A method and system for dynamically determining the impact of construction projects on hydrological monitoring.

Abstract

This invention discloses a method and system for dynamically determining the impact of construction projects on hydrological monitoring, relating to the field of hydrological monitoring technology. The specific steps are as follows: constructing an evaluation index system that couples static and dynamic indicators; assigning static weights to static indicators using the analytic hierarchy process (AHP) and dynamic weights to dynamic indicators using the entropy weight method; calculating the total weight based on the static and dynamic weights; constructing and training a machine learning model; inputting the static, dynamic, and total weight data into the trained machine learning model; and outputting the impact level; automatically matching response measures based on the impact level and trend prediction results. This invention proposes a dynamic weight correction algorithm based on project stage and hydrological conditions, applying machine learning algorithms to determine the impact level of construction projects on hydrological monitoring, achieving a closed-loop linkage from level determination to measure matching, and solving the problems of singular and disconnected evaluation in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of hydrological monitoring technology, and in particular to a method and system for determining the dynamic level of the impact of construction projects on hydrological monitoring. Background Technology

[0002] During the construction and operation of hydrological monitoring projects in the river sections measured by hydrological stations, hydrological monitoring work can be affected by altering river morphology, disturbing water flow and sediment, and interfering with monitoring facilities and equipment. Determining the degree of impact is a core basis for project approval, scheduling, and the formulation of mitigation measures. Existing methods rely on fixed parameters for one-time assessments, failing to consider the dynamic changes during construction and hydrological rhythms. This leads to significant discrepancies between the assessment results and the actual impact. Furthermore, the use of fixed weights to calculate impact levels cannot adapt to changes in indicator priorities under different scenarios. Manual integration of multi-dimensional data, including data on the river section, facilities and equipment, and hydrological data, is inefficient, subjective, and prone to bias due to differences in the experience of assessment personnel. When the impact level escalates due to changes in project or hydrological conditions, the response is delayed, failing to meet the real-time scheduling needs of the project. Therefore, for those skilled in the art, how to dynamically and efficiently assess the degree of hydrological impact is an urgent problem to be solved. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for dynamically determining the impact of construction projects on hydrological monitoring, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following solution: On one hand, a method for dynamically determining the impact of construction projects on hydrological monitoring, the specific steps of which include the following:

[0005] Construct an evaluation index system that couples static and dynamic indicators. Static indicators include station characteristic indicators and engineering design indicators, while dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators.

[0006] The static indicators are assigned static weights using the analytic hierarchy process, and the dynamic indicators are assigned dynamic weights using the entropy weight method. The total weight is calculated based on the static weights and the dynamic weights.

[0007] Build and train a machine learning model, input the static indicators, the dynamic indicators and the total weight data into the trained machine learning model, and output the influence level and trend prediction results.

[0008] Response measures will be automatically matched based on the impact level and trend prediction results.

[0009] Preferably, the station characteristic indicators include the station function level, the test element type, and the monitoring data accuracy level.

[0010] Preferably, the engineering design indicators include the distance between the project and the monitoring station, and the type of river-related project.

[0011] Preferably, the construction dynamic indicators include the proportion of water-blocking area of ​​river-related structures and the intensity of construction movement disturbance; the construction dynamic indicators are collected in real time through Internet of Things sensors.

[0012] Preferably, the hydrological dynamic indicators include real-time water level / flow rate / turbidity change rate, which are collected at a fixed frequency, and the collected hydrological dynamic indicator data are transmitted to the server via a communication module.

[0013] By adopting the above technical solution, the following beneficial technical effects can be achieved: a static and dynamic coupled evaluation index system can be constructed, which solves the problem that the existing technical indicators are too single and do not cover dynamic influencing factors, and provides comprehensive data support for dynamic judgment.

[0014] Preferably, the specific steps for training the machine learning model are as follows:

[0015] Retrieve historical engineering evaluation case databases, extract indicator data and actual grade data, construct a sample set, and divide it into training set and test set according to proportions;

[0016] A machine learning model is built by inputting a feature matrix of indicators and total weights, minimizing the log loss function by adaptively adjusting the learning rate and combining it with an early stopping strategy, and iteratively generating gradient boosting decision trees until the accuracy of the validation set reaches a set threshold, thus completing the training of the machine learning model.

[0017] By adopting the above technical solution, the following beneficial technical effects can be achieved: automatic level determination, elimination of manual intervention, and improvement of determination accuracy and real-time performance.

[0018] Preferably, the method also includes transmitting the impact level, trend curve, and response measures to the engineering scheduling terminal via a communication module, and automatically updating the response measures when the impact level changes.

[0019] On the other hand, a dynamic level determination system for the impact of construction projects on hydrological monitoring is provided, including a data acquisition module, a weight calculation module, a level determination module, and a measure matching module; among which,

[0020] The data acquisition module is used to construct an evaluation index system that couples static and dynamic indicators. The static indicators include station characteristic indicators and engineering design indicators, while the dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators.

[0021] The weight calculation module is used to assign static weights to the static indicators using the analytic hierarchy process, assign dynamic weights to the dynamic indicators using the entropy weight method, and calculate the total weight based on the static weights and the dynamic weights.

[0022] The level judgment module is used to construct and train a machine learning model, input the static indicators, the dynamic indicators and the total weight data into the trained machine learning model, and output the influence level and trend prediction results.

[0023] The measure matching module is used to automatically match response measures based on the impact level and trend prediction results.

[0024] Preferably, it also includes a report generation module, used to generate a level determination report based on the impact level output by the level determination module.

[0025] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0026] (1) Strong dynamic adaptability and high accuracy of judgment: The indicator system covers dynamic factors throughout the entire construction cycle of the project. The weights are adjusted in real time according to the project stage and hydrological conditions. The machine learning model can capture the time series data pattern, which solves the problem of large deviation between the static judgment of the existing technology and the actual impact.

[0027] (2) Improved judgment efficiency without human intervention: Compared with existing technologies, the efficiency is greatly improved, while eliminating subjective bias caused by human experience and ensuring the objectivity of the judgment results. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of the method of the present invention;

[0030] Figure 2 This is a system structure diagram of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] The purpose of this invention is to provide a method for dynamically determining the impact of construction projects on hydrological monitoring, such as... Figure 1 As shown, the specific steps include the following:

[0033] S1. Construct an evaluation index system that couples static and dynamic indicators. Static indicators include station characteristic indicators and engineering design indicators, while dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators.

[0034] S2. For static indicators, the analytic hierarchy process is used to assign static weights, and for dynamic indicators, the entropy weight method is used to assign dynamic weights. The total weight is calculated based on the static weights and dynamic weights.

[0035] S3. Build and train a machine learning model. Input static indicators, dynamic indicators and total weight data into the trained machine learning model and output the influence level and trend prediction results.

[0036] S4. Automatically match response measures based on the impact level and trend prediction results.

[0037] The impact levels are divided into four levels: virtually no impact, slight impact, moderate impact, and severe impact; the trend prediction results refer to the trend prediction and level determination of the impact of construction projects on hydrological monitoring by using machine learning models to extract time-series features.

[0038] Furthermore, an evaluation index system coupling static and dynamic indicators is constructed in S1. Static and dynamic indicators are collected, with static indicators remaining consistent throughout the process, while dynamic indicators are adjusted according to the characteristics of different periods, ensuring the continuity and relevance of the impact assessment throughout the entire construction cycle. Specifically, static indicator collection involves directly inputting engineering design indicators (including distance between the project and the monitoring station, type of river-related project, etc.) and simultaneously reading station characteristic indicator data (including station functional level, test element type, monitoring data accuracy, etc.) from the hydrological station management system via HTTP / HTTPS protocol. Dynamic indicator collection...

[0039] The system collects construction dynamic indicators (proportion of water-blocking area of ​​river-related structures, intensity of construction and operation disturbance) and hydrological dynamic indicators (real-time water level, flow rate, maximum change rate of turbidity change rate within one hour) in real time through various IoT sensors (such as distributed high-precision fiber optic vibration sensors, Doppler current meters, and optical backscattering turbidimeters). The collection frequency is 5 minutes per time. The data is transmitted to the server via a 5G / BeiDou short message communication module, and the indicators are assigned values ​​according to preset rules.

[0040] Furthermore, in S2, the dynamic total weight is calculated based on the "entropy weight-analytic hierarchy process". This specifically includes the following steps:

[0041] S21. Static weight calculation (Analytic Hierarchy Process)

[0042] S211. Hierarchical Structure Construction. A two-level hierarchical structure is constructed: Target Layer: Static subsystem score of the impact of the construction project on hydrological monitoring (denoted as...). ); Indicator layer: static indicators sorted according to a preset order. In this embodiment, five static indicators are selected: distance between the project and the station, type of river-related project, functional level of the station, type of test element, and accuracy level of monitoring data.

[0043] S212, Judgment Matrix Construction and Consistency Check

[0044] For the indicators at the indicator layer, pairwise importance is determined using the 1-9 scale (1 = both indicators are equally important, 3 = the former is slightly more important than the latter, 5 = the former is significantly more important than the latter, 7 = the former is strongly more important than the latter, 9 = the former is extremely more important than the latter, and the inverse is taken for reverse comparison). A 5×5 judgment matrix A is constructed, in the following form:

[0045] ;

[0046] Among them, a ij Let be the importance scale of the i-th indicator relative to the j-th indicator, and satisfy a ij =1 / a ji a ii =1.

[0047] The server calculates the largest eigenvalue λ of the judgment matrix A. max The consistency check is performed using the following formula:

[0048]

[0049]

[0050] In the formula, n is the number of static indicators. In this embodiment, n is 5, and RI = 1.12 (the average random consistency index when n = 5). If CR < 0.1, the matrix is ​​judged to meet the consistency requirements; otherwise, the matrix is ​​readjusted until the consistency conditions are met.

[0051] S213, Static Weights and Subsystem Score Calculation

[0052] The server uses the eigenvector method to find the largest eigenvector of the judgment matrix A, and then normalizes this eigenvector to obtain a static weight vector sorted by indicators.

[0053] ;

[0054] Based on the actual scores xi of the static indicators (i=1~5, corresponding to the scores of the 5 static indicators), the score of the static subsystem is calculated according to the following formula:

[0055]

[0056] In the formula, The value range is [1,4], which is consistent with the scoring range of individual indicators, ensuring a unified quantitative scale.

[0057] S22. Calculation of weights for dynamic subsystems (entropy weight method)

[0058] S221. The data acquisition frequency is 5 minutes / time, and the cumulative number of valid data sets m≥200. In this embodiment, the dynamic indicators selected are: construction and operation disturbance intensity, real-time water level, flow rate and turbidity 1-hour moving average change rate with a 5-minute step.

[0059] Construct an m×3 dynamic data matrix X=(x ij (i=1~m represents the data collection time points, j=1~3 represents the three dynamic indicators), the data is standardized using the following formula (to eliminate differences in units and numerical ranges):

[0060]

[0061] In the formula, min(x) j ), max(x) j ) represent the minimum and maximum values ​​of the j-th dynamic indicator in the m sets of data, respectively, x ′ ij ∈[0,1].

[0062] S222, Information Entropy and Dynamic Weight Calculation

[0063] Calculate the information entropy of the j-th dynamic indicator using the following formula. :

[0064] ;

[0065] in ,for Item No. The proportion of each data point; The natural logarithm of the weight is used to measure the amount of information contained in a single data point. The entropy contribution of a single data point. The value of this item is 0.

[0066] The closer it is to 0, the greater the difference in the data and the richer the information.

[0067] Calculate the dynamic weight vector using the following formula. ,satisfy :

[0068] ;

[0069] S223, Real-time score calculation for dynamic subsystem

[0070] Obtain the standardized values ​​of three dynamic indicators at the real-time acquisition time point t. The real-time score of the dynamic subsystem is calculated using the following formula:

[0071] ;

[0072] In the formula, The value range of is [0,4], which is consistent with the score range of the static subsystem, ensuring that there is no scale bias in the coupled calculation.

[0073] S23, Total Weight Adjustment and Total Impact Index Synthesis

[0074] S231. Setting of composite coefficients during the engineering phase:

[0075] The server determines whether the project is in the construction or operation phase using the phase identification module, and sets the dynamic subsystem weight ratio coefficient B:

[0076] During the construction period: B=0.3, the weight ratio of the static subsystem is 1−B=0.7;

[0077] During operation: B=0.7, the weight ratio of static subsystems is 1−B=0.3.

[0078] S232, Threshold-triggered weight correction:

[0079] The server presets the following dynamic indicator warning thresholds: Rank 1 (hydrological change rate): ≥20%; Rank 2 (water-blocking area ratio): ≥10%; Rank 3 (disturbance intensity): ≥0.8cm / s;

[0080] If the real-time value of any dynamic indicator exceeds the corresponding warning threshold, the weight of that indicator is adjusted according to the following formula, and the sum of the dynamic weights after adjustment is still 1:

[0081]

[0082] In the formula, To adjust the weight of the j-th dynamic indicator, The weights of all dynamic indicators before correction are calculated; the 1.3 in the numerator is 1+0.3=1.3, which means "the weight of dynamic indicators exceeding the threshold is increased by 30% on the original basis", and the 0.3 in the denominator is the quantitative compensation for the "weight increase increment", ensuring that the sum of the weights of all dynamic indicators after correction is still 1 (satisfying the weight normalization requirement).

[0083] S233, Synthesis of Total Impact Index

[0084] The real-time total impact index is synthesized using the following formula:

[0085] ;

[0086] The specific formulas for combining the construction period and the operation period are as follows:

[0087] Construction period: The value range is [0.7, 4].

[0088] Operation period: The value range is [0.3, 4].

[0089] S24, Overall Impact Index Level Mapping

[0090] The server will have a total impact index F total (t) The following interval mapping is used to define four levels of impact, providing a quantitative basis for subsequent dynamic level determination: Essentially no impact: Slight impact: Moderate impact: Serious impact: .

[0091] Furthermore, in S3, dynamic grade determination and trend prediction are performed based on machine learning models, including CatBoost, XGBoost, LightGBM, ExtraTrees, and GRANDE. In this embodiment, the machine learning model is a CatBoost multi-class joint model. Utilizing the CatBoost algorithm's low sample fit, high interpretability, and efficient inference capabilities, accurate mapping of "feature → grade" samples, remedial measure matching, and trend prediction are achieved. The specific steps for training the CatBoost multi-class joint model are as follows:

[0092] S31. Sample set construction:

[0093] Extract a complete sample consisting of "static indicator data + dynamic indicator data + impact level and remedial measures" to ensure that the sample covers typical scenarios throughout the entire construction cycle of a construction project. The specific settings are as follows:

[0094] Input characteristics: Static indicators include 5 items: distance between the project and the station, accuracy level of monitoring data, functional level of the station, type of river-related project, and type of test element; dynamic indicators include 3 items: real-time water level / flow / turbidity 5-minute step 1-hour moving average change rate, water-blocking area ratio of water-related structures, and construction and operation disturbance intensity. Their statistical characteristics are the mean, maximum value, and standard deviation of each indicator.

[0095] Output labels: a combination of two labels (Level 1 label: impact level, divided into four levels: virtually no impact, slight, moderate, and severe; Level 2 label: remedial measures, corresponding one-to-one with the level, based on impact reduction measures in the field of engineering hydrological monitoring).

[0096] Sample splitting: Stratified sampling at an 8:2 ratio (ensuring that the sample ratio of each impact level and corresponding remedial measures is consistent) to obtain a training set (80%) and a test set (20%), with a total sample size of ≥300 sets to meet the model training needs in low-sample scenarios.

[0097] S32, Model Training:

[0098] A CatBoost multi-class joint model is built to achieve synchronous mapping of "input features → impact level + remedial measures". The specific training process is as follows:

[0099] Model parameter configuration: Specify the river-related project type, station function level, monitoring data accuracy level, and test element type as discrete features, take the "total weight of indicators" feature matrix as input, set the core parameters, and enable the built-in ordered lifting and L2 regularization mechanism to suppress overfitting in low-sample scenarios;

[0100] Training objectives: A multi-label classification optimization strategy is adopted, with "accuracy of impact level determination" and "accuracy of remedial measure matching" as dual optimization objectives. Iterative training is carried out until the impact level accuracy on the test set is ≥95% and the remedial measure matching accuracy is ≥93%.

[0101] Model validation: The generalization ability of the model is validated through 5-fold stratified cross-validation to ensure that the consistency between grade determination and measure matching is ≥90% under different engineering scenarios and hydrological conditions, and to avoid model bias caused by single scenario samples.

[0102] After completing the training of the CatBoost multi-class joint model, the trained model is used for real-time data input → simultaneous output of rank and measure:

[0103] Data preprocessing: Real-time collection of static indicator data (one-time input) and dynamic indicator data (collected every 5 minutes, with real-time calculation of statistical characteristics such as mean, maximum, and standard deviation), and data adaptation according to the standardization rules of the training set;

[0104] Model inference: Substitute the preprocessed feature vectors into the trained CatBoost joint model, and the model simultaneously outputs a primary label (impact level, taking the highest probability value) and a secondary label (corresponding remedial measures).

[0105] Result calibration: If dynamic indicators exceed the warning threshold, the model automatically calls the remedial measures optimization logic of similar high-risk cases in history and fine-tunes the output measures (such as adjusting the priority of relocation stations or the rectification time limit in case of severe impact, based on the type of indicators exceeding the threshold).

[0106] Furthermore, based on the statistical characteristic sequence of dynamic indicators over the past 24 hours, a rolling forecast of the impact level change curve and corresponding remedial measures adjustment suggestions are provided for the next 7 days:

[0107] Trend prediction: By using the sliding window method (window size is 12 sets of data, corresponding to 1 hour) to capture the changing trend of dynamic indicators, the rise and fall of the impact level in the next 7 days can be predicted.

[0108] Warning Trigger: When the predicted impact level is upgraded to level ≥1 (e.g., slight → moderate, moderate → severe), a graded warning will be automatically triggered, and a complete plan of "current level + predicted level + current remedial measures + corresponding measures after upgrade" will be pushed to the engineering dispatch terminal simultaneously.

[0109] Updated Measures: If real-time monitoring data shows an actual change in the impact level, the model will automatically re-infer and update the remedial measures, forming a closed-loop linkage of "monitoring, judgment, measures, and early warning".

[0110] By training complete samples and constructing joint models for "characteristic-level measures", the system achieves integrated output of "automatic level determination + precise remedial measure matching + early trend warning" for the impact of hydrological monitoring on construction projects. This completely eliminates manual intervention, solves the pain point of "disconnect between determination and measures" in traditional methods, and is adapted to non-time-series structured data and low-sample scenarios. It balances determination accuracy, logical interpretability, and engineering feasibility, thereby improving the automation and intelligence level of hydrological monitoring impact control for construction projects.

[0111] Furthermore, a closed-loop linkage between the assessment level and impact reduction measures is achieved in S4. Based on real-time assessment levels and trend prediction results, impact reduction measures in the field of hydrological monitoring impact assessment for construction projects are automatically matched (e.g., "basically no impact" corresponds to "no additional measures required," "minor" corresponds to "optimize construction period," "moderate" corresponds to "adjust overall project layout + renovate testing facilities," and "severe" corresponds to "relocate monitoring station"). Measures are pushed out: the assessment results, trend curves, and suggested measures are pushed to the project scheduling terminal via the communication module. When the assessment level changes, the measures are automatically updated, resolving the disconnect between assessment and measures, and enabling real-time project scheduling response.

[0112] This invention also provides a dynamic level determination system for the impact of construction projects on hydrological monitoring, such as... Figure 2 As shown, it includes a data acquisition module, a weight calculation module, a level judgment module, and a measure matching module; among which,

[0113] The data acquisition module is used to construct an evaluation index system that couples static and dynamic indicators. Static indicators include station characteristic indicators and engineering design indicators, while dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators.

[0114] The weight calculation module is used to assign static weights to the static indicators using the analytic hierarchy process, assign dynamic weights to the dynamic indicators using the entropy weight method, and calculate the total weight based on the static weights and the dynamic weights.

[0115] The level judgment module is used to build and train a machine learning model. It inputs the static indicators, the dynamic indicators, and the total weight data into the trained machine learning model and outputs the influence level.

[0116] The measure matching module is used to automatically match response measures based on the impact level and trend prediction results.

[0117] Furthermore, the data acquisition module includes a hydrological data acquisition unit, a station basic data acquisition unit, and a construction dynamic index acquisition unit. The construction dynamic index acquisition unit consists of distributed high-precision fiber optic vibration sensors and real-time video sensors for real-time acquisition. The hardware is installed at the construction site and temporary structures. The software module supports data format conversion. Its function is to collect the construction and operation disturbance intensity every 5 minutes during the construction period and the percentage of water-blocking area of ​​river-related structures once a day. The hydrological data acquisition unit consists of a float-type water level gauge, a Doppler current meter, an optical backscattering turbidimeter, a hydrological early warning system interface, and a 5G / BeiDou short message / fiber optic communication module (automatically switching to BeiDou short message when signal strength < -85dBm). Sensors are installed at the hydrological station's test section and around the water level self-recording station. Its function is to collect dynamic indicators such as the hydrological dynamic change rate every 5 minutes during the operation period. The station basic data acquisition unit consists of an engineering parameter input terminal and a hydrological station management system interface. Its function is to collect the static indicators in S1.

[0118] The weight calculation module includes a standardization unit, a weighting unit, and an outlier removal unit. The standardization unit is used to normalize the dynamic indicator data and eliminate the influence of dimensions. The weighting unit is used to perform the analytic hierarchy process and the entropy weighting method to calculate the dynamic total weight. The outlier removal unit removes abnormal data of dynamic indicators based on the 3σ principle (|x-μ|>3σ is judged as outlier) to ensure data reliability.

[0119] It also includes a report generation module, which generates a level determination report based on the impact level output by the level determination module.

[0120] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for dynamically determining the impact of construction projects on hydrological monitoring, characterized in that, The specific steps include the following: Construct an evaluation index system that couples static and dynamic indicators. Static indicators include station characteristic indicators and engineering design indicators, while dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators. The static indicators are assigned static weights using the analytic hierarchy process, and the dynamic indicators are assigned dynamic weights using the entropy weight method. The total weight is calculated based on the static weights and the dynamic weights. Build and train a machine learning model, input the static indicators, the dynamic indicators and the total weight data into the trained machine learning model, and output the influence level and trend prediction results. Response measures will be automatically matched based on the impact level and trend prediction results.

2. The method for determining the dynamic level of the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, The station characteristic indicators include the station function level, the type of test element, and the accuracy level of monitoring data.

3. The method for determining the dynamic level of the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, The engineering design indicators include the distance between the project and the monitoring station, and the type of river-related project.

4. The method for determining the dynamic level of the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, The construction dynamic indicators include the proportion of water-blocking area of ​​river-related structures and the intensity of construction movement disturbance; these construction dynamic indicators are collected in real time through IoT sensors.

5. The method for determining the dynamic level of the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, The hydrological dynamic indicators include real-time water level, flow rate, and one-hour moving average change rate of turbidity. These are collected at a fixed frequency, and the collected hydrological dynamic indicator data are transmitted to the server via a communication module.

6. The method for dynamically determining the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, The specific steps for training the machine learning model are as follows: Retrieve historical engineering evaluation case databases, extract indicator data and actual grade data, construct a sample set, and divide it into training set and test set according to proportions; A machine learning model is built by inputting a feature matrix of indicators and total weights. The learning rate is adaptively adjusted and an early stopping strategy is used to minimize the log loss function. A map is iteratively generated to improve the decision tree until the accuracy of the validation set reaches a set threshold, thus completing the training of the machine learning model.

7. The method for determining the dynamic level of the impact of construction projects on hydrological monitoring according to claim 1, characterized in that, It also includes transmitting the impact level, trend curve, and response measures to the engineering scheduling terminal via a communication module, and automatically updating the response measures when the impact level changes.

8. A dynamic level determination system for the impact of construction projects on hydrological monitoring, characterized in that, It includes a data acquisition module, a weight calculation module, a level judgment module, and a measure matching module; among which, The data acquisition module is used to construct an evaluation index system that couples static and dynamic indicators. The static indicators include station characteristic indicators and engineering design indicators, while the dynamic indicators include real-time acquired construction dynamic indicators and hydrological dynamic indicators. The weight calculation module is used to assign static weights to the static indicators using the analytic hierarchy process, assign dynamic weights to the dynamic indicators using the entropy weight method, and calculate the total weight based on the static weights and the dynamic weights. The level judgment module is used to construct and train a machine learning model, input the static indicators, the dynamic indicators and the total weight data into the trained machine learning model, and output the influence level and trend prediction results. The measure matching module is used to automatically match response measures based on the impact level and trend prediction results.

9. A dynamic level determination system for the impact of construction projects on hydrological monitoring according to claim 8, characterized in that, It also includes a report generation module, which is used to generate a level determination report based on the impact level output by the level determination module.

Images