JPH04133981A - Controller for elevator - Google Patents

Abstract

PURPOSE:To eliminate unbalance of services and attenuate confusion by predicting change in coming traffic flow with the use of data of the transition with time and change cycle of a predetermined set traffic demand. CONSTITUTION:At the time of starting elevator from a lobby or at the time elapse of 10 seconds since the last traffic flow data is prepared, the content of a table T129W is analyzed, and traffic flow transition detecting tables T129-1 to T129-360 are renewly prepared. Next, the initializing of a traffic demand transition detecting WK table T129W is executed. A traffic demand transition recording table T129 at the present is analyzed to predict change in learning characteristic mode. As a result, when it is judged that a time zone in a typical traffic flow comes to be completed, any one of the characteristic tables T256 are renewed, thus initializing the learning table group. After that, the service status for each elevator is judged so that the traffic confusion and service result are detected which are accumulated in the table T129W or the like.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an elevator control device, and more particularly, to an elevator control device that can manage elevator operation so as not to cause congestion. Regarding.

[Prior Art] Various technologies as described below are known as prior art related to elevator control devices.

For example, the conventional technology described in Japanese Patent Publication No. 31-3966 changes the departure interval depending on the number of elevator cars on the departure floor or a floor near the departure floor, or It has a function to change the departure interval according to the elevator car, and further has a function to perform departure promotion control for the first elevator when the load in the car exceeds a predetermined value.

This prior art attempts to reduce congestion in the car.

In addition, the prior art described in Japanese Patent Publication No. 43-13935 is based on the difference between the elapsed time from the departure standard floor of the ascending car and the elapsed operating time of the descending car. This invention relates to a device for adjusting departure intervals.

In addition, the conventional technology described in Japanese Patent Application Laid-Open No. 57-121569 counts the number of calls on each floor, calculates the priority of each floor in each time period from that value and past data, and uses this to determine the standard floor. This invention relates to a switching device.

In addition, the prior art described in Japanese Patent Publication No. 0], -30753 promotes the departure of the first elevator when the predicted boarding load on the standard floor exceeds the capacity.

Furthermore, the conventional technology described in Japanese Patent Application Laid-open No. 59-48369 has a function of learning the number of elevator users getting on and off each floor over a long period of time, and extracting a building-specific traffic flow mode.
Equipped with a function to generate the optimal driving program for each traffic flow mode, during actual driving, the driving program is selected and predicted from the current traffic flow data and representative traffic flow data learned for each traffic flow mode. This system creates traffic flow data and performs elevator group management based on this data.

Furthermore, in the prior art disclosed in Japanese Patent Application Laid-Open No. 63-51290, the door of the next elevator car is opened when a predetermined period of time has elapsed after the departure of the preceding elevator car.

[Problems to be Solved by the Invention] The prior art described above does not perform driving control that captures temporal changes in traffic demand over a short period of time.

For example, during a dispatch, the amount of traffic entering a building from other means of transportation may change periodically and repeatedly, especially when a large road traffic signal is operating near the building.
．． A situation arises in which passengers arrive at the elevator hall every five minutes. In such a case, the above-mentioned conventional technology has the disadvantage that congestion and quietness are alternately repeated.

In addition, during lunch hours, the traffic volume within the building is complicated due to measures to reduce crowding in the cafeteria, such as the systematic use of cafeterias at different times and changes in service menus depending on the time of day, as well as congestion avoidance actions by users. However, the pattern of this change in traffic volume may change depending on the weather.
It may also show some degree of constant temporal change from day to day.

The conventional technology described above does not control elevators in anticipation of these changes, so immediately after a car with only a few passengers leaves, a large number of new users arrive, and a large number of passengers This has the problem of frequently causing imbalances in the congestion situation, such as long waiting times in the hall, crowded cars, and operation that is similar to that of each floor.

An object of the present invention is to solve the problems of the prior art, and to manage the operation of elevators in response to traffic demand that changes to double action during short poems 141 during dispatch, lunch time, etc.
An object of the present invention is to provide an elevator control device that can eliminate service imbalance and alleviate congestion.

[Means for Solving the Problems] According to the present invention, the above object is to predict future changes in traffic flow using preset traffic demand temporal trends or change period data, control congestion floor vehicle allocation, This is achieved by performing operation management controls such as congested zone dispatch control, congested floor standby control, and congested floor departure control control, and by controlling car operation in accordance with the predicted timing of passenger arrival.

In addition, the purpose is to predict changes in traffic demand.
By extracting the change cycle from the change record data of past elevator usage status, the change time of traffic flow detected from the change record data of the current elevator usage status and the preset change cycle of traffic flow. By creating traffic demand change information, it is possible to predict the next time the traffic volume will change, the traffic demand after a fixed time, or the next time the traffic flow will change, and based on these, the elevator This is achieved through operation management.

Furthermore, the above object is to predict changes in traffic demand by learning the characteristics of traffic demand and its changes by detecting passengers or detecting arrivals at halls, and
Using preset traffic demand temporal trends or change period data, predict future changes in traffic flow, provide predictive guidance for congestion relief, congested floor vehicle allocation control, congested zone distributed vehicle allocation control, congested floor standby control By performing operation management control such as mixed floor departure suppression control, and guidance control according to the predicted passenger frequency, we can also predict the current traffic demand, perform hall call allocation control, and This is achieved by predicting traffic demand at a change point or after a certain period of time and managing the operation of multiple elevators.

[Function] The present invention captures the characteristics of temporal changes in traffic demand, such as temporal changes in traffic volume and periodic repeated changes, within the same traffic flow mode or within the time period of traffic management, and handles complex changes. Outputs information on predicted changes in traffic demand based on knowledge obtained from past learning about changes in traffic demand over time and the change record data on the day, and uses this to execute traffic management control to prevent car congestion. It is something to do.

As a result, the present invention suppresses the departure of cars with only a few passengers on floors where congestion is predicted to occur, and prevents some of the many new users who will arrive from arriving. ■Predicting the occurrence of congestion in advance and ensuring that the number of cars is just the right amount at the predicted time depending on the level of congestion. timely dispatch of vehicles to congested floors or congested zones to alleviate congestion.
It is possible to carry out preventive control operations such as:

[Embodiment] Hereinafter, an embodiment of an elevator control device according to the present invention will be described in detail with reference to FIGS. 1 to 11.

First, the overall operation management control based on characteristic data of changes in traffic demand in the present invention will be explained.

FIG. 1 is a block diagram showing an embodiment of an elevator control device according to an embodiment of the present invention.

In FIG. 1, M is a group management control device, SL is an event reservation service device, S2 is a rule setting device, S4 is an information control device, and K1- and 8 are car number control devices.

In the embodiment of the present invention shown in FIG. 1, the group management control device M can be configured by a program such as a microcomputer (hereinafter referred to as microcomputer), or can be configured by a logic circuit. It is. For convenience of explanation, the present invention will be described assuming that both are mixed, but the present invention is not limited to this and can be easily modified by those skilled in the art.

The input information to this group management control device M includes the position, speed, and door of each elevator car received by each car control device, which is given from 8 to each car control device Kl~ via a common transmission path L1. An elevator consisting of opening/closing status, status signals such as car load signal, car call, driving direction, etc., and management signals such as operation method number and failure detection number.
Control input data, reservation registration data from the event reservation service device S1 for reserving usage and date and time for special use, and extraction rules for generating traffic flow mode and change characteristics from the rule setting device S2; Furthermore, there is time correction data etc. from the maintenance device S3.

Hall calls and the number of people waiting in the hall are also input to the car control device from 1 to 8.

In addition, the number of people in the car and the number of people getting on and off instead of the load in the car mentioned above, the number of people arriving at the hall instead of the number of people waiting in the hall, and the hall congestion degree signal around the date of departure and departure of each elevator are input to the group management control device. You can also do that. Further, these inputs can also be directly taken into the group management control device M via an input/output circuit.

Although not shown, the car of each car is equipped with a car call button for registering the destination Koyo, and an in-car load detection device for detecting the load inside the car.In addition, in the hall on each floor, A hall call button and a device to detect the number of people waiting in the hall are provided. These signals are respectively taken into the machine control device from 1 to 8.

The rule setting device S2 can be configured integrally with an individualization support device that supports the setting of control targets such as congestion in the car and waiting conditions in the hall, and instead of being directly connected to the group management control device M, If necessary, it can also be used to indirectly connect to the group management control device M by writing the resulting data into a predetermined table, which is an electrically erasable non-volatile memory, via the maintenance device S3.

The group management control device M performs the control described below based on the input data as described above.

Since the group management control device M is composed of various functional circuits M1 to M8, the operation of the group management control device M will be described as control by these functional circuits.

The control circuit Ml calculates the number of people getting off and getting on the floor before and after the departure of the elevator based on the change in the load inside the car and the average weight data of passengers getting on and off the elevator, and the traffic flow detection circuit M2
The characteristics of the traffic flow are determined as shown in Fig. 2 (b) by cumulatively adding these values for each floor direction until a predetermined condition is met, and then this is sequentially recorded for a predetermined period of time to create a traffic flow learning circuit. M3 determines the temporal transition or change period for each traffic flow mode, and creates or updates characteristic data of changes in traffic demand, an example of which is shown in FIG. 10(b).

The driving control specification creation circuit M4 uses the current traffic flow data output from the traffic flow detection circuit M2 and the traffic flow learning circuit M3.
Traffic demand prediction data and traffic demand change prediction data are created from the representative traffic flow data that is predicted to occur in the future and the characteristic data of the changes, and further,
While giving top priority to operation according to specific specifications based on data from the event prediction circuit M8, control specifications for operation management and hall call allocation are created using the various data mentioned above.

The control circuit M1 allocates the generated hall call to one of the above-mentioned machines or a plurality of machines according to the commanded call assignment specifications, and transmits the assignment signal to the corresponding machine control device Kl-8 through the common transmission line Ll. Send via.

Then, the car control device 1 to 8 lights up the reservation information on the hall lanterns (not shown) provided at the landings on each floor. In addition, if the inside of the car is more crowded than the target value, the car control device 1 to 8 temporarily suspends the hall call service and , performs operation information guidance control that notifies congestion or fullness using hall information equipment installed in the corresponding hall, full-occupancy lights, etc.

Further, the traffic demand change prediction data outputted from the operation control specification creation circuit M4 is outputted to the information control device S4,
Information is transmitted to the in-building guidance system, building management system, etc. that are connected to the end, such as passage information such as ``The elevator is waiting. Please do not rush to board. It can be used to provide operational management information to building managers such as J.

The information control device S4 is also connected to a control system for escalators, subways, buses, etc., or a road traffic signal control system, and receives information related to the operating status of transportation means other than elevators and traffic demand within the building from these. This information can be used for learning control, creation of operating specifications, etc.

Next, prediction of changes in traffic demand will be explained.

In general, the demand for elevator traffic varies greatly depending on the layout within a building, the conditions of use of each floor, the location of the building, as well as the weather and surrounding transportation conditions.

For example, in the case of an elevator in a building near a station on the Loop Line, a situation occurs where the lobby becomes crowded every five minutes, which is the time interval between train arrivals.

As a typical example, FIG. 3(a) shows the overall transition of traffic demand at the time of dispatch in a building where a road traffic signal is nearby.

From this figure, ■ The dispatch time is one hour from 8:20 to 9:20, ■ The first half, 4/10, is 2/2 of the average.
3 or less, ■ The middle part of the 3/10 hour is as large as 7/4 of the average traffic volume, and ■ The latter half of the 3/1 hour hour has a linear decrease in traffic demand. You can understand the characteristics of temporal changes.

Figure 3 (b) is the time period TW with the maximum traffic demand mentioned above.
Taking I as an example, the change in the number of passengers from the lobby, which is the busiest floor, is plotted by magnifying the change over time.

The characteristics of the changes in the dispatch time during this time period TWI are as follows:
By finding the average value of the times t22 to t27 between peaks indicating the maximum number of passengers arriving at the lobby, data on the period of change can be obtained.

Note that storing the characteristics of temporal changes as sentences, rule data, etc. makes the handling of these sentences, data, etc. complicated both when they are created and when they are used. Therefore, in the embodiment of the present invention, as shown in FIG. 4(a), the time period in which driving is performed in the same traffic flow mode is divided into 15 sections, the proportion of traffic volume in each section is calculated, and the ratio of traffic volume between these sections is calculated. We will create temporal transition data in advance. Note that this FIG. 4(a) corresponds to FIG. 3(a).

Next, Figure 4 (b) and (C) show an example of learning the temporal trend data of lunchtime traffic in a building with a cafeteria on the first or second basement floor and a 30-minute time difference. This is explained by:

Figure 4(b) shows the increasing traffic volume depending on the number of people getting off at the first crowded floor, that is, the floor where the cafeteria is located.
Figure (c) shows the descending traffic volume depending on the number of people boarding from that floor.

From these figures, it can be seen that (1) the upward traffic volume and the downward traffic volume are changing separately, and (2) each peak occurs twice.

Using this property, traffic demand prediction data can be created as follows.

■ Calculate the average number of people moving per unit hour AP (j, f) from the traffic volume for getting on and off learned for each floor direction. ■ Predicting traffic demand from the traffic volume ratio Pu, Pd by direction in the current section. Data YP(j, f) is obtained using the following equation.

YP(○, f)=AP(0,f)XPu/l 5...
...(1) yp(1,f)=AP(1,f)xPd/15...
...(2), i="O" indicates the upward direction, and j=II II indicates the downward direction.

With this configuration, whereas in the past, driving control was performed using four traffic flow modes by focusing only on differences in traffic flow modes, in the 191 embodiment of the present invention, only one traffic flow mode is used. ■ - By reducing the daily traffic flow mode, it becomes possible to perform segmented learning by day (weather, season, specific day).

■In online control, it is possible to perform continuous elevator operation management control that predicts changes.

■Intelligent group management systems that use simulation functions to generate driving programs adapted to traffic flow based on learned data can shorten the development time.

■Since the number of times control targets are set in individualized group management can be reduced, maintenance safety and efficiency, including the number of meetings with customer-side managers, can be improved.

In the above, we explained that the cafeteria floor is on the basement floor, but if the cafeteria floor is on the upper floor and the traffic volume with the lobby is divided into two, the boarding and alighting traffic volume by direction on the first congested floor and the second congested floor. The same effect as described above can be obtained by creating temporal transition data.

Figure 4 (()) shows some of the changes in traffic volume during normal times. From this figure, it can be seen that traffic volume increases approximately every 30 minutes.
It is possible to extract the characteristic that the traffic volume is heavy at 3:00 pm.

FIG. 2 is a traffic demand transition table Tl29 created by the traffic flow detection circuit M2, which is shown in FIG. It is created in step P22. This table 29 can register the traffic demand every 10 seconds for one hour, and is updated with new data sequentially when the table becomes full.

As shown in Figure 2 (b), the data created each time includes the time at which the data was created, the total number of people getting on, the total number of people getting off, and the total number of people in the hall. These include the floor number for each crowded floor, the number of people getting on and off on that floor, and the number of people in the hall on that floor.

Figure 3 is a table T1 of traffic demand trends during dispatch hours.
The data recorded in 29 is shown for explanation purposes.As already explained, the QIA in Fig. 3(a) shows the temporal change in traffic volume, and the QIA in Fig. 3(b) shows the temporal change in traffic volume. teeth,
The ○ mark indicates the number of people arriving at the first crowded floor, and the X mark indicates the number of passengers boarding from the first crowded floor.

The arrival cycle tpx, when passengers arrive at the lobby in large numbers, can be easily determined if an automatic ticket gate is installed in the hall and the number of arriving passengers can be accurately detected, but it can be accurately determined using the data used to detect the number of people waiting in the hall using the hall congestion sensor. I can't. Therefore, in the embodiment of the present invention, the number of passengers who have boarded the elevator is detected from the amount of change in the detected value of the car load sensor, and the change period is determined from that value.

'Figures 4(a) to (d) show the first learning for each traffic flow mode.
Traffic flow mode temporal transition data table T shown in Figure 0
The data recorded on the 256A3 is illustrated for explanation. As already explained, these are shown in Fig. 4 (a
) transition curve Q1 shows the transition at the time of dispatch as shown in Figure 4(b),
The curve Q2 in FIG. 4(c) shows the transition during lunch time, and the curve Q in FIG. 4(d) shows the transition during normal times.

FIG. 5 is a processing flowchart illustrating the operation of the traffic flow detection circuit M2. Hereinafter, the operation of the traffic flow detection circuit M2 will be explained using this flow.

(1) Take the elevator from the lobby, or
It is determined that 10 seconds have passed since the previous traffic flow data creation. In this judgment, if the condition is not satisfied, step P
Traffic flow and usage status are collected from 29 onwards (step P21).

(2) If the conditions are met in the determination of step P21,
The contents of table T129W, which is a WK child table for detecting traffic demand transitions shown in FIG. 2, are analyzed, and traffic flow transition detection tables Tl29-1 to Tl29-360 are updated and created in order. When table T129-360 is reached, table T129-1 is updated next. By doing so, it is possible to always trace back the traffic flow trends up to one hour ago in units of 10 seconds and analyze the change layer IJI (step P22).

(3) Next, from the in-car load signal that the group management operation control circuit Ml takes in from 1 to 8 to each unit control device, the traffic demand trend is calculated to store the number of people getting on and off at each floor until the next learning section. The detection WK child table 129W is initialized (step P23).

(4) Analyze the traffic demand transition record table T129 up to Ozai, predict changes in the learning feature mode, and as a result of this prediction, it is determined that the time period with typical traffic flow is coming to an end. If so, update one of the feature mode tables T256 and initialize the learning table group for detecting this (steps P25, P26, P28).
).

(5) Thereafter, as described above, the service status of each elevator is sequentially judged, traffic demand and service results are detected and accumulated in the work table T129W and others (steps P29, P2O).

By doing as described above, the embodiment of the present invention has the advantage that it is not necessary to include the situation at the turning point of traffic flow in the important traffic flow mode learning data as in the conventional case.

FIG. 6 is a flowchart illustrating the specific operation of step P26 of the feature mode table update process shown in FIG. 5, and this will be explained below.

(1) Traffic flow and usage status are learned, and the traffic flow data table T256A2 and usage status data table T256A4 shown in FIG. 10 are updated using an exponential smoothing method or the like. In addition, service status data table T256A5
is updated, and it is determined whether there has been a periodic change in traffic flow (steps P261 to P263).

(2) If periodic changes in traffic flow are recognized in the determination in step P263, a change characteristic data table that records the characteristics of temporal changes in traffic demand within the same traffic flow mode, which is important for the present invention. T256A6 is updated and the short periodic changes in traffic flow shown in FIG. 3(b) are removed to create large traffic demand transition data (steps P264 and P265).

(3) Also, if it is determined in step P263 that no periodic changes in traffic flow are recognized, the traffic flow change data is deleted and the traffic demand transition table T129 shown in FIG.
The data portion belonging to the corresponding traffic flow mode is aggregated into 15 categories as shown in FIG. 4, and updated processing is performed such as recording it in a vacant area of the transition data table T256A3. For example, if there is already transition data that is recognized as being in the same time period, a process is performed to create updated data from this data and the current measurement data (step P266,
P267).

(4) In what day of the week, weather, and time of day this typical traffic flow mode occurs? Table T256A6 (
b) and (C), and issues a request to update the control knowledge to the administrator of the diagnostic system and rule setting device installed remotely (steps P268 and P269).

FIG. 7 is a flowchart illustrating the operation of the group management operation control circuit M4 to create operation specifications, and this will be explained below.

(1) First, characteristic data of current traffic demand is created from table T129 and table T256W (step P41).

(2) Using the data created in step 41, identify which of the learned representative traffic flow modes is approximated, and extract one or more modes. Here, the description will be made assuming that modes opml and opm2 have been extracted (step P42).

(3) The confidence level for each of the traffic flow modes extracted in step P42 is determined, and predicted traffic flow data is thereby created (steps P43 and P44).

(5) Thereafter, based on the control knowledge, processes such as selecting a call allocation method suitable for these predicted usage conditions and setting its parameter values are executed (steps P45 to
P424).

FIG. 8 is a specific flowchart of the process of creating predicted traffic flow data in step P44 of FIG.

(1) First, create the elapsed time NOMTM after detecting a new traffic flow mode (steps P441 to P44
3).

(2) Determine whether the traffic flow mode that has been used until now has completely disappeared, and calculate the elapsed time NOMTM of the first operation traffic flow mode that has been used as an aid until then in the second operation mode.
The elapsed time of the operation traffic flow mode is substituted into PMTM, and the time is inherited (steps P444, P446).

(3) In order to respond to traffic flow change cycles that occur repeatedly in short time periods, traffic demand 4a) Transfer table T129
The period g of the change period, which is a part of the characteristic data table T256A6 of the change in traffic demand learned as the traffic flow mode, and the degree of periodicity data h consisting of the degree of variation and confidence. From this, the off-peak of the current traffic demand is detected, and the elapsed time PETRFTM from the off-peak required for predicting the next peak is created. This process may be performed by counting the elapsed time since the previous peak detection (steps P447 to
P449).

(4) The period g of the change period, which is part of the traffic demand change characteristic data table T256A6, learned as the second operation traffic flow mode, and the degree of periodicity data h consisting of its degree of variation and confidence. , the next scheduled peak time NPETIM is created, and the predicted traffic demand data NTRF (i.

j, f,). Furthermore, based on this forecast data, current traffic demand, and the number of operating elevators, the atmospheric number request specification 5C for the congested floor or congested band is
Create KN (j, f) (steps P450, P2
S5).

Note that the processing in steps P450 to P452 described above may be executed every time in order to adapt to traffic demand that changes from moment to moment. However, if the number of elevators in the air and the request time data change too closely If this happens, the operation control system will not be able to follow this and the performance of the elevator system may deteriorate, so care must be taken.

(5) Next, current traffic demand data PTRF is created using the following formula (step P453).

CRK+n2 Km2 Note that k in the above formula (1) is as shown in (2) below.

k=Kml+Km2+C=...(2) (1) above
, (2), the meanings of the symbols and the names of the data are as follows.

i...Getting in/getting off J...Ascending direction/descending direction Kml...Approximation degree K of the first traffic flow mode for driving
m2... Approximation CI of the second traffic flow mode for driving
...Weight constant between current traffic flow and learning data GTRF 1 ... Traffic flow data of the first traffic flow mode for driving (data selected from data table T2 56A2 to 256T2) ) GTRF2... Traffic flow data for the second traffic flow mode for driving MTRF... Current traffic flow data (Table T
256M2) Based on these data, select the intensive service floor and decide its degree. Furthermore, data on passenger density on each floor is obtained and these operating parameters are provided to the group management operation control circuit M1.

(6) Next, in order to control the elevator departure interval, first, using equation (3), the current average elevator interval KT
Find M.

KTM-Σ [RTTU (K) +RTTD (K)
) / SVK ...... (3) k = 1 In Equation (3), SV K is the number of elevators in operation, and RTTU and RTTD are the number of elevators per revolution of each elevator. These data are shown in tables T427, T421 and T4 in Figure 11.
22, the value stored in step P30 of the traffic flow detection circuit M2 is used as the measured value each time. That is, the lobby floor given to the operation control circuit M1 is determined based on the next scheduled peak time obtained in steps P441 to P453 described above, the two traffic demand data, and the current elevator spacing KTM obtained using equation (3). , Elevators on crowded floors such as cafeteria floors, floors where service is particularly important.
The departure interval adjustment specification 5C3T is created based on the driving specification creation rule set from the event reservation service and the rule setting device S2 (step P454).

FIG. 9 is a flowchart specifically explaining operations of departure interval control from a crowded floor and control of the number of waiting cars on a crowded floor, which are one of the operation management controls performed by the group management operation control circuit M1.

In this flow, departure control, which will be described below, is performed for the floors where each elevator is located, from No. 1 to No. 8 in order, through steps P140 and P156.

(1) First, the waiting time is updated (counted) on the floor where the own car is located, and it is determined whether there is a parameter specifying a crowded floor. If there is a specified parameter for a crowded floor, the supply number φK indicates how many other Nirevators can arrive and wait at that floor within the requested time, that is, at the time of the next peak occurrence. (J, f
) is determined (steps P141 to P143). So (2
) In-car weight or in-car congestion level CWG of own car
(k) is the car congestion determination parameter UC8W on the relevant floor
G (kj, kf) is determined, and if the congestion level is not high, the number of waiting cars on the floor concerned φK (
kj, kf) satisfies the required number of units 5C8T (kj, kf). In addition, here kj and kf
indicates the service schedule direction of the elevator of the own machine and the floor of the own machine (Steps P144, P145)
.

(3) The car of the own car serves the current floor,
The elapsed time SWTM (k
) determines whether or not the departure interval adjustment time has elapsed, and if it has elapsed, the long waiting prediction hall call allocation permission level signal CLTWTM is used to give permission for hall call allocation on another floor and accompanying immediate reservation guidance. O to (k)
(Steps P146, P150).

(4) On the other hand, control data CT for suppressing hall call assignment or service guidance for an empty elevator car serving a floor that does not correspond to any of steps P142, P144, ]46.

A value determined by the method given from the rule setting device S2 is determined and substituted into TWTM (k) (step P14
7).

(5) Check whether a long waiting time is predicted for the reservation information hall call of the own machine due to the occurrence of a new elevator call, etc. based on the specified long waiting time UCLWT (j, f). If it has occurred, clear the hall call allocation suppression data CAS DTM (k),
Further, a command ECL for promoting automatic departure and validating the door opening button is set (steps P148, P2S5, P2S5).

(6) Conversely, for elevators that need to wait on that floor, the allocation is suppressed and the door opening button is disabled (Step P) 4
9, P2S5).

(7) Next, in order to respond to the next peak traffic demand, it is determined that there are no in-car calls or hall calls for which reservations have been announced.
Command door closed standby (step P154, step P1
55).

In addition, in the case of processing using the conventional technology, the above-mentioned control generally performs departure interval control for each floor direction, but even in the case of 15 floors, which is the average number of service floors, the same procedure is performed 28 times. It is necessary to perform processing. On the other hand, in the case of the embodiment of the present invention described above, each elevator
Since the processing is based on the floor where the floor is located, the processing time can be significantly reduced.

Figure 10 shows the learning table created for each traffic flow mode.b) shows the traffic demand change characteristic data table T25.
Details of 6A6 are shown.

For the group management operation control circuit M1 and traffic flow detection circuit M2, a table T256M of the same size as this table T256A for storing current traffic demand and a table 256W for learning traffic demand etc. are provided.

FIG. 11 shows part of the control table used in the group management operation control circuit M], and FIG. 11(a) shows the table T400 created by the operation specification creation circuit M4 and the group management operation control A table T420 created by circuit Ml is shown. In addition, FIG. 11(b) shows a rule setting device S2, a maintenance device S3, or an event reservation service device S1.
A table T500 is shown that stores judgment data such as in-cage congestion and long waiting times, which are set based on control targets set by the above, in the direction of each floor.

FIG. 12 is a block diagram illustrating the configuration of another embodiment of the present invention. .

This other embodiment of the present invention has almost no difference from FIG. 1, except that the circuit M22 detects a new traffic flow and generates a traffic flow mode, and the circuit M23 provides a simulation function and an expert system. This embodiment differs from the embodiment shown in FIG. 1 in that control knowledge is acquired in advance. The operation specification creation circuit M4 is simply a block diagram expressing the basic functions of the operation specification creation flowchart shown in FIG. 7, and has no particularly new configuration or function.

[Effects of the Invention] As explained above, according to the present invention, the following effects can be achieved.

By delaying the operation of cars with only a few people on board heading to floors where congestion is expected to occur, or by restricting departures from crowded floors, services with a large number of new users may be prevented. can eliminate the imbalance.

In addition, we can predict the occurrence of congestion in advance and, depending on the degree,
In addition, since preventive control is carried out to timely allocate just the right number of cars to congested floors or congested zones at the predicted time, it is possible to alleviate the occurrence of congestion and reduce its frequency.

In addition, the preset time trends or change cycles of traffic demand are determined by learning past changes in traffic flows and analyzing their characteristics, so they can be easily detected based on changes in the building environment or the situation inside the building. Elevator operation can be controlled in accordance with long-term changes in traffic demand.

[Brief explanation of drawings]

FIG. 1 is a plot diagram showing the configuration of an embodiment of the present invention, FIG. 2 is a diagram explaining a traffic demand transition table created by the traffic flow detection circuit M2, and FIG. 3 is a diagram showing the traffic demand during dispatch time. FIG. 4 is a diagram explaining the data recorded in the transition table, FIG. 4 is an explanatory diagram of change transition data, FIG. 5 is a processing flowchart of the traffic flow detection circuit M2, and FIG. 6 is a specific example of the feature mode table update process. Flowchart, FIG. 7 is a flowchart of the specification creation process by the group management operation control circuit M1, FIG. 8 is a specific flowchart of the predicted traffic flow data creation process,
FIG. 9 is a diagram showing part of the control table used in the departure interval control from the congested floor, FIG. 11 is a diagram showing a part of the control table used in the group management operation control circuit M1, and FIG. 12 is a block diagram showing the configuration of another embodiment of the present invention. It is a diagram. K1- to 8... Unit control device, M...
Group management control device, Ml... Group management operation control circuit, M2... Traffic flow detection circuit, M3...
Traffic flow learning circuit, M4... Operation control specification creation circuit, Ml... Calendar time phase circuit, S4...
...Information control device. Agent: Patent Attorney Junji Takeshi Department (1 other person)

Claims (1)

[Scope of Claims] 1. In an elevator system that includes an elevator that serves multiple floors and an elevator control device that controls the elevator, it is possible to a traffic demand change prediction device that predicts changes in traffic flow and outputs traffic demand change information; a congestion prediction and guidance device that predicts congestion or predicts congestion relief status based on the traffic demand change information; Floor dispatch control, congested zone dispatch control, congested floor standby control,
It is equipped with a traffic management control device that performs one or more of the following: congested floor departure suppression control, congested floor passage suppression control, and arrival time adjustment control to congested floors, and provides guidance control or control that matches the predicted passenger generation cycle. An elevator control device characterized by controlling operation. 2. The elevator according to claim 1, wherein the traffic demand change prediction device outputs, as change information, a temporal change cycle of the usage status extracted from change record data of the usage status of the elevator. control device. 3. The traffic demand change prediction device is configured to detect changes in traffic volume (peak or off-peak) detected from change record data of the current usage status of the elevators, and automatically set by learning in advance or externally set. 2. The elevator control device according to claim 1, wherein the elevator control device outputs traffic demand change information based on a traffic flow change cycle set by the information. 4. The traffic demand change prediction device is configured to detect changes in traffic volume (peak or off-peak) detected from change record data of the current usage status of the elevators, and automatically set by learning in advance or externally set. 2. The elevator control device according to claim 1, wherein the elevator control device outputs traffic demand change information including the next traffic flow change time based on the traffic flow change period set by the information. 5. The elevator according to claim 1, wherein the traffic demand change prediction device outputs predicted traffic demand information that predicts the next traffic volume change point or the traffic demand after a certain period of time. Control device. 6. Claim 1, wherein the traffic demand change prediction device learns the traffic demand and the characteristics of its change based on the results of detecting passengers in the elevator or detecting passengers arriving at the hall. Control device for the elevator described. 7. Multiple elevators serving multiple floors;
In an elevator system that includes an elevator control device that controls this, an allocation control device that assigns a generated hall call to one of the elevators, and one or more of preset traffic demand temporal trends or change period data are used. a traffic demand change prediction device that predicts future changes in traffic flow and outputs traffic demand change information; a congestion relief guidance device that predicts and guides congestion relief situations based on the traffic demand change information; a congestion floor vehicle allocation control; Equipped with a traffic management control device that performs one or more of the following: distributed vehicle allocation control in congested zones, standby control on congested floors, control to suppress departures from congested floors, and control to suppress passing through congested floors, and provides guidance according to the predicted passenger frequency. An elevator control device characterized by performing control and group management control operation. 8. The traffic demand change prediction device includes first predicted traffic demand information that predicts the current traffic demand of the elevator;
and second predicted traffic demand information that predicts the next traffic volume change point or traffic demand after a certain period of time, and the call allocation control device controls the elevator based on the first predicted traffic demand. 8. The elevator control device according to claim 7, wherein the operation management device performs elevator operation management based on the second predicted traffic demand. 9. The traffic demand change prediction device is equipped with a traffic demand change feature extraction means that learns the temporal transition or change cycle of traffic demand for each traffic flow mode, and uses the extracted traffic demand change feature data as the characteristic of the change. 8. The elevator control device according to claim 7, wherein the elevator control device is set in advance as data. 10. The traffic demand change prediction device includes a means for detecting traffic demand for the elevator, a means for generating a traffic flow mode having at least an identification element regarding a change in traffic demand, and a means for generating a traffic flow mode for each generated traffic flow mode. Claim 7: characterized in that the transportation demand change feature extraction means is configured to learn the change in traffic demand or the change cycle, and the extracted traffic demand change feature data is set in advance as the change feature data. Elevator control device as described in section.