JPH06100249A - Elevator group management control system - Google Patents

Abstract

(57) [Summary] [Purpose] Detect the dumpling operation status of the group management control elevator, manage the detection result so that it can be used easily by the person in charge of design and maintenance, and further improve the dumpling operation status. Control. [Configuration] An operation state detection unit that calculates an index value indicating the degree of deviation of the elevator from the data of the current position and traveling direction of each elevator, and the index value is determined by the operation state data in the ascending direction and the descending direction. A means for correcting and calculating, a means for determining a tentative traveling direction for improving the index value and a means for determining a standby position for the standby state machine having no call, respectively, in addition to the dumpling A means for outputting the transition of the degree of operation over time in combination with an operation diagram showing the operation status of each unit, and a means for determining the time range in which the operation efficiency is decreasing using the index value. Each unit is provided with means for selecting and outputting the operating state of the elevator in the time range.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an elevator group management control system, and more particularly to an elevator group management control system suitable for detecting and improving a state in which operation efficiency is lowered.

[0002]

2. Description of the Related Art In elevator group management control,
The biggest challenge is to make the waiting time of users as short and uniform as possible. For that purpose, it is desirable that the time intervals of the elevator cars are evenly distributed. In the conventional group management control system, a "assignment evaluation formula" method is mainly used to determine the number of the elevator that will serve the user waiting in the hall based on the estimated arrival time of each elevator to each floor. This method is required to improve the accuracy of arrival time prediction, and is disclosed in Japanese Patent Laid-Open No. 58-636.
As described in Japanese Patent No. 68, the learning function was introduced. In addition to the estimated arrival time, several methods have been devised that take the current elevator operation status into consideration. In particular, even if the control by this basic allocation evaluation formula is performed, the positions of several units may sometimes be biased, so-called "dango operation". The technique described in Japanese Patent Laid-Open No. 62-12186 is proposed.

[0003]

In the above prior art,
Although a predicted value such as “estimated arrival time” is used as a basic evaluation value for control, the group management control is intended for humans, so the prediction often goes wrong. For example, if the user continues to press the door-open button for several tens of seconds, the arrival prediction will be significantly wrong. Even when the learning function described above is used, the effect of such a sudden user operation cannot be taken into consideration, and the position of the succeeding vehicle becomes too close, which may cause dango driving or the like. This misalignment of prediction is an unavoidable property to some extent in a system for predicting a human subject by a probabilistic method. However, it is necessary to surely detect the inefficient operation state caused by the deviation, especially the remarkable dumpling operation state, and promptly solve it. Further, once in the dumpling operation state, the arrival time of each unit and other evaluation values become similar, and it becomes difficult to recover from the dumpling operation to a good state only by the call allocation control by the allocation evaluation formula. for that reason,
It is important to detect a dango operating state or a tendency to reach a dango operating state, and to use this as a trigger for performing a special control method. The above-mentioned JP-A-62
In Japanese Patent Laid-Open No. 112186, the degree of driving dumpling is calculated based on the one-round time using the estimated arrival time to the reference floor, but the above-mentioned prediction error is considered. Absent. In particular, this misalignment of prediction has a greater effect as the prediction time becomes longer, and the accuracy of the calculated value based on the predicted value also becomes a problem in the calculation method that requires one round time of the entire machine. The object of the present invention is to detect the dumpling operating state in which the operating state of the group management control elevator is in a deteriorated state, and to manage the detection result in an easy-to-use manner, and further to improve the detected apple operating state. An object of the present invention is to provide an elevator group management control system for performing control.

[0004]

In order to achieve the above object, there is provided an operation state detecting means for calculating an index value indicating the degree of deviation of the elevator from the data of the current position and traveling direction of each elevator car, and the ascending movement. Equipped with means for calculating and correcting the index value that indicates the degree of deviation based on the operating state data of the direction and the descending direction, and further improving the index value that indicates the degree of deviation for a standby state machine that does not have a call And a means for determining a standby position. In addition, a means for outputting the transition of the index value indicating the degree of deviation over time in combination with the operation diagram showing the operation state of each elevator, and the operation efficiency is reduced by using the index value indicating the degree of deviation. And a means for selecting and outputting the operating state of the elevator within the determined time range.

[0005]

Each function of the elevator group management control system operates as follows. The operation state detecting means for calculating the index value indicating the degree of deviation of the elevator calculates the sum of squares of the position intervals from the data of the current position and the traveling direction of each elevator, and the "dango driving degree" which is the index value of the degree of deviation. Is calculated. Here, the data on the current position and traveling direction of each elevator is the “result of the group management control performed up to the present time”. Therefore, it is not necessary to make an inaccurate prediction. It can be said that the data is suitable for use in calculating the index value for representing the state of. From the above data, the spatial bias can be directly obtained. Next, in order to obtain the target temporal deviation, the dango driving degree correction means calculates the temporal deviation from the spatial deviation based on the ratio of the traveling times in the ascending direction and the descending direction. Determine the correction amount of. The output means outputs the transition of the dango driving degree over time in combination with the operation diagram of each elevator. Further, the time range in which the operation efficiency is decreasing is determined using the dango operation degree, and the operation state of the elevator in the time range is selected and output. The provisional traveling direction determination means calculates and compares the dango driving degree in each of the ascending / descending cases with respect to the standby state machine having no call, and determines a temporary traveling direction for improving the dango driving degree. The assigning means adds an evaluation value to the assignment evaluation formula at the time of call assignment, which promotes the assignment in the direction that coincides with the temporary traveling direction of the standby state machine and adds the evaluation value that suppresses the assignment in the direction different from the temporary traveling direction, and the dango operation. Suppress that. The standby position determining means calculates and compares the dango driving degree on all floors in each traveling direction with respect to the waiting state vehicle, and determines the standby position that most improves the dango driving degree.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. First, an outline of the system configuration and the operation of each unit will be described. FIG. 1 is an overall configuration diagram of an embodiment of the present invention. The system includes an operation state detection device 1, a group management control device 2, an elevator number 3, and a hall call button 4. In the operation state detection device 1, the determination unit 1-1,
There is a recording unit 1-2 and an output unit 1-3, and the group management control device 2
There are an allocation unit 2-1 and a learning unit 2-2 inside. Also,
The respective parts are connected to each other by a data transmission path. The allocation unit 2-1 in the group management control device 2 selects an elevator car 3 to be serviced to a newly arrived user and issues an allocation command. The learning unit 2-2 learns the data indicating the current operation state from each elevator number 3 and the hall call button 4, determines the corresponding traffic flow characteristic mode, and instructs the allocation unit 2-1. . Further, necessary data is selected / added from the data indicating the current operation state, and the determination unit 1-1 and the recording unit 1- of the operation state detection device 1 are selected.
Send to 2. Judgment unit 1-1 in the operation state detection device 1
Determines the dango driving degree using the operation state data and transmits it to the allocating unit 2-1 that executes the group management control. The recording unit 1-2 records the operation state data and the dango driving degree determined by the determination unit 1-1 together, and outputs necessary parts to the output unit 1-3 according to an instruction from the determination unit 1-1. To do. The output unit 1-3 controls and outputs a device such as a display or a printer, or outputs data which a maintenance person or the like analyzes with another device through a magnetic disk or a transmission line. In addition, in this embodiment, the operation state detection device 1 and the group management control device 2 are described as separate devices by dividing them into functions, but the function of the operation state detection device 1 is incorporated into the group management control device 2 to provide a single device. Even if it is configured as a device, the present invention can be carried out without trouble. Also, if necessary,
The number of elevators is N, and the number of floors of the building is L. Next, the basic idea of dango driving degree calculation will be described using an example. FIG. 2 shows an example of an elevator arrangement and its simplified ring representation as preparation for calculating the dango driving degree.

In FIG. 2, the number of elevators is four from A to D (N = 4), and the number of floors is 10th floor (L = 10th floor).
= 10) will be described as an example. In addition, FIG.
The white triangles indicate the call, the double circles indicate the destination floor, and the black triangles above and below each unit indicate the direction of travel. Figure 2
From the state of (a), the position of each unit is shown in the order of the lowest floor (1st floor), the ascending direction, the uppermost floor (10th floor), and the descending direction, ignoring the incoming call and the destination floor. It can be represented by the ring representation of b). Here, the cars are ordered in order from the lowest floor in the traveling direction. Further, for example, since it is possible to go up and down from the 3rd floor, a U signifying an ascending and a D signifying a descending are respectively attached to the 3rd floor so as to be distinguished as 3U and 3D. The floor position when this direction is taken into consideration will be referred to as “direction floor” below, and is represented by the symbol P. Since the elevator does not go from the bottom floor to the bottom and the top floor to the top, using the number of floors L, the actual service location is

[Equation 1] In the following, this will be referred to as the service direction floor S. From this, as the value of P indicating the direction floor, the lowest floor is set to 1, and the value is obtained by adding 1 to each floor in the ascending direction, the top floor, and the descending direction. At this time, the value of the direction floor P of the ascending direction j floor is

[Equation 2] The value of the direction floor P of the descending direction j floor is

[Equation 3] Becomes For example, in the descending direction on the 7th floor, P = 2 × 10
-7 = 13. When the position of the _i-th vehicle is represented by P _i , the position of the (i + 1) -th vehicle is represented by P _{i + 1} , and the position interval measured by the direction rank thereof is represented by d _i , the value defines the function diff of (Equation 4). Can be obtained by

[Equation 4] Also, at this time, the intervals between the respective units have the relationships of (Equation 5) and (Equation 6).

[Equation 5]

[Equation 6] The basic procedure for calculating the dango driving degree described below is as follows: 1) Obtain the time to arrive at each direction floor based on the position (direction floor) of each unit. 2) Obtain the expected value (average value) of the arrival time on all floors. 3) Zero the arrival time in an ideal state in which all units are evenly distributed.
Set the arrival time of 100 points for a complete dango state where all units are gathered in one place. 4) Calculate the dango degree from the expected value of the arrival time obtained earlier. Is. First, assuming that the importance of each floor, each direction, and the one-round time R are constant, the traveling time r per floor can also be regarded as constant.

[Equation 7] Considering the time a _{ij at} which Unit i arrives at the j-direction floor using the traveling time r per floor,

[Equation 8] Therefore, when this arrival time a _ij is expressed in the example of FIG.
It becomes FIG. That is, since Unit 1 is located on the 2nd floor, it is 0 hours, 3 hours is r hours, and 4U is 2r hours.
Time: Indicates what it takes to arrive. Given that the arrival time of the first car arriving at the j-direction floor out of N cars is the waiting time w _j on the j-direction floor, the expected value (average value) m of the waiting times on all floors is

[Equation 9]

[Equation 10] The waiting time w _j on the j-direction floor is also clear from FIG.
This is the arrival time of the car that is closest to the j-direction floor. Considering this from the No. side, the No. _i arrives at the d _i floor from the current P _i direction floor to the (P _{i + 1} -1) direction floor before the first floor where the next no. It will be. here,

[Equation 11]

[Equation 12] Therefore, the expected value m of the waiting time of the omnidirectional floor is that the omnidirectional floor S is divided by N

[Equation 13] Becomes From the result of (Equation 13), the expected value m of the waiting time on all floors can be calculated without obtaining the waiting time w _j on each floor.
It can be calculated directly from the interval d _{i of} each car. Since the expected value m of waiting time on all floors is a function of d _i ,
The minimum value is set when all the machines are evenly distributed, and the maximum value is set when all the machines are completely dumped. The minimum value is the position interval d _i = S / N for all the machine i, and the value is

[Equation 14] The maximum value is when d ₁ = ... = d _N-1 = 0 and d _N = S, and the value is

[Equation 15] Becomes However, although the position interval d _i is an integer, since the theoretical minimum value d _i = S / N does not always become an integer, the actual minimum value becomes a value slightly larger than (Equation 14). From (Equation 14) and (Equation 15), the dango driving degree detection function Gd based on the position interval d _i is

[Equation 16]

[Equation 17] Set to meet. From (Equation 13), the expected value m of the waiting time on all floors is a function of the square of the position interval d _i.

[Equation 18] Assuming that, by substituting (Equation 16) and (Equation 17), the coefficient a,
If we ask for b,

[Formula 19] Becomes Here, when the dango driving degree of the arrangement illustrated in FIG. 2 is calculated,

[Equation 20] Can be determined. That is, when Gd = 25.4, it can be determined that the driving state is substantially good, since it is assumed that the ideal state is 0 point and the perfect dumpling state is 100 points.
Furthermore, by substituting (Equation 13) into (Equation 19),

[Equation 21] Therefore, the one-round time R can be erased in about a minute. From the result of (Equation 21), the dango driving degree Gd due to the position interval d _i can be coped with by changing the number of floors in the service direction S (number of floors L) and the number of elevators N even if 1) the building is different. 2) The operation data used is only the position interval d _i between the units. 3) It can be calculated by simple arithmetic operations such as addition, subtraction, multiplication and division and the square root, and without depending on the one-round time R. It has the feature. Further, since the square root is a monotonically increasing function, if it is used for comparison or determination during control, the value within the root sign may be used, and only simpler addition / subtraction / multiplication / division operations can be performed. The method of calculating the dango driving degree Gd is basically as described above, but the assumption of "the importance of each floor, each direction, and the constant one-round time R" does not always hold, and the actual control Occasionally, there are cases where a call is in-hand, and there is a waiting state machine that does not have a direction of travel, so it may not be possible to use it as it is.

Therefore, an example of a flow chart in the case of actually using the above-mentioned breakfast operation degree Gd will be described with reference to FIGS. 4 to 8. FIG. 4 is a flowchart of the main module showing the outline of the processing of the dango operation detection program 10 executed by the determination unit 1-1 of the operation state detection device. Dango driving detection targets a series of traffic flow feature modes (time zones) as a unit. That is, if the traffic flow mode is the same, the same correction value and the like are used, and the total dango degree as a time zone is obtained using the dango driving degree Gd calculated at each moment. In this embodiment, the program 10 is started each time the traffic flow characteristic mode determined in learning 2-2 is changed, and is continuously calculated until the characteristic mode is changed next time. First, the initialization processing module 20 initializes variables. next,
Dunk driving degree calculation module 30 dango driving degree Gd
Is calculated and based on the dumpling operation degree Gd determined by the dumpling operation time addition module 40, the time during which the dumpling is in operation is added. Further, the dumpling operation degree Gd calculated by the dumpling operation degree calculation module 30 is recorded by the recording unit 1-2 and used by the allocation unit 2-1. Step 10-1
If there is no change in the traffic flow mode in the above judgment, the process returns to the dango driving degree calculation module 30 at a prescribed interval (every 1 second, every time a call occurs, etc.), and the process is repeated. In this embodiment,
This interval is 5 seconds. When the traffic flow mode is changed in the determination in step 10-1, in step 10-2, "dango degree Gt in time zone", which is the ratio of the total gtime of the dumpling driving time to the duration ttime of the corresponding traffic flow mode. Is calculated by (Equation 22), and the dango driving detection process in the traffic flow mode is ended.

[Equation 22] Actually, at the same time as the above-mentioned processing ends, the processing for the next changed traffic flow mode continues to be activated.

FIG. 5 is a flowchart of the initial setting processing module 20. In step 20-1, learning unit 2-
Read the condition calculated from 2. The conditions are the number of elevators N, the number of service floors L, and the upward / downward correction amounts ku and kd. Here, the correction amounts ku and kd are correction values used at the time of actual group management control execution in which the assumption of the floor, the direction, and the constant one-round time R does not hold. The content is an integer ratio of the traveling time in the ascending direction and the descending direction in the corresponding traffic flow mode. The value is, for example, the traveling time in the ascending direction is 25
Minutes, when going down to work for 5 minutes, Ku = 5, kd
= 1. The value learned by the learning unit 2-2 is used. The correction amounts ku and kd are, in the sense, equivalent to apparently multiplied by ku and kd. In step 20-2, variables are initialized.
Here, due to the above correction, the service direction floor S of (Equation 1) is regarded as being ku times in the ascending direction and kd times in the descending direction,

[Equation 23] To change. a and b are the coefficient and the constant term of (Equation 21). Also, flag represents whether or not the dango is in operation, ttime is a duration for the corresponding traffic flow mode, and gtime is a variable for calculating the total dango operation time.

FIG. 6 shows a dango driving degree calculation module 30.
It is a flowchart of. First, in step 30-1, the contents of the work area indicating the number of units in each direction on each floor are cleared. In step 30-2, the position pos _i and the direction dir _i of the _i- th car are read. In step 30-3, a variable di_ssum indicating the sum of squares of position intervals is initialized. next,
A subroutine 31 for calculating the sum of squares is called. Here, the argument 1 of the subroutine calc_di () indicates that the process is started from the first machine. In the sum of squares calculation subroutine 31, the sum of squares of the position intervals is the variable di_s.
Store in sum. In step 30-4, (Equation 21)
Calculate the dango driving rate Gd at each time by
-2, to the group management unit 2-1.

FIG. 7 is a flow chart of the subroutine for calculating the sum of squares of positional intervals. Subroutine calc
The argument j of _di () indicates that the current machine to be processed is the j-th machine. The subroutine calc_di () is recursively called and used. In step 31-1, the processing is divided depending on whether the target number j is smaller than the total number N of units or not. In other words, the former is a flow that determines the traveling direction of each unit,
The latter is a flow for calculating the positional interval d _i between each car. When it is small, it is checked in step 31-2 whether the machine is a stand-by machine with no direction, and if it is directional, the number of machines in the work area indicated by the position pos _{j of the} machine _j and the direction dir _j is calculated in step 31-3. Add one. Then, in order to proceed to the processing of the (j + 1) th machine, the subroutine calc_di
Call () with the argument (j + 1). Step 31-2
If the number j is no direction, the temporary movement direction is assumed to be the ascending direction at step 31-4 and the position pos _{j of the} number _j and the number of the number of the work area indicated by the ascending direction are incremented by 1, and the subroutine calc_di Call () with the argument (j + 1). Then, at step 31-5, the position pos _j and the number of machines in the work area indicated by the ascending direction are subtracted by one. Next, assuming that the traveling direction is the descending direction, the same processing is performed in step 31-
6. Repeat with subroutine 31, step 31-7.
From this, for each car, once if the traveling direction of the corresponding car is fixed, twice if there is no direction,
The subroutine calc_di () will be called. Since the standby unit is in an advantageous state in which the group management control device can determine the direction according to the operating state, it is appropriate to set an advantageous temporary traveling direction for the standby unit that can reduce the dango operation degree. There is. Therefore, the traveling direction with the smaller sum of squares obtained in the process described later, that is,
The dunnage driving degree is calculated by setting the traveling direction of the one having the smaller dango driving degree as a temporary traveling direction. Returning to the explanation of step 31-1, when the target machine number j is larger than the total machine number N, information on the position and the traveling direction of all the machines or the tentative traveling direction is set in the work area by the above-mentioned processing. Therefore, the process shifts to the process of obtaining the sum of squares of the position intervals. First, in step 31-8, the variable d _i representing the position interval is initialized. At this time, in order to simplify the calculation of the position interval from the N-th machine to the first machine, the variables are initialized from the 0-th machine. At step 31-9, 0 is set to the variable i representing the machine number. Steps 31-10 and 31-12 are loop processing in the ascending direction, and the variable k indicates the floor position. In step 31-11 in the loop, ku corresponding to the correction amount per floor in the ascending direction is added to the position interval d _i , and the number of target units i is increased by the number of units in the ascending direction on the kth floor (there is no unit. At that time, it does not change because it increases by 0). Steps 31-13, 14, and 15 are similar processing in the descending direction. Next, step 3
At 1-16, d ₀ is added to d _N , which represents the position interval from the N-th vehicle to the first vehicle. This processing is necessary to start the loop concerning the floor from the lowest floor, not from the first machine position. Next, step 31
In -17, the calculated sum of squares of di is substituted for the variable dssum. If dssum is smaller than di_ssum obtained so far in step 31-18, that is, if it is a value that improves the dango operation degree, step 31-19.
Then, dssum is assigned to di_ssum. By recursively repeating the above processing, the sum of squares of the position intervals d
i_ssum can be calculated.

8 and 9 show examples of outputs obtained by the above-mentioned processing, in comparison with the actual operating state, regarding the degree of operation of the apple. 8 and 9 are examples of the output method of the output unit 1-3. FIG. 8 is an example of a change in the operation diagram and the dango driving degree Gd with the passage of time when operating in the dango driving state, and FIG. 9 shows that the dango driving state disappears in the same traffic flow. It is an example of the change of the operation diagram and the dango driving degree Gd during the time. condition is,
In both cases, the number of elevators N = 3, the number of floors L = 16, the time zone is normal, and the correction amounts are 1 for both Ku and Kd. In FIG. 8, a typical dumpling operation state in which three elevators move together around 11:05 and the dumpling operation degree Gd is as high as 70 points or more.
On the contrary, in FIG. 9, the dumpling operation is almost eliminated, and the dumpling operation degree Gd is also low. In addition, when examining the section in the dumpling operation state from this operation diagram based on empirical judgment, the part where the dumpling operation degree Gd is 70 points or more can be considered to be the dumpling operation state, and the dumpling operation degree Gd is Good operation is under 50 points.
The area between 50 and 70 points is an area where the judgment can be made differently depending on the circumstances. Therefore, in this embodiment, the dango operation section is set to "50 after the dango operation degree Gd exceeds 70 points.
It is possible to make it almost coincide with the judgment of human beings by defining "until the point is exceeded".

FIG. 1 is a flow chart of the dango driving time addition module 40 for obtaining the time zone dango degree in the relevant traffic flow mode by the definition of the dango driving section.
It shows in 0. In step 40-1, the variable flag indicating the dango operation is checked, and if the flag is 0, step 4
If the dango driving degree Gd is 0 or more at 0-2, the flag is set to 1 at step 40-3. If the flag is 1, and the dango operation degree Gd is 50 points or less in step 40-4, the flag is set to 0 in step 40-5. In step 40-6, the calculation period (here, 5 seconds) is added to the duration ttime of the relevant traffic flow mode. In step 40-7, the dlag operation state flag is set to 1
If so, the calculation cycle (here, 5 seconds) is also added to the total gtime of the dango operation time in step 40-8. Thereby, in step 10-2 of the main routine, the dango degree of the time zone can be calculated. As described above, FIG. 4 to FIG.
The processing described above makes it possible to detect the dumpling operation state.

According to this embodiment, the following effects can be obtained. In order to calculate the dango driving degree, only the information on the position and the traveling direction is required as the data at each moment, and the data such as the number of calls and the number of people are not required, so that the amount of data to be recorded may be small.
Since the calculation is simple, the dumpling operating state can be detected and monitored without increasing the load of the group management control process. It is possible to selectively record and output only the data of the time when there is a problem in the operation state based on the dango of the time zone.
Further, by selecting the output range, the analysis target time for improvement can be easily known, which saves labor. By combining and outputting the operation diagram and the dango driving degree,
Visual recognition of dumpling operation becomes easier. Further, the present invention is not limited to the embodiment described above, but is as follows.
There are some implementation methods that are variations of this (each method is not specifically shown). Recording unit 1-2 or learning unit 2-2
Records the operation data for one day, and makes a dumpling operation judgment collectively at a time when the group management control load is low, such as at night. In this method, there is no need to separately provide a state detection device, and there is an effect that the system can be configured with a single microcomputer. The provisional traveling direction of the waiting machine is reflected in the processing of the allocation unit 2-1 (a bonus value equivalent to 5 seconds is given to a call in the same traveling direction as the provisional traveling direction, and 30 seconds in the opposite direction. Give a considerable penalty value). As a result, the stand-by unit, which is advantageous for group management control, can be utilized so as not to be in a dumpling operating state. Figure 7
In steps 31-4 to 31-7, the processing is performed assuming not only the traveling direction but also all positions. As a result, it becomes possible to obtain the optimum standby position of the standby machine.

[0015]

According to the present invention, it is possible to detect a state in which the operation state of the group control elevator is deteriorated from the data of the current position of each elevator and the traveling direction by the above-mentioned simple calculation. . In addition, manage the detection result of the dumpling operation state in an easy-to-use manner and output the time transition of the dumpling operation degree in combination with the operation diagram to make it easier to report to the person in charge of design / maintenance or the building manager. Will be possible. Further, by surely detecting the dumpling operation state, the detection is used as a trigger to perform control for improving the state, and it is possible to promptly solve the problem.

[Brief description of drawings]

FIG. 1 is a diagram showing a system configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing an elevator arrangement and its ring representation.

FIG. 3 is a diagram showing an example of arrival time calculation to each floor.

FIG. 4 is a flowchart showing a main module of a dango operation detection program.

FIG. 5 is a flowchart showing an initial setting processing module of a dumpling operation detection program.

FIG. 6 is a flowchart showing a dango degree calculation processing module of a dango operation detection program.

FIG. 7 is a flowchart showing a subroutine for calculating a sum of squares of position intervals of the dango operation detection program.

FIG. 8 is a diagram illustrating an operation diagram during dango operation and an example of changes in dango operation over time.

FIG. 9 is a diagram showing an example of an operation diagram when the dumpling operation is canceled and a change in the dumpling operation degree over time.

FIG. 10 is a flowchart showing a dumpling operation time addition module of the dumpling operation detection program.

[Explanation of symbols]

 1 Operation state detection device 1-1 Judgment part 1-2 Recording part 1-3 Output part 2 Group management control device 2-1 Allocation part 2-2 Learning part 3 Elevator machine 4 Hall call button 10 Main module of dumpling operation detection program 20 Initial Setting Processing Module 30 Dango Degree Calculation Processing Module 31 Position Interval Sum of Squares Calculation Subroutine 40 Dango Operating Time Addition Module

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Naofumi Nakata 1070 Igemo, Katsuta City, Ibaraki Prefecture, Hitachi Ltd. Mito Plant (72) Inventor Kenji Yoneda 1070 Ige, Katsuta City, Ibaraki Hitachi Ltd. Mito Plant, Ltd. (72) Inventor Kiyoshi Niimura 1070, Moe, Katsuta-shi, Ibaraki Prefecture Hitachi Ltd. Mito Plant

Claims (8)

[Claims] 1. The operation of a plurality of elevators is managed,
In the group management control system that selects the elevator machine to serve the customers in the elevator hall,
An elevator group management control system, comprising: an operation state detection unit that calculates an index value indicating the degree of deviation of elevators from the positional relationship between elevator cars. 2. The elevator group management control system according to claim 1, wherein the index value indicating the degree of deviation is calculated from data of the current position of each car and the traveling direction of each car. 3. The elevator group management control system according to claim 1 or 2, wherein the index value indicating the degree of deviation is calculated by correcting the index value based on the operating state data in the ascending direction and the descending direction. . 4. When the index value indicating the degree of bias is calculated, a provisional means is provided for tentatively determining the traveling direction of the waiting-state machine that does not have the call-in-call command to the direction in which the index value is improved. Item 1. An elevator group management control system according to Item 1. 5. An allocating means for promoting allocation of a direction that coincides with a provisional traveling direction of the standby state vehicle when selecting a service number for a passenger in an elevator hall, and suppressing allocation in a direction different from the provisional traveling direction. The elevator group management control system according to claim 1 or 4, characterized in that 6. A stand-by unit that does not have an incoming call command at the time of calculating the index value indicating the degree of bias,
The elevator group management control according to claim 1, 4 or 5, further comprising: a means for calculating a standby position that most improves the index value, and a means for moving the standby state machine to the standby position. system. 7. The elevator according to claim 1, further comprising means for outputting the transition of the index value indicating the degree of bias over time in combination with an operation diagram showing the operation state of each elevator. Group management control system. 8. Using an index value indicating the degree of bias,
The elevator group management control system according to claim 1, further comprising: a unit that determines a time range in which the operating efficiency is decreasing, and a unit that selects and outputs an operating state of the elevator within the determined time range. .