Ecological space water allocation strategy optimization method and system

Abstract

The invention discloses an ecological space water allocation strategy optimization method and system, and relates to the technical field of ecological water. Comprising the following steps: collecting and preprocessing an infrared spectrogram of a plant with known chlorophyll content, and establishing a neural network model to predict the chlorophyll content. Dividing the ecological space of the to-be-distributed water source into a plurality of sub-regions, randomly selecting plants, collecting an infrared spectrogram, and inputting the infrared spectrogram into the prediction model to obtain a chlorophyll content prediction value. Calculating a plant water demand influence index according to the chlorophyll content and the plant growth characteristic parameters, and obtaining soil characteristic parameters to calculate a soil water source preservation coefficient. The method comprises the following steps: acquiring environment characteristic parameters, calculating an environment water supply influence index, generating a water supply demand index by combining a soil water source preservation coefficient and a plant water demand influence index, comparing the water supply demand index with a water supply demand threshold range, generating a corresponding water use allocation strategy, and improving the overall health of an ecological system and the growth efficiency of plants.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological water use, and in particular to an optimization method and system for water allocation strategies for ecological space water use. Background Art

[0002] In modern agriculture and ecological environment management, the effective allocation and management of water resources are crucial. With the intensification of global climate change and human activities, the problem of water resource shortage has become increasingly serious, especially in arid and semi-arid regions, where the contradiction between the water demand of plants and the available water resources has become more prominent. This problem not only affects the growth and yield of crops but also has a negative impact on the stability of the ecosystem. Traditional water resource allocation methods often rely on experience or static models, ignoring the dynamic relationship between the growth state of plants and soil characteristics. Therefore, there is an urgent need for a scientific water resource allocation strategy based on plant growth characteristics and environmental parameters to achieve the optimal allocation of water resources.

[0003] In the prior art, using chlorophyll content as an indicator of plant growth status and combining remote sensing technology and spectral analysis methods for ecological monitoring has gradually attracted the attention of scientific research and agricultural practice. However, the existing methods usually have problems such as inaccurate data processing, poor model adaptability, and limited service areas, resulting in unsatisfactory effects in practical applications. At the same time, ignoring the dynamic relationship between the growth state of plants and soil characteristics greatly reduces the accuracy and efficiency of water resource allocation and cannot meet the specific needs of different ecological spaces. Therefore, it is particularly important to develop a new type of water allocation strategy for ecological space, which provides accurate water resource allocation plans for different sub-regions through an efficient prediction model combined with multi-dimensional characteristic parameters.

[0004] In the prior art, the authorized announcement number CN116777037B discloses an optimization method for ecological water distribution strategies. Specifically, different decision-making units are taken as the research objects, and multi-objective programming, bilevel programming, and fuzzy-interval credibility constraint programming methods are introduced into the optimization model. Considering the link relationship of water-food-carbon-cultivated land-ecology-nutrition, a bilevel multi-objective fuzzy-interval credibility constraint programming model is developed to solve the problems of agricultural and ecological water resource allocation in arid agricultural areas of northwest China under uncertain conditions. In this method, the total system benefit, ecological satisfaction, and nutrient water productivity are the upper-level objectives, while the economic benefit and grain yield are the lower-level objectives. The coordination variables between the upper and lower levels are the agricultural and ecological water allocation amounts. During the decision-making process, the upper-level optimization results will affect the lower-level objectives and constraints, and the lower-level optimization results also need to be fed back to the upper-level decision-makers to achieve the interaction between the upper and lower levels and obtain the optimal decision. However, in this solution, the actual characteristic parameters of plants (such as the average plant height, average plant leaf area, and average main stem diameter of plants) are not considered, which may lead to unreasonable water resource allocation. If the water resource allocation fails to be adjusted in time, it may cause water shortage in plants during the critical growth stage, thereby affecting the yield and quality. At the same time, the physical and chemical properties of the soil (such as permeability, water content, nutrient status, etc.) will change with time and environmental conditions. Ignoring the dynamic changes of soil properties may lead to incorrect assessments of water retention capacity and water supply capacity. Therefore, the accuracy and effectiveness of the allocated water sources are reduced.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a coal mine supervision method and system based on video images to solve the problems raised in the above background art.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] An optimization method for water distribution strategies in ecological space, the specific steps include:

[0009] Collect infrared spectrograms of several plants with known chlorophyll contents, preprocess the infrared spectrograms of the plants to obtain training sample infrared spectrograms, establish a neural network model based on the training sample infrared spectrograms, use the training sample infrared spectrograms as inputs, and use the corresponding chlorophyll contents as labels to train the neural network model to obtain a chlorophyll content prediction model;

[0010] The ecological space of the water source to be allocated is divided into several sub-regions with equal areas. A number of plants are randomly selected in each sub-region, and their infrared spectrograms are collected. The collected infrared spectrograms are input into the trained chlorophyll content prediction model to obtain the predicted chlorophyll content values of each plant. Based on the average value of the predicted chlorophyll content values of each plant, the chlorophyll content of the plants in this sub-region is characterized;

[0011] According to the chlorophyll content of the plants in the sub-region and combined with the plant growth characteristic parameters of this sub-region, the plant water demand influence index is calculated. At the same time, the soil characteristic parameters in each sub-region are obtained. Based on the collected soil characteristic parameters, the soil water source preservation coefficient is calculated. The soil characteristic parameters include the average soil permeability, the average soil organic matter content, and the average soil particle size. The plant growth characteristic parameters include the average plant height, the average plant leaf area, and the average main stem diameter of the plant;

[0012] The environmental characteristic parameters of each sub-region are obtained. Based on the environmental characteristic parameters, the environmental water supply influence index is calculated. According to the obtained environmental water supply influence index, combined with the soil water source preservation coefficient and the plant water demand influence index, the corresponding water supply demand index of this sub-region is comprehensively calculated. The water supply demand index of each sub-region is compared with the set water supply demand threshold range. According to the comparison result, the corresponding water use allocation strategy is generated. The environmental characteristic parameters include the average environmental temperature, the average humidity, the average rainfall, and the average wind speed.

[0013] Furthermore, preprocessing is performed on the plant infrared spectrogram, where the preprocessing includes image denoising preprocessing and image normalization preprocessing;

[0014] Among them, based on the long short-term memory network model (LSTM model), a chlorophyll content prediction model is established. An activation function and an optimization algorithm are selected. The Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm for the LSTM model; The formula of the Tanh function is:

[0015]

[0016] In the formula, f(r) represents the Tanh function, and the independent variable r represents the input weighted sum of the neuron, that is, the result after the input received by the neuron from the previous layer is weighted and summed;

[0017] At the same time, the hyperparameters of the LSTM model are set. The hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing quantity, and the number of neurons in the hidden layer;

[0018] Among them, the number of network layers is set to a 3-layer network structure, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch processing quantity is set to 256, and the number of neurons in the hidden layer is 32;

[0019] The input of the chlorophyll content prediction model after training is the infrared spectrogram of the plant, and the output is the predicted value of the chlorophyll content of the plant.

[0020] Furthermore, according to the chlorophyll content of the plants in the sub-region and combined with the plant growth characteristic parameters of the sub-region, the plant water requirement influence index is calculated. The specific formula based on which the plant water requirement influence index is calculated is:

[0021]

[0022] In the formula, EIC is the plant water requirement influence index, C is the chlorophyll content of the plants in the sub-region, HL is the average height of the plants in the sub-region, Mt is the average leaf area of the plants in the sub-region, and DH is the average diameter of the main stems of the plants in the sub-region.

[0023] Furthermore, based on the collected soil characteristic parameters, the soil water source preservation coefficient is calculated. The specific formula based on which the soil water source preservation coefficient is calculated is:

[0024]

[0025] In the formula, SCI represents the soil water source preservation coefficient, OM is the average content of soil organic matter in the sub-region, K is the average soil permeability in the sub-region, D mean is the average soil particle size in the sub-region;

[0026] The specific method for obtaining the soil permeability K is as follows: Select the soil sample to be tested and place it in a cylindrical permeameter. The sample needs to be fully saturated before being placed. Set a constant water level difference, that is, maintain a constant height difference between the water source in the permeameter and the water level above the soil sample. Record the amount of water flowing through the soil sample within the time period t. Based on the recorded data, calculate the soil permeability. The specific formula based on which the soil permeability K is calculated is:

[0027]

[0028] In the formula, Q is the amount of water flowing through the soil sample within the time period t, L is the thickness of the soil sample in the permeameter, PF is the cross-sectional area of the soil sample in the permeameter, and h is the set water level difference.

[0029] Furthermore, obtain the environmental characteristic parameters of each sub-region, and calculate the environmental water supply influence index based on the environmental characteristic parameters. The specific formula based on which the environmental water supply influence index is calculated is:

[0030]

[0031] Wherein, WEC is the environmental water supply impact index, RH is the average environmental humidity, T is the average environmental temperature, V is the average environmental wind speed, RH 0 is the reference humidity, T 0 is the reference environmental temperature, Rs is the average rainfall, specifically the annual average rainfall.

[0032] Furthermore, based on the obtained environmental water supply impact index, combined with the soil water source preservation coefficient and the plant water demand impact index, the water supply demand index corresponding to the sub-region is comprehensively calculated and generated. The formula for calculating the water supply demand index is as follows:

[0033]

[0034] Wherein, ZH is the water supply demand index, SCI represents the soil water source preservation coefficient, ω 1 , ω 2 and ω 3 are the weight coefficients of the environmental water supply impact index, the plant water demand impact index, and the soil water source preservation coefficient respectively, where ω 2 > ω 3 ≥ ω 2 , ω 1 + ω 2 + ω 3 = 1, and ω 1 , ω 2 and ω 3 are all greater than 0;

[0035] The water supply demand index of each sub-region is compared with the set water supply demand threshold range. According to the comparison result, the corresponding water use allocation strategy is generated. The specific judgment logic is as follows:

[0036] When yz 1 ≤ ZH ≤ yz 2 , it is judged that the water supply of the sub-region does not need to be adjusted;

[0037] When ZH < yz 1 , it is judged that the water demand of the sub-region is small, and the initial water supply should be reduced;

[0038] When ZH > yz 2 , it is judged that the water demand of the sub-region is large, and the initial water supply should be increased;

[0039] Wherein, yz 1 and yz 2 are respectively the lower and upper limits of the preset water supply demand threshold range, where yz2 >yz 1 。

[0040] The present invention also provides an optimization system for the water allocation strategy in the ecological space. The optimization system for the water allocation strategy in the ecological space is used to execute the above-mentioned optimization method for the water allocation strategy in the ecological space, and includes:

[0041] A prediction network training module, which is used to collect infrared spectrograms of several plants with known chlorophyll contents, preprocess the infrared spectrograms of the plants to obtain training sample infrared spectrograms, establish a neural network model based on the training sample infrared spectrograms, use the training sample infrared spectrograms as inputs, and use the corresponding chlorophyll contents as labels to train the neural network model to obtain a chlorophyll content prediction model;

[0042] A regional plant information prediction module, which is used to equally divide the ecological space of the water source to be allocated into several sub-regions, randomly select several plants in each sub-region, collect their infrared spectrograms, input the collected infrared spectrograms into the trained chlorophyll content prediction model to obtain the predicted chlorophyll content values of each plant, and characterize the chlorophyll content of the plants in the sub-region based on the average value of the predicted chlorophyll content values of each plant;

[0043] A soil characteristic analysis module, which is used to calculate a plant water demand influence index according to the chlorophyll content of the plants in the sub-region and in combination with the plant growth characteristic parameters of the sub-region, and at the same time obtain the soil characteristic parameters in each sub-region, and calculate and generate a soil water source preservation coefficient based on the collected soil characteristic parameters, wherein the soil characteristic parameters include the average soil permeability, the average soil organic matter content, and the average soil particle size, and the plant growth characteristic parameters include the average plant height, the average plant leaf area, and the average main stem diameter of the plant;

[0044] A water source precise allocation module, which is used to obtain the environmental characteristic parameters of each sub-region, calculate an environmental water supply influence index based on the environmental characteristic parameters, and comprehensively calculate and generate a corresponding water supply demand index for the sub-region according to the obtained environmental water supply influence index, in combination with the soil water source preservation coefficient and the plant water demand influence index, compare the water supply demand index of each sub-region with the set water supply demand threshold range, and generate a corresponding water allocation strategy according to the comparison result, wherein the environmental characteristic parameters include the average environmental temperature, the average humidity, the average rainfall, and the average wind speed.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] First, by collecting the infrared spectra of plants with known chlorophyll contents and establishing a prediction model based on deep learning, the prediction of chlorophyll content becomes more accurate. This data-driven model can not only reflect the growth status of plants in real time but also respond promptly to environmental changes. Secondly, through the systematic division of the ecological space, the water source area to be allocated is subdivided into several sub-areas, making the water resource allocation more targeted. By statistically analyzing the predicted chlorophyll content values of plants in each sub-area, a more accurate assessment of the water requirements of plants can be made. At the same time, by combining soil characteristic parameters and environmental characteristic parameters, all factors affecting plant growth can be comprehensively considered. This comprehensive data analysis and model calculation greatly improve the scientificity and rationality of water resource allocation. In addition, based on the comprehensive calculation of the plant water demand impact index, the soil water source preservation coefficient, and the environmental water supply impact index, a water supply demand index for each sub-area can be formed. By comparing the set threshold with the water supply demand index, corresponding water use allocation strategies are generated to ensure that each sub-area obtains the required water, thereby improving the overall health of the ecosystem and the growth efficiency of plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a schematic flow chart of the overall method of the present invention;

[0048] Figure 2 is a schematic structural diagram of the overall system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0050] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those of ordinary skill in the field to which the present invention pertains. The "first", "second", and similar terms used in the present invention do not denote any order, quantity, or importance, but are only used to distinguish different components. The terms "comprising" or "including" and the like mean that the elements or items appearing before this term cover the elements or items listed after this term and their equivalents, without excluding other elements or items. The terms "connected" or "coupled" and the like are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0051] Embodiment:

[0052] Please refer to Figure 1 , the present invention provides a technical solution:

[0053] An optimization method for the water allocation strategy of ecological space, the specific steps include:

[0054] Step 1: Collect the infrared spectrograms of several plants with known chlorophyll contents, preprocess the infrared spectrograms of the plants to obtain the training sample infrared spectrograms, and based on the training sample infrared spectrograms, establish a neural network model. Use the training sample infrared spectrograms as the input and the corresponding chlorophyll contents as the labels to train the neural network model to obtain a chlorophyll content prediction model.

[0055] Preprocess the infrared spectrograms of the plants, where the preprocessing includes image denoising preprocessing and image normalization preprocessing;

[0056] Among them, the denoising method of wavelet transform is used to denoise the image. The specific steps of the wavelet transform denoising method include: decomposing the image after distortion correction through wavelet transform to obtain the wavelet coefficients of the image at different scales and directions; performing threshold processing on the wavelet coefficients, setting the wavelet coefficients with low amplitudes to zero, and retaining the wavelet coefficients with high amplitudes; performing inverse transform on the wavelet coefficients after threshold processing to reconstruct the processed coefficients into an image to complete the image denoising process.

[0057] Image normalization is a common preprocessing method, and the specific method steps are not elaborated here.

[0058] Among them, based on the long short-term memory network model (LSTM model), a chlorophyll content prediction model is established, and an activation function and an optimization algorithm are selected. Among them, the Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm for the LSTM model; the formula of the Tanh function is:

[0059]

[0060] In the formula, f(r) represents the Tanh function, and the independent variable r represents the input weighted sum of the neuron, that is, the result after the input received by the neuron from the previous layer is weighted and summed;

[0061] At the same time, set the hyperparameters of the LSTM model. The hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing quantity, and the number of neurons in the hidden layer;

[0062] Among them, the number of network layers is set to a 3-layer network structure, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch processing quantity is set to 256, and the number of neurons in the hidden layer is 32;

[0063] The input of the trained chlorophyll content prediction model is the infrared spectrogram of the plant, and the output is the predicted value of the chlorophyll content of the plant.

[0064] Through its unique structure, the LSTM network can effectively learn and memorize long-term and short-term dependencies. This is crucial for chlorophyll content prediction because the changes in chlorophyll are affected by multiple time points, especially during the growing season, and the impact on plant growth is often time-related.

[0065] Traditional recurrent neural networks (RNNs) are prone to problems such as vanishing gradients and exploding gradients when dealing with long sequence data. The LSTM effectively alleviates these problems by introducing a gating mechanism (input gate, forget gate, and output gate), enabling the model to better learn and predict long sequences.

[0066] Step 2: Divide the ecological space of the water source to be allocated into several sub-regions of equal area. Randomly select several plants in each sub-region, collect their infrared spectrograms, input the collected infrared spectrograms into the trained chlorophyll content prediction model to obtain the predicted chlorophyll content values of each plant, and characterize the chlorophyll content of the plants in the sub-region based on the average value of the predicted chlorophyll content values of each plant.

[0067] The method steps for collecting the infrared spectrograms of the selected plants are as follows: Clean the surface of the plant leaves, remove dust, soil, and other attachments, align the probe of the infrared spectrometer with the leaf surface, press the start button to collect the infrared spectral data on the leaf surface, and record the sample number;

[0068] Use spectral analysis software to plot the processed spectral data into an infrared spectral image, preprocess the infrared spectral image, and input the preprocessed image into the trained chlorophyll content prediction model to obtain the predicted chlorophyll content values of each plant;

[0069] Calculate the average value of the predicted chlorophyll content values of all the selected plants, and use this average value as the chlorophyll content of the plants in the sub-region.

[0070] The absorption peaks of chlorophyll correspond to the chemical bonds present in its pigment structure. By analyzing the intensity and position of these peaks, the chlorophyll content can be indirectly estimated. The characteristic absorption peaks of chlorophyll are usually located in the near-infrared region (about 700 - 1400 nm) and the mid-infrared region (about 1500 - 1800 nm).

[0071] Step 3: According to the chlorophyll content of the plants in the sub-region, combined with the plant growth characteristic parameters of the sub-region, calculate the plant water demand impact index. At the same time, obtain the soil characteristic parameters within each sub-region, and calculate and generate the soil water source preservation coefficient based on the collected soil characteristic parameters. The soil characteristic parameters include the average soil permeability, the average soil organic matter content, and the average soil particle size, and the plant growth characteristic parameters include the average plant height, the average plant leaf area, and the average main stem diameter of the plant.

[0072] Soil organic matter is the general term for organic substances in the soil, and its content can be measured by the following methods: Randomly select multiple sampling points within the sub-region and collect soil samples. Usually, soil samples at a depth of 0-20 cm are collected, and both horizontal and vertical stratification can be carried out. Air-dry the collected soil samples and remove stones and impurities through a sieve (usually a 2-mm sieve). By the Walkley method: This is a commonly used method for measuring the content of soil organic matter. Through an oxidation-reduction reaction to consume organic substances, after calculating the organic carbon content and multiplying by a coefficient (usually 1.724), the soil organic matter content can be obtained. Calculate the average value based on the soil organic matter content of each sampling point, and use this average value as the average soil organic matter content of the sub-region.

[0073] Adopt the sedimentation principle. According to the difference in sedimentation rates of particles with different particle sizes in a liquid, measure the particle size distribution through specific equipment (such as a particle size analyzer) to obtain the average soil particle size of each sampling point. Calculate the average value based on the average soil particle size of each sampling point, and use this average value as the average soil particle size of the sub-region.

[0074] Randomly select several plants within the sub-region as sampling plants and measure them using a ruler, measuring tape, or laser rangefinder. A laser rangefinder can improve the measurement accuracy and efficiency. For taller plants, use professional plant height measurement tools such as a plant height meter. Calculate the average value based on the height data of each sampling plant, and use this average value as the average plant height of the sub-region. Use professional leaf area measurement instruments (such as LI-3000 or LI-3100) to directly measure the leaf area; or take photos with a camera and analyze the images through computer software to estimate the leaf area. Calculate the average value based on the leaf area of each sampling plant, and use this average value as the average plant leaf area of the sub-region.

[0075] Use a vernier caliper for precise measurement. A diameter gauge is suitable for thicker stems, and a measuring line can be used for quick measurement. Measure at the base of the main stem of the plant (about 5-10 cm above the ground), ensuring that the measurement position is consistent. Use a vernier caliper to directly measure the main stem diameter of each sampling plant and record the main stem diameter value of each sampling plant. Calculate the average value based on the main stem diameter of each sampling plant, and use this average value as the average main stem diameter of the plants in the sub-region.

[0076] Based on the chlorophyll content of the plants in the sub-region and combined with the plant growth characteristic parameters of the sub-region, calculate the plant water requirement impact index. The specific formula for calculating the plant water requirement impact index is as follows:

[0077]

[0078] In the formula, EIC is the plant water demand impact index, C is the chlorophyll content of plants in the sub-region, HL is the average height of plants in the sub-region, Mt is the average leaf area of plants in the sub-region, and DH is the average diameter of the main stems of plants in the sub-region.

[0079] It should be noted that the plant water demand impact index EIC is used to characterize the different water demands of plants under different structural characteristics and different chlorophyll contents. Among them, the larger the value of the plant water demand impact index EIC, the higher the water demand of the plant.

[0080] Among them, the chlorophyll content is a direct indicator of the photosynthetic ability of plants. Generally, the higher the chlorophyll content, the better the photosynthetic efficiency of plants, the more vigorous the growth, and the higher the water demand may be. Therefore, the chlorophyll content C of plants in the sub-region is proportional to the plant water demand impact index EIC. Through the exponential function e -C It indicates that the impact of the chlorophyll content on the different water demands of water resources shows exponential growth, emphasizing the strong dependence of photosynthesis on chlorophyll.

[0081] The plant height has a direct relationship with its water demand. Usually, tall plants need more water to maintain the physiological processes required for their growth. Therefore, the average height of plants in the sub-region is proportional to the plant water demand impact index EIC. In the form of a square root to smooth the impact of height on water demand and avoid over-amplifying the impact of height.

[0082] The average leaf area Mt of plants in the sub-region and the average diameter DH of the main stems of plants in the sub-region are both in the same way as the impact of plant height. The leaf area directly affects the photosynthetic capacity and transpiration. A larger leaf area usually means that the plant has higher photosynthesis and transpiration water loss. Therefore, the increase in leaf area will increase the water demand of the plant. At the same time, the main stem diameter is usually related to the health status and structural strength of the plant. A thicker stem may mean a stronger water transport capacity. Therefore, the average leaf area Mt of plants in the sub-region and the average diameter DH of the main stems of plants in the sub-region are both proportional to the plant water demand impact index EIC. It is processed by taking the square root to reduce the impact of extreme values on the overall calculation and reduce the over-shrinking of the impact on water demand. Through the logarithmic function to better handle the non-linear growth, especially in the complex relationship between plant growth and water demand.

[0083] Based on the collected soil characteristic parameters, the soil water source preservation coefficient is calculated. The formula for calculating the soil water source preservation coefficient is as follows:

[0084]

[0085] In the formula, SCI represents the soil water source preservation coefficient, OM is the average soil organic matter content in this sub-region, K is the average soil permeability in this sub-region, and D mean is the average soil particle size in this sub-region;

[0086] It should be noted that the larger the value of the soil water source preservation coefficient SCI, the stronger the water preservation ability of the soil area, so the less water resources are required correspondingly. On the contrary, more water resources are needed to make up for the lost water resources.

[0087] Among them, organic matter is an important factor in the water retention ability of the soil. A higher organic matter content usually means better water retention ability of the soil because organic matter can adsorb and store water. Therefore, the average soil organic matter content OM in this sub-region is directly proportional to the soil water source preservation coefficient SCI. The form of squaring OM 2 emphasizes the strong positive impact of organic matter on the water preservation ability. The use of the squaring operation indicates that the increase in organic matter contributes non-linearly to soil water preservation. As the organic matter content increases, the improvement of water retention ability will increase significantly.

[0088] Permeability refers to the ability of soil water to pass through the soil. A higher permeability means that water is likely to be lost, which may reduce the water retention ability of the soil. Therefore, the average soil permeability K in this sub-region is inversely proportional to the soil water source preservation coefficient SCI. The exponential function e K represents the exponential level of the permeability impact. High permeability will significantly reduce the SCI value, indicating a decline in the soil's ability to preserve water sources.

[0089] The particle size has an important impact on the water retention characteristics of the soil. Smaller particle sizes usually provide a higher specific surface area, thus enhancing the water retention ability. Therefore, the average soil particle size D in this sub-region mean is inversely proportional to the soil water source preservation coefficient SCI. The form of D mean 2 squared indicates that the impact of particle size is non-linear, and larger particle size values will significantly affect the water retention ability of the soil. Using the form of logarithm ln(1 + D mean 2 ) makes the impact of extremely large particle sizes undergo a certain smoothing process to avoid excessive amplification of the SCI value.

[0090] The specific method for obtaining the soil permeability K is as follows: Select the soil sample to be tested, put it into a cylindrical permeameter. The sample needs to be fully saturated before being put in. Set a constant water level difference, that is, maintain a constant height difference between the water source of the permeameter and the water level above the soil sample. Record the amount of water flowing through the soil sample within the time period t. Calculate the soil permeability based on the recorded data. Among them, the specific formula for calculating the soil permeability K is:

[0091]

[0092] In the formula, Q is the amount of water flowing through the soil sample within the time period t, L is the thickness of the soil sample in the permeameter, PF is the cross-sectional area of the soil sample in the permeameter, and h is the set water level difference.

[0093] Step 4: Obtain the environmental characteristic parameters of each sub-region, calculate the environmental water supply impact index based on the environmental characteristic parameters, and according to the obtained environmental water supply impact index, combine the soil water source preservation coefficient and the plant water demand impact index to comprehensively calculate and generate the water supply demand index corresponding to the sub-region. Compare the water supply demand index of each sub-region with the set water supply demand threshold range, and generate the corresponding water use allocation strategy according to the comparison result. The environmental characteristic parameters include the environmental average temperature, average humidity, average rainfall, and average wind speed.

[0094] Obtain the environmental characteristic parameters of each sub-region, and calculate the environmental water supply impact index based on the environmental characteristic parameters. The specific formula for calculating the environmental water supply impact index is:

[0095]

[0096] In the formula, WEC is the environmental water supply impact index, RH is the environmental average humidity, T is the environmental average temperature, V is the environmental average wind speed, RH 0 is the reference humidity, T 0 is the reference environmental temperature, and Rs is the average rainfall, specifically the annual average rainfall.

[0097] It should be noted that the larger the value of the environmental water supply impact index WEC, the more sufficient the environmental water supply and the less water resources need to be allocated.

[0098] Among them, the environmental humidity directly affects the availability of water. Higher humidity usually means a rich water content in the atmosphere, which is conducive to water retention and supply. Therefore, the environmental average humidity RH is directly proportional to the environmental water supply impact index WEC. The greater the difference from the reference humidity represented by the exponential function the greater the impact on the environmental water supply impact index WEC.

[0099] The annual average rainfall is the direct source of water supply. It is in the numerator, indicating that the higher the rainfall, the stronger the water supply capacity. Therefore, the average rainfall Rs is directly proportional to the environmental water supply impact index WEC.

[0100] Temperature has an important impact on water evaporation and plant transpiration. Higher temperatures usually increase the rate of water evaporation, thus reducing the available water. Therefore, the environmental average temperature T is inversely proportional to the environmental water supply impact index WEC, through It shows that the influence of temperature is non - linear, avoiding too drastic an impact on the WEC when the temperature changes significantly.

[0101] Wind speed affects the evaporation rate. The greater the wind speed, the stronger the evaporation and transpiration usually are, which reduces the water supply to the soil and vegetation. Therefore, the average environmental wind speed V is also inversely proportional to the environmental water supply impact index WEC, V 3 Emphasizes the influence of wind speed on the water supply capacity, especially at high wind speeds, where the impact may increase sharply. The cubic treatment makes the influence of wind speed on water supply more prominent under high - wind - speed conditions.

[0102] Among them, the average environmental humidity, average environmental temperature, and average environmental wind speed are specifically the means of the environmental parameters corresponding to all the locations of the selected plants in the sub - region, and the data at a height of 2 m above the ground are uniformly measured. Refer to the environmental temperature T 0 Generally 25 °C, refer to the humidity RH 0 Generally 25% to 30%.

[0103] According to the obtained environmental water supply impact index, combined with the soil water source preservation coefficient and the plant water demand impact index, the water supply demand index corresponding to the sub - region is comprehensively calculated. The formula based on which the water supply demand index is calculated is:

[0104]

[0105] In the formula, ZH is the water supply demand index, SCI represents the soil water source preservation coefficient, ω 1 、ω 2 and ω 3 are the weight coefficients of the environmental water supply impact index, plant water demand impact index, and soil water source preservation coefficient respectively. Among them, ω 2 >ω 3 ≥ω 2 ,ω 1 +ω 2 +ω 3 =1, and ω 1 、ω 2 and ω 3 are all greater than 0;

[0106] It should be noted that the larger the value of the water supply demand index ZH, the more water supply is required in the sub - region, and more water resources need to be allocated.

[0107] Since the proportional relationship between the environmental water supply impact index, plant water demand impact index, and soil water source preservation coefficient and water supply demand has been described above, it will not be elaborated here.

[0108] Among them, through the square of EIC 2in the form of, indicating the significant impact of the plant water demand impact index on the water supply demand index ZH, and is represented by the logarithmic function ln(1+ω 1 *WEC), indicating that as the environmental water supply impact index increases, the impact on the water supply demand index ZH gradually decreases.

[0109] This setting means that the plant water demand impact index has a higher priority in water supply demand, reflecting that the water demand of plants occupies an important position in the overall water supply demand. Plants are the main water consumers in the ecosystem, so an increase in their water demand will directly affect the water supply demand of the region. The water retention capacity of the soil determines the efficiency of water retention and utilization. If the soil can effectively retain water, even if the environmental water supply conditions are high, the availability of water resources will still be affected by the soil properties. Therefore, ω 2 >ω 3 ≥ω 2 and ω 1 、ω 2 and ω 3 are all greater than 0.

[0110] Compare the water supply demand index of each sub-region with the set water supply demand threshold range, and generate corresponding water use allocation strategies according to the comparison results. The specific logic for judgment is as follows:

[0111] When yz 1 ≤ZH≤yz 2 , it is judged that the water supply of this sub-region does not need to be adjusted;

[0112] When ZH<yz 1 , it is judged that the water demand of this sub-region is small, and the initial water supply should be reduced;

[0113] When ZH>yz 2 , it is judged that the water demand of this sub-region is large, and the initial water supply should be increased;

[0114] where yz 1 and yz 2 are respectively the lower and upper limits of the pre-set water supply demand threshold range, where yz 2 >yz 1 .

[0115] Among them, yz 1 and yz 2 can be set by combining the actual environmental situation with expert experience.

[0116] Please refer to Figure 2 , the present invention also provides an optimization system for the water use allocation strategy of the ecological space. The optimization system for the water use allocation strategy of the ecological space is used to execute the above-mentioned optimization method for the water use allocation strategy of the ecological space, and includes:

[0117] A prediction network training module, which is used to collect infrared spectrograms of several plants with known chlorophyll contents, preprocess the infrared spectrograms of the plants to obtain training sample infrared spectrograms, establish a neural network model based on the training sample infrared spectrograms, use the training sample infrared spectrograms as inputs, and use the corresponding chlorophyll contents as labels to train the neural network model to obtain a chlorophyll content prediction model;

[0118] A regional plant information prediction module, which is used to equally divide the ecological space of the water source to be allocated into several sub-regions, randomly select several plants in each sub-region, collect their infrared spectrograms, input the collected infrared spectrograms into the trained chlorophyll content prediction model to obtain the predicted chlorophyll content values of each plant, and characterize the chlorophyll content of the plants in the sub-region based on the average value of the predicted chlorophyll content values of each plant;

[0119] A soil characteristic analysis module, which is used to calculate a plant water demand influence index according to the chlorophyll content of the plants in the sub-region and in combination with the plant growth characteristic parameters of the sub-region, and at the same time obtain the soil characteristic parameters in each sub-region, and calculate and generate a soil water source preservation coefficient based on the collected soil characteristic parameters, wherein the soil characteristic parameters include the average soil permeability, the average soil organic matter content, and the average soil particle size, and the plant growth characteristic parameters include the average plant height, the average plant leaf area, and the average main stem diameter of the plant;

[0120] A water source precise allocation module, which is used to obtain the environmental characteristic parameters of each sub-region, calculate an environmental water supply influence index based on the environmental characteristic parameters, and comprehensively calculate and generate a corresponding water supply demand index for the sub-region according to the obtained environmental water supply influence index, in combination with the soil water source preservation coefficient and the plant water demand influence index, compare the water supply demand index of each sub-region with the set water supply demand threshold range, and generate a corresponding water use allocation strategy according to the comparison result, wherein the environmental characteristic parameters include the average environmental temperature, the average humidity, the average rainfall, and the average wind speed.

[0121] All the above formulas are calculated by taking the numerical values after dimensionless, and the formulas are obtained by software simulation of collecting a large amount of data to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0122] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0123] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] As described above, the specific implementation manners of the present application are only described, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application.

Claims

1. A method for optimizing water allocation strategy for ecological space, characterized in that: The specific steps include: Collect several infrared spectra of plants with known chlorophyll content, pre-process the infrared spectra of the plants to obtain training sample infrared spectra, establish a neural network model based on the training sample infrared spectra, take the training sample infrared spectra as input and the corresponding chlorophyll content as label, train the neural network model, and obtain a chlorophyll content prediction model; The ecological space of the water source to be allocated is divided into several sub-areas of equal area, several plants are randomly selected in each sub-area, their infrared spectra are collected, and the collected infrared spectra are input into the trained chlorophyll content prediction model to obtain the chlorophyll content prediction value of each plant, and the chlorophyll content of the plants in the sub-area is characterized based on the average value of the chlorophyll content prediction value of each plant; According to the chlorophyll content of the plants in the sub-region, combined with the plant growth characteristic parameters of the sub-region, the plant water demand impact index is calculated, and the soil characteristic parameters in each sub-region are obtained at the same time. Based on the collected soil characteristic parameters, the soil water conservation coefficient is calculated, wherein the soil characteristic parameters include the average soil permeability, the average soil organic matter content and the average soil particle size, and the plant growth characteristic parameters include the average plant height, the average plant leaf area and the average diameter of the plant main stem; The environmental characteristic parameters of each sub-region are obtained, and the environmental water supply impact index is calculated based on the environmental characteristic parameters. According to the obtained environmental water supply impact index, combined with the soil water conservation coefficient and the plant water demand impact index, the water supply demand index corresponding to the sub-region is comprehensively calculated and generated. The water supply demand index of each sub-region is compared with the set water supply demand threshold range. According to the comparison results, the corresponding water allocation strategy is generated, where the environmental characteristic parameters include the average ambient temperature, average humidity, average rainfall and average wind speed.

2. The method for optimizing water allocation strategy for ecological space according to claim 1, characterized in that: Preprocessing the infrared spectrum of the plant, wherein the preprocessing includes image denoising preprocessing and image normalization preprocessing; Based on the long short-term memory network model LSTM model, a chlorophyll content prediction model is established, and the activation function and optimization algorithm are selected. The Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm of the LSTM model; the formula of the Tanh function is: In the formula, f(r) represents the Tanh function, and the independent variable r represents the weighted sum of the neuron's input, that is, the result of the weighted summation of the input received by the neuron from the previous layer; At the same time, the hyperparameters of the LSTM model are set, and the hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons; The number of network layers is set to 3 layers, the number of iterations is set to 200, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, the batch size is set to 256, and the number of hidden layer neurons is 32; The input of the trained chlorophyll content prediction model is the infrared spectrum of the plant, and the output is the predicted value of the chlorophyll content of the plant.

3. The method for optimizing water allocation strategy for ecological space according to claim 2, characterized in that: Based on the chlorophyll content of plants in a sub-region and combined with the plant growth characteristic parameters of that sub-region, a plant water demand impact index is calculated. The specific formula for calculating the plant water demand impact index is as follows: In the formula, EIC is the plant water demand impact index, C is the chlorophyll content of plants in the sub-region, HL is the average height of plants in the sub-region, Mt is the average leaf area of plants in the sub-region, and DH is the average diameter of the main stems of plants in the sub-region.

4. The method for optimizing water allocation strategy for ecological space according to claim 1, characterized in that: Based on the collected soil characteristic parameters, a soil water source preservation coefficient is calculated. The specific formula for calculating the soil water source preservation coefficient is as follows: In the formula, SCI represents the soil water conservation coefficient, OM represents the average content of soil organic matter in the sub-region, K represents the average permeability of the soil in the sub-region, and D mean is the average particle size of soil in the sub-region; The specific method for obtaining the soil permeability K is as follows: Select a soil sample to be tested and place it in a cylindrical permeameter. The sample needs to be fully saturated before being placed. Set a constant water level difference, that is, maintain a constant height difference between the water source in the permeameter and the water level above the soil sample. Record the amount of water flowing through the soil sample within the time period t. Calculate the soil permeability based on the recorded data. The specific formula for calculating the soil permeability K is as follows: In the formula, Q is the amount of water flowing through the soil sample within the time period t, L is the thickness of the soil sample in the permeameter, PF is the cross-sectional area of the soil sample in the permeameter, and h is the set water level difference.

5. The method for optimizing water allocation strategy for ecological space according to claim 3, characterized in that: Obtain the environmental characteristic parameters of each sub-region. Based on the environmental characteristic parameters, an environmental water supply impact index is calculated. The specific formula for calculating the environmental water supply impact index is as follows: In the formula, WEC is the environmental water supply impact index, RH is the average environmental humidity, T is the average environmental temperature, V is the average environmental wind speed, RH0 is the reference humidity, T0 is the reference environmental temperature, and Rs is the average rainfall, specifically the annual average rainfall.

6. The method for optimizing water allocation strategy for ecological space according to claim 5, characterized in that: Based on the obtained environmental water supply impact index, combined with the soil water source preservation coefficient and the plant water demand impact index, a water supply demand index corresponding to the sub-region is comprehensively calculated. The specific formula for calculating the water supply demand index is as follows: In the formula, ZH is the water supply demand index, SCI represents the soil water source preservation coefficient, and ω1, ω2, and ω3 are the weight coefficients of the environmental water supply impact index, the plant water demand impact index, and the soil water source preservation coefficient respectively. Among them, ω2 > ω3 ≥ ω2, ω1 + ω2 + ω3 = 1, and ω1, ω2, and ω3 are all greater than 0; Compare the water supply demand index of each sub-region with the set water supply demand threshold range. Based on the comparison result, a corresponding water use allocation strategy is generated. The specific logic for judgment is as follows: When yz1 ≤ ZH ≤ yz2, it is judged that the water supply of this sub-region does not need to be adjusted; When ZH < yz1, it is judged that the water demand of this sub-region is small, and the initial water supply should be reduced; When ZH > yz2, it is judged that the water demand of this sub-region is large, and the initial water supply should be increased; Among them, yz1 and yz2 are respectively the lower limit and the upper limit of the pre-set water supply demand threshold range, where yz2 > yz1.

7. An ecological space water allocation strategy optimization system, characterized by: The described ecological space water use allocation strategy optimization system is used to execute an ecological space water use allocation strategy optimization method according to any one of claims 1-6, including: The prediction network training module is used to collect infrared spectra of several plants with known chlorophyll content, pre-process the infrared spectra of the plants to obtain infrared spectra of training samples, establish a neural network model based on the infrared spectra of the training samples, take the infrared spectra of the training samples as input, and take the corresponding chlorophyll content as a label to train the neural network model to obtain a chlorophyll content prediction model; The regional plant information prediction module is used to divide the ecological space of the water source to be allocated into several sub-regions, randomly select several plants in each sub-region, collect their infrared spectra, input the collected infrared spectra into the trained chlorophyll content prediction model, obtain the chlorophyll content prediction value of each plant, and characterize the chlorophyll content of the plants in the sub-region based on the average value of the chlorophyll content prediction value of each plant; A soil characteristic analysis module is used to calculate the plant water demand impact index based on the chlorophyll content of the plants in the sub-region and the plant growth characteristic parameters of the sub-region, and to obtain the soil characteristic parameters in each sub-region at the same time, and to calculate and generate the soil water conservation coefficient based on the collected soil characteristic parameters, wherein the soil characteristic parameters include the average soil permeability, the average soil organic matter content and the average soil particle size, and the plant growth characteristic parameters include the average plant height, the average plant leaf area and the average diameter of the main stem of the plant; The water source precise allocation module is used to obtain the environmental characteristic parameters of each sub-area, calculate the environmental water supply impact index based on the environmental characteristic parameters, and use the obtained environmental water supply impact index combined with the soil water conservation coefficient and the plant water demand impact index to comprehensively calculate and generate the water supply demand index corresponding to the sub-area. The water supply demand index of each sub-area is compared with the set water supply demand threshold range, and the corresponding water use allocation strategy is generated according to the comparison results, where the environmental characteristic parameters include the average ambient temperature, average humidity, average rainfall and average wind speed.

Images