Tunnel intelligent management and control method and device based on big data technology and medium

Abstract

The invention relates to the technical field of tunnel intelligent management and control, in particular to a tunnel intelligent management and control method and device based on a big data technology and a medium. Comprising the following steps that S1, safety data in a tunnel are collected in real time, position information is matched for the safety data, and the safety data comprise environment data, traffic data, equipment data and tunnel structure monitoring data; s2, calculating a danger score of the historical safety data of each position, and evaluating the danger degree of each position in combination with a preset danger threshold value; s3, respectively constructing an ARIMA danger prediction model based on the historical safety data under each danger degree; and S4, obtaining predicted danger degrees of different positions based on the danger prediction model, and controlling safety equipment at different positions in the tunnel to be started for danger elimination based on the predicted danger degrees. Safety risk conditions of different positions of the tunnel are accurately identified, and timeliness and effectiveness of tunnel safety management and control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent tunnel management and control, and in particular to an intelligent tunnel management and control method, device and medium based on big data technology. Background Art

[0002] Tunnels are important load-bearing structures of transportation facilities, which make transportation more convenient, save land, and do not occupy farmland. As a part of transportation facilities, if tunnels are damaged, it will seriously affect people's daily work and life. Therefore, monitoring tunnels has become increasingly important.

[0003] Most of the existing tunnel monitoring technologies use different safety monitoring equipment to monitor various safety issues. Since the environment in which the tunnel is located is diverse, the tunnel monitoring data changes all the time. It is difficult to determine whether the monitoring point has changed and how to predict the time of the monitoring point change, so as to prepare in advance to reduce safety hazards. In addition, the continuous advancement of science and technology has prompted the continuous improvement of tunnel construction technology.

[0004] The historical data of many existing tunnels cannot be used as reference data. For new tunnels, it becomes extremely difficult to conduct real-time monitoring of tunnel detection points and to conduct real-time control and safety warning of changes in tunnel detection points. Summary of the invention

[0005] In order to solve the problem that the historical data of many existing tunnels cannot serve as reference data, and it becomes extremely difficult for new tunnels to conduct real-time monitoring of tunnel detection points and to conduct real-time control and safety warning of tunnel detection points, the present invention provides a tunnel intelligent management and control method, equipment and medium based on big data technology.

[0006] In the first aspect, the present invention provides a tunnel intelligent management and control method based on big data technology, which adopts the following technical solutions: A tunnel intelligent management and control method based on big data technology includes the following steps: S1, real-time collection of safety data in the tunnel and matching of location information for the safety data, wherein the safety data includes environmental data, traffic data, equipment data and tunnel structure monitoring data; S2, calculates the danger score of the historical safety data of each location, and evaluates the danger level of each location in combination with the preset danger threshold; S3, constructing ARIMA hazard prediction models based on historical safety data at each hazard level; S4, based on the danger prediction model, obtain the predicted danger level at different locations, and based on the predicted danger level, control the opening of safety equipment at different locations in the tunnel to eliminate danger.

[0007] Optionally, step S2 specifically includes: Obtain environmental data security scores, traffic data security scores, equipment data security scores, and tunnel structure monitoring data security scores; The different safety scores are weighted and combined to obtain the hazard score; The risk levels include low, medium and high.

[0008] Optionally, the specific steps of obtaining the environmental data security score include: Calculate the degree of deviation between the actual value of environmental data in different periods and the safety standard:

[0009]

[0010] in, is the environmental data security score at time j, is the safety standard value of environmental data, is the actual AQI value of the ith pollutant at time j, is the AQI upper limit of the ith pollutant in the classification standard, is the lower limit of AQI of the ith pollutant in the classification standard, is the upper limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the lower limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the actual concentration of the ith pollutant at time j; The weights are configured according to the characteristics of environmental data at different times, and the environmental data security scores at different times are weighted and integrated to obtain the environmental data security score.

[0011] Optionally, obtaining the traffic data safety score specifically includes: Calculate the traffic data safety score based on vehicle speed and flow data:

[0012] in, is the average speed over a period of time, is the preset safe speed threshold, is the standard deviation of speed variation, and C is an additional term for flow and congestion factors.

[0013] Optionally, obtaining the device data security score specifically includes: Based on the status information of the device, evaluate whether the device is in normal working condition to calculate the device data security score:

[0014] in, is the number of devices currently working normally, is the total number of devices.

[0015] Optionally, obtaining the tunnel structure monitoring data safety score specifically includes: the tunnel structure monitoring data includes displacement sensor data, stress sensor data, and deformation sensor data, and the tunnel structure monitoring data safety score is calculated after evaluating each sensor data:

[0016] in, Provide security scores for tunnel structure monitoring data. is the real-time monitoring value of the i-th sensor data, is the standard value of the i-th sensor data, is the safety threshold of the i-th type of sensor data.

[0017] Optionally, step S3 includes: Divide all historical safety data into low-risk data sets, medium-risk data sets, and high-risk data sets; The corresponding ARIMA hazard prediction models were constructed based on the three data sets respectively; The safety data of the current location at the current moment is input into the ARIMA hazard prediction model, and the model outputs the predicted hazard level of the current location at the next moment.

[0018] Optionally, constructing a risk prediction model specifically includes: After smoothing the historical safety data, the data sets are constructed and divided into training sets and validation sets; Determine the autoregressive order, difference order, and moving average order of the ARIMA model; Fitting historical safety data in the training set based on the ARIMA model; The ARIMA model was evaluated using residual analysis based on the validation set. When the residual was white noise, the ARIMA hazard prediction model was obtained. When the residual was not white noise, the autoregressive order, difference order and moving average order of the model were adjusted and fitted again.

[0019] In the second aspect, a tunnel intelligent management and control device based on big data technology includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the memory is coupled to the processor, and when the processor executes the computer program, it implements the tunnel intelligent management and control platform based on big data technology as mentioned above.

[0020] In a third aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned intelligent tunnel management and control platform based on big data technology.

[0021] In summary, the present invention has the following beneficial technical effects: By collecting safety data including environment, traffic, equipment and tunnel structure monitoring in real time and matching location information, a comprehensive and detailed safety status monitoring of the tunnel is achieved. By calculating the danger score of historical safety data at each location and evaluating the danger level in combination with the preset threshold, the safety risk status at different locations of the tunnel can be accurately identified. An ARIMA danger prediction model with different danger levels is constructed, and the data value is fully mined with the help of big data technology to predict the danger level at different locations. This data-driven prediction method improves the accuracy and scientificity of the prediction. Finally, the opening of safety equipment at different locations is controlled according to the prediction results to eliminate dangers, which not only improves the timeliness and effectiveness of tunnel safety management and control, but also helps to optimize the use of safety equipment, reduce energy consumption and operating costs, thereby comprehensively improving the intelligent management level and safety assurance capabilities of the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of a tunnel intelligent management and control method based on big data technology in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following is combined with Figure 1 The present invention is described in further detail.

[0024] Example 1 Reference Figure 1 , a tunnel intelligent management and control method based on big data technology in this embodiment includes the following steps: S1, real-time collection of safety data in the tunnel and matching of location information for the safety data, wherein the safety data includes environmental data, traffic data, equipment data and tunnel structure monitoring data; The tunnel is divided into sections, and more accurate and densely arranged equipment is arranged at locations where danger is more likely to occur for data collection. Different sections of the tunnel may face different geological conditions. By monitoring the tunnel structure (such as lining deformation, stress and strain, etc.) in sections, once a certain section has structural abnormalities (such as lining cracks, excessive deformation, etc.), the section where the problem is located can be accurately determined, rather than indiscriminately checking the entire tunnel. Due to factors such as ventilation effects and vehicle exhaust emission distribution, a certain section may have a high concentration of harmful gases or abnormal humidity and temperature changes. Segmented monitoring of environmental data (such as carbon monoxide, carbon dioxide concentration, temperature, humidity, etc.) can timely discover environmental problems in specific sections. The equipment in the tunnel (such as lighting equipment, ventilation equipment, drainage equipment, etc.) is distributed in different sections. Segmented monitoring of equipment data (such as equipment operating status, energy consumption, etc.) can reasonably allocate maintenance resources according to the actual situation of each section of equipment. For example, if the lighting equipment in a certain section ages faster (as determined by monitoring the power, voltage and other data of the lighting equipment in this section), the lighting equipment in this section can be updated or repaired first, rather than uniformly maintaining the lighting equipment in the entire tunnel, thereby saving maintenance costs and improving the overall operating efficiency of the equipment. If a section of the tunnel has heavy traffic and a high risk of accidents (determined through traffic data monitoring), the traffic management police force can be increased or more safety warning signs can be installed in that section, without wasting too many management resources on sections with light traffic.

[0025] S2, calculates the risk score of the historical safety data of each location, and evaluates the risk level of each location based on the preset risk threshold; specifically includes: Obtain environmental data security scores, traffic data security scores, equipment data security scores, and tunnel structure monitoring data security scores; The different safety scores are weighted and combined to obtain the risk score R, which is calculated as follows:

[0026] Among them, E is the environmental data security score, T is the traffic data security score, D is the equipment data security score, and S is the tunnel structure monitoring data security score. , , , are all weight factors and their sum is 1. The weights are determined based on the main factors affecting the current location. For example, when traffic influence accounts for a large proportion, Set larger to get a more objective rating.

[0027] The risk levels include low, medium and high.

[0028] The specific steps to obtain the environmental data security score include: Calculate the degree of deviation between the actual value of environmental data at different time periods and the safety standard: Environmental data includes temperature, humidity, visibility and other data. After calculating the AQI of each environmental data, select the maximum value to calculate the environmental data safety score at the current moment to effectively quantify the concentration of different pollutants. The calculation is as follows:

[0029]

[0030] in, is the environmental data security score at time j, is the safety standard value of environmental data, is the actual AQI value of the ith pollutant at time j, is the AQI upper limit of the ith pollutant in the classification standard, is the lower limit of AQI of the ith pollutant in the classification standard, is the upper limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the lower limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the actual concentration of the ith pollutant at time j; Configure weights according to the characteristics of environmental data at different times, and perform weighted synthesis of environmental data security scores at different times to obtain environmental data security scores. For example, if the air quality data on certain days is more important for the assessment, different weights can be assigned to the scores on different days, and the weighted average can be calculated. During peak periods such as holidays, air quality is more critical to the assessment and can be assigned a higher weight.

[0031] Obtaining traffic data safety scores specifically includes: Calculate the traffic data safety score based on vehicle speed and flow data:

[0032] in, is the average speed over a period of time, is the preset safe speed threshold, is the standard deviation of speed variation, C is an additional term for flow and congestion factors; The calculation of additional flow and congestion factors needs to take into account the tunnel capacity. If the flow , we can set C = 1, which means that the traffic is relatively smooth and has no negative impact on the safety score.

[0033] if , set to C = 0.8, indicating a certain degree of congestion impact.

[0034] if , set to C = 0.5, indicating severe congestion, which has a greater impact on the safety score.

[0035] Obtaining the device data security score specifically includes: Based on the status information of the equipment, such as lighting brightness, ventilation efficiency, and fire extinguishing system availability, the equipment is evaluated to see whether it is in normal working condition to calculate the equipment data security score:

[0036] in, is the number of devices currently working normally, is the total number of devices.

[0037] The specific method for obtaining the tunnel structure monitoring data safety score includes: the tunnel structure monitoring data includes displacement sensor data, stress sensor data, and deformation sensor data. After evaluating each sensor data, the tunnel structure monitoring data safety score is calculated:

[0038] in, Provide safety scores for tunnel structure monitoring data. is the real-time monitoring value of the i-th sensor data, is the standard value of the i-th sensor data, is the safety threshold of the i-th sensor data; The displacement sensing data mainly detects the structural displacement at different locations in the tunnel, such as the horizontal and vertical displacement of the tunnel wall and arch; Stress sensing data Stress monitoring is used to assess the magnitude of compressive, tensile or shear stresses in various parts of the tunnel structure (such as the vault and tunnel wall). Stress exceeding the design standard or remaining at a high level for a long time may cause the structure to crack or fail; Deformation sensing data Deformation monitoring evaluates local deformation at different locations of the tunnel, including compression or tension deformation of the vault, walls and bottom of the tunnel structure. Severe deformation may cause structural cracking or even failure, so deformation monitoring is crucial; Therefore, the most serious score among the above three is taken as the overall score of the tunnel structure monitoring data to ensure that these situations in the tunnel can be well identified.

[0039] S3, constructing ARIMA hazard prediction models based on historical safety data at each hazard level; Divide all historical safety data into low-risk data sets, medium-risk data sets, and high-risk data sets; The corresponding ARIMA hazard prediction models were constructed based on the three data sets respectively; The safety data of the current location at the current moment is input into the ARIMA hazard prediction model, and the model outputs the predicted hazard level of the current location at the next moment.

[0040] The construction of risk prediction model specifically includes: After the historical safety data is processed smoothly, the data sets are constructed and divided into training sets and validation sets. Specifically, the data sets include: Statistical methods, such as the 3-sigma principle (data values ​​outside the range of ±3 times the standard deviation of the mean are considered outliers), are used to identify outliers in the data, and the average or median of the adjacent values ​​is used to replace the outliers; for example, sudden large or small carbon monoxide concentration values ​​caused by sensor failure may interfere with the construction of the model.

[0041] Use the difference method to process the data smoothly. For example, if the original time series data shows a clear upward or downward trend, the data can be first-order differentiated (i.e., the current data minus the data at the previous moment). If the data after the first-order difference is still not stable, the second-order difference can be continued. At the same time, the stability of the data can be preliminarily judged by observing the autocorrelation function (ACF) and partial autocorrelation function (PACF) of the data. The ACF and PACF of stable data will decay quickly.

[0042] Determine the autoregressive order, differencing order, and moving average order of the ARIMA model; determine the autoregressive order p by observing the partial autocorrelation function (PACF). If the PACF suddenly cuts off after a certain lag order (starting from this lag order, the value of the PACF is within the confidence interval and close to 0), then this lag order is a possible value of p. For example, if the PACF suddenly cuts off after lag 3, then p = 3 is preliminarily determined. Determine d based on the results of the previous data stationarity treatment. If the data reaches stationarity after the first-order difference, d = 1; if it reaches stationarity only after the second-order difference, d = 2. Determine the moving average order q by observing the autocorrelation function (ACF). If the ACF suddenly cuts off after a certain lag order, then this lag order is a possible value of q. For example, if the ACF cuts off after lag 2, q = 2 is preliminarily determined.

[0043] Fitting historical safety data in the training set based on the ARIMA model; The ARIMA model is evaluated using residual analysis based on the validation set. When the residual is white noise, the ARIMA hazard prediction model is obtained. When the residual is not white noise, the autoregressive order, difference order and moving average order of the model are adjusted and fitted again. It can be judged by observing the autocorrelation function (ACF) of the residual and the Ljung-Box test. If the autocorrelation coefficients of all lag orders in the ACF diagram of the residual are within the confidence interval, and the p-value of the Ljung-Box test is greater than the set significance level (such as 0.05), the residual can be considered to be white noise.

[0044] S4, based on the danger prediction model, obtain the predicted danger level at different locations, and based on the predicted danger level, control the opening of safety equipment at different locations in the tunnel to eliminate danger.

[0045] After confirming that the model is effective, you can use it to make predictions. For short-term predictions (such as predicting carbon monoxide concentrations or accident frequency in the next few hours or days), you can directly apply the model to the time point after the last known data point for prediction.

[0046] By predicting the danger level of different locations in the tunnel through the danger prediction model, we can know in advance which areas may be in danger. For example, when it is predicted that the carbon monoxide concentration in a certain section of the tunnel is about to exceed the standard (based on the prediction of factors such as traffic flow and ventilation conditions), the ventilation equipment at that location is turned on in time to avoid the harm caused by excessive carbon monoxide concentration to personnel and vehicles, effectively improving the overall safety of the tunnel.

[0047] With regard to tunnel structures, if it is predicted that a certain area may face the risk of structural deformation or collapse due to geological changes (based on model predictions of structural monitoring data and geological conditions, etc.), reinforcement measures can be taken in advance, such as starting the grouting equipment at that location or adjusting safety equipment such as support pressure, to ensure the stability of the tunnel structure.

[0048] There is no need to indiscriminately activate the safety equipment of the entire tunnel, but to activate the equipment in a targeted manner according to the predicted danger level of different locations. For example, the drainage pump at a certain location is only activated when it is predicted that there is a risk of water accumulation (based on the prediction of drainage equipment data, precipitation and other factors), which avoids unnecessary energy consumption and equipment wear and improves the efficiency of the use of safety equipment.

[0049] Once the danger level is predicted to increase, a quick response can be made and the safety equipment at the corresponding location can be activated in time. For example, when a fire risk is predicted in a certain section of the tunnel (based on the prediction of ambient temperature, smoke concentration trend, etc.), the fire-fighting equipment at that location (such as fire extinguishers, fire sprinklers, etc.) is immediately activated to put out the fire in its infancy and reduce the damage to the tunnel and passing vehicles.

[0050] In the event of sudden natural disasters such as earthquakes and landslides, if the degree of danger at different locations in the tunnel can be predicted in advance, relevant safety equipment can be controlled in advance to reduce the impact of the disaster. For example, if it is predicted that the mountain above a certain section of the tunnel may collapse, the early warning device at that location can be activated in advance to notify passing vehicles to avoid it, and protective equipment (such as guardrails, buffer devices, etc.) can be activated to reduce the severity of the accident.

[0051] Example 2 The difference between this embodiment and Embodiment 1 is that this embodiment provides a tunnel intelligent management and control device based on big data technology, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the memory is coupled to the processor, and when the processor executes the computer program, it implements the tunnel intelligent management and control platform based on big data technology as mentioned above.

[0052] Example 3 The difference between this embodiment and embodiment 1 is that this embodiment provides a computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned tunnel intelligent management and control platform based on big data technology.

[0053] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0054] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0055] The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto. Any non-creative changes that can be thought of by a person skilled in the art, as well as several improvements and modifications made without departing from the principle of the present invention, should fall within the protection scope of the present invention.

Claims

1. A tunnel intelligent management and control method based on big data technology, characterized in that: The following steps are involved: S1, real-time collection of safety data in the tunnel and matching of location information for the safety data, wherein the safety data includes environmental data, traffic data, equipment data and tunnel structure monitoring data; S2, calculates the danger score of the historical safety data of each location, and evaluates the danger level of each location in combination with the preset danger threshold; S3, constructing ARIMA hazard prediction models based on historical safety data at each hazard level; S4, based on the danger prediction model, obtain the predicted danger level at different locations, and based on the predicted danger level, control the opening of safety equipment at different locations in the tunnel to eliminate danger.

2. The intelligent tunnel management and control method based on big data technology according to claim 1 is characterized in that: Step S2 specifically includes: Obtain environmental data security scores, traffic data security scores, equipment data security scores, and tunnel structure monitoring data security scores; The different safety scores are weighted and combined to obtain the hazard score; The risk levels include low, medium and high.

3. The intelligent tunnel management and control method based on big data technology according to claim 2 is characterized in that: The specific steps to obtain the environmental data security score include: Calculate the degree of deviation between the actual value of environmental data in different periods and the safety standard: in, is the environmental data security score at time j, is the safety standard value of environmental data, is the actual AQI value of the ith pollutant at time j, is the AQI upper limit of the ith pollutant in the classification standard, is the lower limit of AQI of the ith pollutant in the classification standard, is the upper limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the lower limit of the concentration of the interval where the ith pollutant in the classification standard is located, is the actual concentration of the ith pollutant at time j; The weights are configured according to the characteristics of environmental data at different times, and the environmental data security scores at different times are weighted and integrated to obtain the environmental data security score.

4. The intelligent tunnel management and control method based on big data technology according to claim 2 is characterized in that: Obtaining traffic data safety scores specifically includes: Calculate the traffic data safety score based on vehicle speed and flow data: in, is the average speed over a period of time, is the preset safe speed threshold, is the standard deviation of speed variation, and C is an additional term for flow and congestion factors.

5. The intelligent tunnel management and control method based on big data technology according to claim 2 is characterized in that: Obtaining the device data security score specifically includes: Based on the status information of the device, evaluate whether the device is in normal working condition to calculate the device data security score: in, is the number of devices currently working normally, is the total number of devices.

6. The intelligent tunnel management and control method based on big data technology according to claim 2 is characterized in that: The specific method for obtaining the tunnel structure monitoring data safety score includes: the tunnel structure monitoring data includes displacement sensor data, stress sensor data, and deformation sensor data. After evaluating each sensor data, the tunnel structure monitoring data safety score is calculated: in, Provide safety scores for tunnel structure monitoring data. is the real-time monitoring value of the i-th sensor data, is the standard value of the i-th sensor data, is the safety threshold of the i-th type of sensor data.

7. The intelligent tunnel management and control method based on big data technology according to claim 2 is characterized in that: Step S3 includes: Divide all historical safety data into low-risk data sets, medium-risk data sets, and high-risk data sets; The corresponding ARIMA hazard prediction models were constructed based on the three data sets respectively; The safety data of the current location at the current moment is input into the ARIMA hazard prediction model, and the model outputs the predicted hazard level of the current location at the next moment.

8. The intelligent tunnel management and control method based on big data technology according to claim 7 is characterized in that: The construction of risk prediction model specifically includes: After smoothing the historical safety data, the data sets are constructed and divided into training sets and validation sets; Determine the autoregressive order, difference order, and moving average order of the ARIMA model; Fitting historical safety data in the training set based on the ARIMA model; The ARIMA model was evaluated using residual analysis based on the validation set. When the residual was white noise, the ARIMA hazard prediction model was obtained. When the residual was not white noise, the autoregressive order, difference order and moving average order of the model were adjusted and fitted again.

9. A tunnel intelligent management and control device based on big data technology, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the memory is coupled to the processor, and when the processor executes the computer program, it implements the tunnel intelligent management and control platform based on big data technology as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the tunnel intelligent management and control platform based on big data technology as described in any one of claims 1 to 8.

Images