A Hazard Source Isolation and Disposal Method for a Lift and Transverse Movement New Energy Stereo Garage

Abstract

The present invention discloses a method for isolating and disposing of hazard sources in a lifting and traversing type new energy stereo garage, including: constructing a two-dimensional fire protection layout mapping model of the stereo garage, and establishing an explosion time prediction constraint equation based on vehicle temperature state parameters; introducing a deep reinforcement learning algorithm to calculate the optimal evacuation path of hazardous vehicles, defining a state space and an action space, and designing a multi-objective reward function that integrates the moving cost of hazardous vehicles and the scheduling cost of obstacle vehicles; evaluating the scheduling difficulty of obstacle vehicles, and using a greedy algorithm to schedule according to the priority of moving costs; realizing the adaptive generation of equipment control instructions through a real-time state feedback mechanism. By using the present invention, the efficiency of isolating and evacuating hazard sources and the system safety can be significantly improved on the premise of ensuring the timeliness of disposal, and the technical bottlenecks of solidified evacuation paths and lagging resource scheduling in the garage can be effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety control of stereo garages, and in particular to a method for isolating and disposing hazardous sources in a lifting and transversely moving new energy stereo garage. Background Art

[0002] With the popularity of new energy vehicles, stereo garages are widely used due to their high space utilization rate. Among them, lifting and sliding stereo garages are famous for their "economical and convenient".

[0003] However, the risk of battery spontaneous combustion in new energy vehicles poses a severe challenge to garage safety. At present, the fire protection system of stereo garages mainly relies on fixed sprinkler fire extinguishing devices and gas explosion suppression systems.

[0004] For example, the Chinese patent document with publication number CN111734183A discloses a three-dimensional parking garage with a fire extinguishing function, which can prevent vehicles in the three-dimensional parking garage from igniting each other and reduce losses. At the same time, water is sprayed through the sprinkler head to cool and extinguish the fire. However, this method fails to fully utilize the space scheduling capacity inside the garage in the early stage of the fire, resulting in the inability to isolate the danger source in a timely and effective manner.

[0005] The Chinese patent document with publication number CN113668921A discloses a charging stereo garage and a supporting fire control method, which detects the temperature of each charging device through a temperature detection device, and judges whether the charging is abnormal based on the detected temperature value. When the temperature is abnormal, the corresponding vehicle loading plate of the abnormal charging device is moved to the fire truck position together with the vehicle for isolation, and the charging device is controlled to be powered off and stopped, and the dry powder spraying device is controlled to spray dry powder on the vehicle in the fire isolation space to extinguish the fire. However, the control method does not mention how to move the vehicle with abnormal temperature to the fire truck position in the most efficient way.

[0006] Therefore, how to quickly evacuate high-risk vehicles within a limited time and effectively dispatch resources for emergency response has become a technical problem that needs to be solved urgently. Summary of the invention

[0007] The present invention provides a method for isolating and disposing hazardous sources in a lifting and transversely moving new energy stereo garage, which can significantly improve the efficiency of hazardous source isolation and evacuation and system safety while ensuring the timeliness of disposal, and effectively solve the technical bottlenecks of solidified garage evacuation paths and delayed resource scheduling.

[0008] A method for isolating and disposing hazardous sources of a lifting and transversely moving new energy stereo garage comprises the following steps:

[0009] (1) Construct a two-dimensional fire protection layout mapping model for a three-dimensional parking garage, collect vehicle temperature state parameters and mark dangerous vehicles, and establish an explosion time prediction constraint equation;

[0010] (2) Define the state space and action space, design a multi-objective reward function that combines the movement cost of dangerous vehicles and the scheduling cost of obstacle vehicles, train a path planning model based on the reinforcement learning algorithm, and calculate multiple evacuation path plans for dangerous vehicles;

[0011] (3) Evaluate the difficulty of obstacle vehicle scheduling, dynamically adjust the positions of obstacle vehicles based on the greedy algorithm to vacate the fire passage, optimize the total scheduling time and meet the hard constraint of the explosion time;

[0012] Select the path with the minimum transfer time from multiple evacuation paths as the optimal evacuation path, and calculate the moving order of obstacle vehicles on the optimal evacuation path;

[0013] (4) Based on the optimal evacuation path and the moving order of obstacle vehicles on the optimal evacuation path, realize the adaptive generation of equipment control instructions through a real-time state feedback mechanism to ensure that dangerous vehicles are transferred to the fire parking space within the explosion time constraint.

[0014] Further, in step (1), the two-dimensional fire layout mapping model maps the three-dimensional space of the stereoscopic garage into a two-dimensional grid plane, and defines the set of parking space coordinates as

[0015] ;

[0016] In the formula, represents the coordinate of the -th parking space, represents the total number of parking spaces;

[0017] Obtain the vehicle temperature of the -th vehicle through a temperature sensor. If , it is marked as a dangerous vehicle, represents the critical temperature threshold; the explosion time prediction constraint equation is defined as:

[0018] ;

[0019] In the formula, represents the explosion time prediction constraint, is the material combustion explosion coefficient.

[0020] Further, in step (2), define the state space , where is the coordinate of an ordinary vehicle; represents the coordinate of the dangerous vehicle, represents the coordinate of the fire parking space; limit the action space to , where represents the lateral movement direction, represents the lifting direction.

[0021] Further, in step (2), a multi-objective reward function that combines the moving cost of the dangerous vehicle and the scheduling cost of the obstacle vehicle is designed, and the formula is:

[0022] ;

[0023] In the formula, is the fixed reward for reaching the fire truck parking space, and it is only when the dangerous vehicle reaches the fire truck parking space , otherwise it is 0. , represent the weight coefficients used to balance the moving cost of the dangerous vehicle and the scheduling cost of the obstacle vehicle. is the actual moving step length of the dangerous vehicle. is the scheduling cost of the obstacle vehicle on the evacuation path. The calculation formula of

[0024] ;

[0025] In the formula, represents the obstacle density parameter, which is used to represent the number of vehicles in the four directions of lifting and traversing of the obstacle vehicle.

[0026] Further, in step (2), based on the deep reinforcement learning algorithm, the path planning model is trained, and the loss function for training is defined as:

[0027] ;

[0028] In the formula, represents the loss value. represents the reward obtained after taking the action for the dangerous vehicle in the state and the next state . is the discount factor. represents the maximum value of all actions in the next state . represents the value of the current state and the action . and are the parameters of the Q-network and the target network of the deep reinforcement learning algorithm respectively;

[0029] The present invention uses the DQN (Deep Q Network) algorithm for training. The DQN model includes a Q network, a target network, and an experience replay unit. The Q network is an agent trained to generate optimal state-action values. Q, that is, the Q-table (a value table), represents the expected reward obtained by taking action a in state s at a certain moment. The environment will feedback the corresponding reward r according to the action of the agent. Therefore, the action that can obtain the maximum reward is selected according to the Q value. The role of the experience replay unit is to interact with the environment and generate data to train the Q network. The target network is used to calculate the maximum reward value that can be obtained by estimating the next state corresponding to this action. The target network and the Q network are exactly the same initially. The experience replay method is used to select samples, and the transfer samples obtained in each iteration are stored in the experience pool. Each time, a part of the data is taken out for training. After a certain number of training rounds are completed, the parameters of the Q network will be synchronized to the target network. The action selection adopts the greedy strategy. At the beginning of training, actions are randomly selected with a high probability to explore more state spaces. As the training progresses, the probability of random selection is gradually reduced, and actions are more dependent on the current Q value.

[0030] Furthermore, the specific process of step (3) is as follows:

[0031] Extract the multiple evacuation paths output in step (2) and screen the set of obstacle vehicles , calculate the distances from the obstacle vehicles to the dangerous vehicle and sort them in ascending order. Based on the principle of giving priority to clearing the obstacle vehicle closest to the dangerous vehicle, calculate the transfer time of the dangerous vehicle under each evacuation path;

[0032] Select the evacuation path with the minimum transfer time as the optimal evacuation path, and calculate the moving order of the obstacle vehicles on the optimal evacuation path through the greedy algorithm.

[0033] Furthermore, calculate the transfer time of the dangerous vehicle under each evacuation path. The formula is:

[0034] ;

[0035] In the formula, is the moving distance for dispatching the obstacle vehicle, is the equipment speed constant, is the dispatching cost of the obstacle vehicle on the evacuation path, is the evacuation time cost of the dangerous vehicle path. The evacuation process needs to meet the following explosion time constraint conditions:

[0036] ;

[0037] In the formula, Indicates the transfer time of dangerous vehicles under each evacuation path. Indicates the prediction constraint of the explosion time.

[0038] Furthermore, in step (4), the adaptive generation of device control instructions is realized through a real-time state feedback mechanism, specifically as follows:

[0039] Parse the optimal evacuation path and the moving order of obstacle vehicles on the optimal evacuation path into a device control instruction sequence, update the real-time state matrix after completing a single instruction, and monitor the moving state of dangerous vehicles until they reach the fire truck parking space.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] By constructing a two-dimensional fire layout mapping model of the stereo garage, combining the deep reinforcement learning algorithm and the greedy scheduling optimization strategy, the present invention realizes efficient isolation of hazard sources and vehicle evacuation. The hazard source isolation and disposal method proposed by the present invention not only improves the safety of the garage, but also optimizes the resource scheduling efficiency, overcoming the deficiencies of the prior art in terms of fixed evacuation paths and lagging scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of the hazard source isolation and disposal method for a lifting and traversing new energy stereo garage according to the present invention.

[0043] Figure 2 It is a schematic diagram of the two-dimensional fire layout mapping in the embodiment of the present invention.

[0044] Figure 3 It is a schematic diagram of multiple evacuation paths calculated by the path planning model in the embodiment of the present invention.

[0045] Figure 4 It is the DQN network architecture of the path planning model in the embodiment of the present invention.

[0046] Figure 5 It is the scheduling flowchart of the greedy algorithm in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0047] The present invention will be further described in detail below with reference to the drawings and embodiments. It should be noted that the following embodiments are intended to facilitate the understanding of the present invention and do not limit it in any way.

[0048] As Figure 1 shown, a hazard source isolation and disposal method for a lifting and traversing new energy stereo garage includes the following steps:

[0049] Step S1, construct a two-dimensional fire layout mapping model of the stereo garage, and define the set of parking space coordinates as:

[0050] ;

[0051] Coordinates of fire truck parking spaces , see the layout of vehicles and fire truck parking spaces in Figure 2 , set the critical temperature threshold for dangerous vehicles .

[0052] When the temperature of the vehicle collected by the temperature sensor exceeds the status parameter, it is marked as a dangerous vehicle, and the combustion explosion coefficient of the battery is obtained by querying the vehicle model , vehicle temperature Establish an explosion time prediction constraint equation:

[0053] ;

[0054] In the formula, is the combustion explosion coefficient of the material.

[0055] Step S2, construct a path planning model based on the Deep Q-Network (DQN), calculate the optimal evacuation path of the dangerous vehicle, and define the state space , where is the coordinate of the ordinary vehicle, which is used as an obstacle vehicle on the evacuation path; the action space is limited to , discrete actions of lateral movement ( direction) or lifting ( direction), considering the movement cost of the dangerous vehicle and the scheduling cost of the obstacle vehicle, define the multi-objective reward function as:

[0056] ;

[0057] In the formula, is the fixed reward for reaching the fire truck parking space, which is only valid when the dangerous vehicle reaches the fire truck parking space , otherwise it is 0, , represent the weight coefficients, which are used to balance the movement cost of the dangerous vehicle and the scheduling cost of the obstacle vehicle, is the actual movement step of the dangerous vehicle, is the scheduling cost of the obstacle vehicle on the evacuation path. In order to estimate the scheduling time of the obstacle vehicle, set the obstacle density parameter , which is used to represent the number of vehicles around the obstacle vehicle (in the four directions of lifting and lateral movement), and evaluate the scheduling difficulty of the vehicle, then The calculation method is:

[0058] .

[0059] As Figure 4 shown, use the DQN network structure (i.e., the deep reinforcement learning algorithm) to train the model, and the loss function is defined as:

[0060] ;

[0061] Wherein, represents the loss value, represents taking an action on the dangerous vehicle in state and the reward obtained after that, and the next state , is the discount factor, represents the maximum value of all actions in the next state , represents the current state and the action of value, and are the parameters of the Q-network and the target network respectively.

[0062] The adopted DQN model includes a Q-network, a target network, and an experience replay unit. The Q-network is an agent trained to generate the optimal state-action value. Q, that is, the Q-table (value table), is the expected return obtained by taking action a in state s at a certain moment. The environment will feedback the corresponding reward r according to the agent's action. Therefore, the action that can obtain the maximum return is selected according to the Q value; the role of the experience replay unit is to interact with the environment and generate data to train the Q-network; the target network is used to calculate the maximum reward value that can be obtained by estimating the next state corresponding to this action. The target network is exactly the same as the Q-network initially. The experience replay method is adopted to select samples, and the transition samples obtained in each iteration are stored in the experience pool. Each time, a part of the data is taken out for training. After a certain number of training rounds are completed, the parameters of the Q-network will be synchronized to the target network. The action selection adopts the greedy strategy. At the beginning of training, actions are randomly selected with a high probability to explore more state spaces. As the training progresses, the probability of random selection is gradually reduced, and actions are more dependent on the current Q value.

[0063] As Figure 3 shown, the top 3 optimal evacuation paths are initially calculated using the path planning model trained in step S2.

[0064] Step S3, the greedy algorithm is used to dynamically adjust the positions of the obstacle vehicles to vacate the fire lane. In order to screen out the optimal evacuation path among them, it is necessary to calculate the scheduling time of the obstacle vehicles on the evacuation path, extract multiple evacuation paths output in the S2 path planning step, and screen the set of obstacle vehicles ; The strategy of the greedy algorithm is to sort according to the priority of the moving cost, and give priority to clearing the obstacle vehicle with the greatest impact (that is, the obstacle vehicle closest to the dangerous vehicle is the top priority for clearing), and calculate the transfer time of the dangerous vehicle in the path case:

[0065] ;

[0066] In the formula, is the moving distance for scheduling the blocking vehicle, is the total cost of obstacle vehicle scheduling, which is used to adjust the impact of scheduling difficulty on the total transfer time of the dangerous vehicle, is the time cost of the dangerous vehicle path evacuation, is the equipment speed constant, selected from a certain model of lifting and traversing stereo garage.

[0067] ;

[0068] The evacuation process needs to meet the explosion time constraint condition:

[0069] ;

[0070] The scheduling process of the greedy algorithm is as Figure 5 shown, and it is calculated that is the most evacuation path. Through the greedy algorithm, it is calculated that on path 3, the obstacle vehicles {(2, 3), (3, 1), (5, 1)} are scheduled in the moving order.

[0071] Step S4, parse the S2 path planning and S3 scheduling optimization results into a device control instruction sequence , and the instruction format is {device ID, action type, target coordinate, timestamp}; update the real-time status matrix after completing a single instruction , and monitor the moving status of the dangerous vehicle until it reaches the fire fighting parking space.

[0072] The above embodiments have described in detail the technical solutions and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the present invention. Any modifications, supplements, and equivalent replacements made within the principle scope of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for isolating and disposing of hazard sources of a lifting and traversing new energy three-dimensional garage, characterized in that, It includes the following steps: (1) Construct a two-dimensional fire layout mapping model of the stereo garage, collect vehicle temperature status parameters and mark dangerous vehicles, and establish an explosion time prediction constraint equation; (2) Define the state space and action space, design a multi-objective reward function that combines the movement cost of dangerous vehicles and the scheduling cost of obstacle vehicles, train a path planning model based on the deep reinforcement learning algorithm, and calculate multiple evacuation path plans for dangerous vehicles; (3) Evaluate the difficulty of scheduling obstacle vehicles, dynamically adjust the positions of obstacle vehicles based on the greedy algorithm to vacate the fire passage, optimize the total scheduling time and meet the hard constraint of the explosion time; Select the path with the minimum transfer time from multiple evacuation paths as the optimal evacuation path, and calculate the moving order of obstacle vehicles on the optimal evacuation path; The specific process includes: Multiple evacuation routes output by the extraction step (2) , and filter the set of obstacle vehicles , calculate the distance from the obstacle vehicle to the dangerous vehicle and sort it in ascending order. Based on the principle of preferentially clearing the obstacle vehicle closest to the dangerous vehicle, calculate the transfer time of the dangerous vehicle under each evacuation route; Select the evacuation path with the minimum transfer time as the optimal evacuation path, and calculate the moving order of obstacle vehicles on the optimal evacuation path through the greedy algorithm; Calculate the transfer time of dangerous vehicles under each evacuation path. The formula is: ； Wherein, is the moving distance of the scheduled obstacle vehicle, is the equipment speed constant, is the scheduling cost of the obstacle vehicle on the evacuation path, is the evacuation time cost of the dangerous vehicle path. The evacuation process needs to satisfy the following explosion time constraint conditions: ； In the formula, represents the transfer time of dangerous vehicles under each evacuation path, represents the explosion time prediction constraint; (4) Based on the optimal evacuation path and the moving order of obstacle vehicles on the optimal evacuation path, realize the adaptive generation of equipment control instructions through the real-time state feedback mechanism to ensure that dangerous vehicles are transferred to the fire parking space within the explosion time constraint.

2. The method for isolating and disposing of hazard sources of the lifting and traversing new energy stereo garage according to claim 1, characterized in that, In step (1), the two-dimensional fire layout mapping model maps the three-dimensional space of the stereo garage into a two-dimensional grid plane, and defines the set of parking space coordinates as ； In the formula, Indicates The coordinates of the parking space, Indicates the total number of parking spaces; Obtain the vehicle temperature of the th vehicle . If , then mark it as a dangerous vehicle. represents the critical temperature threshold. The explosion time prediction constraint equation is defined as: ； In the formula, represents the explosion time prediction constraint, is the material combustion explosion coefficient.

3. The hazard source isolation and disposal method of the lifting and traversing new energy stereo garage according to claim 1, characterized in that, In step (2), define the state space , where is the coordinate of an ordinary vehicle; represents the coordinate of a dangerous vehicle, represents the coordinate of a fire truck parking space; the action space is limited to , where represents the lateral movement direction, represents the lifting direction.

4. The hazard source isolation and disposal method of the lifting and traversing new energy stereo garage according to claim 3, characterized in that, In step (2), design a multi-objective reward function that combines the movement cost of dangerous vehicles and the scheduling cost of obstacle vehicles. The formula is: ； Wherein, is the fixed reward for the dangerous vehicle arriving at the fire truck parking space, and it is only valid when the dangerous vehicle arrives at the fire truck parking space , otherwise it is 0. , represents the weight coefficient, which is used to balance the movement cost of the dangerous vehicle and the scheduling cost of the obstacle vehicle. is the actual movement step of the dangerous vehicle. is the scheduling cost of the obstacle vehicle on the evacuation path. The calculation formula of ； In the formula, represents the obstacle density parameter, which is used to represent the number of vehicles in the four directions of the lifting and horizontal movement of the obstacle vehicle.

5. The hazard source isolation and disposal method for the vertical-lifting and horizontal-shifting new energy stereo garage according to claim 4, characterized in that, In step (2), train a path planning model based on the deep reinforcement learning algorithm. The defined loss function for training is: ； Wherein, represents the loss value, represents taking an action on the dangerous vehicle in the state and obtaining the reward and the next state , is the discount factor, represents the maximum value of all actions in the next state , represents the value of the current state and the action , and are the parameters of the Q-network and the target network of the deep reinforcement learning algorithm, respectively.

6. The method for isolating and disposing of hazard sources of the lifting and traversing new energy three-dimensional garage according to claim 1, characterized in that, In step (4), realize the adaptive generation of equipment control instructions through the real-time state feedback mechanism. Specifically: Parse the optimal evacuation path and the moving order of obstacle vehicles on the optimal evacuation path into a sequence of equipment control instructions, update the real-time state matrix after completing a single instruction, and monitor the moving state of dangerous vehicles until they reach the fire parking space.

Images