogen production system according to any one of claims 1 to 7 are implemented.

Images