Piston wind and renewable energy collaborative energy supply ventilation method and system in tunnel

Abstract

The invention discloses a piston wind and renewable energy collaborative energy supply ventilation method and system in a tunnel. The method comprises the following steps that S1, system initialization and multi-source data collection and fusion are conducted; S2, tunnel ventilation requirements in a future time window are calculated through a prediction model based on collected data; s3, aiming at meeting the ventilation requirement, deploying energy according to the power distribution priority, and generating an initial control strategy containing fan operation parameters and an electric energy deploying scheme; s4, the initial control strategy is executed, and the actual ventilation effect in the tunnel is monitored in real time; s5-S7, performing multi-level evaluation and decision feedback, and updating the decision until the actual ventilation effect reaches the standard; and S8, data archiving and model updating: storing a final successful strategy and related data in the control period into a database, and updating and optimizing the prediction model. Intelligent energy allocation of tunnel ventilation is realized, the green energy utilization rate is improved, a multi-level decision feedback mechanism is established, and the control precision is improved.

Description

Technical Field

[0001] This invention patent relates to the field of tunnel engineering technology, specifically to a method and system for ventilation in tunnels that uses piston wind and renewable energy sources for coordinated power supply. Background Technology

[0002] Tunnel ventilation systems are crucial for ensuring air quality within tunnels and controlling smoke during fires. Currently, tunnel ventilation relies primarily on high-powered fans powered by the municipal power grid. This traditional approach has two significant drawbacks: First, fans are major energy consumers, generating high electricity costs and substantial indirect carbon emissions during continuous operation, which is incompatible with environmental protection strategies. Second, traditional ventilation control strategies are often based on fixed schedules or simple concentration thresholds, lacking foresight and prone to "over-ventilation" or "under-ventilation," wasting energy and potentially endangering traffic safety. It is worth noting that the piston effect generated when trains or vehicles travel at high speeds within tunnels creates strong piston winds. This wind energy is enormous, but its kinetic energy has long been uncaptured and utilized, merely being naturally discharged through wind towers, representing a significant energy waste. Therefore, there is an urgent need in this field for an innovative technological solution to transform the current passively energy-consuming ventilation system into a highly efficient system that actively captures and utilizes renewable energy sources such as piston winds, and achieves refined operation through intelligent prediction—a system integrating energy saving, low carbon emissions, and intelligence. Summary of the Invention

[0003] The purpose of this invention is to solve the problems of huge energy consumption, single energy structure, crude control strategy and lack of adaptive capability in tunnel ventilation systems, and to provide a tunnel ventilation method and system that uses piston wind and renewable energy to provide energy in synergy.

[0004] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows: A method for coordinating piston wind and renewable energy supply for ventilation in tunnels includes the following steps: S1: System initialization and multi-source data acquisition and fusion: synchronously acquire environmental data, traffic flow data and energy status data within the tunnel; S2: Ventilation demand prediction. Based on the data collected in step S1, the tunnel ventilation demand in the future time window is calculated using a prediction model. S3: Multi-objective optimization control strategy generation: Taking the ventilation demand as the objective, energy is allocated according to the priority order of piston wind energy in the tunnel > solar and wind energy at the tunnel entrance > energy storage > mains power, and an initial control strategy including fan operating parameters and power allocation scheme is generated. S4: Strategy Execution and Real-time Monitoring: Execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time; S5: Effect evaluation and decision-making, determine whether the actual ventilation effect meets the preset standard; if it does, proceed to step S8; if it does not, proceed to step S6. S6: Primary parameter adjustment, within the preset adjustment range, performs the first optimization adjustment of key parameters in the initial control strategy; S7: Secondary assessment and decision-making, reassess whether the actual ventilation effect after the primary parameter adjustment meets the standard; if it meets the standard, proceed to step S8; if it does not meet the standard, proceed to step S7a; S7a: Strategy reconstruction triggers data re-collection and demand re-forecasting. Based on the updated data and forecast results, the control strategy is regenerated, and then step S7b is executed. S7b: Level 3 assessment and decision-making, determining whether the actual ventilation effect after strategy reconstruction meets the standard; if it meets the standard, proceed to step S8; if it does not meet the standard, proceed to step S7c. S7c: Level 3 emergency adjustment, activate the expert system based on the case library for decision-making, obtain emergency control strategies, and proceed to step S8; S8: Data archiving and model updating. The successful strategies and related data within this control cycle are stored in the database and used to update and optimize the prediction model.

[0005] As a preferred technical solution, in step S1, the environmental data includes CO concentration, NO concentration, and NO concentration. x Concentration and visibility; the traffic flow data includes vehicle volume and average vehicle speed; the energy status data includes piston wind power generation, photovoltaic power generation, and energy storage SOC value.

[0006] As a preferred technical solution, the prediction model is a two-layer LSTM prediction model trained based on historical environmental data and traffic flow data.

[0007] As a preferred technical solution, in step S3, the wind turbine operating parameters include wind turbine speed, start / stop status, and number of operating units; the power allocation scheme is the power supply ratio of different energy sources.

[0008] As a preferred technical solution, the key parameters include the core adjustment items in the fan operating parameters and the power distribution coefficient in the power allocation scheme, and the preset adjustment range is determined based on the tunnel ventilation design standards and equipment operating limits; A ventilation system for tunnels, utilizing piston wind and renewable energy in a coordinated power supply, is provided to implement the aforementioned method. The system comprises: a distributed sensing module, an intelligent decision-making module, an actuator module, and a coordinated power supply module. The distributed sensing module is used to collect real-time tunnel environmental parameters, traffic flow parameters, and the power generation and energy storage status of the renewable energy system. The intelligent decision-making module is used to execute decision-making processes. The actuator module is used to execute ventilation control commands issued by the intelligent decision-making module. The coordinated power supply module supplies power to the actuator module according to a preset priority.

[0009] As a preferred technical solution, the distributed sensing module includes an environmental sensing unit, a traffic sensing unit, and an energy monitoring unit. The environmental sensing unit includes a CO concentration sensor, a NOx concentration sensor, a visibility meter, and an ultrasonic anemometer. The traffic sensing unit includes a microwave vehicle detector and a video recognition subunit. The energy monitoring unit includes a power sensor and a battery management subunit.

[0010] As a preferred technical solution, the intelligent decision-making module includes: The demand forecasting unit is communicatively connected to the distributed sensing module and is used to calculate the tunnel ventilation demand within a future time window through a forecasting model. The strategy generation unit is communicatively connected to the demand prediction unit and is used to generate an initial control strategy by allocating energy according to the priority order of piston wind energy in the tunnel > solar and wind energy at the tunnel entrance > energy storage > mains power, with the goal of meeting ventilation demand. The execution monitoring unit is communicatively connected to the strategy generation unit and is used to execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time. The assessment and adjustment unit is communicatively connected to the execution monitoring unit and is used to determine whether the actual ventilation effect meets the standard, and sequentially perform primary parameter adjustment, strategy reconstruction or level 3 emergency adjustment; The data archiving and model update unit is communicatively connected to the evaluation and adjustment unit, and is used to store the final successful strategy and related data, and to update and optimize the prediction model.

[0011] As a preferred technical solution, the actuator module includes a variable frequency speed control fan group and an intelligent air valve array.

[0012] As a preferred technical solution, the collaborative energy supply module includes a piston wind power generation unit inside the tunnel, a renewable energy unit at the tunnel entrance, an energy storage system, and a smart microgrid manager. The piston wind power generation unit inside the tunnel includes a miniature vertical axis wind turbine generator deployed inside the tunnel, and the renewable energy unit at the tunnel entrance includes monocrystalline silicon solar photovoltaic panels arranged on the slope of the tunnel entrance and a horizontal axis wind turbine generator installed in the open area at the tunnel entrance. The piston wind power generation unit inside the tunnel, the renewable energy unit at the tunnel entrance, and the energy storage system are electrically connected, and the smart microgrid manager is used to execute the power dispatching scheme.

[0013] The beneficial effects of this invention are: This invention discloses a ventilation method and system for tunnels that utilizes piston wind and renewable energy for synergistic energy supply. Based on environmental data, traffic flow data, and energy status data within the tunnel, ventilation requirements are determined. With the goal of meeting these requirements, the system prioritizes the use of piston wind power generation and renewable energy electricity. The system regulates the wind turbine operating parameters and power allocation scheme, and establishes a multi-level decision feedback mechanism. The system can automatically select the optimal processing strategy based on different deviation levels, improving control accuracy by over 40%. Employing an intelligent energy allocation strategy, the green energy utilization rate reaches over 75%, achieving energy savings of over 60% compared to traditional systems.

[0014] The present invention discloses a ventilation method and system for tunnels that utilizes piston wind and renewable energy for synergistic energy supply. Through online updating and optimization of the prediction model, the prediction accuracy of the system improves by about 2% every 30 days of operation, demonstrating continuous evolution capability.

[0015] The system of this invention employs multiple security protection mechanisms to ensure that the system can still maintain basic operation under abnormal conditions such as equipment failure and communication interruption, thereby improving reliability by more than 35%. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of a ventilation method that uses piston wind and renewable energy to provide energy in a tunnel. Figure 2 This is a structural diagram of a ventilation system in a tunnel that uses piston wind and renewable energy for coordinated power supply. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0019] A tunnel ventilation method that utilizes piston wind and renewable energy for synergistic power supply, such as Figure 1 As shown, it includes the following steps: S1: System Initialization and Multi-Source Data Acquisition and Fusion: Synchronously acquire environmental data within the tunnel (including CO concentration, NO...). x The system state vector is generated from data on concentration, visibility, traffic flow (including traffic volume and average vehicle speed) and energy status (including piston wind power generation, photovoltaic power generation, and energy storage SOC value).

[0020] S2: Ventilation Demand Forecasting. Based on the data collected in step S1, the tunnel ventilation demand within a future time window is calculated using a prediction model. Preferably, the future time window is within the next 15 minutes, and the prediction time interval is 3 minutes. The prediction model is a two-layer LSTM prediction model trained based on historical ventilation data and traffic flow data. The two-layer LSTM prediction model has 128 and 64 hidden layer units, respectively, and a dropout rate of 0.2. The model input features include historical environmental data, traffic flow data, and weather forecast data, and the output is the predicted ventilation demand value for the next 15 minutes. The model is automatically updated every 24 hours, and the root mean square error of the prediction is ≤5%.

[0021] S3: Multi-objective Optimization Control Strategy Generation: Based on ventilation demand and current energy status, a preliminary ventilation control strategy and power allocation scheme are generated. With the goal of meeting the ventilation demand, energy is allocated according to the priority order: piston wind power inside the tunnel > solar and wind power at the tunnel entrance > energy storage > mains power. An initial control strategy including fan operating parameters and a power allocation scheme is generated. The fan operating parameters include fan speed, start / stop status, and number of fans in operation. The power allocation scheme is a power supply ratio scheme for different energy sources. The multi-objective optimization includes three objectives: minimizing energy consumption, minimizing carbon emissions, and maximizing ventilation stability. The average solution time on an Intel i7 processor is 8.5 seconds. The multi-objective optimization algorithm uses an improved NSGA-II framework, with a population size of 100, 500 iterations, a crossover probability of 0.9, and a mutation probability of 0.1.

[0022] S4: Strategy Execution and Real-time Monitoring: Execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time; S5: Effect evaluation and decision-making, determine whether the actual ventilation effect meets the preset standard; if it does, proceed to step S8; if it does not, proceed to step S6; specifically, calculate the effect deviation ΔD, when ΔD≤5%, it is determined to meet the standard and jump to S8, otherwise proceed to S6; S6: Primary parameter adjustment. Within a preset adjustment range, the key parameters in the initial control strategy are optimized for the first time. These key parameters include the core adjustment items in the fan operating parameters and the power allocation coefficient in the power distribution scheme. The preset adjustment range is determined based on tunnel ventilation design standards and equipment operating limits; for example, adjusting the control parameter weights within ±15%. If a parameter exceeds the limit, proceed directly to S7a. Primary parameter adjustment involves establishing an adjustable parameter library containing 25 key parameters. Each parameter has an upper limit of 115% of its initial value and a lower limit of 85%. Using a gradient descent method combined with empirical rules, different adjustment step sizes are adopted based on parameter sensitivity analysis results: the adjustment step size for parameters with high sensitivity is ≤5%, and the adjustment step size for parameters with low sensitivity is ≤15%.

[0023] S7: Secondary assessment and decision-making, re-evaluate whether the actual ventilation effect after the primary parameter adjustment meets the standard; if it meets the standard, proceed to step S8; if it does not meet the standard, proceed to step S7a; specifically, calculate the effect deviation ΔD, if ΔD≤8%, jump to S8, otherwise proceed to S7a; S7a: Strategy reconstruction triggers data re-acquisition and demand re-prediction. Based on the updated data and prediction results, a new control strategy is generated, followed by step S7b. Preferably, a robust optimization method is introduced during strategy reconstruction, using interval numbers to describe uncertain parameters, with the interval width dynamically determined based on the prediction confidence level. During reconstruction, the ventilation demand safety factor is increased from 1.0 to 1.1, and the equipment power margin is increased from 10% to 20%, ensuring strategy reconstruction is completed within 3 minutes.

[0024] S7b: Level 3 assessment and decision-making, determining whether the actual ventilation effect after strategy reconstruction meets the standard; if it meets the standard, proceed to step S8; if it does not meet the standard, proceed to step S7c; specifically, if ΔD≤10%, proceed to S8, otherwise proceed to S7c; S7c: Level 3 emergency adjustment, activating an expert system based on a case library for decision-making, obtaining emergency control strategies, and proceeding to step S8; specifically, matching historical solutions with a similarity >85% in the case library or activating the maximum safe airflow mode; the case library stores control strategies corresponding to historical ventilation compliance cases, and the expert system outputs emergency control strategies based on similar scenario matching. Preferably, the case library stores no less than 1000 sets of historical operation cases, case matching uses an improved k-nearest neighbor algorithm, feature weights are determined by the entropy weight method, and the similarity threshold is set to 85%. The rule base contains more than 200 empirical rules, covering various abnormal operating conditions.

[0025] S8: Data Archiving and Model Update. The successful strategies and related data for this control period are stored in the database and used to update and optimize the prediction model. The prediction model update and optimization employs an online incremental learning algorithm, triggering a model update every 24 hours. A sliding time window is used to manage training data, with a window size of 90 days. A mini-batch gradient descent method is used, with a batch size of 32 and an exponentially decaying learning rate, initially set to 0.001.

[0026] The present invention also provides a system for implementing the above method, such as... Figure 2 As shown, it includes: a distributed sensing module, an intelligent decision-making module, an actuator module, and a collaborative energy supply module; the distributed sensing module is used to collect tunnel environmental parameters, traffic flow parameters, and the power generation and energy storage status of the renewable energy system in real time; the intelligent decision-making module is used to execute the decision-making process of the above method; the actuator module is used to execute the ventilation control commands issued by the intelligent decision-making module; the collaborative energy supply module supplies power to the actuator module according to a preset priority.

[0027] The distributed sensing module includes an environmental sensing unit, a traffic sensing unit, and an energy monitoring unit.

[0028] The environmental sensing unit is used to collect tunnel environmental parameters in real time, including CO concentration sensors and NO sensors. x Concentration sensors and visibility meters are installed on the tunnel sidewalls at a height of 3.5m above the ground, with a longitudinal spacing of 50m. The system also includes ultrasonic anemometers (accuracy ±0.1m / s), installed at 1 / 3 and 2 / 3 height of the tunnel cross-section. The CO concentration sensor uses the NDIR principle, with a range of 0-200ppm and an accuracy of ±1ppm. NO... x The concentration sensor uses an electrochemical principle, with a measurement range of 0-50ppm and an accuracy of ±2%; the visibility meter has a measurement range of 5-10000m and an accuracy of ±5%. All sensors are equipped with temperature compensation to ensure measurement accuracy in ambient temperatures ranging from -20℃ to +60℃.

[0029] The traffic sensing unit is used to collect traffic flow parameters in real time, including a microwave vehicle detector and a video recognition subunit. The microwave vehicle detector operates at a frequency of 24.125 GHz and has a detection accuracy of ≥98%. The video recognition subunit uses a 1080P high-definition camera and supports automatic vehicle classification. The data from the two detection devices are mutually verified through a fusion algorithm, maintaining a detection accuracy of over 95% even under adverse weather conditions.

[0030] The energy monitoring unit detects energy status data, including a power sensor and a battery management subunit. The power sensor has an accuracy class of 0.5 and a response time ≤100ms, used to collect real-time data on piston wind power generation and solar power generation. The battery management subunit monitors parameters such as SOC and SOH in real time, with an SOC estimation error ≤3%. The energy monitoring unit is equipped with an ambient temperature sensor for temperature compensation of battery performance.

[0031] All data is transmitted via industrial Ethernet to the intelligent decision-making module for spatiotemporal alignment and fusion, generating a system state vector S(t)=[E(t),T(t),P(t)]. Where: E(t) is the environmental state equivalent, T(t) is the traffic flow state equivalent, and P(t) is the energy state equivalent.

[0032] Furthermore, the intelligent decision-making module includes: The demand forecasting unit is communicatively connected to the distributed sensing module and is used to calculate the tunnel ventilation demand within a future time window through a forecasting model. The strategy generation unit is communicatively connected to the demand prediction unit and is used to generate an initial control strategy by allocating energy according to the priority order of piston wind energy in the tunnel > solar and wind energy at the tunnel entrance > energy storage > mains power, with the goal of meeting ventilation demand. The execution monitoring unit is communicatively connected to the strategy generation unit and is used to execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time. The assessment and adjustment unit is communicatively connected to the execution monitoring unit and is used to determine whether the actual ventilation effect meets the standard. It sequentially performs primary parameter adjustment, strategy reconstruction, or level 3 emergency adjustment. It includes a parameter adjustment subunit, a strategy reconstruction subunit, and an emergency adjustment subunit, which correspond to the primary parameter adjustment, strategy reconstruction, and level 3 emergency adjustment functions, respectively. The data archiving and model update unit is communicatively connected to the evaluation and adjustment unit, and is used to store the final successful strategy and related data, and to update and optimize the prediction model.

[0033] The actuator module includes a variable frequency speed-regulating wind turbine group and an intelligent air valve array. The collaborative energy supply module includes a piston wind power generation unit inside the tunnel, a renewable energy unit at the tunnel entrance, an energy storage system, and an intelligent microgrid manager. The piston wind power generation unit inside the tunnel includes a miniature vertical axis wind turbine generator set deployed inside the tunnel. The miniature vertical axis wind turbine generator set has a starting wind speed ≤1.5m / s, a rated wind speed of 8m / s, a rotor diameter of 0.8m, a single unit rated power of 800W, and a power generation efficiency ≥15%. The generator set casing is made of aluminum alloy with an anodized surface and an IP54 protection rating. The renewable energy unit at the tunnel entrance includes monocrystalline silicon solar photovoltaic panels arranged on the slope of the tunnel entrance and a horizontal axis wind turbine generator installed in the open area at the tunnel entrance. The monocrystalline silicon solar photovoltaic panels have a conversion efficiency ≥21.5%, and the horizontal axis wind turbine generator has a rated power of 5kW, a cut-in wind speed of 3m / s, and adopts a three-bladed upwind design. The piston wind power generation unit inside the tunnel and the renewable energy unit at the tunnel entrance are integrated into the energy storage system through an intelligent combiner box, with a maximum power point tracking efficiency ≥98%. The smart microgrid manager is used to execute power dispatching schemes.

[0034] Furthermore, the system is equipped with a safety protection module, which employs a three-level alarm system: Level 1 alarms are for parameter exceeding limits, Level 2 alarms are for equipment malfunctions, and Level 3 alarms are for system emergencies. Each alarm level corresponds to a different contingency plan, and the contingency plan library contains over 50 standard handling procedures. These multiple safety protection mechanisms ensure that the system can maintain basic operation even in abnormal situations such as equipment failures and communication interruptions, significantly improving reliability.

[0035] The following is a specific example: A method for coordinating piston wind and renewable energy supply for ventilation in tunnels includes the following steps: S1: System initialization and multi-source data acquisition and fusion. A sensing network is deployed within a 1.2km long urban highway tunnel: a CO concentration sensor (range 0-200ppm, accuracy ±1ppm) and a NO... xConcentration sensors (range 0-50ppm, accuracy ±2%) are installed on the tunnel sidewall at a height of 3.5m above the ground, with a longitudinal spacing of 50m; ultrasonic anemometers (accuracy ±0.1m / s) are installed at 1 / 3 and 2 / 3 height of the tunnel cross-section; traffic flow detection uses a microwave vehicle detector (operating frequency 24.125GHz) and a video recognition system; the energy monitoring module collects the power generation of piston wind turbines, photovoltaic power generation, and energy storage SOC value in real time. All data are transmitted to the central controller via industrial Ethernet for spatiotemporal alignment and fusion to generate the system state vector S(t)=[E(t),T(t),P(t)].

[0036] S2: Ventilation Demand Prediction. The state vector is input into a pre-trained two-layer LSTM prediction model (128-64 hidden neurons, dropout=0.2), which outputs a sequence D of ventilation demand at 3-minute intervals for the next 15 minutes. pre =[D(t+3),...,D(t+15)]. In the example, the current traffic flow Q = 1200 vehicles / hour, CO concentration = 35ppm, and the predicted ventilation demand at time t+3 is D(t+3) = 185m³ / s.

[0037] S3: Generation of multi-objective optimization control strategies. Establishment of optimization objectives: min[0.4·P total +0.3·C emission +0.3·(D actual -D pre )²] Among them, P total C represents the total power consumption of the fan and control system. emission To calculate the carbon dioxide emission equivalent based on the amount of electricity consumed, D actual D represents the actual ventilation volume. pre To predict ventilation demand, constraints include SOC ∈ [20%, 90%] and fan speed ∈ [30%, 100%] of rated speed. An improved NSGA-II algorithm (population size 100, 500 iterations) is used to solve the problem, outputting an initial control strategy, including the optimal control sequence U. opt =[75%,82%,...] (fan speed percentage) and energy distribution scheme [P wind 65%, P solar [20%, SOC: 15%], P wind P represents the power supply ratio of the piston wind capture unit inside the tunnel. solar The power supply ratio of the solar and wind power units at the tunnel entrance is represented by SOC, which represents the power supply ratio of the energy storage battery.

[0038] S4: Strategy execution and real-time monitoring.

[0039] Execute the first instruction of the control sequence U opt At (t+3), the speed of fan No. 1 is adjusted to 75% of its rated speed; energy is simultaneously allocated according to the priority order of piston wind power inside the tunnel > solar and wind power at the tunnel entrance > energy storage > mains power. The actual ventilation effect is monitored at a frequency of 5Hz, and the actual air volume D at time t+3 is measured. actual (t+3)=192m³ / s.

[0040] S5: Effect Evaluation and Decision Making.

[0041] The calculated deviation is ΔD = |192 - 185| / 185 × 100% = 3.8%. Since ΔD ≤ 5%, the standard is met, and the process proceeds directly to S8.

[0042] To fully demonstrate the process of this invention, it is assumed that under another operating condition, ΔD = 15.0% > 5%, then proceed to S6.

[0043] S6: Primary parameter adjustment.

[0044] Adjust the control parameters within ±15% of the initial value: adjust the parameter α, which affects the CO concentration weight in the ventilation demand prediction model, from 0.35 to 0.38, and adjust the weight λ1 of the energy consumption target in the multi-objective optimization function from 0.4 to 0.42. Generate the adjusted strategy and return to S4 for execution.

[0045] S7: Secondary assessment and decision-making.

[0046] After parameter adjustment, ΔD' was remeasured and found to be 11.0% > 8%, which still did not meet the standard, so the process proceeded to S7a.

[0047] S7a: Strategy Restructuring.

[0048] The system triggers a data re-acquisition and re-prediction of demand, extending the prediction window from 15 minutes to 25 minutes and recalculating the ventilation demand using a safety factor K=1.1. Based on the new prediction results, the control strategy is reconstructed, and execution returns to S4.

[0049] S7b: Three-level assessment and decision-making.

[0050] After the reconstruction strategy is executed, ΔD'' = 10.5% > 10%, which is still not up to standard, so proceed to S7c.

[0051] S7c: Level 3 emergency adjustment.

[0052] Match historical solutions with 87% similarity to the current state in the case library and adopt the following emergency control strategy: increase the fan speed to 95% and activate the mains auxiliary power supply. After execution, jump to S8.

[0053] S8: Data archiving and model updates.

[0054] Record the final valid data for this cycle. The monthly false alarm rate (FAR) is 13% < 15%, maintaining traffic flow weights; the missed detection rate (MDR) is 8% < 10%, maintaining a safety margin. Start incremental model learning (learning rate 0.001, batch size 32) and update the LSTM model parameters. After completion, return to S1 to begin the next control cycle.

[0055] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A method for coordinating piston wind and renewable energy power supply for ventilation in tunnels, characterized in that: Includes the following steps: S1: System initialization and multi-source data acquisition and fusion: synchronously acquire environmental data, traffic flow data and energy status data within the tunnel; S2: Ventilation demand prediction. Based on the data collected in step S1, the tunnel ventilation demand in the future time window is calculated using a prediction model. S3: Multi-objective optimization control strategy generation: Taking the ventilation demand as the objective, energy is allocated according to the priority order of piston wind energy in the tunnel > solar and wind energy at the tunnel entrance > energy storage > mains power, and an initial control strategy including fan operating parameters and power allocation scheme is generated. S4: Strategy Execution and Real-time Monitoring: Execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time; S5: Effect evaluation and decision-making, determine whether the actual ventilation effect meets the preset standard; if it does, proceed to step S8; if it does not, proceed to step S6. S6: Primary parameter adjustment, within the preset adjustment range, performs the first optimization adjustment of key parameters in the initial control strategy; S7: Secondary assessment and decision-making, reassess whether the actual ventilation effect after the primary parameter adjustment meets the standard; if it does, proceed to step S8; If the standard is not met, proceed to step S7a; S7a: Strategy reconstruction triggers data re-collection and demand re-forecasting. Based on the updated data and forecast results, the control strategy is regenerated, and then step S7b is executed. S7b: Level 3 assessment and decision-making, determining whether the actual ventilation effect after strategy reconstruction meets the standard; if it does, proceed to step S8; If the standard is not met, proceed to step S7c; S7c: Level 3 emergency adjustment, activate the expert system based on the case library for decision-making, obtain emergency control strategies, and proceed to step S8; S8: Data archiving and model updating. The successful strategies and related data within this control cycle are stored in the database and used to update and optimize the prediction model.

2. The method for synergistic energy supply and ventilation of piston wind and renewable energy in tunnels according to claim 1, characterized in that: In step S1, the environmental data includes CO concentration, NO concentration, etc. x Concentration and visibility; the traffic flow data includes vehicle volume and average vehicle speed; the energy status data includes piston wind power generation, photovoltaic power generation, and energy storage SOC value.

3. The method for synergistic energy supply and ventilation of piston wind and renewable energy in tunnels according to claim 1, characterized in that: The prediction model is a two-layer LSTM prediction model trained based on historical environmental data and traffic flow data.

4. The method for coordinating piston wind and renewable energy power supply for ventilation in tunnels according to claim 1, characterized in that: In step S3, the wind turbine operating parameters include wind turbine speed, start / stop status, and number of operating turbines; the power allocation scheme is the power supply ratio of different energy sources.

5. The method for synergistic energy supply and ventilation of piston wind and renewable energy in tunnels according to claim 1, characterized in that: The key parameters include the core adjustment items in the fan operating parameters and the power distribution coefficient in the power allocation scheme. The preset adjustment range is determined based on the tunnel ventilation design standards and the equipment operating limits.

6. A ventilation system for tunnels that combines piston wind with renewable energy for power supply, characterized in that, The method for implementing the coordinated energy supply ventilation method according to any one of claims 1-5 includes: a distributed sensing module, an intelligent decision-making module, an actuator module, and a coordinated energy supply module; the distributed sensing module is used to collect tunnel environmental parameters, traffic flow parameters, and the power generation and energy storage status of the renewable energy system in real time; the intelligent decision-making module is used to execute the decision-making process; the actuator module is used to execute the ventilation control commands issued by the intelligent decision-making module; and the coordinated energy supply module supplies power to the actuator module according to a preset priority.

7. The tunnel ventilation system for combined piston wind and renewable energy supply according to claim 6, characterized in that, The distributed sensing module includes an environmental sensing unit, a traffic sensing unit, and an energy monitoring unit. The environmental sensing unit includes a CO concentration sensor, a NOx concentration sensor, a visibility meter, and an ultrasonic anemometer. The traffic sensing unit includes a microwave vehicle detector and a video recognition subunit. The energy monitoring unit includes a power sensor and a battery management subunit.

8. The tunnel ventilation system for combined piston wind and renewable energy supply according to claim 6, characterized in that, The intelligent decision-making module includes: The demand forecasting unit is communicatively connected to the distributed sensing module and is used to calculate the tunnel ventilation demand within a future time window through a forecasting model. The strategy generation unit is communicatively connected to the demand prediction unit and is used to generate an initial control strategy by allocating energy according to the priority order of piston wind energy in the tunnel > solar and wind energy at the tunnel entrance > energy storage > mains power, with the goal of meeting ventilation demand. The execution monitoring unit is communicatively connected to the strategy generation unit and is used to execute the initial control strategy and monitor the actual ventilation effect in the tunnel in real time. The assessment and adjustment unit is communicatively connected to the execution monitoring unit and is used to determine whether the actual ventilation effect meets the standard, and sequentially perform primary parameter adjustment, strategy reconstruction or level 3 emergency adjustment; The data archiving and model update unit is communicatively connected to the evaluation and adjustment unit, and is used to store the final successful strategy and related data, and to update and optimize the prediction model.

9. The tunnel ventilation system for combined piston wind and renewable energy supply according to claim 6, characterized in that, The actuator module includes a variable frequency speed control fan group and an intelligent air valve array.

10. The tunnel ventilation system for combined piston wind and renewable energy supply according to claim 6, characterized in that, The collaborative energy supply module includes a piston wind power generation unit inside the tunnel, a renewable energy unit at the tunnel entrance, an energy storage system, and a smart microgrid manager. The piston wind power generation unit inside the tunnel includes a miniature vertical axis wind turbine generator deployed inside the tunnel, and the renewable energy unit at the tunnel entrance includes monocrystalline silicon solar photovoltaic panels arranged on the slope of the tunnel entrance and a horizontal axis wind turbine generator installed in the open area at the tunnel entrance. The piston wind power generation unit inside the tunnel, the renewable energy unit at the tunnel entrance, and the energy storage system are electrically connected, and the smart microgrid manager is used to execute the power dispatching scheme.

Images