A tunnel temperature and humidity dynamic prediction and early warning method based on a multi-domain coupling monitoring network

Abstract

The present application belongs to the field of tunnel construction safety and intelligent monitoring and early warning, and discloses a tunnel temperature and humidity dynamic prediction and early warning method based on a multi-domain coupling monitoring network, which solves the problems that the existing tunnel environment monitoring technology cannot realize water-rock-gas-heat multi-domain monitoring point space correlation, lacks unified coupling state quantitative indicators, can only realize real-time threshold alarm and cannot predict future temperature and humidity and identify risk causes. The present application first lays out multiple types of monitoring nodes in the tunnel to collect water-rock-gas-heat multi-domain monitoring data; constructs a multi-domain coupling monitoring network through node space coding and registration distance determination; then calculates four types of sub-item indexes and fuses to obtain a coupling state index; then inputs the coupling index, historical temperature and humidity and construction control information into a prediction model to predict future temperature and humidity and output a graded early warning; finally determines the main cause of the risk based on the contribution rate of the sub-item index and matches the corresponding control suggestion.

Description

Technical Field

[0001] This invention belongs to the field of tunnel construction safety and intelligent monitoring and early warning, specifically involving a method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network. Background Technology

[0002] As underground engineering projects such as railway tunnels, highway tunnels, hydraulic tunnels, mine roadways, underground powerhouses, and underground transportation corridors gradually develop towards deeper burial, longer lengths, and more complex geological conditions, the risks associated with tunnel construction, including high-temperature surrounding rock, high-temperature hot water, high-humidity air, limited ventilation, and high-temperature and high-humidity operations, are becoming increasingly prominent. In high-temperature or deeply buried tunnels, heat from the deep surrounding rock continuously releases into the tunnel, while seepage water from fissures or high-temperature hot water constantly replenishes moisture and heat. The air inside the tunnel undergoes heat and water vapor transport under ventilation, causing significant spatiotemporal dynamic changes in tunnel temperature and humidity. Therefore, changes in tunnel temperature and humidity are not simply variations in air parameters, but rather a water-rock-air-heat coupled evolution process formed by the combined effects of seepage conditions, surrounding rock temperature, air temperature and humidity, and heat exchange conditions. It is necessary to conduct monitoring, prediction, and early warning systems for this water-rock-air-heat coupled process.

[0003] Existing tunnel environmental monitoring methods only deploy monitoring sections along the tunnel axis, separately arranging sensors for surrounding rock temperature, tunnel air temperature and humidity, and wind speed. This approach can only collect real-time thermal environment parameters for local areas and has significant limitations: First, monitoring data is stored separately according to sensor type and construction section, without establishing spatial registration relationships between seepage, surrounding rock, air, and heat exchange monitoring points. This makes it impossible to construct a multi-domain coupled monitoring chain and trace the spatial source and transmission path of thermal and humidity anomalies. Second, threshold alarms rely solely on single indicators such as air temperature and humidity, wall temperature, and wind speed, without constructing unified quantitative indicators to characterize the water-rock-air-heat coupling evolution state. This makes it impossible to transform multi-source monitoring data into effective features for temperature and humidity prediction. Third, the early warning mode is only a real-time parameter over-limit alarm, which can only reflect the current environmental risk and cannot predict future temperature and humidity trends. Furthermore, it cannot distinguish the core causes of high temperature, high humidity, and combined risks, making it difficult to provide targeted guidance for on-site ventilation, drainage, and cooling control operations. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network, which solves the problems of existing tunnel environmental monitoring technology that cannot achieve spatial correlation of multi-domain measuring points of water-rock-gas-heat, lack of unified coupling state quantitative indicators, can only provide real-time threshold alarms and cannot predict future temperature and humidity and identify risk causes.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0006] A method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network includes the following steps:

[0007] S1. Deploy multiple monitoring nodes inside the tunnel to collect multi-domain monitoring data on water, rock, gas, and heat in the tunnel;

[0008] S2. Unify the spatial coding of all monitoring nodes and complete the time synchronization of monitoring data. Calculate the spatial registration distance based on the node spatial coding. Construct a coupled monitoring chain based on the registration distance threshold to form a multi-domain coupled monitoring network.

[0009] S3. Based on the multi-domain monitoring data in the multi-domain coupled monitoring network, the seepage state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index are calculated respectively, and then fused to obtain the multi-domain coupled state index.

[0010] S4. Input the multi-domain coupling state index, historical temperature and humidity data, and construction control information into the pre-established temperature and humidity prediction model to predict the future air temperature and humidity of the tunnel, calculate the risk index based on the predicted temperature and humidity, and output the early warning result.

[0011] S5. Calculate the contribution rates of the permeability state index, surrounding rock thermal reservoir index, air thermal and moisture transport index, and heat exchange index to the multi-domain coupled state index, identify the main causes of risk based on the contribution rates, and output corresponding auxiliary decision-making suggestions.

[0012] Furthermore, in step S1, multiple monitoring nodes are deployed inside the tunnel to collect multi-domain monitoring data on water, rock, gas, and heat within the tunnel, specifically including:

[0013] Multiple monitoring sections are arranged along the tunnel axis and radial direction. Seepage status nodes, surrounding rock temperature nodes, air status nodes and heat exchange nodes are set up at each monitoring section.

[0014] The seepage temperature and flow rate are collected through the seepage status node; the surface temperature and deep temperature of the surrounding rock are collected through the surrounding rock temperature node; the air temperature, relative humidity and air velocity are collected through the air status node; and the convective heat transfer, evaporative latent heat, condensation latent heat and ventilation heat transfer are collected through the heat exchange node.

[0015] Furthermore, in step S1, the collected tunnel water-rock-gas-heat multi-domain monitoring data are represented as follows:

[0016] ;

[0017] in, For a moment Multi-domain monitoring data; For a moment Seepage status data; For a moment Surrounding rock condition data; For a moment Air condition data; For a moment Heat exchange status data;

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] in, The seepage temperature; This refers to the seepage flow rate; These are parameters related to the wetted state. Temperature at the depth of the surrounding rock. The temperature of the surrounding rock surface; This represents the change in temperature of the surrounding rock. Air temperature; Relative humidity of the air; Air velocity; For convective heat transfer; It is the latent heat of vaporization; The latent heat of condensation; For ventilation and heat exchange.

[0023] Furthermore, in step S2, a unified spatial encoding is applied to all monitoring nodes, and monitoring data time synchronization is completed. This specifically includes:

[0024] Each monitoring node is assigned a unique spatial code:

[0025] ;

[0026] in, For the first Spatial coding of each monitoring node; The tunnel mileage; This indicates the orientation of the cross-section. Radial depth; Height from the bottom plate; The node types include seepage nodes, surrounding rock nodes, air nodes, heat exchange nodes, and construction control nodes;

[0027] The spatial code, monitoring value, and data quality status of the monitoring node are combined to form the node status:

[0028] ;

[0029] in, For the first Each monitoring node at time The node state; For node monitoring values; This represents the data quality status, indicating whether the data is normal, missing, abnormal, drifted, or offline.

[0030] Monitoring data with different sampling frequencies are synchronized to the same time reference.

[0031] Furthermore, in step S2, the spatial registration distance is calculated based on the node spatial encoding, and a coupled monitoring chain is constructed according to the registration distance threshold to form a multi-domain coupled monitoring network, specifically including:

[0032] Calculate the spatial registration distance between nodes based on the node spatial encoding:

[0033] ;

[0034] in, For the first The monitoring node and the first Spatial registration distance of each monitoring node; , The first The monitoring node and the first Spatial coding of each monitoring node;

[0035] Determine whether two nodes have established a spatial association based on their spatial registration distance and node type:

[0036] ;

[0037] in, For the first The monitoring node and the first Spatial registration relationship of monitoring nodes; when Less than or equal to the preset spatial registration threshold, and When the combination of monitoring node types that are allowed to be associated is used, the node is determined. and nodes Establish spatial associations; otherwise, do not establish spatial associations.

[0038] The seepage nodes, surrounding rock nodes, air nodes, and heat exchange nodes that have completed spatial correlation are grouped into the same local influence unit, and a coupled monitoring chain is constructed, represented as:

[0039] ;

[0040] in, For the first A coupled monitoring chain; This is a seepage point; For surrounding rock nodes; An air node; It serves as a heat exchange node; As a construction control point;

[0041] The coupled monitoring chain is organized according to tunnel mileage, cross-sectional location, node correlation, and construction section to form a coupled monitoring network, which is used to characterize the thermal and moisture influence transmission relationship between different local influence units:

[0042] ;

[0043] in, For a multi-domain coupled monitoring network; for the first A coupled monitoring chain, This represents the total number of coupled monitoring chains.

[0044] Furthermore, in step S3, the calculation methods for the permeability state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index include:

[0045] ;

[0046] ;

[0047] ;

[0048] ;

[0049] in, It is a seepage state index, which characterizes the degree of influence of seepage temperature, seepage flow rate and humidity state on the thermal and humid environment of the tunnel. The surrounding rock thermal reservoir index represents the ability of the surrounding rock to release heat into the cavern; The air heat and moisture transport index characterizes the air's ability to transport heat and moisture. The heat exchange index characterizes the intensity of heat exchange between the surrounding rock, seepage water, and air.

[0050] This is the normalized value for the seepage temperature; This represents the normalized value of the seepage flow rate. This is the normalized value for the wet state;

[0051] This is the normalized value of the surrounding rock surface temperature; This represents the normalized value of the deep temperature of the surrounding rock. This is the normalized value of the change in surrounding rock temperature.

[0052] This represents the normalized value of convective heat transfer. This is the normalized value of the latent heat of vaporization; This is the normalized value of the latent heat of condensation; This is the normalized value for ventilation heat exchange. , , , , , , , , , These are the weighting coefficients.

[0053] Furthermore, in step S3, the multi-domain coupling state index is calculated as follows:

[0054] ;

[0055] in, It is a multi-domain coupling state index; , , , These are the weighting coefficients.

[0056] Furthermore, in step S4, the future air temperature and humidity in the tunnel are expressed as follows:

[0057] ;

[0058] in, This is a forecast of future temperature and humidity. The predicted air temperature for a future time. This is a forecast of relative humidity at a future time. For the predicted duration;

[0059] The risk index is calculated based on predicted temperature and humidity, and early warning results are output, specifically including:

[0060] Calculate the risk index:

[0061] ;

[0062] in, Temperature and humidity risk index; This is the normalized predicted future air temperature value; This is the normalized forecast value for future relative humidity. Temperature weighting coefficient; Humidity weighting coefficient;

[0063] The risk level is determined based on the magnitude of the risk index, and a risk warning result is output:

[0064] ;

[0065] in, This is a risk warning result, including the risk level. Risk location Time of risk occurrence and risk type The risk levels are divided into four levels: no risk, general risk, medium risk, and high risk; the risk types include high temperature risk, high humidity risk, and combined high temperature and high humidity risk.

[0066] Furthermore, in step S5, the contribution rates of the permeability state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index to the multi-domain coupling state index are expressed as follows:

[0067] ;

[0068] in, For the first The contribution rate of each state index to the multi-domain coupled state index. These correspond to the permeability state index, the surrounding rock thermal reservoir index, the air heat and moisture transport index, and the heat exchange index, respectively. Indicates the first State index;

[0069] The risk cause identification results are expressed as follows:

[0070] ;

[0071] in, The risk causative factor corresponding to the state index with the largest contribution rate; Contribution rate to the seepage state; Contribution rate to the surrounding rock thermal reservoir; Contribution rate to air heat and moisture transport; Contribution rate to heat exchange.

[0072] Furthermore, in step S5, the auxiliary decision-making suggestions include: ventilation optimization, cooling, spray adjustment, drainage dehumidification, and personnel operation scheduling.

[0073] The beneficial effects of this invention are:

[0074] (1) This invention collects monitoring nodes with heat and moisture interaction relationships of seepage, surrounding rock, air and heat exchange by unifying spatial coding and spatial registration distance judgment rules, constructs a coupled monitoring chain and integrates them into a multi-domain coupled monitoring network, and incorporates the entire process of seepage heat and moisture supply, surrounding rock geothermal release, air heat and moisture transport and inter-field heat exchange into the same spatial analysis system. It can clearly locate the heat and moisture source and track the diffusion path of heat and water vapor in the tunnel space, fully characterize the water-rock-air-heat coupling evolution law, and improve the ability of monitoring data to characterize and interpret the mechanism of heat and moisture environment change in the tunnel.

[0075] (2) This invention constructs four sub-indices in different dimensions: seepage state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index, and obtains a multi-domain coupling state index by weighted fusion. Each sub-index independently quantifies the driving intensity of a single physical field on the temperature and humidity inside the cave. The coupling index uniformly characterizes the intensity of heat and moisture coupling in a local area, solving the problem of inconsistent dimensions of heterogeneous monitoring parameters and difficulty in fusion modeling. It provides complete and standardized input features for temperature and humidity prediction models, effectively reduces prediction errors, and improves the accuracy of predicting future temperature and humidity trends.

[0076] (3) This invention uses the multi-domain coupling state index, historical temperature and humidity time series data, and construction control information as input to the prediction model, and outputs the prediction results of tunnel air temperature and humidity for multiple time periods in the future; based on the predicted value, a standardized risk index is calculated, and four levels of risk are automatically divided and three types of risk are distinguished: high temperature, high humidity, and high temperature and high humidity composite. The complete early warning information including the risk location, the expected occurrence time, and the risk level is output simultaneously, which facilitates the early implementation of ventilation, drainage, cooling and other pre-control operations on site, and actively avoids the safety hazards of high temperature and high humidity operations.

[0077] (4) This invention calculates the contribution rate of each physical field by weighted sub-index, quantitatively determines the four main causes of risk: water seepage, surrounding rock heat release, insufficient ventilation, and abnormal heat exchange, and realizes digital diagnosis of risk root causes; automatically matches different treatment plans such as ventilation optimization, cooling, drainage and dehumidification, spray adjustment, and personnel operation scheduling for different main causes, and directly outputs feasible construction decision suggestions. Attached Figure Description

[0078] Figure 1 This is a flowchart of the tunnel temperature and humidity dynamic prediction and early warning method based on a multi-domain coupled monitoring network in an embodiment of the present invention.

[0079] Figure 2 This is a schematic diagram of an exemplary water-rock-air-heat multi-domain monitoring setup; in the diagram, 1 is a flow meter; 2-5 are water temperature meters; 6-8 are air condition (temperature, humidity, wind speed) meters; 9-11 are heat exchange condition meters; and 12-26 are surrounding rock temperature meters.

[0080] Figure 3 This is a schematic diagram illustrating dynamic prediction and risk warning of temperature and humidity in tunnels.

[0081] Figure 4 This diagram illustrates risk causation identification and decision support. Detailed Implementation

[0082] This invention aims to provide a dynamic prediction and early warning method for tunnel temperature and humidity based on a multi-domain coupled monitoring network. It addresses the problems of existing tunnel environmental monitoring technologies, such as the inability to achieve spatial correlation of multi-domain measuring points (water, rock, gas, and heat), the lack of unified quantitative indicators for coupled states, and the inability to provide real-time threshold alarms while failing to predict future temperature and humidity or identify risk causes. Its core idea is:

[0083] First, four types of monitoring nodes are deployed inside the tunnel to collect multi-domain monitoring data on water, rock, gas, and heat. A multi-domain coupled monitoring network is constructed based on unified spatial coding and spatial registration rules to establish spatial interaction relationships among the monitoring points of each physical field. Then, based on the monitoring data collected by the network, a hierarchical and sub-item state index is constructed and fused to obtain a unified multi-domain coupled state index, which quantitatively characterizes the intensity of local thermal and humidity coupling evolution. Subsequently, the coupled state index, historical temperature and humidity, and construction control information are input into a pre-constructed prediction model to predict future temperature and humidity changes inside the tunnel and output risk warnings in a tiered manner. Finally, the main causes of risk are quantitatively identified by the contribution rate of the sub-item index, and targeted construction control suggestions are output accordingly.

[0084] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0085] This embodiment provides a method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network. (See also...) Figure 1 It includes the following implementation process:

[0086] S1. Multi-domain monitoring of water, rock, gas, and heat:

[0087] In this step, multiple monitoring nodes are deployed within the tunnel to collect multi-domain monitoring data on water, rock, air, and heat. The aim is to establish a fundamental monitoring system for predicting tunnel temperature and humidity, and to obtain key monitoring parameters affecting changes in air temperature and relative humidity. Simultaneously, the system records construction control information such as ventilation, cooling, drainage, dehumidification, sprinkler systems, machinery operations, and personnel scheduling, which can be incorporated into subsequent coupled monitoring chains as construction control nodes.

[0088] In one exemplary implementation, firstly, multiple monitoring sections are deployed radially and axially along the tunnel, with denser deployment in high-temperature surrounding rock areas, high-temperature hot water areas, areas with strong seepage, poorly ventilated areas, near the tunnel face, and high-temperature and high-humidity risk areas. The monitoring diagram is shown below. Figure 2As shown in the figure, seepage status nodes, surrounding rock temperature nodes, air status nodes, and heat exchange nodes are set up in each monitoring section. Among them, the seepage status nodes are used to obtain seepage temperature and seepage flow rate; the surrounding rock temperature nodes are used to obtain the surface temperature and deep temperature of the surrounding rock; the air status nodes are used to obtain air temperature, relative humidity, and wind speed; and the heat exchange nodes are used to obtain or estimate the convective heat transfer, latent heat of vaporization, latent heat of condensation, and ventilation heat transfer status.

[0089] The collected multi-domain monitoring data of water, rock, gas, and heat in the tunnel are represented as follows:

[0090] ;

[0091] in, For a moment Multi-domain monitoring data; For a moment Seepage status data; For a moment Surrounding rock condition data; For a moment Air condition data; For a moment Heat exchange status data;

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] in, The seepage temperature; This refers to the seepage flow rate; These are parameters related to the wetted state. Temperature at the depth of the surrounding rock. The temperature of the surrounding rock surface; This represents the change in temperature of the surrounding rock. Air temperature; Relative humidity of the air; Air velocity; For convective heat transfer; It is the latent heat of vaporization; The latent heat of condensation; For ventilation and heat exchange.

[0097] S2. Spatial registration and coupling monitoring chain construction:

[0098] In this step, all monitoring nodes are uniformly spatially coded and their monitoring data is synchronized in time. Based on the node spatial codes, spatial registration distances are calculated, and coupled monitoring chains are constructed according to the registration distance thresholds to form a multi-domain coupled monitoring network. The purpose is to unify the dispersed seepage nodes, surrounding rock nodes, air nodes, and heat exchange nodes into the same tunnel spatial system, forming a coupled monitoring chain that characterizes the "seepage-surrounding rock-air-heat exchange" process.

[0099] In one exemplary implementation, this step is performed as follows:

[0100] First, a unique spatial code is assigned to each monitoring node:

[0101] ;

[0102] in, For the first Spatial coding of each monitoring node; The tunnel mileage; This indicates the orientation of the cross-section. Radial depth; Height from the bottom plate; The node types include seepage nodes, surrounding rock nodes, air nodes, heat exchange nodes, and construction control nodes;

[0103] Next, the spatial code, monitoring value, and data quality status of the monitoring node are combined to form the node status:

[0104] ;

[0105] in, For the first Each monitoring node at time The node state; For node monitoring values; This represents the data quality status, indicating whether the data is normal, missing, abnormal, drifted, or offline.

[0106] Then, the monitoring data with different sampling frequencies are time-synchronized to the same time reference.

[0107] Next, the spatial registration distance between nodes is calculated based on the node spatial encoding:

[0108] ;

[0109] in, For the first The monitoring node and the first Spatial registration distance of each monitoring node; , The first The monitoring node and the first Spatial coding of each monitoring node;

[0110] Subsequently, based on the spatial registration distance and node type, it is determined whether a spatial association has been established between the two nodes:

[0111] ;

[0112] in, For the first The monitoring node and the first Spatial registration relationship of monitoring nodes; when Less than or equal to the preset spatial registration threshold, and When the combination of monitoring node types that are allowed to be associated is used, the node is determined. and nodes Establish spatial associations; otherwise, do not establish spatial associations.

[0113] Next, the seepage nodes, surrounding rock nodes, air nodes, and heat exchange nodes that have completed spatial correlation are grouped into the same local influence unit, and a coupled monitoring chain is constructed, represented as:

[0114] ;

[0115] in, For the first A coupled monitoring chain; This is a seepage point; For surrounding rock nodes; An air node; It serves as a heat exchange node; As a construction control point;

[0116] Finally, the coupled monitoring chain is organized according to tunnel mileage, cross-sectional location, node correlation, and construction section to form a coupled monitoring network, which is used to characterize the thermal and moisture influence transmission relationship between different local influence units:

[0117] ;

[0118] in, For a multi-domain coupled monitoring network; for the first A coupled monitoring chain, This represents the total number of coupled monitoring chains.

[0119] S3. Calculation of the water-rock-gas-thermal coupling state index:

[0120] In this step, based on multi-domain monitoring data within the multi-domain coupled monitoring network, the seepage state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index are calculated separately and then fused to obtain the multi-domain coupled state index. The purpose is to transform the seepage state, surrounding rock thermal reservoir, air heat and moisture transport, and heat exchange state into a unified water-rock-gas-heat coupled state index, comprehensively characterizing the comprehensive intensity of the multi-field coupled evolution of water-rock-gas-heat in a local area. This provides standardized and interpretable input features for subsequent temperature and humidity prediction models, and also provides a basic quantitative basis for the subsequent quantitative identification of risk causes.

[0121] In one exemplary implementation, this step is performed as follows:

[0122] To eliminate the influence of dimensional differences in monitoring parameters on the calculation of state indices, the relevant monitoring parameters are standardized or dimensionless before calculating the permeability state index, surrounding rock reservoir index, air heat and moisture transport index, and heat exchange index. Dimensionlessness can be achieved through extreme value normalization, standard deviation normalization, or other data standardization methods. The dimensionless expressions for the permeability state index, surrounding rock reservoir index, air heat and moisture transport index, and heat exchange index are as follows:

[0123] ;

[0124] ;

[0125] ;

[0126] ;

[0127] in, It is a seepage state index, which characterizes the degree of influence of seepage temperature, seepage flow rate and humidity state on the thermal and humid environment of the tunnel. The surrounding rock thermal reservoir index represents the ability of the surrounding rock to release heat into the cavern; The air heat and moisture transport index characterizes the air's ability to transport heat and moisture. The heat exchange index characterizes the intensity of heat exchange between the surrounding rock, seepage water, and air.

[0128] This is the normalized value for the seepage temperature; This represents the normalized value of the seepage flow rate. This is the normalized value for the wet state;

[0129] This is the normalized value of the surrounding rock surface temperature; This represents the normalized value of the deep temperature of the surrounding rock. This is the normalized value of the change in surrounding rock temperature.

[0130] This represents the normalized value of convective heat transfer. This is the normalized value of the latent heat of vaporization; This is the normalized value of the latent heat of condensation; This is the normalized value for ventilation heat exchange. , , , , , , , , , These are the weighting coefficients.

[0131] Finally, the multi-domain coupling state index is calculated as follows:

[0132] ;

[0133] in, It is a multi-domain coupling state index; , , , These are the weighting coefficients.

[0134] S4. Dynamic Prediction and Risk Warning of Tunnel Temperature and Humidity:

[0135] In this step, the multi-domain coupling state index, historical temperature and humidity data, and construction control information are input into a pre-established temperature and humidity prediction model to predict the future air temperature and humidity in the tunnel. Based on the predicted temperature and humidity, a risk index is calculated and an early warning result is output. The purpose is to utilize the quantitative characteristics of multi-physics coupling effects combined with historical time-series data and on-site human control conditions to extrapolate the future trend of tunnel air temperature and humidity changes, enabling early prediction of high temperature and high humidity hazards; to quantify the degree of environmental hazard through a standardized risk index, classify risk levels and types, and clarify the risk location and occurrence time, outputting complete early warning information, providing a preliminary basis for on-site ventilation, cooling, drainage, and other control operations, and achieving proactive prevention and control of high temperature and high humidity risks.

[0136] In one exemplary implementation, the process of this step is described in [reference needed]. Figure 3 The details are as follows:

[0137] First, using the multi-domain coupling state index, historical temperature and humidity data, and construction control information as input features, and inputting them into a pre-established temperature and humidity prediction model, the predicted future air temperature and humidity of the tunnel is expressed as follows:

[0138] ;

[0139] in, This is a forecast of future temperature and humidity. The predicted air temperature for a future time. This is a forecast of relative humidity at a future time. For the predicted duration.

[0140] Temperature and humidity prediction models can employ machine learning models, artificial intelligence models, physical constraint models, or combinations thereof.

[0141] Next, a risk index is constructed based on the predicted temperature and humidity results (the predicted temperature and humidity can be normalized or standardized before being included in the risk index calculation):

[0142] ;

[0143] in, Temperature and humidity risk index; This is the normalized predicted future air temperature value; This is the normalized forecast value for future relative humidity. Temperature weighting coefficient; Humidity weighting coefficient;

[0144] Finally, the risk level is determined based on the risk index, and a risk warning result is output:

[0145] ;

[0146] in, This is a risk warning result, including the risk level. Risk location Time of risk occurrence and risk type The risk levels are divided into four levels: no risk, general risk, medium risk, and high risk; the risk types include high temperature risk, high humidity risk, and combined high temperature and high humidity risk.

[0147] S5. Risk Cause Identification and Decision Support:

[0148] In this step, the contribution rates of the seepage state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index to the multi-domain coupling state index are calculated respectively. Based on the contribution rates, the main causes of risk are identified, and corresponding auxiliary decision-making suggestions are output. The purpose is to accurately locate the core driving factors causing excessive temperature and humidity in tunnels by quantifying the contribution ratio of each physical field sub-index to the overall coupling state, and to achieve digital differentiation of risk root causes; and to match specific engineering treatment plans for different main causes, forming a complete closed loop from risk warning to cause diagnosis to on-site construction control.

[0149] In one exemplary implementation, the process of this step is described in [reference needed]. Figure 4 The details are as follows:

[0150] The contribution rate is calculated based on the ratio of each state index (permeability state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index) to the multi-domain coupled state index:

[0151] ;

[0152] in, For the first The contribution rate of each state index to the multi-domain coupled state index. These correspond to the permeability state index, the surrounding rock thermal reservoir index, the air heat and moisture transport index, and the heat exchange index, respectively. Indicates the first State index;

[0153] The risk cause identification results are expressed as follows:

[0154] ;

[0155] in, The risk causative factor corresponding to the state index with the largest contribution rate; Contribution rate to the seepage state; Contribution rate to the surrounding rock thermal reservoir; Contribution rate to air heat and moisture transport; Contribution rate to heat exchange.

[0156] The auxiliary decision is set in advance according to different risk causes, so as to achieve the matching of different risk causes with corresponding auxiliary decisions. For example: (1) For the seepage-dominated risk, the core cause is that the high temperature seepage continuously replenishes heat and water vapor into the tunnel, and the humidity and local temperature rise. The corresponding auxiliary decision can be: prioritize drainage and dehumidification, increase the drainage of seepage points and seal the cracks; adjust the parameters of the spray system at the same time to reduce the additional water vapor evaporation in the tunnel. (2) For the surrounding rock heat release-dominated risk, the core cause is that the high temperature surrounding rock continuously releases geothermal heat, forming a basic high temperature environment. The corresponding auxiliary decision can be: turn on the local cooling equipment and set up the cooling device at fixed points in the high temperature surrounding rock section; shorten the continuous working time of underground personnel and increase the rotation frequency. (3) For the ventilation-insufficient risk, the core cause is that the ventilation volume in the tunnel is insufficient and the airflow organization is poor, so that heat and water vapor cannot be discharged in time. The supporting auxiliary decision-making can be: to carry out ventilation optimization, increase the operating air volume of the fan, adjust the layout of the ventilation pipeline, and clear the airflow channels in the working face and high-temperature section. (4) For the risk dominated by abnormal heat exchange, the core cause is the imbalance of evaporation, condensation and convection heat exchange in the tunnel, and the inability to diffuse the accumulated heat and moisture. The supporting auxiliary decision-making can be: to adjust the opening of the sprinkler system and control the release of latent heat of evaporation in the tunnel; and to improve the heat exchange conditions between the sites in conjunction with ventilation regulation.

[0157] Although embodiments of the present invention have been described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, and all such changes and alterations shall not depart from the protection scope of the present invention.

Claims

1. A method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network, characterized in that, Includes the following steps: S1. Deploy multiple monitoring nodes inside the tunnel to collect multi-domain monitoring data on water, rock, gas, and heat in the tunnel; S2. Unify the spatial coding of all monitoring nodes and complete the time synchronization of monitoring data. Calculate the spatial registration distance based on the node spatial coding. Construct a coupled monitoring chain based on the registration distance threshold to form a multi-domain coupled monitoring network. S3. Based on the multi-domain monitoring data in the multi-domain coupled monitoring network, the seepage state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index are calculated respectively, and then fused to obtain the multi-domain coupled state index. S4. Input the multi-domain coupling state index, historical temperature and humidity data, and construction control information into the pre-established temperature and humidity prediction model to predict the future air temperature and humidity of the tunnel, calculate the risk index based on the predicted temperature and humidity, and output the early warning result. S5. Calculate the contribution rates of the permeability state index, surrounding rock thermal reservoir index, air thermal and moisture transport index, and heat exchange index to the multi-domain coupled state index, identify the main causes of risk based on the contribution rates, and output corresponding auxiliary decision-making suggestions.

2. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 1, characterized in that, In step S1, multiple monitoring nodes are deployed inside the tunnel to collect multi-domain monitoring data on water, rock, gas, and heat in the tunnel, specifically including: Multiple monitoring sections are arranged along the tunnel axis and radial direction. Seepage status nodes, surrounding rock temperature nodes, air status nodes and heat exchange nodes are set up at each monitoring section. The seepage temperature and flow rate are collected through the seepage status node; the surface temperature and deep temperature of the surrounding rock are collected through the surrounding rock temperature node; the air temperature, relative humidity and air velocity are collected through the air status node; and the convective heat transfer, evaporative latent heat, condensation latent heat and ventilation heat transfer are collected through the heat exchange node.

3. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 2, characterized in that, In step S1, the collected tunnel water-rock-gas-heat multi-domain monitoring data are represented as follows: ； in, For a moment Multi-domain monitoring data; For a moment Seepage status data; For a moment Surrounding rock condition data; For a moment Air condition data; For a moment Heat exchange status data; ； ； ； ； in, The seepage temperature; This refers to the seepage flow rate; These are parameters related to the wetted state. Temperature at the depth of the surrounding rock. The temperature of the surrounding rock surface; This represents the change in temperature of the surrounding rock. Air temperature; Relative humidity of the air; Air velocity; For convective heat transfer; It is the latent heat of vaporization; The latent heat of condensation; For ventilation and heat exchange.

4. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 1, characterized in that, In step S2, a unified spatial encoding is applied to all monitoring nodes, and the monitoring data is synchronized over time. This specifically includes: Each monitoring node is assigned a unique spatial code: ； in, For the first Spatial coding of each monitoring node; The tunnel mileage; This refers to the cross-sectional orientation; Radial depth; Height from the base plate; The node types include seepage nodes, surrounding rock nodes, air nodes, heat exchange nodes, and construction control nodes; The spatial code, monitoring value, and data quality status of the monitoring node are combined to form the node status: ； in, For the first Each monitoring node at time The node state; For node monitoring values; This represents the data quality status, indicating whether the data is normal, missing, abnormal, drifted, or offline. Monitoring data with different sampling frequencies are synchronized to the same time reference.

5. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 4, characterized in that, In step S2, the spatial registration distance is calculated based on the node spatial encoding, and a coupled monitoring chain is constructed according to the registration distance threshold to form a multi-domain coupled monitoring network, specifically including: Calculate the spatial registration distance between nodes based on the node spatial encoding: ； in, For the first The monitoring node and the first Spatial registration distance of each monitoring node; , The first The monitoring node and the first Spatial coding of each monitoring node; Determine whether two nodes have established a spatial association based on their spatial registration distance and node type: ； in, For the first The monitoring node and the first Spatial registration relationship of monitoring nodes; when Less than or equal to the preset spatial registration threshold, and When the combination of monitoring node types that are allowed to be associated is used, the node is determined. and nodes Establish spatial associations; otherwise, do not establish spatial associations. The seepage nodes, surrounding rock nodes, air nodes, and heat exchange nodes that have completed spatial correlation are grouped into the same local influence unit, and a coupled monitoring chain is constructed, represented as: ； in, For the first A coupled monitoring chain; This is a seepage point; For surrounding rock nodes; An air node; It serves as a heat exchange node; As a construction control point; The coupled monitoring chain is organized according to tunnel mileage, cross-sectional location, node correlation, and construction section to form a coupled monitoring network, which is used to characterize the thermal and moisture influence transmission relationship between different local influence units: ； in, For a multi-domain coupled monitoring network; for the first A coupled monitoring chain, This represents the total number of coupled monitoring chains.

6. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 1, characterized in that, In step S3, the calculation methods for the permeability state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index include: ； ； ； ； in, It is a seepage state index, which characterizes the degree of influence of seepage temperature, seepage flow rate and humidity state on the thermal and humid environment of the tunnel. The surrounding rock thermal reservoir index represents the ability of the surrounding rock to release heat into the cavern; The air heat and moisture transport index characterizes the air's ability to transport heat and moisture. The heat exchange index characterizes the intensity of heat exchange between the surrounding rock, seepage water, and air. This is the normalized value for the seepage temperature; This represents the normalized value of the seepage flow rate. This is the normalized value for the wet state; This represents the normalized value of the surrounding rock surface temperature. This represents the normalized value of the deep temperature of the surrounding rock. This is the normalized value of the change in surrounding rock temperature. This represents the normalized value of convective heat transfer. This is the normalized value of the latent heat of vaporization; This is the normalized value of the latent heat of condensation; This is the normalized value for ventilation heat exchange. , , , , , , , , , These are the weighting coefficients.

7. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 6, characterized in that, In step S3, the multi-domain coupling state index is calculated as follows: ； in, It is a multi-domain coupling state index; , , , These are the weighting coefficients.

8. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 7, characterized in that, In step S4, the future air temperature and humidity in the tunnel are expressed as follows: ； in, This is a forecast of future temperature and humidity. The predicted air temperature for a future time. This is a forecast of relative humidity at a future time. For the predicted duration; The risk index is calculated based on predicted temperature and humidity, and early warning results are output, specifically including: Calculate the risk index: ； in, Temperature and humidity risk index; This is the normalized predicted future air temperature value; This is the normalized forecast value for future relative humidity. Temperature weighting coefficient; Humidity weighting coefficient; The risk level is determined based on the magnitude of the risk index, and a risk warning result is output: ； in, This is a risk warning result, including the risk level. Risk location Time of risk occurrence and risk type The risk levels are divided into four levels: no risk, general risk, medium risk, and high risk; the risk types include high temperature risk, high humidity risk, and combined high temperature and high humidity risk.

9. The method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in claim 8, characterized in that, In step S5, the contribution rates of the permeability state index, surrounding rock thermal reservoir index, air heat and moisture transport index, and heat exchange index to the multi-domain coupling state index are expressed as follows: ； in, For the first The contribution rate of each state index to the multi-domain coupled state index. These correspond to the permeability state index, the surrounding rock thermal reservoir index, the air heat and moisture transport index, and the heat exchange index, respectively. Indicates the first State index; The risk cause identification results are expressed as follows: ； in, The risk causative factor corresponding to the state index with the largest contribution rate; Contribution rate to the seepage state; Contribution rate to the surrounding rock thermal reservoir; Contribution rate to air heat and moisture transport; Contribution rate to heat exchange.

10. A method for dynamic prediction and early warning of tunnel temperature and humidity based on a multi-domain coupled monitoring network as described in any one of claims 1 to 9, characterized in that, In step S5, the auxiliary decision-making suggestions include: ventilation optimization, cooling, sprinkler adjustment, drainage dehumidification, and personnel operation scheduling.

Images