A method, device, equipment and medium for evacuating people based on flood spreading
By obtaining the static pressure cloud diagram and prediction model of the water flow in the building, combined with finite element simulation and Pix2pix model, the problem of prediction and evacuation of the flow field distribution in the flood spread is solved, and the accurate prediction of the water flow field distribution in the underground space and the pedestrian evacuation exit are achieved. Rapid allocation improves evacuation efficiency and safety.
Patent Information
- Application Number
- CN202410249786.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-05
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-03-05
AI Technical Summary
The existing technology is difficult to effectively predict the distribution of static pressure of underground water flow and guide personnel evacuation in the case of flood proliferation, resulting in imperfect evacuation plans and the inability to deal with changes in flood proliferation in a timely manner.
By obtaining the static pressure cloud diagram of the water flow in the building, input the preset water level change prediction model, calculate the water depth and height, determine the movement speed of the personnel, and use the road search algorithm to guide the personnel to the evacuation exit, combining finite element simulation and Pix2pix model for water flow field Distribution prediction to achieve rapid allocation and guidance for pedestrian evacuation exits.
Accurate prediction of the distribution of water flow fields in underground space is achieved, and the efficiency of distribution and guidance of pedestrian evacuation exits in water invasion is improved, ensuring personnel safety.
Smart Images

Figure CN118278588B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of safe evacuation, and in particular to a method, device, equipment and medium for evacuating people based on flood spreading. Background Art
[0002] Predicting the flood situation in underground space in advance, formulating evacuation strategies in advance based on the prediction results, and guiding pedestrian evacuation are of great significance to saving the lives and property of pedestrians. Summary of the invention
[0003] The purpose of the present invention is to provide a method, device, equipment and medium for evacuating people based on flood spreading, so as to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:
[0004] In a first aspect, the present application provides a method for evacuating people based on flood spreading, comprising:
[0005] Obtain a water flow static pressure cloud map of the target building at an initial moment, wherein the water flow static pressure cloud map is used to characterize the water flow pressure at various locations inside the target building;
[0006] Inputting the water flow static pressure cloud map into a preset water level change prediction model to obtain a water flow static pressure matrix at a target time;
[0007] Calculate the water depth at each location in the target building based on the water flow static pressure matrix;
[0008] Obtaining the target positions of all target persons in the target building;
[0009] Determine the moving speed of each target person based on the target position and the water depth height at each position;
[0010] Determine the target evacuation exit for each target person based on the moving speed, the target position and the path-finding algorithm;
[0011] The target personnel are guided to perform corresponding evacuation operations based on the target evacuation exit.
[0012] In a second aspect, the present application also provides a personnel evacuation device based on flood spreading, comprising:
[0013] A first acquisition unit is used to acquire a water flow static pressure cloud map of a target building at an initial moment, wherein the water flow static pressure cloud map is used to characterize the water flow pressure at various positions inside the target building;
[0014] A first input unit is used to input the water flow static pressure cloud map into a preset water level change prediction model to obtain a water flow static pressure matrix at a target time;
[0015] A first calculation unit, configured to calculate the water depth height at each position in the target building based on the water flow static pressure matrix;
[0016] A second acquisition unit is used to acquire the target positions of all target persons in the target building;
[0017] A first determining unit is used to determine the moving speed of each target person based on the target position and the water depth height at each position;
[0018] A second determining unit, configured to determine a target evacuation exit for each target person based on the moving speed, the target position and a path finding algorithm;
[0019] An evacuation unit is used to guide the target personnel to perform corresponding evacuation operations based on the target evacuation exit.
[0020] In a third aspect, the present application also provides a personnel evacuation device based on flood spreading, including:
[0021] Memory for storing computer programs;
[0022] A processor is used to implement the steps of the method for evacuating people based on flood spread when executing the computer program.
[0023] In a fourth aspect, the present application further provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for evacuating personnel based on flood spread are implemented.
[0024] The beneficial effects of the present invention are:
[0025] The present invention simulates the hydrostatic pressure cloud map of underground space under various working conditions through finite element simulation software, and constructs a one-to-one corresponding data set. The improved pix2pix model is trained and calibrated based on the constructed data set. The trained pix2pix model can accurately predict the flow field distribution of the underground space after water intrusion, and establish a pedestrian evacuation scenario based on the prediction results, ultimately achieving the rapid allocation and guidance of pedestrian evacuation exits in advance under water intrusion conditions.
[0026] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.
[0028] Figure 1 It is a schematic flow chart of a method for evacuating people based on flood spreading described in an embodiment of the present invention;
[0029] Figure 2 A cloud diagram of the static ground pressure distribution of a certain underground space described in an embodiment of the present invention;
[0030] Figure 3 It is a grayscale cloud map of the ground static pressure distribution of a certain underground space described in an embodiment of the present invention;
[0031] Figure 4 It is a schematic diagram of the overall network architecture of the generator described in an embodiment of the present invention;
[0032] Figure 5 Schematic diagram of the model architecture of the discriminator described in an embodiment of the present invention;
[0033] Figure 6 A schematic diagram of generating an evacuation distance based on the A* pathfinding algorithm described in an embodiment of the present invention;
[0034] Figure 7 It is a schematic structural diagram of a personnel evacuation device based on flood spreading described in an embodiment of the present invention;
[0035] Figure 8 It is a schematic diagram of the structure of personnel evacuation equipment based on flood spreading described in an embodiment of the present invention.
[0036] Markings in the figure:
[0037] 10. First acquisition unit; 20. First input unit; 30. First calculation unit; 40. Second acquisition unit; 50. First determination unit; 60. Second determination unit; 70. Evacuation unit; 201. Third acquisition unit; 202. Selection unit; 203. Simulation unit; 204. Correspondence unit; 205. Division unit; 206. Bring-in unit; 207. Verification unit; 208. Test unit; 209. Second calculation unit; 210. Processing unit; 211. Second input unit; 2061. Third input unit; 2062. First expansion unit; 2063. Splicing unit; 2064. Second expansion unit; 2065. First repetition unit; 2066. Composition unit; 2067. First extraction unit; 2068. Second extraction unit; 2069. Fourth input unit; 20610. Adjustment unit; 2091, first inverse processing unit; 2092, second inverse processing unit; 2093, eleventh calculation unit; 2094, twelfth calculation unit; 2095, thirteenth calculation unit; 2096, fifth acquisition unit; 2097, fourteenth calculation unit; 2098, fifteenth calculation unit; 501, third calculation unit; 502, fourth calculation unit; 503, fourth acquisition unit; 504, comparison unit; 505, fifth calculation unit; 506, sixth calculation unit; 601, third determination unit; 602, fourth determination unit; 603, fifth determination unit; 604, sixth determination unit; 605, seventh determination unit; 606, eighth determination unit; 607, seventh calculation unit; 608, eighth calculation unit; 609, ninth calculation unit; 610, tenth calculation unit; 611, second repeating unit;
[0038] 800. Personnel evacuation equipment based on flood spreading; 801. Processor; 802. Memory; 803. Multimedia component; 804. I / O interface; 805. Communication component. DETAILED DESCRIPTION
[0039] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0040] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0041] Embodiment 1:
[0042] This embodiment provides a method for evacuating people based on flood spread.
[0043] See also Figure 1 , the figure shows that the method includes step S10, step S20, step S30, step S40, step S50, step S60 and step S70.
[0044] Step S10. Obtaining a water flow static pressure cloud map of the target building at the initial moment, the water flow static pressure cloud map is used to characterize the water flow pressure at various locations inside the target building;
[0045] Specifically, when a target building begins to suffer from flooding, due to the diversified structure inside the target building, the water depth at each location inside the building is different, resulting in different water flow static pressure at each location. In order to more accurately know the current water flow static pressure at each location inside the target building, it can be measured by installing multiple pressure sensors on the ground inside the building, such as Figure 2 As shown in the figure, it is a static pressure distribution cloud map of a certain underground space. In the static pressure distribution cloud map, different color depths represent different pressure values, which can accurately reflect the changes in the water level height of the underground space, thereby guiding pedestrians to choose the evacuation route.
[0046] Considering that the grayscale cloud image can greatly reduce the model recognition calculation time compared with the RGB three-channel color cloud image, in order to reduce the calculation time, the acquired static pressure data is normalized and drawn into a 256*256 grayscale image, as shown in Figure 3 shown.
[0047] Step S20. Input the water flow static pressure cloud map into the preset water level change prediction model to obtain the water flow static pressure matrix at the target time;
[0048] Specifically, the obtained water flow static pressure cloud map reflects the flood spread at the current moment. The evacuation plan formulated based on this as a reference fails to take into account the subsequent flood spread, resulting in an incomplete evacuation plan. Therefore, the flood spread situation after a certain period of time can be reasonably inferred based on the current flood spread situation, and this can be used as a reference to give priority to people with higher risk factors and ensure their safety.
[0049] Specifically, step S20 specifically includes step S201, step S202, step S203, step S204, step S205, step S206, step S207, step S208, step S209, step S2010 and step S2011:
[0050] Step S201. Obtain multiple preset water flow heights, multiple preset flow rates, and multiple preset water inlets and outlets;
[0051] Step S202. Randomly select any preset water flow height, any preset flow rate and any preset water inlet and outlet to combine and obtain multiple target sets;
[0052] Step S203: simulating the target set based on finite element simulation to obtain a corresponding water flow static pressure data image;
[0053] Step S204. Based on a preset time interval, the water flow static pressure data images are matched one by one to obtain a plurality of data image pairs;
[0054] Specifically, the present application adopts a simulation model to simulate various working conditions, wherein the working condition parameters include the initial water flow height, the initial flow velocity and the water inlet and outlet. The present application adopts 5 working conditions with the initial water flow height of 10 cm, 12 cm, 15 cm, 18 cm and 20 cm respectively, and the initial flow velocity of 1 m / s, 1.2 m / s, 1.5 m / s, 1.8 m / s and 2 m / s respectively, and then the inlets and outlets of the three underground spaces A, B and C are arranged and combined to obtain 7 kinds of water inlet and outlet configurations. Each working condition is calculated for a total of 600 seconds, and each second corresponds to a static pressure cloud map, which is converted into a one-to-one corresponding data set based on a time interval of 300 seconds, that is, 300 pairs of data maps are obtained for each working condition, and the total There are 5*5*7*300 pictures in total. For example, the initial water flow height is 10cm, the initial flow velocity is 1m / s, water intrusion occurs at inlets A and B, and C is the outlet as a set of data, and each set of data contains a static pressure flow field distribution cloud map of the ground inside the target building from 0 seconds to 600 seconds, such as the static pressure flow field distribution cloud map corresponding to the underground space at the 125th second. The flow field distribution cloud map contains the static pressure value distribution of the underground space at different positions at the 125th second under this working condition. The specific number of simulated working conditions and the specific values of each parameter in the working condition can be selected according to the actual situation. This is only for illustration and is not particularly limited. The simulation models include but are not limited to Fluent, CFX and COMSOL.
[0055] Step S205. Divide the data image pairs obtained by simulating all target sets into a plurality of training image pairs, a plurality of verification image pairs and a plurality of test image pairs;
[0056] Step S206. Bring multiple training image pairs into the Pix2pix model for training to obtain a trained initial prediction model;
[0057] Specifically, in the present application, the prediction interval is set to 300 seconds, that is, images between 1 second and 300 seconds are used as conditional images, and images between 301 seconds and 600 seconds are used as target images. The conditional images and the target images correspond one-to-one according to the prediction interval to form image pairs, that is, images of 1 second correspond to images of 301 seconds, and images of 100 seconds correspond to images of 400 seconds. All image pairs are randomly divided into training sets, validation sets, and test sets in proportion, and the image pairs in the training set are input into the Pix2pix model for training, thereby obtaining a trained prediction model.
[0058] Specifically, step S206 specifically includes step S2061, step S2062, step S2063, step S2064, step S2065, step S2066, step S2067, step S2068, step S2069 and step S20610:
[0059] Step S2061. Input the first image in the training image pair into the encoder to perform a layer-by-layer feature extraction operation to generate multiple feature images, wherein the feature images gradually reduce width and height and gradually increase the number of channels based on the generation order;
[0060] Step S2062: Input the last generated feature image into the decoder to perform a size enlargement operation to obtain a size image;
[0061] Step S2063. Splicing operation: splicing the size image with the feature image of the same width, height and number of channels based on the skip connection to obtain a spliced image;
[0062] Step S2064. Enlargement operation: input the spliced image into the decoder to perform a size enlargement operation to obtain an updated size image;
[0063] Step S2065. Repeat the stitching operation and the enlargement operation until the updated size image has the same width, height and number of channels as the first image, and use the updated size image as the generated image;
[0064] Specifically, the Pix2pix network generator uses the U-net network framework. U-Net applies the image enhancement method and can still obtain relatively good accuracy when the number of image pairs is limited. The U-Net structure mainly includes an encoder, a decoder and a jump connection. The overall network architecture of the generator is as follows: Figure 4As shown, the left side is the downsampling operation, that is, the encoder part. After the convolution layer, batch normalization layer (Batch Normalization, BN) and activation function, features are extracted layer by layer, the feature map size is reduced, and the depth is increased. Finally, the width * height * number of channels of the input image size is changed from 256*256*1 to 1*1*512. The right side is the upsampling operation, that is, the decoder part. Each upsampling will double the height and width of the image. The upsampling is performed through the transposed convolution operation to enlarge the feature map size and reduce the depth. The jump connection layer is the part of the middle arrow, which means that the feature map of the corresponding part of the upsampling on the left is centrally cropped to obtain a feature map of the same size as the right, and then spliced with the right to obtain a feature map of 2*2*1024 (the first upsampling layer). Finally, the upsampling of this layer obtains a feature map of size 4*4*512. The upsampling operation is continuously performed, and the size of the output map is 256*256*1, which is consistent with the size of the input map. This will reduce the information loss of the image during the compression process and ensure better generation effect.
[0065] Step S2066. All generated images, first images in all training image pairs, and second images in all training image pairs are respectively formed into a first set, a second set, and a third set;
[0066] Step S2067. Extract one image from the first set and one image from the second set respectively as a first sample pair;
[0067] Step S2068. Extract one image from the second set and one image from the third set respectively as a second sample pair;
[0068] Step S2069. Input the plurality of first sample pairs and the plurality of second sample pairs into the discriminator to obtain a target matrix;
[0069] Specifically, the discriminator in this application uses the PatchGAN network framework. Unlike the traditional GAN discriminant model, which only judges the single probability value of the output image to be true, the output of PatchGAN is a matrix of size N*N. Each element in the output matrix is actually the probability of judging the features at different positions of the image to be true. The authenticity is judged from the features at different positions of the image instead of making an overall evaluation of the entire image. The more "demanding" requirements of the discriminant model can enhance the local texture details of the image generated by the generative model to a certain extent. The model architecture of the discriminator is as follows: Figure 5 shown.
[0070] Step S20610. Adjust the parameters in the Pix2pix model based on the target matrix to obtain an initial prediction model;
[0071] Specifically, the input of the generator is the conditional image in the test image pair, and the output is the generated image. The input of the discriminator is the training image pair or the image pair consisting of the conditional image and the generated image, and the output is a matrix of size NxN, so that the generator and the discriminator can be trained adversarially and optimized together, and finally reach the Nash equilibrium state. That is, when the discriminator cannot judge whether the image is real or fake, the training calibration of the Pix2pix model is completed.
[0072] Step S207: Bring multiple verification image pairs into the initial prediction model for verification, and perform feasibility calculation based on the output prediction results;
[0073] Step S208. When the feasibility of the initial prediction model is greater than a first set threshold, the first image of the plurality of test image pairs is brought into the initial prediction model for testing to obtain a plurality of output prediction images;
[0074] Step S209: Calculate the similarity evaluation index between the predicted image and the second image in the corresponding test image pair to obtain the target index;
[0075] Specifically, the feasibility of the model prediction function is verified based on the divided verification image pairs, and the model that has passed the feasibility verification is tested using test images. The generated prediction images are compared with the target images in the test data pairs, and the prediction accuracy of the model is determined based on the comparison results.
[0076] Specifically, step S209 specifically includes step S2091, step S2092, step S2093, step S2094, step S2095, step S2096, step S2097 and step S2098:
[0077] Step S2091. Perform data inversion processing on the second image in the test image pair to obtain a first static pressure value;
[0078] Step S2092. Perform data inverse processing on the predicted image to obtain a second static pressure value;
[0079] Step S2093. Calculate the difference between the second static pressure value and the first static pressure value to obtain a target difference;
[0080] Step S2094. Calculate the absolute value of the ratio of the target difference to the first static pressure value to obtain a target ratio;
[0081] Step S2095. Calculate the target ratios between the second image and the corresponding predicted image in all test image pairs to obtain a summation result;
[0082] Step S2096. Obtain the number of test sample pairs as the test number;
[0083] Step S2097. Calculate the inverse of the test quantity to obtain the target result;
[0084] Step S2098. Calculate the product of the sum result and the target result to obtain the target indicator;
[0085] Specifically, in order to facilitate image comparison, the generated image can be processed through data inversion to obtain the data form of the static pressure value, and then the mean absolute percentage MAPE is used as the data evaluation index between the generated image and the target image. The calculation formula of the mean absolute percentage MAPE is:
[0086]
[0087] Among them, y i is the actual value of the target image, is the predicted value of the generated image, n is the number of test image pairs; MAPE is the mean absolute percentage.
[0088] Step S2010. When the target index is less than the second set threshold, the initial prediction model is a water level change prediction model;
[0089] Specifically, the target indicator is a percentage. When its value is less than 10, it indicates that the model can well predict the trend of underground water level changes.
[0090] Step S2011. Input the water flow static pressure cloud map into the water level change prediction model to obtain the water flow static pressure matrix at the target time;
[0091] Specifically, the water level change prediction model will predict and output the water flow static pressure cloud map at the target time based on the water flow static pressure cloud map at the current time, and at the same time perform data inverse processing on the predicted water flow static pressure cloud map to obtain the water flow static pressure matrix corresponding to the predicted cloud map.
[0092] Step S30. Calculate the water depth at each location in the target building based on the water flow static pressure matrix;
[0093] Specifically, according to the relationship between pressure and water level, the static pressure matrix can be converted into a water level height matrix, so as to know the water level height at the specific location of the underground space. The water level calculation formula is:
[0094]
[0095] Among them, P is the static pressure of water flow; h is the water level depth; g is the acceleration of gravity; ρ is the density of water;
[0096] At the same time, referring to the geometric structure dimensions of the simulation model, a cellular automaton model is used to construct a two-dimensional underground space evacuation scene under water intrusion. The grid division of this scene refers to the grid division result of the simulation model and is consistent with it. At this time, a ground evacuation scene that matches the number and size of grids in the water level height matrix can be obtained. The data in the water level height matrix is paired one by one with the grids in the evacuation scene to obtain a pedestrian evacuation scene with water level height field attributes.
[0097] Step S40. Obtain the target positions of all target persons in the target building;
[0098] Step S50. Determine the moving speed of each target person based on the target position and the water depth height at each position;
[0099] Specifically, the number and location information of personnel are obtained through the wireless local area network in the venue and the connected mobile devices, and the distribution information is combined with the evacuation scene with water level height field attributes to know the water depth at the location of the personnel. The movement speed of the personnel will be affected by the water depth.
[0100] Specifically, step S50 specifically includes step S501, step S502, step S503, step S504, step S505 and step S506:
[0101] Step S501. Calculate the square of the water depth height corresponding to the target position to obtain a first result;
[0102] Step S502. Calculate the ratio of the first result to the third set threshold value to obtain a second result;
[0103] Step S503. Obtaining a preset critical specific flood force;
[0104] Step S504. Compare the critical specific flood force with the second result. When the second result is less than the critical specific flood force, the moving speed of the target person is the fourth set threshold;
[0105] Step S505. When the second result is greater than the critical specific flood force, the fifth set threshold power of the second result is calculated to obtain a third result;
[0106] Step S506. Calculate the product of the third result and the sixth set threshold value to obtain the moving speed of the target person;
[0107] Specifically, the calculation formula for personnel movement speed is:
[0108]
[0109] Where V is the target personnel moving speed; M is the specific flood force per unit width; M *is the critical specific flood force;
[0110] The formula for calculating the specific flood pressure per unit width is:
[0111]
[0112] Among them, M is the specific flood force per unit width; h is the water level depth.
[0113] Step S60. Determine the target evacuation exit for each target person based on the moving speed, target location and path finding algorithm;
[0114] Specifically, the shortest path from each person to each safety exit is calculated. Considering that the shortest path allocation method usually calculates the straight-line distance from the pedestrian's location to the safety exit when calculating the shortest path, but in reality, there may be obstacles on the evacuation path that hinder the movement of pedestrians, so this method does not conform to objective reality. Therefore, the A* pathfinding algorithm is used in this application to determine the shortest evacuation distance from pedestrians to the safety exit. Its principle is as follows Figure 6 As shown, obstacles are automatically bypassed when searching for the shortest evacuation path.
[0115] Taking into account that when all evacuees choose the nearest exit for evacuation, it is easy to cause uneven utilization of the exits, which increases the overall evacuation time. In order to ensure that the evacuees strive to minimize the evacuation time in the process of moving from the current position to the safe exit, after using the A* pathfinding algorithm to find the shortest evacuation path from pedestrians to each exit, it is necessary to calculate the time for the pedestrian to reach each safe exit based on the improved Togawa evacuation time empirical formula.
[0116] Specifically, step S60 specifically includes step S601, step S602, step S603, step S604, step S605, step S606, step S607, step S608, step S609 and step S6010:
[0117] Step S601. Determine any target person from all target persons as a person to be evacuated;
[0118] Step S602. First determination operation: determine any one exit from all exits as a target exit, and obtain the width of the target exit as a target width;
[0119] Step S603. Second determination operation: determine the shortest evacuation distance between the personnel to be evacuated and the target exit based on the path finding algorithm as the target distance;
[0120] Step S604. The third determination operation: determining the shortest evacuation distance of the remaining target personnel from the target exit based on the path-finding algorithm, and obtaining multiple comparison distances;
[0121] Step S605. Fourth determination operation: determining the number of distances smaller than the target distance from the multiple comparison distances to obtain the target number;
[0122] Step S606. The fifth determination operation: determining the flow rate of the target exit based on the moving speed of the personnel to be evacuated;
[0123] Step S607. First calculation operation: calculate the ratio of the target distance to the moving speed to obtain a fourth result;
[0124] Step S60. Second calculation operation: Calculate the product of the outlet flow rate and the target width to obtain a fifth result;
[0125] Step S608. The third calculation operation: calculate the ratio of the target number to the fifth result to obtain a sixth result;
[0126] Step S609. Fourth calculation operation: Calculate the sum of the fourth result and the sixth result to obtain the evacuation time required for the evacuees to reach the target exit;
[0127] Step S6010. Repeat the first determination operation, the second determination operation, the third determination operation, the fourth determination operation, the fifth determination operation, the first calculation operation, the second calculation operation, the third calculation operation, and the fourth calculation operation until the shortest evacuation time of the personnel to be evacuated does not change, and the target exit corresponding to the shortest evacuation time is the target evacuation exit of the target object;
[0128] Specifically, the improvement of the Togawa evacuation time empirical formula in this application is mainly to consider the impact of different water levels on pedestrian evacuation time. The improved Togawa evacuation time calculation formula is:
[0129]
[0130] Among them, T n is the evacuation time required for the target person to reach exit n; V is the moving speed of the target person; D n N is the shortest evacuation distance from the target person to exit n calculated based on the A* pathfinding algorithm; n is the number of people closer to the nth exit than the target person; C is the exit flow rate; W n is the width of outlet n;
[0131] The formula for calculating the outlet flow rate is:
[0132]
[0133] Among them, V is the moving speed of the target person;
[0134] By comparing the evacuation time required for each pedestrian to reach different safety exits, the exit with the shortest time is considered the safety exit selected by the pedestrian, and the above steps are repeated until each pedestrian no longer changes the exit selection result, that is, the corresponding safety exit is the target evacuation exit.
[0135] Step S70: guiding the target personnel to perform corresponding evacuation operations based on the target evacuation exit;
[0136] Specifically, different exits in the venue need to be marked with different numbers in advance. Through the information transmission between the wireless LAN and mobile devices, the corresponding exit number assigned to each pedestrian will be sent to the personal mobile device for pedestrians to refer to and evacuate independently, so as to improve the efficiency of evacuation, avoid risks in advance, and minimize the loss of personnel and property.
[0137] Embodiment 2:
[0138] like Figure 7 As shown, this embodiment provides a personnel evacuation device based on flood spreading, and the device includes:
[0139] The first acquisition unit 10 is used to acquire a water flow static pressure cloud map of the target building at an initial moment, where the water flow static pressure cloud map is used to characterize the water flow pressure at various locations inside the target building;
[0140] The first input unit 20 is used to input the water flow static pressure cloud map into a preset water level change prediction model to obtain the water flow static pressure matrix at a target time;
[0141] A first calculation unit 30 is used to calculate the water depth height at each position in the target building based on the water flow static pressure matrix;
[0142] A second acquisition unit 40 is used to acquire the target positions of all target persons in the target building;
[0143] A first determining unit 50 is used to determine the moving speed of each target person based on the target position and the water depth height at each position;
[0144] A second determination unit 60 is used to determine a target evacuation exit for each target person based on the moving speed, the target position and the path finding algorithm;
[0145] The evacuation unit 70 is used to guide the target personnel to perform corresponding evacuation operations based on the target evacuation exit.
[0146] In a specific implementation disclosed in the present application, the first input unit 20 includes:
[0147] The third acquisition unit 201 is used to acquire a plurality of preset water flow heights, a plurality of preset flow rates, and a plurality of preset water inlets and outlets;
[0148] A selection unit 202 is used to randomly select any preset water flow height, any preset flow rate and any preset water inlet and outlet for combination to obtain multiple target sets;
[0149] A simulation unit 203 is used to simulate the target set based on finite element simulation to obtain a corresponding water flow static pressure data image;
[0150] A corresponding unit 204 is used to make a one-to-one correspondence between the water flow static pressure data images based on a preset time interval to obtain a plurality of data image pairs;
[0151] A division unit 205 is used to divide the data image pairs obtained by simulating all target sets into a plurality of training image pairs, a plurality of verification image pairs and a plurality of test image pairs;
[0152] The input unit 206 is used to input multiple training image pairs into the Pix2pix model for training to obtain a trained initial prediction model;
[0153] A verification unit 207, used to bring multiple verification image pairs into the initial prediction model for verification, and perform feasibility calculation based on the output prediction results;
[0154] A testing unit 208, configured to, when the feasibility of the initial prediction model is greater than a first set threshold, bring a first image of the plurality of test image pairs into the initial prediction model for testing, and obtain a plurality of output prediction images;
[0155] A second calculation unit 209 is used to calculate a similarity evaluation index between the predicted image and the second image in the corresponding test image pair to obtain a target index;
[0156] The processing unit 210 is used for, when the target index is less than a second set threshold value, the initial prediction model is a water level change prediction model;
[0157] The second input unit 211 is used to input the water flow static pressure cloud map into the water level change prediction model to obtain the water flow static pressure matrix at the target time.
[0158] In a specific embodiment disclosed in the present application, the bringing-in unit 206 includes:
[0159] The third input unit 2061 is used to input the first image in the training image pair into the encoder to perform a layer-by-layer feature extraction operation to generate multiple feature images, wherein the feature images gradually reduce width and height and gradually increase the number of channels based on the generation order;
[0160] The first enlargement unit 2062 is used to input the last generated feature image into the decoder to perform an enlargement operation to obtain a size image;
[0161] The splicing unit 2063 is used for splicing operation: splicing the size image with the feature image with the same width, height and number of channels based on the skip connection to obtain a spliced image;
[0162] The second enlarging unit 2064 is used for enlarging the size of the spliced image into the decoder to obtain an updated size image;
[0163] A first repeating unit 2065 is used to repeat the splicing operation and the enlarging operation until the updated size image has the same width, height and number of channels as the first image, and the updated size image is used as the generated image;
[0164] A forming unit 2066, configured to form all generated images, first images in all training image pairs, and second images in all training image pairs into a first set, a second set, and a third set respectively;
[0165] A first extraction unit 2067, configured to extract an image from the first set and the second set respectively as a first sample pair;
[0166] A second extraction unit 2068 is used to extract an image from the second set and the third set respectively as a second sample pair;
[0167] A fourth input unit 2069 is used to input the plurality of first sample pairs and the plurality of second sample pairs into the discriminator to obtain a target matrix;
[0168] The adjustment unit 20610 is used to adjust various parameters in the Pix2pix model based on the target matrix to obtain an initial prediction model.
[0169] In a specific implementation manner disclosed in the present application, the first determining unit 50 includes:
[0170] The third calculation unit 501 is used to calculate the square of the water depth height corresponding to the target position to obtain a first result;
[0171] The fourth calculation unit 502 is used to calculate the ratio of the first result and the third set threshold value to obtain a second result;
[0172] The fourth acquisition unit 503 is used to acquire a preset critical specific flood force;
[0173] A comparison unit 504 is used to compare the critical specific flood force with the second result. When the second result is less than the critical specific flood force, the moving speed of the target person is a fourth set threshold.
[0174] A fifth calculation unit 505, configured to, when the second result is greater than the critical specific flood force, calculate the second result to the power of a fifth set threshold value to obtain a third result;
[0175] The sixth calculation unit 506 is used to calculate the product of the third result and the sixth set threshold value to obtain the moving speed of the target person.
[0176] In a specific implementation disclosed in the present application, the second determining unit 60 includes:
[0177] The third determining unit 601 is used to determine any target person from all target persons as a person to be evacuated;
[0178] The fourth determining unit 602 is used for a first determining operation: determining any exit from all exits as a target exit, and obtaining a width of the target exit as a target width;
[0179] The fifth determining unit 603 is used for the second determining operation: determining the shortest evacuation distance between the evacuees and the target exit based on the path finding algorithm as the target distance;
[0180] The sixth determining unit 604 is used for the third determining operation: determining the shortest evacuation distance between the remaining target personnel and the target exit based on the path finding algorithm, and obtaining a plurality of comparison distances;
[0181] A seventh determining unit 605 is used for a fourth determining operation: determining the number of distances smaller than the target distance from the multiple comparison distances to obtain the target number;
[0182] An eighth determining unit 606 is used for a fifth determining operation: determining a flow rate of a target exit based on a moving speed of the person to be evacuated;
[0183] The seventh calculation unit 607 is used for the first calculation operation: calculating the ratio of the target distance to the moving speed to obtain a fourth result;
[0184] An eighth calculation unit 608 is used for a second calculation operation: calculating the product of the outlet flow rate and the target width to obtain a fifth result;
[0185] A ninth calculation unit 609, configured for a third calculation operation: calculating a ratio of the target quantity to the fifth result to obtain a sixth result;
[0186] A tenth calculation unit 610 is used for a fourth calculation operation: calculating the sum of the fourth result and the sixth result to obtain the evacuation time required for the evacuees to reach the target exit;
[0187] The second repetition unit 611 is used to repeat the first determination operation, the second determination operation, the third determination operation, the fourth determination operation, the fifth determination operation, the first calculation operation, the second calculation operation, the third calculation operation, and the fourth calculation operation until the shortest evacuation time of the persons to be evacuated no longer changes, and the target exit corresponding to the shortest evacuation time is the target evacuation exit of the target object.
[0188] In a specific implementation disclosed in the present application, the second calculation unit 209 includes:
[0189] A first inverse processing unit 2091 is used to perform data inverse processing on the second image in the test image pair to obtain a first static pressure value;
[0190] A second inverse processing unit 2092 is used to perform data inverse processing on the predicted image to obtain a second static pressure value;
[0191] an eleventh calculating unit 2093, configured to calculate a difference between the second static pressure value and the first static pressure value to obtain a target difference;
[0192] A twelfth calculation unit 2094 is used to calculate the absolute value of the ratio of the target difference value to the first static pressure value to obtain a target ratio;
[0193] A thirteenth calculation unit 2095 is used to sum the target ratios between the second image and the corresponding prediction image in all the test image pairs to obtain a summation result;
[0194] A fifth obtaining unit 2096 is used to obtain the number of test sample pairs as the test number;
[0195] A fourteenth calculation unit 2097, used to calculate the inverse of the test quantity to obtain a target result;
[0196] The fifteenth calculation unit 2098 is used to calculate the product of the sum result and the target result to obtain the target indicator.
[0197] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.
[0198] Embodiment 3:
[0199] Corresponding to the above method embodiment, this embodiment also provides a personnel evacuation device based on flood spread. The personnel evacuation device based on flood spread described below and the personnel evacuation method based on flood spread described above can refer to each other.
[0200] Figure 8 FIG. 8 is a block diagram of a flood-based personnel evacuation device 800 according to an exemplary embodiment. Figure 8 As shown, the personnel evacuation device 800 based on flood spreading may include: a processor 801 and a memory 802. The personnel evacuation device 800 based on flood spreading may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.
[0201] The processor 801 is used to control the overall operation of the personnel evacuation device based on flood spread 800 to complete all or part of the steps in the personnel evacuation method based on flood spread. The memory 802 is used to store various types of data to support the operation of the personnel evacuation device based on flood spread 800, and these data may include, for example, instructions for any application or method operated on the personnel evacuation device based on flood spread 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the flood-based personnel evacuation device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.
[0202] In an exemplary embodiment, the flood-spreading based personnel evacuation device 800 can be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned flood-spreading based personnel evacuation method.
[0203] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned personnel evacuation method based on flood spread are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the personnel evacuation device based on flood spread 800 to complete the above-mentioned personnel evacuation method based on flood spread.
[0204] Embodiment 4:
[0205] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the personnel evacuation method based on flood spread described above can refer to each other.
[0206] A readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the personnel evacuation method based on flood spread of the above method embodiment are implemented.
[0207] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.
[0208] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
[0209] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for evacuating people based on flood spreading, characterized in that: include: Obtain a water flow static pressure cloud map of the target building at an initial moment, wherein the water flow static pressure cloud map is used to characterize the water flow pressure at various locations inside the target building; Inputting the water flow static pressure cloud map into a preset water level change prediction model to obtain a water flow static pressure matrix at a target time; Calculate the water depth at each location in the target building based on the water flow static pressure matrix; Obtaining the target positions of all target persons in the target building; Determine the moving speed of each target person based on the target position and the water depth height at each position; Determine the target evacuation exit for each target person based on the moving speed, the target position and the path-finding algorithm; Guiding the target personnel to perform corresponding evacuation operations based on the target evacuation exit; The water flow static pressure cloud map is input into a preset water level change prediction model to obtain a water flow static pressure matrix at the target time, including: Obtain multiple preset water flow heights, multiple preset flow rates, and multiple preset water inlets and outlets; Randomly select any preset water flow height, any preset flow rate, and any preset water inlet and outlet to combine and obtain multiple target sets; Simulating the target set based on finite element simulation to obtain a corresponding water flow static pressure data image; The water flow static pressure data images are matched one by one based on a preset time interval to obtain a plurality of data image pairs; The data image pairs obtained by simulating all target sets are divided into a plurality of training image pairs, a plurality of verification image pairs and a plurality of test image pairs; Bring multiple training image pairs into the Pix2pix model for training to obtain the trained initial prediction model; Bringing multiple verification image pairs into the initial prediction model for verification, and performing feasibility calculation based on the output prediction results; When the feasibility of the initial prediction model is greater than a first set threshold, bringing a first image of a plurality of test image pairs into the initial prediction model for testing to obtain a plurality of output prediction images; Calculating a similarity evaluation index between the predicted image and the second image in the corresponding test image pair to obtain a target index; When the target indicator is less than a second set threshold, the initial prediction model is the water level change prediction model; Inputting the water flow static pressure cloud map into the water level change prediction model to obtain the water flow static pressure matrix at the target time; Wherein, based on the target position and the water depth height at each position, the moving speed of each target person is determined, including: Calculate the square of the water depth height corresponding to the target position to obtain a first result; Calculating a ratio of the first result to a third set threshold value to obtain a second result; Obtaining a preset critical specific flood force; Comparing the critical specific flood force with the second result, when the second result is less than the critical specific flood force, the moving speed of the target person is a fourth set threshold; When the second result is greater than the critical specific flood force, the second result is raised to the fifth set threshold power to obtain a third result; Calculate the product of the third result and the sixth set threshold to obtain the moving speed of the target person; The calculation formula for personnel movement speed is: Where V is the target personnel moving speed; M is the specific flood force per unit width; M * is the critical specific flood force; The formula for calculating the specific flood pressure per unit width is: Among them, M is the specific flood force per unit width; h is the water level depth.
2. The method for evacuating people based on flood spreading according to claim 1, characterized in that The Pix2pix model includes a generator and a discriminator, and the generator includes an encoder and a decoder; multiple training image pairs are brought into the Pix2pix model for training to obtain an initial prediction model after training, including: Inputting the first image in the training image pair into the encoder to perform a layer-by-layer feature extraction operation to generate a plurality of feature images, wherein the feature images gradually reduce width and height and gradually increase the number of channels based on a generation order; Inputting the finally generated feature image into the decoder to perform a size enlargement operation to obtain a size image; Splicing operation: splicing the size image with the feature image of the same width, height and number of channels based on the skip connection to obtain a spliced image; Enlargement operation: inputting the spliced image into the decoder to perform a size enlargement operation to obtain an updated size image; Repeat the stitching operation and the enlargement operation until the updated size image has the same width, height and number of channels as the first image, and use the updated size image as the generated image; All generated images, first images in all training image pairs, and second images in all training image pairs are respectively formed into a first set, a second set, and a third set; Extracting one image from the first set and one image from the second set respectively as a first sample pair; Extracting one image from the second set and one image from the third set respectively as a second sample pair; Inputting a plurality of first sample pairs and a plurality of second sample pairs into the discriminator to obtain a target matrix; The parameters in the Pix2pix model are adjusted based on the target matrix to obtain the initial prediction model.
3. A personnel evacuation device based on flood spreading, characterized in that: include: A first acquisition unit is used to acquire a water flow static pressure cloud map of a target building at an initial moment, wherein the water flow static pressure cloud map is used to characterize the water flow pressure at various positions inside the target building; A first input unit is used to input the water flow static pressure cloud map into a preset water level change prediction model to obtain a water flow static pressure matrix at a target time; A first calculation unit, configured to calculate the water depth height at each position in the target building based on the water flow static pressure matrix; A second acquisition unit is used to acquire the target positions of all target persons in the target building; A first determining unit is used to determine the moving speed of each target person based on the target position and the water depth height at each position; A second determining unit, configured to determine a target evacuation exit for each target person based on the moving speed, the target position and a path finding algorithm; An evacuation unit, used to guide the target personnel to perform corresponding evacuation operations based on the target evacuation exit; Wherein, the first input unit comprises: A third acquisition unit, used to acquire a plurality of preset water flow heights, a plurality of preset flow rates, and a plurality of preset water inlets and outlets; A selection unit, used for randomly selecting any preset water flow height, any preset flow velocity and any preset water inlet and outlet for combination to obtain multiple target sets; A simulation unit, used for simulating the target set based on finite element simulation to obtain a corresponding water flow static pressure data image; A corresponding unit, used for performing one-to-one correspondence between the water flow static pressure data images based on a preset time interval to obtain a plurality of data image pairs; A division unit, used for dividing the data image pairs obtained by simulating all target sets into a plurality of training image pairs, a plurality of verification image pairs and a plurality of test image pairs; The import unit is used to import multiple training image pairs into the Pix2pix model for training to obtain the trained initial prediction model; A verification unit, used for bringing a plurality of verification image pairs into the initial prediction model for verification, and performing feasibility calculation based on the output prediction results; A testing unit, configured to, when the feasibility of the initial prediction model is greater than a first set threshold, bring a first image of a plurality of test image pairs into the initial prediction model for testing, and obtain a plurality of output prediction images; A second calculation unit, used to calculate a similarity evaluation index between the predicted image and the second image in the corresponding test image pair to obtain a target index; A processing unit, configured to, when the target indicator is less than a second set threshold, use the initial prediction model as the water level change prediction model; A second input unit is used to input the water flow static pressure cloud map into the water level change prediction model to obtain the water flow static pressure matrix at a target time; Wherein, the first determining unit includes: A third calculation unit, used for calculating the square of the water depth height corresponding to the target position to obtain a first result; a fourth calculation unit, configured to calculate a ratio of the first result to a third set threshold value to obtain a second result; A fourth acquisition unit, used to acquire a preset critical specific flood force; A comparison unit, used for comparing the critical specific flood force with the second result, and when the second result is less than the critical specific flood force, the moving speed of the target person is a fourth set threshold; A fifth calculation unit, configured to, when the second result is greater than the critical specific flood force, calculate the second result to the power of a fifth set threshold value to obtain a third result; a sixth calculation unit, configured to calculate the product of the third result and a sixth set threshold value to obtain the moving speed of the target person; The calculation formula for personnel movement speed is: Where V is the target personnel moving speed; M is the specific flood force per unit width; M * is the critical specific flood force; The formula for calculating the specific flood pressure per unit width is: Among them, M is the specific flood force per unit width; h is the water level depth.
4. The personnel evacuation device based on flood spreading according to claim 3 is characterized in that: The Pix2pix model includes a generator and a discriminator, the generator includes an encoder and a decoder, and the input unit includes: A third input unit is used to input the first image in the training image pair into the encoder to perform a layer-by-layer feature extraction operation to generate a plurality of feature images, wherein the feature images gradually reduce width and height and gradually increase the number of channels based on a generation order; A first enlargement unit, used for inputting the last generated feature image into the decoder to perform an enlargement operation to obtain a size image; A splicing unit, used for a splicing operation: splicing the size image with a feature image of the same width, height and number of channels based on a skip connection to obtain a spliced image; A second enlarging unit is used for enlarging the size of the spliced image by inputting it into the decoder to obtain an updated size image; A first repeating unit is used to repeat the stitching operation and the enlargement operation until the updated size image has the same width, height and number of channels as the first image, and the updated size image is used as the generated image; A composition unit, used to respectively form all generated images, first images in all training image pairs, and second images in all training image pairs into a first set, a second set, and a third set; A first extraction unit, configured to extract one image from each of the first set and the second set as a first sample pair; A second extraction unit, configured to extract one image from each of the second set and the third set as a second sample pair; a fourth input unit, configured to input a plurality of first sample pairs and a plurality of second sample pairs into the discriminator to obtain a target matrix; An adjustment unit is used to adjust various parameters in the Pix2pix model based on the target matrix to obtain the initial prediction model.
5. A personnel evacuation device based on flood spreading, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the method for evacuating people based on flood spread as described in any one of claims 1 to 2 when executing the computer program.
6. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for evacuating people based on flood spread as claimed in any one of claims 1 to 2 are implemented.
Citation Information
Patent Citations
Underground space flooding situation rapid deduction method, device, equipment and medium
CN116070549A
Method and device for predicting evacuation capability of evacuation platform
CN117541083A
