An Adaptive Control Method for Rapidly Deployed Emergency Rescue Training Containers for Chemical Oil and Gas Tank Fires

Abstract

This invention discloses an adaptive control method for a rapidly deployable emergency rescue training cabin for chemical and oil / gas tank fires, belonging to the field of emergency rescue technology. The method includes the following steps: Step S1: Multi-source sensing data acquisition: Through a sensor network deployed inside and outside the training cabin, multi-dimensional environmental parameters and fire source status parameters of the training scenario are collected in real time; the multi-dimensional environmental parameters include the ambient temperature, thermal radiation intensity, smoke concentration, wind direction, and wind speed of the training area; the fire source status parameters include fire source power, flame height, and temperature distribution of the combustion area. This invention aims to address the significant shortcomings of existing emergency rescue training equipment for chemical and oil / gas tank fires in terms of scenario adaptability, intelligent control, and rapid deployment capabilities.

Description

Technical Field

[0001] This invention relates to the field of emergency rescue technology, and in particular to an adaptive control method for a rapidly deployable emergency rescue training container for chemical oil and gas tank fires. Background Technology

[0002] Chemical oil and gas tanks (including crude oil storage tanks, refined oil storage tanks, liquefied hydrocarbon spherical tanks, etc.) are core storage facilities in the petrochemical industry. Their stored media possess dangerous characteristics such as flammability, explosiveness, volatility, and toxicity, making chemical oil and gas tank fires particularly unique and extremely dangerous. Tank fires, especially those in large tanks, involve high temperatures, strong heat radiation, and wide coverage areas. Once a major leak and explosion occurs, a single tank fire can trigger a chain reaction of fires, including multiple tank fires, surface pool fires, flowing fires, fires in adjacent facilities, and water pollution, resulting in casualties, significant property damage, environmental pollution, and international repercussions. Various simulation training devices and systems have emerged in the existing technology. For example, existing technologies disclose fire extinguishing training systems based on display devices, which simulate the fire extinguishing process through image display and sensing equipment, capable of simulating complex and ever-changing real fire extinguishing scenarios. Other technologies disclose foam fire extinguishing training devices, which actively control the fire source power through a fire source control program that couples the interaction between the extinguishing agent and the oil pool fire, dynamically simulating the fire extinguishing process. However, the above-mentioned existing technologies still have the following shortcomings in practical applications:

[0003] In existing technologies, current emergency rescue training equipment for chemical and oil / gas tank fires lacks the ability to adaptively adjust scenarios specific to the characteristics of oil / gas tank fires. Chemical and oil / gas tank fires have unique characteristics such as large fire source area, high heat radiation intensity, rapid fire development, and long combustion duration. These characteristics place completely different demands on the parameters of training equipment, such as fire source power, fire pattern, and temperature distribution, compared to ordinary building fires. Existing training devices mostly use preset fixed fire source power and combustion parameters, making it difficult to dynamically adjust training scenario parameters according to changes in the training object (such as different tank types, different fire types, and different fire extinguishing tactics) and the training environment (such as ambient temperature, wind direction, and wind speed). This results in insufficient matching between the training scenario and real oil / gas tank fires. The control systems of existing training equipment are mostly open-loop or semi-open-loop structures, lacking a closed-loop adaptive control mechanism based on multi-source sensing data. This shortcoming is prominent in the field of training equipment. During training, there is a lack of real-time evaluation of the training scenario status and intelligent feedback on the trainees' operational responses, making it difficult to achieve dynamic adaptive adjustment of training difficulty and scenario parameters, and failing to effectively support the needs of hierarchical and phased progressive training.

[0004] In summary, existing emergency rescue training equipment for chemical and oil / gas tank fires has significant shortcomings in terms of scenario adaptability, intelligent control, and rapid deployment capabilities. It is difficult to meet the urgent needs of emergency rescue teams in the new era for "high simulation, multiple scenarios, adjustability, and rapid response" in their combat-oriented training. Therefore, developing an emergency rescue training cabin for chemical and oil / gas tank fires that can adaptively adjust training scenario parameters according to training needs and environmental changes, and that has rapid deployment capabilities and a closed-loop control mechanism, along with its adaptive control method, has significant theoretical and engineering application value. Summary of the Invention

[0005] The purpose of this invention is to address the significant shortcomings of existing emergency rescue training equipment for chemical and oily gas tank fires in terms of scenario adaptability, intelligent control, and rapid deployment capabilities. Therefore, this invention proposes an adaptive control method for a rapidly deployable emergency rescue training cabin for chemical and oily gas tank fires.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] An adaptive control method for a rapidly deployable emergency rescue training container for chemical oil and gas tank fires includes the following steps:

[0008] Step S1: Multi-source sensing data acquisition: Through a sensor network deployed inside and outside the training cabin, multi-dimensional environmental parameters and fire source status parameters of the training scene are collected in real time; the multi-dimensional environmental parameters include the ambient temperature, thermal radiation intensity, smoke concentration, wind direction and wind speed of the training area; the fire source status parameters include fire source power, flame height and temperature distribution of the combustion area.

[0009] Step S2: Adaptive modeling of training scenario: Based on the multi-source sensing data, combined with the preset chemical oil and gas tank fire types and fire extinguishing tactical requirements, dynamically generate or match the target fire source power curve, heat radiation distribution model and smoke diffusion model of the current training scenario.

[0010] Step S3: Closed-loop control decision: Compare the real-time collected parameters with the target model, and use an adaptive control algorithm to calculate the control commands for the fire source system, environmental system and assessment system;

[0011] Step S4: Multi-system collaborative execution: According to the control instructions, dynamically adjust the burner power, fuel supply rate and fire pattern of the fire source system, adjust the ventilation volume and temperature and humidity of the environmental system, and update the training scoring criteria of the evaluation system in real time.

[0012] Step S5: Effect Feedback and Iterative Optimization: Collect environmental parameters and fire source status after execution, perform deviation analysis with the target model, and input the deviation value as feedback to step S3 to form a closed-loop adaptive control until the training scene parameters meet the preset accuracy requirements.

[0013] Preferably, the sensor network in step S1 is arranged redundantly, including a wireless infrared thermal imaging sensor, a thermocouple array, a flue gas concentration sensor, and an ultrasonic anemometer; the sensor network has self-testing and fault self-diagnosis functions, and automatically forms a network when the mobile cabin is rapidly deployed.

[0014] Preferably, the preset types of chemical oil and gas tank fires in step S2 include at least tank top fire, tank wall fire, sealing ring fire, flowing fire, and three-dimensional fire; the fire extinguishing tactics requirements include one or more combinations of foam fire extinguishing, dry powder fire extinguishing, and water mist cooling.

[0015] Preferably, the adaptive control algorithm in step S3 adopts fuzzy PID control or model predictive control; the control instructions include the burner start-stop sequence, fuel regulating valve opening degree and ignition timing of the fire source system, the exhaust fan speed and fresh air valve opening degree of the environmental system, and the difficulty level and assessment index weight of the evaluation system.

[0016] Preferably, adjusting the fire pattern of the fire source system in step S4 includes controlling the spread direction, coverage area and pulsation frequency of the flames to simulate the combustion dynamic characteristics of a real chemical oil and gas tank fire.

[0017] Preferably, the training cabin adopts a modular and rapid deployment structure, including independent fire source cabin modules, environmental control cabin modules, and command and assessment cabin modules; the modules are quickly connected through standardized mechanical and communication interfaces, and automatically execute the adaptive control process of steps S1 to S5 after deployment.

[0018] Preferably, the trainees' operational actions, fire extinguishing agent spray trajectory, and movement path are collected through visual sensors and wearable devices, and the behavioral data is matched with the scoring criteria of the evaluation system to dynamically adjust the difficulty of the training scenario and the intensity of the fire source response.

[0019] Preferably, the preset accuracy requirements in step S5 include: the temperature deviation of key points in the training area does not exceed ±10% of the target value, the thermal radiation intensity deviation does not exceed ±15%, and the flue gas concentration deviation does not exceed ±20%; when the deviation exceeds the threshold, the high-speed iterative adjustment in steps S3 to S5 is triggered, and the adjustment cycle is no more than 2 seconds.

[0020] Preferably, the method further includes a safety interlock protection step: when the real-time monitored heat radiation intensity or smoke concentration exceeds the safety threshold, the fire source system is automatically shut off, the smoke exhaust system is fully opened, and an audible and visual alarm is triggered, while maintaining a survivable environment for personnel inside the shelter.

[0021] Preferably, the training cabin is also equipped with a virtual reality fusion module, which is used to synchronously display the actual fire scene parameters and the digital twin model, and provide the commander with a visual auxiliary decision-making interface; the adaptive control method simultaneously corrects the closed-loop control parameters according to the manual intervention instructions input by the commander through the interface.

[0022] Compared with existing technologies, this invention provides an adaptive control method for a rapidly deployable emergency rescue training cabin for chemical oil and gas tank fires, which has the following beneficial effects:

[0023] 1. This invention, through multi-source sensing data acquisition and adaptive modeling of training scenarios, can dynamically generate target fire source power curves, thermal radiation distribution models, and smoke diffusion models that are highly matched with real fires, based on the typical types of chemical oil and gas tank fires and different fire-fighting tactical requirements. Compared with the training devices in the prior art that use preset fixed fire source parameters, this invention overcomes the defect of insufficient matching between training scenarios and real oil and gas tank fires, enabling trainees to train in a highly realistic environment of thermal radiation, smoke diffusion, and dynamic changes in fire intensity, significantly improving the effectiveness of combat training.

[0024] 2. This invention adopts a closed-loop adaptive control architecture: by collecting environmental parameters and fire source status parameters in real time and comparing them with the target model, the adaptive control algorithm is used to calculate the control command and drive the fire source system, environmental system and evaluation system to execute collaboratively. Finally, a closed loop is formed through effect feedback and iterative optimization. This mechanism solves the problem that most existing training equipment is open-loop or semi-open-loop control and lacks dynamic feedback adjustment based on multi-source sensing data. It can automatically correct the fire source power, fire pattern, environmental ventilation and scoring criteria according to the actual response during the training process, realize the dynamic adaptive adjustment of training difficulty and scene parameters, and effectively support the needs of hierarchical and phased progressive training.

[0025] 3. The training cabin in this invention adopts a modular design, including independent fire source cabin modules, environmental control cabin modules, and command and assessment cabin modules. These modules are quickly connected via standardized mechanical and communication interfaces and automatically execute adaptive control processes after deployment. This design overcomes the shortcomings of existing training equipment, such as long scenario transition times, lengthy preparation cycles, and difficulty in flexibly switching between various fire types. It can quickly respond to the needs of different training tasks, significantly improving the mobility and efficiency of training equipment. This invention incorporates safety interlock protection steps: when the real-time monitored heat radiation intensity or smoke concentration exceeds a safety threshold, it automatically executes emergency shutdown of the fire source system, full opening of the smoke exhaust system, and audible and visual alarms, while maintaining a survivable environment for personnel within the cabin. Simultaneously, the sensor network has self-checking and fault self-diagnosis functions, enabling automatic networking and monitoring of its own operational status during rapid deployment. These safety mechanisms effectively prevent secondary accidents caused by uncontrolled fire sources or environmental deterioration during training, ensuring the safety of trainees and equipment. Attached Figure Description

[0026] Figure 1 This is an overall flowchart of the adaptive control method of the rapid deployment emergency rescue training container for chemical oil and gas tank fires according to the present invention.

[0027] Figure 2 This is a schematic diagram of the module composition of a rapid deployment emergency rescue training cabin, which is an adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to the present invention.

[0028] Figure 3 This is a schematic diagram of the sensor network layout for an adaptive control method of a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to the present invention.

[0029] Figure 4 This is an adaptive control closed-loop control block diagram of an adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to the present invention.

[0030] Figure 5 This is a flowchart illustrating the safety interlock protection logic of an adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires, as described in this invention.

[0031] Figure 6 This is a schematic diagram illustrating the rapid switching between multiple training scenarios in an adaptive control method for a rapidly deployable emergency rescue training cabin for chemical oil and gas tank fires, as described in this invention. Detailed Implementation

[0032] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0033] Example 1

[0034] This embodiment provides an adaptive control method for a rapidly deployable emergency rescue training cabin for chemical oil and gas tank fires. The training cabin adopts a modular design, including a fire source cabin module, an environmental control cabin module, and a command and assessment cabin module. The modules are quickly connected via standardized mechanical interfaces and quick-pluggable communication interfaces (such as RJ45 industrial Ethernet interfaces and RS-485 buses). The overall dimensions of the cabin conform to standard container dimensions (6.058m long × 2.438m wide × 2.591m high), facilitating road, rail, or sea transportation. On-site deployment and system self-testing can be completed within 30 minutes.

[0035] I. Acquisition of Multi-Source Sensing Data (Step S1)

[0036] Deploy a sensor network within the training area. Specifically:

[0037] Infrared thermal imaging sensors: Three wireless infrared thermal imagers (model FLIR A70, temperature measurement range -20℃~1200℃) are installed on the top and side walls of the training cabin to collect the heat radiation intensity distribution and flame temperature field of the fire source area in real time.

[0038] Thermocouple array: K-type thermocouples are arranged in a grid pattern (1m spacing) at heights of 0.5m, 1.5m and 2.5m above the ground in the training area, totaling 24 measuring points, to measure the vertical and horizontal distribution of ambient temperature.

[0039] Smoke concentration sensors: Electrochemical smoke sensors (detecting CO, CO2, H2S, etc.) and laser smoke density meters are installed at the smoke outlet and the four corners of the training area to monitor smoke concentration and visibility in real time.

[0040] Ultrasonic anemometer: An integrated ultrasonic anemometer (model LufftWS600) is installed on the top of the outer side of the cabin to collect environmental wind speed and direction data. At the same time, a second anemometer is installed at the inlet of the internal circulation ventilation duct to monitor the internal airflow organization.

[0041] All sensors support the IEEE 802.11ac wireless communication protocol and have self-testing and fault self-diagnosis functions. After the shelter is powered on, each sensor automatically sends a registration signal to the central controller (PLC or embedded industrial computer, Siemens S7-1500) in the command and evaluation module. The central controller automatically constructs a sensor topology map and verifies the validity of the data. When a sensor fails, the system automatically uses data from neighboring sensors for interpolation compensation and issues a replacement prompt on the display interface.

[0042] II. Adaptive Modeling of Training Scenarios (Step S2)

[0043] The central controller incorporates a database of chemical and oil / gas tank fire scenarios. This database includes fire source power curves, heat radiation attenuation models, and smoke generation rate models for typical fire types (tank top fire, tank wall fire, sealing ring fire, flowing fire, and three-dimensional fire) under different environmental conditions (temperature, wind speed). The database is built based on real fire experimental data and CFD simulation results.

[0044] Before training begins, the instructor selects training scenario parameters through the human-computer interaction interface (touchscreen or portable tablet) of the command and evaluation module, for example:

[0045] Fire type: Fire on the sealing ring of a 100,000 cubic meter external floating roof crude oil storage tank;

[0046] Firefighting tactics: Use foam fire suppression combined with water mist cooling of the tank walls;

[0047] Training difficulty level: Intermediate (corresponding to peak fire source power of 2.5MW, peak heat radiation intensity of 8kW / m², and peak smoke concentration of 500ppm).

[0048] Environmental conditions: Initial ambient temperature 25℃, no external wind.

[0049] Based on the above selection, the central controller retrieves the corresponding target fire source power curve P_target(t) (a time function simulating the process of a fire from its occurrence and development to stable combustion), target thermal radiation distribution model Q_target(x,y,z), and target smoke concentration distribution model C_target(x,y,z,t) from the database. Simultaneously, it uses an interpolation algorithm to make the scene parameters in the discrete database continuous, generating a smooth target curve.

[0050] III. Closed-loop control decision-making (step S3)

[0051] The central controller receives the multi-dimensional sensing data acquired in step S1 in real time and calculates the actual fire source power P_actual(t), actual thermal radiation intensity Q_actual(x,y,z), and actual smoke concentration C_actual(x,y,z,t) at the current moment. The actual values ​​are compared with the target values ​​to obtain the deviation e(t) = P_target(t) - P_actual(t) and the spatial deviation distribution of the thermal radiation field and the smoke concentration field.

[0052] This embodiment employs a fuzzy PID control algorithm for regulation decisions. The fuzzy PID controller takes the deviation e and the rate of change of deviation ec as inputs, and tunes the PID parameters Kp, Ki, and Kd online using a preset fuzzy rule table (including membership functions for temperature deviation and rate of change of deviation, and fuzzy inference rules). The controller outputs three sets of regulation commands:

[0053] Ignition source system commands include the opening degree of the fuel regulating valve of the main burner (0~100%), the start-stop sequence of the auxiliary burner, the ignition timing sequence and the flame pulsation frequency;

[0054] Environmental system instructions include exhaust fan speed (0~3000rpm), fresh air electric regulating valve opening (0~100%), air conditioning system cooling / heating mode switching, and air supply angle.

[0055] Evaluation system instructions include the difficulty level adjustment value of the current training scenario and the weight coefficient update value of each assessment indicator (for example, when the actual fire source power lags behind the target value by more than 15%, the weight of the "rapid response" indicator is automatically increased).

[0056] IV. Multi-system collaborative execution (Step S4)

[0057] Fire Source System: The fire source module houses a multi-nozzle gas burner array (using propane or natural gas as fuel, with a maximum total power of 5MW). Each burner nozzle is independently controlled, with fuel regulating valves and air regulating valves driven by stepper motors. The fuel regulating valve opening command output from the central controller is converted from digital to analog (D / A) and then drives the proportional regulating valve with a 4~20mA current signal. Simultaneously, by controlling the start / stop and injection angle of nozzles at different positions, the system simulates the flame spread direction, coverage area, and pulsation frequency. For example, in simulating a sealing ring fire, 12 nozzles arranged circumferentially along the tank wall are ignited sequentially, with an ignition interval of 0.2 seconds between adjacent nozzles, creating the effect of flames spreading around the tank wall.

[0058] Environmental System: The environmental control module includes one centrifugal exhaust fan (15,000 m³ / h), one make-up air fan, and one precision air conditioning unit (20 kW cooling capacity, 15 kW heating capacity). The central controller regulates the exhaust fan speed via a frequency converter, controlling the exhaust rate and static pressure difference in the training area; it also regulates the fresh air valve opening via an electric actuator, controlling the ratio of fresh air to return air, thereby adjusting the smoke concentration and oxygen content. The air conditioning system maintains the average temperature in the training area within the range of 20-30°C (except for areas near fire sources) based on feedback from ambient temperature sensors, ensuring the thermal and humidity comfort of trainees within their protective equipment.

[0059] Evaluation System: The command evaluation module has a built-in evaluation server that updates instructions based on the scoring criteria issued by the central controller, dynamically adjusting the scoring weights of each assessment item. For example, for intermediate training, the initial stage (0-30 seconds) emphasizes the weight of "fire reconnaissance and alarm" (30%), the intermediate stage (30-120 seconds) emphasizes the weight of "cooling and fire extinguishing tactics execution" (50%), and the later stage emphasizes the weight of "preventing reignition and site cleanup" (20%). Scoring data is displayed in real time on the commander's terminal and the trainees' wearable displays.

[0060] V. Results Feedback and Iterative Optimization (Step S5)

[0061] After executing step S4, the central controller again collects the real-time data from step S1 and calculates the deviation between the current actual parameters and the target model. Accuracy requirements are set as follows: temperature deviation at key points in the training area (eight measuring points within a 5m radius and 1.5m height centered on the fire source) ≤ ±10% (e.g., if the target temperature is 400℃, the actual temperature should be between 360~440℃), thermal radiation intensity deviation ≤ ±15%, and flue gas concentration deviation ≤ ±20%. When the deviation is within the allowable range, the system maintains the current control parameters; when the deviation exceeds the limit, steps S3 to S5 are re-executed, with an adjustment cycle of no more than 2 seconds, until the deviation returns to within the threshold.

[0062] If the deviation fails to converge after 5 consecutive cycles, the system will determine that the scenario is abnormal, automatically reduce the power of the fire source to a safe value (not exceeding 30% of the maximum power), and issue an audible and visual alarm to the commander, prompting him to check the fuel supply, fan operation, or sensor status.

[0063] VI. Specific Implementation of Additional Features

[0064] (1) Trainer behavior recognition

[0065] Four industrial-grade depth cameras (Intel RealSense D455) were installed around the training cabin, and wearable inertial measurement units (IMUs, model MPU9250) and UWB positioning tags were integrated into the trainees' fire helmets, firefighting gloves, and fire boots. The cameras captured human skeletal movements (such as gun-holding posture, direction of movement, and bending over to dodge), the IMUs captured arm movement trajectories and the aiming point of the sprayed extinguishing agent, and the UWB tags provided three-dimensional spatial position (accuracy ±10cm). The central controller compared the above data with the preset standard operating procedure (SOP) and calculated the action deviation value. For example, if the trainee did not activate the fixed water cannon to cool the adjacent tank wall within 10 seconds of the fire starting, the evaluation system automatically deducted the "cooling timeliness" score, and dynamically adjusted the intensity of the subsequent fire source (reducing the power by 5% to simulate the degree of fire spread caused by operational delay in a real scenario).

[0066] (2) Safety interlock protection steps

[0067] The system monitors heat radiation intensity and smoke concentration in real time. Safety thresholds are set as follows: heat radiation intensity exceeding 15 kW / m² (for 5 seconds) or smoke concentration exceeding 1000 ppm (for 10 seconds). If these limits are exceeded, the central controller directly cuts off the power to the main fuel solenoid valve and ignition transformer of the fire source system via a hardware safety relay. Simultaneously, it switches the exhaust and makeup air fans to full-speed operation, activates the cabin spray cooling system (pre-action water spray, flow rate 60 L / min), and triggers the audible and visual alarms (alarm bell and red flashing light). At the same time, the system broadcasts instructions to trainees to evacuate through the emergency exits on both sides. The safety interlock response time is no more than 200 ms, ensuring a survivable environment for personnel inside the cabin under any circumstances (temperature ≤ 60℃, heat radiation ≤ 2 kW / m², CO concentration ≤ 200 ppm).

[0068] (3) Virtual reality integration and visualization-assisted decision-making

[0069] The command and assessment module is equipped with a graphics workstation (NVIDIA RTX A6000 GPU) running digital twin software (developed based on the Unity 3D engine). This software receives real-time fire source parameters, sensor data, and trainee location data from the central controller, synchronously displaying the actual fire scene status (flame color, smoke diffusion path, thermal radiation cloud map) and the trainee's virtual avatar in a 3D virtual scene. Commanders can select areas and set temporary intervention points in the virtual scene using a stylus or gestures (e.g., "enhance the fire on the left" or "reduce smoke concentration"). The system converts these manual intervention commands into corresponding control commands (e.g., increasing the fuel valve opening of the left burner by 5% or increasing the exhaust fan speed by 200 rpm), which are then fed forward to the PID controller output in step S3, achieving human-machine collaborative adaptive control.

[0070] VII. Examples of switching between multiple training scenarios

[0071] Taking the transition from "sealed ring fire" to "flowing fire" training as an example:

[0072] The instructor selects "Scene Switch → Flowing Fire (Crude Oil Leak, Area 20m²)" on the command interface. The system executes automatically:

[0073] Close the burner nozzle corresponding to the original sealing ring flame, stop the fuel supply, and turn on the exhaust fan to purge for 30 seconds;

[0074] Activate the ground-based flowing fire combustion disc (located in the center of the training area, consisting of 9 independently controlled combustion units, each with a power of 0.3MW);

[0075] The central controller loads the flowing fire target curve from the database (the fire source power rapidly rises to 2.7MW, and the heat radiation is mainly concentrated at a height of 0~1m above the ground).

[0076] Adjust the area of ​​interest of the sensor network (automatically align the infrared thermal imager with the ground area, and focus the thermocouples on the near-ground measurement points).

[0077] Update assessment criteria (focusing on "flowing fire containment and foam coverage" skills).

[0078] The entire switching process is completed automatically within 5 minutes, without the need for manual rearrangement of pipelines or sensors, which greatly improves training efficiency.

[0079] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0080] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0081] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. An adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires, characterized in that, Includes the following steps: Step S1: Multi-source sensing data acquisition: Through a sensor network deployed inside and outside the training cabin, multi-dimensional environmental parameters and fire source status parameters of the training scene are collected in real time; the multi-dimensional environmental parameters include the ambient temperature, thermal radiation intensity, smoke concentration, wind direction and wind speed of the training area; the fire source status parameters include fire source power, flame height and temperature distribution of the combustion area. Step S2: Adaptive modeling of training scenario: Based on the multi-source sensing data, combined with the preset chemical oil and gas tank fire types and fire extinguishing tactical requirements, dynamically generate or match the target fire source power curve, heat radiation distribution model and smoke diffusion model of the current training scenario. Step S3: Closed-loop control decision: Compare the real-time collected parameters with the target model, and use an adaptive control algorithm to calculate the control commands for the fire source system, environmental system and assessment system; Step S4: Multi-system collaborative execution: According to the control instructions, dynamically adjust the burner power, fuel supply rate and fire pattern of the fire source system, adjust the ventilation volume and temperature and humidity of the environmental system, and update the training scoring criteria of the evaluation system in real time. Step S5: Effect Feedback and Iterative Optimization: Collect environmental parameters and fire source status after execution, perform deviation analysis with the target model, and input the deviation value as feedback to step S3 to form a closed-loop adaptive control until the training scene parameters meet the preset accuracy requirements.

2. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 1, characterized in that, The sensor network in step S1 is redundantly arranged, including a wireless infrared thermal imaging sensor, a thermocouple array, a flue gas concentration sensor, and an ultrasonic anemometer; the sensor network has self-testing and fault self-diagnosis functions, and automatically forms a network when the mobile cabin is rapidly deployed.

3. The adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to claim 1, characterized in that, The preset types of chemical oil and gas tank fires in step S2 include at least tank top fire, tank wall fire, sealing ring fire, flowing fire, and three-dimensional fire; the fire extinguishing tactics requirements include one or more combinations of foam fire extinguishing, dry powder fire extinguishing, and water mist cooling.

4. The adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to claim 1, characterized in that, The adaptive control algorithm in step S3 adopts fuzzy PID control or model predictive control; the control commands include the burner start-stop sequence, fuel regulating valve opening degree and ignition timing of the fire source system, the exhaust fan speed and fresh air valve opening of the environmental system, and the difficulty level and assessment index weight of the evaluation system.

5. The adaptive control method for a rapid deployment emergency rescue training cabin for chemical oil and gas tank fires according to claim 1, characterized in that, In step S4, adjusting the fire pattern of the fire source system includes controlling the direction of flame spread, coverage area, and pulse frequency to simulate the dynamic combustion characteristics of a real chemical oil and gas tank fire.

6. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 1, characterized in that, The training cabin adopts a modular and rapid deployment structure, including independent fire source cabin modules, environmental control cabin modules, and command and assessment cabin modules; the modules are quickly connected through standardized mechanical and communication interfaces, and automatically execute the adaptive control process of steps S1 to S5 after deployment.

7. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 1, characterized in that, The system collects the trainees' operational actions, extinguishing agent spray trajectory, and movement path using visual sensors and wearable devices. The behavioral data is then matched with the scoring criteria of the evaluation system to dynamically adjust the difficulty of the training scenario and the intensity of the fire source response.

8. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 1, characterized in that, The preset accuracy requirements in step S5 include: the temperature deviation of key points in the training area does not exceed ±10% of the target value, the thermal radiation intensity deviation does not exceed ±15%, and the flue gas concentration deviation does not exceed ±20%; when the deviation exceeds the threshold, the high-speed iterative adjustment of steps S3 to S5 is triggered, and the adjustment cycle is no more than 2 seconds.

9. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 1, characterized in that, The method also includes a safety interlock protection step: when the real-time monitored heat radiation intensity or smoke concentration exceeds the safety threshold, the fire source system is automatically shut off, the smoke exhaust system is fully opened and an audible and visual alarm is triggered, and the living environment of the personnel in the cabin is maintained.

10. The adaptive control method for a rapid deployment emergency rescue training container for chemical oil and gas tank fires according to claim 9, characterized in that, The training cabin is also equipped with a virtual reality fusion module, which is used to synchronously display actual fire scene parameters with digital twin models and provide commanders with a visual auxiliary decision-making interface; the adaptive control method also corrects the closed-loop control parameters according to the manual intervention instructions input by the commander through the interface.

Images