Climate change response type modular micro-landscape self-adaptive monitoring system

Abstract

The invention discloses a climatic change response type modular micro-landscape self-adaptive monitoring system, which comprises a micro-landscape module, a climatic change monitoring module, a data processing module and a self-adaptive adjustment module, the micro-landscape module is made of a sustainable material subjected to multi-dimensional testing and is integrated with climatic change sensitive type indicative organisms; the climatic biological monitoring module is used for collecting climatic environment parameter data and indicative biological quantitative characteristic data; the data processing module is used for receiving climate environment parameter data and indicative biological quantitative characteristic data acquired by the climate biological monitoring module, predicting climate trend and micro-landscape growth state pre-judgment through a CNN-LSTM network, and outputting an optimization scheme in combination with a genetic algorithm; and the adaptive module dynamically adjusts the module structure and generates environmental parameters, thereby realizing accurate response to climate change and micro-landscape ecological efficiency optimization.

Description

Technical Field

[0001] This invention belongs to the field of landscape engineering and environmental monitoring, specifically a climate change-responsive modular micro-landscape adaptive monitoring system. Background Technology

[0002] In the existing technology, domestic micro-landscape research focuses on interior decoration, landscaping and the creation of vernacular landscapes, while foreign research focuses on microclimate regulation and human thermal perception optimization. Modular design has been applied to products, buildings and other fields. Micro-landscape materials need to pass physical, chemical and biological performance tests. Climate change monitoring mainly relies on sensors and some indicator organisms.

[0003] However, the existing solutions have obvious shortcomings: micro-landscape and climate change monitoring lack deep integration, modular theoretical research lags behind practice, a complete system integrating indicator organisms and adaptive regulation has not been formed, and there is a lack of unified ecological assessment standards and practical implementation paths, making it difficult to accurately respond to dynamic climate conditions and meet the actual needs of urban ecological sustainable development. Summary of the Invention

[0004] To address the shortcomings of the existing technologies, the present invention aims to provide a climate change-responsive modular micro-landscape adaptive monitoring system to solve the technical problems of the lack of deep integration between existing micro-landscape and climate change monitoring, the disconnect between modular theory and practice, the absence of a complete system including indicator organisms and adaptive regulation, and the lack of unified ecological performance evaluation standards and practical implementation paths, which makes it difficult to accurately respond to dynamic climate conditions and continuously optimize the urban micro-ecological environment.

[0005] To achieve the above objectives, this invention discloses a climate change-responsive modular micro-landscape adaptive monitoring system, the system comprising: a micro-landscape module, a climate biological monitoring module, a data processing module, and an adaptive adjustment module;

[0006] The micro-landscape modules are made from selected sustainable materials and integrate climate change-sensitive indicator organisms;

[0007] The climate biological monitoring module is used to collect climate and environmental parameter data and indicative biological quantitative characteristic data;

[0008] The data processing module is used to receive the climate and environmental parameter data and the indicative biological quantitative characteristic data, analyze them, and output the control plan;

[0009] The adaptive adjustment module dynamically adjusts the micro-landscape module according to the control scheme to adapt to climate change, while ensuring the stable generation of indicator organisms and the validity of monitoring data.

[0010] Furthermore, the micro-landscape module is made of materials selected through multi-dimensional performance testing and ecological effectiveness assessment;

[0011] The multi-dimensional performance tests include compressive strength testing, durability testing, and corrosion resistance testing;

[0012] The compressive strength test evaluates the material by calculating its compressive strength; the formula for calculating the compressive strength test is:

[0013]

[0014] in, It is compressive strength; It is the pressure value applied to the material; It is the actual area of ​​the material subjected to force;

[0015] The durability test is evaluated based on the aging rate; the formula for the aging rate is:

[0016]

[0017] in, It refers to the material aging rate; It represents the change in mass of the material during accelerated aging tests; t is the duration of the aging test.

[0018] The corrosion resistance test is evaluated using an electrochemical corrosion rate; the formula for the electrochemical corrosion rate is:

[0019]

[0020] in, It is the corrosion rate; It is the current value during the corrosion process; It is the number of electrons transferred; It is the surface area of ​​the material that participates in the corrosion reaction;

[0021] The ecological effectiveness assessment employs landscape function analysis methods for transect layout and observation. These methods assess the landscape function through two main categories: landscape structure and soil surface characteristics. The landscape structure includes a first single indicator and a first composite indicator. The soil surface characteristics include a second single indicator and a second composite indicator.

[0022] The first single indicator includes the total vegetation area intercepted by the transect. Maximum coverage area The total length of the area covered by the transect line and profile length The first composite index includes ground-based indicators. and landscape structure indicators ;

[0023] The ground index The calculation formula is:

[0024]

[0025] The landscape structure index The calculation formula is:

[0026] ;

[0027] The second single index includes surface cover rate C, number of dead leaves and leaf layers D, coverage rate of cryptogamic plants E, quantitative value of crust fragmentation F, quantitative value of soil erosion grade G, sediment accumulation H, resistance to disturbance coefficient J, quantitative value of disintegration test results Q, coverage rate of perennial plants L, quantitative value of surface roughness T, and soil texture parameter W; the second composite index includes soil stability index Sa, water infiltration index If, and nutrient cycling index Ne.

[0028] The formula for calculating the soil stability index Sa is as follows:

[0029]

[0030] in, Let be the weight coefficient of the i-th sample; The total number of samples involved in the calculation; Let be the land cover rate of the i-th sample; Let be the number of layers of dead leaves in the i-th sample; Let be the coverage rate of cryptogams in the i-th sample; This is the quantified value of the degree of crust breakage for the i-th sample; This represents the quantified value of the soil erosion level for the i-th sample. Let be the amount of sediment accumulation in the i-th sample; Let be the anti-interference capability coefficient of the i-th sample; This is the quantified value of the disintegration test result for the i-th sample;

[0031] The formula for calculating the water permeability index If is:

[0032]

[0033] in, Let be the plant coverage of the i-th sample; Let be the quantified value of the surface roughness of the i-th sample; Let be the soil texture parameters for the i-th sample;

[0034] The formula for calculating the nutrient cycling index Ne is as follows:

[0035] .

[0036] Furthermore, the material screening for the micro-landscape module also includes tensile testing, abrasion resistance testing, and durability testing in physical performance testing; corrosion resistance testing and fire resistance testing in chemical performance testing; biodegradation testing and antibacterial performance testing in biological performance testing; and the screening of material adaptability and sustainability is completed by combining systematic clustering, TOPSIS method, and expert scoring weighting method.

[0037] Furthermore, the climate biological monitoring module includes indicator organisms and monitoring equipment; the indicator organisms are selected from climate change-sensitive organisms identified through literature screening in the fields of ecology, botany, agronomy, and hydrology, and use the micro-landscape module as a growth carrier; the monitoring equipment includes a temperature sensor and a hygrometer; the temperature sensor is used to record temperature data and calculate the average temperature over a certain period using an average temperature formula; the average temperature formula is expressed as:

[0038]

[0039] in, It is the average temperature over a certain period of time; It refers to the number of monitoring sessions within a specific time period; This is a single temperature reading;

[0040] The hygrometer is used to record humidity data and analyze the trend of humidity change using a humidity change rate formula; the humidity change rate formula is expressed as:

[0041]

[0042] in, It is the rate of change of humidity; It is the change in humidity during the monitoring period; For monitoring duration.

[0043] Furthermore, the climate and environmental parameter data includes the average temperature and humidity change rates over a certain period of time;

[0044] The indicative biological quantitative data include at least one of the following: growth rate of the organism, coefficient of variation of leaf morphology, and content of physiological metabolites.

[0045] Furthermore, the data processing module is used to receive the climate and environmental parameter data and the indicative biological quantitative characteristic data, construct a modular model of the micro-landscape using GIS or related software, integrate deep learning algorithms to obtain climate trend prediction and micro-landscape growth status prediction, calculate the changes in micro-landscape adaptation effect through comparative analysis and effect evaluation formula, and optimize the design of the micro-landscape module using optimization algorithms to output a control scheme.

[0046] Furthermore, the effect evaluation formula is expressed as follows:

[0047]

[0048] in, It is the amount of change in effect; These are the indicator values ​​after the system has been running; These are the indicator values ​​before the system is running.

[0049] Furthermore, the deep learning algorithm includes a hybrid model of LSTM and CNN, CNN-LSTM, for climate trend prediction and micro-landscape growth status prediction.

[0050] The CNN-LSTM consists of an input layer, a CNN spatial feature extraction module, an LSTM temporal feature capture module, a feature fusion module, a fully connected layer, and an output layer.

[0051] The optimization algorithm is a genetic algorithm, used for the optimized design of the micro-landscape module.

[0052] The genetic algorithm achieves multi-objective collaborative optimization of micro-landscape module combination, climate adaptation strategy and ecological efficiency through hybrid encoding, initial population generation, fitness function construction, selection-crossover-mutation genetic operations and multiple termination condition determination, and outputs Pareto optimal solution set to support the implementation of modular practice path.

[0053] The adaptive adjustment module dynamically adjusts the combination structure and generation environmental parameters of each modular micro-landscape unit according to the control scheme output by the data processing module. It continuously collects micro-landscape adaptation data under different climatic conditions through dynamic intervention experiments, and iteratively optimizes the control logic based on the feedback results to achieve a response to multi-dimensional climate change.

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] (1) This invention constructs a system from material screening to biological indicators to intelligent monitoring and finally to dynamic regulation, realizing the integration of multiple disciplines; it deeply integrates landscape function analysis (LFA) ecological effectiveness assessment, deep learning algorithms and modular micro-landscapes, solving the pain points of lack of integration between micro-landscape and climate change monitoring and the disconnect between modular theory and practice in the existing technology, and realizing the synergistic optimization of ecological adaptability and climate responsiveness.

[0056] (2) This invention uses a CNN-LSTM hybrid network and a genetic / simulated annealing algorithm to achieve accurate prediction of climate trends and micro-landscape growth status. Combined with the dynamic regulation of the adaptive adjustment module, it solves the limitation of existing technologies that cannot accurately respond to dynamic climate conditions and improves the adaptation efficiency and stability of micro-landscapes to multi-dimensional climate changes such as temperature and humidity.

[0057] (3) This invention establishes a standardized modular development and evaluation path for micro-landscapes. It screens sustainable materials through multi-dimensional performance testing and integrates indicator organisms of appropriate scale. It not only forms a unified ecological performance evaluation standard, but also has the flexibility and practicality of modular assembly. It fills the gap of existing technologies lacking practical implementation paths and provides a replicable technical solution for urban microclimate improvement and the construction of a healthy China. Attached Figure Description

[0058] Figure 1 This is an overall structural diagram of the system of the present invention;

[0059] Figure 2 This is a schematic diagram of the CNN-LSTM model used in the system of this invention;

[0060] Figure 3 This is a flowchart of the genetic algorithm optimization process used in the system of this invention; Detailed Implementation

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below. 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.

[0062] Please see Figure 1 , Figure 2 and Figure 3 This application provides a detailed description of the technical solutions provided in each embodiment.

[0063] This application provides a climate change-responsive modular micro-landscape adaptive monitoring system, such as... Figure 1 Specifically, it includes: a micro-landscape module, a climate and biological monitoring module, a data processing module, and an adaptive adjustment module;

[0064] The application scenario was defined as a green space in an old community in Hohhot, Inner Mongolia, with an area of ​​approximately 18 square meters. The average daily temperature in this area is 30-33°C in summer and -15°C in winter. The humidity fluctuates greatly and the soil has poor water retention. A modular micro-landscape system is needed to achieve climate change monitoring and micro-ecological optimization.

[0065] For core resources and materials, we selected PLA biodegradable plastics, recycled concrete, and bamboo fiber composites.

[0066] For testing equipment, prepare a universal testing machine, accelerated aging test chamber, electrochemical workstation, temperature sensor, hygrometer, Martindale abrasion tester, GIS software, computer, 3D printer, ruler, image recognition equipment and pressure sensor;

[0067] The indicator organisms were selected through a literature review, screening literature in the fields of ecology and botany, and moss and sea buckthorn seedlings were chosen, meeting the requirements of suitable scale, high safety and strong adaptability.

[0068] Multi-dimensional performance testing of materials, including physical performance testing, chemical performance testing and biological performance testing;

[0069] The physical performance tests include compressive strength test, tensile test, abrasion resistance test and durability test;

[0070] The compressive strength test involves processing the three types of materials into standard 10cm×10cm×10cm test blocks, and applying pressure at a constant speed using a pressure sensor. According to the calculation formula for compressive strength test:

[0071]

[0072] in, It is compressive strength; It is the pressure value applied to the material; It is the actual area of ​​the material subjected to force;

[0073] The tensile test was performed on PLA biodegradable plastic, recycled concrete and bamboo fiber composite material using a universal testing machine to measure elongation.

[0074] The abrasion resistance test was performed on PLA biodegradable plastic, recycled concrete and bamboo fiber composite material using a Martindale abrasion tester.

[0075] The durability test simulates extreme climates using an accelerated aging chamber, and the durability is evaluated based on the aging rate; the formula for the aging rate is:

[0076]

[0077] in, It refers to the material aging rate; It represents the change in mass of the material during accelerated aging tests; t is the duration of the aging test.

[0078] The chemical performance tests include corrosion resistance tests and fire resistance tests;

[0079] The corrosion resistance test was conducted using an electrochemical workstation with NaCl solution as the corrosion medium. The corrosion resistance was assessed based on the electrochemical corrosion rate; the formula for the electrochemical corrosion rate is:

[0080]

[0081] in, It is the corrosion rate; It is the current value during the corrosion process; It is the number of electrons transferred; It is the surface area of ​​the material that participates in the corrosion reaction;

[0082] The fire resistance test was conducted according to GB / T 8624-2012 standard, with PLA biodegradable plastic classified as B1 grade, recycled concrete as A grade, and bamboo fiber composite material as B2 grade.

[0083] The biological performance tests include biodegradation tests and antibacterial performance tests;

[0084] The biodegradation test involves burying the three materials in community soil and measuring the degradation rate after 6 months.

[0085] The antibacterial performance test was performed on the three materials using the antibacterial zone method.

[0086] The ecological effectiveness assessment employs landscape function analysis methods for transect layout and observation. These methods assess the landscape function through two main categories: landscape structure and soil surface characteristics. The landscape structure includes a first single indicator and a first composite indicator. The soil surface characteristics include a second single indicator and a second composite indicator.

[0087] The first single indicator includes the total vegetation area intercepted by the transect. Maximum coverage area The total length of the area covered by the transect line and profile length The first composite index includes ground-based indicators. and landscape structure indicators ;

[0088] The ground index The calculation formula is:

[0089]

[0090] The landscape structure index The calculation formula is:

[0091] ;

[0092] The second single index includes surface cover rate C, number of dead leaves and leaf layers D, coverage rate of cryptogamic plants E, quantitative value of crust fragmentation F, quantitative value of soil erosion grade G, sediment accumulation H, resistance to disturbance coefficient J, quantitative value of disintegration test results Q, coverage rate of perennial plants L, quantitative value of surface roughness T, and soil texture parameter W; the second composite index includes soil stability index Sa, water infiltration index If, and nutrient cycling index Ne.

[0093] The formula for calculating the soil stability index Sa is as follows:

[0094]

[0095] in, Let be the weight coefficient of the i-th sample; The total number of samples involved in the calculation; Let be the land cover rate of the i-th sample; Let be the number of layers of dead leaves in the i-th sample; Let be the coverage rate of cryptogams in the i-th sample; This is the quantified value of the degree of crust breakage for the i-th sample; This represents the quantified value of the soil erosion level for the i-th sample. Let be the amount of sediment accumulation in the i-th sample; Let be the anti-interference capability coefficient of the i-th sample; This is the quantified value of the disintegration test result for the i-th sample;

[0096] The formula for calculating the water permeability index If is:

[0097]

[0098] in, Let be the plant coverage of the i-th sample; Let be the quantified value of the surface roughness of the i-th sample; Let be the soil texture parameters for the i-th sample;

[0099] The formula for calculating the nutrient cycling index Ne is as follows:

[0100] .

[0101] The screening process combines systematic clustering, TOPSIS, and expert scoring and weighting to screen materials for adaptability and sustainability.

[0102] Two types of standard modules are manufactured using 3D printers: basic modules and biological carrier modules.

[0103] The basic module measures 40cm×40cm×15cm and has a reserved concave-convex splicing interface to ensure the independence and synergy between modules;

[0104] The biological support module measures 40cm×40cm×20cm, has a built-in 10cm thick nutrient substrate, and is reserved with irrigation holes and sensor mounting slots.

[0105] Indicator organisms were inoculated using micro-landscape modules as growth carriers. Moss was evenly spread in the biological carrier modules, and three Crassulaceae plants were planted in each module. After inoculation, the plants were placed in an environment of 25°C and 60% humidity for 7 days to allow them to recover and ensure that the roots were fully integrated with the substrate.

[0106] The climate biological monitoring module has five sets of monitoring devices deployed in the center and four corners of the micro-landscape module group. The temperature sensor and hygrometer are installed at a height of 1.5m, avoiding direct sunlight and rain splash, and are connected to the data acquisition terminal via data cable.

[0107] The temperature sensor is used to record temperature data and calculate the average temperature over a certain period using an average temperature formula; the average temperature formula is expressed as:

[0108]

[0109] in, It is the average temperature over a certain period of time; It refers to the number of monitoring sessions within a specific time period; This refers to a single temperature reading; the monitoring period is set to 1 day, with temperature data collected every 30 minutes. Range: 25℃-33℃.

[0110] The hygrometer is used to record humidity data and analyze the trend of humidity change using a humidity change rate formula; the humidity change rate formula is expressed as:

[0111]

[0112] in, It is the rate of change of humidity; It is the change in humidity during the monitoring period; The monitoring duration is 12 hours.

[0113] The indicative biological quantitative characteristic data collection includes at least one of the following: growth rate of organisms, leaf morphology variation coefficient, and content of physiological metabolites.

[0114] The growth rate was measured every 7 days using a ruler to determine the height of the Crassulaceae plant. The growth rate was calculated as (height in week 2 - height in week 1) / 7 days.

[0115] The leaf morphology variation coefficient is calculated by taking pictures of the leaves with an image recognition device and calculating the aspect ratio. The variation coefficient is calculated as: (Aspect ratio of the second week - Aspect ratio of the first week) / Aspect ratio of the first week × 100%.

[0116] The content of physiological metabolites was determined by HPLC, and the content of flavonoids in the leaves of Crassulaceae plants was detected. The time-series data such as temperature, humidity change rate and biological growth rate collected in the climate biological monitoring module were normalized by Min-Max, and the samples were divided by sliding window method to construct training set, validation set and test set according to 7:2:1.

[0117] A 3D model of the micro-landscape was constructed using GIS software, and spatial feature parameters were extracted to form a 64×64×3 spatial feature matrix, which corresponds to length, width, and feature dimension, respectively.

[0118] like Figure 2 As shown, a CNN-LSTM hybrid model is used for climate trend prediction and micro-landscape growth status prediction.

[0119] The CNN-LSTM hybrid model consists of an input layer, a CNN spatial feature extraction module, an LSTM temporal feature capture module, a feature fusion module, a fully connected layer, and an output layer.

[0120] The input layer receives three types of channel data: channel 1 is time-series climate data: temperature and humidity change rate; channel 2 is time-series biological data: growth rate, leaf morphology variation coefficient; and channel 3 is a spatial feature matrix: 64×64×3.

[0121] The CNN spatial feature extraction module has five layers. The first layer is a 3×3 convolutional layer with 32 kernels, a stride of 1, "same" padding, ReLU activation function, and an output dimension of 64×64×32. The second layer is a pooling layer containing 2×2 max pooling with a stride of 2 and an output dimension of 32×32×32. The third layer is a 3×3 convolutional layer with 64 kernels, a stride of 1, "same" padding, ReLU activation function, and an output dimension of 32×32×64. The fourth layer is a pooling layer containing 2×2 max pooling with a stride of 2 and an output dimension of 16×16×64. The fifth layer is a flattening layer that converts the three-dimensional feature map into a one-dimensional vector with an output dimension of 16×16×64=16384.

[0122] The LSTM temporal feature capture module comprises a temporal data concatenation layer, two LSTM layers, and two Dropout layers. The temporal data concatenation layer concatenates the data from channel 1 and channel 2 into a 24×4 temporal sequence. The first LSTM layer has 128 hidden units, an activation function of tanh, return_sequences=True, and an output dimension of 24×128. This is followed by a Dropout layer with a dropout rate of 0.2 and an output dimension of 24×128. The second LSTM layer has 64 hidden units, an activation function of tanh, return_sequences=False, and an output dimension of 64. Finally, there is a second Dropout layer with a dropout rate of 0.2 and an output dimension of 64.

[0123] The feature fusion and fully connected layer are formed by concatenating the CNN output vector with the LSTM output vector to create a 16448-dimensional fusion vector. This vector is then passed through a first fully connected layer with 256 neurons, a ReLU activation function, and an output dimension of 256. Next, a Dropout layer with a dropout rate of 0.2 and an output dimension of 256 is passed through it. Finally, a second fully connected layer with 128 neurons, a ReLU activation function, and an output dimension of 128 is passed through it.

[0124] The output layer adopts a dual-branch output: the climate trend branch outputs the average temperature and humidity change rate for the next 7 days; the growth status branch outputs the growth rate and leaf morphology variation coefficient of Crassulaceae plants for the next 7 days.

[0125] The model was trained using the Adam optimizer with a learning rate of 0.001 and a joint loss function of MSE1+MSE2, where MSE1 is the climate prediction loss and MSE2 is the growth state prediction loss, with 150 iterations.

[0126] Using GIS software, a modular model was constructed to simulate three combination schemes: row-coordinate, distributed, and nested.

[0127] The columnar design uses 3 basic modules and 2 biological support modules, spliced ​​together in a 2×3 layout with a module spacing of 12cm, providing excellent ventilation.

[0128] The distributed system uses modules randomly arranged at intervals of 8-15cm, resulting in high space utilization.

[0129] The embedded module is a biological support module embedded in the center of the basic module with a spacing of 10cm, which provides strong stability.

[0130] The optimal combination is selected through a Lowest Performance Assessment (LFA); such as... Figure 3As shown, a genetic algorithm is used for optimization. The genetic algorithm achieves multi-objective collaborative optimization of micro-landscape module combination, climate adaptation strategy and ecological efficiency through hybrid encoding, initial population generation, fitness function construction, selection-crossover-mutation genetic operations and multiple termination condition determination, and outputs Pareto optimal solution set to support the implementation of modular practice path.

[0131] The hybrid coding includes discrete variable coding and continuous variable coding; the discrete variable coding adopts binary, with a module combination of 3 basic modules and 2 biological support modules; the code is 10101, where 1 represents a basic module and 0 represents a biological support module; the continuous variable coding includes irrigation amount, light duration and module spacing, and is coded as [irrigation amount value, light duration value, module spacing value];

[0132] Initial population generation: The population size is 100. Individuals that meet the constraints are randomly generated. After feasibility screening, 95 valid individuals are retained.

[0133] Based on the LFA composite index and the optimization objective, the fitness function is defined as follows:

[0134]

[0135] in, It is an ecological efficiency value. ; It is a climate adaptability value. , It is a score for climate response speed. It is the score of biological growth fitness rate; Practicality value, , Standardized scores for modules, For generality score; after calculating the fitness value of each individual in the population using the fitness function F, optimization is completed according to the following steps, and the final control scheme is output:

[0136] A roulette wheel selection plus elite retention strategy is adopted. First, the roulette wheel selection probability is allocated according to the proportion of the individual fitness value to the total fitness of the population. The higher the fitness value of an individual, the greater the probability of it being selected to participate in reproduction. At the same time, the top 10% of the best individuals in the population are directly retained to enter the next generation to avoid the loss of excellent genes and ensure that the optimization direction does not deviate from the core objective.

[0137] Execution is categorized based on the characteristics of mixed encoding:

[0138] The discrete variable encoding adopts a single-point crossover method, randomly selecting the 3rd bit of the binary encoding string as the crossover point, swapping the corresponding segments of the two parents, and obtaining the crossover probability.

[0139] The continuous variables are obtained using the arithmetic crossover method, with weighting coefficients set. =0.4, according to the formula Calculate the offspring parameters to ensure they remain within a reasonable range; whereby... For irrigation amount of offspring; and These are the irrigation amounts for parent generation 1 and parent generation 2, respectively.

[0140] To avoid premature convergence of the population, mutation is performed based on the population type:

[0141] The bit-flipping mutation method is used for discrete variables. One bit of the binary encoded string is randomly selected and flipped. After mutation, the module combination constraints are verified. If they are violated, the mutation is repeated.

[0142] For continuous variables, Gaussian mutation method is used to introduce random disturbances, and the mutation amplitude is controlled within a reasonable range of ±10% to avoid parameter failure.

[0143] The termination condition is determined by monitoring the optimal fitness value of the population in real time during the iteration process. The iteration stops when any of the following conditions are met:

[0144] (1) The number of iterations reaches the preset maximum value of 100 generations.

[0145] (2) The variation range of the optimal fitness value of the population over 15 consecutive generations is ≤0.002.

[0146] Finally, the optimal result is output. When the iteration terminates, the top 20% of individuals in terms of fitness are extracted to form the Pareto optimal solution set.

[0147] Substituting the above optimal solution into the effect evaluation formula for verification, the effect evaluation formula is expressed as follows:

[0148]

[0149] in, It is the amount of change in effect; These are the indicator values ​​after the system has been running; These are the indicator values ​​before the system is running.

[0150] The data processing module encapsulates the optimal solution into a formal control scheme, which includes a module combination diagram, parameter settings, and execution timing table, and transmits it to the adaptive adjustment module through a data interface.

[0151] After receiving the control scheme, the adaptive adjustment module completes the initial adjustment in two steps: the first step is to adjust the module combination structure; the second step is to control the environmental parameters, start the automatic irrigation system, and set to spray twice a day at 6:00 and 18:00 to ensure that the growth needs of Crassulaceae plants and mosses are matched.

[0152] By continuously collecting micro-landscape adaptation data under different climatic conditions through dynamic intervention experiments, and iteratively optimizing the control logic based on feedback results, a response to multi-dimensional climate change can be achieved.

[0153] During the implementation of the control plan, based on the dynamic intervention experiment, system operation data was collected every 10 days: Regarding climate and environmental parameters, real-time temperature and humidity change rates were collected every 30 minutes using sensors; regarding indicative biological quantitative characteristics, the growth rate of Crassulaceae plants was measured using a ruler, leaf morphology variation coefficients were calculated using image recognition technology, and flavonoid content was detected by HPLC; regarding LFA composite indicators, three 10m transects were re-laid, and measurements were taken and calculated according to the corresponding formulas. , The data collected (Sa, If, and Ne) is fed back to the data processing module to update the CNN-LSTM model training set and the fitness function parameters of the genetic algorithm. If the actual temperature exceeds the predicted value by 1.5℃ for three consecutive days, the adaptive module automatically triggers an emergency adjustment, reducing the module spacing to 12cm to improve heat dissipation efficiency and increasing the irrigation volume to 28ml per day to supplement water evaporation. If the moss coverage is found to be below 55%, the humidity change rate threshold is adjusted from 2.5%RH / h to 2.0%RH / h. When the humidity change rate is below this threshold, the spray humidification is automatically activated to maintain a suitable humidity environment. The model is iterated and the optimization algorithm is updated every 30 days to ensure that the control logic matches the dynamic climate conditions and biological growth status in real time.

[0154] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A climate change-responsive modular micro-landscape adaptive monitoring system, characterized in that, The system includes: a micro-landscape module, a climate and biological monitoring module, a data processing module, and an adaptive adjustment module; The micro-landscape modules are made from selected sustainable materials and integrate climate change-sensitive indicator organisms; The climate biological monitoring module is used to collect climate and environmental parameter data and indicative biological quantitative characteristic data; The data processing module is used to receive the climate and environmental parameter data and the indicative biological quantitative characteristic data, analyze them, and output the control plan; The adaptive adjustment module dynamically adjusts the micro-landscape module according to the control scheme to adapt to climate change, while ensuring the stable generation of indicator organisms and the validity of monitoring data.

2. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 1, characterized in that, The micro-landscape module is made of materials selected through multi-dimensional performance testing and ecological effectiveness assessment. The multi-dimensional performance tests include compressive strength testing, durability testing, and corrosion resistance testing; The compressive strength test evaluates the material by calculating its compressive strength; the durability test evaluates the material based on its aging rate; and the corrosion resistance test evaluates the material based on its electrochemical corrosion rate. The ecological effectiveness assessment employs landscape function analysis methods for transect layout and observation. These methods assess the landscape function through two main categories: landscape structure and soil surface characteristics. The landscape structure includes a first single indicator and a first composite indicator. The soil surface characteristics include a second single indicator and a second composite indicator.

3. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 2, characterized in that, The first single indicator includes the total vegetation area intercepted by the transect. Maximum coverage area The total length of the area covered by the transect line and profile length The first composite index includes ground-based indicators. and landscape structure indicators ; The ground index The calculation formula is: The landscape structure index The calculation formula is: ； The second single index includes surface cover rate C, number of dead leaves and leaf layers D, coverage rate of cryptogamic plants E, quantitative value of crust fragmentation F, quantitative value of soil erosion grade G, sediment accumulation H, resistance to disturbance coefficient J, quantitative value of disintegration test results Q, coverage rate of perennial plants L, quantitative value of surface roughness T, and soil texture parameter W; the second composite index includes soil stability index Sa, water infiltration index If, and nutrient cycling index Ne. The formula for calculating the soil stability index Sa is as follows: in, Let be the weight coefficient of the i-th sample; The total number of samples involved in the calculation; Let be the land cover rate of the i-th sample; Let be the number of layers of dead leaves in the i-th sample; Let be the coverage rate of cryptogams in the i-th sample; This is the quantified value of the degree of crust breakage for the i-th sample; This represents the quantified value of the soil erosion level for the i-th sample. Let be the amount of sediment accumulation in the i-th sample; Let be the anti-interference capability coefficient of the i-th sample; This is the quantified value of the disintegration test result for the i-th sample; The formula for calculating the water permeability index If is: in, Let be the plant coverage of the i-th sample; Let be the quantified value of the surface roughness of the i-th sample; Let be the soil texture parameters for the i-th sample; The formula for calculating the nutrient cycling index Ne is as follows: 。 4. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 2, characterized in that, The material selection for the micro-landscape module also includes tensile testing, abrasion resistance testing, and durability testing in physical performance testing; corrosion resistance testing and fire resistance testing in chemical performance testing; and biodegradation testing and antibacterial performance testing in biological performance testing. The adaptability and sustainability of materials were screened by combining systematic clustering, TOPSIS, and expert scoring weighting methods.

5. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 1, characterized in that, The climate biological monitoring module includes indicator organisms and monitoring equipment. The indicator organisms are selected from climate change-sensitive organisms identified through literature screening in the fields of ecology, botany, agronomy, and hydrology, and use the micro-landscape module as a growth carrier. The monitoring equipment includes a temperature sensor and a hygrometer. The temperature sensor is used to record temperature data and calculate the average temperature over a certain period using an average temperature formula. The hygrometer is used to record humidity data and analyze the humidity change trend using a humidity change rate formula.

6. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 1, characterized in that, The climate and environmental parameter data include the average temperature and humidity change rates over a certain period of time; The indicative biological quantitative data include at least one of the following: growth rate of the organism, coefficient of variation of leaf morphology, and content of physiological metabolites.

7. The climate change-responsive modular micro-landscape adaptive monitoring system according to claim 1, characterized in that, The data processing module is used to receive the climate and environmental parameter data and the indicative biological quantitative characteristic data, construct a modular micro-landscape model using GIS or related software, integrate deep learning algorithms to obtain climate trend prediction and micro-landscape growth status prediction, calculate the changes in micro-landscape adaptation effect through comparative analysis and effect evaluation formula, and optimize the design of the micro-landscape module using optimization algorithms to output a control scheme.

8. A climate change-responsive modular micro-landscape adaptive monitoring system according to claim 7, characterized in that, The effect evaluation formula is expressed as follows: in, It is the amount of change in effect; These are the indicator values ​​after the system has been running; These are the indicator values ​​before the system is running.

9. A climate change-responsive modular micro-landscape adaptive monitoring system according to claim 7, characterized in that, The deep learning algorithm uses a hybrid model of LSTM and CNN, CNN-LSTM, for climate trend prediction and micro-landscape growth status prediction; the optimization algorithm is a genetic algorithm, used for optimizing the design of micro-landscape modules.

10. A climate change-responsive modular micro-landscape adaptive monitoring system according to claim 1, characterized in that, The adaptive adjustment module dynamically adjusts the combination structure and generation environment parameters of each modular micro-landscape unit according to the control scheme output by the data processing module. It continuously collects micro-landscape adaptation data under different climatic conditions through dynamic intervention experiments, and iteratively optimizes the control logic based on the feedback results to achieve a response to multi-dimensional climate change.

Images