CN111311028A - Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm - Google Patents

Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm Download PDF

Info

Publication number
CN111311028A
CN111311028A CN202010207394.3A CN202010207394A CN111311028A CN 111311028 A CN111311028 A CN 111311028A CN 202010207394 A CN202010207394 A CN 202010207394A CN 111311028 A CN111311028 A CN 111311028A
Authority
CN
China
Prior art keywords
evacuation
node
nodes
particle swarm
personnel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010207394.3A
Other languages
Chinese (zh)
Inventor
陈俊风
王家豪
钱煜晖
谢迎娟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changzhou Campus of Hohai University
Original Assignee
Changzhou Campus of Hohai University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changzhou Campus of Hohai University filed Critical Changzhou Campus of Hohai University
Priority to CN202010207394.3A priority Critical patent/CN111311028A/en
Publication of CN111311028A publication Critical patent/CN111311028A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72403—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality
    • H04M1/72406—User interfaces specially adapted for cordless or mobile telephones with means for local support of applications that increase the functionality by software upgrading or downloading
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M1/00—Substation equipment, e.g. for use by subscribers
    • H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
    • H04M1/724—User interfaces specially adapted for cordless or mobile telephones
    • H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
    • H04M1/72457—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to geographic location
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Human Computer Interaction (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Environmental & Geological Engineering (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Alarm Systems (AREA)

Abstract

The invention discloses a large-scale indoor personnel evacuation system based on an improved particle swarm optimization algorithm, which improves the particle swarm optimization algorithm in a swarm intelligence algorithm, optimizes a multi-input multi-output path and provides a real-time updated indoor personnel evacuation path planning system and method by combining various sensor sensing environment information. The invention finds possible emergency in the shortest time by the characteristics of flexibility and high efficiency, provides the most reasonable and most efficient evacuation path for people through any electronic equipment, puts life safety at the head, and can save personal and social losses to the utmost extent before and when the emergency happens.

Description

Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm
Technical Field
The invention relates to a large-scale indoor personnel evacuation system based on an improved particle swarm optimization algorithm, and belongs to the field of swarm intelligent path planning.
Background
With the development of Chinese urbanization and the increasing of people stream density, the scale-to-grade ratio of high-rise buildings and the internal structure of the buildings gradually tend to be diversified and complicated, and the indoor environment area is large and the functions are various. In the face of an emergency, indoor personnel often miss the best escape opportunity because the severity of the event is unknown, or cannot select the best escape route because the personnel are not familiar with the internal structure of the building. The gathered people face the dangerous accidents of sudden and serious events, crowding and trampling, and the like, and even cause irrecoverable life and property loss.
In order to solve the problem of low escape efficiency at present, more and more projects and technicians in recent years pay attention to the research of the optimal crowd evacuation method, and various path planning schemes are proposed to avoid the terrible result of crowded treading caused by judgment errors.
Chinese patent No.: CN201610867890 discloses an indoor multi-exit pedestrian flow simulation method considering evacuation guidance, which focuses on guiding action based on cellular automata, introduces guiding action of guiding personnel into a simulation model, constructs a pedestrian flow simulation model considering multi-exit evacuation guidance conditions, and quantifies guiding acting force in a multi-exit evacuation scene.
Chinese patent No.: CN201810151107 discloses a crowd evacuation method and system, in which the time required for the crowd to evacuate in each civil air defense project is calculated according to the flow of the crowd passing through the entrance of each civil air defense project, then the position of the civil air defense project with the shortest evacuation time is obtained based on the time, and the position is sent to a display terminal for displaying, so that the staff can guide the crowd of other civil air defense projects near the civil air defense project to the civil air defense project for evacuation.
However, the manual guidance of the simulated human behavior greatly consumes evacuation time, and is increasingly complicated and tedious in emergency events such as earthquake and fire. Meanwhile, the shortest time position is obtained by calculating the crowd flow and is displayed on an output terminal until a worker guides the process, so that a large amount of information is delayed. Therefore, the real-time monitoring of the geographic position, the efficient indication of the most reasonable path, the adaptability under different environments and the like are still unsolved problems.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a large indoor personnel evacuation system based on an improved particle swarm optimization algorithm, which improves the particle swarm optimization algorithm in a swarm intelligence algorithm, optimizes a multi-input multi-output path and provides a real-time updated indoor personnel evacuation path planning system and method by combining various sensor sensing environment information.
The invention mainly adopts the technical scheme that:
as shown in fig. 1, a large indoor people evacuation system based on an improved particle swarm optimization algorithm comprises an environment construction module and a planned path module, wherein,
the environment construction module comprises the following specific construction steps:
step 1-1: before entering a large indoor building, a mobile terminal is used for downloading a related software program, the software program allows a current user to access GPS positioning information and update the position information in real time, the software program sends the position information to a processing terminal at intervals, the processing terminal adopts an improved algorithm to perform data processing, the mobile terminal has a voice broadcasting function, and when danger occurs, the mobile terminal performs voice broadcasting for constantly prompting personnel of an individual optimal path;
step 1-2: the method comprises the following steps that a plurality of sensors of different types are arranged at every position in a building at certain intervals and used for predicting different emergencies, detection data of the sensors are updated in real time, the detection data of the sensors and GPS positioning information of crowds are sent to a processing terminal in real time, when a certain numerical value of the sensors exceeds a preset safety threshold value, the processing terminal gives an alarm in advance to remind an evacuation system and workers of safety prejudgment, and evacuation route planning is facilitated before crowd evacuation;
step 1-3: each evacuation node is divided, each evacuation path is composed of a plurality of arc sections, each evacuation node on each floor of the building is simplified into an evacuation source point, the building is connected from the evacuation source point to an exit through the evacuation nodes, new nodes are formed by combining the evacuation nodes to reduce the number of network nodes, and relevant characteristic parameters are correspondingly modified after the new nodes are formed.
The path planning module comprises the following specific construction steps:
step 2-1: establishing a multi-input source point-multi-output evacuation model based on graph theory, and reasonably distributing people of each evacuation source point to different paths so as to minimize the evacuation time of an evacuation network, wherein the model is established by the following relational expression:
Figure BDA0002421606000000021
Figure BDA0002421606000000031
wherein minT is min (maxT)ij) To ensure that the time between node i and node j is minimized,
Figure BDA0002421606000000032
indicating the required evacuation time, C, for node i and node j in different environments resulting in different speedsi(t)≤CmaxiThe capacity threshold value is used for ensuring that the number of people at the node i at the time t cannot exceed the person density of the node;
Figure BDA0002421606000000033
for ensuring that everyone on node i is evacuated, PijMore than or equal to 0 is used to ensure that the people flow value is positive;
step 2-2: an improved particle swarm algorithm is adopted to optimize path planning, in the traditional PSO algorithm, each particle has a certain adaptive value to all targets, the dimension of each target is n, and the position of the particle i after iteration for k times is represented as
Figure BDA0002421606000000034
The velocity is expressed as
Figure BDA0002421606000000035
Optimizing the particles iPosition is denoted as Pi n(k) The optimum position of the entire particle population is denoted as Pn(k) I.e. speed update by Pi n(k) And Pn(k) Are co-performed so that the particles are closer to the optimum position, Pi n(k) The position information is updated as follows:
Figure BDA0002421606000000039
at the same time
Figure BDA00024216060000000310
The speed information is updated as:
Figure BDA00024216060000000311
wherein w is the inertial weight; c. C1And c2Is a learning factor, usually c1＝c2＝2；r1And r2Is [0,1 ]]The random number of (2);
describing the inertia weight by adopting a function for controlling hormone regulation, and then giving the design of the inertia weight w based on the rule of a biological hormone regulation mechanism as follows:
Figure BDA00024216060000000312
wherein, wmaxRepresents the maximum value of the inertial weight; w is aminRepresents the minimum value of the inertia weight; w is a0Representing an initial value of inertia; k represents the current iteration number; k represents a threshold value, and K > 0; q represents the Hill coefficient, Q is more than or equal to 1, and a new particle speed updating formula is provided as follows based on the formula:
Figure BDA0002421606000000041
step 2-3, performing indoor overall path planning based on the improved particle swarm algorithm, and refreshing the GPS information of the personnel at intervals by a software program so as to give an optimal escape route based on the improved particle swarm algorithm in real time;
step 2-4: the specific steps of constructing the bottleneck point detection module aiming at the special period are as follows:
calculating a bottleneck point and a node capacity threshold, identifying a potential bottleneck point limiting exit movement by using linear programming, calculating the node capacity threshold by taking a as a personnel density threshold, and when the personnel density of a node exceeds a, considering that the bottleneck point starts to appear, namely the personnel movement speed is rapidly reduced and even stopped, wherein the specific expression is as follows:
Figure BDA0002421606000000042
each evacuation node
Figure BDA0002421606000000043
Is obtained by linear programming if
Figure BDA0002421606000000044
The node is considered to be a bottleneck point.
2. The large indoor personnel evacuation system based on the improved particle swarm optimization algorithm according to claim 1, wherein the steps 1-3 are preliminarily set for the following two special positions:
(1) when the evacuation nodes in the building channel intersection are connected to the surrounding evacuation intersection nodes, dividing the evacuation nodes into one evacuation node respectively;
(2) due to the irregular flow of people, evacuation nodes at the T-shaped intersection and the crossroad are not merged.
Preferably, the evacuation system is characterized by the following specific evacuation steps:
3-1: the environment construction module collects GPS positioning information, structural composition and entrance and exit information of an indoor large building where the personnel are located;
3-2: setting nodes indoors based on graph theory, and performing addition and deletion modification on nodes of special sections;
3-3: determining correct and wrong evacuation nodes according to the type of the emergency, and primarily determining a feasible route;
3-4: establishing a multi-input source point-multi-output evacuation model by utilizing an improved particle swarm algorithm, and finding out the most appropriate evacuation route;
3-5: pre-judging possible bottleneck points through the model result in the step 3-4;
3-6: judging whether congestion occurs at a certain node, if so, opening other channels which are not commonly used or selecting other nodes closest to the bottleneck point for escape, and then turning to the step 3-4 to re-plan an escape route; otherwise, continuing to the step 3-7;
3-7: and determining an evacuation scheme, finishing the algorithm and outputting an optimal planned path.
Preferably, the solution of the bottleneck point in the steps 2-4 is as follows:
case a: when people are evacuated, if the situation that people jam at an evacuation node occurs, a bottleneck point occurs, an uncommon channel or an employee channel closest to the bottleneck point is considered to be opened so as to reduce the pressure of the people at the node. But because the entrance of the staff is narrow and unsafe, when the bottleneck point is no longer crowded or the density of the staff is less than or equal to a, measures are immediately taken to close the entrance of the staff so as to ensure the safety of the crowd.
Case B: if a bottleneck point appears under the condition that no redundant channel is available in the large building, another unblocked channel closest to the bottleneck point is recommended to be used through a software program, and an escape route is re-planned, so that the escape efficiency is improved.
Preferably, the emergency events in the steps 1-2 include fire, earthquake, theft, robbery and air pollution, and the sensors include a temperature sensor, an ambient light sensor, a humidity sensor, an air pressure sensor, a GPS, a noise sensor, a vibration sensor, a gravity sensor and an air sensitive sensor.
Has the advantages that: the invention provides a large-scale indoor personnel evacuation system based on an improved particle swarm optimization algorithm, which is a multi-source multi-output large-scale indoor personnel evacuation system based on the improved particle swarm optimization algorithm, finds out possible emergency events in the shortest time by virtue of the characteristics of flexibility and high efficiency, provides the most reasonable and most efficient evacuation path for a group through any electronic equipment, puts life safety at the head, and can save personal and social losses to the maximum extent before and when the emergency event occurs. Its advantages are as follows:
(1) the practicability is wide. The system can be applied to various complex large-scale indoor building structures, the more complex the structure is, the better the evacuation effect is, and the less the possibility of bottleneck points is;
(2) reducing casualties and property loss. On one hand, the addition of the sensor can provide possibility for predicting the occurrence of fire, earthquake and other emergencies in advance, thereby not only providing precious time for crowd evacuation, but also enabling a decision maker to make a decision in advance and saving precious property; on the other hand, the efficient evacuation path reduces the evacuation time and the investment of rescuers.
Drawings
Fig. 1 is an overall frame diagram of the evacuation system of the present invention;
FIG. 2 is a graph of different incidents with various sensors of the present invention;
FIG. 3 is one of the architectural structure special location node partitioning diagrams of the present invention;
FIG. 4 is one of the architectural structure special location node partitioning diagrams of the present invention;
FIG. 5 is a structural view of the third floor of the Luo-Pong case building of the present invention;
fig. 6 is a flow chart based on the improved particle swarm optimization algorithm.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
A large indoor personnel evacuation system based on an improved particle swarm optimization algorithm comprises an environment construction module and a planned path module, wherein,
the environment construction module comprises the following specific construction steps:
step 1-1: before entering a large indoor building, electronic equipment (including a mobile terminal) is used for downloading a related software program, the software program allows a current user to access GPS information, updates position information in real time, sends the position information to a processing terminal (computer) at intervals, and carries out data processing by the processing terminal (computer) through an improved algorithm;
step 1-2: arranging a plurality of sensors of different types at certain intervals in each position of the indoor of a building (the arrangement of specific sensors is set by a person skilled in the art according to actual conditions) for predicting different emergencies, updating the detection data of the sensors in real time, and sending the detection data of the sensors and the GPS positioning information of the evacuated crowd to a processing terminal in real time, (equivalently, at certain intervals, the real-time detection data of the sensors of different types and the current position positioning information of each person are packaged and sent to the processing terminal, wherein the information of the sensors is uniformly sent to a base station through radio, the GPS information is determined by GPS satellite positioning and then sent to a computer terminal from the base station and the GPS satellite, the sensor information is only used as a pre-warning of a danger alarm, and the subsequent path planning is only related to the GPS positioning information.) when a certain value of the sensors exceeds a preset safety threshold value, the processing terminal gives an alarm in advance to remind a user of making safety prejudgment, so that the evacuation route planning can be conveniently made before the crowd is evacuated;
step 1.3: each evacuation node is divided, each evacuation path is composed of a plurality of arc sections, each evacuation node on each floor of the building is simplified into an evacuation source point, the building is connected from the evacuation source point to an exit through the evacuation nodes, new nodes are formed by combining the evacuation nodes to reduce the number of network nodes, and relevant characteristic parameters are correspondingly modified after the new nodes are formed.
The path planning module comprises the following specific construction steps:
step 2-1: establishing a multi-input source point-multi-output evacuation model based on graph theory, and reasonably distributing people of each evacuation source point to different paths so as to minimize the evacuation time of an evacuation network, wherein the model is established by the following relational expression:
Figure BDA0002421606000000071
Figure BDA0002421606000000072
wherein minT is min (maxT)ij) To ensure that the time between node i and node j is minimized as much as possible,
Figure BDA0002421606000000073
indicating the required evacuation time, C, for node i and node j in different environments resulting in different speedsi(t)≤CmaxiThe capacity threshold value is used for ensuring that the number of people at the node i at the time t cannot exceed the person density of the node;
Figure BDA0002421606000000074
for ensuring that everyone on node i is evacuated, PijMore than or equal to 0 is used to ensure that the people flow value is positive;
step 2-2: an improved particle swarm algorithm is adopted to optimize a planned path, each particle has a certain adaptive value to all targets in the traditional PSO algorithm, the dimension of each target is n, and the position of the particle i after iteration for k times is represented as
Figure BDA0002421606000000075
The velocity is expressed as
Figure BDA0002421606000000076
The optimal position of the particle i is denoted as Pi n(k) The optimum position of the entire particle population is denoted as Pn(k) Each particle adjusts its speed and position according to its current best position and the best position shared by the group, i.e. the speed update is Pi n(k) And Pn(k) Are co-performed so that the particles are closer to the optimum position, Pi n(k) The position information is updated as follows:
Figure BDA00024216060000000710
at the same time
Figure BDA0002421606000000081
The speed information is updated as:
Figure BDA0002421606000000082
wherein w is the inertial weight; c. C1And c2Is a learning factor, usually c1＝c2＝2；r1And r2Is [0,1 ]]The random number of (2);
describing the inertia weight by adopting a function for controlling hormone regulation, and then giving the design of the inertia weight w based on the rule of a biological hormone regulation mechanism as follows:
Figure BDA0002421606000000083
wherein, wmaxRepresents the maximum value of the inertial weight; w is aminRepresents the minimum value of the inertia weight; w is a0Representing an initial value of inertia; k represents the current iteration number; k represents a threshold value, and K > 0; q represents the Hill coefficient, and Q is more than or equal to 1, and based on the formula, a new particle speed updating formula is provided as follows:
Figure BDA0002421606000000084
step 2.3, performing indoor overall path planning based on the improved particle swarm algorithm, and refreshing the GPS positioning information of the personnel at intervals by a software program for giving an optimal escape route based on the improved particle swarm algorithm in real time;
step 2.4: the specific steps of constructing the bottleneck point detection module aiming at the special period are as follows:
calculating bottleneck points and node capacity thresholds, identifying potential bottleneck points limiting exit movement by using linear programming, calculating the node capacity thresholds by taking a as a personnel density threshold, wherein the value of a depends on different building conditions, and when the personnel density of a node exceeds a, the bottleneck points are considered to be started to appear, namely the personnel movement speed is rapidly reduced or even stopped, and the specific expression is as follows:
Figure BDA0002421606000000085
each evacuation node
Figure BDA0002421606000000086
Is obtained by linear programming if
Figure BDA0002421606000000087
The node is considered to be a bottleneck point.
Preferably, the steps 1 to 3 are preliminarily set for two specific positions:
(1) when the evacuation nodes in the building channel intersection are connected to the surrounding evacuation intersection nodes, dividing the evacuation nodes into one evacuation node respectively;
(2) due to the irregular flow of people, evacuation nodes at the T-shaped intersection and the crossroad are not merged.
Preferably, the specific steps of the indoor overall path planning in step 2.3 are as follows:
3-1: collecting the geographical position, the structural composition and the entrance and exit information of the indoor large building where the personnel are located;
3-2: setting nodes indoors based on graph theory, and performing addition and deletion modification on nodes of special sections;
3-3: determining correct and wrong source points according to the types of the emergency events, and primarily determining a feasible route;
3-4: establishing a multi-input source point-multi-output evacuation model by utilizing an improved particle swarm algorithm, and finding out the most appropriate evacuation route;
3-5: pre-judging possible bottleneck points through the model result in the step 3-4;
3-6: judging whether congestion occurs at a certain node, if so, opening other channels which are not commonly used or selecting other nodes closest to the bottleneck point for escape, and then turning to the step 3-4 to re-plan an escape route; otherwise, continuing to the step 3-7;
3-7: and determining an evacuation scheme, finishing the algorithm and outputting an optimal planned path.
Preferably, the solution of the bottleneck point in step 2.4 is as follows:
case a: when people are evacuated, if the situation that people jam at an evacuation node occurs, a bottleneck point occurs, an uncommon channel or an employee channel closest to the bottleneck point is considered to be opened so as to reduce the pressure of the people at the node. But because the entrance of the staff is narrow and unsafe, when the bottleneck point is no longer crowded or the density of the staff is less than or equal to a, measures are immediately taken to close the entrance of the staff so as to ensure the safety of the crowd.
Case B: if a bottleneck point appears under the condition that no redundant channel is available in the large building, another unblocked channel closest to the bottleneck point is recommended to be used through a software program, and an escape route is re-planned, so that the escape efficiency is improved.
Preferably, the emergency events in the steps 1-2 include fire, earthquake, theft and robbery, and air pollution, and the sensors include a temperature sensor, an ambient light sensor, a humidity sensor, an air pressure sensor, a GPS, a noise sensor, a vibration sensor, a gravity sensor, and an air sensitive sensor.
As shown in fig. 2, in the present invention, the association between the emergency and the sensor may be a temperature sensor, an ambient light sensor, a humidity sensor, and an air pressure sensor corresponding to a fire; the earthquake corresponds to a GPS, a noise sensor and a vibration sensor; the theft and robbery correspond to the gravity sensor; air pollution corresponds to gas sensors, etc.
Example 1:
environment building module
In order to verify and apply the emergency evacuation model to the rupo palace, before entering the palace to visit, each tourist and staff are ensured to be equipped with one or more of a smart phone, a smart watch, a computer (or a tablet), and the like, a specific software program is downloaded, and a simulation test is carried out, so that the number of visitors and the average number of visitors at present are provided for the tourists.
A plurality of sensors of different types are arranged at every place in a room at certain intervals so as to achieve the effect of predicting different emergencies such as earthquakes, fires and the like. The data of the operation is updated in real time, and is sent to the processing terminal together with the GPS positioning information of the crowd to be evacuated, but the data does not participate in the processing of the follow-up improved algorithm, when a certain numerical value of the sensor exceeds a preset safety threshold value, an emergency possibly does not occur at the moment, the terminal gives an alarm in advance to remind of making a safety prejudgment, and the evacuation route planning is convenient to be made before the crowd evacuation premise.
Dividing each evacuation node: the construction of the step is mainly to simplify the process from the evacuation starting point to the node terminal point, one evacuation path is composed of a plurality of arc segments, each node of each floor of the building can be simplified into an evacuation source point, and the building is connected from the source point to the terminal point, namely an exit, through the nodes. To avoid the system running time from being prolonged due to too many nodes or even unnecessary repetitive operations, it is considered to reasonably merge the nodes that can be merged to reduce the number of network nodes. The main principle is to merge all nodes connected to the same node, and the relevant characteristic parameters need to be modified correspondingly after a new node is formed.
For special positions, in particular the two shown by fig. 3 and 4:
with respect to fig. 3, when nodes in an intersection are connected to nodes of surrounding intersections, they are respectively divided into one node.
With respect to fig. 4, at T-junctions and intersections, because of irregular flow of people, it is not considered to merge them.
Path planning module
According to the Luo palace website, the museum is closed every Tuesday and in public holidays. Assuming that visitors do not stay for more than half a day, considering that the lupo palace has received 810 ten thousand visitors in 2017, we calculate the average number of visitors per moment to be 13106. Before entering a large indoor building, the electronic device is used to download a relevant software program which allows access to the current user's GPS information and updates the location information in real time, sending the information to the processing terminal at intervals for processing by a subsequent computer using an improved algorithm. In addition, when danger happens, the voice broadcasting function is provided, so that personnel can be prompted to have the best path all the time.
Fig. 5 is a structure diagram of the third floor of the luo pumice leisojou museum (Richelieu), which has three exhibition halls, namely the syrian museum, the german museum and the leisojou museum, wherein four floors of each exhibition hall are respectively a first floor, a second floor and a third floor. Three exhibition halls are connected together in the underground floor, all the entrances and exits are in the bottom floor, and VIP channels are opened in the ground floor, namely the ground floor. Due to the fact that the three exhibition halls are repaired respectively and have different opening times, the Musai residence halls are often opened to the outside independently, and the Musai residence halls are considered to be used as an independent evacuation system for research. As can be seen from fig. 5, the numeral signs represent different exhibition halls, storing works with different artistic features or styles, the downward stair symbol beside the numeral 855 represents a green safety channel, and since the floor is a top floor, only the nodes are evacuated downward; meanwhile, the escalator identification beside the number 835 represents the escalator, potential safety hazard is considered when an unexpected event occurs, people are not recommended to be evacuated from the escalator, and similarly, the vertical elevator is omitted from the drawing because the safety problem is not considered. In fig. 3, circle numbers represent division and combination of nodes according to the graph theory principle, arrow directions represent evacuation directions which can be selected by people at the nodes, triangular numbers represent exits of a floor, can be upward evacuation exits or downward evacuation exits, each evacuation node is divided according to floors, and the nodes are reasonably arranged for the luohuogong.
Escape model relates to parameter introduction
Step 1: equivalent length of evacuation and evacuation speed, using em for different emergencies, denoted as
Figure BDA0002421606000000111
Evacuation route wijEquivalent length of
Figure BDA0002421606000000115
Is shown as
Figure BDA0002421606000000112
Wherein f (K)sRho, T) is the activity index of the action of the personnel under the influence of fire smoke;
Figure BDA0002421606000000116
is wijThe geometric length of (a).
Wherein f (K)s,ρ,T)＝f1(Ks)·f2(ρ)·f3(T)，f1(Ks) For the visibility-influencing factor, KsTo reduce the light coefficient
Figure BDA0002421606000000113
f2(ρ) is toxic gas concentration influence coefficient:
Figure BDA0002421606000000114
where ρ is the concentration of CO (%); t is exposure time (min). f. of3(T) is the influence coefficient of the flue gas temperature:
Figure BDA0002421606000000121
wherein T issIs the temperature of the fire field, v0Is a human beingNormal walking speed.
Step 2: the calculation formulas of the evacuation speed and the flow coefficient of different channels are as follows, wherein v is the personnel evacuation speed, and ρ is the people flow density.
(1) The evacuation velocity and the personnel flow coefficient on the horizontal channel are ^ 1.427-0.3549 rho;
(2) the evacuation speed and the personnel flow coefficient descending the stairs are ^ v ═ 0.6502-0.0972 ρ;
(3) the evacuation velocity and people flow coefficient up stairs is ═ v ═ 0.564-0.0765 ρ,
step 3: simplifying the model, selecting the flow C of people at the time of the node ti(t) and node spacing
Figure BDA0002421606000000122
The density of the stream of people is calculated as,
Figure BDA0002421606000000123
(1) the method is characterized in that people of each source point are reasonably distributed to different paths under the condition that the limit of the node capacity cannot be exceeded, so that the evacuation time of an evacuation network is minimized. The following relationships are established:
Figure BDA0002421606000000124
wherein the content of the first and second substances,
Figure BDA0002421606000000125
for ensuring that everyone on node i is evacuated, PijA value of > 0 is used to ensure that the value of the flow of personnel is positive.
(2) Bottleneck point detection and solution
Case a: the building has employee access. When people are evacuated, if the situation that people at a certain node are congested occurs, a bottleneck point is formed, an uncommon channel or a staff channel closest to the bottleneck point is opened to reduce the pressure of the people at the node. But because the entrance of the staff is narrow and unsafe, when the bottleneck point is no longer crowded or the density of the staff is less than or equal to a, measures are immediately taken to close the entrance of the staff so as to ensure the safety of the crowd.
Case B: there is no staff access to the building. If the bottleneck point appears under the condition that no redundant channel is available in the large building, other smooth channels nearest to the bottleneck point are recommended to be used through a software program, and the escape route is re-planned, so that the escape efficiency is improved.
(3) Improved particle swarm optimization path
And optimizing the planned path by adopting an improved particle swarm algorithm. In the conventional PSO algorithm, each particle has a certain adaptation value to all targets, and the dimension of each target is n. The position of the particle i after k iterations is represented as
Figure BDA0002421606000000131
The velocity is expressed as
Figure BDA0002421606000000132
The optimal position of the particle i is denoted as Pi n(k) In that respect The speed is updated by Pi n(k) And Pn(k) Are co-performed so that the particles are closer to the optimum position, Pi n(k) The position information is updated as follows:
Figure BDA0002421606000000136
at the same time
Figure BDA0002421606000000137
The speed information is updated as:
Figure BDA0002421606000000138
wherein w is the inertial weight; c. C1And c2Is a learning factor, usually c1＝c2＝2；r1And r2Is [0,1 ]]The random number of (2).
However, in the conventional PSO design process, the inertia weight is generally described by a linear function or a constant function, and the design is difficult to avoid the algorithm from falling into local optimization. The function controlling hormone regulation (i.e., the Hill coefficient) is a non-linear function that is non-negative and monotonic. The inertia weight w given based on the rule of the bio-hormone regulation mechanism is designed as:
Figure BDA0002421606000000139
wherein wmaxRepresents the maximum value of the inertial weight; w is aminRepresents the minimum value of the inertia weight; w is a0Representing an initial value of inertia; k represents the current iteration number; k represents a threshold value, and K > 0; q represents a Hill function, and Q ≧ 1. Therefore, based on the above thought, a new velocity update formula for the particles is proposed:
Figure BDA0002421606000000141
an indoor overall path planning is performed based on an improved particle swarm optimization algorithm, and the specific steps are as shown in fig. 6:
3-1: collecting information of geographical position, structural composition, entrance and exit of the indoor large building where the personnel are located;
3-2: setting nodes indoors based on graph theory, and performing addition and deletion modification on nodes of special sections;
3-3: determining correct and wrong evacuation nodes according to the type of the emergency, and primarily determining a feasible route;
3-4: establishing a multi-input source point-multi-output evacuation model by utilizing an improved particle swarm algorithm, and finding out the most appropriate evacuation route;
3-5: pre-judging possible bottleneck points through the model result in the step 3-4;
3-6: judging whether congestion occurs at a certain node, if so, opening other channels which are not commonly used or selecting other nodes closest to the bottleneck point for escape, and then turning to the step 3-4 to re-plan an escape route; otherwise, continuing to the step 3-7;
3-7: and determining an evacuation scheme, finishing the algorithm and outputting an optimal planned path.
The software program refreshes the geographical position information of the personnel at each time to give the optimal escape route based on the improved particle swarm algorithm at any time.
According to the flow of the improved particle swarm optimization, the avgWide is set to be 15m, all parameters are brought into the system, and when P is the valueiWhen the average value of (a) is 100, T is 293s, which is about five minutes, the specific results of the evacuation system are shown in table 1:
TABLE 1 evacuation System specific results
Figure BDA0002421606000000142
Figure BDA0002421606000000151
Comparing the results in table 1, on the premise that the VIP channel is opened and the employee channel is not opened, no bottleneck point which may exist is found. Therefore, with these parameters, a strategy of opening only the VIP channel is feasible for the lipsotrichum, which achieves better results.
The related software programs related to the present invention are designed by those skilled in the art according to actual situations, belong to conventional technical means, and are not described in detail.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (5)

1. A large indoor personnel evacuation system based on an improved particle swarm optimization algorithm is characterized by comprising an environment construction module and a planned path module, wherein,
the environment construction module comprises the following specific construction steps:
step 1-1: before entering a large indoor building, a mobile terminal is used for downloading a related software program, the software program allows a current user to access GPS positioning information and update the position information in real time, the software program sends the position information to a processing terminal at intervals, the processing terminal adopts an improved algorithm to perform data processing, the mobile terminal has a voice broadcasting function, and when danger occurs, the mobile terminal performs voice broadcasting for constantly prompting personnel of an individual optimal path;
step 1-2: the method comprises the following steps that a plurality of sensors of different types are arranged at every position in a building at certain intervals and used for predicting different emergencies, detection data of the sensors are updated in real time, the detection data of the sensors and GPS positioning information of crowds are sent to a processing terminal in real time, when a certain numerical value of the sensors exceeds a preset safety threshold value, the processing terminal gives an alarm in advance to remind an evacuation system and workers of safety prejudgment, and evacuation route planning is facilitated before crowd evacuation;
step 1-3: dividing each evacuation node, wherein each evacuation path consists of a plurality of arc sections, each evacuation node of each floor of the building is simplified into an evacuation source point, the building is connected from the evacuation source point to an exit through the evacuation nodes, new nodes are formed by combining the evacuation nodes to reduce the number of network nodes, and relevant characteristic parameters are correspondingly modified after the new nodes are formed;
the path planning module comprises the following specific construction steps:
step 2-1: establishing a multi-input source point-multi-output evacuation model based on graph theory, and reasonably distributing people of each evacuation source point to different paths so as to minimize the evacuation time of an evacuation network, wherein the model is established by the following relational expression:
min T＝min(max Tij)，
Figure FDA0002421605990000011
Figure FDA0002421605990000012
wherein min T is min (max T)ij) To ensure that the time between node i and node j is minimized,
Figure FDA0002421605990000021
indicating the required evacuation time, C, for node i and node j in different environments resulting in different speedsi(t)≤C maxiThe capacity threshold value is used for ensuring that the number of people at the node i at the time t cannot exceed the person density of the node;
Figure FDA0002421605990000022
Figure FDA0002421605990000023
for ensuring that everyone on node i is evacuated, PijMore than or equal to 0 is used to ensure that the people flow value is positive;
step 2-2: an improved particle swarm algorithm is adopted to optimize path planning, in the traditional PSO algorithm, each particle has a certain adaptive value to all targets, the dimension of each target is n, and the position of the particle i after iteration for k times is represented as
Figure FDA0002421605990000024
The velocity is expressed as
Figure FDA0002421605990000025
The optimal position of the particle i is denoted as Pi n(k) The optimum position of the entire particle population is denoted as Pn(k) I.e. speed update by Pi n(k) And Pn(k) Are co-performed so that the particles are closer to the optimum position, Pi n(k) The position information is updated as follows:
Figure FDA0002421605990000026
at the same time
Figure FDA0002421605990000027
The speed information is updated as:
Figure FDA0002421605990000028
wherein w is the inertial weight; c. C1And c2Is a learning factor, usually c1＝c2＝2；r1And r2Is [0,1 ]]The random number of (2);
describing the inertia weight by adopting a function for controlling hormone regulation, and then giving the design of the inertia weight w based on the rule of a biological hormone regulation mechanism as follows:
Figure FDA0002421605990000029
wherein, wmaxRepresents the maximum value of the inertial weight; w is aminRepresents the minimum value of the inertia weight; w is a0Representing an initial value of inertia; k represents the current iteration number; k represents a threshold value, and K > 0; q represents the Hill coefficient, Q is more than or equal to 1, and a new particle speed updating formula is provided as follows based on the formula:
Figure FDA00024216059900000210
step 2-3, performing indoor overall path planning based on the improved particle swarm algorithm, and refreshing the GPS positioning information of the personnel at intervals by a software program for giving an optimal escape route based on the improved particle swarm algorithm in real time;
step 2-4: the specific steps of constructing the bottleneck point detection module aiming at the special period are as follows:
calculating a bottleneck point and a node capacity threshold, identifying a potential bottleneck point limiting exit movement by using linear programming, calculating the node capacity threshold by taking a as a personnel density threshold, and when the personnel density of a node exceeds a, considering that the bottleneck point starts to appear, namely the personnel movement speed is rapidly reduced and even stopped, wherein the specific expression is as follows:
Figure FDA0002421605990000031
each evacuation node
Figure FDA0002421605990000032
Is obtained by linear programming if
Figure FDA0002421605990000033
The node is considered to be a bottleneck point.
2. The large indoor personnel evacuation system based on the improved particle swarm optimization algorithm according to claim 1, wherein the steps 1-3 are preliminarily set for the following two special positions:
(1) when the evacuation nodes in the building channel intersection are connected to the surrounding evacuation intersection nodes, dividing the evacuation nodes into one evacuation node respectively;
(2) due to the irregular flow of people, evacuation nodes at the T-shaped intersection and the crossroad are not merged.
3. The large indoor personnel evacuation system based on the improved particle swarm optimization algorithm according to claim 1, wherein the specific evacuation steps of the evacuation system are as follows:
3-1: the environment construction module collects GPS positioning information, structural composition and entrance and exit information of an indoor large building where the personnel are located;
3-2: setting nodes indoors based on graph theory, and performing addition and deletion modification on nodes of special sections;
3-3: determining correct and wrong evacuation nodes according to the type of the emergency, and primarily determining a feasible route;
3-4: establishing a multi-input source point-multi-output evacuation model by utilizing an improved particle swarm algorithm, and finding out the most appropriate evacuation route;
3-5: pre-judging possible bottleneck points through the model result in the step 3-4;
3-6: judging whether congestion occurs at a certain node, if so, opening other channels which are not commonly used or selecting other nodes closest to the bottleneck point for escape, and then turning to the step 3-4 to re-plan an escape route; otherwise, continuing to the step 3-7;
3-7: and determining an evacuation scheme, finishing the algorithm and outputting an optimal planned path.
4. A large indoor people evacuation system based on improved particle swarm optimization algorithm according to claim 1, wherein the solution of the bottleneck point in the step 2-4 is as follows:
case a: when people are evacuated, if the situation that people jam at an evacuation node occurs, a bottleneck point occurs, an uncommon channel or an employee channel closest to the bottleneck point is considered to be opened so as to reduce the pressure of the people at the node. But as the entrance of the staff is narrow and unsafe, when the bottleneck point is not crowded any more or the staff density is less than or equal to a, immediately taking measures to close the entrance of the staff so as to ensure the safety of the crowd; case B: if a bottleneck point appears under the condition that no redundant channel is available in the large building, another unblocked channel closest to the bottleneck point is recommended to be used through a software program, and an escape route is re-planned, so that the escape efficiency is improved.
5. A large indoor personnel evacuation system based on improved particle swarm optimization algorithm according to claim 1, wherein the emergencies in the steps 1-2 comprise fire, earthquake, theft, robbery and air pollution, and the sensors comprise temperature sensor, ambient light sensor, humidity sensor, air pressure sensor, GPS, noise sensor, vibration sensor, gravity sensor and air-sensitive sensor.
CN202010207394.3A 2020-03-23 2020-03-23 Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm Pending CN111311028A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010207394.3A CN111311028A (en) 2020-03-23 2020-03-23 Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010207394.3A CN111311028A (en) 2020-03-23 2020-03-23 Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm

Publications (1)

Publication Number Publication Date
CN111311028A true CN111311028A (en) 2020-06-19

Family

ID=71158823

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010207394.3A Pending CN111311028A (en) 2020-03-23 2020-03-23 Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm

Country Status (1)

Country Link
CN (1) CN111311028A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113901727A (en) * 2021-11-16 2022-01-07 枫树谷(成都)科技有限责任公司 Crowd evacuation simulation method based on molecular dynamics and machine learning
CN115759463A (en) * 2022-12-02 2023-03-07 武汉理工大学三亚科教创新园 Passenger ship passenger escape path dynamic planning method
CN116030607A (en) * 2023-03-27 2023-04-28 中国电力工程顾问集团西南电力设计院有限公司 Intelligent power plant safety supervision reminding and early warning system
CN116187612A (en) * 2023-04-28 2023-05-30 交通运输部水运科学研究所 Dynamic flow distribution-based personnel evacuation path optimization management method and system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106611474A (en) * 2015-10-24 2017-05-03 重庆聆益机械有限公司 Mechanical processing and manufacturing safety protection alarm system
CN109655068A (en) * 2019-01-31 2019-04-19 山东师范大学 Path replanning method and device, system based on the event of fire that happens suddenly in building

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106611474A (en) * 2015-10-24 2017-05-03 重庆聆益机械有限公司 Mechanical processing and manufacturing safety protection alarm system
CN109655068A (en) * 2019-01-31 2019-04-19 山东师范大学 Path replanning method and device, system based on the event of fire that happens suddenly in building

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG, JIAHAO等: "Multi-input source and Multi-destination Evacuation Model", 《2019 IEEE 9TH ANNUAL INTERNATIONAL CONFERENCE ON CYBER TECHNOLOGY IN AUTOMATION, CONTROL, AND INTELLIGENT SYSTEMS (CYBER)》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113901727A (en) * 2021-11-16 2022-01-07 枫树谷(成都)科技有限责任公司 Crowd evacuation simulation method based on molecular dynamics and machine learning
CN115759463A (en) * 2022-12-02 2023-03-07 武汉理工大学三亚科教创新园 Passenger ship passenger escape path dynamic planning method
CN115759463B (en) * 2022-12-02 2023-11-17 武汉理工大学三亚科教创新园 Dynamic planning method for escape path of passenger on passenger ship
CN116030607A (en) * 2023-03-27 2023-04-28 中国电力工程顾问集团西南电力设计院有限公司 Intelligent power plant safety supervision reminding and early warning system
CN116030607B (en) * 2023-03-27 2023-06-09 中国电力工程顾问集团西南电力设计院有限公司 Intelligent power plant safety supervision reminding and early warning system
CN116187612A (en) * 2023-04-28 2023-05-30 交通运输部水运科学研究所 Dynamic flow distribution-based personnel evacuation path optimization management method and system

Similar Documents

Publication Publication Date Title
CN111311028A (en) Large-scale indoor personnel evacuation system based on improved particle swarm optimization algorithm
CN105631555B (en) A kind of method for pushing and device for evacuating path
Zheng et al. Simulation of pedestrian evacuation considering emergency spread and pedestrian panic
JP2000113357A (en) Comprehensive disaster prevention and belief system
CN114445984A (en) Fire-fighting emergency evacuation system and method for urban large building
Bukowski et al. Egress concepts and design approaches
Zeng et al. Effects of initial distribution ratio and illumination on merging behaviors during high-rise stair descent process
CN115752490B (en) Safe trip path optimization method and system based on big data and positioning technology
CN112488401A (en) Fire escape route guiding method and system
CN116822764A (en) Visual intelligent emergency command system of fire control
Zuo et al. Simulation and optimization of pedestrian evacuation in high-density urban areas for effectiveness improvement
CN112488423A (en) Method for planning escape path of trapped personnel in fire scene
JP4915550B2 (en) Evacuation behavior prediction system and evacuation behavior prediction method
Li et al. Research on OREMS-based large-scale emergency evacuation using vehicles
CN116993917A (en) Fire simulation evacuation drilling method based on three-dimensional model
Osaragi et al. Simulation model of evacuation behavior following a large-scale earthquake that takes into account various attributes of residents and transient occupants
CN106600862A (en) Fire escape assistance method and system
CN117114222A (en) Intelligent emergency evacuation line optimization method and system based on real-time data of Internet of things
Zhou et al. Modeling of crowd dynamics with emergency signs via modified social force model
KR102592101B1 (en) A system and methods for guiding evacuation route based on artificial intelligence
Franzese et al. A methodology for the assessment of traffic management strategies for large-scale emergency evacuations
Jiang et al. Emergency evacuation based on long range communication technology
SATIR et al. A review of evacuation of high-rise buildings
CN114067065A (en) Escape route visualization system and escape method
Wei et al. Integrated urban evacuation planning framework for responding to human-caused disasters over a surface transportation network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20200619

RJ01 Rejection of invention patent application after publication