JPS59138577A - Controller for group of elevator - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an elevator group management device, and more particularly to an elevator group management device equipped with a control means suitable for providing priority service to crowded floors and preventing overcrowding.

Conventionally, in order to give preferential service to crowded floors, etc., the service priority level for each floor was determined based on the number of hall calls per unit time, the number of people boarding the elevator on each floor, etc. However, the above information is information that is greatly affected by the elevator group management algorithm, and there is a drawback that if the elevator group management algorithm is changed to another group management algorithm, the crowded floor will be changed to another floor.

Recently, in order to find a service priority level that is not affected by elevator group management algorithms, service priority levels have been developed based on information indicating the essential degree of congestion that is not affected by group management algorithms, such as the boarding order from each floor. There is a way to calculate.

However, with this method, it is not affected by the group management algorithm; instead, it is
The service priority level value is not optimal.

An object of the present invention is to provide means for calculating optimal service priority parameters for individual group management algorithms.

A feature of the present invention is that group management simulations are performed multiple times using multiple group management algorithms for in-building traffic that is not affected by the elevator group management algorithm, thereby providing optimal services for each individual group management algorithm. By calculating the priority level, a service priority level indicating the degree of congestion on each floor can be calculated more efficiently than conventional service priority level calculation means.

Hereinafter, the present invention will be explained in detail with reference to FIGS. 1 to 19. In the description of the embodiment, first, the hardware configuration for realizing the present invention will be described, then the overall software configuration and its control concept will be described, and finally, the explanation will be given using a flowchart for realizing this control concept.

FIG. 1 shows the overall hardware configuration of an embodiment of the present invention.

The elevator group management control device MA includes a microcomputer Ml that controls elevator operation and a microcomputer M2 that controls simulation. ! Data is communicated between =M2 by the serial communication processor 5DAC via the communication line CMc. Further, detailed configuration and operation explanation regarding this SDA can be found in Japanese Patent Application Laid-Open No. 56-37972 and Japanese Patent Application Laid-open No. 56
-37973.

The microcomputer M1, which controls elevator operation, receives the call signal HC from the hall call device tHD through a parallel input/output circuit Pi.
What are the machine control microcomputers E and -E (here, it is assumed that there are n elevators) that are called in through A and also control the elevators such as opening/closing doors and commanding car acceleration/deceleration? , a serial communication processor SDA similar to the above, ~
8DA, and are connected via communication lines CM and -CMA.

On the other hand, the microcomputer M2 has a setting device PD that provides information necessary for determining the optimum operation control parameters for simulation.
The signal PM from the input/output circuit PiA is inputted via the parallel input/output circuit PiA.

The machine control microcomputers E, ~E, still have the control input/output element Eo, which consists of call information necessary for control, various elevator safety limit switches, relays, and response lamps.
o, ~EiO,.

Note that 5DAL is a serial communication processor, 810. ~5I
OB is a signal line.

FIG. 2 shows the overall configuration of the software, which is roughly divided into operation control system software SF1 and simulation system software SF2.

The operation control system software S'FI is an operation control program 5F14 that directly commands and controls elevator group management control such as call assignment processing and elevator distributed standby processing.
Consists of. The input information for this program is the elevator position, direction, etc. sent from the machine control program.
Elevator control data table 8F11 for car calls, etc.
These include a hall call height table SF'12, an elevator specification table 5F13 such as the number of elevators to be managed, and optimal operation control parameters calculated and output by the simulation software SF2.

On the other hand, the simulation software SF2 consists of the following processing programs. That is, (1) Data collection program 5F20: A program that samples the contents of hall call and elevator control data tables online at regular intervals and collects data for simulation, and mainly collects traffic demand by destination floor. do.

(2) Simulation data calculation program 5
F22: This is probably a program that calculates simulation data by taking into account the contents of the online sampling data table 8F21 collected from the data collection program and the contents of the sampling data table of past time periods. (3) Traffic demand Classification program 8F33... Inputs traffic demand by destination and time information obtained from simulation data table 5F24, calculates traffic volume within the building,
The program divides traffic demand into eight categories: before lunch, during lunch, after lunch, normal, normal congestion, leaving work, and quiet.

(4) Simulation execution program 5F35・
...Simulation data table 5F24, traffic demand classification table 5F34, and elevator specification table 5F
25, perform one line of simulation, and save the results to the simulation statistical processing data table 5F3.
Output to 6.

(5) Various curve calculation programs based on simulation 5F23: A statistical processing data table 5F36 based on simulation is input, a simulation is performed for each of a plurality of predetermined parameters, and various curve data tables 5F26 are calculated and output. Various curve data table 5
F26 is, for example, an average waiting time curve table, a power consumption curve table, etc.

(6) Optimal operation control parameter calculation program 5F2
7...Target value table 5 set from the various curve data tables 5F26 and the externally provided target setting device PD
Inputs F28 and calculates and outputs the optimal operation control parameter 5F29 according to power saving. (7) Statistical processing calculation program 5F32... Calculates stop probability, fullness prediction, etc. from the statistical processing data table 5F36 by simulation. Statistics table 5F3
Output to 7.

Next, a method of calculating optimal operation control parameters using simulation will be explained.

A stop call evaluation function is used as a call allocation method.

The concept of this stop call evaluation function was published in Japanese Patent Application Laid-Open No. 52-4724.
No. 9, and JP-A-52-126845.

The stop call evaluation function is: φ=TI, 1. It is expressed as llαTα Tα=ΣβS. Here, T-8 is the evaluation value of waiting time, Tα
is a stopped call evaluation value, and α is a weighting coefficient between the waiting time evaluation value T and the stopped call evaluation value Tα. This α is called an area priority parameter. Further, β is a weighting coefficient for the stop call S (call to be serviced) on the floor adjacent to the generated hall call, and is set to, for example, 0 to 20.

FIG. 3 shows the table structure of the operation control system software used in one embodiment of the present invention, which is roughly divided into blocks of an elevator control table 5F11, a hall call table 5F12, and an elevator specification table 5F13. The tables in each block will be described each time the operation control program described below is explained.

First, the operation control system program will be explained, and then the simulation system program will be explained. It is assumed that the programs described below are managed under a system program that divides programs into a plurality of tasks and performs effective control, that is, an operating system (O8).

Therefore, programs can be started freely from the system timer or from other programs.

Now, FIGS. 4 to 8 show flowcharts of the operation control program. In addition, the especially important calculation program for predicted elevator arrival time among the operation control programs, service priority level program, fullness prediction program, and call assignment program will be explained.

FIG. 4 is a flowchart for calculating the predicted arrival time of an elevator to an arbitrary floor, which is to be used as basic data for calculating the waiting time evaluation value. This program is activated periodically, for example, every 1 second, and calculates the predicted arrival time from the current position of the elevator to the desired floor for all floors, all elevators, and all directions.

In FIG. 4, step E10 is the number of loops in the direction of elevators, step E20 is the number of loops in the number of elevators, and steps EIO, E20, and E120 are for loop processing for all numbers and directions of elevators. show. In step E30, the floor is set to the elevator position. Next, in step E40, initial values are set in the time table T for work. This initial value may include the number of seconds left after the door is opened or closed before the elevator can depart, or the time required to start the elevator when it is stopped. Next, the program advances one floor (step E50), and determines whether the set floor is the same as the elevator position (step E60). If they are the same, it means that the predicted arrival time table for one elevator has been calculated, and the process jumps to step E120 and repeats the same process for the other elevators. On the other hand, if 'NO' in step E60, the time table T
The first floor traveling time T7 is added to (step E70).

Then, this time table T is set in the arrival time table (step E80).

Next, it is determined whether there is a corner call or an assigned hall call, that is, a call that should be serviced by the elevator of interest.
Floor standard stop time T6 and door time control parameter Di for i floor j direction of door time control parameter table Tt
Modify the time table T using j. This method is
A value obtained by multiplying the door time control parameter Dij by the first floor standard stopping time Tll is added to the time table (step E100). Next, the process goes to step E50 and the above process is repeated for all floors and directions.

On the other hand, if it is 7NO in step E90, the stop probability Pij for the i-floor j direction in the stop probability table Tp, the door time control parameter Dij for the i-floor j direction in the door time control parameter table Tt, and the 1st floor standard stop time. Ts is used to modify the time table T. In this method, the product of the stop probability PIj and the door time control parameter Dij multiplied by the first floor standard stop time Tg is added to the time table T (step EIIO ).Next, jump to step E50 and repeat the above process for all floors and directions. Note that step E70 and step E10
0 and the first floor running time T at step E110, and 1
The floor standard stopping time Ts is given as one of the optimal operation control parameters by the simulation system software, and is based on the door time control parameter Dij and the stopping probability P.
ij is given from the statistical table 5F37.

FIG. 5 is a flowchart of the service priority level program, which is activated when a hall call occurs. In step F10, the floors are looped and it is determined on which floor the hall call has occurred (step F20)
. Next, determine the direction of the hall call (step F30
and step F40). Then, the service priority level Sij for floor i and direction j determined above in the service dimming level table TU is added to the hall call elapsed time for floor i and direction j in the hall call elapsed time table shown in FIG. Step F50). Furthermore, the service priority level Sij is given from the statistical table 5F37 (FIG. 2) of the simulation software.

FIG. 6 is a flowchart of the crowd prediction program. First, we will explain how to start this program. The presence or absence of a hall call is checked by starting every 0.5 seconds, for example. Activation of this full-occupancy prediction program is performed in step 0 above.
It takes place a little faster than 5 seconds. That is, immediately after the fullness prediction program is started, the program for checking whether a hall call has occurred is started. 0.5 above
If the time is set to 0.1 seconds, for example, it is safe to assume that there are no passengers during this time.

Therefore, the time between starting the fullness prediction program and starting the program for checking whether a hall call has occurred is made as short as possible.

Now, in the flowchart of the crowd prediction program,
Make initial settings for the elevator using the mass and step GIO. Here, it is set to =1. Next, the number of people in the elevator car is set to PK (step G20), and the floor at the elevator position is set to i (step G30). Then, it is determined whether there is a hall call on the i floor (step G40).
)Q If there is no car call, it is determined whether there is a car call on the i floor.Step G50)) If there is a car call on the i floor, the i floor j of the fullness prediction table TI is determined from the corner number FK.
The predicted fullness of the call in the direction Fij (C) is subtracted, (step G60) f, and the floor i in step G100 is advanced by one. If there is no car call on the i floor in step G50, the process jumps to step G100.

If there is a hall call on the i floor in step G40, it is determined whether there is a car call on the third floor (step G70).
Add the predicted hall call fullness value F ij (H) in the direction of floor j, and further add the above hall call fullness predicted value F ij (C)
To subtract step G30), jump to step G100. If there is no car call on the first floor in step G70, the predicted hall call fullness value F ij (H) is added to the number FK of empty cars (step G90), and the i floor is advanced by one in step G100.

Next, it is determined whether floor i is the top floor or the bottom floor (step Gl 10). If "No", the process jumps to step G40; if "YES", the value of the number of white people in the car FK is equal to the rated number of people in the car. It is determined whether it is within 90% of (step G120). If it is within 90 degrees, the elevator K is disabled for service (step G130), and it is determined whether all elevators are finished (step Gl 40).
'l. If step G120 is "YES", the above step G140 is determined. If the program has been completed for all elevators, this program ends; if not for all elevators, the program moves to the next elevator K (step G150) and then jumps to step G20 to perform the above processing. Note that step G60. G80. G90
The car call fullness prediction Frj (C) and the hall call fullness prediction Fij (H) are given from the statistical table.

Figure 7 is a flowchart of the call allocation program.
The program is started when a hall call occurs. In this program, the call allocation algorithm is in step H.
50 is a long-waiting call minimization call allocation algorithm. When a hall call occurs, first, in step HIO, the generated hall call is read from the outside. and,
Steps H20 and H80, and steps H30 and ) (70) perform loop processing for floors and directions. Step H40 determines whether there is a generated hall call. If not, jumps to step I (70, and all Step H40 is “YES”.
``If so, the long-waiting call minimization call allocation algorithm of step H50 is performed to allocate the call to the optimal elevator (step H60).

FIG. 8 is a processing flowchart of the long-waiting call minimization call allocation algorithm. In order to determine which elevator is optimal, loop processing is performed in steps H50-1 and H3O-6 using the elevators determined to be serviceable by the crowd prediction program or the like. The processing inside the loop is −
First, in step H3O-2, the maximum predicted waiting time T□8 of the allocated hall calls on the front floor including the generated hall call is calculated. Note that the predicted waiting time is the sum of the hall call elapsed time indicating the elapsed time from the occurrence of the hall call to the present time and the predicted arrival time. Next step H3O-3
Then, a stop call evaluation value Tα is calculated from stop calls on predetermined floors before and after the hall call that has occurred, and a stop call evaluation function φ (φ
""T mat-αTα) is calculated (step H3
O-4), and selects the smallest elevator among this evaluation function φ (step H3O-5). If the above process is executed for all serviceable elevators KK, the elevator K with the optimal evaluation value will be selected by the calculation in step H3O-5.

Next, FIGS. 9 and 11 show table configurations of the simulation software.

FIG. 9 shows the optimum operation control parameter 8F29, various curve data tables 5F26, target value table 5F28, sampling data table 8F21, simulation,
The configuration of the cheetah table 5F24 for cars, the elevator specification table 5F25 (not shown since it is the same as in FIG. 3), and the traffic demand classification table 5F34 is shown. Figure 10 is a statistical processing day using simulation! Table 5F36
This table is composed of an elevator direction reversal count table TD, an elevator stop count table Tax, a hall call count TII, a car call count table Tc, a table for the number of people boarding the elevator TI, and an elevator neighbor a table To. Figure 11 is a statistical table, 8 F 37
This table shows the stop probability Chiful Tp, 7
R member prediction table Tt, f-bis priority level table T
It consists of a door time control parameter table T.

Next, the simulation software program will be explained. First, the data collection program is started at regular intervals (for example, 1 second) and for a fixed period of time (for example, 1 second).
When data is collected (for example, for 10 minutes), the data is stored in the sampling data table SF21 shown in FIG. tr, 1
Data such as the standard stop time of 1 s are collected. Travel time for the first floor of the above elevator 1. and one standard stop count t8
After the sampling time is over, the travel time for the first floor can be calculated by dividing the travel time by the number of floors traveled, and the standard stop time for i times is calculated from the number of elevator stops and the time between doors (stop time). Can calculate. In addition, the collected data is
When the sampling time ends, the above calculation is performed,
And sampling data table 5F210 in FIG.
The data are stored in an online measurement table and a time zone table. This online total i+J cheetah table adds the subscript "new" to the item name like Qll@W*trno, tan*v, and the time zone table has Qota + tr. td+ts. o, 5 like t6
It is written with the subscript d added.

The 8F22 simulation data calculation program is activated periodically, and the simulation data is predictively calculated using online measured data and past data by adding an appropriate coupling variable γ.

For example, the destination traffic volume is calculated using the following formula.

QPro”γQ, , + (1−γ)Q, 1° Therefore, the larger the joint transformation γ is, the greater the weight of the destination traffic volume data measured online becomes.

Note that subscripts pr and e are added to the predicted data. Similarly to the above, the predicted data of the first floor running time and one standard stopping time j rPr* + j *Pre
is also calculated. Also, this ttPra + j sPr
The data marked with * is the optimum operation control parameter 8F shown in Figure 9.
29 T.

It is set in the Ts table. Then, a simulation execution program is started based on the predicted data calculated by this program.

Furthermore, based on the above forecast data, the forecast data of destination traffic volume is further divided into eight types of traffic demand based on time information: dispatch, before lunch, during lunch, after lunch, normal, normal congestion, leaving work, and quiet. This is a transportation demand classification program.

FIG. 12 is a flowchart of the simulation execution program. There is an area priority parameter which is a weighting coefficient as a simulation parameter, and a simulation is executed for each parameter case. - First, simulation data such as destination traffic volume is set (step 5 CIO). Next, step 5C
In step 8C30, the area priority parameter is set and a simulation is executed (step 8C30). Note that the area priority parameter α is, for example, 0, 1, 2°3.4.5
It is. The simulated results for each case are then stored for each parameter (step 5C50).

In addition, the simulation results are stored as average waiting time, power consumption value, longevity rate, fullness prediction harameter, and door time fl.
These include ill control parameters, service priority level parameters, and outage probability parameters. After completing the simulation for all the above cases (step 5C40),
The "optimal area" parameter, the service performance, and various control parameters are calculated (step 5C60).

The execution of the simulation of Stella 7'5C30 will be described in detail using the flowchart of FIG. 13.

First, input processing of the area priority parameter α is performed (step Al0). Next, simulation variables are initialized (step A20).

For example, this includes the initial setting of random numbers for passenger generation processing, which will be described later, and the initial setting of a hall call table. Step A30
Now let's initialize the statistical processing variables. Here, in step A40 for initializing the statistical table, etc., the time is set to zero, and in step A90, the time is added to a predetermined value, which is set to 1 here. ), it is determined whether this time exceeds a predetermined time (step A100). From step A50 until the above time exceeds the predetermined time, Stella 7
'Perform the process of A90. In step A5'O, a passenger generation process is performed, and step A60 is a group management process in which a hall call is assigned when a hall call is generated. Perform machine number processing such as. Step A80
is a statistical data collection process that collects statistical data.

Here, steps A50 to A70 will be explained. The passenger generation process in step A50 is
The passenger generation floor i and the passenger destination floor 12 are determined by random numbers based on the destination traffic volume prediction data obtained by the John data calculation program 5F22. Furthermore, the number of passengers generated from the 12th floor to the 12th floor is determined using the above random number, and a hall call is generated on the 11th floor. Next, step A
The group management process 60 performs call assignment if the hall call occurs. The call assignment method is the same as that explained in the operation control program, and the car number processing in step A70 includes processing such as the running state of the elevator, the stopped state, door opening/closing, and car call generation.

Next, regarding the statistical data collection process in step A80,
This will be explained using the flowchart shown in FIG. Step A
30-1 to A30-4-1: Then, elevator direction j1 area completion priority parameter α, traffic demand classification M, floor i
is the number of loops from step A30-6 to A30-
Up to 9, the loop completion determination for each of the above-mentioned j and α+M+i is performed. Step A30-5 collects statistical data (the number of stops of the car, the number of hall calls, the number of car calls, the number of people getting on, the number of people getting off, etc.) for each of the above j, α1M, +.

FIG. 15 to FIG. 19 are flowcharts regarding the stop probability, fullness prediction, service priority level, and door time control parameters of the statistical processing calculation program. FIG. 15 is a flowchart of the stop probability calculation program, in which the number of loops for floor i is set (step BPIO)
. Next, set the number of loops in direction j (step BP
20).

Then, the stop probability P ij of the i-th floor and the j-direction is calculated from the number of direction reversals d in the elevator direction reversal table TD and the number of elevator stops eI (step BP30). The stop probability Pij is calculated using the following equation.

Here, α is a predetermined coefficient.

FIG. 16 is a flowchart of the fullness prediction calculation program, in which the number of loops for floor i is set (step BFIO
). Next, set the number of loops in direction j (step B
P20). Then, the i floor j of the hall's faulty table T□
The predicted fullness value F ij (H) of the hall call on the i floor in the j direction is calculated from the number of hall calls in the direction fl and the number of people on the i floor in the j direction of the elevator boarding number table T+. Still, the number of car calls for the i floor in the j direction of the car call number table Tc, and the number of people getting off the elevator table T. Calculate the predicted fullness value F ij (C) for the car call from the i floor j direction of the passenger RCij to the 1st floor (step BP'30
). The predicted full-occupancy value Fij (H) for hall calls and the predicted full-occupancy value Fij (C) for car calls are calculated by the following equations.

Here, α2 and α are predetermined coefficients.

m171dl-, ]:, Elevator fullness prediction value GK
In the flowchart of the calculation program, first, the number of loops in the elevator is set (step BFao-1).
. Next, a predicted elevator fullness value GK is calculated (step BP'30-2>. The calculation formula is as follows.

GK = ΣpH1 revision - Σp, 2. (c) I2 Here, i,: The assigned hall call floor 12 that is served by elevator K: The call floor j that is served by elevator K: The direction of the elevator F 111 (H) and FI21 (C) are the fullness prediction team.
Included in F/I/T.

FIG. 18 shows a service priority level calculation program. First, floor i is set to the number of loops (step B5l0
). Next, set direction J to loop rotation (BS20)
. Then, the service priority level Sij for the i-floor j direction is calculated from the number of hall calls f1 for the i-floor j direction of the hall call table Tg and the number of passengers Rh 14 for the i-floor j direction of the elevator boarding number table T. Step B5
30).

The 1-bis priority level SIj is expressed by the following equation.

Here, α4 is a predetermined coefficient.

FIG. 19 shows the door time control parameter calculation program. First, floor i is set to the number of loops (step B
DIO). Next, the direction j is set to the number of loops (step BD20). Then, the hall call number f++ on the i floor in the j direction of the hall call number table TI and the elevator number of people table T+.

The number of people boarding the car on the i floor in the j direction, the car call number g on the i floor in the j direction of the car call number table Tc, and the number of people getting off the elevator table T. The door time control parameter Di is calculated from the number of people getting off on the i floor in the j direction (step BD).
30). Door time? The aL wholesale parameter is expressed by the following equation.

Here α, 2 and α are sufficient coefficients.

The above statistical processing is performed and each of the above values is used in the operation control program.

According to the present invention, (1) the optimal service priority level value for the 1A elevator group management algorithm, for example, the minimum waiting time group management algorithm, the optimal service (f first bell value) for the energy saving group management algorithm; Since it can be calculated, the congestion degree of a crowded floor can be calculated in accordance with the group management algorithm, and efficient elevator group management is possible.

(2) Since the service priority level is calculated for each floor rather than for each direction, the degree of congestion can be calculated with high accuracy, making it possible to efficiently manage groups of elevators.

[Brief explanation of drawings]

1 to 19 are all drawings of the present invention,
Figure 1 is the overall configuration diagram of the group management control system, Figure 2 is the overall software diagram, Figure 3 is the table configuration diagram of the group management operation control system, and Figure 4 is the calculation of the predicted arrival time table. Fig. 5 is a flowchart for calculating the hall call elapsed time table, Fig. 6 is a flowchart for calculating fullness prediction, Fig. 7 is a flowchart for call allocation calculation, and Fig. 8 is for calculating long waiting minimization call allocation. Flowcharts, Figures 9 to 11 are table configuration diagrams of the simulation system, Figure 12 is a flowchart for calculating optimal area priority parameters by simulation, and Figure 13.
Figure 14 is a flowchart for collecting statistical data by simulation, Figure 15 is a flowchart for calculating stop probability by simulation, Figure 16 is a flowchart for calculating predicted fullness value by simulation, and Figure 17 is a flowchart for calculating statistical data by simulation. Fig. 118 is a flowchart for calculating the service priority level by simulation, and Fig. 118 is a flowchart for calculating the door time system φ parameter by simulation. MA... Elevator group management control device, HD... Hall call device, Ml... Elevator group management operation control microcomputer, M2... Simulation microcomputer, SD
A...Processor dedicated to serial communication between microcontrollers, E1~
Ea... Microcomputer for machine control, PM... Target setter output signal. No. 7 l No. 8 Concave No. 9 Prisoner 10 Figure $ll Figure No. 12 Concave 73 Zuhou /4 Country No. 7.5 No. 17 Kuni No. 7g

Claims (1)

[Scope of Claims] 1. A plurality of elevators operating between multiple floors, a hall call means provided at each floor landing, and a hall call means provided in each elevator car for indicating the destination floor; means for collecting and accumulating in-building traffic; means for selecting an elevator to service the hall call according to an evaluation function with variable parameters; and a group for assigning the hall call to a selected one of the plurality of elevators. In an elevator group management apparatus comprising a management control means, a means for selecting the elevator to be serviced, and a means for simulating the group management control means, the simulation means performs a process for each floor or direction for each variable parameter. An elevator group management device characterized by calculating a service priority level. 2. The elevator group management device according to claim 1, wherein the service priority level is a parameter indicating the degree of congestion. 3. The elevator according to claim 1, wherein selection of the elevator to be serviced and group management control are performed based on a service priority level that is optimal for the variable parameters evaluated and determined by the simulation means. Group management device.