A carbon sink dynamic evaluation and optimization method and system for urban ecology

Abstract

This invention relates to the field of urban ecological monitoring and optimization control technology, and discloses a method and system for dynamic assessment and optimization of urban carbon sinks. The method includes: acquiring multi-source green space characteristic data; constructing a light energy utilization model to output a photosynthetic carbon fixation sequence; using a long short-term memory network model based on a weighted trend-maintaining loss function and a DropBlock mechanism; setting scenario influence factors to construct multi-scenario evolution results; and using a particle swarm optimization algorithm to generate optimal green space management and control parameters. Compared to existing technologies, especially under conditions of frequent human management, which struggle to achieve high-frequency dynamic carbon sink assessment and optimization control for urban green spaces of different types, this application improves the real-time performance, accuracy, and efficiency of urban carbon sink monitoring and control optimization by introducing a trend-maintaining loss mechanism and time-segment DropBlock regularization to enhance model generalization ability, and by combining multi-scenario perturbation simulation.

Description

Technical Field

[0001] This invention relates to the field of urban ecological monitoring and optimization control technology, and in particular to a method and system for dynamic assessment and optimization of carbon sequestration in urban ecology. Background Technology

[0002] Currently, the assessment and management of carbon sinks in urban ecosystems is gradually becoming an important component of achieving the strategic goal of "carbon neutrality." Traditional carbon sink assessment methods mainly rely on vegetation index models (such as NDVI and EVI) constructed from remote sensing imagery and meteorological data, or on light energy utilization models (such as the CASA model) for estimation. These methods are typically based on fixed time scales, regional average parameters, and the assumption of static green space, making it difficult to adapt to the typical characteristics of urban green space, such as complex structures, diverse management behaviors, and frequent environmental disturbances. For example, street-side green belts, vertical greening, and park green spaces differ significantly in morphological structure, light acquisition, temperature and humidity fluctuations, and frequency of human intervention, but traditional models usually lack classification modeling mechanisms and cannot dynamically respond to these heterogeneous characteristics. At the same time, existing models generally rely on annual or monthly average data, failing to achieve real-time updates at daily or even higher frequencies, and are unable to predict carbon sink change trends. Furthermore, current urban green space management behaviors (such as irrigation, pruning, and replanting) have a significant impact on carbon sink capacity, but existing technologies rarely incorporate management behaviors as modeling variables into the prediction framework, leading to significant deviations between prediction results and reality. Meanwhile, existing technologies generally lack effective feedback optimization mechanisms and cannot dynamically generate management strategies based on carbon sink prediction results. Therefore, there is an urgent need for a carbon sink assessment and optimization method that integrates heterogeneous urban green spaces, multi-source data fusion, high-frequency dynamic prediction, and intelligent regulation, capable of accurately assessing carbon sink changes, predicting trends, and optimizing strategies even in situations involving complex green space types and frequent human intervention. Summary of the Invention

[0003] To address the aforementioned technical shortcomings, the purpose of this invention is to propose a dynamic assessment and optimization method for carbon sequestration in urban ecology. This method aims to solve the technical problem that existing technologies, especially under conditions of frequent human management, struggle to achieve high-frequency dynamic carbon sequestration assessment and optimization control for urban green spaces of different types.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for dynamic assessment and optimization of carbon sinks for urban ecology.

[0005] The proposed method for dynamic assessment and optimization of carbon sequestration in urban ecology includes:

[0006] Step S10: Obtain the current day's time through a preset multi-source sensor network and remote sensing data source. Green space feature dataset Green space feature dataset Including NDVI value, temperature Soil moisture Light intensity Management behavior variables and time index ;

[0007] Step S20: Construct a light energy utilization model and use the green space characteristic dataset. As input to the light energy utilization efficiency model, the output yields a sequence of photosynthetic carbon fixation. ;

[0008] Step S30: Sequence of photosynthetic carbon fixation As input to the pre-trained Long Short-Term Memory (LSTM) network, the output is a predicted sequence of photosynthetic carbon fixation for the next day; the LSM network pre-training process incorporates a weighted trend-preserving loss function and employs a DropBlock mechanism to prevent overfitting.

[0009] Step S40: Obtain the current green space type, set the scenario impact factor according to the current green space type, map the scenario impact factor to the photosynthetic carbon fixation prediction sequence, and perform cluster analysis to obtain the multi-scenario evolution results;

[0010] Step S50: Based on the multi-scenario evolution results, set the multi-objective function for carbon sink optimization, and use the particle swarm optimization algorithm to solve the multi-objective function for carbon sink optimization. Each particle represents a set of candidate control strategies u(t). By updating the particle position and velocity, iteratively approximate the optimal solution to obtain the optimal management and control parameters for each green space type in the next day.

[0011] Preferably, in step S40, the current green space type includes street green belt, park green belt and vertical greening; the scenario influencing factors include temperature rise influencing factor, precipitation change influencing factor, management frequency influencing factor and green space area expansion influencing factor.

[0012] Preferably, in step S20, the light energy utilization model includes a three-layer shallow structure, specifically including:

[0013] The first layer is a radiation absorption estimation layer, used to extract the actual absorbed photosynthetically active radiation characteristics based on NDVI and light intensity.

[0014] The second layer, the environmental adaptation and regulation layer, is used to extract light energy utilization efficiency characteristics based on variables such as air temperature, soil moisture, and management behavior.

[0015] The third layer, the carbon sink output calculation layer, is used to perform product integral-based calculations based on the characteristics of photosynthetically active radiation and extracted light energy utilization efficiency. Combined with time indexing, it outputs a time-series format of photosynthetic carbon fixation sequence.

[0016] Preferably, step S30, the step of introducing a weighted trend-preserving loss function during the pre-training process of the Long Short-Term Memory network, specifically includes:

[0017] Training sample pairs were constructed based on time series data of photosynthetic carbon fixation and green space characteristic datasets. ,in, For the time of day forward Green space feature dataset of the sky For the first The output of the time series of photosynthetic carbon fixation will be As a predicted carbon sink value sequence ;

[0018] Construct a corresponding real carbon sink value sequence based on preset historical data. According to the true carbon sink value sequence and predicted value sequence Construct a weighted trend preservation loss function. The formula used for the weighted trend preservation loss function is as follows:

[0019]

[0020] in, Preserve the loss function for weighted trend; , ; These are trend constraint weighting coefficients, used to control the sensitivity to trend changes; For the time of day The true carbon sink value sequence, For the first The true carbon sink value sequence for each day. For the time of day The predicted carbon sink value sequence, For the first The predicted carbon sink value sequence for the day; ( ) is the mean squared error function between the actual trend change and the predicted trend change.

[0021] Preferably, step S30, which employs the DropBlock mechanism to prevent overfitting during the pre-training process of the Long Short-Term Memory network, specifically includes:

[0022] Two parameters for the DropBlock mechanism are pre-defined: the masking probability p, which represents the probability of the DropBlock mechanism being applied in each training batch; and the masking block length. , used to represent the number of consecutively masked time steps in the time series dimension;

[0023] For each green space feature dataset in the training sample pairs Sample a starting position in its time dimension The construction length is the length of the shielding block. Window shield And construct the masking matrix. , where the i-th element in the masking matrix ; Masking matrix Application to green space feature dataset Obtain a masked green space feature dataset to prevent overfitting. , ,in This indicates element-wise multiplication.

[0024] Preferably, in step S40, the multi-scenario evolution results include a carbon sink response elasticity index, a sensitive factor contribution ranking table, and a green space type scenario adaptation map; wherein, the carbon sink response elasticity index includes absolute response magnitude, relative carbon sink change rate, and change trend shift; the sensitive factor contribution ranking table is obtained by using a single scenario influence factor while keeping other factors constant, and is used to analyze the marginal change of carbon sink output; the green space type scenario adaptation map is used to support green space priority classification.

[0025] Preferably, in step S50, the carbon sink optimization multi-objective function includes a carbon sink achievement objective function and a control behavior cost objective function; the candidate control strategy u(t) is set through a remote cloud platform, and the influence weight of the candidate control strategy u(t) on the action space in the particle swarm algorithm is preset by expert experience; the optimal management and control parameters include the optimal irrigation frequency, the optimal pruning frequency, and the optimal green space deployment personnel configuration.

[0026] This invention also provides a dynamic assessment and optimization system for carbon sequestration in urban ecology, comprising:

[0027] The green space feature acquisition module is used to obtain the time of day through a preset multi-source sensor network and remote sensing data source. Green space feature dataset Green space feature dataset Including NDVI value, temperature Soil moisture Light intensity Management behavior variables and time index ;

[0028] The light energy utilization modeling module is used to build a light energy utilization model and to process green space feature datasets. As input to the light energy utilization efficiency model, the output yields a sequence of photosynthetic carbon fixation. ;

[0029] The carbon sink prediction module is used to sequence photosynthetic carbon fixation. As input to the pre-trained Long Short-Term Memory (LSTM) network, the output is a predicted sequence of photosynthetic carbon fixation for the next day; the LSM network pre-training process incorporates a weighted trend-preserving loss function and employs a DropBlock mechanism to prevent overfitting.

[0030] The multi-scenario evolution simulation module is used to obtain the current green space type, set scenario influence factors based on the current green space type, map the scenario influence factors to the photosynthetic carbon fixation prediction sequence, and obtain the multi-scenario evolution results.

[0031] The optimization and control strategy generation module is used to set a multi-objective function for carbon sink optimization based on the evolution results of multiple scenarios. The particle swarm optimization algorithm is used to solve the multi-objective function for carbon sink optimization. Each particle represents a set of candidate control strategies u(t). By updating the particle position and velocity, the optimal solution is iteratively approximated to obtain the optimal management and control parameters for each green space type in the next day.

[0032] The present invention also provides a device for dynamic assessment and optimization of carbon sinks for urban ecology, comprising: a memory, a processor, and a program for dynamic assessment and optimization of carbon sinks for urban ecology stored in the memory and executable on the processor. When the program for dynamic assessment and optimization of carbon sinks for urban ecology is executed by the processor, a method for dynamic assessment and optimization of carbon sinks for urban ecology is implemented.

[0033] The present invention also provides a computer program product, including a dynamic assessment and optimization program for carbon sinks in urban ecology, which, when executed by a processor, implements the aforementioned dynamic assessment and optimization method for carbon sinks in urban ecology.

[0034] The beneficial effects of this invention are as follows: Compared with the technical problems in the prior art, especially under the condition of frequent human management, it is difficult to achieve high-frequency dynamic carbon sink assessment and optimization control for urban green spaces of different types. This application improves the real-time performance, accuracy and regulation optimization efficiency of urban carbon sink monitoring by introducing a trend-maintaining loss mechanism and time segment DropBlock regularization to enhance the model's generalization ability, and by combining multi-scenario perturbation simulation. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart illustrating a method for dynamic assessment and optimization of carbon sequestration for urban ecology, based on the present invention.

[0037] Figure 2 This is a schematic diagram of the structure of a dynamic assessment and optimization device for carbon sequestration in urban ecology according to the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Example 1: As Figure 1 The diagram shown is a flowchart of the method for dynamic assessment and optimization of carbon sinks for urban ecology according to the present invention, and an embodiment of the method for dynamic assessment and optimization of carbon sinks for urban ecology according to the present invention is presented.

[0040] In Example 1, the method for dynamic assessment and optimization of carbon sequestration for urban ecology includes:

[0041] Step S10: Obtain the current day's time through a preset multi-source sensor network and remote sensing data source. Green space feature dataset Green space feature dataset Including NDVI value, temperature Soil moisture Light intensity Management behavior variables and time index ;

[0042] It should be noted that the multi-source sensor network may include micro-meteorological sensors installed around the green space, soil moisture monitoring nodes, light intensity acquisition modules, and personnel input terminals for collecting information on management activities. Remote sensing data sources can be multispectral satellite imagery (such as Sentinel-2) with a resolution better than 10 meters to obtain daily NDVI vegetation index values, and the location of the green space can be spatially registered using a geographic information system (GIS). The collection of green space characteristic data does not rely on a single device or a single moment, but rather integrates sources from different time points, scales, and data types to form a structured dataset in a unified format, providing a data foundation for subsequent carbon sink modeling.

[0043] It should be understood that management behavior variables not only include explicit operational behaviors such as irrigation, pruning, and fertilization, but can also be quantified through log records, automated work orders, or maintenance frequency estimation to characterize the intensity of human intervention. The introduction of these variables significantly enhances the model's ability to model non-ecological factors.

[0044] Step S20: Construct a light energy utilization model and use the green space characteristic dataset. As input to the light energy utilization efficiency model, the output yields a sequence of photosynthetic carbon fixation. ;

[0045] It should be noted that in step S20, the light energy utilization model includes a three-layer shallow structure, specifically: the first layer is a radiation absorption estimation layer, used to extract the actual absorbed photosynthetically active radiation characteristics based on NDVI and light intensity; the second layer is an environmental adaptation regulation layer, used to extract light energy utilization efficiency characteristics based on temperature, soil moisture, and management behavior variables; and the third layer is a carbon sink output calculation layer, used to perform calculations based on product integrals according to the photosynthetically active radiation characteristics and the extracted light energy utilization efficiency characteristics, and combined with a time index, outputs a time series of photosynthetic carbon fixation.

[0046] Understandably, light intensity and NDVI primarily determine the basic photosynthetic capacity of vegetation, while temperature, humidity, and human management practices influence the final carbon fixation by regulating photosynthetic efficiency. Therefore, the model extracts light energy utilization efficiency characteristics in the second layer based on variables such as temperature, soil moisture, and management practices. This light energy utilization model is not only applicable to conventional urban greening units such as street-side green belts and park green spaces, but can also be extended to special types of green spaces such as vertical greening and rooftop greening.

[0047] Step S30: Sequence of photosynthetic carbon fixation As input to the pre-trained Long Short-Term Memory (LSTM) network, the output is a predicted sequence of photosynthetic carbon fixation for the next day; the LSM network pre-training process incorporates a weighted trend-preserving loss function and employs a DropBlock mechanism to prevent overfitting.

[0048] It should be noted that step S30, the step of introducing a weighted trend-preserving loss function during the pre-training process of the Long Short-Term Memory network, specifically includes: constructing training sample pairs based on the time series of photosynthetic carbon fixation and the green space feature dataset. ,in, For the time of day forward Green space feature dataset of the sky For the first The output of the time series of photosynthetic carbon fixation will be As a predicted carbon sink value sequence Construct a corresponding real carbon sink value sequence based on preset historical data. According to the true carbon sink value sequence and predicted value sequence Construct a weighted trend preservation loss function. The formula used for the weighted trend preservation loss function is as follows: ,in, Preserve the loss function for weighted trend; , ; These are trend constraint weighting coefficients, used to control the sensitivity to trend changes; For the time of day The true carbon sink value sequence, For the first The true carbon sink value sequence for each day. For the time of day The predicted carbon sink value sequence, For the first The predicted carbon sink value sequence for the day; ( ) represents the mean squared error function between the actual trend change and the predicted trend change. Step S30, which employs the DropBlock mechanism to prevent overfitting during the pre-training process of the Long Short-Term Memory network, specifically includes: pre-setting two parameters for the DropBlock mechanism, including: the masking probability p, representing the probability of the DropBlock mechanism being applied in each training batch; and the masking block length. , used to represent the number of consecutively masked time steps in the time series dimension; for each green space feature dataset in the training sample pair Sample a starting position in its time dimension The construction length is the length of the shielding block. Window shield And construct the masking matrix. , where the i-th element in the masking matrix ; Masking matrix Application to green space feature dataset Obtain a masked green space feature dataset to prevent overfitting. , ,in This indicates element-wise multiplication.

[0049] Understandably, the weighted trend-preserving loss function not only penalizes prediction errors but also explicitly considers the accuracy of carbon sink growth or decline trends; this is particularly important for short-term prediction scenarios of urban green spaces under sudden climate changes and unexpected maintenance actions. Meanwhile, the DropBlock mechanism improves generalization performance by randomly masking continuous time segments, preventing the model from overfitting to inputs from a single time period.

[0050] It should be understood that, compared with traditional LSTM training methods that only use the MSE loss function or ordinary Dropout mechanism, the training mechanism constructed in this implementation method can better enhance the model's ability to learn complex carbon sink dynamics, especially when there are large differences in the performance of different types of green spaces in the city and strong environmental fluctuations, it can still maintain high trend consistency and prediction accuracy.

[0051] Step S40: Obtain the current green space type, set the scenario impact factor according to the current green space type, map the scenario impact factor to the photosynthetic carbon fixation prediction sequence, and perform cluster analysis to obtain the multi-scenario evolution results;

[0052] It should be noted that in step S40, the multi-scenario evolution results include carbon sink response elasticity indicators, a sensitive factor contribution ranking table, and a green space type scenario adaptability map. Among them, the carbon sink response elasticity indicators include absolute response magnitude, relative carbon sink change rate, and trend shift. The sensitive factor contribution ranking table is obtained by using a single scenario influence factor while keeping other factors constant, and is used to analyze the marginal changes in carbon sink output. The green space type scenario adaptability map is used to support green space priority classification.

[0053] Understandably, by using scenario simulation and response indicator extraction, we can not only identify the trend of carbon sink changes under the current management plan, but also assess in advance the potential impact of future climate change or policy intervention on the carbon sink system.

[0054] It should be understood that this invention is not limited to the prediction of static carbon sink values, but obtains the dynamic behavior mapping of the carbon sink system under different external disturbances through multi-scenario analysis, thereby providing input-based management strategies for subsequent optimization, rather than relying solely on empirical rules, which significantly improves the scientific nature and adaptability of the control strategy.

[0055] Step S50: Based on the multi-scenario evolution results, set the multi-objective function for carbon sink optimization, and use the particle swarm optimization algorithm to solve the multi-objective function for carbon sink optimization. Each particle represents a set of candidate control strategies u(t). By updating the particle position and velocity, iteratively approximate the optimal solution to obtain the optimal management and control parameters for each green space type in the next day.

[0056] It should be noted that in step S50, the carbon sink optimization multi-objective function includes the carbon sink achievement objective function and the control behavior cost objective function; the candidate control strategy u(t) is set through a remote cloud platform, and the influence weight of the candidate control strategy u(t) on the action space in the particle swarm algorithm is preset by expert experience; the optimal management and control parameters include the optimal irrigation frequency, the optimal pruning frequency, and the optimal green space deployment personnel configuration.

[0057] Understandably, incorporating expert knowledge into the weighting of control variables can avoid carbon sink deviations or resource waste caused by blind searches. Meanwhile, the multi-scenario weighted processing approach makes the generated strategy more robust, maintaining stable carbon sink performance even under conditions of climate disturbances and data uncertainty.

[0058] It should be understood that this optimization strategy not only outputs a single optimal solution, but can also form a set of strategy candidates, which can be allocated in different green space management systems according to budget, cycle or scheduling priority. It is suitable for application scenarios such as hierarchical deployment, regional coordination and rolling updates, and has strong engineering expansion value.

[0059] Example 2: Furthermore, the present invention provides a dynamic assessment and optimization system for urban ecological carbon sinks, employing a method for dynamic assessment and optimization of urban ecological carbon sinks as described in the above embodiments, which can solve the technical problem of dynamic assessment and optimization of urban ecological carbon sinks. Compared with the prior art, the beneficial effects of the dynamic assessment and optimization system for urban ecological carbon sinks provided by the present invention are the same as those of the method for dynamic assessment and optimization of urban ecological carbon sinks provided in the above embodiments, and other technical features of the dynamic assessment and optimization system for urban ecological carbon sinks are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0060] Example 3: This invention provides a device for dynamic assessment and optimization of carbon sequestration in urban ecology. Please refer to... Figure 2A carbon sequestration dynamic assessment and optimization device for urban ecology includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to execute the carbon sequestration dynamic assessment and optimization method for urban ecology described in Embodiment 1 above. The carbon sequestration dynamic assessment and optimization device for urban ecology in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. This carbon sequestration dynamic assessment and optimization device for urban ecology is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the invention. A carbon sequestration dynamic assessment and optimization device for urban ecology may include a processor 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the carbon sequestration dynamic assessment and optimization device for urban ecology. The processor 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows a carbon sequestration dynamic assessment and optimization device for urban ecology to exchange data with other devices wirelessly or via wired communication. Although the figure shows a carbon sequestration dynamic assessment and optimization device for urban ecology with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0061] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for dynamic assessment and optimization of carbon sinks for urban ecology. The computer program product provided by this invention can solve the technical problem of dynamic assessment and optimization of carbon sinks for urban ecology. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the method for dynamic assessment and optimization of carbon sinks for urban ecology provided in the above embodiments, and will not be repeated here.

[0062] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processor 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0063] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0064] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamic assessment and optimization of carbon sequestration in urban ecology, characterized in that, The methods include: Step S10: Obtain the current day's time through a preset multi-source sensor network and remote sensing data source. Green space feature dataset Green space feature dataset Including NDVI value, temperature Soil moisture Light intensity Management behavior variables and time index ; Step S20: Construct a light energy utilization model and use the green space characteristic dataset. As input to the light energy utilization efficiency model, the output yields a sequence of photosynthetic carbon fixation. ; Step S30: Sequence of photosynthetic carbon fixation As input to the pre-trained Long Short-Term Memory (LSTM) network, the output is a predicted sequence of photosynthetic carbon fixation for the next day; the LSM network pre-training process incorporates a weighted trend-preserving loss function and employs a DropBlock mechanism to prevent overfitting. Step S40: Obtain the current green space type, set scenario impact factors based on the current green space type, map the scenario impact factors to the photosynthetic carbon fixation prediction sequence, and perform cluster analysis to obtain multi-scenario evolution results; among which, the multi-scenario evolution results include carbon sink response elasticity indicators, sensitive factor contribution ranking tables, and green space type scenario adaptability maps; carbon sink response elasticity indicators include absolute response magnitude, relative carbon sink change rate, and trend shift; the sensitive factor contribution ranking table is obtained by using a single scenario impact factor while keeping other factors constant, and is used to analyze the marginal changes in carbon sink output; the green space type scenario adaptability map is used to support green space priority classification; Step S50: Based on the multi-scenario evolution results, a multi-objective function for carbon sink optimization is set, and a particle swarm optimization algorithm is used to solve the multi-objective function for carbon sink optimization. Each particle represents a set of candidate control strategies u(t). By updating the particle position and velocity, the optimal solution is iteratively approximated to obtain the optimal management and control parameters for each green space type in the next day. Among them, the multi-objective function for carbon sink optimization includes a carbon sink achievement objective function and a control behavior cost objective function. The candidate control strategies u(t) are set through a remote cloud platform, and the influence weight of the candidate control strategies u(t) on the action space in the particle swarm optimization algorithm is preset by expert experience. The optimal management and control parameters include the optimal irrigation frequency, the optimal pruning frequency, and the optimal green space deployment personnel configuration.

2. The method for dynamic assessment and optimization of carbon sequestration for urban ecology as described in claim 1, characterized in that, In step S40, the current green space types include street green belts, park green belts, and vertical greening; the scenario influencing factors include temperature rise influencing factors, precipitation change influencing factors, management frequency influencing factors, and green space area expansion influencing factors.

3. The method for dynamic assessment and optimization of carbon sequestration for urban ecology as described in claim 1, characterized in that, In step S20, the light energy utilization model includes a three-layer shallow structure, specifically including: The first layer is a radiation absorption estimation layer, used to extract the actual absorbed photosynthetically active radiation characteristics based on NDVI and light intensity. The second layer, the environmental adaptation and regulation layer, is used to extract light energy utilization efficiency characteristics based on variables such as air temperature, soil moisture, and management behavior. The third layer, the carbon sink output calculation layer, is used to perform product integral-based calculations based on the characteristics of photosynthetically active radiation and extracted light energy utilization efficiency. Combined with time indexing, it outputs a time-series format of photosynthetic carbon fixation sequence.

4. The method for dynamic assessment and optimization of carbon sequestration for urban ecology as described in claim 1, characterized in that, Step S30, the step of introducing a weighted trend-preserving loss function during the pre-training process of the Long Short-Term Memory network, specifically includes: Training sample pairs were constructed based on time series data of photosynthetic carbon fixation and green space characteristic datasets. ,in, For the time of day forward Green space feature dataset of the sky For the first The output of the time series of photosynthetic carbon fixation will As a predicted carbon sink value sequence ; Construct a corresponding real carbon sink value sequence based on preset historical data. According to the true carbon sink value sequence and predicted value sequence Construct a weighted trend preservation loss function. The formula used for the weighted trend preservation loss function is as follows: ； in, Preserve the loss function for weighted trend; , ; These are trend constraint weighting coefficients, used to control the sensitivity to trend changes; For the time of day The true carbon sink value sequence, For the first The true carbon sink value sequence for each day. For the time of day The predicted carbon sink value sequence, For the first The predicted carbon sink value sequence for the day; ( ) is the mean squared error function between the actual trend change and the predicted trend change.

5. A method for dynamic assessment and optimization of carbon sequestration for urban ecology as described in claim 4, characterized in that, Step S30, which involves using the DropBlock mechanism to prevent overfitting during the pre-training process of the Long Short-Term Memory network, specifically includes: Two parameters for the DropBlock mechanism are pre-defined: the masking probability p, which represents the probability of the DropBlock mechanism being applied in each training batch; and the masking block length. , used to represent the number of consecutively masked time steps in the time series dimension; For each green space feature dataset in the training sample pairs Sample a starting position in its time dimension The construction length is the length of the shielding block. Window mask And construct the masking matrix. , where the i-th element in the masking matrix ; Masking matrix Application to green space feature dataset Obtain a masked green space feature dataset to prevent overfitting. , ,in This indicates element-wise multiplication.

6. A dynamic assessment and optimization system for carbon sequestration in urban ecology, applied to the dynamic assessment and optimization method for carbon sequestration in urban ecology as described in any one of claims 1-5, characterized in that, The aforementioned dynamic assessment and optimization system for carbon sequestration in urban ecology includes: The green space feature acquisition module is used to obtain the time of day through a preset multi-source sensor network and remote sensing data source. Green space feature dataset Green space feature dataset Including NDVI value, temperature Soil moisture Light intensity Management behavior variables and time index ; The light energy utilization modeling module is used to build a light energy utilization model and to process green space feature datasets. As input to the light energy utilization efficiency model, the output yields a sequence of photosynthetic carbon fixation. ; The carbon sink prediction module is used to sequence photosynthetic carbon fixation. As input to the pre-trained Long Short-Term Memory (LSTM) network, the output is a predicted sequence of photosynthetic carbon fixation for the next day; the LSM network pre-training process incorporates a weighted trend-preserving loss function and employs a DropBlock mechanism to prevent overfitting. The multi-scenario evolution simulation module is used to obtain the current green space type, set scenario influence factors based on the current green space type, map the scenario influence factors to the photosynthetic carbon fixation prediction sequence, and perform cluster analysis to obtain multi-scenario evolution results. These results include a carbon sink response elasticity index, a sensitive factor contribution ranking table, and a green space type scenario adaptability map. The carbon sink response elasticity index includes absolute response magnitude, relative carbon sink change rate, and trend shift. The sensitive factor contribution ranking table is obtained by using a single scenario influence factor while keeping other factors constant, and is used to analyze the marginal changes in carbon sink output. The green space type scenario adaptability map is used to support green space priority classification. The optimization and control strategy generation module is used to set a multi-objective function for carbon sink optimization based on the evolution results of multiple scenarios. The particle swarm optimization algorithm is used to solve the multi-objective function for carbon sink optimization. Each particle represents a set of candidate control strategies u(t). By updating the particle position and velocity, the optimal solution is iteratively approximated to obtain the optimal management and control parameters for each green space type in the next day. The multi-objective function for carbon sink optimization includes a carbon sink achievement objective function and a control behavior cost objective function. The candidate control strategies u(t) are set through a remote cloud platform. The influence weight of the candidate control strategies u(t) on the action space of the particle swarm algorithm is preset by expert experience. The optimal management and control parameters include the optimal irrigation frequency, the optimal pruning frequency, and the optimal personnel configuration for green space deployment.

7. A device for dynamic assessment and optimization of carbon sequestration in urban ecology, characterized in that, The device for dynamic assessment and optimization of carbon sinks for urban ecology includes: a memory, a processor, and a program for dynamic assessment and optimization of carbon sinks for urban ecology stored in the memory and executable on the processor. When the program for dynamic assessment and optimization of carbon sinks for urban ecology is executed by the processor, it implements a method for dynamic assessment and optimization of carbon sinks for urban ecology as described in any one of claims 1 to 5.

8. A computer program product, characterized in that, The computer program product includes a dynamic assessment and optimization program for carbon sinks in urban ecology. When the dynamic assessment and optimization program for carbon sinks in urban ecology is executed by the processor, it implements a dynamic assessment and optimization method for carbon sinks in urban ecology as described in any one of claims 1 to 5.

Images