CN-122022106-A - Dynamic guide identifier optimization method

CN122022106ACN 122022106 ACN122022106 ACN 122022106ACN-122022106-A

Abstract

The invention relates to a dynamic guiding mark optimizing method which comprises the steps of dividing an internal evacuation space of a subway station into a plurality of cells, constructing a space discrete model according to an actual space structure, a preset channel connection relation and preset evacuation, defining the number of passengers in each cell and the transmission flow of passengers between adjacent cells in the current time step, deducing the evacuation process of the passengers in the plurality of cells based on preset supply and demand constraints, constructing a cell transmission model, mapping fire environment parameters to a cell layer, correcting the throughput of the cells in the fire environment, constructing a dynamic guiding mark optimizing model, embedding uncertainty in the fire environment into the dynamic guiding mark optimizing model to form a dynamic guiding mark robust optimizing model, solving the robust optimizing model, and obtaining an optimal dynamic guiding mark pointing strategy through iterative updating. The intelligent level of evacuation guidance in the fire environment is remarkably improved, and an effective technical support is provided for emergency management of large-scale subway stations.

Inventors

YANG XIAOXIA
ZHANG PING
LI QINGYUN
XIN LIPING
CHENG SEN
PEI XIAOJUAN
LI JIHONG
WEI JINLI
ZHAO LINGYAN
QU DAYI
PAN FUQUAN
ZHANG WENBO
DING MINGYUE
REN LING
YIN YIN
YANG FAZHAN
MA HAO
WANG JIJUN
TIAN YANBING
ZHOU MING
CUI CHUNYUE

Assignees

青岛理工大学
中国铁建电气化局集团有限公司
中铁建电气化局集团第三工程有限公司
中铁通信信号勘测设计院有限公司

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (10)

1. A dynamic guide identifier optimization method, comprising: Dividing an internal evacuation space of a target subway station into a plurality of cells, and establishing a communication relationship among the cells according to an actual space structure, a preset channel connection relationship and preset evacuation of the target subway station to construct a space discrete model of the target subway station, wherein the cells are used for representing a local evacuation area in the subway station; defining the number of passengers in each cell and the passenger transmission flow between adjacent cells in the current time step, deducing the evacuation process of the passengers in the cells based on preset supply and demand constraint, and constructing a cell transmission model; Acquiring fire environment parameters at different moments in the target subway station, mapping the fire environment parameters to cell layers, and correcting the traffic capacity of each cell in the cell transmission model under the fire environment; Constructing a dynamic guiding mark optimizing model aiming at the accumulated evacuation time of passengers, the exposure level of fire risks and the density balance of key areas, wherein a dynamic guiding mark decision variable in the dynamic guiding mark optimizing model represents the pointing state of each guiding mark in the current time step and is used for controlling the flow path of passengers among cells; determining uncertainty in the fire environment according to the fire environment parameters, and embedding the uncertainty into the dynamic guide mark optimization model to form a dynamic guide mark robust optimization model; And solving the dynamic guide identifier robust optimization model, and obtaining the optimal dynamic guide identifier pointing strategy of the target subway station in the fire uncertain environment through iterative updating.
2. The method of claim 1, wherein the plurality of cells includes a first cell and a second cell, the first cell being an adjacent upstream cell of the second cell, the defining the number of passengers in each cell and the traffic of passengers between adjacent cells in the current time step, and deriving evacuation processes of passengers in the plurality of cells based on preset supply and demand constraints, constructing a cell transfer model comprising: calculating the current time step according to the physical attribute parameters of the cells and the free movement speed of the passengers in the free movement state; Defining the number of passengers in each cell in the current time step; Calculating the supply capacity of the first unit cell in the time step according to the number of passengers in the first unit cell, the cell area and a first preset passing condition, wherein the supply capacity is used for representing the maximum number of passengers output by the first unit cell to the second unit cell in the current time step; calculating the receiving capacity of the second cell in the time step according to the remaining available space of the second cell, the cell area and a second preset traffic condition, wherein the receiving capacity is used for representing the maximum number of passengers received by the second cell from the first cell in the current time step; according to the free movement speed, the preset cell critical density, the preset cell congestion density and the preset congestion movement speed, deducing the evolution characteristics of the intra-cell crowd transmission flow along with time in the passenger evacuation process to obtain the cell actual density; And defining passenger transmission flow between adjacent cells in the current time step according to the actual density of the cells, the supply capacity and the receiving capacity, and constructing a cell transmission model.
3. The method of claim 2, wherein defining passenger traffic between adjacent cells for the current time step comprises: Calculating evacuation field values corresponding to the second cells when the first cells are communicated with the plurality of second cells in the current time step, wherein the evacuation field values are used for representing evacuation superiority in the downstream direction; Determining the distribution proportion of the crowd transmission flow distributed to each second binary cell according to the evacuation field value corresponding to each second binary cell; And defining passenger transmission flow between adjacent cells in the current time step according to the evacuation priority of each second cell based on the distribution proportion of each second cell.
4. The method of claim 1, wherein the acquiring fire environment parameters at different times in the target subway station, mapping the fire environment parameters to cell layers, and correcting the throughput of each cell in the cell transmission model in the fire environment, comprises: Acquiring fire environment parameters at different moments in the target subway station, and mapping the fire environment parameters to cell layers of the cell transmission model to acquire fire environment states corresponding to each cell under different time steps; constructing a fire environment influence factor based on the fire environment state, wherein the fire environment influence factor is used for quantifying the inhibition effect of the fire environment on passenger traffic capacity; and correcting the traffic capacity of each cell in the free circulation state according to the fire environment influence factors to obtain a corrected cell transmission model, wherein the corrected cell transmission model considers the effective traffic capacity of cells after the fire environment influence.
5. The method of claim 4, wherein said constructing a fire environment influence factor based on said fire environment status comprises: Calculating to obtain a quantized value of the visibility at the cell according to a preset weight coefficient of the visibility, an actual value of the visibility at the cell in the current time step and an initial value of the visibility at the cell in the current time step, and calculating to obtain a visibility factor according to the quantized value of the visibility at the cell, wherein the visibility factor is used for representing damage of the limited vision caused by smoke to the crowd mobility; Calculating to obtain a quantized value of the CO concentration at the cell according to a preset weight coefficient of the CO concentration, an actual value of the CO concentration at the cell in the current time step and an initial value of the CO concentration at the cell in the current time step, and calculating to obtain a CO concentration factor according to the quantized value of the CO concentration at the cell, wherein the CO concentration factor is used for representing the influence of poisoning risks caused by the CO exposure accumulation effect on the crowd mobility; Calculating to obtain a quantized value of the temperature at the cell according to a preset weight coefficient of the temperature, an actual value of the temperature at the cell in the current time step and an initial value of the temperature at the cell in the current time step, and calculating to obtain a temperature factor according to the quantized value of the temperature at the cell, wherein the temperature factor is used for representing the constraint of heat radiation damage and heat resistance limit caused by a high-temperature environment on crowd movement capacity; And constructing a fire environment influence factor according to the visibility factor, the CO concentration factor and the temperature factor.
6. The method of claim 1, wherein the constructing a dynamic guide sign optimization model targeting passenger cumulative evacuation time, fire risk exposure level, and critical zone density equalization comprises: accumulating the residence time of the passengers in each cell in the evacuation process to obtain accumulated residence time of the passengers; Normalizing the quantized value of the visibility of the cells, the quantized value of the concentration of CO at the cells and the quantized value of the temperature at the cells and superposing the quantized values according to weights to obtain a nominal risk value at the cells in the current time step, wherein the nominal risk value is used for representing the environmental risk level born by passengers in the evacuation process; Constructing a passenger density variance function of the key area according to the actual density of cells in the key area in the target subway station in the current time step and the average value of the densities of all the key areas in the current time step; And constructing a dynamic guide identification optimization model according to the accumulated residence time of the passengers, the nominal risk value and the passenger density variance function.
7. The method of claim 1, wherein the determining the uncertainty in the fire environment based on the fire environment parameters and embedding the uncertainty in the dynamic guide identifier optimization model to form a dynamic guide identifier robust optimization model comprises: setting the fire environment parameters as uncertain parameters, wherein the uncertain parameters are used for representing the influence of fire development and environmental disturbance on evacuation operation of subway stations; Constructing an uncertainty set for the uncertainty parameter, wherein the uncertainty set is used for describing a fluctuation interval of the uncertainty parameter within an allowable range, and the uncertainty set at least comprises a box type uncertainty set and an intersection budget uncertainty set; under the constraint condition of the uncertain set, carrying out robustness processing on the dynamic guide mark optimization model to form a dynamic guide mark robust optimization model, so that the dynamic guide mark robust optimization model still meets evacuation operation constraint conditions under the condition of the least adverse situation; the dynamic guided identification robust optimization model is transformed to transform the worst case conditions contained therein into a set of deterministic constraints of equal order.
8. The method of claim 7, wherein said translating the dynamic guided identification robust optimization model to translate the least favorable case conditions contained therein into a set of deterministic constraints of equal order comprises: Constructing an inner layer maximization problem function according to the crowd distribution state of the cells, the uncertain parameters and the least unfavorable deviation value; constructing a Lagrangian function according to the non-negative Lagrangian multiplier and the maximized problem function; Solving a one-dimensional maximization problem of the Lagrangian function about the least favorable deviation value to obtain an optimal value of the least favorable deviation; Substituting the optimal value into the maximized problem function, and obtaining the total evacuation risk value increment under the least adverse condition according to the optimal deviation contribution of all cells and Lagrangian multiplier terms; substituting the total evacuation risk value increment into a pre-constructed minimum problem function to obtain the dynamic guide identifier robust optimization model containing a group of deterministic constraint conditions.
9. The method of claim 7, wherein solving the dynamic guide identifier robust optimization model and obtaining an optimal dynamic guide identifier pointing strategy of the target subway station in the fire uncertainty environment through iterative updating comprises: Coding a pointing strategy of the dynamic guide identifier, which is given by the dynamic guide identifier robust optimization model, in the current time step into a discrete decision sequence as an individual representation of each individual in a genetic algorithm; initializing genetic algorithm parameters, generating an initial population, and calculating the fitness value of the guide identifier corresponding to each individual in the initial population according to the deterministic optimization model; performing selection, crossing and mutation operations on population individuals based on the fitness value, and generating a candidate dynamic guide identifier pointing strategy; And in the iterative search process of the discrete solution space, updating the candidate dynamic guiding identification pointing strategy based on a kangaroo escape optimization algorithm to obtain the optimal dynamic guiding identification pointing strategy of the target subway station in the fire uncertain environment.
10. The method of claim 9, wherein updating the candidate dynamic guide identifier pointing policy based on a kangaroo escape optimization algorithm in the iterative search process of the discrete solution space to obtain an optimal dynamic guide identifier pointing policy of the target subway station in a fire uncertainty environment comprises: screening partial individuals from the population of the genetic algorithm, and inputting a guide mark pointing decision sequence corresponding to the partial individuals as an initial solution into a kangaroo escape optimization algorithm; Constructing a direction vector pointing to a current optimal solution, wherein the direction vector is used for ensuring that an individual has the capability of moving to a better area; Applying a random rotation to the direction vector and calculating a rotation angle; Generating a rotated escape direction according to the unit vector treated with the direction vector and the rotation angle; based on the escape direction and a preset jump intensity parameter, nonlinear offset in the individual searching process is realized by setting a Gaussian disturbance term, and jump updating is carried out on an individual towards the current optimal solution direction by introducing a direction guide term between the optimal individual and the current individual; After continuous disturbance update is carried out on an individual, the updated individual is mapped back to a preset discrete domain through a nearest neighbor projection mode, so that the feasibility constraint of a dynamic guiding mark pointing strategy is met while the escape searching characteristic is kept for the updated disturbance; Comprehensively evaluating the evacuation efficiency and the safety performance of the candidate dynamic guide identifier pointing strategy, selecting and updating the population based on the comprehensive evaluation result, and continuing the next iteration until the iteration is stopped when the termination condition is met, so as to obtain the optimal dynamic guide identifier pointing strategy.

Description

Dynamic guide identifier optimization method Technical Field The invention relates to the technical field of urban rail transit safety and emergency evacuation, in particular to a dynamic guiding identification optimizing method. Background With the rapid development of urban rail transit, subways have become an important component of urban traffic. However, the subway station is usually located underground, and has the characteristics of closed space, complex structure, high personnel density and the like. In case of fire, the smoke diffuses rapidly, the visibility drops sharply, congestion is easily caused, and the life safety of passengers is seriously threatened. Guide signs are key facilities for guiding people to evacuate. Currently, a subway station generally adopts a static guiding sign, and the direction of the static guiding sign is fixed at the initial stage of construction. However, in a dynamically evolving fire scene, the fire, smoke and crowd distribution all change in real time, and the static guiding mark is difficult to respond in time, so that passengers can be continuously guided to a dangerous area, and the evacuation risk is increased. Therefore, a dynamic guiding system capable of adjusting the direction of indication according to the real-time environment becomes a potential means for improving evacuation efficiency. However, dynamic steering faces a number of technical challenges in practical applications. The fire environment has strong uncertainty, and parameters such as temperature, visibility and the like are changed rapidly, so that the guiding decision is extremely complex. Meanwhile, the subway station has a plurality of evacuation nodes, and the guiding decision presents high-dimensional discrete characteristics. Therefore, it is needed to construct a subway station passenger evacuation dynamic refinement method oriented to fire uncertain environments, so that evacuation efficiency and personnel safety are cooperatively ensured under the condition of the least unfavorable disaster, and reliability and engineering applicability of a subway station emergency evacuation system are comprehensively improved. Disclosure of Invention In order to solve the technical problems, an embodiment of the present disclosure provides a dynamic guide identifier optimization method. In a first aspect, an embodiment of the present disclosure provides a dynamic guide identifier optimization method, including: Dividing the internal evacuation space of the target subway station into a plurality of cells, establishing a communication relationship among the cells according to the actual space structure, the preset channel connection relationship and the preset evacuation of the target subway station, and constructing a space discrete model of the target subway station, wherein the cells are used for representing the local evacuation area inside the subway station; Defining the number of passengers in each cell and the passenger transmission flow between adjacent cells in the current time step, deducing the evacuation process of the passengers in a plurality of cells based on preset supply and demand constraint, and constructing a cell transmission model; acquiring fire environment parameters at different moments in a target subway station, mapping the fire environment parameters to cell layers, and correcting the traffic capacity of each cell in a cell transmission model under the fire environment; Constructing a dynamic guiding mark optimizing model aiming at the accumulated evacuation time of passengers, the exposure level of fire risks and the density balance of key areas, wherein a dynamic guiding mark decision variable in the dynamic guiding mark optimizing model represents the pointing state of each guiding mark in the current time step and is used for controlling the flow path of passengers among cells; Determining uncertainty in the fire environment according to the fire environment parameters, and embedding the uncertainty into a dynamic guide mark optimization model to form a dynamic guide mark robust optimization model; And solving the dynamic guide identifier robust optimization model, and obtaining the optimal dynamic guide identifier pointing strategy of the target subway station in the fire uncertain environment through iterative updating. In a second aspect, an embodiment of the present disclosure provides a dynamic guide identifier optimizing apparatus, including: the first model construction unit is used for dividing the internal evacuation space of the target subway station into a plurality of cells, and establishing a communication relationship among the cells according to the actual space structure, the preset channel connection relationship and the preset evacuation of the target subway station to construct a space discrete model of the target subway station, wherein the cells are used for representing the local evacuation area in the subway station; The second model constructio