Plain reservoir ecological slope protection evaluation method and system based on random forest model

Abstract

The invention relates to the technical field of environment monitoring and ecological environment protection, in particular to a plain reservoir ecological slope protection evaluation method and system based on a random forest model, and the method comprises the following steps: building a health evaluation index system based on health evaluation indexes of plain reservoir ecological slope protection; grading the health evaluation indexes to obtain health grading standards of the plain reservoir ecological slope protection; obtaining health evaluation index weight distribution of the plain reservoir ecological slope protection; determining a health evaluation simulation scene of the plain reservoir ecological slope protection according to the health evaluation indexes; and establishing a health evaluation model according to a random forest model, and performing health evaluation on the health evaluation simulation scene so as to realize evaluation of the plain reservoir ecological slope protection. The health evaluation model is constructed based on the random forest algorithm, the multi-index evaluation system is used for ecological health evaluation, and the health condition of the plain reservoir ecological slope protection is objectively reflected.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and ecological environment protection, in particular to an evaluation method and system for the ecological slope protection of plain reservoirs based on a random forest model. Background Art

[0002] Urban plain reservoirs or artificial lakes are not only water source projects for water resource regulation and utilization, but also important elements of the urban ecological landscape pattern. Therefore, plain reservoirs (artificial lakes) are one of the key points in the construction of urban water resource infrastructure, and at the same time, they are also the main way to create water conservancy project scenic spots. In the process of operating, managing, and restoring urban lakes and reservoirs, it is necessary to evaluate the health status of the water ecosystem of urban lakes and reservoirs. The ecological slope protection project is an important part of lakes and reservoirs, and its health status directly affects the water ecological health of lakes and reservoirs. Therefore, it is very necessary to construct a practical and feasible health evaluation method and system for reservoir ecological slope protection and strengthen the research on its related technologies to scientifically formulate effective protection measures.

[0003] At present, the health evaluation of water ecosystems in China is still in the initial stage of development. The common health evaluation methods in research mainly include the indicator species method and the index system method. The indicator species method mainly diagnoses the health status of surface water ecosystems by the changes in the biomass, productivity, certain functions, and physiological indicators of single or multiple species in the ecosystem; the index system method selects multiple relevant indicators considering various characteristics required for healthy surface water bodies, aiming to comprehensively reflect the changes in the structure, function, service capacity, etc. of surface water ecosystems such as lakes and reservoirs. It is the currently more commonly used method for evaluating the health of surface water ecosystems. The present invention also adopts the index system method to evaluate the health of the ecological slope protection of plain reservoirs.

[0004] After the traditional reservoir ecological slope protection project is completed, there are problems such as fewer or missing items of concern or monitoring of the health evaluation indicators involved by researchers or management departments, and insufficient relevant data, which are unfavorable for implementing the health evaluation of ecological slope protection. For newly built ecological slope protection projects, the same problem of lack of actual monitoring data also exists. Therefore, although researchers in this field have proposed various water ecological health evaluation methods, most of them are established for natural rivers, lakes and other water bodies and are not applicable to the evaluation of ecological slope protection of urban plain reservoirs or artificial lakes. Therefore, after establishing an evaluation index system for the ecological slope protection of plain reservoirs based on the analytic hierarchy process, the evaluation indicators are assigned values and input into a supervised random forest model for training after normalization processing, and relevant simulation scenarios are designed to carry out simulation evaluation, and finally the health evaluation results of the ecological slope protection are obtained for decision-makers to refer to. Summary of the Invention

[0005] Aiming at the deficiencies in the prior art, the present invention provides an evaluation method and system for the ecological slope protection of plain reservoirs based on a random forest model.

[0006] To achieve the above object, in the first aspect, the present invention provides an evaluation method for the ecological slope protection of plain reservoirs based on a random forest model. The method includes the following steps: establishing a health evaluation index system based on the health evaluation indexes of the ecological slope protection of plain reservoirs; dividing the health evaluation indexes into grades to obtain the health grade division standard of the ecological slope protection of plain reservoirs; obtaining the weight distribution of the health evaluation indexes of the ecological slope protection of plain reservoirs; determining the health evaluation simulation scenario of the ecological slope protection of plain reservoirs according to the health evaluation indexes; establishing a health evaluation model according to the random forest model, and conducting a health evaluation on the health evaluation simulation scenario to achieve the evaluation of the ecological slope protection of plain reservoirs. By constructing a health evaluation index system, combining the health grade division standard and the weight distribution of health evaluation indexes, constructing a health evaluation simulation scenario, and using the random forest model to construct a health evaluation model to evaluate the health evaluation simulation scenario, the present invention improves the accuracy and objectivity, effectively reduces the interference of human factors, provides a reliable evaluation means for the health status of the ecological slope protection of plain reservoirs, helps to timely discover and solve potential problems, and ensures the health and stability of the ecological slope protection system of plain reservoirs.

[0007] Optionally, establishing a health evaluation index system based on the health evaluation indexes of the ecological slope protection of plain reservoirs includes: constructing the health comprehensive index of the ecological slope protection of plain reservoirs as the target layer; obtaining the evaluation index types according to the characteristics of the ecological slope protection of plain reservoirs to establish the criterion layer; selecting the health evaluation indexes based on the criterion layer to obtain the index layer; and establishing the health evaluation index system of the ecological slope protection of plain reservoirs by combining the target layer, the criterion layer and the index layer. By constructing the health evaluation index system of the ecological slope protection of plain reservoirs, the present invention realizes the hierarchical and structured management from specific health evaluation indexes to comprehensive evaluation indexes; the health comprehensive index of the target layer provides an intuitive and quantitative basis for the overall evaluation, the classification of the criterion layer clearly defines the attributes and functions of different evaluation indexes, and the detailed establishment of the index layer ensures the comprehensiveness and pertinence of the evaluation; improves the accuracy and scientificity of the evaluation, and helps to guide the subsequent health management and optimization work of the ecological slope protection of plain reservoirs.

[0008] Optionally, the method of grading the health evaluation indicators to obtain the health grading standard for the ecological slope protection of the plain reservoir includes: obtaining the grading of the health evaluation indicators for the ecological slope protection of the plain reservoir; obtaining the grading result of the health evaluation indicators based on the grading of the health evaluation indicators; and establishing the health grading standard for the ecological slope protection of the plain reservoir based on the grading result. Through the grading of health evaluation indicators, the present invention provides a clear definition standard for the health status of the ecological slope protection of the plain reservoir, enables the evaluation indicators to be quantitatively evaluated, ensures the objectivity and consistency of the evaluation results, and establishes a health grading standard based on the grading result, providing a scientific basis for the health management of the ecological slope protection, helping to timely discover potential ecological health problems, and promoting the health management of the ecological system of the plain reservoir.

[0009] Optionally, the step of obtaining the grading result of the health evaluation indicators based on the grading of the health evaluation indicators includes: normalizing the original data of the health evaluation indicators to obtain standardized data for constructing the grading result. The normalization process includes positive indicator normalization and negative indicator normalization. The positive indicator normalization and the negative indicator normalization respectively satisfy the following relationships: ; ; where is the standardized data, is the original data, is 0.1 times the lower threshold of the health evaluation indicator, is 2 times the upper threshold of the health evaluation indicator. Through the standardization process of positive and negative indicators, the present invention effectively solves the problems of difference and incomparability of the original data of health evaluation indicators. By setting reasonable multiples of the lower threshold and the upper threshold, the accuracy and reasonableness of the standardized data are ensured, providing a reliable basis for the grading of health evaluation indicators, enhancing the objectivity and scientific nature of the evaluation results, and helping to more accurately evaluate the health status of the ecological slope protection of the plain reservoir.

[0010] Optionally, obtaining the weight distribution of the health evaluation indicators for the ecological slope protection of the plain reservoir includes: constructing a first judgment matrix for the criterion layer; assigning relative importance to the evaluation indicator types according to the first judgment matrix to obtain a first assignment result; obtaining a first weight vector for the evaluation indicator types based on the first assignment result; constructing a second judgment matrix for the indicator layer; assigning relative importance to the health evaluation indicators according to the second judgment matrix to obtain a second assignment result; obtaining a second weight vector for the health evaluation indicators based on the second assignment result; and combining the first weight vector and the second weight vector to obtain the weight distribution of the health evaluation indicators. By constructing judgment matrices and assigning relative importance to the evaluation indicators, the present invention determines the weight distribution of the health evaluation indicators for the ecological slope protection of the plain reservoir. The establishment of the first judgment matrix and the second judgment matrix provides a data basis for weight calculation. By combining the first weight vector and the second weight vector, an accurate and reasonable weight distribution is obtained, effectively improving the accuracy and credibility of the health evaluation, and providing strong support for the health management of the ecological slope protection of the plain reservoir.

[0011] Optionally, combining the first weight vector and the second weight vector to obtain the weight distribution of the health evaluation indicators includes: performing consistency verification on the first weight vector and the second weight vector to obtain a consistency ratio, and the consistency verification satisfies the following relationship: ; ; where is the consistency index, is the maximum eigenvalue, is the matrix order of the judgment matrix, is the consistency ratio, is the average random consistency index; obtaining the effective first weight vector and the effective second weight vector based on the consistency ratio to determine the weight distribution of the health evaluation indicators. Through consistency verification, the present invention ensures the rationality and reliability of the first weight vector and the second weight vector. By using the consistency index and the consistency ratio, combined with the average random consistency index, the consistency degree of the judgment matrix is effectively evaluated, improving the accuracy and scientificity of weight allocation, enhancing the rationality of the weight distribution of the health evaluation indicators, laying a reliable data basis for subsequent health evaluation, and ensuring the objectivity and credibility of the evaluation results.

[0012] Optionally, determining the health evaluation simulation scenario of the ecological slope protection of the plain reservoir according to the health evaluation index includes: assigning values to the health evaluation index according to the weight distribution of the health evaluation index to construct a health evaluation simulation scenario. The present invention assigns values to the evaluation index according to the weight distribution of the health evaluation index, ensures that the health evaluation simulation scenario can truly reflect the health status of the ecological slope protection of the plain reservoir, and highlights the influence of key indicators. Through scientific value assignment, the health evaluation simulation scenario is closer to the actual situation, improving the accuracy and practicability of the health evaluation, and providing strong support for the health management and maintenance of the ecological slope protection of the plain reservoir.

[0013] Optionally, establishing the health evaluation model of the ecological slope protection of the plain reservoir according to the random forest model includes: obtaining health evaluation samples according to the health grade division standard; normalizing the health evaluation samples to obtain normalized health evaluation samples; based on the random forest model, combining the normalized health evaluation samples to obtain a health evaluation model. The present invention effectively improves the quality and usability of data by obtaining health evaluation samples and normalizing them. Combining with the random forest model and using the normalized health evaluation samples for model construction improves the accuracy and stability of the health evaluation model, enhances the generalization ability of the model, provides reliable technical support for the health evaluation of the ecological slope protection of the plain reservoir, and helps to achieve accurate and efficient health evaluation and management.

[0014] Optionally, conducting a health evaluation on the health evaluation simulation scenario includes: using the health evaluation index of the health evaluation simulation scenario as the input data of the health evaluation model; based on the input data, combining with the health evaluation model to obtain the health comprehensive index of the health evaluation simulation scenario as the output data to achieve the health evaluation of the health evaluation simulation scenario. The present invention uses the health evaluation index of the health evaluation simulation scenario as the input data, and obtains the health comprehensive index of the health evaluation simulation scenario through the health evaluation model, realizing the accurate health evaluation of the health evaluation simulation scenario. It not only improves the objectivity and scientificity of the evaluation, but also can comprehensively reflect the health status of the ecological slope protection system. Taking the health comprehensive index as the model output, engineering personnel can intuitively understand the health status of the health evaluation simulation scenario of the ecological slope protection of the plain reservoir. On this basis, according to the actual monitoring data and relevant statistical materials of the ecological slope protection, it is beneficial to complete the health status evaluation of the operating project by the established evaluation system of the ecological slope protection of the plain reservoir, providing a strong basis for formulating targeted health management and optimization measures, and helping to promote the health and sustainable development of the ecological system of the plain reservoir.

[0015] Second aspect, the present invention provides an ecological slope protection evaluation system for plain reservoirs based on a random forest model. The system executes the ecological slope protection evaluation method for plain reservoirs based on a random forest model provided by the present invention. The system includes an input device, an output device, a processor, and a memory. The advantage lies in that: the hardware facilities integrated in the present invention have excellent performance. The input device, the output device, the processor, and the memory are interconnected with each other, and the information transmission between each component is smooth. Through the interaction of multiple hardware facilities, an efficient information processing system is constructed. The system provided by the present invention optimizes the data processing flow as a whole, improves the evaluation efficiency and accuracy, provides strong hardware facility support for the health evaluation of the ecological slope protection of plain reservoirs, and helps to realize the scientific evaluation of the health of the ecological slope protection of plain reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of the ecological slope protection evaluation method for plain reservoirs based on a random forest model according to an embodiment of the present invention;

[0017] Figure 2 It is a framework diagram of the ecological slope protection evaluation method for plain reservoirs based on a random forest model according to an embodiment of the present invention;

[0018] Figure 3 It is a system flowchart of the ecological slope protection evaluation system for plain reservoirs according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described here are only for illustrative purposes and are not used to limit the present invention. In the following description, in order to provide a thorough understanding of the present invention, a large number of specific details are set forth. However, it will be apparent to those of ordinary skill in the art that: it is not necessary to employ these specific details to practice the present invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the present invention.

[0020] Throughout the specification, the reference to "an embodiment", "embodiment", "an example", or "example" means that: the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment of the present invention. Thus, the phrases "in an embodiment", "in embodiments", "an example", or "example" that appear throughout the specification do not necessarily all refer to the same embodiment or example. In addition, the specific features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments or examples. In addition, those of ordinary skill in the art should understand that the diagrams provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0021] Please refer to Figure 1, an embodiment of the present invention provides an evaluation method for the ecological slope protection of plain reservoirs based on a random forest model, and the method includes the following steps:

[0022] S1. Establish a health evaluation index system based on the health evaluation indexes of the ecological slope protection of plain reservoirs.

[0023] Among them, S1 specifically includes the following steps:

[0024] S11. Construct the health comprehensive index of the ecological slope protection of the plain reservoir as the target layer.

[0025] Specifically, comprehensively evaluate and analyze the economic characteristics, ecological characteristics, landscape characteristics, safety characteristics and construction characteristics of the ecological slope protection of the plain reservoir to obtain the health comprehensive index of the ecological slope protection system of the plain reservoir, and the health comprehensive index is used as the target layer.

[0026] S12. Obtain the evaluation index types according to the characteristics of the ecological slope protection of the plain reservoir to establish the criterion layer.

[0027] In this embodiment, classify the health evaluation indexes to obtain the evaluation index types, and then construct the criterion layer. The evaluation index types should be able to reflect the economic characteristics, ecological characteristics, landscape characteristics, safety characteristics and construction characteristics of the ecological slope protection of the plain reservoir.

[0028] S13. Select the health evaluation indexes based on the criterion layer to obtain the index layer.

[0029] In this embodiment, a total of 23 health evaluation indexes are selected as the index layer. The health evaluation indexes are divided into quantitative indexes and qualitative indexes, with quantitative indexes as the main and qualitative indexes as the supplement.

[0030] It should be noted that the selection of the health evaluation indexes of the ecological slope protection of the plain reservoir needs to meet the principles of scientificity, systematicness, dynamics, feasibility and hierarchy.

[0031] S14. Combine the target layer, the criterion layer and the index layer to establish the health evaluation index system of the ecological slope protection of the plain reservoir.

[0032] In this embodiment, combine the index layer, the criterion layer and the target layer, and explain the meanings of the health evaluation indexes, and then establish the health evaluation index system of the ecological slope protection of the plain reservoir. The health evaluation index system is shown in Table 1:

[0033] Table 1: Health evaluation index system of ecological slope protection of plain reservoir

[0034] S2. Classify the health evaluation indicators to obtain the health grade classification standard for the ecological slope protection of the plain reservoir.

[0035] Among them, S2 specifically includes the following steps:

[0036] S21. Obtain the classification grades of the health evaluation indicators for the ecological slope protection of the plain reservoir.

[0037] In this embodiment, in the health evaluation index system, the quantitative indicators include construction cost, maintenance cost, water permeability and water retention, water quality purification function, vegetation coverage rate, soil organic matter content, water quality status, plant adaptability rate, vegetation coverage degree, structural stability and anti-scouring property. The quantitative indicators are calculated through on-site measurement, field tests and query of existing data; other health evaluation indicators in the health evaluation index system are qualitative indicators, which need to be quantified by means of expert investigation and scoring.

[0038] Specifically, the classification grades of the health evaluation indicators for the quantitative indicators include good, relatively good, general, relatively poor and poor, as shown in Table 2:

[0039] Table 2: Classification grades of health evaluation indicators for quantitative indicators

[0040] It should be noted that C in the construction cost and maintenance cost indicators is the average cost per unit area after actual engineering research; η in the water quality purification function indicator is the average removal rate of the indicator with the highest removal rate among the water quality evaluation indicators; the water quality status score refers to the qualitative indicator, the score for Class I water quality is 8, the score for Class II water quality is 6, and so on.

[0041] Specifically, for the quantitative research of qualitative indicators, first determine the comment set and standard set of the qualitative indicators. The comment set and standard set satisfy the following relationship:

[0042]

[0043]

[0044] Among them, is the comment set, is the comment, is the standard set, is the standard.

[0045] In this embodiment, , the comment set and standard set satisfy the following relationship:

[0046]

[0047]

[0048] Among them, is the comment set, is the comment, is the standard set.

[0049] Furthermore, according to the health evaluation indicators of qualitative indicators, the grading includes good, relatively good, general, relatively poor, and poor. At the same time, based on the existing engineering data, each qualitative indicator is judged and scored respectively.

[0050] S22. Obtain the grading result of the health evaluation indicators according to the grading of the health evaluation indicators.

[0051] In this embodiment, there are problems such as inconsistent data scales and dimensions in the original data of the health evaluation indicators. In order to eliminate the influence of different dimensions of each indicator, it is necessary to normalize the original data; the data range after normalization is 0-1, which is convenient for subsequent model training; for the health evaluation indicators, the health evaluation indicators with larger values, higher grades, and more ideal evaluation results are called positive indicators, and vice versa are negative indicators.

[0052] Specifically, the normalization processing of positive indicators and negative indicators respectively satisfies the following relationships:

[0053]

[0054]

[0055] Among them, is the standardized data, is the original data, is 0.1 times of the lower threshold of the health evaluation indicator, is 2 times of the upper threshold of the health evaluation indicator.

[0056] Furthermore, obtain the standardized data according to the standardization processing, and then obtain the grading result of the health evaluation indicators.

[0057] S23. Establish the health grading standard of the ecological slope protection of the plain reservoir based on the grading result.

[0058] In this embodiment, construct the health grading standard of the ecological slope protection of the plain reservoir based on the grading results of quantitative indicators and qualitative indicators, as shown in Table 3:

[0059] Table 3: Health grading standard of ecological slope protection of plain reservoir

[0060] S3. Obtain the weight distribution of the health evaluation indicators of the ecological slope protection of the plain reservoir.

[0061] Specifically, first construct the first judgment matrix of the criterion layer. Subsequently, according to the first judgment matrix, use the 1-9 scale method to assign relative importance values to the evaluation index types of the ecological slope protection of the plain reservoir to obtain the first assignment result. Finally, based on the first assignment result, use the summation method to obtain the first weight vector of the evaluation index types.

[0062] Specifically, first construct the second judgment matrix of the index layer. Subsequently, according to the second judgment matrix, use the 1-9 scale method to assign relative importance values to the health evaluation indexes of the ecological slope protection of the plain reservoir to obtain the second assignment result. Finally, based on the second assignment result, use the summation method to obtain the second weight vector of the health evaluation indexes.

[0063] It should be noted that when constructing the judgment matrix, the elements at the same level are compared pairwise in terms of their importance relative to a certain element in the previous level. The 1-9 scale method is adopted. The larger the number, the higher the relative importance of the former compared to the latter. The pairwise comparison results are organized into a matrix form. Taking the elements of the criterion layer as an example, if there are n criteria, an n×n matrix is constructed, and the element a ij represents the importance scale value of the i-th criterion relative to the j-th criterion.

[0064] In addition, when calculating the weight vector, first add up the elements in each column of the judgment matrix to obtain the total sum. Then divide each element by the total sum of its column to obtain the normalized matrix. Next, add up the elements in each row of the normalized matrix, and then normalize the obtained result to get the weight vector.

[0065] Furthermore, perform consistency verification on the first weight vector and the second weight vector to obtain the consistency ratio. The consistency verification satisfies the following relationship:

[0066]

[0067]

[0068] Among them, is the consistency index, is the maximum eigenvalue, is the matrix order of the judgment matrix, is the consistency ratio, is the average random consistency index.

[0069] Judge the effectiveness of the first weight vector and the second weight vector according to the consistency ratio. If , the weight vector is valid; otherwise, the judgment matrix needs to be adjusted and the weight vector needs to be recalculated.

[0070] According to the determined valid first weight vector and second weight vector, the final weight of each health evaluation index at the index layer is obtained by multiplying the weight of each health evaluation index by the weight of its corresponding criterion layer, so as to determine the weight distribution of the health evaluation indexes, as shown in Table 4:

[0071] Table 4: Weight Distribution of Health Evaluation Indexes for the Ecological Slope Protection of Plain Reservoirs

[0072] S4. Determine the health evaluation simulation scenario of the ecological slope protection of the plain reservoir according to the health evaluation indexes.

[0073] Specifically, assign values to the health evaluation indexes according to the weight distribution of the health evaluation indexes to construct a health evaluation simulation scenario; different values are assigned to the health evaluation indexes to construct different health evaluation simulation scenarios.

[0074] In an optional embodiment, among the quantitative indexes, the levels of the five quantitative indexes of structural stability, anti-scour ability, water permeability and water retention, plant coverage rate, and construction and maintenance costs are set as "poor", and the levels of the remaining quantitative indexes are set as "good", and the level of the qualitative index is set as "good". Based on the above settings, Scenario 1 of the health evaluation simulation scenario is constructed, as shown in Table 5:

[0075] Table 5: Health Evaluation Index Values of the Health Evaluation Simulation Scenario for the Ecological Slope Protection of Plain Reservoirs (Scenario 1)

[0076] In an optional embodiment, among the quantitative indexes, the levels of the five indexes of structural stability, anti-scour ability, water permeability and water retention, plant coverage rate, and construction and maintenance costs are set as "good", and the levels of the remaining quantitative indexes are set as "poor", and the qualitative index is set as "poor". Based on the above settings, Scenario 2 of the health evaluation simulation scenario is constructed, as shown in Table 6:

[0077] Table 6: Health Evaluation Index Values of the Health Evaluation Simulation Scenario for the Ecological Slope Protection of Plain Reservoirs (Scenario 2)

[0078] In an optional embodiment, the levels of the quantitative indexes are all set as "medium", and the level of the qualitative index is set as "good". Based on the above settings, Scenario 3 of the health evaluation simulation scenario is constructed, as shown in Table 7:

[0079] Table 7: Health Evaluation Index Values of the Health Evaluation Simulation Scenario for the Ecological Slope Protection of Plain Reservoirs (Scenario 3)

[0080] S5. Establish a health evaluation model based on the random forest model to conduct a health evaluation on the health evaluation simulation scenario, so as to realize the evaluation of the ecological slope protection of the plain reservoir.

[0081] Among them, S5 specifically includes the following steps:

[0082] S51. Obtain health evaluation samples according to the health level division standard.

[0083] In this embodiment, according to the threshold ranges of each health evaluation index in the health level division standard, 600 groups of samples are randomly generated as health evaluation samples, among which 120 groups of samples are randomly generated for each health level, and a total of 600 groups of samples for the five health levels; the health evaluation samples include training samples and test samples, and the first 500 groups are selected as training samples, and the last 100 groups are selected as test samples, as shown in Table 8:

[0084] Table 8: Design of health evaluation samples and expected outputs for the ecological slope protection of the plain reservoir

[0085] S52. Normalize the health evaluation samples to obtain normalized health evaluation samples.

[0086] In this embodiment, the positive indicators include service life (C2), water permeability and water retention (C3), hydrological regulation function (C4), soil and water conservation benefit (C5), soil hydrophilicity (C6), soil organic matter content (C7), water quality purification function (C8), water quality status (C9), biodiversity (C10), vegetation coverage rate (11), plant suitability rate (C12), sustainability (C13), landscape diversity (C14), water bank landscape suitability (C15), flood control (C16), waterlogging drainage (C17), structural stability (C18) and construction speed (C23), a total of 18 positive indicators, and the rest of the health evaluation indicators are negative indicators.

[0087] Furthermore, use the positive indicator normalization relation formula and negative indicator normalization relation formula in step S22 to normalize the health evaluation samples to obtain normalized health evaluation samples. The normalized health evaluation samples include normalized training samples and normalized test samples. After normalization, the values of the health evaluation indicators at the index layer under each health evaluation simulation scenario are shown in Table 9:

[0088] Table 9: Values of health evaluation indicators at the index layer under each health evaluation simulation scenario after normalization

[0089] S53. Based on the random forest model, obtain a health evaluation model in combination with the normalized health evaluation samples.

[0090] In this embodiment, a health evaluation model for the ecological slope protection of plain reservoirs with "multiple inputs and one output" is established based on a random forest model using normalized training samples, and the health evaluation model is trained, tested, and optimized based on the normalized test samples, so as to obtain a health evaluation model with excellent performance.

[0091] S54. Use the health evaluation indicators of the health evaluation simulation scenario as the input data of the health evaluation model.

[0092] Specifically, select the normalized training samples of the health evaluation indicators of the health evaluation simulation scenario as the input data of the health evaluation model.

[0093] S55. Based on the input data, combine the health evaluation model to obtain the health comprehensive index of the health evaluation simulation scenario as the output data, so as to realize the health evaluation of the health evaluation simulation scenario.

[0094] Specifically, input the input data into the health evaluation model, and use the health comprehensive index of the target output in the corresponding health evaluation sample group as the model output of the health evaluation model to obtain the ecological health evaluation result of the health evaluation simulation scenario of the ecological slope protection of the plain reservoir, as shown in Table 10:

[0095] Table 10: Ecological health evaluation results of the health evaluation simulation scenario

[0096] It should be noted that for the ecological slope protection project of newly built reservoirs (artificial lakes), several operation scenarios can be set for simulation evaluation according to the purpose of health evaluation; for the ecological slope protection project of reservoirs (artificial lakes) that have been built and actually put into operation, methods such as on-site monitoring and sampling and expert scoring can be adopted according to the type of evaluation indicators for measurement, and the ecological health status evaluation can be completed based on the established health evaluation system developed based on the random forest model.

[0097] S6. Analyze the reliability of the ecological health evaluation results.

[0098] Specifically, based on Table 10, compare and analyze the influence degrees of different criterion layer indicators and the actual situation of the ecological slope protection of the reservoir, so as to provide suggestions and support for the future proposed technical and engineering measures; for the ecological slope protection project of newly built plain reservoirs, set three health evaluation simulation scenarios, and their evaluation results are "good", "poor", and "good" respectively; thus, it can be seen that the health status of most health evaluation indicators determines the health status of the ecological slope protection of the plain reservoir. In order to make the health status of the ecological slope protection reach a good level, it is necessary to ensure that more than half of the health evaluation indicators are in a good health status.

[0099] It should be noted that as a newly built ecological slope protection for plain reservoirs, when conducting construction and health assessment, more attention should be paid to ensuring that various health assessment indicators at the safety and ecological levels are at the "good" level to guarantee its safety, usability, and ecological health. In the future operation and maintenance of the ecological slope protection, attention should also be paid to ensuring that various ecological and safety indicators are in a maintainable and repairable state, so as to achieve an ideal state in the ecological civilization construction of the plain reservoir ecological slope protection.

[0100] Please refer to Figure 2 , in an optional embodiment, in order to efficiently execute the evaluation method for the plain reservoir ecological slope protection based on the random forest model provided by the present invention, the present invention provides an evaluation system for the plain reservoir ecological slope protection based on the random forest model. The system includes an input device, an output device, a processor, and a memory, and these hardware facilities are interconnected. Among them, the memory is used to store a computer program, and the computer program includes program instructions. The processor is configured to call the program instructions to execute the specific steps of the related embodiments of the evaluation method for the plain reservoir ecological slope protection based on the random forest model provided by the present invention. The evaluation system for the plain reservoir ecological slope protection based on the random forest model provided by the present invention has a complete, objective, and stable structure, which improves the overall applicability and practical application ability of the present invention.

[0101] Please refer to Figure 3 , which shows the system flow chart of the evaluation system for the plain reservoir ecological slope protection based on the random forest model, elaborately demonstrates the operation steps of the plain ecological slope protection evaluation system, and describes the health evaluation process of the system for the plain reservoir ecological slope protection.

[0102] In summary, the evaluation method and system for the plain reservoir ecological slope protection based on the random forest model provided by the method of the present invention fill the technical gap in the health evaluation of the plain reservoir ecological slope protection, have strong operability and applicability, and can obtain reasonable, scientific, and practical index relative importance information through the training of the machine learning model for samples in the simulated scenario, ensuring the practicability of the health evaluation of the newly built ecological slope protection project; it also has good applicability to the health evaluation of the ecological slope protection project of urban artificial lakes; it can comprehensively, accurately, and objectively reflect the health status of the plain reservoir ecological slope protection system. The method of the present invention is easy to understand, simple to calculate, with less workload, and is convenient for engineering application, providing a theoretical basis and technical support for the further development of environmental monitoring and ecological environmental protection technologies.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. An ecological slope protection evaluation method for plain reservoirs based on a random forest model, characterized in that, It includes the following steps: Establish a health evaluation index system based on the health evaluation indexes of the ecological slope protection of the plain reservoir; Classify the health evaluation indexes to obtain the health grade classification standard of the ecological slope protection of the plain reservoir; Obtain the weight distribution of the health evaluation indexes of the ecological slope protection of the plain reservoir; Determine the health evaluation simulation scenario of the ecological slope protection of the plain reservoir according to the health evaluation indexes; Establish a health evaluation model based on the random forest model, and conduct a health evaluation on the health evaluation simulation scenario to realize the evaluation of the ecological slope protection of the plain reservoir.

2. The ecological slope protection evaluation method for plain reservoirs based on the random forest model according to claim 1, wherein The establishment of the health evaluation index system based on the health evaluation indexes of the ecological slope protection of the plain reservoir includes: Construct the health comprehensive index of the ecological slope protection of the plain reservoir as the target layer; Obtain the evaluation index types according to the characteristics of the ecological slope protection of the plain reservoir to establish the criterion layer; Select the health evaluation indexes based on the criterion layer to obtain the index layer; Establish the health evaluation index system of the ecological slope protection of the plain reservoir by combining the target layer, the criterion layer and the index layer.

3. The ecological slope protection evaluation method for plain reservoirs based on the random forest model according to claim 1, wherein The classification of the health evaluation indexes to obtain the health grade classification standard of the ecological slope protection of the plain reservoir includes: Obtain the classification grades of the health evaluation indexes of the ecological slope protection of the plain reservoir; Obtain the grade classification results of the health evaluation indexes according to the classification grades of the health evaluation indexes; Establish the health grade classification standard of the ecological slope protection of the plain reservoir based on the grade classification results.

4. The ecological slope protection evaluation method for plain reservoirs based on a random forest model according to claim 3, characterized in that The obtaining of the grade classification results of the health evaluation indexes according to the classification grades of the health evaluation indexes includes: Normalize the original data of the health evaluation indexes to obtain standardized data to construct the grade classification results. The normalization process includes positive index normalization and negative index normalization; The positive index normalization and the negative index normalization respectively satisfy the following relationships: ； ； Among them, is the standardized data, is the original data, is 0.1 times the lower threshold of the health evaluation index, is 2 times the upper threshold of the health evaluation index.

5. The ecological slope protection evaluation method for plain reservoirs based on a random forest model according to claim 2, wherein The obtaining of the weight distribution of the health evaluation indexes of the ecological slope protection of the plain reservoir includes: Construct the first judgment matrix of the criterion layer; Assign relative importance to the evaluation index types according to the first judgment matrix to obtain the first assignment result; Obtain the first weight vector of the evaluation index types based on the first assignment result; Construct the second judgment matrix of the index layer; Assign relative importance to the health evaluation indexes according to the second judgment matrix to obtain the second assignment result; Obtain the second weight vector of the health evaluation indexes based on the second assignment result; Combine the first weight vector and the second weight vector to obtain the weight distribution of the health evaluation indexes.

6. The ecological slope protection evaluation method for plain reservoirs based on a random forest model according to claim 5, characterized in that, The combination of the first weight vector and the second weight vector to obtain the weight distribution of the health evaluation indexes includes: Conduct consistency verification on the first weight vector and the second weight vector to obtain the consistency ratio. The consistency verification satisfies the following relationship: ； ； Among them, is the consistency index, is the maximum eigenvalue, is the matrix order of the judgment matrix, is the consistency ratio, is the average random consistency index; Obtain the effective first weight vector and the effective second weight vector according to the consistency ratio to determine the weight distribution of the health evaluation indexes.

7. The ecological slope protection evaluation method for plain reservoirs based on the random forest model according to claim 1, wherein The determination of the health evaluation simulation scenario of the ecological slope protection of the plain reservoir according to the health evaluation indexes includes: Assign values to the health evaluation indicators according to the weight distribution of the health evaluation indicators to determine a health evaluation simulation scenario.

8. The ecological slope protection evaluation method for plain reservoirs based on the random forest model according to claim 1, characterized in that, The establishment of the health evaluation model for the ecological slope protection of the plain reservoir according to the random forest model includes: Obtain health evaluation samples according to the health level division standard; Normalize the health evaluation samples to obtain normalized health evaluation samples; Based on the random forest model, obtain a health evaluation model in combination with the normalized health evaluation samples.

9. The ecological slope protection evaluation method for plain reservoirs based on a random forest model according to claim 2, wherein The health evaluation of the health evaluation simulation scenario includes: Use the health evaluation indicators of the health evaluation simulation scenario as the input data of the health evaluation model; Based on the input data, obtain the health comprehensive index of the health evaluation simulation scenario as the output data in combination with the health evaluation model to achieve the health evaluation of the health evaluation simulation scenario.

10. An ecological slope protection evaluation system for plain reservoirs based on a random forest model, characterized in that, The system includes an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. Among them, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the evaluation method for the ecological slope protection of the plain reservoir based on the random forest model according to any one of claims 1-9.

Images