Combustion chemical reaction kinetics data processing method and system for ADN-based liquid propellants

Abstract

The present invention discloses a combustion chemical reaction kinetics data processing method and system for ADN-based liquid propellants, and belongs to the field of data processing technology. The present invention first obtains the combustion chemical reaction kinetics vector and the primitive state linkage vector, so as to comprehensively capture the key information in the combustion process. The primitive state linkage vector is structured semantically mapped to obtain the reaction state structured vector and the strategy association requirement vector, which is helpful to accurately classify and locate the combustion reaction related information. The first and second data integration confidence weights are determined, and the degree of association between the primitive reaction data and the visual association labels under different integration strategies can be quantified, making the analysis of the combustion process more scientific and meticulous. The final determined visual monitoring results can intuitively present the characteristics of the primitive reaction data under different strategies, and provide clear and accurate data support for in-depth understanding of the combustion mechanism of ADN-based liquid propellants, optimizing propellant performance, and ensuring safe use.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a combustion chemical reaction kinetics data processing method and system for ADN-based liquid propellants. Background Art

[0002] In the research and application of ADN-based liquid propellants, in-depth understanding and effective monitoring of the combustion process have always been core technical issues. Traditional research methods often lack systematicity and accuracy when processing combustion process data. For example, it is difficult to effectively classify and integrate a large amount of elementary reaction data in the combustion process, and it is impossible to accurately judge the importance and interrelationships of each elementary reaction in the entire combustion process. At the same time, because the combustion process involves the interaction of multiple complex chemical reactions, environmental factors, and the characteristics of the propellant itself, the existing data processing methods are difficult to fully and accurately present these complex relationships, resulting in many challenges in combustion mechanism research, propellant performance optimization, and safe use. Summary of the invention

[0003] The present invention provides a combustion chemical reaction kinetics data processing method and system for ADN-based liquid propellants, which can solve or partially solve the technical problems involved in the above-mentioned background technology.

[0004] The embodiment of the present invention provides a combustion chemical reaction kinetics data processing method for ADN-based liquid propellants, which is applied to a data processing system. The method comprises: in response to a combustion process monitoring task, obtaining a combustion chemical reaction kinetics vector of the ADN-based liquid propellant corresponding to the combustion process monitoring task and a primitive state linkage vector corresponding to a plurality of primitive reaction data obtained according to the combustion process monitoring task; performing structured semantic mapping on the primitive state linkage vector of each primitive reaction data in the plurality of primitive reaction data to obtain a reaction state structured vector corresponding to at least one type of set combustion chemical reaction tracking event and a strategy association requirement vector corresponding to a plurality of set data integration strategies; based on the strategy association requirement vector corresponding to the original data integration strategy and each of the primitive state linkage vectors, a structured semantic mapping is performed on the primitive state linkage vector of each primitive reaction data in the plurality of primitive reaction data to obtain a reaction state structured vector corresponding to at least one type of set combustion chemical reaction tracking event and a strategy association requirement vector corresponding to each of the plurality of set data integration strategies; based on the strategy association requirement vector corresponding to the original data integration strategy and each of the primitive state linkage vectors, a structured semantic mapping is performed on the primitive state linkage vector of each primitive reaction data in the plurality of primitive reaction data. Reaction state structured vector, determine the first data integration confidence weight of the targeted elementary reaction data corresponding to at least one first set visualization association label in the original data integration strategy; based on the strategy association requirement vector corresponding to the selected data integration strategy, each of the reaction state structured vectors and the combustion chemical reaction kinetics vector, determine the second data integration confidence weight of the targeted elementary reaction data corresponding to at least one second set visualization association label in the selected data integration strategy; based on at least one first data integration confidence weight and at least one second data integration confidence weight corresponding to the several elementary reaction data, determine the visualization monitoring results of the several elementary reaction data in the original data integration strategy and the selected data integration strategy respectively.

[0005] An embodiment of the present invention provides a data processing system, comprising at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the above method.

[0006] An embodiment of the present invention provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above method are implemented.

[0007] In an embodiment of the present invention, by first obtaining the combustion chemical reaction kinetics vector and the primitive state linkage vector, the key information in the combustion process can be fully captured, laying the foundation for subsequent in-depth analysis. The primitive state linkage vector is structured semantically mapped to obtain the reaction state structured vector and the strategy association requirement vector, which helps to accurately classify and locate the information related to the combustion reaction, and improve the accuracy and pertinence of data processing. By determining the first and second data integration confidence weights, the degree of association between the primitive reaction data and the visual association labels under different integration strategies can be quantified, making the analysis of the combustion process more scientific and detailed. The final visual monitoring results can intuitively present the characteristics of the primitive reaction data under different strategies, providing clear and accurate data support for in-depth understanding of the combustion mechanism of ADN-based liquid propellants, optimizing propellant performance, and ensuring safe use. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 The present invention provides a flowchart of a method for processing combustion chemical reaction kinetics data for ADN-based liquid propellants in accordance with an embodiment of the present invention.

[0009] Figure 2 A structural schematic diagram of a data processing system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0010] Figure 1 A method for processing combustion chemical reaction kinetics data of ADN-based liquid propellants is shown, which is applied to a data processing system. The method includes the following steps 101 to 105.

[0011] The present invention mainly focuses on data processing during the combustion process of ADN-based liquid propellant. Through a series of complex operations, the visual monitoring results of elementary reaction data under different data integration strategies are determined. These results are of great significance for in-depth understanding of the combustion mechanism of ADN-based liquid propellant, optimizing performance and ensuring safe use.

[0012] Step 101: In response to a combustion process monitoring task, a data processing system obtains a combustion chemical reaction kinetic vector of an ADN-based liquid propellant corresponding to the combustion process monitoring task and primitive state linkage vectors corresponding to a plurality of primitive reaction data obtained according to the combustion process monitoring task.

[0013] First, the data processing system responds to the combustion process monitoring task. This combustion process monitoring task focuses on the combustion process of ADN-based liquid propellants. The combustion chemical reaction kinetic vector is a vector that comprehensively reflects the kinetic characteristics of various chemical reactions in the combustion process of ADN-based liquid propellants. For example, it may contain components related to the chemical reaction rate in the combustion process, such as the rate value of the ADN decomposition reaction at a specific temperature. For example, at 300K, the ADN decomposition reaction rate is 0.01 mol / (L·s), which may be used as an element in the kinetic vector. It may also contain components related to the change in the concentration of the reactant over time, such as the initial concentration of ADN is 1 mol / L. As the combustion reaction proceeds, the concentration gradually decreases, and information such as its concentration change rate will also be reflected in this vector.

[0014] There are many elementary reactions in the combustion process of ADN-based liquid propellant. Each elementary reaction data has a corresponding elementary state linkage vector. This vector covers various state information of the elementary reaction. For example, for the redox elementary reaction of ADN and another reactant, its elementary state linkage vector may contain the pressure information when the reaction occurs, such as some characteristics of the reaction when the pressure is 1atm, and may also contain the initial temperature information of the reaction system, such as the energy state of the reaction system at 250K. These elementary reaction data are many basic data obtained through monitoring and analysis of the combustion process, and they are the cornerstone of subsequent processing.

[0015] Step 102: The data processing system performs structured semantic mapping on the primitive state linkage vector of each primitive reaction data among the plurality of primitive reaction data to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event and strategy association requirement vectors corresponding to multiple types of set data integration strategies.

[0016] In this step, the data processing system performs structured semantic mapping on the primitive state linkage vector of each primitive reaction data. This process is similar to reorganizing and classifying the original primitive state linkage vector containing various messy information. For example, for a primitive reaction involving the decomposition of ADN and the production of nitrogen and oxygen, its original primitive state linkage vector may contain some numerical values ​​and state labels that are difficult to understand directly. Through structured semantic mapping, information related to the state of the reactant, such as the chemical bond breaking state of nitrogen atoms and oxygen atoms during the decomposition of ADN, is mapped to specific components in the reaction state structured vector. If this reaction is regarded as an event, then the reaction state structured vector is the state description of this event under a specific semantic structure.

[0017] At the same time, strategy-associated demand vectors corresponding to multiple types of set data integration strategies are also obtained. Data integration strategies may include integration according to reaction type, integration according to the temperature range in which the reaction occurs, etc. For example, the strategy-associated demand vector integrated according to reaction type may have different demand tags for different types of reactions such as redox reactions and decomposition reactions. For redox reactions, more attention may be paid to the state information related to electron transfer reflected in the primitive state linkage vector, while for decomposition reactions, more attention may be paid to the information such as the order of decomposition of the reaction substances represented in the vector. These demand vectors provide a basis for the subsequent determination of the confidence weight of data integration.

[0018] Step 103: The data processing system determines, based on the strategy association requirement vector corresponding to the original data integration strategy and each of the reaction state structured vectors, a first data integration confidence weight of the targeted primitive reaction data corresponding to at least one first set visualization association label in the original data integration strategy.

[0019] Further, the data processing system determines the weight based on the strategy-associated demand vector and each reaction state structured vector corresponding to the original data integration strategy. For example, the original data integration strategy is to integrate the primitive reaction data according to the temperature range. If the reaction state structured vector of a primitive reaction data indicates that the reaction mainly occurs in the high temperature range (e.g., 500-600K), and the strategy-associated demand vector corresponding to the high temperature range integration strategy has a high degree of attention to the reaction in this temperature range, then the first data integration confidence weight of this primitive reaction data corresponding to a first set visual association label (e.g., "high temperature reaction dominant") in the original data integration strategy may be high, for example, 0.8. If the reaction state structured vector of a primitive reaction data shows that its reaction temperature range does not match the key temperature range of the original data integration strategy, for example, the reaction is mainly in 300-400K, and the strategy focuses on 500-600K, then its corresponding first data integration confidence weight may be low, for example, 0.3. This confidence weight reflects the degree of association between the primitive reaction data and the specific visual association label under the original data integration strategy.

[0020] Step 104: The data processing system determines, based on the strategy association requirement vector corresponding to the selected data integration strategy, each of the reaction state structured vectors and the combustion chemical reaction kinetics vector, a second data integration confidence weight of the targeted elementary reaction data corresponding to at least one second set visualization association label in the selected data integration strategy.

[0021] Next, the data processing system determines the weight based on the strategy association requirement vector corresponding to the selected data integration strategy, each reaction state structured vector and the combustion chemical reaction kinetics vector. For example, the selected data integration strategy is to integrate according to the chemical activity of the reactants. For a primitive reaction data, if its reaction state structured vector shows that the reactant has a high chemical activity, such as ADN reacts with a highly active additive, and the combustion chemical reaction kinetics vector shows that this reaction has a large impact on the rate of the entire combustion process, and it is highly matched with the strategy association requirement vector integrated according to the chemical activity, then the second data integration confidence weight of this primitive reaction data corresponding to a second set visual association label (such as "high activity substance reaction key") in the selected data integration strategy may be very high, for example, 0.9. On the contrary, if the reaction state structured vector shows that the chemical activity of the reactant is low and the matching degree with the strategy association requirement vector is low, then the corresponding second data integration confidence weight may be low, such as 0.2. This weight reflects the closeness of the association between the primitive reaction data and the specific visual association label under the selected data integration strategy.

[0022] Step 105: The data processing system determines the visualization monitoring results of the plurality of elementary reaction data in the original data integration strategy and the selected data integration strategy, respectively, based on at least one first data integration confidence weight and at least one second data integration confidence weight respectively corresponding to the plurality of elementary reaction data.

[0023] Finally, the data processing system determines the visual monitoring results in the original data integration strategy and the selected data integration strategy based on at least one first data integration confidence weight and at least one second data integration confidence weight corresponding to the plurality of elementary reaction data. For example, for a set of elementary reaction data, if under the original data integration strategy (integration according to temperature range), the first data integration confidence weights corresponding to the first set visual association label of "high temperature reaction dominant" of multiple elementary reaction data are all high, then in the visual monitoring results, these elementary reaction data will be highlighted as the features related to high temperature reaction dominant in the visual display integrated according to temperature range (such as temperature-reaction rate relationship diagram). Similarly, under the selected data integration strategy (integration according to chemical activity), if the second data integration confidence weights corresponding to the second set visual association label of "high activity substance reaction key" of some elementary reaction data are high, then in the visual display integrated according to chemical activity (such as chemical activity-reaction contribution relationship diagram), these elementary reaction data will be highlighted as the parts related to high activity substance reaction key. In this way, the visualized monitoring results under different data integration strategies can clearly present the different characteristics and correlations of the elementary reaction data during the combustion process of ADN-based liquid propellants, providing an accurate and orderly data basis for in-depth research on combustion mechanisms, optimization of propellant performance, and safe use of ADN-based liquid propellants.

[0024] It can be seen that based on the above steps, the elementary reaction data in the combustion process of ADN-based liquid propellants can be deeply processed and analyzed, so as to obtain meaningful visual monitoring results under different data integration strategies, which provides important data support and decision-making basis for the research and application related to ADN-based liquid propellants.

[0025] In the data processing process related to the combustion of ADN-based liquid propellants, multiple vectors are involved. The following is an exemplary description of these vectors.

[0026] 1. Combustion Chemical Reaction Kinetics Vector

[0027] 1. Chemical reaction rate component: The ADN decomposition reaction rate is different at different temperatures. For example, at 300K, the ADN decomposition reaction rate is 0.01mol / (L·s); at 400K, the rate may become 0.05mol / (L·s); at 500K, the rate increases to 0.1mol / (L·s). These reaction rate values ​​at different temperatures together constitute the characteristic vector part of the chemical reaction rate dimension.

[0028] 2. Reactant concentration component: For example, the initial concentration of ADN is 1 mol / L, and as the combustion reaction proceeds, the concentration gradually decreases. In addition to ADN, there may be other reactants, such as the initial concentration of the fuel additive that reacts with ADN is 0.5 mol / L. During the reaction, the concentration of ADN changes to 0.8 mol / L at a certain moment, and the concentration of the additive changes to 0.3 mol / L. Moreover, the rates of change of concentrations of different reactants are also different. The rate of change of ADN concentration is -0.05 mol / (L·min) in a certain period of time, and the rate of change of additive concentration is -0.1 mol / (L·min). These values ​​together constitute the characteristic vector of the dimension of reactant concentration, which reflects the consumption of substances during the combustion reaction.

[0029] 3. Reaction heat effect component: The combustion reaction is accompanied by a heat effect. For the combustion of ADN-based liquid propellant, for example, the heat released in a certain reaction stage is 100 kJ / mol, while in another stage, the heat released is 150 kJ / mol due to different reaction degrees. At the same time, the heat release rate is also changing. For example, the heat release rate in the initial stage is 5 kJ / (mol·s), which may become 10 kJ / (mol·s) as the reaction proceeds. These values ​​of the reaction heat effect constitute another dimension of the combustion chemical reaction kinetic vector.

[0030] 2. Primitive State Linkage Vector

[0031] 1. Primitive state linkage vector related to redox primitive reactions

[0032] For the redox reaction of ADN with another reactant, the pressure information when the reaction occurs is one dimension. For example, some characteristics of the reaction when the pressure is 1atm will be reflected in the vector, and when the pressure becomes 2atm, the reaction will behave differently.

[0033] The initial temperature information of the reaction system is also a dimension. For example, the energy state of the reaction system at 250 K is different from that at 350 K. At 250 K, for example, the activation energy of the system is 50 kJ / mol, while at 350 K, the activation energy becomes 40 kJ / mol.

[0034] The molar ratio of the reactants is also a dimension. For example, the initial molar ratio of ADN to the reactants is 1:2. As the reaction proceeds, the molar ratio will change, such as to 1:1.5. These values ​​together constitute the primitive state linkage vector.

[0035] 3. Reaction state structured vector (obtained through structured semantic mapping)

[0036] 1. Reaction state structured vector of ADN decomposition reaction

[0037] For the elementary reaction of ADN decomposition to produce nitrogen and oxygen, the chemical bond breaking state of nitrogen and oxygen atoms during ADN decomposition can be described from multiple aspects. For example, the change in bond energy between nitrogen atoms and adjacent atoms is one dimension, such as from the initial 100 kJ / mol to 80 kJ / mol during the reaction; the degree of chemical bond polarization of oxygen atoms is another dimension, with an initial polarization rate of 0.1 and a change to 0.2 during the reaction.

[0038] The ratio of the generation rate of the reaction products is also a dimension. For example, the ratio of the nitrogen generation rate to the oxygen generation rate is initially 1:1.2, and changes to 1:1.5 as the reaction proceeds. These values ​​together constitute the reaction state structured vector, which reflects the representation of information related to the state of the reaction substances under a specific semantic structure.

[0039] 4. Strategy-related demand vector (related to data integration strategy)

[0040] 1. Strategy association demand vector integrated by reaction type (taking redox reaction and decomposition reaction as examples)

[0041] For the strategy-related demand vector integrated by reaction type, in terms of redox reactions, the number of electron transfers is one dimension. For example, when the number of electron transfers is 2, the corresponding weight value is 0.5. The direction of electron transfer is another dimension. If it is transferred from an atom in ADN to an atom in another reactant, the value 1 is used to represent this direction. If it is the opposite direction, -1 is used to represent it. The weight value corresponding to this direction information is 0.3.

[0042] In terms of decomposition reactions, the order of decomposition of the reactants is one dimension. For example, if ADN decomposes before another substance, it is represented by a value of 0 (indicating that it occurs first), and the corresponding weight value is 0.6; the type of decomposition products is another dimension. If there are three types of decomposition products, it is represented by a value of 3, and the corresponding weight value is 0.2. These values ​​together constitute the strategy association demand vector, which reflects the degree of attention paid to information related to different reaction types.

[0043] 5. First data integration confidence weight (under the original data integration strategy)

[0044] 1. Example of first data integration confidence weight by temperature range integration (raw data integration strategy)

[0045] For example, the original data integration strategy is to integrate the elementary reaction data according to the temperature range. If the reaction state structured vector of an elementary reaction data indicates that the reaction mainly occurs in the high temperature range (e.g., 500-600K), the strategy-associated demand vector corresponding to the high temperature range integration strategy has a higher degree of attention to the reaction in this temperature range.

[0046] From the temperature dimension, the weight value corresponding to this temperature range is 0.4; from the dimension of the change of reaction rate in this temperature range, the reaction rate increases rapidly with the increase of temperature, and the weight value corresponding to this feature is 0.3; from the characteristic dimension of the change of reaction substance concentration in this temperature range, the concentration change conforms to the expected pattern of high temperature reaction, and the corresponding weight value is 0.1. Combining the weights of these dimensions, the first data integration confidence weight of the primitive reaction data corresponding to a first set visualization association label (such as "high temperature reaction dominated") in the original data integration strategy is 0.8.

[0047] On the contrary, if the reaction state structured vector of a primitive reaction data shows that its reaction temperature range does not match the focus temperature range of the original data integration strategy, for example, the reaction is mainly at 300-400K, while the strategy focuses on 500-600K.

[0048] From the temperature dimension, the weight value corresponding to this temperature range is 0.1; from the dimension of the change of reaction rate in this temperature range, the reaction rate changes relatively smoothly, and the corresponding weight value is 0.1; from the characteristic dimension of the change of reaction substance concentration in this temperature range, the concentration change does not have the typical characteristics of high temperature reaction, and the corresponding weight value is 0.1. In summary, the corresponding first data integration confidence weight may be lower, such as 0.3. This value indicates the degree of association between the primitive reaction data and the specific visualization association label under the original data integration strategy.

[0049] 6. Second data integration confidence weight (under selected data integration strategy)

[0050] 1. Example of confidence weights for the second data integration based on the chemical activity of the reaction substances (selected data integration strategy)

[0051] For example, the selected data integration strategy is to integrate according to the chemical activity of the reactants. For a primitive reaction data, if its reaction state structured vector shows that the reactant has a high chemical activity, such as ADN reacting with a highly active additive.

[0052] From the perspective of the activity of the reactants, the weight value corresponding to high activity is 0.5; from the perspective of the combustion chemical reaction kinetic vector indicating the impact of this reaction on the rate of the entire combustion process, the reaction greatly increases the combustion rate, and the weight value corresponding to this feature is 0.3; from the perspective of the promotion of the reaction products to subsequent reactions, the reaction products can effectively promote other reactions, and the corresponding weight value is 0.1. Combining the weights of these dimensions, the second data integration confidence weight of the elementary reaction data corresponding to a second set visualization association label (such as "high activity substance reaction key") in the selected data integration strategy is 0.9.

[0053] On the contrary, if the reaction state structured vector shows that the chemical activity of the reaction substances is low and the matching degree with the strategy-related demand vector is low.

[0054] From the perspective of the activity of the reactants, the weight value corresponding to low activity is 0.1; from the perspective of the rate impact of the reaction on the entire combustion process indicated by the combustion chemical reaction kinetic vector, the reaction has little effect on the combustion rate, and the corresponding weight value is 0.1; from the perspective of the promotion effect of the reaction products on the subsequent reactions, the reaction products have no obvious promotion effect on the subsequent reactions, and the corresponding weight value is 0.0. In summary, the corresponding second data integration confidence weight may be low, such as 0.2. This value reflects the close association between the primitive reaction data and the specific visualization association label under the selected data integration strategy.

[0055] In some preferred embodiments, the primitive state linkage vector of the primitive reaction data is subjected to structured semantic mapping to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event and strategy association requirement vectors corresponding to multiple types of set data integration strategies, including: performing temporal semantic mapping on the primitive state linkage vector of the primitive reaction data to obtain a temporal reaction state structured vector and a temporal strategy association requirement vector; performing a first structural knowledge encoding on the temporal reaction state structured vector to obtain a reaction state structured vector corresponding to at least one type of set combustion chemical reaction tracking event; performing a second structural knowledge encoding on the temporal strategy association requirement vector to obtain a strategy association requirement vector corresponding to the original data integration strategy, and a strategy association requirement vector corresponding to the selected data integration strategy.

[0056] On the basis of the above-mentioned preferred embodiment, the first structural knowledge encoding is performed on the time-series reaction state structured vector to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event, including: performing multiple rounds of structural knowledge encoding on the time-series reaction state structured vector to obtain several first structured state knowledge encodings; determining the attention coefficients corresponding to the at least one type of set combustion chemical reaction tracking event; for each of the at least one type of set combustion chemical reaction tracking event, the several first structured state knowledge encodings are strengthened with the knowledge vector by the attention coefficient corresponding to the set combustion chemical reaction tracking event to obtain a reaction state structured vector corresponding to the set combustion chemical reaction tracking event.

[0057] On the basis of the above-mentioned preferred embodiment, the second structural knowledge encoding is performed on the timing strategy associated demand vector to obtain a strategy associated demand vector corresponding to the original data integration strategy and a strategy associated demand vector corresponding to the selected data integration strategy, including: performing multiple rounds of structural knowledge encoding on the timing strategy associated demand vector to obtain a number of second structured state knowledge encodings; respectively determining the attention coefficient corresponding to the original data integration strategy and the attention coefficient corresponding to the selected data integration strategy; strengthening the knowledge vectors of the several second structured state knowledge encodings through the attention coefficient corresponding to the original data integration strategy to obtain a strategy associated demand vector corresponding to the original data integration strategy; strengthening the knowledge vectors of the several second structured state knowledge encodings through the attention coefficient corresponding to the selected data integration strategy to obtain a strategy associated demand vector corresponding to the selected data integration strategy.

[0058] In the data processing of the ADN-based liquid propellant combustion process, there is an exemplary operation process for the structured semantic mapping of the primitive state linkage vector of the primitive reaction data. The first step is to perform temporal semantic mapping to obtain the temporal reaction state structured vector and the temporal strategy association requirement vector.

[0059] In this process, consider the primitive state linkage vector, such as the primitive state linkage vector of the redox primitive reaction of ADN and other substances mentioned earlier, which contains multi-dimensional information such as pressure (such as 1atm), initial temperature (such as 250K), and molar ratio of reactants (such as 1:2). When performing temporal semantic mapping on such primitive state linkage vectors, this information is reorganized from the perspective of time series. For example, as the combustion reaction proceeds, the pressure may fluctuate around 1atm at different times, the initial temperature of 250K may gradually increase, and the molar ratio of reactants will also change accordingly. The temporal reaction state structured vector and the temporal strategy association demand vector obtained through temporal semantic mapping are like organizing these dynamic changes in chronological order.

[0060] Then, the first structural knowledge encoding is performed on the sequential reaction state structured vector. This process is a multi-round structural knowledge encoding, for example, after three rounds of encoding. The first round of encoding may be encoding the information related to the concentration of the reactants in the sequential reaction state structured vector. For example, the initial ADN concentration is 1 mol / L. As time changes, the concentration change is converted into an exemplary encoding form according to certain rules in the first round of encoding; the second round of encoding may be for information related to the reaction rate, such as the ADN decomposition reaction rate of 0.01 mol / (L·s) at 300K, and its relationship with the rate at other temperatures is encoded; the third round of encoding is for information related to the reaction thermal effect, such as the case where the heat released at a certain reaction stage is 100 kJ / mol. After these multiple rounds of encoding, several first structural state knowledge encodings are obtained.

[0061] Then determine the attention coefficients corresponding to at least one type of set combustion chemical reaction tracking event. For example, for the combustion chemical reaction tracking event that ADN decomposes to produce nitrogen and oxygen, the attention coefficient is set to 0.6 according to its importance, and for the combustion chemical reaction tracking event that ADN reacts with additives to generate intermediate products, the attention coefficient is set to 0.4. For each set combustion chemical reaction tracking event, the knowledge vectors of several first structured state knowledge codes are strengthened by the corresponding attention coefficients. For example, for the event that ADN decomposes to produce nitrogen and oxygen, the attention coefficient of 0.6 is applied to several first structured state knowledge codes obtained previously, and the relevant codes such as the concentration of reactants, reaction rate, and reaction heat effect are weighted and strengthened, so as to obtain the reaction state structured vector corresponding to this event.

[0062] The processing of the timing strategy association requirement vector is similar. Multiple rounds of structural knowledge encoding are performed on it, for example, three rounds. The first round encodes the information related to the reaction type in the timing strategy association requirement vector, such as the number of electron transfers in the redox reaction; the second round encodes the information related to the range of conditions for the reaction, such as the temperature range (such as 300-400K) related to the strategy association requirements; the third round encodes the strategy association requirements related to the activity of the reactant. Several second structured state knowledge encodings are obtained.

[0063] The attention coefficients corresponding to the original data integration strategy and the selected data integration strategy are determined respectively. For example, for the original data integration strategy according to the temperature range, the attention coefficient is set to 0.7, and for the selected data integration strategy according to the chemical activity of the reaction material, the attention coefficient is set to 0.8. Then, the knowledge vector of several second structured state knowledge encodings is strengthened by the attention coefficient corresponding to the original data integration strategy to obtain the strategy association demand vector corresponding to the original data integration strategy. For example, in the process of strengthening the encoding of the original data integration strategy, the encoding parts such as reaction type, reaction condition range, and reaction material activity are weighted according to the attention coefficient of 0.7 to form the strategy association demand vector corresponding to the original data integration strategy. Similarly, the knowledge vector of several second structured state knowledge encodings is strengthened by the attention coefficient corresponding to the selected data integration strategy to obtain the strategy association demand vector corresponding to the selected data integration strategy. In this process, the encoding part is weighted according to the attention coefficient of 0.8.

[0064] In this way, in the entire data processing process, the reaction state structured vector related to the combustion chemical reaction tracking event and the strategy association requirement vector corresponding to different data integration strategies can be more accurately extracted from the primitive state linkage vector of the primitive reaction data. These vectors play a key role in the subsequent determination of data integration confidence weights and other operations. For example, when determining the first data integration confidence weight, based on the strategy association requirement vector corresponding to the original data integration strategy and each reaction state structured vector, if the reaction state structured vector indicates that the reaction is carried out within a specific temperature range (such as 500-600K), and the original data integration strategy focuses on this temperature range and has a higher weight in the strategy association requirement vector, then the first data integration confidence weight will be higher (such as 0.8). When determining the second data integration confidence weight, based on the strategy association demand vector corresponding to the selected data integration strategy, the structured vectors of each reaction state, and the combustion chemical reaction kinetics vector, if the reaction state structured vector shows that the chemical activity of the reactant is high, the selected data integration strategy focuses on the chemical activity and has a high weight in the strategy association demand vector, and the combustion chemical reaction kinetics vector shows that the reaction has a great influence on the combustion rate, then the second data integration confidence weight will be very high (such as 0.9). The final visualization monitoring results can accurately reflect the various characteristics of the ADN-based liquid propellant combustion process, whether in terms of reaction type, temperature range, and reaction material activity.

[0065] Therefore, by performing operations such as temporal semantic mapping, structural knowledge encoding, and knowledge vector enhancement on the primitive state linkage vector of the primitive reaction data, vectors related to combustion chemical reaction tracking events and data integration strategies can be extracted more accurately. This helps to improve the accuracy in determining the confidence weight of data integration, so that the visual monitoring results can more accurately reflect the various characteristics of the ADN-based liquid propellant combustion process, providing more reliable data support for in-depth research on combustion mechanisms, optimizing propellant performance, and safely using ADN-based liquid propellants.

[0066] In an exemplary design concept, the at least one type of set combustion chemical reaction tracking event corresponds to at least one first set visualization association label in the original data integration strategy respectively; the first data integration confidence weight of the primitive reaction data corresponding to at least one first set visualization association label in the original data integration strategy is determined based on the strategy association requirement vector corresponding to the original data integration strategy and each of the reaction state structured vectors, including: for each of the at least one first set visualization association label, the strategy association requirement vector corresponding to the original data integration strategy and the reaction state structured vector corresponding to the first set visualization association label are enhanced with knowledge vectors to obtain a first structured state knowledge encoding corresponding to the first set visualization association label; the first structured state knowledge encoding corresponding to the first set visualization association label is cyclically fully connected to obtain a first association label semantic vector corresponding to the first set visualization association label; and the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy is determined based on the first association label semantic vector corresponding to the first set visualization association label.

[0067] On the basis of this design idea, the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy is determined based on the first association label semantic vector corresponding to the first set visualization association label, including: determining the first target association label involved in the association with the first set visualization association label; the first target association label includes the first set visualization association label other than the first set visualization association label in the at least one first set visualization association label; performing first knowledge vector enhancement on the first association label semantic vector corresponding to the first target association label and the first association label semantic vector corresponding to the first set visualization association label to obtain the first association label enhanced semantic vector corresponding to the first set visualization association label; determining the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy based on the first association label enhanced semantic vector corresponding to the first set visualization association label.

[0068] Based on this design idea, the first target association label also includes a correlation tracking event indicator, and the correlation tracking event indicator corresponds to the correlation tracking event in the original data integration strategy; the first association label semantic vector corresponding to the correlation tracking event indicator is obtained by performing a cyclic full-connection processing on the reaction state structured vector corresponding to the correlation tracking event; the correlation tracking event includes the correlation relationship between the primitive reaction and the ambient temperature, and the correlation relationship between the primitive reaction and the pressure.

[0069] In the technical solution for processing ADN-based liquid propellant combustion data, a combustion chemical reaction tracking event is set to have a one-to-one correspondence with the first set visualization association label in the original data integration strategy.

[0070] When determining the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy, the operation flow is as follows. For each first set visualization association label, the knowledge vector enhancement operation is first performed. For example, if the original data integration strategy is integrated according to the temperature range, there is a first set visualization association label of "high temperature reaction dominant". The strategy association demand vector corresponding to the original data integration strategy may contain multi-dimensional weight information such as the temperature range, the change of reaction rate in the temperature range, and the characteristics of the change of reaction substance concentration in the temperature range. For example, the weight value corresponding to the temperature range is 0.4, the weight of the reaction rate change is 0.3, and the weight of the concentration change feature is 0.1. The reaction state structured vector corresponding to "high temperature reaction dominant" contains information such as the chemical bond breaking state of the reaction substance at high temperature (500-600K) and the ratio of the reaction product generation rate. The two are enhanced with knowledge vectors to obtain the first structured state knowledge encoding corresponding to "high temperature reaction dominant".

[0071] Then, the first structured state knowledge encoding is subjected to a cyclic full connection process. Taking "high temperature reaction dominance" as an example, the cyclic full connection process will synthesize the relationship between the various elements in the encoding, such as the potential relationship between the chemical bond breaking state of the reactant and the ratio of the reaction product generation rate, so as to obtain the first associated label semantic vector corresponding to "high temperature reaction dominance".

[0072] Then, the first data integration confidence weight is determined based on the first association label semantic vector. This process involves determining the first target association label that is associated with the first set visualization association label. All other first set visualization association labels except the original first set visualization association label may be the first target association label. For example, in the original data integration strategy integrated according to the temperature range, in addition to "high temperature reaction dominance", there may be labels such as "medium temperature reaction transition" and "low temperature reaction start". In addition, the first target association label also includes correlation tracking event indicators, which correspond to the correlation tracking events in the original data integration strategy, such as the correlation relationship between the primitive reaction and the ambient temperature, and the correlation relationship between the primitive reaction and the pressure. For the first association label semantic vector corresponding to the correlation tracking event indicator, it is obtained by performing a cyclic full connection process on the reaction state structured vector corresponding to the correlation tracking event. For example, for the correlation relationship between the primitive reaction and the ambient temperature, the reaction state structured vector contains information such as the reaction rate and the state of the reaction substance at different temperatures, and the corresponding first association label semantic vector is obtained through a cyclic full connection process.

[0073] Afterwards, the first associated label semantic vector corresponding to the first target associated label and the first associated label semantic vector corresponding to the first set visualization associated label (such as "high temperature reaction dominant") are reinforced with the first knowledge vector. For example, the first associated label semantic vector corresponding to "medium temperature reaction transition" contains coding elements related to information such as the activity of the reactant at medium temperature and the thermal effect of the reaction, and the first associated label semantic vector corresponding to "low temperature reaction initiation" contains coding elements related to information such as the initial state of the reactant at low temperature. These vectors are reinforced with the first associated label semantic vector corresponding to "high temperature reaction dominant" to obtain the first associated label reinforced semantic vector corresponding to "high temperature reaction dominant".

[0074] Finally, the first data integration confidence weight corresponding to "dominance of high temperature reaction" in the original data integration strategy is determined based on this first associated label enhanced semantic vector. If the elements in the first associated label enhanced semantic vector indicate that the reaction characteristics of the primitive reaction at high temperature are very obvious, and the fusion with the semantic vectors related to other temperature ranges further highlights the dominance of high temperature reaction, then the first data integration confidence weight may be very high, such as 0.8; if the elements in the vector show that the high temperature reaction characteristics are not significant or the relationship with other temperature ranges is not clear, then the first data integration confidence weight may be low, such as 0.3.

[0075] In this way, through a series of complex and interrelated operations, the first data integration confidence weight is determined from multiple perspectives. Operations such as knowledge vector enhancement, cyclic full connection processing, and consideration of multiple associated labels and correlation tracking events enable the determined first data integration confidence weight to more comprehensively and accurately reflect the degree of association between the primitive reaction data and the first set visual associated label under the original data integration strategy, providing a more accurate data processing basis for subsequent in-depth research, performance optimization, and safe use of the ADN-based liquid propellant combustion process.

[0076] In an alternative technical idea, the at least one type of set combustion chemical reaction tracking event corresponds one-to-one to at least one second set visualization association tag in the selected data integration strategy; the second data integration confidence weight of the targeted elementary reaction data corresponding to at least one second set visualization association tag in the selected data integration strategy is determined based on the strategy association requirement vector corresponding to the selected data integration strategy, each of the reaction state structured vectors and the combustion chemical reaction kinetic vector, including: for each of the at least one second set visualization association tag, the strategy association requirement vector corresponding to the selected data integration strategy, the reaction state structured vector corresponding to the targeted second set visualization association tag and the combustion chemical reaction kinetic vector are strengthened in the first round of knowledge vectors to obtain a basic second structured state knowledge encoding;

[0077] A second round of knowledge vector enhancement is performed based on the basic second structured state knowledge code and the combustion chemical reaction kinetics vector to obtain an advanced second structured state knowledge code; the advanced second structured state knowledge code is used as the basic second structured state knowledge code in the next round of knowledge vector enhancement; the second structured state knowledge code obtained by the last round of knowledge vector enhancement is used as the second associated label semantic vector corresponding to the second set visual associated label; based on the second associated label semantic vector corresponding to the second set visual associated label, the second data integration confidence weight of the targeted primitive reaction data corresponding to the targeted second set visual associated label in the selected data integration strategy is determined.

[0078] Further, the method of determining the second data integration confidence weight of the targeted primitive reaction data corresponding to the second set visualization association label in the selected data integration strategy based on the second association label semantic vector corresponding to the targeted second set visualization association label includes: determining the second target association label involved in the association with the targeted second set visualization association label; the second target association label includes the second set visualization association label other than the targeted second set visualization association label in the at least one second set visualization association label; performing knowledge vector enhancement on the second association label semantic vector corresponding to the second target association label and the second association label semantic vector corresponding to the targeted second set visualization association label to obtain the second association label enhanced semantic vector corresponding to the targeted second set visualization association label; determining the second data integration confidence weight of the targeted primitive reaction data corresponding to the targeted second set visualization association label in the selected data integration strategy based on the second association label enhanced semantic vector corresponding to the targeted second set visualization association label.

[0079] Under this technical idea, the combustion chemical reaction tracking events are set to correspond one-to-one with the second set visualization association labels in the selected data integration strategy. This correspondence constitutes the basic framework of the entire process of determining the second data integration confidence weight.

[0080] When determining the second data integration confidence weight of the primitive reaction data corresponding to the second set visualization association label in the selected data integration strategy, a series of complex and orderly operation processes are involved.

[0081] First, for each second-setting visualization association label, the first round of knowledge vector enhancement operation is performed. Taking the selected data integration strategy as an example, the integration is based on the chemical activity of the reaction substances. Each element under this strategy plays a key role in the entire data processing process. For example, if one of the second-setting visualization association labels is "key to the reaction of highly active substances", the comprehensive effect of multiple vectors needs to be considered. The strategy-associated demand vector corresponding to the selected data integration strategy contains a lot of key information, such as the weight information for judging the chemical activity of reactants. If the chemical activity of reactants is high, the corresponding weight value may be set to 0.5. This weight value reflects the importance of reactants with high chemical activity under this strategy. The reaction state structured vector also contains rich information, such as the chemical bond state of reactants. For example, the chemical bond energy of a certain reactant is 100 kJ / mol. This value reflects the stability of the chemical bond. The information such as the reaction product generation rate is also an important part of the reaction state structured vector. For example, the reaction product generation rate is 0.1 mol / (L·s). This rate value is closely related to the degree and efficiency of the reaction. The combustion chemical reaction kinetics vector contains core information such as reaction rate value and change in reactant concentration. For example, under specific conditions, the reaction rate is 0.05 mol / (L·s), and the initial concentration of the reactant is 1 mol / L, which changes to 0.8 mol / L as the reaction proceeds. These values ​​intuitively reflect the dynamic process of the reaction.

[0082] The three vectors mentioned above, namely the strategy-associated demand vector corresponding to the selected data integration strategy, the reaction state structured vector corresponding to the "key to high-activity substance reaction", and the combustion chemical reaction kinetics vector, are subjected to the first round of knowledge vector reinforcement. This reinforcement process is not a simple numerical addition or combination, but a deep fusion of the elements in these vectors based on exemplary algorithms and rules. For example, the chemical activity weight in the strategy-associated demand vector will be adjusted according to the chemical bond state in the reaction state structured vector. If the chemical bond energy is low (meaning that the chemical activity may be high), the weight may be further increased; at the same time, the reaction rate value and the change in the concentration of the reactant will also affect the weight ratio of the chemical activity of the reactant in the entire vector. Through such a complex fusion process, the basic second structured state knowledge encoding is obtained. This basic second structured state knowledge encoding is a new vector that integrates multiple vector information. The elements in it are no longer a simple list of the original vector elements, but the result of deep integration.

[0083] Then, the second round of knowledge vector reinforcement is carried out based on the basic second structured state knowledge encoding and the combustion chemical reaction kinetics vector, so as to obtain the advanced second structured state knowledge encoding. In this round of reinforcement process, the elements in the basic second structured state knowledge encoding obtained in the first round interact deeply with the combustion chemical reaction kinetics vector again. For example, the information related to the chemical activity of the reactants, the information related to the reaction product generation rate, and the partial integrated information of the reaction rate and concentration change in the basic second structured state knowledge encoding after the first round of reinforcement are reinforced again with the information such as the reaction rate value and the reaction substance concentration change in the combustion chemical reaction kinetics vector. In this process, the intrinsic connection between these information will be further explored. For example, the influence of the chemical activity of the reactants on the reaction rate and the change of the concentration of the reactants will be more accurately reflected in the advanced second structured state knowledge encoding. If the chemical activity of the reactants is judged to be high after the first round of reinforcement, in the second round of reinforcement, the effect of this high activity on the reaction rate and the influence on the reaction substance concentration reduction rate will be more carefully quantified and integrated into the advanced second structured state knowledge encoding. Moreover, this advanced second structured state knowledge encoding will serve as the basic second structured state knowledge encoding in the next round of knowledge vector reinforcement. If there are multiple rounds of reinforcement, each round will conduct deeper mining and integration of information based on the previous round, thereby continuously improving this comprehensive vector.

[0084] After completing multiple rounds of knowledge vector reinforcement, the second structured state knowledge obtained by the last round of knowledge vector reinforcement is encoded as the second associated label semantic vector corresponding to the second set visual associated label (such as "high-activity material reaction key"). This second associated label semantic vector is the result of multiple rounds of deep integration, and it contains information after comprehensive consideration from multiple aspects. From the perspective of the chemical activity of the reactant, it not only contains the initial judgment value of the chemical activity, but also contains comprehensive quantitative information on its impact on the reaction rate, the change in the concentration of the reactant, and the rate of reaction product generation; from the perspective of the reaction rate, it covers the change law of the reaction rate at different stages and the correlation between it and the chemical activity of the reactant; from the perspective of the change in the concentration of the reactant, it includes the initial concentration, the change trend, and the mutual influence relationship with other factors (such as the chemical activity of the reactant, the rate of reaction product generation, etc.). This vector forms a comprehensive semantic description for the visual associated label under the selected data integration strategy.

[0085] Subsequently, the second data integration confidence weight is determined based on the semantic vector of the second association label. This process first determines the second target association label that is associated with the second set visualization association label. In the selected data integration strategy, other second set visualization association labels except the second set visualization association label may become the second target association label. For example, in the selected data integration strategy for integration according to the chemical activity of the reactants, in addition to "high-activity material reaction is critical", there may also be labels such as "medium-activity material reaction is auxiliary" and "low-activity material reaction is secondary". These labels each correspond to different reaction characteristics and role positioning in the entire combustion process.

[0086] The second associated label semantic vectors corresponding to these second target associated labels and the second associated label semantic vectors corresponding to the second set visualization associated labels (such as "high-activity substance reaction key") are enhanced with knowledge vectors. Taking "medium-activity substance reaction auxiliary" as an example, its corresponding second associated label semantic vector contains coding elements related to information such as reaction heat effect and reaction substance state change during the reaction of medium-activity substance. For example, the reaction heat effect of medium-activity substance reaction is 50kJ / mol, and the reaction substance state change is manifested as a specific change in the molecular distance. These elements reflect the characteristics of medium-activity substance in the reaction; for the second associated label semantic vector corresponding to "low-activity substance reaction secondary", it contains coding elements related to information such as the relatively low reaction rate (such as 0.01mol / (L·s)) and the small number of reaction products during the reaction of low-activity substance. Strengthening these vectors with the second associated label semantic vector corresponding to "high-activity substance reaction key" is a complex information fusion process. In this process, the reaction characteristics corresponding to different active substances influence and correlate with each other. For example, the high reaction rate of highly active substances may affect the reaction heat effect of medium-active substances, and the small amount of reaction products of low-active substances may have a certain inhibitory or promoting effect on the reaction process of highly active substances. These relationships will be taken into account in the process of knowledge vector enhancement, thereby obtaining the second associated label enhanced semantic vector corresponding to "high-active substance reaction key".

[0087] Finally, the second data integration confidence weight corresponding to the "key to high-activity substance reaction" in the selected data integration strategy is determined based on the second associated label reinforced semantic vector. If the elements in the second associated label reinforced semantic vector indicate that the primitive reaction has a significant impact on the entire combustion process when the chemical activity of the reactant is high, and the criticality of the high-activity substance reaction is further highlighted after being fused with the semantic vectors related to other active substances, then the second data integration confidence weight may be very high, such as 0.9. This means that when the selected data integration strategy is integrated according to the chemical activity of the reactant, this primitive reaction is highly identified as a "key to high-activity substance reaction" reaction, and this primitive reaction will be the focus of subsequent data processing, analysis, and research related to the combustion process. On the contrary, if the elements in the vector show that the characteristics of the high-activity substance reaction are not significant or the relationship with other active substances is not clear, then the second data integration confidence weight may be low, such as 0.2. This shows that under the selected data integration strategy, the degree of association of this primitive reaction under the associated label of "key to high-activity substance reaction" is low, and its importance is relatively small in the overall data processing and analysis.

[0088] With this design, the second data integration confidence weight is determined through multiple rounds of knowledge vector reinforcement and consideration of multiple associated labels. The multiple rounds of reinforcement process enable the information in each vector to be continuously integrated and deepened, and can more comprehensively, meticulously and accurately reflect the degree of association between the elementary reaction data and the second set visual associated label under the selected data integration strategy. Considering multiple associated labels further optimizes the determination of this degree of association, and comprehensively considers the various relationships of the elementary reaction data in the entire combustion reaction system, thereby providing more accurate and reliable data processing results for the research, performance optimization and safe use of the ADN-based liquid propellant combustion process. This helps to deeply understand the combustion mechanism of ADN-based liquid propellants, can provide an accurate basis for the targeted optimization of propellant performance, and provides important data support for ensuring the safe use of ADN-based liquid propellants.

[0089] In another preferred embodiment, the visual monitoring results of the several elementary reaction data in the original data integration strategy and the selected data integration strategy are determined based on at least one first data integration confidence weight and at least one second data integration confidence weight respectively corresponding to the several elementary reaction data, including: for each of the several elementary reaction data, from at least one first data integration confidence weight and at least one second data integration confidence weight of the elementary reaction data, extracting at least one first initial data integration confidence weight of the elementary reaction data in the original data integration strategy and at least one second initial data integration confidence weight of the elementary reaction data in the selected data integration strategy; According to at least one first initial data integration confidence weight of the targeted elementary reaction data, weighted summation is performed to obtain a first target data integration confidence weight corresponding to the targeted elementary reaction data and the original data integration strategy; based on the first target data integration confidence weight, a first visual monitoring result of the targeted elementary reaction data in the original data integration strategy is generated; according to at least one second initial data integration confidence weight of the targeted elementary reaction data, weighted summation is performed to obtain a second target data integration confidence weight corresponding to the targeted elementary reaction data and the selected data integration strategy; based on the second target data integration confidence weight, a second visual monitoring result of the targeted elementary reaction data in the selected data integration strategy is generated.

[0090] In the technical solution system for ADN-based liquid propellant combustion data processing, for each of a number of elementary reaction data, the entire operation process is a step-by-step in-depth and highly information-integrated process. First, from at least one first data integration confidence weight and at least one second data integration confidence weight of the elementary reaction data, at least one first initial data integration confidence weight in the original data integration strategy and at least one second initial data integration confidence weight in the selected data integration strategy are extracted.

[0091] Taking the original data integration strategy of integration according to the temperature range as an example, consider a primitive reaction data. Under this strategy, there are multiple first data integration confidence weights associated with different first setting visualization association labels. For example, the first data integration confidence weight corresponding to the association label "high temperature reaction dominated" may be 0.8. The determination of this value is based on the results of a series of previous complex operations. For example, when processing the primitive state linkage vector of the primitive reaction data, if the reaction state structured vector indicates that the reaction rate increases significantly at high temperature (such as 500-600K), and the change in the concentration of the reaction substance at high temperature conforms to the expected pattern of "high temperature reaction dominated", and at the same time, the strategy association demand vector corresponding to "high temperature reaction dominated" in the original data integration strategy has a higher weight allocation to these features, these factors combined make the first data integration confidence weight corresponding to "high temperature reaction dominated" reach 0.8.

[0092] Let's look at the first data integration confidence weight corresponding to the associated label "medium-temperature reaction transition", which may be 0.3. For example, in the medium temperature range (such as 300-400K), the reaction rate changes relatively smoothly, and the concentration change of the reactant does not have the typical characteristics of high-temperature reactions. In addition, the strategy association demand vector related to the medium temperature range under the original data integration strategy assigns a relatively low weight to this situation, resulting in the first data integration confidence weight corresponding to "medium-temperature reaction transition" being 0.3. These values ​​such as 0.8 and 0.3 are the first initial data integration confidence weights extracted from multiple first data integration confidence weights of elementary reaction data.

[0093] For the selected data integration strategy, for example, integration is performed according to the chemical activity of the reactants. Similarly, for this elementary reaction data, the corresponding value is extracted from the second data integration confidence weight associated with different second set visualization association labels. For example, the second data integration confidence weight corresponding to the association label "key to high-activity substance reaction" may be 0.9. The formation of this value stems from the previous rounds of vector processing operations. If the reaction state structured vector shows that the reactant has a high chemical activity, such as the chemical bonds of the reactant are easy to break, and its chemical bond energy is relatively low, and the combustion chemical reaction kinetics vector shows that this reaction has a greater impact on the rate of the entire combustion process, and the strategy association requirement vector corresponding to "key to high-activity substance reaction" under the selected data integration strategy is highly matched to these features and assigned a higher weight, these combined situations make the second data integration confidence weight corresponding to "key to high-activity substance reaction" reach 0.9.

[0094] The second data integration confidence weight corresponding to the associated label "Medium-active substance reaction assistance" may be 0.2. For example, the reaction state structured vector shows that the chemical activity of the reactant is at a medium level, and the reaction has little effect on the combustion rate. In addition, the strategy-associated demand vector corresponding to "Medium-active substance reaction assistance" under the selected data integration strategy assigns a lower weight to this situation, resulting in the second data integration confidence weight corresponding to "Medium-active substance reaction assistance" being 0.2. These values ​​such as 0.9 and 0.2 are the second initial data integration confidence weights extracted from multiple second data integration confidence weights of the primitive reaction data.

[0095] Then, weight summation is performed based on at least one first initial data integration confidence weight of the targeted primitive reaction data to obtain the first target data integration confidence weight corresponding to the original data integration strategy. Taking the first initial data integration confidence weights of 0.8 and 0.3 extracted under the original data integration strategy as an example, the summation operation is performed to obtain 1.1. This summation process is not a simple numerical addition, but a comprehensive quantification of the confidence levels of the primitive reaction data related to different associated tags under the original data integration strategy. The significance of this first target data integration confidence weight is that it comprehensively reflects the overall confidence level of the primitive reaction data under the original data integration strategy.

[0096] Based on the first target data integration confidence weight, the first visualization monitoring result of the targeted elementary reaction data in the original data integration strategy is generated. If the first target data integration confidence weight is high, such as 1.1 (for example, when the overall weight value range is 0-2), then in the visualization monitoring result of the original data integration strategy, this elementary reaction data will be presented in a more prominent way in the relevant visualization display (such as the temperature-reaction rate relationship diagram) integrated according to the temperature range. For example, in the relationship diagram, the line representing this elementary reaction data may be drawn thicker and the color is more vivid and eye-catching to intuitively reflect its higher degree of association under the original data integration strategy. At the same time, if there are other visualization elements, such as the size of the data point, the shape of the marker, etc., they may also be adjusted accordingly according to the first target data integration confidence weight. For example, the size of the data point may be set larger, and the shape of the marker may be set to a special shape so that it is easier to be identified and distinguished among many elementary reaction data.

[0097] Then, the weighted sum is performed based on at least one second initial data integration confidence weight of the targeted elementary reaction data to obtain the second target data integration confidence weight corresponding to the selected data integration strategy. Taking the second initial data integration confidence weights of 0.9 and 0.2 extracted under the selected data integration strategy as an example, the sum is 1.1. This second target data integration confidence weight is also a quantitative representation of the comprehensive confidence level of the elementary reaction data under the selected data integration strategy.

[0098] Based on the second target data integration confidence weight, a second visualization monitoring result of the targeted elementary reaction data in the selected data integration strategy is generated. Similarly, if the second target data integration confidence weight is high, such as 1.1 (for example, when the overall weight value range is 0-2), this elementary reaction data will be highlighted in the visualization display of the chemical activity integration of the reactants (such as the chemical activity-reaction contribution relationship diagram). For example, in the relationship diagram, the area corresponding to this elementary reaction data may be filled with a special color or represented by a larger icon to highlight its importance under the selected data integration strategy. In addition, other visualization elements related to the elementary reaction data, such as the style of the connecting line, the display mode of the text annotation, etc., may also be adjusted according to the second target data integration confidence weight. For example, the connecting line may be bolded or use a different line type, and the text annotation may be displayed in a larger font or a more eye-catching color to better convey the key information of this elementary reaction data under the selected data integration strategy.

[0099] The above complex and orderly process is repeated for each elementary reaction data, and finally the visual monitoring results of all elementary reaction data in the original data integration strategy and the selected data integration strategy are obtained. These visual monitoring results can show the characteristics and correlations of elementary reaction data in the combustion process of ADN-based liquid propellant under different integration strategies in an intuitive, accurate and comprehensive way. Whether it is the influence of temperature range on the reaction or the contribution of chemical activity of reactants to the reaction, these visual monitoring results can be clearly presented.

[0100] It can be seen that the visualization monitoring results are determined by extracting and summing the confidence weights of the elementary reaction data under different strategies. This method enables the importance of each elementary reaction data in the original data integration strategy and the selected data integration strategy to be quantified and intuitively reflected. Accurately displaying its characteristics and correlations in the visualization monitoring results will help researchers understand the combustion process of ADN-based liquid propellants more deeply and meticulously. From a macro perspective, it provides a comprehensive data perspective for in-depth research on the combustion mechanism; from a micro perspective, it provides a clear and accurate data presentation method for optimizing propellant performance and ensuring the safe use of ADN-based liquid propellants, which helps to effectively analyze and manage ADN-based liquid propellants at different levels.

[0101] In an alternative embodiment, before obtaining the combustion chemical reaction kinetics vector of the ADN-based liquid propellant corresponding to the combustion process monitoring task and the primitive state linkage vectors respectively corresponding to the plurality of primitive reaction data obtained according to the combustion process monitoring task, the method further includes: in response to the combustion process monitoring task, obtaining the sensor information collected during the combustion process of the ADN-based liquid propellant corresponding to the combustion process monitoring task; querying in the signal data set according to the sensor information to obtain at least one primitive reaction data; after determining the visualization monitoring results of the plurality of primitive reaction data in the original data integration strategy and the selected data integration strategy respectively, the method further includes: performing a first visualization association output on each primitive reaction data using the visualization monitoring results of the plurality of primitive reaction data in the original data integration strategy; and performing a second visualization association output on each primitive reaction data using the visualization monitoring results of the plurality of primitive reaction data in the selected data integration strategy.

[0102] At the front end of the entire data processing process, that is, before obtaining the combustion chemical reaction kinetic vector and the primitive state linkage vector, there are a series of pre-operations. When the system responds to the combustion process monitoring task, it is crucial to obtain the sensor information collected during the combustion of ADN-based liquid propellant. The sources of these sensor information are rich and varied, and each type of sensor information carries important characteristic information related to the combustion process.

[0103] For example, the temperature sensor continues to work during the combustion process, recording the temperature values ​​at different times. In the initial stage of combustion, the temperature value that may be obtained is 300K, which reflects the thermal state at the beginning of combustion. As the combustion reaction progresses, the temperature will change, such as at a certain critical reaction moment, the temperature rises to 400K. These changes in temperature values ​​are not just simple digital records, they reflect the energy release during the combustion process and the intensity of the reaction. During the combustion process, the pressure sensor is also working synchronously to obtain pressure information. At the beginning of combustion, the pressure may be stable at 1atm, which is the initial pressure condition of the combustion reaction. As the reaction develops, due to factors such as the consumption of reactants and the generation of gaseous products, the pressure may change, for example, in an exemplary reaction process, the pressure becomes 1.5atm. This change in pressure is closely related to the stoichiometric relationship of the reaction, the characteristics of the reaction vessel, and the kinetic process of the reaction.

[0104] In addition, the component sensor is responsible for obtaining information on the concentration of reactants. For ADN-based liquid propellants, for example, the initial concentration of ADN is 1 mol / L, and this initial concentration sets the material basis for the combustion reaction. As the combustion reaction continues, ADN molecules participate in various chemical reactions, and their concentration gradually changes. For example, at a certain moment the concentration becomes 0.8 mol / L. This change in concentration reflects the consumption rate of ADN during the combustion process, and also implies the reaction of other substances that react with ADN. Moreover, in addition to ADN, there may be other reactants, such as oxidants or additives, whose concentration changes are also monitored and recorded by the component sensor. Information such as the changing trends in the concentrations of these different substances and their proportional relationships provide important basis for a deep understanding of the chemical process of the combustion reaction.

[0105] Based on these rich sensor information, querying in the signal data set to obtain at least one elementary reaction data involves complex query algorithms and rules. The signal data set is a large data set containing a lot of information related to the combustion process. For example, it may contain various data records under different combustion conditions and different reaction stages. When querying elementary reaction data, multiple factors need to be considered comprehensively. Taking the sensor information of temperature and pressure as an example, when the temperature increases from 300K to 400K and the pressure changes from 1atm to 1.5atm, this temperature-pressure change pattern may be associated with an exemplary elementary reaction. Combined with the change in the concentration of the reactant, such as the ADN concentration changes from 1mol / L to 0.8mol / L and the corresponding changes in the concentration of other reactants, search and match in the signal data set through an exemplary algorithm.

[0106] This query algorithm may be based on the basic principles of chemical reactions and pre-established models. For example, according to the theory of chemical reaction kinetics, under exemplary temperature and pressure conditions, the concentration changes of certain reaction substances conform to exemplary mathematical relationships. The query algorithm will use these theoretical relationships to filter out the elementary reaction data that meet the conditions in the signal data set. These elementary reaction data contain many important information, such as the precise conditions under which the reaction occurs (not only the values ​​of temperature and pressure, but also factors such as the rate of change of temperature and pressure), detailed information on the reaction substances (such as the types of substances involved in the reaction, their initial concentrations, and the concentration change trend during the reaction), etc. In addition, the elementary reaction data may also contain information on the direction of the reaction, that is, whether the reaction is forward or reverse, and information on the rate of the reaction. Although the formula is not introduced here, the rate information can be reflected by the change in the concentration of the reaction substance within a certain time interval.

[0107] After determining the visual monitoring results of several elementary reaction data in the original data integration strategy and the selected data integration strategy, the subsequent operations further enrich the connotation of data processing. The visual monitoring results of several elementary reaction data in the original data integration strategy are used to perform the first visual correlation output on each elementary reaction data.

[0108] Taking the original data integration strategy of integrating by temperature range as an example, under this strategy, different temperature ranges are set as different categories to analyze the characteristics of elementary reaction data. If a certain elementary reaction data is shown in the visual monitoring results to have a high degree of correlation in the high temperature range (such as 500-600K), then the corresponding first visual correlation output will be presented in a temperature-reaction rate relationship diagram in a carefully designed way. In this relationship diagram, the line color representing this elementary reaction data is set to red. This color is chosen to highlight its importance in the high temperature range, which is in sharp contrast to the line colors of other elementary reaction data that may have different degrees of correlation in different temperature ranges. For example, for those elementary reaction data with a high degree of correlation in the medium temperature range, their line color may be set to blue; while for those elementary reaction data with a high degree of correlation in the low temperature range, the line color may be green.

[0109] The thickness of the line is also an important visualization element. For the elementary reaction data with a high degree of correlation in the high temperature range, the line thickness is set to 2px, so that the relatively thick line is more visually eye-catching and can attract the attention of the observer. For the elementary reaction data with a relatively low degree of correlation, the line thickness may be 1px. This difference in thickness intuitively reflects the different importance of the elementary reaction data in the temperature-reaction rate relationship. In addition, the data point markings in the figure also have special significance. For this elementary reaction data with a high degree of correlation in the high temperature range, its data points are marked in a five-pointed star shape. This unique marking shape is different from the marking shapes of other elementary reaction data (such as circles, triangles, etc.), which further enhances the recognition of this elementary reaction data in the visualization results.

[0110] Similarly, the visualization monitoring results of several elementary reaction data in the selected data integration strategy are used to perform a second visualization correlation output on each elementary reaction data. For example, the selected data integration strategy is integrated according to the chemical activity of the reactants. Under this strategy, the chemical activity of the reactants is regarded as a key factor in analyzing the elementary reaction data. If a certain elementary reaction data is shown to be critically related to the reaction of a highly active substance in the visualization monitoring results, then the corresponding second visualization correlation output will have a unique presentation in the chemical activity-reaction contribution relationship diagram.

[0111] In this relationship diagram, the area corresponding to this elementary reaction data is filled with yellow. This bright color selection is to highlight the key position of this elementary reaction data in the reaction of highly active substances. In contrast, for elementary reaction data related to reactions of low-active substances or not playing a key role in the reaction, the corresponding area may be filled with white. White and yellow form a sharp contrast, allowing observers to quickly distinguish the differences in the chemical activity of different elementary reaction data in the reaction substances. The icon in the area is also an important visualization method. For elementary reaction data that are key to the reaction of highly active substances, the icon is an upward arrow. This upward arrow has multiple meanings. On the one hand, it indicates that the elementary reaction data plays a key role in the reaction of highly active substances. On the other hand, it can also imply that under this highly active reaction, certain indicators (such as reaction rate, amount of reaction product generated, etc.) are on an upward trend. For other elementary reaction data, the icon may be a downward arrow or no arrow. These different icon shapes intuitively reflect the different roles of elementary reaction data in the chemical activity-reaction contribution relationship of the reactant.

[0112] In addition, icon size is also an important visualization element. For this elementary reaction data that is critical to the reaction of highly active substances, the icon size is 20px×20px, so that the relatively large icon is more conspicuous in the figure. For elementary reaction data with a lower degree of relevance, the icon size may be 10px×10px. The difference in icon size further emphasizes the different importance of elementary reaction data under the selected data integration strategy.

[0113] In the entire data processing process, there are close logical connections and mutual influences between various links. Starting from the initial sensor information acquisition link, its accuracy and completeness are like the cornerstone of a building, which directly determines the reliability of subsequent data processing. If there is an error in the temperature sensor, such as the measured temperature value is higher or lower than the actual value, this may cause a deviation in the obtained temperature-pressure-concentration relationship. This deviation may make the query result inaccurate when querying the primitive reaction data based on the sensor information. For example, a primitive reaction that should have been identified as occurring at a specific temperature and pressure may not be correctly queried due to temperature errors, or other unrelated reactions may be mistakenly identified as the target primitive reaction.

[0114] When querying elementary reaction data, if the query algorithm is unreasonable or imperfect, it will also have a serious impact on the entire data processing process. For example, if the query algorithm does not fully consider the multiple possible modes of concentration changes of the reactants, some important elementary reaction data may be missed. Or, if the query algorithm is not properly set to be sensitive to temperature and pressure changes, it may misidentify elementary reaction data and include some reaction data that does not conform to the actual combustion process into the analysis scope.

[0115] The accuracy of the visual monitoring results is connected to the processing of elementary reaction data and the visual correlation output. If the visual monitoring results incorrectly evaluate the correlation degree of elementary reaction data under a certain integration strategy, for example, when the original data integration strategy is integrated according to the temperature range, it is wrongly judged that the correlation degree of a certain elementary reaction data in the high temperature range is low, while in fact the reaction plays a key role at high temperature, then in the first visual correlation output, the elementary reaction data cannot be correctly presented in the form of red lines, thicker lines, and five-pointed star shapes, etc., which leads to the observer's misunderstanding of the role of the elementary reaction data in the combustion process.

[0116] Similarly, under the selected data integration strategy, if the visual monitoring results incorrectly judge the degree of correlation between the elementary reaction data and the highly active substance reaction, then in the second visual correlation output, it will not be possible to accurately highlight the importance of the elementary reaction data through yellow fill areas, upward arrow icons, and larger icon sizes, which will affect the correct interpretation of the chemical activity-related mechanisms of the reactants in the entire combustion process.

[0117] Through such a complete, rigorous and interrelated data processing process, the data of the ADN-based liquid propellant combustion process can be processed and displayed comprehensively and deeply from multiple dimensions. Whether it is a detailed analysis of the combustion process based on a certain factor (such as temperature range) under the original data integration strategy, or an in-depth analysis of the combustion process based on another factor (such as the chemical activity of the reactant) under the selected data integration strategy, it can accurately and effectively present various relevant information of the elementary reaction data, providing solid data support and intuitive visualization basis for further studying the combustion mechanism of ADN-based liquid propellants, optimizing their performance and ensuring their safe use.

[0118] In combination with the above content, the ADN-based liquid propellant combustion data processing flow is comprehensively improved by adding sensor information acquisition and elementary reaction data query operations on the front end, and performing visual correlation output operations on the back end. The front-end operation ensures the accuracy of the data source and the validity of the elementary reaction data, so that subsequent analysis is based on reliable data. The visual correlation output operation on the back end presents the correlation relationship of the elementary reaction data under different integration strategies in an intuitive and accurate way, which helps to deeply understand the influence of different factors (such as temperature, chemical activity, etc.) on the reaction during the combustion process of ADN-based liquid propellants, thereby providing strong support for optimizing the performance of propellants, ensuring their safe use, and in-depth research on combustion mechanisms.

[0119] In an exemplary technical solution, the visualization monitoring results of the several primitive reaction data in the selected data integration strategy are used to perform a second visualization association output on each primitive reaction data, including: in response to a structure association event of the several primitive reaction data corresponding to the original data integration strategy, the primitive reaction data indicated by the structure association event is used as information to be visualized; the original data integration strategy is switched to the selected data integration strategy, and the information to be visualized is output according to the selected data integration strategy; in response to an event adjustment request for the selected data integration strategy, the visualization monitoring results of the several primitive reaction data in the selected data integration strategy are used to output subsequent primitive reaction data having a time-series relationship with the information to be visualized according to the selected data integration strategy.

[0120] In the above exemplary technical solution, for the second visual correlation output using the visual monitoring results of several elementary reaction data in the selected data integration strategy, first respond to the structural correlation event of several elementary reaction data corresponding to the original data integration strategy, and this structural correlation event is the trigger point of the entire process. Under the original data integration strategy, for example, the elementary reaction data are integrated according to the temperature range, and there is a certain structural correlation between the elementary reaction data. When this structural correlation event occurs, the elementary reaction data indicated therein is used as information to be visualized. For example, under the original data integration strategy, according to the temperature range, some elementary reaction data have an exemplary correlation relationship in the temperature range of 300-400K. When the structural correlation event for this part of the elementary reaction data is triggered, these elementary reaction data related to the temperature range of 300-400K become information to be visualized.

[0121] Then switch the original data integration strategy to the selected data integration strategy. Here, for example, the selected data integration strategy is to integrate according to the chemical activity of the reactants. After the switch, output the information to be visualized according to the selected data integration strategy. Under the strategy of integration according to the chemical activity of the reactants, the elementary reaction data that were previously used as information to be visualized (elementary reaction data related to the temperature range of 300-400K under the original data integration strategy) need to be reconsidered and output from the perspective of the chemical activity of the reactants. For example, some reactions in these elementary reaction data involve reactants with low chemical activity. Under the selected data integration strategy, they may be visualized and output in a relatively low-key manner, such as using smaller icons and lighter colors in the chemical activity-reaction contribution relationship diagram; while elementary reaction data involving highly chemically active reactants are presented in a more prominent manner, such as using larger icons and brighter colors.

[0122] Then, in response to the event adjustment request for the selected data integration strategy. This event adjustment request may be caused by the need for further analysis of the combustion process. For example, in the process of analyzing according to the chemical activity integration strategy of the reactants, it is found that a deeper understanding of the subsequent reactions of certain elementary reaction data under a specific time sequence is needed. Using the visual monitoring results of several elementary reaction data in the selected data integration strategy, the subsequent elementary reaction data with a time sequence connection with the information to be visualized is output according to the selected data integration strategy. For example, the previous information to be visualized is elementary reaction data related to a certain chemically active reactant. Under the selected data integration strategy, it is found through the visual monitoring results that, as time goes by, some subsequent elementary reaction data have a time sequence connection with the previous reaction. These subsequent elementary reaction data may be due to the intermediate products produced by the previous reaction participating in the new reaction. When outputting these subsequent elementary reaction data, the rules under the selected data integration strategy are also followed. If the chemical activity of the reactants in the subsequent elementary reaction data is higher, they will be more prominent in the visualization output, such as by making the lines representing these subsequent elementary reaction data thicker or changing their colors in the chemical activity-reaction contribution relationship diagram; if the chemical activity is lower, they will be presented in a less prominent manner.

[0123] Throughout the entire process, data processing and conversion are coherent and logical. Starting from the structural association events under the original data integration strategy, to switching strategies and outputting the information to be visualized according to the new strategy, and then to responding to event adjustment requests to output subsequent elementary reaction data with temporal connections, each link is closely connected. For example, the characteristics of the elementary reaction data involved in the elementary reaction data structural association events under the original data integration strategy will affect the output method under the selected data integration strategy. If these elementary reaction data show a higher reaction rate within the temperature range under the original data integration strategy, when switching to the selected data integration strategy, although the chemical activity is considered, this higher reaction rate may imply that the reactant has a higher chemical activity or is closely related to the highly chemically active substance in the subsequent reaction, which may be reflected in the visualization output under the selected data integration strategy.

[0124] Similarly, when outputting subsequent elementary reaction data, the foundation established by the previous information to be visualized also plays an important role. If the elementary reaction data represented by the previous information to be visualized is determined to have a certain chemical activity influence range under the selected data integration strategy, then the performance of the subsequent elementary reaction data with a temporal connection with it within this influence range will be reflected in the visualization output. For example, if the chemical activity of the previous elementary reaction data causes a certain impact on the reaction of a certain area, if the subsequent elementary reaction data reacts in this area or is associated with this impact, this connection will be displayed in an exemplary manner in the visualization output, such as through line connections or color gradients to represent this temporal and logical relationship.

[0125] Therefore, by responding to structural association events, switching data integration strategies, and responding to event adjustment requests to output elementary reaction data with time-series connections, effective conversion and in-depth analysis of elementary reaction data from the original data integration strategy to the selected data integration strategy are achieved. This method can more comprehensively display the association relationship of elementary reaction data under different strategies, taking into account the time series of data, and helps to more deeply understand the relationship between elementary reaction data under different conditions and time processes during the combustion of ADN-based liquid propellants, providing more detailed and comprehensive data display and analysis methods for combustion mechanism research, propellant performance optimization and safe use.

[0126] Based on the above steps 101 to 105, the combustion process monitoring task includes at least one of a chemical reaction rate monitoring task, a propellant vibration frequency monitoring task or a combustion environment monitoring task; the elementary reaction data includes at least one of chemical reaction state data, propellant vibration frequency state data or combustion environment state data.

[0127] In the technical solution related to ADN-based liquid propellant combustion data processing, based on the framework of steps 101-105, the combustion process monitoring tasks and elementary reaction data have exemplary scope definitions.

[0128] The combustion process monitoring task covers at least one of the chemical reaction rate monitoring task, the propellant vibration frequency monitoring task or the combustion environment monitoring task. The chemical reaction rate monitoring task is of key significance in the entire ADN-based liquid propellant combustion research. For example, during the combustion process, the chemical reaction rate will show different numerical characteristics. For the ADN decomposition reaction, the reaction rate may be 0.01 mol / (L·s) at a certain stage, while in another stage, due to changes in factors such as temperature, pressure or concentration of reactants, the reaction rate may increase to 0.05 mol / (L·s). By monitoring the chemical reaction rate, we can gain an in-depth understanding of the speed of each reaction in the combustion process, which is crucial for mastering the kinetic characteristics of the combustion reaction.

[0129] The task of monitoring the vibration frequency of the propellant is also indispensable. The vibration frequency of the propellant during the combustion process is an important manifestation of its physical properties. For example, in the early stage of combustion, the vibration frequency of the propellant is 100Hz. As the combustion reaction progresses, due to factors such as reaction heat and changes in material structure, the vibration frequency may change, such as to 120Hz. Different vibration frequencies reflect changes in the internal structure and energy state of the propellant, and this change is closely related to the progress of the combustion reaction.

[0130] The combustion environment monitoring task is also an important part of the combustion process monitoring task. The combustion environment involves multiple factors such as temperature, pressure, and gas composition. For example, at the beginning of combustion, the ambient temperature is 300K and the pressure is 1atm. As the combustion reaction proceeds, the temperature may rise to 500K and the pressure may increase to 1.5atm. At the same time, the gas composition in the combustion environment will also change. For example, during the combustion of ADN-based liquid propellant, the initial oxygen content is 21%. As the combustion reaction consumes oxygen and produces other gases, the oxygen content will gradually decrease, while the content of gases such as carbon dioxide and nitrogen will increase accordingly.

[0131] The elementary reaction data includes at least one of the chemical reaction state data, the propellant vibration frequency state data or the combustion environment state data. The chemical reaction state data contains a wealth of information. Taking the redox reaction between ADN and other substances as an example, the chemical reaction state data can reflect the concentration change of the reacting substances. For example, if the initial concentration of ADN is 1 mol / L, its concentration decreases to 0.8 mol / L during the reaction. At the same time, the concentration of the reaction product will increase accordingly. In addition, the energy state change of the reaction also belongs to the category of chemical reaction state data. At a certain reaction stage, the heat released by the reaction is 100 kJ / mol. This energy release value is closely related to the chemical state of the reaction.

[0132] The propellant vibration frequency state data directly reflects the vibration characteristics of the propellant during the combustion process. As mentioned earlier, the propellant vibration frequency will change in different combustion stages, and these different vibration frequency values ​​are the propellant vibration frequency state data. For example, in a certain intermediate combustion stage, the propellant vibration frequency is stable at 110Hz, and this value interacts with factors such as chemical reactions, temperature, and pressure during the combustion process. If the heat released by the chemical reaction increases, it may cause the propellant molecular motion to intensify, thereby increasing the vibration frequency; conversely, if the pressure decreases, the vibration frequency may decrease.

[0133] The combustion environment state data covers various states of environmental factors during the combustion process. For example, the temperature state data, the temperature change value from 300K to 500K records the dynamic process of the combustion environment temperature. In terms of pressure state data, the change from 1atm to 1.5atm reflects the change in the combustion environment pressure. In the gas composition state data, the value of the oxygen content gradually decreasing from 21% and the value of the gas content of carbon dioxide, nitrogen and other gases gradually increasing fully describe the changes in the gas composition in the combustion environment.

[0134] In the entire data processing flow, there is a close internal connection between the various aspects involved in the combustion process monitoring task and the different types of elementary reaction data. For example, the chemical reaction rate monitoring task is closely related to the chemical reaction state data. The change in reaction rate will directly affect the concentration change of the reactant, which is then reflected in the chemical reaction state data. When the reaction rate increases from 0.01mol / (L·s) to 0.05mol / (L·s), the concentration change rate of the reactant will also accelerate. For example, the process of ADN concentration decreasing from 1mol / L to 0.8mol / L will be faster. This relationship between the reaction rate and the concentration change of the reactant is reflected through the chemical reaction state data.

[0135] The propellant vibration frequency monitoring task is more directly related to the propellant vibration frequency status data. The monitored vibration frequency value is part of the propellant vibration frequency status data, and the changes in these data reflect the changes in the physical state of the propellant during the combustion process. Moreover, the propellant vibration frequency status data is also related to other types of data. For example, when the combustion environment temperature increases from 300K to 500K, the propellant vibration frequency may change from 100Hz to 120Hz, which indicates that changes in the combustion environment temperature will affect the vibration frequency status data of the propellant.

[0136] The same is true for the relationship between the combustion environment monitoring task and the combustion environment state data. The temperature, pressure, gas composition and other data obtained by the combustion environment monitoring task directly constitute the combustion environment state data. At the same time, changes in the combustion environment state data will affect the chemical reaction rate, propellant vibration frequency, etc. For example, when the combustion environment pressure increases from 1atm to 1.5atm, the chemical reaction rate may change from 0.01mol / (L·s) to 0.02mol / (L·s), which reflects the interaction between the combustion environment state data and the chemical reaction rate.

[0137] The complex interrelationship between the combustion process monitoring task and the elementary reaction data plays an important role in the entire ADN-based liquid propellant combustion data processing process. Through the in-depth analysis and processing of these relationships, a more comprehensive and in-depth understanding of the combustion mechanism of ADN-based liquid propellants can be achieved.

[0138] The above examples define the specific scope of the combustion process monitoring tasks and elementary reaction data, and reveal the complex internal relationship between them. This helps to more comprehensively acquire and analyze the data of the ADN-based liquid propellant combustion process, deeply understand the mutual influence between the chemical reactions in the combustion process, the physical properties of the propellant, and the changes in the combustion environment, and provides a richer and more accurate data basis for studying the combustion mechanism, optimizing the propellant performance, and ensuring safe use.

[0139] In summary, the present invention first obtains the combustion chemical reaction kinetic vector and the primitive state linkage vector, which can comprehensively capture the key information in the combustion process and lay the foundation for subsequent in-depth analysis. The primitive state linkage vector is structured semantically mapped to obtain the reaction state structured vector and the strategy association requirement vector, which helps to accurately classify and locate the information related to the combustion reaction, and improve the accuracy and pertinence of data processing. By determining the first and second data integration confidence weights, the degree of association between the primitive reaction data and the visual association labels under different integration strategies can be quantified, making the analysis of the combustion process more scientific and meticulous. The final visual monitoring results can intuitively present the characteristics of the primitive reaction data under different strategies, providing clear and accurate data support for in-depth understanding of the combustion mechanism of ADN-based liquid propellants, optimizing propellant performance, and ensuring safe use.

[0140] Further, Figure 2 FIG. 2 is a schematic diagram of a data processing system 200 provided in an embodiment of the present invention. Figure 2 The data processing system 200 shown includes a processor 210, which can call and run a computer program from a memory to implement the method in the embodiment of the present invention. Figure 2As shown, the data processing system 200 may further include a memory 230. The processor 210 may call and run a computer program from the memory 230 to implement the method in the embodiment of the present invention. The memory 230 may be a separate device independent of the processor 210, or may be integrated in the processor 210. Optionally, as Figure 2 As shown, the data processing system 200 may also include a transceiver 220, and the processor 210 may control the transceiver 220 to interact with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices. Optionally, the data processing system 200 may implement the corresponding processes corresponding to the storage engine or the components (such as processing modules) in the storage engine or the device deployed with the storage engine in each method of the embodiment of the present invention, which will not be described in detail for the sake of brevity.

[0141] Based on the above, a readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the above method are implemented.

[0142] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose of the present invention and the scope of protection of the present invention, all of which are within the protection of the present invention.

Claims

1. A method for processing combustion chemical reaction kinetics data of ADN-based liquid propellants, characterized in that: The method is applied to a data processing system, and the method comprises: In response to a combustion process monitoring task, obtaining a combustion chemical reaction kinetic vector of an ADN-based liquid propellant corresponding to the combustion process monitoring task and primitive state linkage vectors corresponding to a plurality of primitive reaction data obtained according to the combustion process monitoring task; Performing structured semantic mapping on the primitive state linkage vector of each primitive reaction data of the plurality of primitive reaction data to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event and strategy association requirement vectors corresponding to multiple types of set data integration strategies; Based on the strategy association requirement vector corresponding to the original data integration strategy and each of the reaction state structured vectors, determining a first data integration confidence weight of the targeted primitive reaction data corresponding to at least one first set visualization association label in the original data integration strategy; Based on the strategy association requirement vector corresponding to the selected data integration strategy, each of the reaction state structured vectors and the combustion chemical reaction kinetics vector, determining a second data integration confidence weight corresponding to at least one second set visualization association label in the selected data integration strategy for the targeted elementary reaction data; Based on at least one first data integration confidence weight and at least one second data integration confidence weight respectively corresponding to the plurality of elementary reaction data, visualization monitoring results of the plurality of elementary reaction data in the original data integration strategy and the selected data integration strategy are determined.

2. The method according to claim 1, characterized in that The structured semantic mapping is performed on the primitive state linkage vector of the primitive reaction data to obtain the reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event and the strategy association requirement vectors corresponding to multiple types of set data integration strategies, including: Performing temporal semantic mapping on the primitive state linkage vector of the targeted primitive reaction data to obtain a temporal reaction state structured vector and a temporal strategy associated demand vector; Performing a first structural knowledge encoding on the time-series reaction state structured vector to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event; The timing strategy associated requirement vector is subjected to second structural knowledge encoding to obtain a strategy associated requirement vector corresponding to the original data integration strategy and a strategy associated requirement vector corresponding to the selected data integration strategy.

3. The method according to claim 2, characterized in that The first structural knowledge encoding is performed on the time-series reaction state structured vector to obtain reaction state structured vectors corresponding to at least one type of set combustion chemical reaction tracking event, including: Performing multiple rounds of structural knowledge encoding on the temporal reaction state structured vector to obtain a plurality of first structural state knowledge encodings; Determining attention coefficients corresponding to the at least one type of set combustion chemical reaction tracking events; For each of the at least one type of set combustion chemical reaction tracking events, the several first structured state knowledge encodings are enhanced by the attention coefficient corresponding to the set combustion chemical reaction tracking event to obtain a reaction state structured vector corresponding to the set combustion chemical reaction tracking event.

4. The method according to claim 2, characterized in that The step of performing a second structural knowledge encoding on the timing strategy associated requirement vector to obtain a strategy associated requirement vector corresponding to the original data integration strategy and a strategy associated requirement vector corresponding to the selected data integration strategy includes: Performing multiple rounds of structural knowledge encoding on the timing strategy associated demand vector to obtain a plurality of second structural state knowledge encodings; respectively determining an attention coefficient corresponding to the original data integration strategy and an attention coefficient corresponding to the selected data integration strategy; By using the attention coefficient corresponding to the original data integration strategy, the plurality of second structured state knowledge encodings are enhanced by knowledge vectors to obtain a strategy-associated demand vector corresponding to the original data integration strategy; The plurality of second structured state knowledge encodings are reinforced with knowledge vectors through the attention coefficient corresponding to the selected data integration strategy, so as to obtain a strategy-associated requirement vector corresponding to the selected data integration strategy.

5. The method according to claim 1, characterized in that The at least one type of set combustion chemical reaction tracking event corresponds to at least one first set visualization association tag in the original data integration strategy respectively; the first data integration confidence weight corresponding to at least one first set visualization association tag in the original data integration strategy for the targeted primitive reaction data is determined based on the strategy association requirement vector corresponding to the original data integration strategy and each of the reaction state structured vectors, including: For each of the at least one first setting visualization association tag, the strategy association requirement vector corresponding to the original data integration strategy and the reaction state structured vector corresponding to the first setting visualization association tag are enhanced with knowledge vectors to obtain a first structured state knowledge encoding corresponding to the first setting visualization association tag; Performing a cyclic full-connection process on the first structured state knowledge encoding corresponding to the first set visualization association label to obtain a first association label semantic vector corresponding to the first set visualization association label; Determining, according to the first associated label semantic vector corresponding to the first set visual associated label, a first data integration confidence weight of the targeted primitive reaction data corresponding to the targeted first set visual associated label in the original data integration strategy; The method of determining the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy based on the first association label semantic vector corresponding to the first set visualization association label includes: determining the first target association label involved in the association with the first set visualization association label; the first target association label includes the first set visualization association label other than the first set visualization association label in the at least one first set visualization association label; performing first knowledge vector enhancement on the first association label semantic vector corresponding to the first target association label and the first association label semantic vector corresponding to the first set visualization association label to obtain the first association label enhancement semantic vector corresponding to the first set visualization association label; determining the first data integration confidence weight of the primitive reaction data corresponding to the first set visualization association label in the original data integration strategy based on the first association label enhancement semantic vector corresponding to the first set visualization association label; The first target association tag also includes a correlation tracking event indicator, which corresponds to the correlation tracking event in the original data integration strategy; the first association tag semantic vector corresponding to the correlation tracking event indicator is obtained by performing a cyclic full-connection processing on the reaction state structured vector corresponding to the correlation tracking event; the correlation tracking event includes the correlation relationship between the primitive reaction and the ambient temperature, and the correlation relationship between the primitive reaction and the pressure.

6. The method according to claim 1, characterized in that The at least one type of set combustion chemical reaction tracking event corresponds to at least one second set visualization association tag in the selected data integration strategy respectively; the second data integration confidence weight corresponding to at least one second set visualization association tag in the selected data integration strategy respectively for the targeted elementary reaction data based on the strategy association requirement vector corresponding to the selected data integration strategy, each of the reaction state structured vectors and the combustion chemical reaction kinetics vector comprises: For each of the at least one second setting visualization association tag, a strategy association requirement vector corresponding to the selected data integration strategy, a reaction state structured vector corresponding to the second setting visualization association tag, and the combustion chemical reaction kinetics vector are subjected to a first round of knowledge vector reinforcement to obtain a basic second structured state knowledge encoding; A second round of knowledge vector enhancement is performed based on the basic second structured state knowledge code and the combustion chemical reaction kinetics vector to obtain an advanced second structured state knowledge code; the advanced second structured state knowledge code is used as the basic second structured state knowledge code in the next round of knowledge vector enhancement; The second structured state knowledge encoding obtained by strengthening the last round of knowledge vectors is used as a second association label semantic vector corresponding to the second set visualization association label; Determining, according to the second associated label semantic vector corresponding to the second set visual associated label, a second data integration confidence weight of the targeted primitive reaction data corresponding to the targeted second set visual associated label in the selected data integration strategy; The determining, based on the second associated label semantic vector corresponding to the second set visual associated label, of the second data integration confidence weight of the primitive reaction data corresponding to the second set visual associated label in the selected data integration strategy comprises: Determine a second target association tag associated with the second set visualization association tag; the second target association tag includes the second set visualization association tag other than the second set visualization association tag in the at least one second set visualization association tag; Performing knowledge vector enhancement on the second association label semantic vector corresponding to the second target association label and the second association label semantic vector corresponding to the second set visualization association label, so as to obtain the second association label enhanced semantic vector corresponding to the second set visualization association label; According to the second association tag enhanced semantic vector corresponding to the second set visualization association tag, a second data integration confidence weight corresponding to the second set visualization association tag in the selected data integration strategy is determined for the primitive reaction data.

7. The method according to claim 1, characterized in that The determining, based on at least one first data integration confidence weight and at least one second data integration confidence weight respectively corresponding to the plurality of elementary reaction data, the visualized monitoring results of the plurality of elementary reaction data in the original data integration strategy and the selected data integration strategy respectively, comprises: For each of the plurality of elementary reaction data, extracting at least one first initial data integration confidence weight of the elementary reaction data in the original data integration strategy and at least one second initial data integration confidence weight of the elementary reaction data in the selected data integration strategy from at least one first data integration confidence weight and at least one second data integration confidence weight of the elementary reaction data; Performing weight summation according to at least one first initial data integration confidence weight of the targeted elementary reaction data to obtain a first target data integration confidence weight corresponding to the targeted elementary reaction data and the original data integration strategy; Based on the first target data integration confidence weight, generating a first visual monitoring result of the targeted primitive reaction data in the original data integration strategy; Performing weight summation according to at least one second initial data integration confidence weight of the targeted elementary reaction data to obtain a second target data integration confidence weight corresponding to the targeted elementary reaction data and the selected data integration strategy; Based on the second target data integration confidence weight, a second visualization monitoring result of the targeted primitive reaction data in the selected data integration strategy is generated.

8. The method according to claim 1, characterized in that Before obtaining the combustion chemical reaction kinetic vector of the ADN-based liquid propellant corresponding to the combustion process monitoring task and the primitive state linkage vectors corresponding to a plurality of primitive reaction data obtained according to the combustion process monitoring task, the method further includes: In response to the combustion process monitoring task, acquiring sensor information collected during the combustion process of the ADN-based liquid propellant corresponding to the combustion process monitoring task; querying in a signal data set according to the sensing information to obtain at least one elementary reaction data; After determining the visualized monitoring results of the plurality of elementary reaction data in the original data integration strategy and the selected data integration strategy respectively, the method further comprises: Using the visualized monitoring results of the plurality of elementary reaction data in the original data integration strategy, a first visualized correlation output is performed on each elementary reaction data; Using the visual monitoring results of the plurality of elementary reaction data in the selected data integration strategy, a second visual correlation output is performed on each elementary reaction data; The step of utilizing the visualized monitoring results of the plurality of elementary reaction data in the selected data integration strategy to perform a second visualized correlation output on each elementary reaction data comprises: In response to a structure association event of a plurality of primitive reaction data corresponding to the original data integration strategy, the primitive reaction data indicated by the structure association event is used as information to be visualized; Switching the original data integration strategy to the selected data integration strategy, and outputting the information to be visualized according to the selected data integration strategy; In response to an event adjustment request for the selected data integration strategy, the visualization monitoring results of the plurality of primitive reaction data in the selected data integration strategy are utilized, and subsequent primitive reaction data having a time-series relationship with the information to be visualized are output according to the selected data integration strategy.

9. The method according to any one of claims 1 to 8, characterized in that The combustion process monitoring task includes at least one of a chemical reaction rate monitoring task, a propellant vibration frequency monitoring task or a combustion environment monitoring task; the elementary reaction data includes at least one of chemical reaction state data, propellant vibration frequency state data or combustion environment state data.

10. A data processing system, characterized in that: The method comprises at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the method according to any one of claims 1 to 9.

Images