A method for evaluating the effect of saline-alkali land improvement based on digital twins

Abstract

The present invention discloses a method for evaluating the effect of soil improvement in saline-alkali land based on digital twins, which relates to the technical field of saline-alkali land soil improvement effect evaluation. The method comprises the following steps: collecting saline-alkali land soil improvement effect monitoring data including multi-dimensional soil data, crop growth data and meteorological data, and preprocessing the collected data, calculating the salinization index, fertility index, porosity and permeability of the soil, and analyzing the saline-alkali land soil improvement effect, thereby obtaining the saline-alkali land soil improvement effect grade, and utilizing the preprocessed crop growth data and meteorological data in combination with a multiple linear regression algorithm to obtain the improvement effect feedback coefficient. The multi-dimensional data acquisition technology, multiple linear regression algorithm, digital twin model construction technology and model optimization technology in the present invention are closely combined with modern information technology, which solves the problems of incomplete analysis data, inaccurate evaluation results and low intelligence level that may exist in traditional evaluation methods.

Description

Technical Field

[0001] The present invention relates to the technical field of saline-alkali land soil improvement effect evaluation, and in particular to a saline-alkali land soil improvement effect evaluation method based on digital twins. Background Art

[0002] Globally, saline-alkali land is widely distributed, posing a severe challenge to agricultural production and the ecological environment. The effectiveness evaluation of saline-alkali land soil improvement work is of great significance. However, traditional evaluation methods have many limitations and mainly rely on manual field sampling and simple laboratory analysis to evaluate the effect of saline-alkali land soil improvement. This method is not only inefficient, but also unable to achieve real-time and dynamic monitoring of large areas of saline-alkali land. With the rise of digital twin technology, its application in various fields continues to expand. In terms of the evaluation of saline-alkali land soil improvement effect, although there have been some attempts based on digital twins, there are still many problems. Therefore, there is an urgent need for a more complete and accurate digital twin-based saline-alkali land soil improvement effect evaluation method to improve the scientificity and effectiveness of saline-alkali land improvement work.

[0003] Although existing technologies have made great progress in the improvement of saline-alkali land soil, there are still some problems that need to be optimized. Traditional saline-alkali land soil improvement makes it difficult to conduct comprehensive analysis of multi-dimensional soil data, crop growth data and meteorological data, and then evaluate the effect of saline-alkali land soil improvement, resulting in the problem of low accuracy of existing technologies in evaluating the effect of saline-alkali land soil improvement. Summary of the Invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for evaluating the effect of saline-alkali land improvement based on digital twins, comprising the following steps:

[0005] Step 1: Collect saline-alkali land soil improvement effect monitoring data, including multi-dimensional soil data, crop growth data, and meteorological data, and pre-process the collected data to provide a data foundation for the implementation of subsequent steps;

[0006] Step 2: Using the pre-treated saline-alkali land soil improvement effect monitoring data, calculate the soil salinization index, fertility index, porosity and permeability, and analyze the saline-alkali land soil improvement effect to obtain the saline-alkali land soil improvement effect grade;

[0007] Step 3: Using the pre-processed crop growth data and meteorological data, combined with a multiple linear regression algorithm, to obtain the feedback coefficient of the improvement effect;

[0008] Step 4: Based on the obtained improvement effect feedback coefficient, analyze the improvement effect feedback level, and use the improvement effect feedback level to update the saline-alkali land soil improvement effect level, thereby reducing the error in evaluating the improvement effect results based solely on saline-alkali land soil data;

[0009] Step 5: Combining the iterative optimization algorithm with the updated saline-alkali land improvement effect level, a digital twin evaluation model was constructed. The digital twin evaluation model was optimized through a deep neural network algorithm, thereby improving the accuracy of the digital twin evaluation model.

[0010] Step 6: Based on the output results of the optimized digital twin evaluation model, a corresponding saline-alkali land soil improvement effect signal is issued, which solves the problem that traditional saline-alkali land soil improvement effect evaluation methods are difficult to conduct comprehensive analysis of multi-dimensional soil data, crop growth data and meteorological data, resulting in low evaluation accuracy of saline-alkali land soil improvement effects in existing technologies.

[0011] A further improvement of the technical solution of the present invention is that in step 1, the process of collecting monitoring data on the improvement effect of saline-alkali land includes:

[0012] The target research area of ​​saline-alkali soil is divided into several small saline-alkali soil areas, which have the same area and are numbered respectively;

[0013] Deploy different data collection equipment to collect multi-dimensional soil data, crop growth data, and meteorological data for each small saline-alkali soil area, including pH meters, conductivity meters, ion chromatographs, soil nitrogen, phosphorus, and potassium sensors, soil corers, mercury intrusion porosimeter, laser particle size analyzer, crop height meter, crop angle stem diameter meter, leaf area meter, crop yield monitor, tipping bucket rain gauge, temperature sensor, and sunshine duration sensor;

[0014] The multi-dimensional soil data includes soil pH, total salt content, salt ion concentration, nitrogen content, phosphorus content, potassium content, total volume, particle size, and volume of interparticle spaces; the crop growth data includes plant height, stem diameter, leaf area, and yield per unit area; the meteorological data includes precipitation, temperature, and sunshine duration;

[0015] Specifically, the pH value, total salt content, salt ion concentration, nitrogen content, phosphorus content, potassium content, total volume, particle size and volume of inter-particle gaps of the soil are collected using a pH meter, a conductivity meter, an ion chromatograph, a soil nitrogen, phosphorus and potassium sensor, a soil core sampler, a mercury intrusion meter and a laser particle size analyzer; the plant height, stem diameter, leaf area and yield per unit area of ​​the crops are collected using a crop height meter, a crop angle stem diameter meter, a leaf area meter and a crop yield monitor. Since the areas of the various small saline-alkali soil areas are the same, the crop yields of the various small saline-alkali soil areas are regarded as the yield per unit area of ​​the crops; the precipitation, temperature and sunshine duration are collected using a tipping bucket rain gauge, a temperature sensor and a sunshine duration sensor.

[0016] A further improvement of the technical solution of the present invention is that in step 1, the process of preprocessing the collected saline-alkali land soil improvement effect monitoring data includes:

[0017] Perform data cleaning and data normalization on the collected improvement effect monitoring data, assign timestamps to the multi-dimensional soil data, crop growth data, and meteorological data, and adjust the timestamps to synchronize the collection time of the multi-dimensional soil data, crop growth data, and meteorological data;

[0018] The multi-dimensional soil data, crop growth data and meteorological data were matched with the numbers of their corresponding small saline-alkali soil areas, and the multi-dimensional soil data, crop growth data and meteorological data were integrated to generate an improvement effect monitoring dataset. The improvement effect monitoring dataset was divided into a training set and a test set, where the ratio of the training set to the test set was 7:3.

[0019] A further improvement of the technical solution of the present invention is that in step 2, the calculation process of the soil salinization index, fertility index, porosity and permeability includes:

[0020] According to the principle of soil salinization, weights are assigned to the soil pH value, total salt content and salt ion concentration respectively, and the soil salinization index is calculated using the weighted average method. The calculation formula is:

[0021]

[0022] Among them, SI is the soil salinization index, , and are soil pH value weight, soil total salt content weight and soil salt ion concentration weight, respectively. , and are soil pH, total salt content, and salt ion concentration;

[0023] The soil nitrogen content, phosphorus content, and potassium content are added to 1 and then the logarithm is taken to avoid meaningless logarithms. The soil fertility index is calculated using the logarithmic method. The calculation process is as follows:

[0024]

[0025] in, is the soil fertility index, is the soil nitrogen content, is the soil phosphorus content, is the soil potassium content;

[0026] Calculate the ratio of the volume of the voids between soil particles to the total volume of the soil to obtain the soil porosity; calculate the average soil particle size and calculate the soil permeability based on the Kozeny-Carman equation. The calculation formula is as follows:

[0027]

[0028] in, is the soil permeability, is the Kozeny constant, is the soil porosity, is the average soil particle size.

[0029] A further improvement of the technical solution of the present invention is that in step 2, the process of obtaining the saline-alkali land soil improvement effect level includes:

[0030] According to the physical properties of soil and the principle of chemical balance, weights are assigned to soil salinization index, soil fertility index, soil porosity and soil permeability respectively;

[0031] The weighted average method is used to calculate the product of the soil salinization index and its weight, the product of the soil fertility index and its weight, the product of the soil porosity and its weight, and the product of the soil permeability and its weight, respectively. The calculated results are then summed to obtain the saline-alkali land soil improvement coefficient.

[0032] The saline-alkali soil improvement coefficient is used to analyze the saline-alkali soil improvement effect and evaluate the saline-alkali soil improvement effect level of each small saline-alkali soil area. The saline-alkali soil improvement effect level includes low improvement effect, medium improvement effect and high improvement effect;

[0033] When the saline-alkali soil improvement coefficient is less than 0.3, the corresponding small saline-alkali soil area has a low improvement effect; when the saline-alkali soil improvement coefficient is between 0.3 and 0.6, the corresponding small saline-alkali soil area has a medium improvement effect; when the saline-alkali soil improvement coefficient is greater than 0.6, the corresponding small saline-alkali soil area has a high improvement effect.

[0034] A further improvement of the technical solution of the present invention is that in step 3, the process of obtaining the improvement effect feedback coefficient includes:

[0035] Extract the crop growth data and meteorological data from the improvement effect monitoring dataset, use the crop growth data and meteorological data in the training set, combine with the multiple linear regression algorithm, take the crop growth data and meteorological data as input, and take the improvement effect feedback coefficient as output, learn the linear relationship between the crop growth data, meteorological data and the improvement effect feedback coefficient, and train the improvement effect feedback model;

[0036] The crop growth data and meteorological data in the test set are input into the improvement effect feedback model to evaluate its performance. The improvement effect feedback model is optimized by adjusting the intercept term and regression coefficient of the improvement effect feedback model to obtain the final improvement effect feedback model. The corresponding improvement effect feedback coefficient is output based on the current crop growth data and meteorological data of each small saline-alkali soil area.

[0037] The expression of the improved effect feedback model is:

[0038]

[0039] in, To improve the effect feedback coefficient, , , , , , and are the regression coefficients of crop height, crop stem diameter, crop leaf area, crop yield per unit area, precipitation, temperature and sunshine duration, , , , , , and They are crop height, crop stem diameter, crop leaf area, crop yield per unit area, precipitation, temperature and sunshine duration. is the intercept term, is the error term.

[0040] A further improvement of the technical solution of the present invention is that in step 4, the process of analyzing the improvement effect feedback level and using the improvement effect feedback level to update the saline-alkali land soil improvement effect level includes:

[0041] Based on the improvement effect feedback coefficient output by the improvement effect feedback model, the improvement effect level of saline-alkali land is analyzed. When the improvement effect feedback coefficient is lower than 0.4, the corresponding small saline-alkali land area has a low improvement effect; when the improvement effect feedback coefficient is between 0.4 and 0.75, the corresponding small saline-alkali land area has a medium improvement effect; when the improvement effect feedback coefficient is higher than 0.75, the corresponding small saline-alkali land area has a high improvement effect, and the improvement effect feedback level is obtained;

[0042] Based on the obtained improvement effect feedback level, when the saline-alkali land soil improvement effect level and the improvement effect feedback level are the same, it will not be updated; when the saline-alkali land soil improvement effect level and the improvement effect feedback level are different, the saline-alkali land soil improvement effect level will be replaced by the improvement effect feedback level to update the saline-alkali land soil improvement effect level;

[0043] The updated saline-alkali land soil improvement effect grade is numbered and the updated saline-alkali land soil improvement effect grade number is integrated into the improvement effect monitoring dataset.

[0044] A further improvement of the technical solution of the present invention is that in step 5, the construction process of the digital twin evaluation model includes:

[0045] Extract the multi-dimensional soil data and updated saline-alkali land soil improvement effect grade number from the improvement effect monitoring dataset;

[0046] Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with an iterative optimization algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The association between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the digital twin evaluation model;

[0047] The multi-dimensional soil data in the test set is input into the digital twin evaluation model, and the error between the output results of the digital twin evaluation model and the actual updated saline-alkali land soil improvement effect grade number is compared to evaluate the performance of the digital twin evaluation model, adjust the parameters of the digital twin evaluation model, improve the performance of the digital twin evaluation model, and construct a digital twin evaluation model.

[0048] A further improvement of the technical solution of the present invention is that in step 5, the optimization process of the digital twin evaluation model includes:

[0049] Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with the deep neural network algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The nonlinear relationship between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the deep neural network model;

[0050] The multi-dimensional soil data in the test set is input into the deep neural network model, and the error between the output of the deep neural network model and the actual updated saline-alkali land soil improvement effect grade number is compared. The deep neural network model parameters are characterized, and the deep neural network model is optimized to obtain the final deep neural network model. The current multi-dimensional soil data of each small saline-alkali land soil area is used to output the corresponding updated saline-alkali land soil improvement effect grade number;

[0051] The digital twin evaluation model is trained using the updated saline-alkali land soil improvement effect grade number output by the deep neural network model, and the parameters of the digital twin evaluation model are adjusted again. The digital twin evaluation model is optimized to obtain the final digital twin evaluation model. Combined with the current multi-dimensional soil data of each small saline-alkali land soil area, the corresponding updated saline-alkali land soil improvement effect grade number is output.

[0052] A further improvement of the technical solution of the present invention is that in step 6, the process of issuing a corresponding saline-alkali land soil improvement effect signal based on the output result of the optimized digital twin evaluation model includes:

[0053] Based on the updated saline-alkali land soil improvement effect grade number output by the optimized digital twin evaluation model, match the saline-alkali land soil improvement effect grade corresponding to the updated saline-alkali land soil improvement effect grade number;

[0054] An electronic display screen is set up in the target research area of ​​saline-alkali soil, and the updated saline-alkali soil improvement effect level of each small saline-alkali soil area is displayed on the screen in text form, including the electronic display screen displaying low words, the electronic display screen displaying medium words and the electronic display screen displaying high words;

[0055] When the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is low improvement effect, the electronic display screen displays the word "low"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is medium improvement effect, the electronic display screen displays the word "medium"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is high improvement effect, the electronic display screen displays the word "high".

[0056] The beneficial effects of the present invention are as follows: a digital twin-based saline-alkali land soil improvement effect evaluation method in the present invention, compared with the traditional digital twin-based saline-alkali land soil improvement effect evaluation method, the multidimensional data acquisition technology, multivariate linear regression algorithm, digital twin model construction technology and model optimization technology in the method of the present invention are closely combined with modern information technology, accurately capturing multidimensional soil data, crop growth data and meteorological data, and then obtaining the soil salinization index, fertility index, porosity, permeability, saline-alkali land soil improvement coefficient and improvement effect feedback coefficient, and obtaining the saline-alkali land soil improvement effect grade through the saline-alkali land soil improvement coefficient, and obtaining the improvement effect feedback coefficient through the improvement effect The improvement effect feedback coefficient is used to obtain the improvement effect feedback level, and the improvement effect feedback level is used to update the saline-alkali land soil improvement effect level, thereby achieving real-time and comprehensive monitoring of the saline-alkali land soil improvement effect, solving the problems of incomplete analysis data, inaccurate evaluation results and low intelligence in traditional evaluation methods, and ensuring that the method in the present invention can be refined within a more precise range for a dynamic monitoring standard for a saline-alkali land soil improvement effect evaluation method based on digital twins, so that the monitored data becomes a more accurate indicator under the same conditions. The development and application of this method have significantly enhanced the intelligence level in the saline-alkali land soil improvement effect evaluation process based on digital twins. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0058] Figure 1 This is a flow chart of a method for evaluating the effect of saline-alkali land improvement based on digital twins of the present invention. DETAILED DESCRIPTION

[0059] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0060] like Figure 1 As shown, the present invention provides a method for evaluating the effect of saline-alkali land soil improvement based on digital twins, which consists of the following steps:

[0061] Step 1: Collect saline-alkali land soil improvement effect monitoring data, including multi-dimensional soil data, crop growth data, and meteorological data, and pre-process the collected data to provide a data foundation for the implementation of subsequent steps;

[0062] Step 2: Using the pre-treated saline-alkali land soil improvement effect monitoring data, calculate the soil salinization index, fertility index, porosity and permeability, and analyze the saline-alkali land soil improvement effect to obtain the saline-alkali land soil improvement effect grade;

[0063] Step 3: Using the pre-processed crop growth data and meteorological data, combined with a multiple linear regression algorithm, to obtain the feedback coefficient of the improvement effect;

[0064] Step 4: Based on the obtained improvement effect feedback coefficient, analyze the improvement effect feedback level, and use the improvement effect feedback level to update the saline-alkali land soil improvement effect level, thereby reducing the error in evaluating the improvement effect results based solely on saline-alkali land soil data;

[0065] Step 5: Combining the iterative optimization algorithm with the updated saline-alkali land improvement effect level, a digital twin evaluation model was constructed. The digital twin evaluation model was optimized through a deep neural network algorithm, thereby improving the accuracy of the digital twin evaluation model.

[0066] Step 6: Based on the output results of the optimized digital twin evaluation model, a corresponding saline-alkali land soil improvement effect signal is issued, which solves the problem that traditional saline-alkali land soil improvement effect evaluation methods are difficult to conduct comprehensive analysis of multi-dimensional soil data, crop growth data and meteorological data, resulting in low evaluation accuracy of saline-alkali land soil improvement effects in existing technologies.

[0067] Preferably, in step 1, the process of collecting monitoring data on the improvement effect of saline-alkali land includes:

[0068] The target research area of ​​saline-alkali soil is divided into several small saline-alkali soil areas, which have the same area and are numbered respectively;

[0069] Deploy different data collection equipment to collect multi-dimensional soil data, crop growth data, and meteorological data from each small saline-alkali soil area. The data collection equipment includes pH meters, conductivity meters, ion chromatographs, soil nitrogen, phosphorus, and potassium sensors, soil corers, mercury intrusion porosimeter, laser particle size analyzer, crop height meter, crop angle stem diameter meter, leaf area meter, crop yield monitor, tipping bucket rain gauge, temperature sensor, and sunshine duration sensor.

[0070] Multi-dimensional soil data includes soil pH, total salt content, salt ion concentration, nitrogen content, phosphorus content, potassium content, total volume, particle size, and volume of interparticle spaces; crop growth data includes plant height, stem diameter, leaf area, and yield per unit area; meteorological data includes precipitation, temperature, and sunshine duration;

[0071] Specifically, the pH value, total salt content, salt ion concentration, nitrogen content, phosphorus content, potassium content, total volume, particle size and volume of inter-particle gaps of the soil are collected using a pH meter, a conductivity meter, an ion chromatograph, a soil nitrogen, phosphorus and potassium sensor, a soil core sampler, a mercury intrusion meter and a laser particle size analyzer; the plant height, stem diameter, leaf area and yield per unit area of ​​the crops are collected using a crop height meter, a crop angle stem diameter meter, a leaf area meter and a crop yield monitor. Since the areas of the various small saline-alkali soil areas are the same, the crop yields of the various small saline-alkali soil areas are regarded as the yield per unit area of ​​the crops; the precipitation, temperature and sunshine duration are collected using a tipping bucket rain gauge, a temperature sensor and a sunshine duration sensor.

[0072] Preferably, in step 1, the process of preprocessing the collected saline-alkali land soil improvement effect monitoring data includes:

[0073] Perform data cleaning and data normalization on the collected improvement effect monitoring data, assign timestamps to the multi-dimensional soil data, crop growth data, and meteorological data, and adjust the timestamps to synchronize the collection time of the multi-dimensional soil data, crop growth data, and meteorological data;

[0074] The multi-dimensional soil data, crop growth data and meteorological data were matched with the numbers of their corresponding small saline-alkali soil areas, and the multi-dimensional soil data, crop growth data and meteorological data were integrated to generate an improvement effect monitoring dataset. The improvement effect monitoring dataset was divided into a training set and a test set, where the ratio of the training set to the test set was 7:3.

[0075] Preferably, in step 2, the calculation process of the soil salinization index, fertility index, porosity and permeability includes:

[0076] According to the principle of soil salinization, weights are assigned to the soil pH value, total salt content and salt ion concentration respectively, and the soil salinization index is calculated using the weighted average method. The calculation formula is:

[0077]

[0078] Among them, SI is the soil salinization index, , and are soil pH value weight, soil total salt content weight and soil salt ion concentration weight, respectively. , and are soil pH, total salt content, and salt ion concentration;

[0079] The soil nitrogen content, phosphorus content, and potassium content are added to 1 and then the logarithm is taken to avoid meaningless logarithms. The soil fertility index is calculated using the logarithmic method. The calculation process is as follows:

[0080]

[0081] in, is the soil fertility index, is the soil nitrogen content, is the soil phosphorus content, is the soil potassium content;

[0082] Calculate the ratio of the volume of the voids between soil particles to the total volume of the soil to obtain the soil porosity; calculate the average soil particle size and calculate the soil permeability based on the Kozeny-Carman equation. The calculation formula is as follows:

[0083]

[0084] in, is the soil permeability, is the Kozeny constant, is the soil porosity, is the average soil particle size.

[0085] Preferably, in step 2, the process of obtaining the saline-alkali land soil improvement effect level includes:

[0086] According to the physical properties of soil and the principle of chemical balance, weights are assigned to soil salinization index, soil fertility index, soil porosity and soil permeability respectively;

[0087] The weighted average method is used to calculate the product of the soil salinization index and its weight, the product of the soil fertility index and its weight, the product of the soil porosity and its weight, and the product of the soil permeability and its weight, respectively. The calculated results are then summed to obtain the saline-alkali land soil improvement coefficient.

[0088] The saline-alkali soil improvement coefficient is used to analyze the saline-alkali soil improvement effect and evaluate the saline-alkali soil improvement effect level of each small saline-alkali soil area. The saline-alkali soil improvement effect level includes low improvement effect, medium improvement effect and high improvement effect.

[0089] When the saline-alkali soil improvement coefficient is less than 0.3, the corresponding small saline-alkali soil area has a low improvement effect; when the saline-alkali soil improvement coefficient is between 0.3 and 0.6, the corresponding small saline-alkali soil area has a medium improvement effect; when the saline-alkali soil improvement coefficient is greater than 0.6, the corresponding small saline-alkali soil area has a high improvement effect.

[0090] Preferably, in step 3, the process of obtaining the improvement effect feedback coefficient includes:

[0091] Extract the crop growth data and meteorological data from the improvement effect monitoring dataset, use the crop growth data and meteorological data in the training set, combine with the multiple linear regression algorithm, take the crop growth data and meteorological data as input, and take the improvement effect feedback coefficient as output, learn the linear relationship between the crop growth data, meteorological data and the improvement effect feedback coefficient, and train the improvement effect feedback model;

[0092] The crop growth data and meteorological data in the test set are input into the improvement effect feedback model to evaluate its performance. The improvement effect feedback model is optimized by adjusting the intercept term and regression coefficient of the improvement effect feedback model to obtain the final improvement effect feedback model. The corresponding improvement effect feedback coefficient is output based on the current crop growth data and meteorological data of each small saline-alkali soil area.

[0093] The expression of the improved effect feedback model is:

[0094]

[0095] in, To improve the effect feedback coefficient, , , , , , and are the regression coefficients of crop height, crop stem diameter, crop leaf area, crop yield per unit area, precipitation, temperature and sunshine duration, , , , , , and They are crop height, crop stem diameter, crop leaf area, crop yield per unit area, precipitation, temperature and sunshine duration. is the intercept term, is the error term.

[0096] Preferably, in step 4, the process of analyzing the improvement effect feedback level and updating the saline-alkali land soil improvement effect level by using the improvement effect feedback level includes:

[0097] Based on the improvement effect feedback coefficient output by the improvement effect feedback model, the improvement effect level of saline-alkali land is analyzed. When the improvement effect feedback coefficient is lower than 0.4, the corresponding small saline-alkali land area has a low improvement effect; when the improvement effect feedback coefficient is between 0.4 and 0.75, the corresponding small saline-alkali land area has a medium improvement effect; when the improvement effect feedback coefficient is higher than 0.75, the corresponding small saline-alkali land area has a high improvement effect, and the improvement effect feedback level is obtained;

[0098] Based on the obtained improvement effect feedback level, when the saline-alkali land soil improvement effect level and the improvement effect feedback level are the same, it will not be updated; when the saline-alkali land soil improvement effect level and the improvement effect feedback level are different, the saline-alkali land soil improvement effect level will be replaced by the improvement effect feedback level to update the saline-alkali land soil improvement effect level;

[0099] The updated saline-alkali land soil improvement effect grade is numbered and the updated saline-alkali land soil improvement effect grade number is integrated into the improvement effect monitoring dataset.

[0100] Preferably, in step 5, the process of building the digital twin evaluation model includes:

[0101] Extract the multi-dimensional soil data and updated saline-alkali land soil improvement effect grade number from the improvement effect monitoring dataset;

[0102] Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with an iterative optimization algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The association between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the digital twin evaluation model;

[0103] The multi-dimensional soil data in the test set is input into the digital twin evaluation model, and the error between the output results of the digital twin evaluation model and the actual updated saline-alkali land soil improvement effect grade number is compared to evaluate the performance of the digital twin evaluation model, adjust the parameters of the digital twin evaluation model, improve the performance of the digital twin evaluation model, and construct a digital twin evaluation model.

[0104] Preferably, in step 5, the optimization process of the digital twin evaluation model includes:

[0105] Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with the deep neural network algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The nonlinear relationship between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the deep neural network model;

[0106] The multi-dimensional soil data in the test set is input into the deep neural network model, and the error between the output of the deep neural network model and the actual updated saline-alkali land soil improvement effect grade number is compared. The deep neural network model parameters are characterized, and the deep neural network model is optimized to obtain the final deep neural network model. The current multi-dimensional soil data of each small saline-alkali land soil area is used to output the corresponding updated saline-alkali land soil improvement effect grade number;

[0107] The digital twin evaluation model is trained using the updated saline-alkali land soil improvement effect grade number output by the deep neural network model, and the parameters of the digital twin evaluation model are adjusted again. The digital twin evaluation model is optimized to obtain the final digital twin evaluation model. Combined with the current multi-dimensional soil data of each small saline-alkali land soil area, the corresponding updated saline-alkali land soil improvement effect grade number is output.

[0108] Preferably, in step 6, the process of issuing a corresponding saline-alkali land soil improvement effect signal based on the output result of the optimized digital twin evaluation model includes:

[0109] Based on the updated saline-alkali land soil improvement effect grade number output by the optimized digital twin evaluation model, match the saline-alkali land soil improvement effect grade corresponding to the updated saline-alkali land soil improvement effect grade number;

[0110] An electronic display screen is set up in the target research area of ​​saline-alkali soil, and the updated saline-alkali soil improvement effect level of each small saline-alkali soil area is displayed on the screen in text form, including the electronic display screen displaying low words, the electronic display screen displaying medium words and the electronic display screen displaying high words;

[0111] When the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is low improvement effect, the electronic display screen displays the word "low"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is medium improvement effect, the electronic display screen displays the word "medium"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is high improvement effect, the electronic display screen displays the word "high".

[0112] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for evaluating the effect of saline-alkali soil improvement based on digital twins, characterized by: The following steps are involved: Step 1: Collect saline-alkali land soil improvement effect monitoring data including multi-dimensional soil data, crop growth data, and meteorological data, and pre-process the collected data; Step 2: Use the pre-processed multi-dimensional soil data to calculate the soil salinization index, fertility index, porosity, and permeability, and analyze the saline-alkali land soil improvement effect to obtain the saline-alkali land soil improvement effect grade. The process of obtaining the saline-alkali land soil improvement effect grade includes: According to the physical properties of soil and the principle of chemical balance, weights are assigned to soil salinization index, soil fertility index, soil porosity and soil permeability respectively; The weighted average method is used to calculate the product of the soil salinization index and its weight, the product of the soil fertility index and its weight, the product of the soil porosity and its weight, and the product of the soil permeability and its weight, respectively. The calculated results are then summed to obtain the saline-alkali land soil improvement coefficient. The saline-alkali soil improvement coefficient is used to analyze the saline-alkali soil improvement effect and evaluate the saline-alkali soil improvement effect level of each small saline-alkali soil area. The saline-alkali soil improvement effect level includes low improvement effect, medium improvement effect and high improvement effect; When the saline-alkali soil improvement coefficient is less than 0.3, the corresponding small saline-alkali soil area has a low improvement effect; when the saline-alkali soil improvement coefficient is between 0.3 and 0.6, the corresponding small saline-alkali soil area has a medium improvement effect; when the saline-alkali soil improvement coefficient is greater than 0.6, the corresponding small saline-alkali soil area has a high improvement effect; Step 3: Using the pre-processed crop growth data and meteorological data, combined with a multiple linear regression algorithm, to obtain the feedback coefficient of the improvement effect; Step 4: Based on the obtained improvement effect feedback coefficient, analyze the improvement effect feedback level, and use the improvement effect feedback level to update the saline-alkali land soil improvement effect level. Based on the improvement effect feedback coefficient output by the improvement effect feedback model, analyze the saline-alkali land soil improvement effect level. When the improvement effect feedback coefficient is lower than 0.4, the corresponding small saline-alkali land soil area has a low improvement effect; when the improvement effect feedback coefficient is between 0.4 and 0.75, the corresponding small saline-alkali land soil area has a medium improvement effect; when the improvement effect feedback coefficient is higher than 0.75, the corresponding small saline-alkali land soil area has a high improvement effect, and the improvement effect feedback level is obtained; Based on the obtained improvement effect feedback level, when the saline-alkali land soil improvement effect level and the improvement effect feedback level are the same, it will not be updated; when the saline-alkali land soil improvement effect level and the improvement effect feedback level are different, the saline-alkali land soil improvement effect level will be replaced by the improvement effect feedback level to update the saline-alkali land soil improvement effect level; Number the updated saline-alkali land soil improvement effect grades and integrate the updated saline-alkali land soil improvement effect grade numbers into the improvement effect monitoring data set; Step 5: Combine the iterative optimization algorithm with the updated saline-alkali land improvement effect level to build a digital twin evaluation model, and optimize the digital twin evaluation model through a deep neural network algorithm; Step 6: Based on the output results of the optimized digital twin evaluation model, a corresponding saline-alkali land soil improvement effect signal is issued.

2. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 1, characterized in that: In step 1, the process of collecting monitoring data on the improvement effect of saline-alkali land includes: The target research area of ​​saline-alkali soil is divided into several small saline-alkali soil areas, which have the same area and are numbered respectively; Deploy different data collection equipment to collect multi-dimensional soil data, crop growth data, and meteorological data from each small saline-alkali soil area, including pH meters, conductivity meters, ion chromatographs, soil nitrogen, phosphorus, and potassium sensors, soil corers, mercury intrusion porosimeter, laser particle size analyzer, crop height meter, crop angle stem diameter meter, leaf area meter, crop yield monitor, tipping bucket rain gauge, temperature sensor, and sunshine duration sensor; The multidimensional soil data includes soil pH value, total salt content, salt ion concentration, nitrogen content, phosphorus content, potassium content, total volume, particle size, and volume of interparticle spaces; the crop growth data includes crop plant height, stem diameter, leaf area, and yield per unit area; and the meteorological data includes precipitation, temperature, and sunshine duration.

3. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 2, characterized in that: In step 1, the process of preprocessing the collected improvement effect monitoring data includes: Perform data cleaning and data normalization on the collected improvement effect monitoring data, assign timestamps to the multi-dimensional soil data, crop growth data, and meteorological data, and adjust the timestamps to synchronize the collection time of the multi-dimensional soil data, crop growth data, and meteorological data; The multi-dimensional soil data, crop growth data and meteorological data are matched with the numbers of their corresponding small saline-alkali soil areas, and the multi-dimensional soil data, crop growth data and meteorological data are integrated to generate an improvement effect monitoring dataset, which is then divided into a training set and a test set.

4. The method for evaluating the effect of soil improvement on saline-alkali land based on digital twins according to claim 3, characterized in that: In step 2, the calculation process of the soil salinization index, fertility index, porosity and permeability includes: According to the principle of soil salinization, weights are assigned to the soil pH value, total salt content and salt ion concentration respectively, and the soil salinization index is calculated using the weighted average method; The soil fertility index was calculated by adding the nitrogen content, phosphorus content and potassium content of the soil to 1 and taking the logarithm. The volume of the voids between soil particles is calculated as a percentage of the total soil volume to obtain the soil porosity. The average soil particle size is calculated and the soil permeability is calculated based on the Kozeny-Carman equation.

5. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 4, characterized in that: In step 3, the process of obtaining the improvement effect feedback coefficient includes: Extract the crop growth data and meteorological data from the improvement effect monitoring dataset, use the crop growth data and meteorological data in the training set, combine with the multiple linear regression algorithm, take the crop growth data and meteorological data as input, and take the improvement effect feedback coefficient as output, learn the linear relationship between the crop growth data, meteorological data and the improvement effect feedback coefficient, and train the improvement effect feedback model; The crop growth data and meteorological data in the test set are input into the improvement effect feedback model to evaluate the performance of the improvement effect feedback model. The improvement effect feedback model is optimized by adjusting the intercept term and regression coefficient of the improvement effect feedback model to obtain the final improvement effect feedback model. The current crop growth data and meteorological data of each small saline-alkali soil area are combined to output the corresponding improvement effect feedback coefficient.

6. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 5, characterized in that: In step 5, the construction process of the digital twin evaluation model includes: Extract the multi-dimensional soil data and updated saline-alkali land soil improvement effect grade number from the improvement effect monitoring dataset; Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with an iterative optimization algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The association between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the digital twin evaluation model; The multi-dimensional soil data in the test set is input into the digital twin evaluation model, and the error between the output results of the digital twin evaluation model and the actual updated saline-alkali land soil improvement effect grade number is compared to evaluate the performance of the digital twin evaluation model, adjust the parameters of the digital twin evaluation model, improve the performance of the digital twin evaluation model, and construct a digital twin evaluation model.

7. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 6, characterized in that: In step 5, the optimization process of the digital twin evaluation model includes: Using the multi-dimensional soil data in the training set and the updated saline-alkali land soil improvement effect grade number, combined with the deep neural network algorithm, the multi-dimensional soil data is used as input and the updated saline-alkali land soil improvement effect grade number is used as output. The nonlinear relationship between the multi-dimensional soil data and the updated saline-alkali land soil improvement effect grade number is learned to train the deep neural network model; The multi-dimensional soil data in the test set is input into the deep neural network model, and the error between the output of the deep neural network model and the actual updated saline-alkali land soil improvement effect grade number is compared. The deep neural network model parameters are characterized, and the deep neural network model is optimized to obtain the final deep neural network model. The current multi-dimensional soil data of each small saline-alkali land soil area is used to output the corresponding updated saline-alkali land soil improvement effect grade number; The digital twin evaluation model is trained using the updated saline-alkali land soil improvement effect grade number output by the deep neural network model, and the parameters of the digital twin evaluation model are adjusted again. The digital twin evaluation model is optimized to obtain the final digital twin evaluation model. Combined with the current multi-dimensional soil data of each small saline-alkali land soil area, the corresponding updated saline-alkali land soil improvement effect grade number is output.

8. The method for evaluating the effect of saline-alkali land improvement based on digital twins according to claim 7, characterized in that: In step 6, the process of issuing a corresponding saline-alkali land soil improvement effect signal based on the output results of the digital twin evaluation model includes: Based on the updated saline-alkali land soil improvement effect grade number output by the optimized digital twin evaluation model, the saline-alkali land soil improvement effect grade corresponding to the updated saline-alkali land soil improvement effect grade number is matched; An electronic display screen is set up in the target research area of ​​saline-alkali soil, and the updated saline-alkali soil improvement effect level of each small saline-alkali soil area is displayed on the screen in text form, including the electronic display screen displaying low words, the electronic display screen displaying medium words and the electronic display screen displaying high words; When the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is low improvement effect, the electronic display screen displays the word "low"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is medium improvement effect, the electronic display screen displays the word "medium"; when the level of the saline-alkali soil improvement effect after the small saline-alkali soil area is updated is high improvement effect, the electronic display screen displays the word "high".

Images