Optimization method of ventilation layout in submarine tunnel based on digital twin
Through digital twin technology, the layout of the ventilation system in the subsea tunnel is optimized, combined with real-time monitoring of the sensor network, the problem of poor adaptability of the ventilation system under high traffic flow and pollutant concentration in the tunnel is solved, and better ventilation effect and system stability are achieved.
Patent Information
- Application Number
- CN202411166919.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-23
AI Technical Summary
The existing undersea tunnel ventilation system layout optimization method has large operating loads under high traffic flow and pollutant concentrations, poor adaptability, difficult to achieve the expected ventilation effect, and lack effective real-time monitoring and optimization solutions.
The ventilation layout optimization method of the subsea tunnel segment based on digital twin is adopted. By generating effect coefficients and optimization degrees, the ventilation system layout is optimized, the operating status is monitored in real time using the sensor network, and maintenance and layout adjustments are carried out in a timely manner to ensure ventilation effect and system stability.
The ventilation system adaptability under different working conditions is achieved, the ventilation effect is improved, the air quality and operating state in the tunnel are normal, the fault frequency is reduced, and the overall performance of the ventilation system is improved.
Smart Images

Figure CN118690474B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of ventilation layout optimization, and in particular to a method for optimizing the ventilation layout of an undersea tunnel interval based on digital twins. Background Art
[0002] The transportation hub building on the artificial island of the cross-sea project has a different geographical location, climatic environment, and complexity in engineering technology and cooperation professions from land buildings. It is difficult to meet the requirements of the project according to conventional architectural design ideas. The tunnel ventilation system is a key engineering facility, and its design, construction and operation must strictly follow relevant standards and specifications. The tunnel ventilation system is responsible for ventilating the interval tunnel and exhausting smoke from the tunnel in the event of a tunnel fire. Its goal is to quickly discharge harmful gases, smoke and heat from the tunnel, ensure that the internal air is clean and the temperature is suitable, and provide a comfortable traffic environment for drivers.
[0003] The spatial layout design of traditional cross-sea tunnels is based on the layout design of ventilation ducts as the main design of the entire building. The ventilation ducts occupy a large amount of physical space. Therefore, it is particularly important to optimize the spatial layout of ventilation ducts in the design of cross-sea tunnels.
[0004] In the process of complex engineering and design, such as transportation hub buildings on artificial islands of cross-sea projects, it is particularly important to predict and analyze their construction progress and judge the impact of their design in advance. As a highly integrated digital representation, the digital twin model simulates the entire life cycle of a physical object or system. Specifically, this method for complex engineering is a method called digital twin. The digital twin model is a virtual "twin" that contains all the key features, performance parameters, operating status and other information of the physical object, and can collect data of the physical object in real time through sensors, the Internet of Things and other technologies to achieve synchronous updates with the physical object. The core value of the digital twin model is that it can simulate the behavior and performance of physical objects in a virtual environment, so as to predict, analyze and optimize. This simulation can help people better understand the operating mechanism of physical objects, discover potential problems, predict future performance changes, and make decisions accordingly. Therefore, optimizing the ventilation layout through digital twins and obtaining the overall optimization of the building space layout is an important purpose of the patent of this invention.
[0005] In a Chinese invention patent with authorization announcement number CN116663412B, a method and system for optimizing the layout of a zero-carbon building based on deep learning technology is disclosed, which relates to the technical field of zero-carbon building layout optimization. A training data set of zero-carbon buildings is collected in advance, and a first deep learning model for predicting light intensity, a second deep learning model for predicting building ventilation, and a third deep learning model for predicting building temperature stability are trained using the training data set. The orientation adjustment step and the size adjustment step, the initial building orientation, and the initial window size are set, the building light intensity, building ventilation, and building temperature stability are calculated, and the gradient descent method is used to find a suitable building layout.
[0006] A tunnel construction ventilation system is disclosed in a Chinese invention patent with application publication number CN114718629A, which relates to the field of tunnel construction. Its technical solution includes a control host, a CMOS camera installed inside the tunnel, a monitoring device and a ventilation device. The control host includes a data system, a central processing system, a cloud storage module, an output system and an image system. The data system includes a gas monitoring module, a CO monitoring module and a dust monitoring module. The data system monitors the data inside the tunnel in real time, and the data system is communicated with the central processing system.
[0007] Combined with the above application and the contents of the prior art:
[0008] After the ventilation system needs to be arranged in the tunnel area, in order to achieve the expected ventilation effect, it is necessary to set a reasonable layout for the ventilation system so that the positions of various components of the ventilation system are more reasonable. The existing ventilation system layout plan is usually designed after the tunnel construction is completed, and then the ventilation system is installed according to the design plan. In order to improve the operation effect of the ventilation system, after obtaining the ventilation layout plan, the layout plan is usually optimized again.
[0009] The existing layout optimization method usually uses a multi-objective optimization algorithm to adjust the positions of various components of the ventilation system after setting the ventilation target to achieve the optimization of the ventilation system. However, this optimization scheme takes fewer factors into consideration. When the ventilation demand in the tunnel is high, for example, when the traffic volume in the tunnel is large and the pollution level is high, the operating load of the ventilation system is also large. The layout optimization scheme has poor adaptability to different working conditions. The original ventilation system layout scheme may be difficult to achieve the expected effect, and the ventilation effect in the tunnel will be affected to a certain extent.
[0010] To this end, the present invention provides a method for optimizing the ventilation layout of submarine tunnel intervals based on digital twins. Summary of the invention
[0011] 1. Technical issues to be resolved
[0012] In view of the deficiencies of the prior art, the present invention provides a method for optimizing the layout of ventilation in submarine tunnels based on digital twins. The method generates an effect coefficient from test data. If the effect coefficient does not exceed the effect threshold, the ventilation system layout optimization scheme is output from the ventilation system layout optimization knowledge graph. The optimized scheme is simulated and tested using the trained tunnel ventilation digital twin model. The optimization degree is generated using the test data, and corresponding processing is taken according to the optimization degree. The operating status data of the ventilation system is collected by the sensor network, and the ventilation system is inspected and maintained. The fault density is generated from the fault data of the ventilation system. If it exceeds expectations, an alarm instruction is issued. It is ensured that its operating status can be maintained normally, and the ventilation effect in the tunnel is guaranteed, thereby solving the technical problems recorded in the background technology.
[0013] (II) Technical solution
[0014] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for optimizing the ventilation layout of an undersea tunnel interval based on digital twins, comprising: collecting environmental data in the tunnel and constructing a target area condition data set, and generating a ventilation demand coefficient in the target area from the regional condition data set. , if the ventilation demand coefficient Exceeding demand expectations, issuing data collection instructions to external parties;
[0015] After collecting ventilation system data, tunnel structure data and environment data in the target area, the data is used as sample data, and the tunnel ventilation digital twin model is generated by training the sample data.
[0016] Set different target test scenarios for the ventilation system. Under different target test scenarios, use the tunnel ventilation digital twin model to test the current layout of the ventilation system, and generate effect coefficients from the test data. , if the effect coefficient Without exceeding the effect threshold, layout optimization instructions are issued to the outside;
[0017] Feature extraction is performed on the test data. Based on the obtained optimization features, the ventilation system layout optimization plan is output from the ventilation system layout optimization knowledge graph. The optimized plan is simulated and tested using the trained tunnel ventilation digital twin model. The optimization degree is generated using the test data. , according to the optimization degree Take appropriate measures;
[0018] The sensor network collects the operating status data of the ventilation system, and generates the operating status coefficient from the operating status data. , where the response time to the acquisition Energy efficiency ratio Perform linear normalization and map the corresponding data values to the interval Then follow the steps below:
[0019] ;
[0020] Weight coefficient: , ;
[0021] If the operating state coefficient If the ventilation system is lower than expected, the system will be inspected and maintained, and the fault density will be generated from the fault data of the ventilation system. , if the obtained fault density When the density threshold is exceeded, an alarm command is issued to the outside.
[0022] Furthermore, environmental data and traffic flow data in the target area are collected, and the above data are aggregated to construct a target area condition data set; the ventilation demand coefficient is generated from the area condition data set , where the pollutant concentration and traffic volume After linear normalization, the corresponding data values are mapped to the interval According to the following formula:
[0023] ;
[0024] in, In the target area i The pollutant concentration at each monitoring node is For the i The traffic flow at each monitoring node is is the weight coefficient, .
[0025] Furthermore, after receiving the data collection instruction, data collection is performed, including: performance data of the ventilation system, tunnel size and structure, and the acquired data are combined with the data in the target area condition data set. After data preprocessing, the first model data set is summarized and constructed; based on the first model data set, simulation analysis software is used to perform fluid dynamics simulation to obtain air flow, temperature distribution and pollutant diffusion data in the target area, which are summarized as the second model data set.
[0026] Furthermore, different operation scenarios are set for the ventilation system in the target area and used as target test scenarios. In the target test scenarios, the ventilation system is tested using the trained tunnel ventilation digital twin model to obtain corresponding test data. The obtained simulation test data are aggregated to construct a ventilation test data set, and the effect coefficient is generated from the test data in the ventilation test data set. .
[0027] Furthermore, after obtaining the temperature gradients in different sub-areas of the target area, and wind energy density After that, linear normalization is performed on both of them, and the corresponding data values are mapped to the interval The effect coefficient is constructed according to the following formula :
[0028] ;
[0029] Weight coefficient: , , is the number of sub-regions; For the i The temperature gradient in each sub-region is is the mean value of the temperature gradient, is the variance of the temperature gradient; For the i The wind energy density in each sub-region is is the mean value of wind energy density, is the variance of wind energy density.
[0030] Furthermore, after receiving the layout optimization instruction, various test data in the ventilation test data set are obtained, and feature extraction is performed on the test data to obtain corresponding optimization features; the optimization of the ventilation system layout in the tunnel is used as the target word, and a ventilation system layout optimization knowledge graph is pre-constructed; the optimization features are used as input, and according to the correspondence between the optimization features and the ventilation system layout optimization scheme, the ventilation system layout optimization scheme is output from the ventilation system layout optimization knowledge graph to optimize the current layout scheme of the ventilation system.
[0031] Furthermore, taking the ventilation system layout optimization plan as input, the trained tunnel ventilation digital twin model was used to simulate the ventilation system layout optimization in the target area, and the corresponding effect coefficients were obtained at several test nodes. Then, according to the effect coefficient Changes in build optimization , if the optimization degree Without exceeding the preset optimization threshold, a solution optimization instruction is issued to the outside.
[0032] Furthermore, the test nodes before and after the ventilation system layout optimization plan are sorted according to the time axis and correspond one by one. The optimization degree is constructed according to the following method: :
[0033] ;
[0034] in, For thei The optimized intermediate value of the test nodes, is its mean value, , is the number of test nodes, and They are the ventilation system layout optimization schemes before and after optimization. i The effect coefficient of the test node, and is the corresponding mean.
[0035] Furthermore, after the ventilation system enters the operating state, the operating state data of the ventilation system is collected by the sensor network in the target area, and the operating state data set is constructed after being aggregated; the operating state coefficient is constructed from the operating state data of the ventilation system. If the obtained operating status coefficient is lower than the preset performance threshold, a maintenance instruction is issued to the outside.
[0036] Furthermore, after receiving the maintenance instruction, the location and time of the ventilation system fault are recorded, and the fault data set of the ventilation system is generated after being summarized; the fault density is generated from the fault data set , as follows:
[0037] ;
[0038] in, is the number of ventilation system failures, It is a malfunction i To failure j The shortest distance, is the average distance, It is a malfunction i To failure j The time interval is the average value of the time interval, and the weight coefficient is: , ; The weight coefficient can be obtained by referring to the hierarchical analysis method.
[0039] (III) Beneficial effects
[0040] The present invention provides a method for optimizing the ventilation layout of submarine tunnel intervals based on digital twins, which has the following beneficial effects:
[0041] 1. Ventilation demand coefficient Judge the ventilation demand in the target area. If the ventilation demand in the tunnel is large, install a ventilation system in the target area to increase the ventilation capacity in the target area and ensure the ventilation effect. The trained tunnel ventilation digital twin model can be used to test the current ventilation effect in the target area and make a judgment on the current ventilation effect.
[0042] 2. Test the ventilation system under different working conditions. Based on the obtained test data, the adaptability of the ventilation system and its layout to different scenarios and the corresponding ventilation effect can be determined; the corresponding effect coefficient is generated from the obtained test data. , with effect coefficient Evaluate and judge whether the current ventilation system layout is feasible, and judge whether the ventilation system layout can achieve the expected effect. If it cannot achieve the expected effect, it means that the current ventilation system layout may be insufficient and needs to be adjusted and optimized in time.
[0043] 3. Extract features from the acquired test data, and output the ventilation system layout optimization plan from the ventilation system layout optimization knowledge graph to optimize the layout of the ventilation system in the target area, optimize the various equipment in the ventilation system and its working parameters, and improve the ventilation effect of the ventilation system in the target area by adjusting and optimizing the ventilation system.
[0044] 4. The tunnel ventilation digital twin model tests the obtained optimization solution and obtains the effect coefficient at each test node. Post-generation optimization , to optimize Evaluate and screen the optimization plan to confirm its feasibility. If the optimization plan is not feasible, re-formulate the optimization plan or re-optimize the optimization plan to optimize the layout of the ventilation system in the tunnel and improve the ventilation effect in the tunnel.
[0045] 5. Based on the operating status coefficient Evaluate the current operating status of the ventilation system to determine whether there are any abnormalities in the current operating status. If there are any abnormalities, perform maintenance on it in a timely manner to ensure that its operating status can remain normal and avoid further deterioration or abnormalities in the operating status of the ventilation system.
[0046] 6. If there is a fault in a component of the ventilation system, the fault density is generated based on the fault location and fault time of the component. , based on the fault density Assess the frequency of ventilation system failures, maintain the ventilation system, and further ensure the ventilation effect in the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 It is a schematic flow chart of the method for optimizing ventilation layout of submarine tunnel intervals based on digital twins of the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] See also Figure 1 The present invention provides a method for optimizing the ventilation layout of an undersea tunnel interval based on digital twins, comprising:
[0050] Step 1: Collect environmental data in the tunnel and construct a target area condition data set, and generate the ventilation demand coefficient in the target area from the regional condition data set , if the ventilation demand coefficient Exceeding demand expectations, issuing data collection instructions to external parties;
[0051] The step 1 includes the following contents:
[0052] Step 101: After selecting the installation area of the ventilation system in the tunnel, take it as the target area, collect environmental data in the target area, including temperature, humidity and pollutant concentration at each monitoring node in the tunnel, traffic flow change data, including vehicle type, speed and flow in different time periods, etc., and summarize the above data to construct a target area condition data set;
[0053] Step 102: Generate ventilation demand coefficient in the target area from the regional condition data set, where the pollutant concentration and traffic volume After linear normalization, the corresponding data values are mapped to the interval According to the following formula:
[0054] ;
[0055] in, In the target area i The pollutant concentration at each monitoring node is For the i The traffic flow at each monitoring node is is the weight coefficient, ;
[0056] Based on historical data and ventilation demand management expectations in the target area, the demand threshold is set in advance; if the ventilation demand coefficient If the demand threshold is exceeded, it means that the ventilation demand in the current target area is relatively large, and the ventilation effect in the target area needs to be adjusted and the ventilation layout needs to be re-optimized. At this time, a data collection instruction is issued to the outside;
[0057] Among them, pollutant concentration and vehicle flow will affect the ventilation demand in the tunnel. This is mainly because the tunnel is a relatively closed space, and its internal air quality is directly affected by external pollution sources, and vehicle emissions are one of the main pollution sources in the tunnel. The following is a specific analysis:
[0058] Impact of pollutant concentration: Automobile exhaust contains a variety of harmful components, such as carbon monoxide, soot, aluminum, phosphide, sulfur, etc., which will pollute the air in the tunnel. When the concentration of these pollutants accumulates to a certain level, it will not only harm human health, such as causing poisoning symptoms and even endangering life, but also affect driving safety, such as reducing visibility.
[0059] According to the relevant specifications of tunnel construction ventilation technology, there are strict concentration control requirements for harmful gases such as carbon monoxide. For example, the volume concentration of carbon monoxide in tunnel construction should be controlled within 0.0024% to ensure the safety of construction workers.
[0060] Impact of traffic volume: Traffic volume is an important determinant of pollutant emissions in tunnels. The concentration of pollutants produced by vehicle exhaust emissions is related to traffic volume and speed. The greater the traffic volume, the higher the pollutant concentration.
[0061] According to the calculation formula of tunnel fan ventilation volume, the ventilation volume (Q) is proportional to the vehicle flow (N). This shows that as the vehicle flow increases, the required ventilation volume will also increase accordingly in order to maintain the air quality in the tunnel.
[0062] Poor ventilation conditions may lead to serious air pollution and safety hazards. Therefore, when designing the tunnel ventilation system, it is necessary to fully consider the impact of vehicle flow to ensure that the ventilation system can meet the ventilation needs under different vehicle flows.
[0063] When using, combine the contents in steps 101 and 102:
[0064] When it is necessary to install a ventilation system inside the tunnel and to lay out the ventilation system, various data are collected in the tunnel, and the ventilation demand coefficient is generated from the collected demand data. , according to the ventilation demand coefficient Determine the ventilation demand in the target area. If the ventilation demand in the tunnel is large, a ventilation system can be installed in the target area to increase the ventilation capacity in the target area and ensure the ventilation effect.
[0065] The existing layout optimization method usually uses a multi-objective optimization algorithm to adjust the positions of various components of the ventilation system after setting the ventilation target to achieve the optimization of the ventilation system. However, this optimization scheme takes fewer factors into consideration. When the ventilation demand in the tunnel is high, for example, when the traffic volume in the tunnel is large and the pollution level is high, the operating load of the ventilation system is also large. The layout optimization scheme has poor adaptability to different working conditions. The original ventilation system layout scheme may be difficult to achieve the expected effect, and the ventilation effect in the tunnel will be affected to a certain extent.
[0066] Step 2: After collecting ventilation system data, tunnel structure data and environment data in the target area, use them as sample data to train and generate a tunnel ventilation digital twin model;
[0067] The step 2 includes the following contents:
[0068] Step 201: after receiving the data collection instruction, data collection is performed, including: performance data of the ventilation system, including air volume, air pressure and energy consumption, etc., tunnel size and structural information, such as the cross section, longitudinal section, entrance and exit design of the tunnel, etc.; the acquired data is combined with the data in the target area condition data set, and after data preprocessing, the first model data set is summarized and constructed;
[0069] Step 202: Based on the first model data set, a simulation analysis software, such as a finite element analysis software, is used to perform a fluid dynamics simulation to obtain air flow, temperature distribution, and pollutant diffusion data in the target area, and the data are summarized as a second model data set.
[0070] Extracting part of the data from the first and second model data sets as sample data, and after constructing an initial model by a machine learning algorithm, training the initial model by the sample data to generate a trained tunnel ventilation digital twin model;
[0071] When using, combine the contents in steps 201 and 202:
[0072] After data collection in the target area, the digital twin model of tunnel ventilation is trained and acquired by combining the performance data of the ventilation system and the current layout plan of the ventilation system in the target area. The trained digital twin model of tunnel ventilation can be used to test the current ventilation effect in the target area and make a judgment on the current ventilation effect.
[0073] Step 3: Set different target test scenarios for the ventilation system. Under different target test scenarios, use the tunnel ventilation digital twin model to test the current layout of the ventilation system and generate the effect coefficient from the test data. , if the effect coefficient Without exceeding the effect threshold, layout optimization instructions are issued to the outside;
[0074] The step three includes the following contents:
[0075] Step 301, setting different operating scenarios for the ventilation system in the target area, including normal operation, vehicle traffic peak hours, fire emergency situations, etc., as target test scenarios;
[0076] In the target test scenario, the ventilation system is tested using the trained tunnel ventilation digital twin model to achieve multi-condition simulation. The response and performance of the ventilation system under different conditions are evaluated based on the simulated test data. The corresponding test data are obtained, including pollutant concentration, relative humidity inside the tunnel and at the entrance / exit, and smoke generation rate. The obtained simulated test data are aggregated to construct a ventilation test data set.
[0077] When in use, by pre-setting a variety of different working conditions, the ventilation system is tested under different working conditions. Based on the test data obtained, the adaptability of the ventilation system and its layout to different scenarios and the corresponding ventilation effects can be determined.
[0078] Step 302: Generate index data such as wind speed distribution, temperature gradient and pollutant concentration under different target test scenarios from the test data in the ventilation test data set; after setting a qualified reference value under the corresponding target test scenario for each index data, generate an effect coefficient from the index data , with effect coefficient Provide feedback on the ventilation effect;
[0079] The temperature gradients in different sub-areas of the target area are obtained. (i.e. the change in temperature per unit distance) and wind energy density After that, linear normalization is performed on both of them, and the corresponding data values are mapped to the interval According to the following formula:
[0080] ;
[0081] Weight coefficient: , , is the number of sub-regions; For the i The temperature gradient in each sub-region is is the mean value of the temperature gradient, is the variance of the temperature gradient; For the i The wind energy density in each sub-region is is the mean value of wind energy density, is the variance of wind energy density;
[0082] Combine historical data and management expectations of ventilation effects to pre-set effect thresholds; if the effect coefficient is obtained If the effect threshold is exceeded, it means that the ventilation system in the target area has achieved the expected effect when it is running. If the effect coefficient is If the effect threshold is not exceeded, it means that the ventilation system in the target area cannot achieve the expected effect during operation and the ventilation performance is insufficient. At this time, a layout optimization instruction is issued to the outside.
[0083] When using, combine the contents in steps 301 and 302:
[0084] After completing the test of the ventilation system of the current layout, the corresponding effect coefficient is generated from the obtained test data , with effect coefficient Evaluate and judge whether the current ventilation system layout is feasible, and judge whether the ventilation system layout can achieve the expected effect. If it cannot achieve the expected effect, it means that the current ventilation system layout may be insufficient and needs to be adjusted and optimized in time.
[0085] Step 4: Extract features from the test data. Based on the obtained optimization features, output the ventilation system layout optimization plan from the ventilation system layout optimization knowledge graph. Use the trained tunnel ventilation digital twin model to simulate and test the optimization plan. Use the test data to generate the optimization degree. , according to the optimization degree Take corresponding measures;
[0086] The step 4 includes the following contents:
[0087] Step 401: after receiving the layout optimization instruction, obtain various test data in the ventilation test data set, and pre-set qualified reference values according to qualified expectations of various parameter indicators; extract features from the test data, and obtain corresponding optimization features according to whether the current ventilation system needs to be optimized;
[0088] Taking optimization of ventilation system layout in tunnel as the target word, after deep retrieval and entity relationship building, a knowledge graph of ventilation system layout optimization is pre-built; taking optimization features as input, according to the correspondence between optimization features and ventilation system layout optimization scheme, the ventilation system layout optimization scheme is output from the ventilation system layout optimization knowledge graph, and the current layout scheme of the ventilation system is optimized by adjusting the layout of ventilation ports, increasing or decreasing the number of fans, changing the fan operation mode, etc., to improve ventilation efficiency and safety;
[0089] During use, when it is necessary to optimize the current ventilation system layout, feature extraction is performed on the acquired test data, and the ventilation system layout optimization plan is output from the ventilation system layout optimization knowledge graph. Based on the acquired test data, the knowledge graph outputs the corresponding optimization plan as a reference plan to optimize the layout of the ventilation system in the target area, optimize the various equipment in the ventilation system and its working parameters, and improve the ventilation effect of the ventilation system in the target area by completing the adjustment and optimization of the ventilation system.
[0090] Step 402: Taking the ventilation system layout optimization plan as input, the trained tunnel ventilation digital twin model is used to simulate and test the ventilation system layout optimization in the target area, and the corresponding effect coefficients are obtained at several test nodes. Then, according to the effect coefficient Changes in build optimization , as follows:
[0091] The test nodes before and after the ventilation system layout optimization plan are sorted according to the timeline and correspond one by one. The optimization degree is constructed according to the following method :
[0092] ;
[0093] in, For the i The optimized intermediate value of the test nodes, is its mean value, , is the number of test nodes, and They are the ventilation system layout optimization schemes before and after optimization. i The effect coefficient of the test node, and is the corresponding mean;
[0094] To obtain the optimization Verify its improvement effect on ventilation performance and the practical feasibility of the optimization plan; pre-set the optimization threshold based on historical data and management expectations of the optimization effect; if the obtained optimization degree If the optimization threshold is exceeded, it means that the current optimization plan is feasible and can be executed;
[0095] If the optimization Without exceeding the optimization threshold, an optimization instruction is issued to the outside to re-optimize the current ventilation system layout optimization plan. For example, a multi-objective optimization algorithm is used to fine-tune the key parameters of the ventilation system, such as the fan installation position, speed, and vent opening, to achieve the best ventilation effect.
[0096] When using, combine the contents in steps 401 and 402:
[0097] After obtaining the optimization plan, after setting up several test nodes, the tunnel ventilation digital twin model tests the obtained optimization plan, and the corresponding effect coefficient is reconstructed based on the test data. , and obtain the effect coefficient on each test node Post-generation optimization , to optimize Evaluate and screen the optimization plan to confirm its feasibility. If the optimization plan is not feasible, re-formulate the optimization plan or re-optimize the optimization plan to optimize the layout of the ventilation system in the tunnel and improve the ventilation effect in the tunnel.
[0098] Step 5: The sensor network collects the operating status data of the ventilation system and generates the operating status coefficient from the operating status data. , if the operating state coefficient If the ventilation system is lower than expected, the system will be inspected and maintained, and the fault density will be generated from the fault data of the ventilation system. , if the obtained fault density When the density threshold is exceeded, an alarm command is issued to the outside;
[0099] The step five includes the following contents:
[0100] Step 501: After the ventilation system enters the operating state, the sensor network in the target area collects the operating state data of the ventilation system, including response time, energy efficiency ratio, noise level, and air quality improvement rate, and then constructs an operating state data set after aggregation;
[0101] Constructing the operating status coefficient from the operating status data of the ventilation system , where the response time to the acquisition Energy efficiency ratio Perform linear normalization and map the corresponding data values to the interval Then follow the steps below:
[0102] ;
[0103] Weight coefficient: , ;The weight coefficient can be obtained by referring to the hierarchical analysis method;
[0104] If the obtained operating status coefficient is lower than the preset performance threshold, it means that the ventilation system may have a fault during operation and needs to be maintained, and a maintenance instruction is issued to the outside;
[0105] When in use, the operating status data of the ventilation system is monitored in the target area, and an operating status coefficient is generated from the operating status data. The current operating status of the ventilation system is evaluated based on the operating status coefficient to determine whether there is any abnormality in the current operating status. If there is any abnormality, it can be maintained in time to ensure that its operating status can be maintained normal, thereby avoiding further deterioration of the operating status of the ventilation system or the occurrence of abnormalities.
[0106] The operating status data of the ventilation system mainly includes the following key parameters, which are essential for evaluating the performance and efficiency of the ventilation system:
[0107] Air flow (air volume): refers to the amount of air flowing through the fan per unit time. Common units are m³ / s, m³ / min, and m³ / h. Air flow is an important indicator of ventilation system performance, reflecting the system's ability to transport air.
[0108] Wind pressure: The wind pressure of the fan is divided into static pressure, dynamic pressure and total pressure. Static pressure: refers to the pressure energy (i.e. pressure potential energy) obtained by a unit volume of gas after it flows through the fan. Dynamic pressure: refers to the kinetic energy obtained by a unit volume of gas after it flows through the fan. Total pressure: refers to the total energy obtained by a unit volume of gas after it flows through the fan. The size of wind pressure reflects the ability of the ventilation system to overcome resistance and is one of the key parameters for evaluating system performance.
[0109] Temperature and humidity: In addition to providing fresh air, the ventilation system also needs to maintain appropriate temperature and humidity levels.
[0110] By measuring the temperature and humidity of indoor air, it is possible to assess the heat and moisture loads of the ventilation system and whether comfort requirements are being met.
[0111] Noise: Ventilation systems generate noise during operation, especially fans and air handling equipment. The noise level has an important impact on the comfort and health of the indoor environment and needs to be measured and evaluated to see if it exceeds the specified limit standards.
[0112] Energy consumption: The energy consumption of a ventilation system is an important indicator for evaluating its economic performance. It is usually evaluated by measuring the power and efficiency of the fan.
[0113] Step 502: After receiving the maintenance instruction, the ventilation system is inspected and maintained. If there is a fault in the ventilation system, the location and time of the fault are recorded, and the fault data set of the ventilation system is generated after being summarized;
[0114] Generate fault density from fault data set , based on the fault density Evaluate the frequency of ventilation system failures and confirm the performance stability of the ventilation system in the following ways:
[0115] ;
[0116] in, is the number of ventilation system failures, It is a malfunction i To failure j The shortest distance, is the average distance, It is a malfunction i To failure j The time interval is the average value of the time interval, and the weight coefficient is: , ;
[0117] Based on historical data and management expectations of ventilation system performance, the density threshold is pre-set; if the fault density obtained is If the density threshold is exceeded, it means that the ventilation system has a high frequency of failures during operation. This may be due to improper layout or other reasons. The ventilation system needs to be reinstalled or replaced. At this time, an alarm command is sent to the outside. After receiving the alarm command, the management personnel adjust the ventilation system or take other treatment measures.
[0118] When using, combine the contents in steps 501 and 502:
[0119] After optimizing the layout of the ventilation system, the ventilation system is inspected and maintained. After marking each component of the ventilation system, if there is a fault in the component of the ventilation system, the fault density is generated based on the fault location and fault time of the component. , based on the fault density Evaluate the frequency of ventilation system failures. If after the layout of the ventilation system is optimized, in order to maintain the current ventilation effect, the frequency of operational failures during ventilation system operation increases, it means that the current ventilation system layout is still insufficient. At this time, the ventilation system layout needs to be optimized and maintained to further ensure the ventilation effect in the tunnel.
[0120] The analytic hierarchy process, or AHP for short, is a decision analysis method that decomposes the elements related to the decision problem into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. The core of this method is to regard a complex multi-objective decision problem as a system, and then decompose the goal into multiple goals or criteria, and then decompose it into several levels of multiple indicators (or criteria, constraints). Through the fuzzy quantification method of qualitative indicators, the hierarchical single ranking (weight) and total ranking can be calculated, thus providing a systematic method for multi-indicator and multi-plan optimization decision-making.
[0121] The construction methods of ventilation system layout optimization knowledge graph can be summarized into three main methods: bottom-up method, top-down method and a hybrid method of the two. You can refer to the following:
[0122] Bottom-up approach:
[0123] Knowledge extraction: Extract entities, attributes and relationships related to ventilation system layout optimization from open linked data sources. This includes automatic or semi-automatic extraction of key information such as ventilation equipment, ventilation parameters, layout requirements, etc. from multi-source data such as text, databases, and expert systems.
[0124] Knowledge fusion: Due to the diversity of data sources, there may be knowledge duplication or inconsistency. Through methods such as entity alignment and attribute value filling, the ambiguity between knowledge from different sources is eliminated to ensure the accuracy and consistency of the integrated knowledge.
[0125] Knowledge processing: abstract the concepts of the constructed data layer and construct the model layer of the knowledge graph. This includes summarizing specific data into higher-level concepts, such as ventilation system type, layout strategy, etc., and evaluating the quality and credibility of the knowledge.
[0126] Top-down method:
[0127] Ontology construction: Starting from the top-level concept of ventilation system layout optimization, the top-level ontology is constructed. Then, according to specific business needs and data conditions, these concepts and their relationships are refined to form a well-structured concept hierarchy tree.
[0128] Data source extraction: Use some data sources to extract ontology, that is, perform ontology learning. This includes extracting key concepts and relationships from existing ventilation system design documents, engineering cases, expert opinions, etc. as the basis of the knowledge graph.
[0129] Model layer construction: Based on the results of ontology construction, the model layer of the knowledge graph is formed. This includes defining the types and constraints of entities, relationships, attributes, and the logical relationships between them.
[0130] A hybrid approach:
[0131] This method combines the advantages of bottom-up and top-down methods. First, the top-level structure and key concepts of the knowledge graph are constructed through the top-down method; then, the bottom-up method is used to extract specific knowledge and details from the actual data and fill them into the knowledge graph.
[0132] In the hybrid approach, the weight and order of the two methods can be flexibly adjusted according to the actual situation. For example, a top-down approach can be used to build a rough framework and key concepts, and then a bottom-up approach can be used to continuously supplement and improve the detailed information.
[0133] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0134] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can 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 this application.
[0135] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0136] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0137] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0138] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0139] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.
[0140] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for optimizing ventilation layout of submarine tunnel intervals based on digital twins, characterized by: include, Collect environmental data in the tunnel and construct a target area condition data set, and generate the ventilation demand coefficient in the target area from the regional condition data set , if the ventilation demand coefficient Exceeding demand expectations, issuing data collection instructions to external parties; After collecting ventilation system data, tunnel structure data and environment data in the target area, the data is used as sample data, and the tunnel ventilation digital twin model is generated by training the sample data. Set different target test scenarios for the ventilation system. Under different target test scenarios, use the tunnel ventilation digital twin model to test the current layout of the ventilation system, and generate effect coefficients from the test data. , if the effect coefficient Without exceeding the effect threshold, layout optimization instructions are issued to the outside; Feature extraction is performed on the test data. Based on the obtained optimization features, the ventilation system layout optimization plan is output from the ventilation system layout optimization knowledge graph. The optimized plan is simulated and tested using the trained tunnel ventilation digital twin model. The optimization degree is generated using the test data. , according to the optimization degree Take corresponding measures; The sensor network collects the operating status data of the ventilation system, and generates the operating status coefficient from the operating status data. , where the response time to the acquisition Energy efficiency ratio Perform linear normalization and map the corresponding data values to the interval Then follow the steps below: ; Weight coefficient: , ; If the operating state coefficient If the ventilation system is lower than expected, the system will be inspected and maintained, and the fault density will be generated from the fault data of the ventilation system. , if the obtained fault density When the density threshold is exceeded, an alarm command is issued to the outside.
2. The method for optimizing ventilation layout of an undersea tunnel according to claim 1, characterized in that: Collect environmental data and traffic flow data in the target area, and aggregate the above data to build a target area condition data set; Generating ventilation demand coefficients from regional condition data sets , where the pollutant concentration and traffic volume After linear normalization, the corresponding data values are mapped to the interval According to the following formula: ; in, In the target area i The pollutant concentration at each monitoring node is For the i The traffic flow at each monitoring node is is the weight coefficient, .
3. The method for optimizing ventilation layout of submarine tunnel according to claim 1, characterized in that: After receiving the data collection instruction, data collection is performed, including: performance data of the ventilation system, tunnel size and structure, and the acquired data is combined with the data in the target area condition data set, and after data preprocessing, the first model data set is summarized and constructed; Based on the first model data set, simulation analysis software is used to perform fluid dynamics simulation to obtain air flow, temperature distribution and pollutant diffusion data in the target area, which are summarized as the second model data set.
4. The method for optimizing ventilation layout of an undersea tunnel according to claim 3 is characterized in that: Different target test scenarios are set for the ventilation system in the target area. In the target test scenarios, the ventilation system is tested using the trained tunnel ventilation digital twin model to obtain the corresponding test data. The obtained simulation test data are aggregated to construct a ventilation test data set, and the effect coefficient is generated from the test data in the ventilation test data set. .
5. The method for optimizing ventilation layout of an undersea tunnel according to claim 4, characterized in that: The temperature gradients in different sub-areas of the target area are obtained. and wind energy density After that, linear normalization is performed on both of them, and the corresponding data values are mapped to the interval The effect coefficient is constructed according to the following formula : ; Weight coefficient: , , is the number of sub-regions; For the i The temperature gradient in each sub-region is is the mean value of the temperature gradient, is the variance of the temperature gradient; For the i The wind energy density in each sub-region is is the mean value of wind energy density, is the variance of wind energy density.
6. The method for optimizing ventilation layout of an undersea tunnel according to claim 5, characterized in that: After receiving the layout optimization instruction, various test data in the ventilation test data set are obtained, and features are extracted from the test data to obtain corresponding optimization features; Taking the optimization of ventilation system layout in tunnel as the target word, a ventilation system layout optimization knowledge graph is pre-constructed; taking the optimization features as input, according to the correspondence between the optimization features and the ventilation system layout optimization scheme, the ventilation system layout optimization scheme is output from the ventilation system layout optimization knowledge graph to optimize the current layout scheme of the ventilation system.
7. The method for optimizing ventilation layout of an undersea tunnel according to claim 6, characterized in that: Taking the ventilation system layout optimization plan as input, the trained tunnel ventilation digital twin model is used to simulate the ventilation system layout optimization in the target area, and the corresponding effect coefficients are obtained at several test nodes. Then, according to the effect coefficient Changes in build optimization , if the optimization degree Without exceeding the preset optimization threshold, a solution optimization instruction is issued to the outside.
8. The method for optimizing ventilation layout of an undersea tunnel according to claim 7, characterized in that: The test nodes before and after the ventilation system layout optimization plan are sorted according to the timeline and correspond one by one. The optimization degree is constructed according to the following method : ; in, For the i The optimized intermediate value of the test nodes, is its mean value, , is the number of test nodes, and They are the ventilation system layout optimization schemes before and after optimization. i The effect coefficient of the test node, and is the corresponding mean.
9. The method for optimizing ventilation layout of an undersea tunnel according to claim 1, characterized in that: After the ventilation system enters the operating state, the sensor network in the target area collects the operating state data of the ventilation system, and then aggregates and constructs an operating state data set; Constructing the operating status coefficient from the operating status data of the ventilation system If the obtained operating status coefficient is lower than the preset performance threshold, a maintenance instruction is issued to the outside.
10. The method for optimizing ventilation layout of an undersea tunnel according to claim 9, characterized in that: After receiving the maintenance instruction, the location and time of the ventilation system fault are recorded, and the fault data set of the ventilation system is generated after being summarized; the fault density is generated from the fault data set , as follows: ; in, is the number of ventilation system failures, It is a malfunction i To failure j The shortest distance, is the average distance, It is a malfunction i To failure j The time interval is the average value of the time interval, and the weight coefficient is: , .
