Intelligent data analysis system and method applied to stable drainage of road

Abstract

The invention relates to the field of highway infrastructure management, in particular to an intelligent data analysis system and method applied to stable drainage of a highway. The method comprises the steps that soil humidity, rainfall and underground water level data of an environment sensor are acquired, data cleaning, format standardization and preliminary fusion are carried out on edge nodes, and an environment preprocessing data sequence is generated; the data sequence is transmitted to a center node, deep fusion is carried out through a multi-source data fusion algorithm, and environment state parameters are generated; based on the parameters and historical data, constructing a dynamic drainage model by using an adaptive algorithm, and adjusting model parameters in real time to generate an optimal drainage strategy; and inputting the optimal strategy into an execution control module, generating a drainage control instruction, and performing closed-loop optimization on the model according to feedback data. Through multi-source data fusion and self-adaptive model dynamic adjustment, the stability and adaptability of the highway drainage system are improved, different environment and climate conditions are effectively coped with, and the maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of highway infrastructure management, and in particular to a data intelligent analysis system and method applied to stable highway drainage. Background Art

[0002] As an important part of infrastructure, the stability of highway drainage system directly affects the service life and driving safety of highway. However, traditional highway drainage system is usually based on static design parameters and fixed drainage schemes, which are difficult to adapt to changing environmental conditions. In particular, when the rainfall intensity, groundwater level and soil permeability environmental factors change, the static drainage system often cannot respond in time, which can easily lead to poor drainage and water accumulation, thus affecting the stability of highway structure.

[0003] With the development of intelligent data analysis technology, using sensor networks to collect multi-source environmental data in real time and combining intelligent models to dynamically adjust the drainage system has gradually become a feasible solution. Although the existing technology has proposed a method of using multi-source data for drainage management, it still has certain shortcomings in how to achieve efficient integration of real-time environmental data, adaptive adjustment of models and closed-loop optimization. Especially under complex geological and extreme climatic conditions, the robustness and real-time performance of the existing drainage system still need to be improved.

[0004] In order to solve the above problems, the present invention proposes a data intelligent analysis system, which can dynamically adapt to environmental changes and generate optimal drainage strategies through multi-source data fusion and adaptive model adjustment technology, thereby significantly improving the stability and efficiency of the highway drainage system. Summary of the invention

[0005] The present invention provides a data intelligent analysis system and method for stable highway drainage, so as to solve the problem of how to construct and adaptively adjust a highway drainage model in real time based on multi-source environmental data such as soil moisture, rainfall, and groundwater level, so as to cope with the impact of geological and climate changes on the stability of the highway drainage system.

[0006] In order to solve the above technical problems, the present invention provides a data intelligent analysis method applied to stable drainage of highways, comprising: Obtain soil moisture, rainfall, and groundwater level data from environmental sensors, and perform cleaning, format standardization, and preliminary fusion processing at the edge node to obtain an environmental preprocessing data sequence; Transmitting the pre-processed environmental data from the edge node to the central node, deeply fusing the data sequence based on a multi-source data fusion algorithm, and generating environmental state parameters; Based on the environmental state parameters and historical data, a dynamic drainage model is constructed using an adaptive algorithm, and the model parameters are adjusted in real time to obtain an optimal drainage strategy sequence; The optimal drainage strategy sequence is input into the execution control module to generate drainage control instructions, and the drainage model is closed-loop optimized according to the drainage feedback data to update the drainage strategy.

[0007] Furthermore, the step of acquiring environmental sensor data includes: Core data on soil moisture, rainfall and groundwater level are collected in real time through a network of environmental sensors deployed; The collected data is subjected to transmission verification to ensure the integrity of the data and obtain valid preliminary data.

[0008] Furthermore, the step of performing data cleaning at the edge node includes: Performing integrity check on the collected preliminary data and filtering invalid data and abnormal values; The missing data are interpolated and supplemented, and the three sigma rule is applied to detect anomalies in the data to form cleaned data.

[0009] Furthermore, the format standardization and preliminary fusion processing steps include: Standardizing the format of the cleaned data to generate standardized soil moisture, rainfall and groundwater level data series; The standardized data sequence is preliminarily fused based on the weighted average method to obtain an environmental preprocessing data sequence.

[0010] Furthermore, the step of transmitting the pre-processed data from the edge node to the central node includes: Transmitting the environmental preprocessing data sequence to a central node; Data packet verification is performed during the transmission process to ensure the integrity and accuracy of data transmission.

[0011] Furthermore, the step of deeply fusing data based on a multi-source data fusion algorithm includes: Using the set feature weights to perform deep fusion on the preprocessed data sequence to generate a unified environmental state parameter; The environmental state parameters are stored as a unified environmental data set for model construction.

[0012] Furthermore, the step of constructing a dynamic drainage model using an adaptive algorithm includes: Based on the environmental state parameters and historical data, an initial dynamic drainage model is constructed using an adaptive algorithm; The initial water discharge is generated by weighting the historical data.

[0013] Furthermore, the step of adjusting the model parameters in real time includes: The key parameters of the dynamic drainage model are updated according to the real-time environmental status parameters to make the model output adapt to environmental changes.

[0014] Furthermore, the step of generating drainage control instructions and closed-loop optimization includes: The optimal drainage strategy sequence is transmitted to the execution control module to generate control instructions, and the closed-loop optimization adjustment of the model parameters is performed according to the operating status fed back by the drainage equipment.

[0015] Furthermore, a data intelligent analysis system for stable drainage of highways includes: Data acquisition module, used to obtain soil moisture, rainfall and groundwater level data, and perform data cleaning, format standardization and preliminary fusion processing; A data processing and fusion module is used to transmit the pre-processed data sequence to a central node and perform deep fusion of multi-source data to generate environmental state parameters; A dynamic model building and adaptive adjustment module, used to build a dynamic drainage model based on the environmental state parameters and historical data and perform real-time parameter adjustment; The execution control and feedback optimization module is used to convert the optimal drainage strategy sequence into drainage control instructions and perform closed-loop optimization on the drainage model.

[0016] The key innovative features of the present invention include: (1) Multi-source data fusion algorithm realizes efficient fusion of real-time environmental data.

[0017] (2) The adaptive algorithm adjusts the drainage model parameters in real time to generate the optimal drainage strategy.

[0018] (3) The closed-loop feedback mechanism continuously optimizes the drainage strategy based on equipment and environmental data.

[0019] The following are its main beneficial effects: The present invention provides a data intelligent analysis system and method for stable highway drainage. In view of the problem of unstable drainage caused by changes in environmental conditions during highway drainage, multi-source environmental data such as soil moisture, rainfall and groundwater level are obtained in real time, and data cleaning, format standardization and preliminary fusion processing are performed at the edge node, which significantly improves the accuracy and transmission efficiency of the data. Then, the data is deeply fused at the central node to generate unified environmental state parameters, providing a reliable basis for subsequent drainage decisions. Compared with traditional drainage methods, the present invention uses an adaptive algorithm to construct a dynamic drainage model, and makes real-time adjustments based on environmental state parameters and historical data to cope with changes in geology and climate. After the optimal drainage strategy sequence generated by this method is input into the execution module, closed-loop optimization is used to ensure that the drainage strategy maintains efficient response in the dynamic changes of the environment, thereby effectively improving the stability and adaptability of the system and reducing maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a data intelligent analysis method for stable drainage of highways provided in an embodiment of the present application; Figure 2 A structural block diagram of a data intelligent analysis system for stable drainage of highways provided in an embodiment of the present application; DETAILED DESCRIPTION

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first" and "second" in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0022] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0023] Example 1: Reference Figure 1 , is a flow chart of a data intelligent analysis method for stable drainage of highways provided by an embodiment of the present invention, and the flow chart may at least include steps S100-S400: S100, obtaining soil moisture, rainfall and groundwater level data from environmental sensors, and performing cleaning, format standardization and preliminary fusion processing at the edge node to obtain an environmental preprocessing data sequence; S200, transmitting the pre-processed environmental data from the edge node to the central node, performing deep fusion on the data sequence based on a multi-source data fusion algorithm, and generating environmental state parameters; S300, based on the environmental state parameters and historical data, using an adaptive algorithm to construct a dynamic drainage model, and adjusting the model parameters in real time to obtain an optimal drainage strategy sequence; S400: Input the optimal drainage strategy sequence into an execution control module to generate a drainage control instruction, and perform closed-loop optimization processing on the drainage model according to drainage feedback data to update the drainage strategy.

[0024] Step S100 at least includes steps S110-S130: S110, obtaining real-time data of soil moisture, rainfall and groundwater level collected by environmental sensors, and ensuring normal transmission of each data source.

[0025] Specifically, the deployed environmental sensor network collects core data indicators in real time, including soil moisture (set as ), rainfall (set to ) and the groundwater level (set to ), which data is the basic input for adaptive adjustment of the dynamic model.

[0026] To ensure the validity and continuity of the data, firstly, each sensor is checked for normal transmission. The check condition is that the data interval is less than the preset time threshold. , that is, if the data update frequency of any sensor satisfies - ≤ , the transmission is considered normal.

[0027] The preliminary data obtained will be passed to the edge computing node as the input of the subsequent preprocessing module.

[0028] S120: Clean the sensor data at the edge node to remove invalid data and abnormal values.

[0029] The edge node receives the real-time data transmitted , , After that, the data integrity check is first performed. If a data value is missing or abnormal (such as exceeding the normal threshold range set by the sensor), it is marked as invalid data and interpolation or historical mean is used. , , To complete.

[0030] Furthermore, the three sigma rule is used to detect anomalies in the data. Assume that the mean of the data set is and the standard deviation is , then the condition for judging whether the data x is an outlier is:

[0031] The values ​​that meet this condition will be filtered. Reassemble the cleaned data into a continuous data sequence , , to ensure that data quality meets standards and provide a basis for the next step of format standardization.

[0032] S130, perform format standardization and preliminary fusion processing on the cleaned data to generate an environmental preprocessing data sequence as input for subsequent deep fusion.

[0033] Specifically, for the cleaned data sequence , , Perform format standardization. Let the standardization function be f(x), then the standardization formula for each data value x is:

[0034] Where min(X) and max(X) are the minimum and maximum values ​​of sequence X respectively, and the obtained standardized data is , , .

[0035] At the edge node, the standardized data sequence , , Perform preliminary fusion processing, use weighted average method to combine the influence weights of each data source (for example, soil moisture has a greater impact on permeability, so a higher weight is given) to generate an environmental preprocessing data sequence The weighted average formula is:

[0036] in , , are the weight coefficients of each data source, and + + =1.

[0037] Environment preprocessing data sequence It will be used as the input of the next module (S200 multi-source data fusion module) for further deep fusion to ensure the basic data quality for the construction of the dynamic drainage model.

[0038] Step S200 at least includes steps S210-S230: S210 , transmitting the environmental preprocessing data sequence from the edge node to the central node for centralized processing.

[0039] Specifically, the environment preprocessing data sequence generated from S130 The data (including cleaned, formatted and preliminarily fused standardized data of soil moisture, rainfall and groundwater level) are transmitted to the central node through the network.

[0040] During the transmission process, a packet verification mechanism is used to ensure the integrity of the transmission and avoid fusion errors caused by data loss. If a packet is missing or damaged, the central node will automatically request the edge node to resend it until the complete data sequence is complete. Totally accepted.

[0041] After the transmission is completed, the central node will receive The data is stored as raw input sequences, ready to enter the multi-source data fusion processing step.

[0042] S220. Based on a multi-source data fusion algorithm, the received environmental preprocessing data is deeply fused, and the characteristics of each data source are combined to form a unified environmental data set.

[0043] The environmental preprocessing data sequence received by the central node Contains standardized soil moisture at each time , rainfall and groundwater level Furthermore, based on a multi-source data fusion algorithm, the pre-processed data sequence is deeply fused to generate a unified environmental data set.

[0044] To achieve deep fusion, set weight parameters , , Corresponding to the characteristic weights of soil moisture, rainfall and groundwater level respectively, where + + = 1. The state of the environment after fusion The calculation formula is:

[0045] in Represents the unified environmental state parameters at the current time t, which will serve as the basic data for the next step of aggregation analysis.

[0046] Deep fusion data The unified environmental data set containing characteristic information of the environmental state at each moment is then stored and enters further aggregation processing.

[0047] S230, further aggregate and statistically analyze the data sequences in the environmental data set to generate environmental state parameters for use by the model building module.

[0048] After completing the unified environmental data set After generation, it is further aggregated and analyzed to generate key environmental status parameters.

[0049] Specifically, the sliding average algorithm is applied to Smooth the time series to eliminate the impact of short-term fluctuations in burst data. The sliding average formula is as follows:

[0050] in, represents the smoothed environment state at time t, and N represents the length of the sliding window.

[0051] Furthermore, based on the smoothed environment state , statistical analysis is performed on the data to extract key environmental status parameters, including average soil moisture, rainfall intensity, and groundwater level, to provide input data for the subsequent model building module (S300).

[0052] Step S300 at least includes steps S310-S330: S310: Based on the environmental state parameters and system historical data, an initial dynamic drainage model is constructed using an adaptive algorithm.

[0053] Specifically, the environmental state parameters received at the central node and historical data of the system (including drainage conditions and response effects under past environmental conditions) as input to build an initial dynamic drainage model .

[0054] The core formula of the dynamic model is established using the adaptive algorithm. Assume that the model output is the current discharge , and depends on the input variables and historical data weight coefficient The displacement calculation formula is:

[0055] Among them, f is the calculation function of the adaptive algorithm, which maps the effective features in the historical data to the current drainage demand.

[0056] By calculating the formula , complete the initial dynamic model Build so that it can be dynamically adjusted according to the real-time environment status parameters in the subsequent steps.

[0057] S320, adjusting the parameters of the initial model, dynamically updating the model parameters according to the real-time changing environmental state parameters, so that the model is always adapted to the current environment.

[0058] Real-time updated environmental status parameters The data are continuously fed into the model through the acquisition and fusion module (S200) to ensure that the drainage model dynamically adapts to the current environment. along with Changes are constantly adjusting parameters.

[0059] The key parameters of the model (represents the adjustment factor of the model to the current state) is updated based on the new environmental state. The parameter adjustment formula is:

[0060] in, are the updated model parameters, is the model parameter of the previous moment, is the learning rate, which is used to control the rate at which the model is adjusted.

[0061] Furthermore, the updated parameters Reapply to the dynamic drainage model so that the model output Reflect current environmental changes in real time and generate implementable drainage strategies.

[0062] S330. After the model is adjusted, an optimal drainage strategy sequence is generated, which includes drainage adjustment schemes of the system under different conditions for the execution control module to call.

[0063] The optimal drainage strategy sequence for different environmental conditions is generated through the adaptively adjusted drainage model. }, the sequence contains drainage scenarios for a variety of environmental states such as rainfall intensity and groundwater level.

[0064] Specifically, according to the real-time environment status and the adjusted model output , generating the optimal drainage , and construct a drainage strategy sequence. The optimal drainage strategy formula is:

[0065] Here, g is a function used to generate the optimal policy based on the current state of the environment and the adjusted model parameters.

[0066] The generated drainage strategy sequence { } will be passed to the next module to control the execution system to perform actual drainage operations.

[0067] Step S400 at least includes steps S410-S430: S410: Input the optimal drainage strategy sequence into the execution control module, and generate specific drainage control instructions according to the strategy.

[0068] The optimal drainage strategy sequence generated by S330 { }Input to the execution control module, the strategy sequence includes the drainage adjustment amount of the system under different environmental conditions.

[0069] Specifically, the execution control module executes each strategy The drainage demand value in generates specific drainage control instructions The instructions include the discharge volume, equipment startup / shutdown time and the equipment configuration that needs to be executed.

[0070] Furthermore, the generated drainage control instruction Store instruction sequence { , ,… } and pass the sequence to the drainage equipment control module to achieve automatic drainage control.

[0071] S420: Send the control instruction to the drainage equipment, trigger the drainage operation and monitor the execution process.

[0072] The drainage control instruction sequence { , ,… }Sent to the execution module of the drainage equipment via wireless or wired communication to ensure accurate arrival of the instructions.

[0073] Specifically, the device execution module receives the instruction After that, the device is triggered to start or stop operation according to the drainage volume and operation configuration in the instruction to realize the drainage operation under the described environment.

[0074] During the equipment execution process, the monitoring system monitors the equipment operation status in real time and execution status to capture operational anomalies or deviations, providing a basis for closed-loop optimization.

[0075] S430: Based on the operating status and environmental change data fed back by the drainage equipment, the drainage model is optimized in a closed loop, and the drainage strategy is updated and adjusted to improve the adaptability and stability of the system.

[0076] After the drainage operation is completed, according to the operating status feedback from the equipment and real-time environmental change data , further analyze the actual drainage effect and environmental response.

[0077] Use feedback information to adjust the current drainage model parameters Adjustments are made to correct the deviation between the initial forecast and the actual execution. Set the correction factor Represents the model update amplitude, and the correction formula is:

[0078] in, Calculated based on feedback data to reduce the error between actual and predicted.

[0079] After completing the parameter update, the optimal drainage strategy sequence is regenerated based on the revised model { }, and transmits this sequence back to the execution control module to form a closed-loop optimization, ensuring that the system can adapt to environmental changes more accurately in subsequent operations.

[0080] The key innovative features of the present invention include: (1) Multi-source data fusion algorithm realizes efficient fusion of real-time environmental data.

[0081] (2) The adaptive algorithm adjusts the drainage model parameters in real time to generate the optimal drainage strategy.

[0082] (3) The closed-loop feedback mechanism continuously optimizes the drainage strategy based on equipment and environmental data.

[0083] The following are its main beneficial effects: The present invention provides a data intelligent analysis system and method for stable highway drainage. In view of the problem of unstable drainage caused by changes in environmental conditions during highway drainage, multi-source environmental data such as soil moisture, rainfall and groundwater level are obtained in real time, and data cleaning, format standardization and preliminary fusion processing are performed at the edge node, which significantly improves the accuracy and transmission efficiency of the data. Then, the data is deeply fused at the central node to generate unified environmental state parameters, providing a reliable basis for subsequent drainage decisions. Compared with traditional drainage methods, the present invention uses an adaptive algorithm to construct a dynamic drainage model, and makes real-time adjustments based on environmental state parameters and historical data to cope with changes in geology and climate. After the optimal drainage strategy sequence generated by this method is input into the execution module, closed-loop optimization is used to ensure that the drainage strategy maintains efficient response in the dynamic changes of the environment, thereby effectively improving the stability and adaptability of the system and reducing maintenance costs.

[0084] Embodiment 2: Figure 2 1 is a structural block diagram of a data intelligent analysis system for stable drainage of highways provided by an embodiment of the present invention. Figure 2 As shown, the system may include the following modules: The data acquisition module 10 is used to collect real-time environmental data related to highway drainage from a variety of sensors, including core data indicators such as soil moisture, rainfall, and groundwater level. Specifically, data is acquired in real time through the sensor network, and preliminary data verification is performed at the edge node to ensure the continuity and effectiveness of transmission. The collected data will serve as the basic input for subsequent processing and model building, improving the system's accurate perception of environmental conditions.

[0085] The data processing and fusion module 20 is used to pre-process and deeply fuse the collected data. Specifically, the environmental data transmitted by the edge nodes are cleaned and formatted in a standardized manner, outliers are removed, and a unified environmental data set is generated through a multi-source data fusion algorithm. Furthermore, a sliding average algorithm is used to aggregate the data to eliminate the impact of short-term fluctuations and generate smooth environmental state parameters. The processed data will provide a stable input for the dynamic drainage model.

[0086] The dynamic model building and adaptive adjustment module 30 is used to build an initial dynamic drainage model based on environmental state parameters and system historical data, and to adaptively adjust the model parameters. Specifically, the adaptive algorithm is used to dynamically update the model parameters according to the current environmental state parameters to generate a drainage strategy that adapts to the real-time environment. This module dynamically adjusts the parameters to adapt to the rapid changes in rainfall and groundwater level environment, thereby maintaining the best effect of the drainage strategy.

[0087] The execution control and feedback optimization module 40 is used to convert the generated optimal drainage strategy into specific drainage control instructions and send them to the drainage equipment. Specifically, after receiving the optimal drainage strategy, the control module generates control instructions, including the start and stop operations of the equipment and the drainage volume configuration. During the execution process, the monitoring system records the operating status and environmental feedback of the equipment, returns the actual drainage effect to the model, and performs closed-loop optimization processing to improve the accuracy and response sensitivity of the model.

[0088] Through the above system structure and functional design, the present invention provides an efficient highway drainage management solution. Through multi-source data fusion and adaptive adjustment of dynamic models, the system realizes intelligent optimization of drainage strategies, adapts to different geological conditions and extreme weather conditions, improves the stability and reliability of highway drainage systems, and effectively reduces maintenance costs.

[0089] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing the embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to effectively replace some of the technical features therein. Any effective structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is also within the scope of patent protection of this application.

Claims

1. A data intelligent analysis method for stable drainage of highways, characterized in that: include: Obtain soil moisture, rainfall, and groundwater level data from environmental sensors, and perform cleaning, format standardization, and preliminary fusion processing at the edge node to obtain an environmental preprocessing data sequence; Transmitting the pre-processed environmental data from the edge node to the central node, deeply fusing the data sequence based on a multi-source data fusion algorithm, and generating environmental state parameters; Based on the environmental state parameters and historical data, a dynamic drainage model is constructed using an adaptive algorithm, and the model parameters are adjusted in real time to obtain an optimal drainage strategy sequence; The optimal drainage strategy sequence is input into the execution control module to generate drainage control instructions, and the drainage model is closed-loop optimized according to the drainage feedback data to update the drainage strategy.

2. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of acquiring environmental sensor data comprises: Core data on soil moisture, rainfall and groundwater level are collected in real time through a network of environmental sensors deployed; The collected data is subjected to transmission verification to ensure the integrity of the data and obtain valid preliminary data.

3. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of performing data cleaning at the edge node includes: Performing integrity check on the collected preliminary data and filtering invalid data and abnormal values; The missing data are interpolated and supplemented, and the three sigma rule is applied to detect anomalies in the data to form cleaned data.

4. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The format standardization and preliminary fusion processing steps include: Standardizing the format of the cleaned data to generate standardized soil moisture, rainfall and groundwater level data series; The standardized data sequence is preliminarily fused based on the weighted average method to obtain an environmental preprocessing data sequence.

5. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of transmitting the pre-processed data from the edge node to the central node comprises: Transmitting the environmental preprocessing data sequence to a central node; Data packet verification is performed during the transmission process to ensure the integrity and accuracy of data transmission.

6. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of deeply fusing data based on a multi-source data fusion algorithm comprises: Using the set feature weights to perform deep fusion on the preprocessed data sequence to generate a unified environmental state parameter; The environmental state parameters are stored as a unified environmental data set for model construction.

7. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of constructing a dynamic drainage model using an adaptive algorithm comprises: Based on the environmental state parameters and historical data, an initial dynamic drainage model is constructed using an adaptive algorithm; The initial water discharge is generated by weighting the historical data.

8. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The step of adjusting the model parameters in real time comprises: The key parameters of the dynamic drainage model are updated according to the real-time environmental status parameters to make the model output adapt to environmental changes.

9. The data intelligent analysis method for stable drainage of highways according to claim 1 is characterized in that: The steps of generating drainage control instructions and closed-loop optimization include: The optimal drainage strategy sequence is transmitted to the execution control module to generate control instructions, and the closed-loop optimization adjustment of the model parameters is performed according to the operating status fed back by the drainage equipment.

10. A data intelligent analysis system for stable drainage of highways, characterized in that: include: Data acquisition module, used to obtain soil moisture, rainfall and groundwater level data, and perform data cleaning, format standardization and preliminary fusion processing; A data processing and fusion module is used to transmit the pre-processed data sequence to a central node and perform deep fusion of multi-source data to generate environmental state parameters; A dynamic model building and adaptive adjustment module, used to build a dynamic drainage model based on the environmental state parameters and historical data and perform real-time parameter adjustment; The execution control and feedback optimization module is used to convert the optimal drainage strategy sequence into drainage control instructions and perform closed-loop optimization on the drainage model.

Images