A dynamic forward modeling method and system for carbon release scenarios oriented to time series

Abstract

The present invention belongs to the field of ecosystem carbon cycle, and specifically relates to a dynamic forward modeling method and system for carbon release scenarios for time series, which aims to solve the problems that the existing carbon release scenario forward modeling cannot capture detailed changes, cannot identify the mixed release of multiple carbon sources, and the number of forward modeling times is limited. The method of the present invention includes: selecting a target curve and determining potential carbon sources; identifying potential fluctuations of the target curve and dividing the stages; generating a carbon release pattern in each stage based on a random sampling method and inputting it into a carbon cycle model for simulation; error calculation and screening of the optimal carbon release pattern. The present invention can efficiently generate a variety of carbon release patterns, expand the coverage of extreme conditions and complex carbon release scenarios, and can achieve high-precision simulation and further optimize the dynamic evolution process of complex carbon release scenarios.

Description

Background Art

[0002] In geological history, there have been many extreme carbon release events, such as the Permian-Triassic Extinction Event, the Paleocene-Eocene Thermal Maximum (PETM), and the Toarcian Oceanic Anoxic Event (T-OAE) in the Early Jurassic. These events were accompanied by large-scale carbon release and climate warming, with obvious global environmental effects. Studying these historical events can not only deepen our understanding of the carbon release mechanism, but also provide a scientific basis for predicting climate change caused by modern anthropogenic carbon emissions.

[0003] Existing technologies mainly explore the specific scenarios (carbon release patterns / scenarios) of past carbon release events by combining with carbon cycle models. Carbon cycle models, such as LOSCAR (Long-term Ocean Sediment CArbon Reservoir Model) and COPSE (Carbon-Oxygen-Phosphorus-Sulfur-Evolution Model), have been widely used in carbon cycle and climate change simulations in geological history. These models derive proxy indicators at that time by setting boundary conditions for a specific geological period and combining numerical simulation methods, and further reflect the potential process of carbon release.

[0004] The core of the carbon cycle model is to simulate the impact of carbon release on the carbon cycle and climate system by adjusting key parameters, such as carbon release amount, release rate, and carbon isotope composition. This method usually adopts the forward simulation method, that is, starting from the known input conditions, the output results are gradually derived. Although the existing carbon cycle model plays an important role in reproducing the carbon release scenarios in geological history, it still has shortcomings in some aspects and there is room for further optimization. The specific analysis is as follows:

[0005] 1. Forward modeling relies on multiple simulations to explore different parameter combinations, but due to limited computing resources, the number and scope of model simulations are limited. This is the result of a trade-off between computational efficiency and resource cost in traditional model design. Although increasing computing power can alleviate this problem, it is still difficult to completely solve the lack of coverage of extreme conditions or rare events.

[0006] 2. The focus of long-time scale simulation is to reveal the overall trend, and the ability to describe short-term, high-frequency changes is weak. The design of the time resolution of existing models often matches the research objectives and tends to capture large-scale environmental changes, while the simulation of detailed changes requires more data support and higher calculation accuracy, which is not a priority when resources are limited.

[0007] 3. Existing models usually adopt a single or limited carbon source release mode in the assumption of carbon release scenarios, such as assuming that volcanic activity, methane hydrate collapse or a single event in the biosphere carbon cycle is dominant. However, carbon release events in geological history often involve mixed releases of multiple carbon sources, and the timing, intensity and isotopic characteristics of these carbon sources may vary significantly. Traditional models find it difficult to simultaneously analyze and simulate the combined effects of multiple carbon sources, resulting in insufficient reproducibility of complex carbon release processes. This is because existing models lack flexibility in parameter setting, and a single forward modeling cannot capture the dynamic interactive characteristics of mixed carbon source release. Summary of the invention

[0008] In order to solve the above problems in the prior art, namely, the existing carbon cycle model cannot grasp the detailed changes in the forward modeling of carbon release scenarios, cannot identify the mixed release of multiple carbon sources, and the number of forward modeling is limited, the first aspect of the present invention proposes a dynamic forward modeling method for carbon release scenarios oriented to time series, which is used for the dynamic forward modeling of carbon release event processes in geological history to obtain the optimal carbon release mode. The method comprises the following steps:

[0009] Step S1, selecting a target curve of a carbon release event to be dynamically forward modeled and identifying the corresponding potential carbon source type; determining the release intensity, rate and isotopic composition of the potential carbon source type;

[0010] Step S2, identifying the curve characteristics of the target curve through an automatic recognition algorithm and a set secondary verification method, analyzing the timing information of the carbon release event, and then dividing the target curve into multiple stages; the curves of each stage have clear boundaries and the characteristics of each stage are consistent with the actual change trend of the target curve;

[0011] Step S3: randomly generate multiple groups of carbon release patterns in each stage curve based on a random sampling method, and input them into the carbon cycle model for numerical simulation one by one:

[0012] Each group of carbon release patterns is composed of multiple parameters, including carbon source type, contribution ratio of carbon source type, release rate of carbon source type and isotope of carbon source; in the numerical simulation process, the evolution process of the carbon release event is dynamically simulated by adjusting the parameters and release pattern of each carbon source;

[0013] Step S4: Compare the carbon release curves generated by the numerical simulation with the target curves respectively, obtain the errors between any two curves, and determine the final carbon release mode according to the errors.

[0014] In some preferred embodiments, the target curve includes a carbon isotope curve, an atmospheric carbon dioxide concentration curve, and a temperature change curve.

[0015] The random sampling method is any one of the Monte Carlo method, Latin hypercube sampling or Bayesian MCMC method.

[0016] In some preferred embodiments, in step S3, multiple groups of combinations are randomly selected in the parameter space according to the characteristics of the target curve and the type of potential carbon source at each stage using the Monte Carlo method to randomly generate multiple groups of carbon release patterns.

[0017] In some preferred embodiments, when a carbon cycle model is input for simulation, different carbon cycle types are selected according to a set carbon release scenario.

[0018] In some preferred embodiments, when the selected target curve is a carbon isotope curve, the carbon cycle model is centered on the carbon isotope mass balance equation to describe the dynamic changes of carbon isotopes in the system. The carbon isotope mass balance equation is expressed as:

[0019] ;

[0020] in, represents the carbon isotope ratio at the next moment, is the ratio at the current moment, Indicates The release rate of a carbon source, For the The carbon isotopic composition of each carbon source, is the total amount of carbon in the system.

[0021] In some preferred embodiments, in step S4, the error between the simulated carbon isotope curve and the target carbon isotope curve is calculated, and the error values ​​are sorted from small to large, and the carbon release pattern of the carbon isotope curve with an error value less than a set threshold is selected as the final carbon release scenario.

[0022] In some preferred embodiments, the final carbon release scenario may be one or more.

[0023] In some preferred embodiments, when quantifying the model fitting effect, the fitting degree of the simulation results is evaluated by calculating the root mean square error, and the error calculation formula is as follows:

[0024] ;

[0025] Among them, RMSE means root mean square error, represents the carbon isotope value simulated by the model, is the actual carbon isotope value at the corresponding time point in the target curve, is the total number of time points.

[0026] In a second aspect of the present invention, a dynamic forward modeling system for carbon release scenarios oriented to a time series is proposed, the system comprising:

[0027] a feature selection module configured to select a target curve and determine the type of potential carbon source corresponding thereto, and determine the intensity, rate and isotopic composition of its release;

[0028] A feature recognition module is configured to recognize the curve features of the target curve, analyze and extract the timing information of the carbon release event, and divide the target curve into stages;

[0029] A parameter generation module is configured to randomly generate multiple groups of carbon release patterns in each stage based on a random sampling method for the target curve after the stage division, so as to obtain multiple carbon release patterns;

[0030] The carbon cycle simulation module is configured to numerically simulate the carbon release patterns one by one, adjust the parameters and release patterns of each carbon source, and dynamically simulate the evolution of carbon isotopes in the system;

[0031] The error screening module is configured to compare the carbon release curve generated by the numerical simulation with the target curve, screen the error between the simulation and the target results, and determine the optimal carbon release mode.

[0032] According to a third aspect of the present invention, an electronic device is provided, comprising:

[0033] at least one processor; and

[0034] a memory communicatively connected to at least one of the processors; wherein,

[0035] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned dynamic forward modeling method of carbon release scenarios oriented to time series.

[0036] Beneficial effects of the present invention:

[0037] (1) The present invention adopts a forward modeling strategy that combines stage division with random sampling simulation technology, randomly generates multiple groups of carbon release patterns in each stage, comprehensively covers possible carbon release scenarios, enhances the exploration capability of parameter space, effectively improves the accuracy and pertinence of forward modeling, and significantly improves the coverage and accuracy of forward modeling of carbon release scenarios; it can efficiently generate multiple carbon release patterns, covering mixed release scenarios of complex carbon sources, and realize accurate simulation of the high-resolution dynamic evolution process of the target curve through stage division and error optimization;

[0038] (2) Through the fluctuation information recognition technology based on the target curve, multiple forward dynamic simulations are realized, achieving high-resolution capture of short-term dynamic characteristics during the carbon release process, further improving the simulation accuracy and dynamic response capability, and solving the problem that the existing technology focuses more on long-term trends in time resolution design and cannot effectively capture short-term high-frequency changes;

[0039] (3) Through detailed analysis of the potential fluctuations of the target curve, the potential release stages and characteristics of different carbon sources are accurately identified, the target curve is divided into multiple specific stages, and a carbon release pattern is generated for each stage. Through dynamic parameter adjustment and multi-scenario forward simulation, the interaction of multiple carbon sources and their comprehensive impact on the carbon cycle are reflected. This improvement breaks through the traditional model's reliance on a single carbon source and improves the simulation capability of complex carbon release scenarios.

[0040] (4) By optimizing the simulation strategy and introducing random parameter combinations, the coverage of extreme conditions and complex carbon release scenarios is expanded, and the ability of the forward model to explore diverse scenarios is enhanced. This effectively solves the problems of limited simulation times and insufficient parameter combination coverage caused by computing resource constraints in existing forward modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0042] Figure 1 It is a flow chart of the dynamic forward modeling method of carbon release scenario for time series of the present invention;

[0043] Figure 2 is a histogram of the release amounts of various carbon sources generated by the Monte Carlo random method in Example 1, wherein Figure 2 (a) is the randomly generated hypothetical volcanic CO2 release, Figure 2 (b) is the randomly generated hypothetical release of thermogenic methane. Figure 2 (c) is the randomly generated hypothetical methane release. Figure 2 (d) in the equation is the total carbon release amount generated randomly;

[0044] Figure 3 : is a result diagram of the RMSE value of the Monte Carlo random simulation result of Example 1, wherein Figure 3 (A) is the root mean square error distribution of the simulated carbon dioxide concentration results. Figure 3 (B) is the root mean square error distribution of the simulated carbon isotope results. Figure 3 (C) in the figure is the normalized distribution of the root mean square error of carbon dioxide and carbon isotope;

[0045] Figure 4 1 is a schematic diagram of three final simulation results and a target carbon isotope curve obtained after error screening in Example 1;

[0046] Figure 5 1 is a schematic diagram of three final simulation results and a target carbon dioxide curve obtained after error screening in Example 1;

[0047] Figure 6 It is the analysis result of the release patterns corresponding to the three carbon release patterns after error screening in Example 1. DETAILED DESCRIPTION

[0048] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0049] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0050] Based on target curve analysis, the present invention identifies potential fluctuations of the target curve and divides it into stages, each of which corresponds to a relatively independent carbon release process; then randomly generates carbon release pattern combinations in each stage and simulates them one by one; finally, the carbon isotope curve generated by numerical simulation is compared with the target carbon isotope curve, and the pattern with the smallest error is selected as the final carbon release scenario. This improves the simulation efficiency and the scientific nature of the results, and provides stronger technical support for the research and prediction of complex carbon release events.

[0051] In order to more clearly explain the dynamic forward modeling method of carbon release scenarios for time series of the present invention, the following is combined with Figure 1 Each step in the embodiment of the present invention is described in detail.

[0052] A time series-oriented carbon release scenario dynamic forward modeling method according to a first embodiment of the present invention is used for dynamic forward modeling of a carbon release event process to obtain a final carbon release pattern, including steps S1 to S4, each of which is described in detail as follows:

[0053] Step S1, selecting a target curve of a carbon release event to be dynamically forward modeled and identifying the potential carbon source type corresponding thereto;

[0054] The present invention takes the target curve as the core reference. The selection of the target curve should be based on the specific research needs, including the carbon isotope curve ( ), atmospheric carbon dioxide concentration curve or temperature change curve, etc.

[0055] In this embodiment, when selecting the target curve, it is necessary to pay attention to:

[0056] The target curve should have clear time series variation characteristics and reflect the main trends and stage characteristics of carbon release events. The selection of the target curve should also be combined with the geological background to ensure the credibility and adaptability of the data.

[0057] A well-defined boundary consists of a finite number of closed intervals, each of which is defined on its own and has well-defined endpoints.

[0058] After determining the target curve, further identify the potential carbon source type corresponding to it:

[0059] Common sources of carbon release in geological history include volcanic activity, methane hydrate collapse, and organic carbon oxidation. Among them, volcanic activity usually releases a large amount of carbon dioxide and sulfide through eruptions, which directly affects atmospheric carbon reserves and global climate; methane hydrate collapse releases a large amount of methane when the temperature rises or the pressure conditions change, which is an important source of rapid carbon release events; organic carbon oxidation may be triggered by climate warming or sea level fluctuations, resulting in a large release of sedimentary organic carbon. In addition, the impact of changes in terrestrial or marine carbon reserves on carbon release should also be considered, such as reduced forest cover or weakened marine carbon pump effect. For different types of carbon sources, the intensity, rate and isotopic composition of their release should be determined to provide important parameter basis for subsequent forward simulations. This process ensures the reasonable linkage between the target curve and the carbon source dynamics, laying the foundation for the restoration of complex carbon release scenarios.

[0060] Step S2, identifying the curve characteristics of the target curve through an automatic recognition algorithm and a set secondary verification method, analyzing the timing information of the carbon release event, and then dividing the target curve into multiple stages;

[0061] In order to improve the recognition accuracy, in this embodiment, a mathematical analysis method (such as the first-order derivative, the second-order derivative method or the gradient change analysis) is used as an automatic recognition algorithm to automatically identify the inflection points of the target curve. These inflection points usually reflect significant changes in the trend of the curve, such as the transition from stable to rapid release or the termination of the release process. The secondary verification method uses manual experience to verify the results of the automatic recognition and adjust the rationality of the inflection point division. Finally, the target curve is decomposed into several stages, such as the initial stable period, the rapid release period and the slow recovery period, and each stage corresponds to a relatively independent carbon release process.

[0062] Furthermore, the phase division needs to follow several principles:

[0063] There should be clear boundaries between each stage to ensure the logic of model construction; the stage characteristics should minimize overlap and fuzzy areas to avoid unnecessary errors in the simulation results; the characteristics of each stage must be consistent with the actual change trend of the target curve.

[0064] In this embodiment, the target curve is selected as the carbon isotope curve, which may appear as follows in the initial stage: The value is stable or fluctuates slowly, while the rapid release period is The above stage division not only clarifies the main time nodes of the carbon release process, but also provides a structural basis for the subsequent segmented forward simulation based on the Monte Carlo method. The accuracy of the stage division directly affects the efficiency of the forward model and the reliability of the simulation results.

[0065] Furthermore, the fluctuation characteristics of the target curve can be identified by a variety of algorithms or means, not limited to the above-mentioned use of mathematical analysis methods as automatic identification algorithms. For example, machine learning methods (such as cluster analysis) can be used to automatically divide the stages, or Bayesian analysis combined with prior information can be used for segmentation. In addition, the target curve can be decomposed into multiple frequency bands based on empirical mode decomposition (EMD) to extract key fluctuation characteristics. However, these methods may be sensitive to data quality and algorithm parameters, and additional verification methods are required to ensure the scientific nature of the division results.

[0066] Step S3, after completing the stage division of the target curve, randomly generate multiple groups of carbon release patterns in each stage based on the random sampling method, compose a carbon release simulation curve and input it into the carbon cycle model for numerical simulation one by one;

[0067] Each group of carbon release patterns is composed of a combination of multiple parameters, including the carbon source type, the contribution ratio of the carbon source type, the release rate of the carbon source type and the isotope of the carbon source; a large number of carbon release pattern combinations are randomly generated to cover various possible carbon release scenarios, and they are simulated one by one in combination with numerical models to evaluate the adaptability of different patterns; during the simulation process, the parameters and release patterns of each carbon source are adjusted to dynamically simulate the evolution of carbon release events in the system.

[0068] The carbon cycle model is based on the carbon isotope mass balance equation, which describes the dynamic changes of carbon isotopes in the system. The carbon isotope mass balance equation is expressed as:

[0069] ;

[0070] in, represents the carbon isotope ratio at the next moment, is the ratio at the current moment, Indicates The release rate of a carbon source, For the The carbon isotopic composition of each carbon source, is the total amount of carbon in the system.

[0071] Preferably, in this embodiment, the carbon release pattern is generated and simulated based on the Monte Carlo method, and the method for generating the carbon release pattern is:

[0072] The carbon release pattern is used as a curve parameter. At each stage, according to the characteristics of the target curve and the potential carbon source type, the Monte Carlo method is used to randomly extract multiple groups of combinations in the parameter space, randomly generate multiple random carbon release patterns, input them into the carbon cycle model one by one for numerical simulation to generate carbon release curves, and determine the global carbon release pattern of each carbon release curve. The parameter space includes a large number of randomly generated parameters, including carbon source type, contribution ratio of carbon source type, release rate of carbon source type, and isotope of carbon source.

[0073] Furthermore, the random sampling method is not limited to the Monte Carlo method mentioned above, and Latin hypercube sampling or Bayesian MCMC method can also be used to generate parameter combinations and combined with the carbon cycle model for simulation.

[0074] Furthermore, when the carbon cycle model is input for simulation, different carbon cycle types are selected according to the set carbon release scenario, and the carbon cycle types include LOSCAR or COPSE.

[0075] As an alternative, various carbon cycle models involving atmosphere-ocean-sediment are available in the art.

[0076] Step S4: Compare the carbon release curve generated by the numerical simulation with the target curve, screen the error between the simulation and the target results, and determine the optimal carbon release mode:

[0077] The error between the simulated carbon isotope curve and the target carbon isotope curve is calculated to quantify the model fitting effect. All simulated carbon release curves are sorted from small to large according to the error value, and the carbon release pattern of the carbon release curve with an error less than the set threshold is selected as the final carbon release scenario.

[0078] Preferably, in this embodiment, when quantifying the model fitting effect, the fitting degree of the simulation result is evaluated by calculating the root mean square error, and the error calculation formula is as follows:

[0079] ;

[0080] in, represents the carbon isotope value simulated by the model, is the actual carbon isotope value at the corresponding time point in the target curve, is the total number of time points. By calculating the RMSE for each set of simulation results, the fitting quality of the global carbon release pattern can be quantitatively evaluated. The smaller the error, the higher the degree of fit between the global carbon release pattern and the target curve. During the screening process, all generated global carbon release patterns are sorted from small to large according to the RMSE value, and the global carbon release pattern with the smallest error is selected as the final carbon release scenario.

[0081] This model not only includes the contribution ratio, release rate and isotopic composition of carbon sources at different stages, but also reflects the dynamic characteristics of the target curve. This error screening method provides a clear quantitative basis for the selection of the optimal carbon release model, and provides scientific support for the dynamic reduction of complex carbon release events. The final results can reveal the main sources of carbon release in a specific time period and its dynamic evolution law, providing a key reference for the study of geological history carbon cycle and modern carbon emission prediction.

[0082] Furthermore, in terms of error calculation, it is not limited to the above-mentioned use of the root mean square error (RMSE). The log-likelihood method can also be used to quantify the model fitting effect, thereby assigning probability weights to different simulation results, which can more intuitively reflect the quality of the model fitting with the target curve. In addition, other statistical indicators (such as AIC or BIC) can be combined to comprehensively evaluate different modes and further select the optimal mode. These alternative methods can provide more flexible model screening basis in different scenarios, but may require additional computing resources.

[0083] Although the various steps in the above embodiment are described in the above-mentioned order, those skilled in the art can understand that in order to achieve the effect of this embodiment, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in a reverse order. These simple changes are within the scope of protection of the present invention.

[0084] This embodiment preferably uses a dynamic forward modeling method and retrograde simulation and optimization analysis of carbon release scenarios for time series to analyze the dramatic carbon emission event in the Toarcian period of the Early Jurassic. The carbon emission event records carbon isotopes ( ) was a large negative shift of about 7%, accompanied by a sharp increase in atmospheric carbon dioxide concentration and global temperature, which became one of the possible driving forces of biological extinction at that time. At the same time, there was a mixed release of multiple carbon sources during the period, including volcanic activity, methane hydrate collapse and organic carbon oxidation.

[0085] One or more target curves can be selected as references. In this embodiment, both the carbon dioxide concentration curve and the carbon isotope curve are used as reference curves. Then, a variety of carbon release patterns are randomly generated using the Monte Carlo method. Figure 2 As shown in the figure, it covers the contribution ratio, release rate and isotopic composition of different carbon sources (such as volcanic activity, methane hydrate collapse, and organic carbon oxidation). These patterns reflect the possible carbon release scenarios of the Toarcian period, ranging from high-intensity, short-term pulse release to slow changes in low-rate, long-term release, providing a comprehensive parameter combination for subsequent model simulations.

[0086] Then, each release mode is simulated by the LOSCAR carbon cycle model, and the various parts are randomly combined to generate a carbon release curve, and the error between the generated carbon release curve and the target result is calculated. The RMSE values ​​of the target curve and the carbon dioxide concentration curve are used to evaluate the degree of fit between the carbon release patterns of different curves and the target curve. Figure 3 As shown, according to Figure 3 The root mean square error distribution of the carbon dioxide concentration results in (A) and Figure 3 The root mean square error distribution of the carbon isotope results in (B) can be used to obtain the standardized distribution of the root mean square error of carbon dioxide and carbon isotopes. The results show that the RMSE value presents a certain distribution characteristic, reflecting the matching ability of different release modes to the target curve; among them, the RMSE values ​​of some modes are relatively low (red points), indicating that these modes can better restore the characteristics of carbon release events in the Toarcian period.

[0087] All global carbon release patterns are sorted from small to large according to the RMSE value, and the global carbon release pattern with the smallest error is selected as the final carbon release scenario. The carbon release curve generated by the numerical simulation is compared with the target curve to screen the error between the simulation and the target results. Figure 4 , Figure 5 The three most consistent simulation results obtained after error screening are compared with the target carbon isotope curve and the target carbon dioxide curve in the line graph. This model not only restores the overall trend of the target curve well, but also captures the significant changes in the rapid release stage and the slow fluctuations in the recovery period.

[0088] Figure 6 The results of the analysis of the three selected carbon release modes are shown in Figure 1, which are three possibilities for the evolution of the carbon release event in the Toarcian period. They may be driven by multiple carbon sources, with volcanic activity accounting for the main contribution, accompanied by the rapid release of methane hydrates and the continuous release of organic carbon oxidation. The combination of carbon sources in different stages is different in the figure. The entire process of the dynamic evolution of the target curve is accurately simulated through stage division and error optimization, thus fully reflecting the evolution process. The mixed release mode of each carbon source successfully explains the evolution of the event. The significant negative shift of the Toarcian period and its evolution over time. This result supports the hypothesis that multiple carbon sources act synergistically to drive Toarcian environmental changes, and provides new evidence for further understanding of complex carbon release events. The proposed method successfully performed a high-precision dynamic reduction of the drastic carbon emission events in the Toarcian period, verifying its strong applicability and scientific value in the study of complex carbon release events.

[0089] A dynamic forward modeling system for carbon release scenarios oriented to a time series according to a second embodiment of the present invention comprises:

[0090] a feature selection module configured to select a target curve and determine the type of potential carbon source corresponding thereto, and determine the intensity, rate and isotopic composition of its release;

[0091] A feature recognition module is configured to recognize the curve features of the target curve, analyze and extract the timing information of the carbon release event, and divide the target curve into stages;

[0092] A parameter generation module is configured to randomly generate multiple groups of carbon release patterns in each stage based on a random sampling method for the target curve after the stage division, so as to obtain multiple carbon release patterns;

[0093] The carbon cycle simulation module is configured to numerically simulate the carbon release patterns one by one, adjust the parameters and release patterns of each carbon source, and dynamically simulate the evolution of carbon isotopes in the system;

[0094] The error screening module is configured to compare the carbon release curve generated by the numerical simulation with the target curve, screen the error between the simulation and the target results, and determine the optimal carbon release mode.

[0095] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0096] It should be noted that the dynamic forward modeling system for carbon release scenarios for time series provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiment can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not regarded as improper limitations on the present invention.

[0097] An electronic device according to a third embodiment of the present invention includes:

[0098] at least one processor; and

[0099] a memory communicatively connected to at least one of the processors; wherein,

[0100] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned dynamic forward modeling method of carbon release scenarios oriented to time series.

[0101] A fourth embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned dynamic forward modeling method for carbon release scenarios oriented to time series.

[0102] Technicians in the relevant technical field can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the electronic device and computer-readable storage medium described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0103] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software, or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0104] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0105] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0106] The terms "first", "second", etc. are used to distinguish similar objects rather than to describe or indicate a particular order or sequence.

[0107] The term "comprise" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus / device.

[0108] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A dynamic forward modeling method for carbon release scenarios for time series, used for dynamic forward modeling of carbon release events in geological history to obtain the optimal carbon release mode, characterized in that: The following steps are involved: Step S1, selecting a target curve of a carbon release event to be dynamically forward modeled and identifying the corresponding potential carbon source type; Determining the release intensity, rate and isotopic composition of the potential carbon source type; Step S2, identifying the curve characteristics of the target curve through an automatic recognition algorithm and a set secondary verification method, analyzing the timing information of the carbon release event, and then dividing the target curve into multiple stages; the curves of each stage have clear boundaries and the characteristics of each stage are consistent with the actual change trend of the target curve; Step S3: randomly generate multiple groups of carbon release patterns in each stage curve based on a random sampling method, and input them into the carbon cycle model for numerical simulation one by one: Each group of carbon release patterns is composed of multiple parameters; the parameters include carbon source type, contribution ratio of carbon source type, release rate of carbon source type and isotope of carbon source; in the numerical simulation process, the evolution process of the carbon release event is dynamically simulated by adjusting the parameters and release pattern of each carbon source; Step S4: Compare the carbon release curves generated by the numerical simulation with the target curves respectively, obtain the errors between any two curves, and determine the final carbon release mode according to the errors.

2. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 1, characterized in that: The target curves include a carbon isotope curve, an atmospheric carbon dioxide concentration curve, and a temperature change curve.

3. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 1, characterized in that: The random sampling method is any one of the Monte Carlo method, Latin hypercube sampling or Bayesian MCMC method.

4. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 3 is characterized in that: In step S3, at each stage, according to the characteristics of the target curve and the type of potential carbon source, a plurality of groups of combinations are randomly selected in the parameter space using the Monte Carlo method to generate a plurality of groups of carbon release patterns.

5. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 1, characterized in that: When inputting the carbon cycle model for simulation, different carbon cycle types are selected according to the set carbon release scenario.

6. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 1, characterized in that: When the selected target curve is the carbon isotope curve, the carbon cycle model is based on the carbon isotope mass balance equation to describe the dynamic changes of carbon isotopes in the system. The carbon isotope mass balance equation is expressed as: ； in, represents the carbon isotope ratio at the next moment, is the ratio at the current moment, Indicates i The release rate of a carbon source, For the i The carbon isotopic composition of each carbon source, is the total amount of carbon in the system.

7. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 6, characterized in that: In step S4, the error between the simulated carbon isotope curve and the target carbon isotope curve is calculated, and the error values ​​are sorted from small to large, and the carbon release mode of the carbon isotope curve with an error value less than a set threshold is selected as the final carbon release scenario.

8. The method for dynamic forward modeling of carbon release scenarios for time series according to claim 7, characterized in that: The fitting degree of the simulation results is evaluated by calculating the root mean square error. The error calculation formula is as follows: ； Among them, RMSE means root mean square error, represents the carbon isotope value simulated by the model, is the actual carbon isotope value at the corresponding time point in the target curve, is the total number of time points.

9. A dynamic forward modeling system for carbon release scenarios oriented to time series, according to a dynamic forward modeling method for carbon release scenarios oriented to time series according to any one of claims 1 to 8, characterized in that: The system includes: a feature selection module configured to select a target curve and determine the type of potential carbon source corresponding thereto, and determine the intensity, rate and isotopic composition of its release; A feature recognition module is configured to recognize the curve features of the target curve, analyze and extract the timing information of the carbon release event, and divide the target curve into stages; A parameter generation module is configured to randomly generate multiple groups of carbon release patterns in each stage based on a random sampling method for the target curve after the stage division, so as to obtain multiple carbon release patterns; The carbon cycle simulation module is configured to numerically simulate the carbon release patterns one by one, adjust the parameters and release patterns of each carbon source, and dynamically simulate the evolution of carbon isotopes in the system; The error screening module is configured to compare the carbon release curve generated by the numerical simulation with the target curve, screen the error between the simulation and the target results, and determine the optimal carbon release mode.

10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement a dynamic forward modeling method for carbon release scenarios oriented to a time series as described in any one of claims 1-8.

Images