JP2574876B2 - Elevator group control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an elevator group management control device.
The present invention relates to a group management control device that performs control that satisfies multiple items of control goals.

[Conventional technology]

Conventionally, the elevator group management control device, with the purpose of improving the operating efficiency of the elevator and improving the service of passengers,
A system has been adopted in which generated hall calls are monitored online, and the average waiting time is shortened by allocating the calls to the optimal elevators in consideration of the service status of all hall calls. Recently, when selecting an elevator to assign a hall call from among a plurality of elevators, a variable parameter is added to an evaluation function for evaluating each elevator, and the variable parameter value is changed according to traffic demand. Japanese Patent Application Laid-Open No. S58-58158 discloses a method in which a parameter value satisfying a preset target value is learned using the obtained result, and call assignment control is performed using the parameter according to the operating condition of the elevator. −52162
And JP-A-58-63668.

In this method, two control target values, that is, waiting time and energy saving, can be set by a level command from a switch or a building management system. The stop call near the generated call is evaluated, and the number of stop calls is large. It introduces a stop that preferentially assigns generated hall calls to elevators and an evaluation index. By appropriately changing the weight coefficient of the stop call evaluation index, a weight coefficient that optimizes the waiting time is obtained, and conversely, if this weight coefficient is increased, an energy saving effect can be obtained. Meanwhile, Japanese Patent Publication No. 62-70,
Japanese Patent Publication No. 62-71 also describes a method in which the waiting time and energy saving are taken into consideration. On the other hand, Japanese Patent Publication No. 58-56709 discloses a method in which the fullness forecast is added to the evaluation index.
The publication describes that at least one of a forecast deviation probability and a fullness probability is added as an evaluation index.

[Problems to be solved by the invention]

In the prior art, which has two control goals of waiting time and energy saving, by reducing the average waiting time, the dissatisfaction of passengers can be eliminated, but still,
There were some problems. Specifically, long waits always occur at the same time on a specific floor, calls are assigned to distant elevators even when a standby elevator is nearby, or large luggage such as wagons Despite the transportation, the elevator was assigned to a congested elevator, and after the elevator departed, various complaints arose, such as having to generate a call again. Had been told.

In addition, control target values other than the waiting time and energy saving, and the waiting time and the probability of fullness (probability of missed forecast) are not taken into consideration. It is difficult for the owner, manager or the like to instruct and perform the operation, and it is impossible to perform multi-target control according to various combinations.

To respond to various complaints in the building once delivered, elevator users such as building owners and managers request modification from the elevator manufacturer, and change or add programs on the elevator manufacturer's side. Natsu ROM had to be reworked. This usually requires a lot of manpower and time, and cannot satisfy all of the diverse requirements of passengers, passengers, building owners, managers, etc. However, there was a problem that it was difficult to show the effect of actually operating the elevator after changing the program.

As described above, there are restrictions on the resources available in each building, the number of elevators, the number of passengers, the elevator speed, and the like, and inputting a desired value that can be realized within these restrictions requires much trial and error and experience. Further, it is difficult to digitize the user's wishes in advance without contradiction. However, in the prior art, there is no idea of inputting a user's request, and therefore no consideration is given to a method of determining the request without contradiction, and it is difficult to realize control that matches the individuality of the building to be installed. It was.

An object of the present invention is to make it possible to set a control target consistent with a wide variety of requests in a user-friendly format in an easy-to-understand format, to directly incorporate the control target into a control system, and immediately reflect the control target in control. It is to solve this problem. It is still another object of the present invention to provide an elevator control device that allows a user to easily add or delete various required items.

[Means for solving the problem]

The above object is to set the control target value by setting the importance for each target at the same time, taking into account the importance, and considering a control method capable of satisfying the target value. Provide a control target setting range determination unit that determines the setting range of, if a target value exceeding the setting range is set, by providing a function to notify the setter of the corresponding direction, the control target value of three or more items Setting can be easily performed, and a control method that satisfies the set target value can be easily determined.

[Action]

In order to determine a control target value while maintaining consistency between a plurality of control target items, the degree of interaction between each control target is determined by:
The range of control target values that can be set is determined in advance in accordance with the control items to be set and their importance (priority), which are obtained in advance by a simulator and recorded in the form of a relevance table.
Since the target setter is notified, it is possible to suppress improper setting. In addition, when the setter sets a target that deviates from the target setting range, a function for notifying the setter of a measure that satisfies the deviated target is added so that each control can be performed under the understanding of the setter. Since the target can be set, the building manager can deal with problems in use.

〔Example〕

Hereinafter, an embodiment of the present invention will be described with reference to FIGS. 1 to 12.

FIG. 1 is a configuration diagram of one embodiment of the present invention. Reference numeral 1 denotes a group management control device that executes operation management control of a plurality of elevators, and reference numeral 2 denotes a group of control units for controlling individual elevators. In this figure, three units A to C (2A to 2C) are used. No. unit control device. A hall call device 3 is installed in each floor elevator hall, and this signal is transmitted directly to the group management control device 1. Further, the input / output device 4 provided separately instructs a change of the control target value and the like. The result of executing the group management control can be output to the input / output device 4. Next, the operation in the group management control device 1 will be described.
Signal of hall call device 3 installed in each floor hall, and
Information such as the position and direction of each elevator, the car call, and the number of passengers is taken into the group management operation execution unit 1-1 from each of the car control devices 2A, 2B, and 2C. If a new rule call signal is transmitted from the hall call device 3, based on the various information transmitted from each of the unit control devices 2A, 2B, and 2C, the current operation status and FIG. The traffic flow pattern shown is determined, and the control method and each control parameter corresponding to the closest traffic flow pattern are selected from the control table shown in FIG. 2, and the driving method and the call method recorded in the selected control table are selected. Perform assignment control. Here, one of the driving methods is to determine a service floor as shown in FIG. The shaded portions in the figure indicate serviceable floors of each elevator. In addition, it means that a plurality of elevators are preferentially dispatched to a floor where users are concentrated, and that a door opening / closing time is to be changed.

Next, the call assignment control method is based on the control target setting unit 1 in advance.
For the control target value set in -4, items evaluated by the call assignment evaluation formula (for example, waiting time, boarding time, car congestion degree, reservation appropriate rate, energy saving, information guide transmission rate, long waiting time) Rate, etc.), multiple evaluation formulas are prepared for each control item, and this evaluation formula is selected based on data such as elevator specifications and traffic flow patterns. Ask for. A weighting factor is added in accordance with the degree of importance set for the obtained evaluation value, and the sum (or difference) thereof is taken as a total evaluation value, which is generated at a unit having the minimum (or maximum) of the total evaluation value. A hall call is assigned.

Here, the traffic flow pattern is obtained by obtaining the state of movement from the data of getting on and off the person, and FIG. 3 shows a typical example. For example, pattern A is a pattern that occurs during work hours,
Upward movement from the base floor is dominant.
On the other hand, the pattern B is a case where the flow of the people going to the reference floor in the state of the leaving time is dominant. The pattern C is a pattern seen in the first half of the lunch time, and is a pattern in which people concentrate on a certain floor of the dining room. In the latter half of lunch, there is also a pattern that this is reversed from a specific floor. The pattern D is seen in a normal time zone, and frequently moves between floors.

Next, the data learning unit 1-2 collected the data in the group management operation execution unit. The number of passengers on each floor (collected in units of 100 people or in units of 5 to 10 minutes), the waiting time of each hall floor, the service time of car calls (the time from the occurrence of a car call to the arrival at the destination floor), Statistical processing is performed on various data corresponding to the control target items shown in Fig. 5, such as the number of passengers in each car, the number of changes in call assignment, the floors and times of passage of the call assignment cars, the floors and times of occurrence of long waits, etc. Is recorded in a learning data table provided for each item.

The control program generation unit 1-3 includes various learning data tables created by the data learning unit and the control target setting unit 1
-4, the control target value table (FIG. 6) created in
A control method that satisfies the control target value recorded in the control target table is selected using the control strategy determination knowledge base and the inference function. Next, using the selected control method, simulation is performed using a simulator built in the same part, and an optimum control parameter is obtained based on the result. The selected control method and control parameters are registered in the control table shown in FIG.

The control target setting section 1-4 classifies and sorts the control target items and their values input from the input / output device 4 according to the use environment of the elevator to create a control target value table.

In the control target setting range determining unit 1-5, the input / output device 4
Instruct the user to select a control target item to be set, and then have the user input the importance (priority) of achieving each target for the selected control target item. In order to input this importance level, as a simple way to set the importance even if the target setter is not familiar with group management control, the importance of the target items between two items is set by a pairwise comparison and created. For example, there is a method of obtaining a unique value of the paired comparison matrix obtained.

The following method can be considered as a method of determining the setting range of each control target according to the input situation. First, in the case of using the conventional evaluation method of assigning a call that preferentially achieves the waiting time, the achievement degree of each input control target item is measured in advance, or obtained by simulation, and the value is set as a standard target achievement value. It is recorded in the control target predicted value table of FIG. Further, the maximum achievement value for each control item in the same state is obtained and recorded in the control target predicted value table in advance.

In the control target value setting range determination unit 1-5, the maximum achievement value of the control target item having the highest input importance is a _max , the standard value of the control target item is a _ave , and the control of the maximum importance is performed. The degree of influence of the target item on the other item B is a _ab . The degree of influence obtained using a simulator or the like is selected from among those recorded in the table shown in FIG. ), The maximum (or minimum) possible set value b _S of the item B can be obtained by the following equation with the standard control target value b _{ave of the} item B.

b _S = b _ave × {1 + a _ab × (a _max / a _ave )} (1) where a _ab ≠ 0 where a _ab = 0, the maximum value b for item B
Considering that _max can be set, the maximum value (or the minimum value) that the other term can take for the target term of B is obtained, and the larger one is set as the maximum settable value. If the maximum value (or the minimum value) is determined, the possible range is a value smaller (or larger) than the maximum value (or the minimum value) of the settable values, and is uniquely determined.

The settable range described above is transmitted to the input / output device 4 and displayed on the device. The target setter sets a value within the set range, thereby preventing an unachievable target value from being set. Also, when a target value exceeding the target settable range is input, the target value set range determining unit determines and notifies the user of a change in importance, a change of a target item, or the like. There is also a function of notifying the target setter that the operation has been performed.

The above operation will be described using a flowchart. Ninth
The figure shows the operation from the input of the control target to the determination of the target value. First, a command for new or change of the control target is transmitted to the group management control device 1 (401). From the group management control device 1 to the input / output device 4, the current building specification (402),
The elevator specification (403) and the like are transmitted, and if necessary, the setter makes corrections, and the group management control device 1
Send those data to. Next, the setter sets a time zone or a traffic flow in which it is necessary to achieve the target, and transmits the information to the group management control device (40).
4) Next, an item for which the setting is to be changed is extracted from the control target items prepared in advance (405). However, in this extraction, the term of the waiting time, which is a basic control item, is always included. The reason for this is that changing the target value of other target items often affects this wait time term, and the wait time becomes extremely bad, and the setter sets how much in advance to avoid complaints from users. This is to know whether it will be a waiting time. In addition, the user is asked to set the importance of each item together with the setting of the control item to be set. When the above processing is completed, the control target value setting range is determined by using the set value and the predicted achievement value table of each control item and the control target value setting range using the above-described method or the like.
Determined by -5. After the setting range is determined, each control target value is input. As a method of inputting the target value, a method of expressing a settable numerical value for each control item and inputting the corresponding value from a keyboard, or a method of using a graphic such as a radar chart can be considered. FIG. 10 shows an example of a display screen when the radar chart is used. In the figure, a standard value (each target item is connected by a solid line) and a settable area between each control target item are indicated by oblique lines. In this example, the demand for the waiting time is high, and the riding time is often slightly longer, so that the riding time is smaller than the standard value.

In the case of radar chart input, input can be simplified by using a mouse or touch panel together.

Next, it is determined whether or not each input control target value is within the set control target value setting storage range (40).
7). If it is out of the settable range, the best method that satisfies the target is searched (408) and presented, and the fact that it is out of the settable range is indicated on the display unit of the input / output device 4 or the sound generation. The notification is made using the section (409).

If the value is within the setting range, the control target setting unit 1-4
The input target value is transmitted to (410).

However, if the instruction is not received in an area other than the target settable area, the processing of 407, 408, and 409 becomes unnecessary, and it is possible to add a determination of whether or not there is a reset before the processing of 410.

The control target setting unit 1-4 determines a traffic flow pattern from the input control target value and a time zone in which the target value is achieved, and changes the content of the corresponding control target value table to the input value, or Create and record a new control target value table. Next, here, the control program generator 1-
3 is notified that the control target value table has been changed or newly installed. Next, in the control program generation unit 1-3,
A control method and a control parameter that satisfy the control target value are obtained. FIG. 11 shows a flowchart of the operation. Here, first, the traffic flow and time zone are read (301). Next, the control target value is read (302). A control method that satisfies the target value is selected by using a preset knowledge table and an inference function (303).

Next, using the selected control method, the simulation is performed by the simulator built in the control program generator 1-5 (304). Here, the simulation is performed several times by changing each control parameter, a result corresponding to each control target item is obtained, and the control parameter having the highest degree of achievement with respect to the target value is selected (305). The result determined in this process is registered in the control table, and the process ends here. The above is the description of the processing flow when information on a new or changed control target value table is transmitted. Here, the processing described below is usually performed. FIG. 11 shows the processing flow.
First, a traffic flow pattern is generated from the traffic flow data created by the data learning unit (310). Next, it is determined whether the generated traffic flow pattern is a new pattern or a pattern registered so far (312). If it is a new traffic flow pattern, the two closest traffic flow patterns are selected (313). In the selected traffic flow pattern, the registered control methods are compared to determine whether they are the same method (314). If the control method is the same, the simulation is performed by changing the control parameters using the new traffic flow pattern and the control method (315). next,
The degree of achievement with respect to the control target is calculated from the control target value set in the same time zone and the simulation result (316).
Next, the respective achievement levels are compared, and the control parameter with the highest achievement level is selected (317), and the control method and parameters are registered for the traffic flow pattern. If two control methods are selected, one of them is set as the first control method, and that method is extracted from the control table (318). Next, the processes of 315 to 317 described above are performed for the extracted method (319). Subsequently, data relating to the second control method is extracted (320), and the processes of 315 to 317 are performed similarly to the case of the first control method (32).
1). Next, the best target achievement value is compared for each control method, and the control method and control parameter with the larger achievement degree are selected (322). The selected control method and control parameters are registered in the control table as the control method and control parameters for the new traffic flow pattern.

If it is determined that the traffic flow pattern is the same as the conventional traffic flow pattern in the determination 312 as to whether the traffic flow pattern is the new traffic flow pattern, the control parameters are changed and the traffic flow is simulated (324) by using the actually measured data. ). Next, the degree of achievement of the set control target is calculated from the result of the simulation (325). The best control parameter is determined from the degree of achievement thus obtained (326), and its control method and control parameters are registered (327).

As described above, according to the present embodiment, when three or more control target values are set, the settable range of each target value can be clarified, so that the target can be easily set and the set target value is satisfied. It is not necessary to prepare many knowledge bases for selecting the control method to be controlled, and the speed of selecting the control method can be improved.

Next, another embodiment of the present invention will be described. First, in the setting of the control target value, a method of designating the settable area was adopted, but as another method, the total settable points for the set control items are obtained, and each time the setter sets each target value, A method of subtracting from the total evaluation point from the set value. If the total evaluation point becomes zero on the way, the target value of the previously set item can be reduced and the target value of the selected control target item can be set. Then, when the setting of all items is completed, a setting completion signal is sent to the control target setting section 1-4 to end the setting.

In the case of using this method, it is necessary to normalize each control target value and weight each to achieve smoothing, and to obtain the degree of relevance between control target items in the same way as the method described above. There is. Also, if there is at least one independent target item having no relevance regarding the total score, it is necessary to treat it as an independent setting item and provide a setting area different from the comprehensive evaluation value.

FIG. 13 shows an example of the setting in the case of using a radar chart. When indicated by the solid line, the total score is 80 points, as indicated by the broken line, with emphasis on wanting to ride quickly (waiting time) and riding a car (the degree of congestion in the car). By slightly improving the notification time (reservation notification time) that the player wants to be notified early, the score becomes zero and the target value cannot be set any more. As described above, the present method is effective when setting a target value in a graphic display such as a ratchet chart, and the target setting can be easily performed.

When obtaining the total evaluation score, the following method can be considered as a method of normalizing each control item. Assuming that the minimum serviceable average waiting time in a traffic flow pattern or time zone for each control item is 25 seconds and the maximum average waiting time is 70 seconds, the minimum average waiting time is 100 and the maximum average waiting time Normalize as 0. Other terms can be normalized similarly.

Up to now, the configuration of this system has been explained by directly inputting control target values. However, it is difficult for a building manager or the like to set individual control target values. As shown in the figure, it is necessary to convert the input target value into an actual control target value so that it can be set by normalization. Therefore, as shown in FIG. 14, a sensitivity target conversion unit 1-6 is provided, where the input sensitivity target value is converted into a control target value, and the above-described processing is performed. In this case, the control target setting range determination unit previously described uses the conversion function or the value recorded in the conversion table to the control specified by using a simulator or the like in advance in accordance with the specifications of the building and the usage environment of the elevator. By adding the maximum (or minimum) settable value from the weighting factor obtained from the degree of association between items, one conversion function can cope with various conditions.

FIG. 15 shows an example of a conversion function. (1) converts a desire to ride quickly into an average waiting time. If this conversion function is f (t ₁ ), the function is set to a minimum value. the congestion time zone is added possible values a ₁ can be determined target value corresponding. For example, if a desired value of 100 is input, a value of 40 seconds is set, and call allocation exceeding 40 seconds is prevented as much as possible. In the case of (2), in order to set the congestion degree in the car in response to the desire to ride an empty car, the setter 10
When 0 is entered, the hotel operates so that calls are not allocated to cars with a congestion degree of 25% or more. In this example, the type of building and the traffic demand are handled separately, but in reality, the restriction by the building type needs to be handled simultaneously with the traffic demand. As described above, if an input is made in the form of sensitivity, it is not necessary to consider restrictions on the input.

〔The invention's effect〕

ADVANTAGE OF THE INVENTION According to this invention, setting of an unrealizable target value can be prevented beforehand, troublesomeness with respect to a target value setting can be reduced, and a desired target value can be set.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention, FIG. 2 is a diagram showing an example of a control table, FIG. 3 is a diagram showing a typical example of a traffic flow pattern, FIG. FIG. 6 is a diagram showing an example of a control target item corresponding to a sensitivity target item, FIG. 6 is a diagram showing an example of a control target value table, FIG.
Fig. 8 is an example diagram of the degree of association table, Fig. 9 is a flowchart for setting the control target, Fig. 10 is an example diagram for presenting a control target setting range by radar chart, and Fig. 11 is a flow for determining a control method corresponding to the control target. FIG. 12 is a flowchart for generating a control program based on a control result, and FIGS. 13 to 15 are diagrams for explaining another embodiment of the present invention. 1 ... group management control device, 1-1 ... group management operation execution unit,
1-2 Data learning unit 1-3 Control program generation unit 1-4 Control target setting unit 1-5 Target setting range determination unit 2, Unit control device group 3, ... hall calling device, 4 ... input / output device.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kiyoshi Nakamura 4026 Kuji-cho, Hitachi City, Ibaraki Prefecture Inside Hitachi, Ltd.Hitachi Research Laboratory, Ltd. Inside the laboratory (72) Inventor Kenji Yoneda 1070 Mo, Katsuta-shi, Ibaraki Pref. Mito Plant, Hitachi, Ltd. (56) References JP-A-1-192682 (JP, A)

Claims (5)

(57) [Claims] 1. An elevator group management control device for controlling a plurality of elevators operating between multiple floors, comprising: a control target setting unit capable of setting target values of three or more control target items; In setting the target value in the setting unit, a control target setting region determining unit that determines a setting region of the target value for each control item, an operation method that satisfies a control target value set in the setting region, An elevator group management control device, comprising: a control program generation unit that selects and determines a control method; and an operation control unit that performs control using the determined operation method and control method. 2. The driving method according to claim 1, wherein the driving method is at least one of determining a service floor, allocating a vehicle preferentially to a specific floor, and changing a door opening / closing time. An elevator group management control device, characterized in that: 3. The control target setting area determining unit according to claim 1, wherein the control target item and the input control target item have a priority with respect to the preset control target item and a building specification, a traffic demand, or a time zone. The group management control device for an elevator, wherein the set area is determined using another degree of association table between control target values. 4. The control target setting area determination unit according to claim 3, wherein a total value is obtained for all of the input control target items using the relevance table, and each control target is set within the range of the total value. An elevator group management control device, wherein a value can be set. 5. The system according to claim 1, wherein the three or more control target items include an average hall waiting time, an average riding time,
Be sure to select at least two other items from among the congestion degree in the car, predictive accuracy rate, long wait occurrence rate, information transmission rate, energy saving, packed passage rate, car call first arrival rate, and transport capacity, including the waiting time. An elevator group management control device, characterized in that: