JPS623075B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to an elevator group management device, and is particularly suitable for an elevator group management and control device using a computer.

[Prior art]

Conventional group management devices do not perform detailed group management control that adapts to the specific traffic volume of a building, making it difficult to predict when elevators will be full.

[Purpose of the invention]

An object of the present invention is to provide an elevator group management device that enables advanced predictive control, such as improving the accuracy of predicting elevator occupancy, by understanding building-specific traffic volume.

[Summary of the invention]

The features of the present invention are to classify traffic demand based on the characteristics of the building, aggregate the number of people getting on and off at each floor for each classified traffic demand, and calculate and learn the average number of people getting on and off.
It is possible to predict the number of people in the car based on the calculated value and the number of people in the car. More specifically, for example, in group management control that allocates hall calls to elevators, when predicting elevator occupancy, the average number of passengers from hall calls that have been allocated by traffic demand, direction, and floor, and the average number of passengers from car calls. Calculate the number of people getting off, the average number of passengers getting on and the average number of people getting off based on the matched calls of allocated hall calls and car calls,
By predicting how full the car will be by taking into account the number of people in the car, it is possible to prevent the car from being filled to capacity, which can irritate waiting customers.

[Embodiments of the invention]

Hereinafter, the present invention will be explained in detail with reference to a specific embodiment shown in FIGS. 1 to 13. In addition, the explanation of the embodiment will first describe the hardware configuration that realizes the present invention, then describe the overall software configuration and its control concept, and finally explain the software that realizes the above control concept using a table configuration diagram and a flowchart. Explain using. FIG. 1 shows the overall hardware configuration of an embodiment of the present invention.

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

The elevator group management device MA has an elevator operation control function, an in-building traffic data collection function, a function to calculate signals that classify traffic demand from characteristics of traffic volume or time signals, etc., and a function to calculate the number of passengers and alighters on each floor. The average number of people boarding an elevator based on an assigned hall call, the average number of people getting off an elevator based on a car call, and the average number of people boarding a call where the assigned hall call and car call match, determined by traffic demand, floor, and direction. There is a microcomputer M that controls the function of calculating and learning the number of people getting off the elevator and the average number of people getting off the elevator, and the function of calculating a predicted value of fullness of the elevator from the number of people in the elevator car.

The microcomputer M has a circuit PIA that inputs and outputs the call signal HC from the hall call device HD in parallel.
(Peripheral Interface Adapter), and also controls individual elevators such as door opening/closing and car acceleration/deceleration commands.
E ₁ ~ _{E o} (Here, it is assumed that there is an elevator number n) means the serial communication processor SDA ₁ ~
It is connected to SDA _o via communication lines CM ₁ to _{CM o} . On the other hand, the signal P _M from the setting device PD, which gives commands necessary for determining the variable parameters described above, is inputted via a circuit PIA that inputs and outputs signals in parallel.

In addition, the machine control microcomputers E ₁ to _{E o} are equipped with car call information necessary for control, car load change information at each floor, and control input/output elements consisting of various elevator safety limit switches, relays, and response lamps. Circuit that inputs and outputs signals in parallel with EIO ₁ to EIO _o
Connected to PIA via signal lines SIO ₁ to SIO _o .

The present invention will be generally explained using FIG. The microcomputer M has a built-in operation control program mainly for call assignment, and this operation control program takes in information necessary for control from the microcomputers E ₁ to E _o for controlling each machine and the hall call HC. Further, the operation control program performs call assignment using the variable parameters described above. For example, this parameter includes the stop probability required to accurately determine the predicted elevator arrival time, which is one element of the evaluation function for call assignment, or the predicted fullness value to accurately predict the fullness.

These calculations are processed in real time according to traffic demand such as when commuting to work, during the first half of lunch, during mid-lunch, during late lunch, during normal times, during normal busy times, when leaving work, during off-peak times, etc.
Outputs the optimal parameters for elevator group management at any given time. Therefore, the elevator group management control according to the present invention changes from time to time. It can adapt to the environmental conditions of the building and greatly contributes to improving elevator group management performance.

Next, the specific hardware configuration of each microcomputer will be shown, but these microcomputers can be easily configured as shown in FIGS. 2 and 3. MPUs (Micro Processing Units), which are the core of microcomputers, are 8-bit, 16-bit, etc., and in particular, 8-bit MPUs are appropriate because they do not require much processing power for the microcomputers _E1 to _Eo for machine control. be. On the other hand, since the microcomputer M for elevator operation control requires complex calculations, it is necessary to use a 16-bit microcomputer with excellent calculation performance.
MPU is appropriate.

Now, as shown in Figures 2 and 3, each microcomputer has a ROM (Read, Only Memory) that stores control programs, etc. on the bus line BUS of the MPU, and a RAM that stores control data and work data.
(Random Access Memory), a circuit PIA that inputs and outputs signals in parallel, and a dedicated processor SDA (Serial Data Memory) that performs serial communication with other microcontrollers.
Adapter (for example, Hitachi HD43370) is connected.

In addition, each microcomputer M, E ₁ to _{E o} smell, RAM,
The ROM is composed of a plurality of elements depending on the size of its control program.

FIG. 2 shows the configuration of the elevator group management device.

In FIG. 3, as elevator control data, for example, the car call button CB, the safety limit switch SW _L , the relay contact SW _Ry , and the car weight Weight are imported into the RAM from the PIA. On the other hand, the data calculated by the MPU is output from the PIA to control output elements such as the response lamp Lamp and the relay Ry.

Here, the hardware configuration of the processor SDA for serial communication between the microcontrollers used in Figures 2 and 3 mainly consists of a transmission buffer as shown in Figure 4.
TX _B , reception buffer RX _B , data parallel/
It is composed of a P/S that performs serial conversion, an S/P that performs inverse conversion, and a controller CNT that controls their timing. The above transmitting buffer TX _B and receiving buffer RX _B can be freely accessed by the microcomputer, and data can be written and read. On the other hand, SDA is the controller CNT
Therefore, it has a function of automatically transmitting the contents of the transmission buffer TX _B to the reception buffer RX _B of another SDA via the P/S. Therefore, the microcomputer does not need to perform any transmission/reception processing, so it can concentrate on other processing. 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.
No. 56-37973.

Next, a software configuration according to an embodiment of the present invention will be described.First, the overall software configuration will be explained with reference to FIG.

As shown in Figure 5, the software consists of software that calculates the predicted fullness value and an operation control program SF13 that performs group management operation control based on the predicted fullness value, both of which are processed by the microcomputer M. be done. However, since the occupancy prediction value calculation software and the operation control software can be easily separated, they can be easily configured using separate microcomputers.

The operation control program SF13 is software that directly commands and controls elevator group management control such as call assignment processing, full-occupancy prediction detection processing, and elevator distributed standby processing. As input information for this program, the elevator control table SF1 including elevator position, direction, car call, etc., which is transmitted from the machine control program (Fig. 1: built in the microcontrollers _E1 to _Eo ), the hall call table SF2, and the prediction of full capacity. The input data is calculated using value calculation software and outputs the fullness prediction value learning parameters.

On the other hand, the full occupancy prediction value calculation software is composed of the following processing program.

(1) In-building traffic collection program SF3: This is a program that samples the contents of the hall call and elevator control tables online at predetermined intervals, and mainly collects the in-building traffic.

(2) Traffic demand classification program SF5...... Inputs the traffic volume table SF4 obtained by the in-building traffic collection program SF3 and time information TI, and calculates the traffic volume in the building for work, before lunch, during lunch, and after lunch. , is a program that divides traffic demand into eight categories: normal, normal congestion, leaving work, and quiet.

(3) Average number of people boarding/disembarking program
SF7……Intra-building traffic collection program SF3
The traffic volume table SF4 obtained by , and the traffic demand classification signal table SF6 have so-called matching call information in which the elevator directions, positions, and assigned hall calls, car calls, and hall calls and car calls for each elevator match. Using the elevator control data table SF1 as input, collect data while the traffic demand signal does not change, and when the traffic demand signal changes, calculate the average number of people boarding and exiting by floor, direction, and call type. This is a program that outputs the table SF8.

(4) Learning program SF9 for the average number of people boarding and average number of people getting off...The same traffic demand signal, same floor, same direction, and same call calculated in the past by inputting the average number of boarding and average number of people getting off table SF8 This is a program that learns the types of and exponential smoothing.

(5) Crowd prediction value learning program SF11……Average number of people boarding, average number of people disembarking learning table SF1
This is a program that calculates a fullness prediction value by inputting 0.

Next, a table configuration used in an embodiment of the present invention will be explained with reference to FIGS. 6 to 8. Figure 6 shows the table structure of the operation control program SF13.
Broadly speaking, it consists of an elevator control table SF1 and a hall call table SF2. The tables within each block will be described each time the operation control program is explained below.

Figures 7 to 8 show the table structure of the fullness prediction value calculation software, including traffic volume table SF4, traffic demand classification table SF6, average number of people boarding, average number of people getting off table SF8, average number of people getting on, and average number of people getting off. It is composed of blocks of a learning table SF10 and a fullness prediction value learning table SF12.

Next, an embodiment of the software of the present invention will be described.

First, the operation control program will be explained, and then the stop probability calculation software will be explained. Note that the programs described below are managed under a system program that divides the program into a plurality of tasks and performs efficient control, that is, an operating system (OS). Therefore, programs can be started freely from the system timer or from other programs.

Now, FIGS. 9 to 12 show the flow of the operation control program. Three of the most important operation control programs, the predicted elevator arrival time table calculation program, the call assignment program, and the fullness prediction detection program, will be explained.

FIG. 9 is a flowchart of a program for calculating the predicted arrival time of an elevator to an arbitrary floor, which is to be the basic data for calculating the waiting time evaluation value. For example, this program is started periodically every second,
The predicted arrival time from the current position of the elevator to any floor is calculated for all floors, using stop probability parameters determined for each floor and each direction for all elevators.

In FIG. 9, steps E10 and E100
indicates that loop processing is performed for all elevator numbers. In step E20, an initial value is first set in the work time table T, and its contents are set in the predicted arrival time table shown in FIG. As an initial value, the number of seconds remaining before departure can be considered based on the open/closed state of the door, or the predetermined time until activation when the elevator is stopped, etc.

Next, the program advances one floor (step E30) and compares whether the floor is the same as the elevator position (step E40). If they are the same, 1
This means that the predicted arrival time table for this elevator has been calculated, so the process jumps to step E100 and repeats the same process for the other elevators. On the other hand, if "NO" in step E40, the first floor running time Tr is added to the time table T (step E50). This time table T is then set as a predicted arrival time table (step E60). Next, it is determined whether there is a car call or an assigned hall call, that is, a call that should be serviced by the elevator of interest, and if so, the elevator stops and the first floor stop time is
Add Ts to the time table (step E8)
0). Next, the process jumps to step E30 and the above process is repeated for all floors. However, if not, the elevator will not stop at this time, but may stop due to a car call etc. after the assigned hall call service.
P ^E _j Ts is added to the time table T using the stop probability P ^E _j determined for each traffic demand, floor, and direction (E90).

FIG. 10 shows the flow of the call allocation program, which is activated when a hall call occurs. In this program, the call allocation algorithm is the long-waiting call minimization call allocation algorithm (described later in FIG. 11) as shown in step A50.
It is.

When a hall call occurs, the first step is step A1.
When set to 0, the generated hall call is read from outside. Then, the following processing is performed in a loop at steps A20 and A80, and steps A30 and A70. That is, if there is a generated hall call, this call is assigned to the elevator that is most suitable in terms of minimizing long-waiting calls (step A60).

FIG. 11 is a processing flow of the long-waiting call minimum call allocation algorithm. First, step A5
0-1, the fullness prediction is detected and elevators excluded from service are determined (described later). Steps A50-2 and A5 are used to determine which elevator is optimal.
0-7 performs loop processing based on the number of elevators.
The processing within the loop begins with step A50-3.
Calculate the maximum predicted waiting time Tmax for the hall calls assigned on the front floor, including the generated hall calls. Note that the predicted waiting time is the sum of the hall call elapsed time (see FIG. 6) indicating the elapsed time from the occurrence of the hall call to the present time. Next step A50
In step A50-4, a stop call evaluation value T _C is calculated from the stop calls on predetermined floors before and after the hall call that has occurred, and an evaluation function φ is calculated using this evaluation value and the aforementioned maximum predicted waiting time Tmax (step A50-4). 5). Then, the smallest elevator in this evaluation function φ is selected (A50-6). When the above process is executed for all elevators, step A50-
By the calculation in step 6, the elevator with the optimal evaluation value is selected.

FIG. 12 is a flowchart of a program for determining service elevator candidates based on predicted fullness values.

First, in step B10, in order to determine whether or not the service is available for all elevators, first
Set the elevator for the machine. In step B20, _L
Set the number of people in the car for the car. In step B30, the hall call floor that has occurred is assigned to the floor variable j.
In step B40, it is determined whether there is a hall call on the j floor, and if there is, in step B50,
The direction of the hall call for the j floor is determined, and if the call is in the UP direction, UP is assigned to the direction variable E, and if the call is in the DOWN direction, DOWN is assigned to the direction variable E (step B70). Next, in step B80, it is determined whether there is a car call on the j floor, and if there is, there is a hall call and a car call in the E direction on the j floor, so the fullness prediction value P of the j-floor E-direction matching call is calculated.
Add ^E _j,K to F _L (step B90). If not, in step B100, there is only a hall call in the E direction on the j floor, so the predicted fullness value P
Add ^E _j,H to the predicted full occupancy detection value F _L .

On the other hand, if there is no hall call on the j floor in step B40, in step B110,
It is determined whether there is a car call on the j floor, and if so, the direction of the elevator is determined in step B120. If the elevator is in the UP direction, UP is assigned to the direction variable E in step B140. However, if the elevator direction is Down, the direction variable E is set in step B130.
Assign Down. As a result, _the predicted full occupancy value ^P
Add to _L. In step B160, 1 is added to the jth floor. In step B170, it is determined whether floor j is the top floor or the bottom floor. In step B180, it is determined whether the predicted fullness detection value F _L is within 90% of the rated number of people in the car. If 90%
If the value exceeds 1, the elevator L is judged to be out of service and excluded (step B190). Otherwise, in step B200, it is determined whether fullness prediction detection has been performed for all elevators, and if it has not been completed, in step B210,
For the next elevator, fullness prediction detection is performed in the same manner as described above. If the process has been completed for all elevators, this program ends.

Above, we have explained the processing flow of the operation control program of the predicted arrival time table calculation program, the call assignment program, and the fullness prediction detection program. There are service processing programs, distributed standby processing programs that cause elevators to wait at predetermined floors during off-peak hours, etc., but their explanations will be omitted.

Next, the program of the fullness prediction value calculation software is shown in FIGS. 13 to 15.

FIG. 13 is a program SF7 for calculating the average number of people getting on board and the average number of people getting off the train. This program is activated when traffic demand classification signals change, and the average number of people boarding per hall call, the average number of people exiting per car call, and the coincidence of hall calls and car calls for each floor and direction in the traffic demand. This is a program that calculates the average number of people boarding and the average number of people getting off per matched call. First, in step C10, the direction E is UP.
In step C20, the floor i is set to 1. Next, in step C30, the traffic table
The average number of people boarding per hall call ^h E _i is calculated from the number of people a ^E _i boarding the elevator due to hall calls in the i-floor direction E of SF4 and the number A ^E _i of hall calls. In the next step C40, in the i-floor direction E
From the number of people getting off the elevator due to car calls b ^E _i and the number of car calls B ^E _i , the average number of people getting off per car call C ^E _i is calculated, and in step C50,
From the number of people boarding the elevator due to so-called matching calls g E _i and the number of people getting off the elevator d ^E i and the number of matching calls D ^E _i where hall calls and car calls in direction E exist on ^the _i floor, one matching call is obtained. The average number of people getting on board i ^E _i and the average number of people getting off per trip j ^E _i are calculated. In step C60, 1 is added to floor i. In step C70, it is determined whether floor i has reached the top floor,
If it is not the top floor, repeat the above process. If it is the top floor, in step C80, the direction E is Down.
If it is in the Down direction, step C
At 90, Down is assigned to the direction E and the above process is repeated. If the direction is Down, the average number of people getting on and getting off has been calculated for all floors and directions, and the process ends.

Next, FIG. 14 shows the flow of a learning program for the average number of people boarding and exiting, in which the direction E is set to UP in step D10, and the floor i is set to 1 in step D20. Next, in step D30, h ^E _i , c ^E _i ,
i ^E _i , j ^E _i and the average number of people boarding and average number of people disembarking. From H ^E _i , C ^E _i , I ^E _i , J ^E _i of the learning table, create a new H ^E
_i , C ^E _i ,
I ^E _i and J ^E _i are calculated and learned. In step D40, 1 is added to i. In step D50, floor i
Determine if is the top floor, and if not, repeat the above process. If it is on the top floor, direction E is Down
If it is not Down, then in step D70, Down is assigned to direction E, and the above process is repeated.
If direction E is Down, end.

FIG. 15 shows a fullness prediction value learning program, which is started after the learning program for the average number of people boarding and the average number of people disembarking is completed. First, in step F10, the direction E is set to UP, and in step F20, the direction E is set to UP.
Assign 1 to . Next, in step F30, H ^E _i ,
From C ^E _i , I ^E _i , J ^E _{i ,} the predicted full occupancy value P ^E _{i, H} , P
^E _i,C ,P
Calculate ^E _i,K . In step F40, 1 is added to floor i, and in step F50, it is determined that floor i is the top floor. If it is not the top floor, repeat the above process. If it is the top floor, it is determined in step F60 that the direction E is Down; if not, Down is assigned to the direction E in step F70, and the above process is repeated. If it is Down, terminate.

In the above manner, predicted crowding values were determined by traffic demand, by floor, by direction, and by hall call, car call, and matching call.

〔Effect of the invention〕

According to the present invention, the number of people getting on and off by traffic demand is aggregated based on the characteristics of the building, the average number of people getting on and off is calculated, and the fullness is predicted from the calculated value and the number of people in the car, so it can be adapted to the building. This enables advanced predictive control, and has the effect of preventing elevators from being filled with people, for example, during energy-saving operation or during times of congestion.

[Brief explanation of the drawing]

1 to 4 are hardware configuration diagrams constituting an example of a group management device according to the present invention, FIG. 5 is an overall software configuration diagram of the present invention, FIGS. 6 to 8 are table configuration diagrams, and FIG. Figures 12 to 12 are flowcharts for explaining the operation control program software.
Figures 1 to 15 show the fullness prediction calculation software. MA...Elevator group management device, M...Microcomputer, HD...Hall calling device, _E1 ~ _Eo ...Machine control microcomputer, _EIO1 ~ _EIOo ...Control input/output element.

Claims (1)

[Scope of Claims] 1. A plurality of elevators operating between multiple floors, a hall call device provided at each landing for calling the elevators, and a hall call device provided in each elevator car for indicating the destination floor. A car calling device, means for allocating hall calls to elevators according to an evaluation function having variable parameters, and fullness prediction means for predicting that each elevator car will be full at each landing, In the apparatus for group management control of the plurality of elevators by varying parameters, means for collecting the traffic volume in the building, means for calculating a signal for classifying traffic demand from the characteristics of the traffic volume or the time signal, and the collecting method. a means for aggregating the number of passengers and alighting passengers on each floor according to the classified traffic demand based on the traffic volume; means for calculating and learning a new average number of passengers and new average number of people getting off the train, and predicting the full capacity from the calculated and learned values and the number of people in the elevator car. . 2. An elevator group management device according to claim 1, characterized in that the aggregating means and the arithmetic learning means are configured to perform aggregating, arithmetic and learning for each ascending and descending direction. 3. In claim 1, the calculation learning means determines whether the average number of people boarding an elevator based on an assigned hall call, the average number of people getting off an elevator based on a car call, and the assigned hall call and the car call match. An elevator group management device characterized by comprising means for calculating the average number of people getting on and the average number of people getting off.