CN117311188B - Control method, system and equipment for crowd diversion railings in fixed places - Google Patents
Control method, system and equipment for crowd diversion railings in fixed places Download PDFInfo
- Publication number
- CN117311188B CN117311188B CN202311249023.1A CN202311249023A CN117311188B CN 117311188 B CN117311188 B CN 117311188B CN 202311249023 A CN202311249023 A CN 202311249023A CN 117311188 B CN117311188 B CN 117311188B
- Authority
- CN
- China
- Prior art keywords
- model
- vector
- output
- matrix
- control
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000004088 simulation Methods 0.000 claims abstract description 34
- 238000005070 sampling Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000033001 locomotion Effects 0.000 claims abstract description 9
- 239000013598 vector Substances 0.000 claims description 88
- 239000011159 matrix material Substances 0.000 claims description 77
- 238000010276 construction Methods 0.000 claims description 12
- 230000001276 controlling effect Effects 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 230000007704 transition Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 5
- 230000001105 regulatory effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 2
- 238000005457 optimization Methods 0.000 description 6
- 208000027744 congestion Diseases 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Computing Systems (AREA)
- Geometry (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Algebra (AREA)
- Automation & Control Theory (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of evacuation safety, in particular to a control method, a system and equipment for crowd diversion railings in fixed places, wherein the method comprises the following steps: 1: individual motion simulation software based on a social force model constructs a three-dimensional simulation scene provided with a fixed entrance and exit place; 2: putting a simulation individual into the three-dimensional simulation scene, simulating a flow guide railing scene with various length sequence changes at the fixed entrance, obtaining passenger flow indexes at bottlenecks at each sampling moment, and constructing an input and output data set; 3: analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure; 4: and constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object to obtain a corresponding optimal length sequence of the guide rail. The invention utilizes the model predictive control method to dynamically regulate and control the diversion rail at the landing mouth of the subway platform layer, and has high predictive accuracy.
Description
Technical Field
The invention relates to the technical fields of pedestrian simulation, evacuation safety and control, in particular to a control method, a system and equipment for crowd diversion railings in fixed places.
Background
The passenger flow volume of the urban subway system is huge, and when the subway is transferred, the phenomenon of congestion easily occurs at the landing entrance of the platform layer, so that the transfer speed of passengers at the subway station is influenced, and the transfer efficiency of the subway station is reduced. This problem increases the risk of subway operation. In addition, once a large number of people need to be evacuated in a subway station, the risk of congestion is greatly increased, the probability of occurrence of safety accidents is also increased, the most common method for relieving the congestion of passengers at the landing entrance of a subway platform layer in the prior art is to manually arrange guide rails to adjust the movement of the passengers, but the method has the problems of low efficiency and high cost, so that a dynamic regulation and control method for the length of the guide rails is needed to effectively relieve the congestion of the passengers.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention aims at: the subway guide rail regulation and control method, system and equipment based on model predictive control are provided, and are used for solving the problems of low efficiency and high cost in the prior art that the guide rail is manually arranged to regulate the moving direction of the hole passenger flow by manually observing the passenger flow.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a method for controlling crowd diversion railings for a fixed location, comprising the steps of:
step 1: individual motion simulation software based on a social force model constructs a three-dimensional simulation scene provided with a fixed entrance and exit place;
step 2: putting a simulation individual into the three-dimensional simulation scene, simulating a flow guide railing scene with various length sequence changes at the fixed entrance, obtaining passenger flow indexes at bottlenecks at each sampling moment, and constructing an input and output data set;
step 3: analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure;
step 4: and constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object to obtain a corresponding optimal length sequence of the guide rail.
In the above control method for crowd diversion railings in fixed places, in the step 1, the fixed entrance place is provided with a subway platform, the building of the three-dimensional simulation scene comprises the platform and stairs connected with a station hall, and the individual comprises a carriage getting-off crowd and a crowd waiting at the platform layer.
In the above control method for crowd diversion railings in fixed places, in the step 2, the input/output data set is that at each time node of the time sequence, the diversion railings at the platform stair opening are subjected to random length change within a predetermined range, so as to obtain a plurality of groups of length sequences of the diversion railings at the subway platform stair opening and passenger flow indexes at the bottleneck of the stair opening at each sampling moment, wherein the passenger flow indexes comprise passenger flow density and/or crowd pressure.
The above control method for crowd diversion railings for fixed places, in the step 3, includes:
step 301: input and output data acquisition and processing, namely, an input and output data set is constructed by defining a length sequence of a guide rail as input and using a passenger flow index at the bottleneck of a stair opening at each sampling moment as output, and the accuracy and usability of the data are realized by removing noise and carrying out data normalization preprocessing;
step 302: selecting and identifying a model structure, and using a linear ARX model as a system identification object, wherein the linear ARX model structure is as follows:where y and u are the output and input of the model, a i And b j Is the regression coefficient of output and input quantity, f and g are the lag time of system output and input, n a And n b Respectively the order of the model, and ζ (t) is modeling error;
step 303: and identifying model parameters, namely solving the structure and each order coefficient of the linear ARX model by adopting a least square method, and obtaining model parameters of different orders and corresponding parameter estimation values through identification so as to determine optimal model parameters.
The above control method for crowd diversion railings in fixed places, in the step 3, further includes:
step 304: model verification, namely, model verification is carried out by using an unused data set, and the fitting capacity and the prediction performance of the model are evaluated by adopting a root mean square error index.
In the above control method for crowd diversion railings in fixed places, in step 302, the model orders are selected to conform to AIC criteria, and the model orders corresponding to the minimum AIC values are found out as the identification model orders by calculating AIC values of a plurality of models.
The above method for controlling crowd diversion railings for fixed places, wherein the step 303 comprises the following steps:
step a: the vector parameters are defined and the vector parameters are defined,
wherein θ is a parameter vector, representing the vector form of the estimated parameter, Φ t-1 The vector form of input and output data before the moment t is represented by the input data vector;
step b: the linear ARX model can be rewritten as an expression of a matrix by the vector parameters:
step c: definition of h m ＝max(n a ,n b ) Defining N as the total number of the identification data, and arranging the sequence data to be identified into a matrix form:
Y＝[y k+1 y k+2 … y N ] T
Φ N ＝[Φ k Φ k+1 … Φ N-1 ] T wherein Y is an output matrix, Φ N Is an input data matrix;
step d: defining an optimal Jie Gong formula:wherein (1)>Representing a least squares estimate of the parameter θ, Φ N And representing the matrix, obtaining an optimal solution of the model according to the optimal solution formula according to the least square method principle, and thus obtaining model parameters of different orders and corresponding parameter estimation values through identification to determine optimal model parameters.
The above control method for crowd diversion railings in fixed places, wherein the step 4 comprises:
step 401: a state space representation of a model, comprising: defining a state vector of a diversion rail system on the basis of the linear ARX model as follows:
x 1,t ＝y(t),h m ＝max(n a ,n b ),h＝2,3,…h m where X (t) is a state vector and t represents a time step, the linear ARX model structure is represented as a state space model as shown below:
wherein x (t+1) represents the state vector of the system at time t+1, x (t) represents the state vector of the system at time t, A represents the state transition matrix, B represents the input matrix, u (t) represents the input vector of the system at time t,
omega is a constant vector, ψ (t+1) is an error term, and C represents an output matrix;
the expression form of each matrix in the state space model is as follows:
step 402: recursively predicting output, the state and the output of each predicted time being calculated based on the state and the output of the last time, comprising: calculating a prediction output based on the state space model, defining an output vector:
wherein (1)>Is a constant combination vector, ">For the optimized control sequence at the current moment, only the control signal at the first moment can be output to the guide rail system, p represents a prediction time domain, q represents a control time domain, the relation between the two is smaller than p, and q represents that the control quantity after the q moment in the optimized control sequence can not change any more, namely u (t+i) =u (t+q) (i is larger than or equal to q), and the control quantity is larger than or equal to q>A predictive state vector representing a system of guide rails, +.>The representation is based on time tA predicted output vector of the diversion railing system is obtained through system state prediction;
performing recursive calculation according to the state space model to obtain a prediction matrix model of a prediction vector (t) as follows:
the expression form of the prediction matrix model C is as follows:
wherein, the method comprises the steps of, wherein,representing a predicted output matrix, E representing an identity matrix, < >>Representing a gain matrix +_>Representing a prediction state transition matrix->Representing a predicted input matrix;
simplifying the predictive output of the linear ARX modelWherein G is a coefficient transfer matrix, y 0 (t) represents a system steady-state output vector;
step 403: designing a control performance target and optimizing and solving, and optimizing control sequences in a prediction modelSolving to obtain a predictive controller based on the linear ARX model, and regulating and controlling a controlled object diversion railing system to obtain an optimal length sequence value of the diversion railing in the evacuation process, wherein the optimal length sequence value is used for controlling actions at the current moment and comprises the following steps: defining a system performance index function:
wherein Q is the difference between the output vectorsIs a unit weight matrix of size p x p, R is a weight matrix of size q x q for the optimal control sequence increment,/o>Representing a tracking set point vector;
obtaining the expected output vectorControl sequence increment->
Wherein the control increment Δu (t) =u (t) -u (t-1), u uper 、u low Vector of maximum and minimum values, deltau, representing control length of the guide rail system uper 、Δu low Maximum and minimum vectors representing system control quantity increment, y uper 、y low Representing the maximum and minimum vectors of the predicted output.
A control system for a crowd diversion rail for a fixed location, the control system applying the control method for a crowd diversion rail for a fixed location as claimed in any one of the preceding claims, comprising: the system comprises a scene construction module, a data set construction module, a system identification module and an MPC control module;
the scene construction module is used for constructing a three-dimensional simulation model of a scene provided with a fixed entrance and exit place, and the fixed entrance and exit place is a subway platform layer;
the data set construction module is used for putting a simulation individual into the three-dimensional simulation model, simulating a plurality of diversion railing scenes with different length sequence changes at the stair opening of the subway platform layer, obtaining passenger flow indexes at the bottleneck of each sampling moment, and constructing an input/output data set;
the system identification module is used for analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure;
and the MPC control module is used for constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object, and controlling the rail length of each time node in the evacuation process.
The control equipment for the crowd diversion railing in the fixed place comprises at least one processor, at least one memory and a bus, wherein the memory and the memory are connected with the processor, the processor and the memory are communicated through the bus, and the processor is used for calling program instructions in the memory so as to execute the control method for the crowd diversion railing in the fixed place.
The control method, the system and the equipment for the crowd diversion railing in the fixed place have the beneficial effects that: the invention utilizes the model predictive control method to dynamically regulate and control the diversion railings at the landing of the subway platform layer, has high prediction precision, fully considers the influence of the length change of the diversion railings under different time nodes on the passenger flow evacuation index, has high effect on the change of the passenger flow, reduces the congestion condition and improves the evacuation safety. And compared with the traditional manual regulation, the method reduces the labor and time cost and improves the regulation effect and the evacuation efficiency.
According to the invention, the simulation software based on the social force model is adopted to obtain the data set of the model, so that the length change scenes of various diversion railings can be simulated, a large amount of training set data in each scene can be obtained in a short time, and the efficiency is improved.
In the step 2, the length of the diversion rail is changed at each time node in the simulation experiment to obtain the input and output data set, the method is to simulate in a three-dimensional simulation scene of the subway platform layer constructed based on the social force model, and the social force model can simulate individual movement more accurately, so that the reliability of the obtained data is high.
Drawings
Fig. 1 is a schematic flow chart of a subway guide rail regulation and control method based on model predictive control in the embodiment of the invention;
FIG. 2 is a schematic diagram of a three-dimensional simulation scene of a subway platform layer with a simulated individual according to an embodiment of the invention;
FIG. 3 is a flowchart of a system identification method according to an embodiment of the invention;
FIG. 4 is a flow chart of a method for designing an MPC controller according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a subway diversion rail regulation and control system controlled by model prediction according to an embodiment of the invention;
FIG. 6 is a schematic diagram of a MPC control module system according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It should be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations. In addition, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
The fixed access places comprise a mall, a high-speed rail station, a subway station, a restaurant, a leisure and entertainment place and the like, and the space is airtight, and the building with the specific access is set.
Example 1:
in this embodiment, the procedure of the control method of the crowd diversion rail for the fixed location is described in detail by taking the diversion rail at the entrance of the subway platform.
Referring to fig. 1, fig. 1 is a flow chart of a control method for crowd guiding rail in a fixed place, and the method mainly controls subway guiding rail based on model prediction, and is described in detail as follows:
step S101: and constructing a three-dimensional simulation scene of the subway platform scene.
Individual motion simulation software based on a social force model is adopted, and the subway platform comprises a platform and stairs connected with a station hall, which correspond to a field scene.
It should be noted that, in the embodiment of the present application, the evacuation environment is a subway platform layer, and the evacuation crowd includes the people getting off the carriage and the people waiting at the platform layer.
The building is performed by using mass motion software, which is professional stream simulation software for simulating and analyzing the movement behavior of personnel in buildings, cities and other spaces. The method is based on an advanced social force model and a path planning algorithm, can provide a highly accurate artificial result of people flow, and helps users make scientific decisions.
When a scene is constructed, the software is opened and a geometric model of the building or space is created using a scene editing tool provided by the software. Building model files can be imported or elements such as walls, floors, stairs and the like of a building can be drawn manually. The space model of the building is ensured to accurately reflect the actual situation.
S102: and placing the simulation individuals into a three-dimensional simulation scene, simulating a plurality of diversion railing scenes with different length sequence changes at the landing entrance of the platform layer, obtaining passenger flow indexes at the bottleneck of each sampling moment, and constructing an input and output data set.
As shown in fig. 2, in the simulation, a simulation individual is put into a simulation scene, and at each time node of the time sequence, the diversion railings at the platform stair opening are subjected to random length change within a predetermined range, so as to obtain a plurality of groups of length sequences of the diversion railings at the subway platform stair opening, and passenger flow indexes such as passenger flow density, crowd pressure and the like at the bottleneck of the stair opening at each sampling moment, namely, an input/output data set.
According to the method, the length of the diversion rail is changed at each time node in the simulation experiment, the input and output data set is obtained, the simulation is carried out in the three-dimensional simulation scene of the subway platform layer constructed based on the social force model, the social force model can simulate individual movement more accurately, and therefore the reliability of the obtained data is high.
S103: and analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure.
Referring to fig. 3, the step S103 further includes S301, S302, S303, S304, where specific details are:
s301, input and output data acquisition and processing.
It can be understood that by defining the length sequence of the guide rail as input, the passenger flow indexes such as passenger flow density and crowd pressure at the bottleneck of the stair opening at each sampling moment are used as output, and an input/output data set is constructed. By removing noise, data normalization and other preprocessing, the accuracy and usability of the data are realized.
S302, selecting and identifying a model structure.
It will be appreciated that this example uses a linear ARX model as the system recognition object, and we assume that the real model of the system has the structure of the linear ARX model. The mathematical formula (1) of the linear ARX model structure is as follows:
where y and u are the linear ARX model output and input, a i And b j Is the regression coefficient of output and input quantity, f and g are the lag time of system output and input, n a And n b Respectively the order of the model, and ζ (t) is the modeling error.
It will be appreciated that the model order is selected in compliance with the AIC criterion, and the model order corresponding to the minimum AIC value is found out as the recognition model order by calculating the AIC values of the plurality of models.
S303, identifying model parameters.
It will be appreciated that this example uses the least squares method to solve the structure of the linear ARX model and the order coefficients. First, the following vector parameters are defined:
wherein θ is a vector parameter, represents a vector form of an estimated parameter represented by a vector formed by regression coefficients of input output quantities in a linear ARX model, and the purpose of the subsequent series of formula derivation is to calculate an optimal value of the vector, Φ t-1 The vector form of the input/output data before the time t is represented as the input data vector.
By the vector parameters described above, the linear ARX model can be rewritten as a matrix expression (2) as follows:
then define h m ＝max(n a ,n b ) Definition N is the total number of recognition data, N is defined as in k above, t=k+1.
Constructing a matrix, wherein the matrix is a rectangular array formed by a plurality of rows and columns, the form of the matrix is generally expressed by a capital letter, and the sequence data to be identified are arranged into a matrix form as follows:
Y＝[y k+1 y k+2 … y N ] T
Φ N ＝[Φ k Φ k+1 … Φ N-1 ] T wherein Y is an output matrix, Φ N For input data matrix, y and u are the output and input of the model, a i And b j Is the regression coefficient of output and input quantity, f and g are the lag time of system output and input, n a And n b Respectively the order of the model.
In the matrix, the upper right corner-1 represents the inverse of the matrix obtained by multiplying the first two matrices. The upper right corner is provided with T to represent the transposed matrix of the matrix, so that the basic symbol in mathematics is used for operation among the mathematic matrices, and the formula is obtained by the principle of least square method in the same way as the existing matrix operation and mathematical operation. The following corner marks provided at the same positions are the same as those explained herein.
Defining an optimal solution formula, and according to the principle of a least square method and an optimal solution formula (3), identifying model parameters with different orders and corresponding parameter estimation values to determine optimal model parameters, wherein the formula (3) is as follows:
wherein,representing a least squares estimate of the parameter θ, Φ N And representing the matrix, obtaining an optimal solution of the model according to the optimal solution formula according to the least square method principle, and thus obtaining model parameters of different orders and corresponding parameter estimation values through identification to determine optimal model parameters.
It can be understood that the optimal model parameters are determined by identifying model parameters of different orders and corresponding parameter estimation values.
S304, model verification.
Model verification is performed by using unused data sets, and indexes such as root mean square error and the like are adopted to evaluate the fitting capacity and the prediction performance of the model.
S104, constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object, and obtaining an optimal length sequence of the corresponding guide rail.
Referring to fig. 4, the step S104 further includes S401, S402, S403, where specifically:
s401, representing a state space of the model.
First, define the system state vector of the diversion rail on the basis of ARX model as follows:
x 1,t ＝y(t),h m ＝max(n a ,n b ),h＝2,3,…h m wherein X (t) is a state vector, the first digit in the lower right corner of each element in the vector represents the number for distinguishing between different states, h m Taking n for the highest order region of the state a n b The maximum of the two, t, represents the time step, since the state equation is in the form of a transition from the linear ARX model described above, the parameters inside are identicalThe parts a and b of (a) represent regression coefficients as in the linear ARX model above, and subscripts distinguish between different order coefficients, and the meanings of the subscripts and symbols are the same without special meaning or explanation.
It will be appreciated that the linear ARX model equation (1) can be expressed as a state space model (4) as follows:
wherein x (t+1) represents a state vector of the system at time t+1, x (t) represents a state vector of the system at time t, a is a constant matrix representing a state transition matrix for describing an evolution of a system state from t to t+1, B is a constant matrix representing an input matrix for describing an influence of an input on the system state, u (t) represents an input vector of the system at time t, Ω is a constant vector representing a constant term in a system model, ψ (t+1) represents an error term for representing an unmodeled dynamic or unknown disturbance in the system, it acts at time t+1, C is a constant matrix representing an output matrix for describing a linear relation between the state vector and the output.
In the above formula, the expression form of each matrix in the state space model is as follows, wherein A, B, C is a constant matrix:
wherein h is m-1 The subscript of the element in the first column of matrix A, which element is from 1 to h m ，a hm-1 Is the penultimate of the first column. The lower right corner of the matrix or vector indicates the dimension, the first element indicates several rows and the second indicates several columns. The meaning is the same in matrix B and matrix C.
S402, recursively predicting and outputting. From the current time, recursive state and output predictions are made using the MPC controller's prediction model. The prediction of the state and output at each predicted time is calculated based on the state and output at the last time.
It will be appreciated that the predicted output is calculated based on the state space model represented by the above linear ARX model structure, defining the following output vector:
in the above-mentioned method, the step of,for the optimal control sequence at the current moment, only the control signal at the first moment will be output to the guide rail system,/for>The elements in (a) represent prediction inputs at different moments, p represents a prediction time domain, q represents a control time domain, the relation between the two is satisfied with q < p, and the control quantity after the moment q in the optimized control sequence is not changed any more, namely u (t+i) =u (t+q) (i is larger than or equal to q)>A predictive state vector representing a system of guide rails, +.>The predicted output vector of the guide rail system based on the system state prediction at the time t is shown, u (t+1) represents the component of the optimal control sequence at the current time at the time t+1, and the same is true->And the component of the predicted output t+2 of the diversion railing system obtained by predicting the system state at the t moment is shown.
It can be understood that, by performing a recursive calculation according to the state space model (4) of the system, a prediction matrix model (5) of the prediction vector (t) can be obtained as follows:
in the following formula, E represents an identity matrix, the expression form of a prediction matrix model is as follows,
wherein (1)>Representing the prediction output matrix,/>Representing a gain matrix +_>Representing a prediction state transition matrix->The predicted input matrix is represented, the upper right corner of each element in the matrix represents the square of that element, and the lower right corner of the matrix represents the dimension.
It will be appreciated that simplifying the predicted output of the linear ARX modelThe following are provided:
wherein G is a coefficient transfer matrix, mapping the predicted input vector to the predicted output vector, y 0 And (t) represents a system steady-state output vector.
In the control of the guide rail, a certain error exists between the predicted output of the model and the output of the guide rail system of the actual controlled object, and the error of the output of the model needs to be compensated to the current predicted output to enable the algorithm to be closed loop. The deviation of the model shown in the formula (6) is compensated by a constant term of the model, and the constant deviation term a in the ARX model (1) after compensation 0 Will be calculated by the following formulaInstead, the matrix in the state space model (2) further transformed by equation (1), the Ω matrix in the state vector will have a corresponding change, and the rest of the computation will be unchanged.
S403, designing a control performance target and optimizing and solving. And calculating the optimal control input at the current moment according to the difference between the predicted output and the target value obtained by the system simulation and the objective function of the optimization problem. This optimal control input will be used for the control action at the current moment.
It can be appreciated that for an optimized control vector in a predictive modelSolving, wherein the optimal control vector is the same as the previous optimal control sequence, and in the optimal control, the optimal control sequence is defined as a sequenceA series of control signals over a period of time is represented as a vector, the elements of which represent the control signals at a certain moment in time. Consider that the system control target is to make passenger flow indexes such as passenger flow density track the set value vector +.>Defining a system performance index function (7) as follows:
in the above equation, the elements in the vector are all components at a certain time,represents the tracking set value vector, Q is about the output vector difference +.>Is a weight matrix of q×q with respect to the increment of the optimal control sequence, and the objective function design does not involve the control amount +.>Otherwise, the elimination of the final control target steady state error is affected. Desired output vector +.>Control sequence increment->The definition is as follows:
wherein the control increment Δu (t) =u (t) -u (t-1), u uper 、u low Vector of maximum and minimum values, deltau, representing control length of the guide rail system uper 、Δu low The maximum value and the minimum value vector of the increment of the control quantity of the system are represented, the increment constraint is set to enable the control process to be more stable, y uper 、y low Representing the maximum and minimum vectors of the predicted output.
y r Setting the tracking value y by the system s Calculated, when the set value y s When the step amplitude is suddenly changed too much, the system can track and respond in time to maintain the controlled target, the control input needs to be adjusted at the same time, and when the control quantity u is changed too much, the system stability is reduced, and the dynamic performance is reduced, so that the set value y needs to be set s Conversion to a gentle reference trajectory y r Equation (8) is as follows:
y r (t+i)＝y s (t+i)-α i (y s (t)-y(t))(i＝1,…,p) (8)。
defining a lower triangular matrix with the following matrix H being all 1 and the dimension being q multiplied by q; u (u) t-1 Representing the control quantity at the last moment of the system, defining a vectorThe relation satisfies (9) the following formula:
it can be appreciated that Q, R in the quadratic objective function (7) is a positive weighting matrix, and the optimization problem is convex optimization according to mathematical optimization principle analysis, so that an optimal solution is necessarily present. In the optimization process, the constraint is not considered, but a post constraint method is adopted, and the optimal control law can be obtained through a matrix inversion methodBy definition (9) can be expressed as formula (10),representing the control sequence->Is the first element in (c).
It can be understood that the prediction controller based on the ARX model is obtained through the definition and the derivation of the system, and the optimal length sequence value of the diversion rail in the evacuation process can be obtained through regulating and controlling the diversion rail system of the controlled object. The obtained optimal control vector is a series of optimal control signals for the controlled object, which are obtained by the control system, and the controlled object is a guide rail system, so that the optimal length sequence value of the guide rail is the practical meaning of the optimal control vector in the control system, and only progressive statement is made here without specific formula operation, so that symbol representation is not needed.
Example 2:
the embodiment discloses a control system for crowd's water conservancy diversion railing in fixed place.
As shown in fig. 5, a control system for crowd diversion railings in a fixed location includes a scene construction module, a data set construction module, a system identification module, and an MPC control module;
a scene construction module configured to: constructing a three-dimensional simulation model of a scene provided with a fixed entrance and exit place, wherein the fixed entrance and exit place is a subway platform layer;
a dataset construction module configured to: placing the simulation individuals into a three-dimensional simulation model, simulating a plurality of diversion railing scenes with different length sequence changes at the landing entrance of a platform floor, obtaining passenger flow indexes at the bottleneck of each sampling moment, and constructing an input and output data set;
a system identification module configured to: analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure;
an MPC control module configured to: and constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object, and controlling the railing length of each time node in the evacuation process.
It should be noted that, referring to fig. 6, the system structure of the MPC control module is schematically shown. The MPC controller module S601 is used for regulating and controlling the diversion railing system module S602. The system comprises a MPC controller module S601, a diversion railing system module S602, a prediction model and constraint conditions are combined, and an optimal control input is calculated and generated by using an optimization solver through inputting a current system state and an external reference signal to realize regulation and control of the module S602. The S602 module includes sensors for measuring system status, actuators for implementing control inputs, and other auxiliary devices.
Example 3:
an object of the present embodiment is to provide a computer-readable storage medium.
A computer-readable storage medium having stored thereon a computer program which when executed by a processor performs the steps in a method of controlling a crowd diversion rail for a stationary venue as described in embodiment 1 of the present disclosure.
Example 4:
an object of the present embodiment is to provide a control device for crowd diversion railings in a fixed location. The control device is an electronic device, and comprises at least one processor, and at least one memory and a bus connected with the processor, wherein the processor and the memory complete communication through the bus, and the processor is used for calling program instructions in the memory to execute the control method for the crowd diversion handrail in the fixed place, which is described in the embodiment 1.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (3)
1. The control method for the crowd diversion railing for the fixed place is characterized by comprising the following steps of:
step 1: the method comprises the steps of constructing a three-dimensional simulation scene provided with a fixed entrance place and a subway platform based on individual motion simulation software of a social force model, wherein the fixed entrance place is built by the three-dimensional simulation scene, the three-dimensional simulation scene comprises a platform and stairs connected with a station hall, and the individual comprises a carriage getting-off crowd and a platform layer waiting crowd;
step 2: putting an analog individual into the three-dimensional simulation scene, simulating a flow guide railing scene with various length sequence changes at the fixed entrance and exit to obtain passenger flow indexes at bottleneck positions of all sampling moments, and constructing an input/output data set, wherein the input/output data set is that the flow guide railing at the landing entrance performs random length changes within a preset range at each time node of a time sequence to obtain the length sequences of a plurality of groups of flow guide railings at the landing entrance of a subway station, and the passenger flow indexes at the bottleneck positions of the landing entrance of each sampling moment, and the passenger flow indexes comprise passenger flow density and/or crowd pressure;
step 3: analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure, wherein the method comprises the following steps:
step 301: input and output data acquisition and processing, namely, an input and output data set is constructed by defining a length sequence of a guide rail as input and using a passenger flow index at the bottleneck of a stair opening at each sampling moment as output, and the accuracy and usability of the data are realized by removing noise and carrying out data normalization preprocessing;
step 302: selecting and identifying a model structure, and using a linear ARX model as a system identification object, wherein the linear ARX model structure is as follows:where y and u are the output and input of the model, a i And b j Is the regression coefficient of output and input quantity, f and g are the lag time of system output and input, n a And n b Respectively, the orders of the models, and ζ (t) is a modeling error, wherein the selection of the orders of the models complies with an AIC criterion, and the model orders corresponding to the minimum AIC values are found out to serve as identification model orders by calculating AIC values of a plurality of models;
step 303: model parameter identification, adopting a least square method to solve the structure and each order coefficient of the linear ARX model, and obtaining model parameters of different orders and corresponding parameter estimation values through identification so as to determine optimal model parameters, wherein the method comprises the following steps of:
step a: the vector parameters are defined and the vector parameters are defined,
wherein θ is a parameter vector, representing the vector form of the estimated parameter, Φ t-1 The vector form of input and output data before the moment t is represented by the input data vector;
step b: the linear ARX model can be rewritten as an expression of a matrix by the vector parameters:
step c: definition of h m ＝max(n a ,n b ) Defining N as the total number of the identification data, and arranging the sequence data to be identified into a matrix form:
wherein Y is an output matrix, phi N Is an input data matrix;
step d: defining an optimal Jie Gong formula:wherein (1)>Representing a least squares estimate of the parameter θ, Φ N The representative matrix is used for obtaining an optimal solution of the model according to the optimal solution formula according to the least square method principle, so that model parameters of different orders and corresponding parameter estimation values are obtained through identification, and the optimal model parameters are determined;
step 304: model verification, namely performing model verification by using an unused data set, and evaluating fitting capacity and prediction performance of the model by adopting a root mean square error index;
step 4: the identification model is used as a prediction model and a controlled object to construct an MPC controller and a control system, and a corresponding optimal length sequence of the guide rail is obtained, which comprises the following steps:
step 401: a state space representation of a model, comprising: defining a state vector of a diversion rail system on the basis of the linear ARX model as follows:
x 1,t ＝y(t),h m ＝max(n a ,n b ),h＝2,3,…h m where X (t) is a state vector and t represents a time step, the linear ARX model structure is represented as a state space model as shown below:wherein x (t+1) represents a state vector of the system at a time t+1, x (t) represents a state vector of the system at a time t, a represents a state transition matrix, B represents an input matrix, u (t) represents an input vector of the system at a time t, Ω is a constant vector, ψ (t+1) is an error term, and C represents an output matrix;
the expression form of each matrix in the state space model is as follows:
step 402: recursively predicting output, the state and the output of each predicted time being calculated based on the state and the output of the last time, comprising: calculating a prediction output based on the state space model, defining an output vector:
wherein (1)>Is a constant combination vector, ">For the optimized control sequence at the current moment, only the control signal at the first moment can be output to the guide rail system, p represents a prediction time domain, q represents a control time domain, the relation between the two is smaller than p, and q represents that the control quantity after the q moment in the optimized control sequence can not change any more, namely u (t+i) =u (t+q) (i is larger than or equal to q), and the control quantity is larger than or equal to q>A predictive state vector representing a system of guide rails, +.>The predicted output vector of the diversion rail system, which is obtained based on the system state prediction at the time t, is shown;
performing recursive calculation according to the state space model to obtain a prediction matrix model of a prediction vector (t) as follows:
the expression form of the prediction matrix model C is as follows:
wherein,representing a predicted output matrix, E representing an identity matrix, < >>Representing a gain matrix +_>Representing a prediction state transition matrix->Representing a predicted input matrix;
simplifying the predictive output of the linear ARX modelWherein G is a coefficient transfer matrix, y 0 (t) represents a system steady-state output vector;
step 403: designing a control performance target and optimizing and solving, and optimizing control sequences in a prediction modelSolving to obtain a predictive controller based on the linear ARX model, and regulating and controlling a controlled object diversion railing system to obtain an optimal length sequence value of the diversion railing in the evacuation process, wherein the optimal length sequence value is used for controlling actions at the current moment and comprises the following steps: defining a system performance index function:
wherein Q is the difference of the output vectorsIs a unit weight matrix of size p x p, R is a weight matrix of size q x q for the optimal control sequence increment,/o>Indicating tracking settingsVector;
obtaining the expected output vectorControl sequence increment->
Wherein the control increment Δu (t) =u (t) -u (t-1), u uper 、u low Vector of maximum and minimum values, deltau, representing control length of the guide rail system uper 、Δu low Maximum and minimum vectors representing system control quantity increment, y uper 、y low Representing the maximum and minimum vectors of the predicted output.
2. A control system for a crowd diversion rail for a fixed location, the control system applying the control method for a crowd diversion rail for a fixed location of claim 1, comprising: the system comprises a scene construction module, a data set construction module, a system identification module and an MPC control module;
the scene construction module is used for constructing a three-dimensional simulation model of a scene provided with a fixed entrance and exit place, and the fixed entrance and exit place is a subway platform layer;
the data set construction module is used for putting a simulation individual into the three-dimensional simulation model, simulating a plurality of diversion railing scenes with different length sequence changes at the stair opening of the subway platform layer, obtaining passenger flow indexes at the bottleneck of each sampling moment, and constructing an input/output data set;
the system identification module is used for analyzing and processing the data set by adopting a system identification method to obtain an identification model of the optimal parameters of the optimal structure;
and the MPC control module is used for constructing an MPC controller and a control system by taking the identification model as a prediction model and a controlled object, and controlling the rail length of each time node in the evacuation process.
3. A controlgear that is used for crowd's water conservancy diversion railing in fixed place, its characterized in that: the control method for the crowd diversion railing for the fixed place comprises at least one processor, at least one memory and a bus, wherein the memory and the memory are connected with the processor, the processor and the memory are communicated through the bus, and the processor is used for calling program instructions in the memory to execute the control method for the crowd diversion railing for the fixed place according to claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311249023.1A CN117311188B (en) | 2023-09-26 | 2023-09-26 | Control method, system and equipment for crowd diversion railings in fixed places |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311249023.1A CN117311188B (en) | 2023-09-26 | 2023-09-26 | Control method, system and equipment for crowd diversion railings in fixed places |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117311188A CN117311188A (en) | 2023-12-29 |
| CN117311188B true CN117311188B (en) | 2024-03-12 |
Family
ID=89236594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311249023.1A Active CN117311188B (en) | 2023-09-26 | 2023-09-26 | Control method, system and equipment for crowd diversion railings in fixed places |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117311188B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117590766A (en) * | 2024-01-19 | 2024-02-23 | 青岛理工大学 | Control method and device for angle adjustment of channel inlet guide rail |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2644794A1 (en) * | 2008-11-18 | 2010-05-18 | Sharon Yetman | Subway barrier wall with enhanced passenger flow |
| CN107065902A (en) * | 2017-01-18 | 2017-08-18 | 中南大学 | UAV Attitude fuzzy adaptive predictive control method and system based on nonlinear model |
| CN114063455A (en) * | 2021-11-11 | 2022-02-18 | 中南大学 | Soil pressure balance shield machine propulsion speed control method, modeling method and device |
| CN115577574A (en) * | 2022-12-08 | 2023-01-06 | 西南交通大学 | Method, device and equipment for calculating position of diversion rail and readable storage medium |
| CN115719294A (en) * | 2022-11-23 | 2023-02-28 | 厦门大学 | Indoor pedestrian flow evacuation control method and system, electronic device and medium |
| WO2023061959A1 (en) * | 2021-10-14 | 2023-04-20 | Hitachi, Ltd. | Control of a complex environment |
| CN116047897A (en) * | 2022-11-01 | 2023-05-02 | 大连理工大学 | Gas turbine predictive control method based on parameter self-adaptive disturbance rejection controller |
-
2023
- 2023-09-26 CN CN202311249023.1A patent/CN117311188B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2644794A1 (en) * | 2008-11-18 | 2010-05-18 | Sharon Yetman | Subway barrier wall with enhanced passenger flow |
| CN107065902A (en) * | 2017-01-18 | 2017-08-18 | 中南大学 | UAV Attitude fuzzy adaptive predictive control method and system based on nonlinear model |
| WO2023061959A1 (en) * | 2021-10-14 | 2023-04-20 | Hitachi, Ltd. | Control of a complex environment |
| CN114063455A (en) * | 2021-11-11 | 2022-02-18 | 中南大学 | Soil pressure balance shield machine propulsion speed control method, modeling method and device |
| CN116047897A (en) * | 2022-11-01 | 2023-05-02 | 大连理工大学 | Gas turbine predictive control method based on parameter self-adaptive disturbance rejection controller |
| CN115719294A (en) * | 2022-11-23 | 2023-02-28 | 厦门大学 | Indoor pedestrian flow evacuation control method and system, electronic device and medium |
| CN115577574A (en) * | 2022-12-08 | 2023-01-06 | 西南交通大学 | Method, device and equipment for calculating position of diversion rail and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117311188A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109816095B (en) | Network flow prediction method based on improved gated cyclic neural network | |
| CN117311188B (en) | Control method, system and equipment for crowd diversion railings in fixed places | |
| CN111257934B (en) | Seismic oscillation peak acceleration prediction method based on second-order neuron deep neural network | |
| CN109839824A (en) | A kind of Delays In Networked Control System compensation method based on PREDICTIVE CONTROL | |
| Alfred | A genetic-based backpropagation neural network for forecasting in time-series data | |
| Tian et al. | Time-delay compensation method for networked control system based on time-delay prediction and implicit PIGPC | |
| CN104899432A (en) | Kernel function combination-based PSO-LSSVM fluctuating wind speed prediction method | |
| CN107704426A (en) | Water level prediction method based on extension wavelet-neural network model | |
| CN105404152A (en) | Flight quality prediction method for simulating subjective evaluation of pilot | |
| CN114819068A (en) | Hybrid target track prediction method and system | |
| Dudzik et al. | The structural reliability analysis using explicit neural state functions | |
| CN109800517B (en) | Improved reverse modeling method for magnetorheological damper | |
| CN108566178A (en) | A kind of random opportunistic network characteristic value filtering method of unstable state | |
| CN106067075B (en) | Building energy load prediction model building and load prediction method and device | |
| CN110779526B (en) | Path planning method, device and storage medium | |
| JPH06208392A (en) | Pattern recognizing method and its device | |
| CN117076887A (en) | Pump station unit running state prediction and health assessment method and system | |
| KR20200105365A (en) | Method for reinforcement learning using virtual environment generated by deep learning | |
| CN115775055A (en) | Method, device, equipment and medium for predicting personnel evacuation time of multi-story building | |
| CN113962431A (en) | Bus load prediction method for two-stage feature processing | |
| CN113112092A (en) | Short-term probability density load prediction method, device, equipment and storage medium | |
| CN108960406B (en) | MEMS gyroscope random error prediction method based on BFO wavelet neural network | |
| CN112446516A (en) | Travel prediction method and device | |
| CN114297940B (en) | Method and device for determining unsteady state reservoir parameters | |
| Wu et al. | Chaotic newton-raphson optimization based predictive control for permanent magnet synchronous motor systems with long-delay |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |