Alkaline hydrogen production system electrolyte flow control method and system based on multi-dimensional knowledge graph

Abstract

The invention provides an alkaline hydrogen production system electrolyte flow control method and system based on a multi-dimensional knowledge graph, and relates to the technical field of alkaline hydrogen production. Comprising the following steps: acquiring basic data of a current alkaline hydrogen production system to construct a knowledge graph; collecting real-time operation data of the current alkaline hydrogen production system, inputting the real-time operation data into the knowledge graph for cross-entity relation reasoning, and outputting a real-time state score of the electrolysis system; obtaining an optimal gas-liquid ratio set value according to the real-time state score of the electrolysis system and a knowledge graph association rule; calculating a theoretical gas yield according to the optimal gas-liquid ratio set value and the total current of the electrolytic cell in combination with a Faraday law so as to determine an electrolyte flow set value; determining a flow deviation according to the electrolyte flow set value and the electrolyte real-time flow; according to the flow deviation, flow speed self-adaption adjustment is conducted through the electrolysis system real-time state score and the threshold value score; according to the invention, the problems of untimely flow adjustment, potential safety hazards and high cost in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of alkaline hydrogen production, and in particular to a method and system for controlling electrolyte flow in an alkaline hydrogen production system based on a multidimensional knowledge graph. Background Art

[0002] Electrolyte flow control in water electrolysis hydrogen and oxygen production systems is crucial for maintaining stable system operation, improving gas quality, reducing energy consumption, and mitigating equipment corrosion. Existing electrolyte flow control in water electrolysis hydrogen and oxygen production systems is manually set and essentially fixed. Existing electrolyte flow control technologies fail to account for the impact of electrolyzer power fluctuations, resulting in significant variations in the gas-liquid ratio within the system and instability in the gas-liquid separation process. This leads to unstable product gas purity and poor system stability. The system is unable to adapt to power fluctuations, such as renewable energy sources, or to grid peak load regulation.

[0003] In this regard, the existing invention patent with patent publication number CN115161707B discloses a method and system for controlling electrolyte flow in a water electrolysis hydrogen and oxygen production system, including: using a data acquisition device to obtain the total current, electrolyte flow, electrolytic cell pressure and electrolytic cell outlet temperature of the electrolytic cell; determining the gas volume flow rate under the working state based on the number of chambers in the electrolytic cell; calculating a set value for the electrolyte volume flow rate according to a set gas-liquid ratio; in the current iteration cycle, comparing the electrolyte volume flow rate set value with the real-time electrolyte flow rate and performing flow adjustment; in each subsequent iteration cycle, determining a new electrolyte volume flow rate set value based on the re-collected actual data and adjusting it. However, the current patent content does not adjust the flow rate in a timely manner when the electrolytic cell load fluctuates or the sensor fails, resulting in substandard gas purity; drastic changes in the electrolytic cell temperature and pressure affect the gas-liquid ratio and flow rate, and no effective countermeasures are provided; insufficient energy consumption optimization leads to low system energy efficiency and increased operating costs. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide an electrolyte flow control method and system for an alkaline hydrogen production system based on a multidimensional knowledge graph. The present invention solves the problems in the existing technology of untimely flow adjustment, potential safety hazards and high cost.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for controlling electrolyte flow in an alkaline hydrogen production system based on a multidimensional knowledge graph, comprising:

[0007] Obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to build a knowledge graph containing a multi-dimensional entity relationship network;

[0008] Collecting real-time operating data of the current alkaline hydrogen production system and inputting the real-time operating data into the knowledge graph for cross-entity relationship reasoning, and outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration;

[0009] Calculating an optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0010] The electrolyte flow rate setting value is determined by calculating the theoretical gas yield according to the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law;

[0011] determining a flow deviation according to the electrolyte flow set value and the electrolyte real-time flow;

[0012] According to the flow deviation, the flow rate is adaptively adjusted using the real-time status score and threshold score of the electrolysis system, wherein the adaptive adjustment method is:

[0013] If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation. If the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

[0014] Preferably, the entities include electrolytic cell current, temperature, pressure, electrolyte flow, gas purity, and energy efficiency index, and the entities are connected through dynamic weight association rules.

[0015] Preferably, the knowledge graph of the multi-dimensional entity relationship network includes:

[0016] Basic entity layer, dynamic relationship layer, weight update mechanism, reasoning and scoring module and compensation mechanism module;

[0017] The basic entity layer is used to determine the basic parameters of the electrolytic cell, the dynamic relationship layer is used to establish the association relationship between different entities, the weight update mechanism is used to determine the association weight of the points between entities through the gradient descent algorithm, the reasoning and scoring module is used to calculate the real-time status score of the electrolytic cell based on the basic parameters of the electrolytic cell, the association relationship between different entities and the association weight of the points between entities, and the supplementary mechanism module is used to generate compensation flow according to the association rules in the knowledge graph.

[0018] Preferably, the calculation process of the real-time status score of the electrolysis system includes:

[0019] Determining a function set that affects the real-time status score of the electrolysis system according to the association rule, wherein the function set includes: a first influence function, a second influence function, a third influence function, a fourth influence function, and a fifth influence function;

[0020] Obtaining a real-time status score of the electrolysis system according to the determined association weight and the corresponding function in the function set;

[0021] The calculation expression of the real-time status score of the electrolysis system is:

[0022] S(t)=f(I(t),Q(t),P(t),T(t),C(t))=w1*f I (I(t))+w2*f Q (Q(t))+w3*f P (P(t))+w4*f T (T(t))

[0023] +w5*f C (C(t));

[0024] Among them, S(t) is the real-time status score of the electrolysis system, f I (I(t)) is the first function, f Q (Q(t)) is the second influence function, f P (P(t)) is the third influence function, f T (T(t)) is the fourth influence function, f C (C(t)) is the fifth influence function, I(t), Q(t), P(t), T(t), C(t) are respectively the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration, w1, w2, w3, w4 and w5 are respectively the weight coefficients corresponding to the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration;

[0025] The expression of the first function is:

[0026]

[0027] The expression of the second function is:

[0028]

[0029] The expression of the third function is:

[0030]

[0031] The expression of the fourth function is:

[0032]

[0033] The expression of the fifth function is:

[0034]

[0035] More specifically, the process of determining association rules is as follows: collecting historical operating data of the electrolytic cell, including relevant parameters such as current, flow, pressure, temperature and hydrogen concentration; cleaning and organizing the collected data to ensure the integrity and consistency of the data for subsequent analysis; using data mining technology, selecting regression analysis to analyze the relationship between different entities (variables); through algorithmic analysis, discovering how the variables in the electrolysis process affect each other in order to establish a basic association framework; based on model analysis, identifying the key factors that have a greater impact on the real-time status score of the electrolysis system; confirming which factors have a significant impact on system performance under different conditions; defining specific influence functions based on the identified key factors, and quantifying the impact of these factors on the performance evaluation of the electrolytic cell. The influence function should be able to effectively reflect the specific contribution of each factor to the electrolysis system under different conditions, and may take a linear or nonlinear form.

[0036] The first influencing function is related to the total current of the electrolyzer. Current is the driving force of the electrolysis reaction and directly affects the rate of hydrogen production.

[0037] When the current value is appropriate, the electrolysis process is efficient; however, if the current exceeds a certain threshold, it may cause the equipment to overheat or reduce efficiency. Therefore, this function evaluates the impact of current changes on system efficiency.

[0038] The second influencing function is related to the real-time flow rate of the electrolyte. The flow rate affects the concentration of reactants in the electrolyzer, thereby affecting the reaction rate and hydrogen production.

[0039] A flow rate that is too high may result in a dilute concentration of reactants in the electrolyte, while a flow rate that is too low may slow down the reaction rate. Therefore, this function reflects the effect of flow rate on hydrogen production and quality.

[0040] The third influencing function is related to the pressure in the electrolytic cell. The pressure changes the behavior of the gas during the electrolysis process and affects the stability of the reaction.

[0041] Within the normal pressure range, the reaction efficiency is high; however, when the pressure deviates from the normal range, the reaction efficiency may be impaired. This function is used to quantify the impact of pressure on electrolytic cell performance.

[0042] The fourth influence function is related to the temperature of the electrolytic cell. Temperature is a key factor affecting the rate of chemical reactions. High temperatures usually accelerate the reaction, while low temperatures may lead to incomplete reactions.

[0043] This function evaluates the effect of temperature changes on hydrogen generation efficiency and helps determine how to adjust operations to maintain the optimal reaction rate under different temperature conditions.

[0044] The fifth influence function is related to the purity of hydrogen. The concentration of hydrogen directly reflects the effect of the electrolysis process.

[0045] Accurately evaluating changes in hydrogen concentration is crucial to ensuring that the system operates in an efficient state, so this function is used to analyze the impact of hydrogen purity on the system status score.

[0046] Preferably, the calculating of the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph includes:

[0047] Determine the preset basic gas-liquid ratio setpoint and target state score;

[0048] Determine the real-time gas-liquid ratio value corresponding to the real-time status score using the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0049] An optimal gas-liquid ratio setting value is determined according to the preset basic gas-liquid ratio setting value, the target state score and the real-time gas-liquid ratio value.

[0050] Preferably, the calculation formula for the optimal gas-liquid ratio setting value is:

[0051] R opt =R base +k*(S(t)-S target )*G -1 (S(t));

[0052] Among them, the R opt is the optimal gas-liquid ratio setting value, k is the adjustment coefficient, R base is the preset basic gas-liquid ratio setting value, S target Score the target state, G -1 (S(t)) is the gas-liquid ratio-efficiency correlation subgraph to determine the real-time gas-liquid ratio value corresponding to the real-time status score.

[0053] Preferably, the step of calculating the theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine the electrolyte flow setting value comprises:

[0054] The number of moles of gas is calculated based on the total current of the electrolytic cell and in combination with Faraday's law;

[0055] determining a theoretical gas yield based on the number of moles of gas;

[0056] Calculate the theoretical gas flow rate based on the theoretical gas output and electrolysis time;

[0057] The electrolyte flow rate setting value is calculated based on the theoretical gas flow rate and the optimal gas-liquid ratio setting value.

[0058] Preferably, the basic data of the electrolytic cell include:

[0059] Electrolyzer structural parameters and historical operating data.

[0060] Preferably, the electrolytic cell structural parameters include:

[0061] Electrolyser geometry, material properties, active electrode area and electrolyte type.

[0062] An electrolyte flow control system for an alkaline hydrogen production system based on a multidimensional knowledge graph, comprising:

[0063] A knowledge graph construction module is used to obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to construct a knowledge graph containing a multi-dimensional entity relationship network;

[0064] A first calculation module is configured to collect real-time operating data of the current alkaline hydrogen production system and input the real-time operating data into the knowledge graph for cross-entity relationship reasoning, thereby outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration;

[0065] A second calculation module is used to calculate the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0066] a third calculation module, configured to calculate a theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine an electrolyte flow setting value;

[0067] a fourth calculation module, configured to determine a flow rate deviation according to the electrolyte flow rate set value and the electrolyte real-time flow rate;

[0068] An adjustment module is configured to adaptively adjust the flow rate based on the flow deviation using the real-time state score and threshold score of the electrolysis system, wherein the adaptive adjustment method is:

[0069] If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation. If the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

[0070] The present invention discloses the following technical effects:

[0071] The present invention provides an electrolyte flow control method and system for an alkaline hydrogen production system based on a multidimensional knowledge graph. The method comprises: obtaining electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to construct a knowledge graph containing a multidimensional entity relationship network; collecting real-time operation data of the current alkaline hydrogen production system and inputting the real-time operation data into the knowledge graph for cross-entity relationship reasoning, and outputting a real-time status score of the electrolysis system, wherein the operation data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration; calculating an optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and a gas-liquid ratio-efficiency association subgraph in the knowledge graph; calculating a theoretical gas production according to the optimal gas-liquid ratio setting value and the total current of the electrolyzer in combination with Faraday's law to determine an electrolyte flow setting value; and determining a flow deviation according to the electrolyte flow setting value and the real-time flow rate of the electrolyte;

[0072] According to the flow deviation, the real-time status score and threshold score of the electrolysis system are used to adaptively adjust the flow rate, wherein the adaptive adjustment method is: if the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation; if the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation. The present invention can infer the status score of the electrolysis system in real time by constructing a knowledge graph of a multi-dimensional entity relationship network and integrating the structural parameters of the electrolytic cell with historical operating data, thereby solving the problem that traditional methods have slow response to the operating status and rely on experience. In addition, based on the dynamic adjustment mechanism of the real-time score and the gas-liquid ratio-efficiency relationship, the pump frequency can be adaptively adjusted, which significantly improves the stability of gas production and purity, ensures the efficiency and economy of hydrogen production, and thus improves the intelligence level and automation level of the alkaline hydrogen production system. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0074] Figure 1 A flow chart of a method for controlling electrolyte flow in an alkaline hydrogen production system based on a multidimensional knowledge graph is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0077] like Figure 1 As shown, the present invention provides an electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph, comprising:

[0078] Step 100: Obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to construct a knowledge graph containing a multi-dimensional entity relationship network;

[0079] Step 200: Collecting real-time operating data of the current alkaline hydrogen production system and inputting the real-time operating data into the knowledge graph for cross-entity relationship reasoning, and outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration;

[0080] Specifically, cross-entity relationship reasoning refers to analyzing the relationships between different entities in a knowledge graph to derive new information or status, thereby providing a comprehensive assessment of the current system status. In an alkaline hydrogen production system, real-time operating data is used as input, and through the analysis of multi-dimensional relationships, a real-time status score of the electrolysis system is output;

[0081] The entities involved mainly include the following categories:

[0082] Total current of electrolytic cell (I): real-time data reflecting the current intensity of electrolyte.

[0083] Real-time electrolyte flow rate (Q): indicates the flow of electrolyte in the system.

[0084] Electrolytic cell pressure (P): The gas or liquid pressure inside the electrolytic cell, which affects the reaction efficiency and safety.

[0085] Electrolytic cell temperature (T): The temperature during the electrolysis reaction directly affects the reaction rate and gas production.

[0086] Outlet hydrogen concentration (C): The concentration of hydrogen discharged from the system after production, indicating the purity of the product and the reaction effect.

[0087] The specific implementation process of cross-entity relationship reasoning includes the following steps:

[0088] Data input: Input the real-time collected operational data (I, Q, P, T, C) into the reasoning module of the knowledge graph. This data reflects the current working status of the system.

[0089] Relationship mapping: In the knowledge graph, the relationships between these entities are defined and established as follows: the relationship between electrolytic cell current and electrolyte flow rate: affecting the gas generation rate; the relationship between temperature and electrolytic cell pressure: increased temperature may lead to increased pressure, affecting the safety and reaction rate of the electrolytic cell; the relationship between hydrogen concentration and flow rate: higher flow rate is generally associated with higher gas purity;

[0090] Step 300: Calculating an optimal gas-liquid ratio setting value based on the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0091] Step 400: Calculating a theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine an electrolyte flow setting value;

[0092] Step 500: determining a flow deviation according to the electrolyte flow set value and the electrolyte real-time flow;

[0093] Step 600: Based on the flow deviation, the flow rate is adaptively adjusted using the real-time status score and the threshold score of the electrolysis system, wherein the adaptive adjustment method is:

[0094] Step 601: If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation; if the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

[0095] Specifically, a knowledge graph is a structured representation of various related information and data. In an electrolysis system, the knowledge graph constructs a correlation model based on key parameters such as pressure, temperature, and flow rate. This model reflects the operating characteristics of the electrolytic cell under different conditions and their impact on electrolyte flow rate. Using these correlation rules, the system can quickly identify and respond to different operating conditions.

[0096] Pressure-Temperature-Flow Rolling Rule:

[0097] Pressure-temperature-flow compensation association rules are generated using data mining and machine learning techniques based on historical electrolyzer operating data. These rules describe how the electrolyte flow rate should be adjusted to achieve the desired electrolysis effect under specific pressure and temperature conditions. The rules state that at a specific temperature and pressure, if the flow rate falls below a certain threshold, the flow rate should be increased to improve hydrogen generation efficiency.

[0098] Furthermore, the entities include electrolytic cell current, temperature, pressure, electrolyte flow, gas purity, and energy efficiency indicators, and the entities are connected through dynamic weight association rules.

[0099] Specifically, dynamic weighted association rules are rules that dynamically adjust the strength (or importance) of relationships between different entities in the knowledge graph. These rules change during operation based on real-time data changes to reflect the latest state of the interaction between entities.

[0100] More specifically, when establishing connections between entities, the following aspects need to be considered:

[0101] Relevance: Whether entities influence each other. There's a direct correlation between electrolytic cell current and electrolyte flow rate; increasing current requires increasing flow rate to maintain reaction efficiency. There's also a close relationship between temperature and electrolytic cell pressure; rising temperature leads to rising pressure. Dynamic Update: Because the actual operating conditions of the system change at any time, the weights between entities cannot be statically maintained but must be adjusted dynamically.

[0102] More specifically, the weight adjustment mechanism: Historical data learning: Based on historical operating data, machine learning algorithms are used to identify the strength of relationships between different entities;

[0103] Dynamic weight calculation: The algorithm uses real-time data to feed the model and calculate the dynamic weight between each entity and other entities in real time. If an increase in electrolyzer current directly improves hydrogen production (gas purity), the weight between current and gas purity can be increased accordingly.

[0104] Apply weights: The obtained dynamic weights are used in subsequent reasoning and calculations to adjust the operating conditions of the electrolyzer in real time by affecting the scores and outputs.

[0105] Furthermore, the knowledge graph of the multi-dimensional entity relationship network includes:

[0106] Basic entity layer, dynamic relationship layer, weight update mechanism, reasoning and scoring module and compensation mechanism module;

[0107] The basic entity layer is used to determine the basic parameters of the electrolytic cell, the dynamic relationship layer is used to establish the association relationship between different entities, the weight update mechanism is used to determine the association weight of the points between entities through the gradient descent algorithm, the reasoning and scoring module is used to calculate the real-time status score of the electrolytic cell based on the basic parameters of the electrolytic cell, the association relationship between different entities and the association weight of the points between entities, and the supplementary mechanism module is used to generate compensation flow according to the association rules in the knowledge graph.

[0108] Furthermore, the basic data of the electrolytic cell include:

[0109] Electrolyzer structural parameters and historical operating data.

[0110] Furthermore, the electrolytic cell structural parameters include:

[0111] Electrolyser geometry, material properties, active electrode area and electrolyte type.

[0112] Specifically, the basic entity layer:

[0113] The basic entity layer is the foundation of the knowledge graph and is primarily used to define and store the basic parameters of the electrolyzer, ensuring that the system's performance evaluation is based on accurate data input. This layer primarily includes the following parameters:

[0114] Electrolyzer structural parameters: These parameters are key elements that affect the operating performance of the electrolyzer, including:

[0115] Geometry: The physical shape of the electrolytic cell, which affects the flow of reactants and current distribution.

[0116] Material properties: The conductivity, corrosion resistance and temperature tolerance of the materials used in the electrolytic cell determine the durability and stability of the electrolytic cell.

[0117] Effective electrode area: The area of ​​the electrode directly affects the efficiency of the electrolysis reaction. A larger electrode area usually increases the amount of hydrogen produced.

[0118] Electrolyte type: The type of electrolyte used. Different electrolytes have different effects on reaction rate and product quality.

[0119] Historical operating data: This data records the electrolyzer's operating status during previous operations, including flow, current, temperature, pressure, and gas purity. The accumulation of historical data facilitates model optimization and dynamic adjustments.

[0120] Dynamic relationship layer:

[0121] The dynamic relationship layer is used to establish relationships between different entities so that they can flexibly respond to parameter changes during operation. This layer determines the degree of mutual influence between entities through analysis and modeling of historical data, mainly involving:

[0122] Entity associations: such as the dynamic relationship between electrolyzer current, temperature, and flow, and their impact on gas purity and energy efficiency.

[0123] Real-time evaluation: Based on real-time data, the association rules between entities are automatically updated and adjusted to reflect the current operating status of the system.

[0124] Weight update mechanism:

[0125] The weight update mechanism is responsible for continuously optimizing the association weights between entities through the gradient descent algorithm, so that the model can timely adjust the influence of each entity based on real-time operation data. Specifically, this mechanism includes:

[0126] Gradient descent method: Gradient descent optimization algorithm is used to continuously adjust weights according to the loss function to reduce errors and improve the accuracy of the model.

[0127] Dynamic adjustment: During each operation cycle, the data changes of each entity are monitored in real time, and the weights are automatically updated based on the input data, so that the knowledge graph always remains adaptable.

[0128] Reasoning and scoring module:

[0129] The reasoning and scoring module is used to generate real-time status scores for the system based on the basic entity layer data, the association rules and weights of the dynamic relationship layer.

[0130] The compensation mechanism module generates compensation traffic based on the association rules in the knowledge graph to ensure that the system can adjust in time when facing unstable operating conditions:

[0131] Utilization of association rules: When real-time data deviations are detected, the pressure-temperature-flow compensation rules defined in the knowledge graph are used to generate the required compensation flow.

[0132] Adaptive adjustment: Based on the calculation results of the compensation flow, the electrolyte flow is adjusted in real time to optimize the hydrogen production and purity.

[0133] Furthermore, the calculation process of the real-time status score of the electrolysis system includes:

[0134] Determining a function set that affects the real-time status score of the electrolysis system according to the association rule, wherein the function set includes: a first influence function, a second influence function, a third influence function, a fourth influence function, and a fifth influence function;

[0135] Obtaining a real-time status score of the electrolysis system according to the determined association weight and the corresponding function in the function set;

[0136] The calculation expression of the real-time status score of the electrolysis system is:

[0137] S(t)=f(I(t),Q(t),P(t),T(t),C(t))=w1*f I (I(t))+w2*f Q (Q(t))+w3*f P (P(t))+w4*f T (T(t))

[0138] +w5*f C (C(t));

[0139] Among them, S(t) is the real-time status score of the electrolysis system, f I (I(t)) is the first function, f Q (Q(t)) is the second influence function, f P (P(t)) is the third influence function, f T (T(t)) is the fourth influence function, f C (C(t)) is the fifth influence function, I(t), Q(t), P(t), T(t), C(t) are respectively the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration, w1, w2, w3, w4 and w5 are respectively the weight coefficients corresponding to the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration;

[0140] The expression of the first function is:

[0141]

[0142] The expression of the second function is:

[0143]

[0144] The expression of the third function is:

[0145]

[0146] The expression of the fourth function is:

[0147]

[0148] The expression of the fifth function is:

[0149]

[0150] Furthermore, the calculating of the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph includes:

[0151] Determine the preset basic gas-liquid ratio setpoint and target state score;

[0152] Determine the real-time gas-liquid ratio value corresponding to the real-time status score using the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0153] An optimal gas-liquid ratio setting value is determined according to the preset basic gas-liquid ratio setting value, the target state score and the real-time gas-liquid ratio value.

[0154] Specifically, in the process of hydrogen production by electrolysis, the gas-liquid ratio (GLR) refers to the ratio of gas volume to liquid volume, which usually affects the yield and purity of hydrogen. The calculation of the optimal gas-liquid ratio setting value aims to dynamically adjust the operating conditions through the association rules of real-time state scoring and knowledge graph, thereby improving the reaction efficiency. The subgraph establishes the relationship between the gas-liquid ratio and system efficiency by analyzing historical operating data and real-time monitoring. The gas-liquid ratio-efficiency association subgraph defines the changes in system efficiency under different gas-liquid ratios.

[0155] Furthermore, the calculation formula for the optimal gas-liquid ratio setting value is:

[0156] R opt =R base +k*(S(t)-S target )*G -1 (S(t));

[0157] Among them, the R opt is the optimal gas-liquid ratio setting value, k is the adjustment coefficient, R base is the preset basic gas-liquid ratio setting value, S target Score the target state, G -1 (S(t)) is the gas-liquid ratio-efficiency correlation subgraph to determine the real-time gas-liquid ratio value corresponding to the real-time status score.

[0158] Specifically, the adjustment coefficient determines the degree of influence between the real-time status score and the optimal gas-liquid ratio. This coefficient can be adjusted based on historical data to ensure the effectiveness of the model.

[0159] Furthermore, the step of calculating the theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine the electrolyte flow setting value includes:

[0160] The number of moles of gas is calculated based on the total current of the electrolytic cell and in combination with Faraday's law;

[0161] determining a theoretical gas yield based on the number of moles of gas;

[0162] Calculate the theoretical gas flow rate based on the theoretical gas output and electrolysis time;

[0163] The electrolyte flow rate setting value is calculated based on the theoretical gas flow rate and the optimal gas-liquid ratio setting value.

[0164] This embodiment also provides an alkaline hydrogen production system electrolyte flow control system based on a multi-dimensional knowledge graph, including:

[0165] A knowledge graph construction module is used to obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to construct a knowledge graph containing a multi-dimensional entity relationship network;

[0166] A first calculation module is configured to collect real-time operating data of the current alkaline hydrogen production system and input the real-time operating data into the knowledge graph for cross-entity relationship reasoning, thereby outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration;

[0167] A second calculation module is used to calculate the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph;

[0168] a third calculation module, configured to calculate a theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine an electrolyte flow setting value;

[0169] a fourth calculation module, configured to determine a flow rate deviation according to the electrolyte flow rate set value and the electrolyte real-time flow rate;

[0170] An adjustment module is configured to adaptively adjust the flow rate based on the flow deviation using the real-time state score and threshold score of the electrolysis system, wherein the adaptive adjustment method is:

[0171] If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation. If the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

[0172] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0173] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for controlling electrolyte flow in an alkaline hydrogen production system based on a multidimensional knowledge graph, characterized in that: include: Obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to build a knowledge graph containing a multi-dimensional entity relationship network; Collecting real-time operating data of the current alkaline hydrogen production system and inputting the real-time operating data into the knowledge graph for cross-entity relationship reasoning, and outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration; Calculating an optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph; The electrolyte flow rate setting value is determined by calculating the theoretical gas yield according to the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law; determining a flow deviation according to the electrolyte flow set value and the electrolyte real-time flow; According to the flow deviation, the flow rate is adaptively adjusted using the real-time status score and threshold score of the electrolysis system, wherein the adaptive adjustment method is: If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation. If the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

2. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 1, characterized in that: The entities include electrolytic cell current, temperature, pressure, electrolyte flow, gas purity, and energy efficiency indicators, and the entities are connected through dynamic weight association rules.

3. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 1, characterized in that: The knowledge graph of the multi-dimensional entity relationship network includes: Basic entity layer, dynamic relationship layer, weight update mechanism, reasoning and scoring module and compensation mechanism module; The basic entity layer is used to determine the basic parameters of the electrolytic cell, the dynamic relationship layer is used to establish the association relationship between different entities, the weight update mechanism is used to determine the association weight of the points between entities through the gradient descent algorithm, the reasoning and scoring module is used to calculate the real-time status score of the electrolytic cell based on the basic parameters of the electrolytic cell, the association relationship between different entities and the association weight of the points between entities, and the supplementary mechanism module is used to generate compensation flow according to the association rules in the knowledge graph.

4. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 3 is characterized in that: The calculation process of the real-time status score of the electrolysis system includes: Determining a function set that affects the real-time status score of the electrolysis system according to the association rule, wherein the function set includes: a first influence function, a second influence function, a third influence function, a fourth influence function, and a fifth influence function; Obtaining a real-time status score of the electrolysis system according to the determined association weight and the corresponding function in the function set; The calculation expression of the real-time status score of the electrolysis system is: S(t)=f(I(t),Q(t),P(t),T(t),C(t))=w1*f I (I(t))+w2*f Q (Q(t))+w3*f P (P(t))+w4*f T (T(t)) +w5*f C (C(t)); Among them, S(t) is the real-time status score of the electrolysis system, f I (I(t)) is the first function, f Q (Q(t)) is the second influence function, f P (P(t)) is the third influence function, f T (T(t)) is the fourth influence function, f C (C(t)) is the fifth influence function, I(t), Q(t), P(t), T(t), C(t) are respectively the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration, w1, w2, w3, w4 and w5 are respectively the weight coefficients corresponding to the total current of the electrolyzer, the real-time flow rate of the electrolyte, the pressure of the electrolyzer, the temperature of the electrolyzer and the outlet hydrogen concentration; The expression of the first function is: The expression of the second function is: The expression of the third function is: The expression of the fourth function is: The expression of the fifth function is:

5. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 3 is characterized in that: The calculating of the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph includes: Determine the preset basic gas-liquid ratio setpoint and target state score; Determine the real-time gas-liquid ratio value corresponding to the real-time status score using the gas-liquid ratio-efficiency association subgraph in the knowledge graph; An optimal gas-liquid ratio setting value is determined according to the preset basic gas-liquid ratio setting value, the target state score and the real-time gas-liquid ratio value.

6. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 5, characterized in that: The calculation formula for the optimal gas-liquid ratio setting value is: R opt =R base +k*(S(t)-S target )*G -1 (S(t)); Among them, the R opt is the optimal gas-liquid ratio setting value, k is the adjustment coefficient, R base is the preset basic gas-liquid ratio setting value, S target Score the target state, G -1 (S(t)) is the gas-liquid ratio-efficiency correlation subgraph to determine the real-time gas-liquid ratio value corresponding to the real-time status score.

7. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 5 is characterized in that: The step of calculating the theoretical gas yield according to the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine the electrolyte flow setting value includes: The number of moles of gas is calculated based on the total current of the electrolytic cell and in combination with Faraday's law; determining a theoretical gas yield based on the number of moles of gas; Calculate the theoretical gas flow rate based on the theoretical gas output and electrolysis time; The electrolyte flow rate setting value is calculated based on the theoretical gas flow rate and the optimal gas-liquid ratio setting value.

8. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 5, characterized in that: The basic data of the electrolytic cell include: Electrolyzer structural parameters and historical operating data.

9. The electrolyte flow control method for an alkaline hydrogen production system based on a multidimensional knowledge graph according to claim 8, characterized in that: The electrolytic cell structural parameters include: Electrolyser geometry, material properties, active electrode area and electrolyte type.

10. An electrolyte flow control system for an alkaline hydrogen production system based on a multidimensional knowledge graph, characterized in that: include: A knowledge graph construction module is used to obtain the electrolyzer structural parameters and historical operation data of the current alkaline hydrogen production system to construct a knowledge graph containing a multi-dimensional entity relationship network; A first calculation module is configured to collect real-time operating data of the current alkaline hydrogen production system and input the real-time operating data into the knowledge graph for cross-entity relationship reasoning, thereby outputting a real-time status score of the electrolysis system, wherein the operating data includes the total current of the electrolyzer, the real-time flow rate of the electrolyte, the electrolyzer pressure, the electrolyzer temperature, and the outlet hydrogen concentration; A second calculation module is used to calculate the optimal gas-liquid ratio setting value according to the real-time status score of the electrolysis system and the gas-liquid ratio-efficiency association subgraph in the knowledge graph; a third calculation module, configured to calculate a theoretical gas yield based on the optimal gas-liquid ratio setting value and the total current of the electrolytic cell in combination with Faraday's law to determine an electrolyte flow setting value; a fourth calculation module, configured to determine a flow rate deviation according to the electrolyte flow rate set value and the electrolyte real-time flow rate; An adjustment module is configured to adaptively adjust the flow rate based on the flow deviation using the real-time state score and threshold score of the electrolysis system, wherein the adaptive adjustment method is: If the real-time status score of the electrolysis system is greater than or equal to the threshold score, the pump frequency is adjusted according to the flow deviation. If the real-time status score of the electrolysis system is less than the threshold score, the pressure-temperature-flow compensation association rule in the knowledge graph is used to generate a compensation flow, and the pump frequency is adjusted according to the compensation flow and the flow deviation.

Images