JPH02138080A - Group management control device for elevator - Google Patents

Abstract

PURPOSE:To respond to various requests from users by receiving inputs from the users with a multi-target inputting means, determining the allocation program with a control method determining means, and predicting the operation results with a simulating means. CONSTITUTION:A multi-target inputting means 1-1 and a control method determining means 1-2 are provided on an individuality support device 1 separately from a group management control device 4, and the ordinary elevator group management control is performed only by the group management control device 4. Inputs from users are received by the multi-target inputting means 1-1, the allocation program is determined by the sorting, extraction, correction and generation by the control method determining means 1-2, the operation results are predicted by a simulator 1-4, whether the results are satisfactory or not is inquired to the users via a display device 2, the target and weight are changed if the results are unsatisfactory, the processing is interactively repeated until the users are satisfied, and the group management control device 4 is set when satisfactory results are obtained. Various requests from the users can be responded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an elevator group management control device, and particularly to an elevator group management control device suitable for use in realizing the various demands of elevator users. This invention relates to a group management control device.

[Conventional technology]

Conventionally, in elevator group management, the control objective that serves as the standard for call assignment is the hall call duration from the time the call button is pressed until the elevator arrives at the landing, which corresponds to the waiting time in the hall. Efforts have been made to shorten this time for many years. As a conventional technique that considers items other than these as control targets, for example, techniques described in Japanese Patent Publication No. 1982-70, Japanese Patent Publication No. 62-71, etc. are known.

The conventional techniques described in these publications are designed to save power consumption of elevators and balance the hall call duration time so as not to degrade the service to waiting customers at the landing.

For this reason, this conventional technology adds not only the hall call duration time but also the predicted value of the dissipation Fcff1 force to the evaluation items when allocating a car to a hall call, thereby saving energy and improving service to customers waiting at the hall. We are trying to balance both
It is excellent in that it is possible to automatically save energy and shorten hall call duration time at the same time.

However, the conventional technology does not include other control target items such as boarding time, congestion level in the car, reservation change rate, etc., and also does not automatically balance energy saving and hall call duration time. It is not possible to reflect the opinions of users.

As a conventional technique that takes into consideration items other than those mentioned above as control items, for example, the technique described in Japanese Patent Publication No. 58-56709 is known.

This conventional technology adds fullness prediction to the control target items, but does not consider other control target items other than hall call duration time, and when balancing with hall call duration time. It is difficult to incorporate the wishes of users into the system.

[Problem to be solved by the invention]

The above-mentioned conventional technology is excellent in that it achieves a balance between the user's individual wishes and the hall call duration, which is the main control target item for group management. However, the above-mentioned prior art does not take into consideration the need to balance three or more conflicting control objectives and determine a control method that satisfies the user. Since it is not possible to set control target values for the control target items, there is a problem in that it is not possible to perform control that satisfies the user's overall wishes.

SUMMARY OF THE INVENTION An object of the present invention is to provide an elevator group management control device that solves the problems of the prior art and can meet various user requests.

[Means to solve the problem]

According to the present invention, the above object is achieved by providing a conventional group management control device with multi-target input means and control target determination means. The purpose is to accept input from the user using the multi-objective input means, determine an assigned program using the classification, extraction, modification, and generation means of the control method determining means, and predict the operation results using the simulation means. The predictive control results are presented to the user, and the user is asked whether or not the results are satisfactory. If the user is dissatisfied, go back to the beginning, change the goals and weights, and interactively repeat the above process until the user is satisfied. This is achieved by repeatedly setting the assignment program in the group management control device if a satisfactory result is obtained.

[Effect]

The multi-target input means and the control method determining means are provided separately from the 11 group management control means, and normal elevator group management control is performed only by the 1 group management control means. The multi-objective input means and the control method determining means are activated by a user's request and work off-line, apart from the operation of the group management control means, until the assignment program is generated. Therefore, the group management control means will not be overloaded, and the generated assignment program is predicted by simulation and is handed over to the group management means after obtaining the user's consent. No changes occur, and there is no problem with group management control.

According to the present invention, as described above, it is possible to provide a group management control device that can meet various user requests.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an elevator group management control system according to the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing the overall configuration of an embodiment of the present invention, FIG. 2 is a schematic flowchart explaining the operation, FIG. 3 is a diagram explaining the configuration of the control method determination knowledge base, and FIG. FIG. 5 is a flowchart illustrating target input processing, FIG. 5 is a flowchart illustrating control method determination processing, FIG. 6 is a flowchart illustrating comprehensive control method generation processing, and FIG. 7 is a flowchart illustrating simulation processing.

Figure 8 is a diagram explaining an example of classification knowledge, Figure 9 is a diagram showing an example of extraction knowledge, Figure 10 is a diagram showing an example of correction knowledge, and Figure 11 is an example of a prediction result display screen. FIG. In FIG. 1, 1 is an individualization support device, 1-1 is a multi-target input section, 1-2 is a control method determining section, and 1-3 is a control method determining knowledge base.

1-4 is a simulator, 1-5 is an environment, a traffic flow database, 2 is a display device, 3 is an input device, 4 is a group management control device, 4-1 is a group management control unit, 4-2 is a car, 4- 3 is a hall call button, and 4-4 is a control method database.

As shown in FIG. 1, a system according to an embodiment of the present invention includes a group management control device 4, which is a group management control means, and a multi-target generation means. The individualization support device 1, which is configured with an offline individualization support device 1, is equipped with a display device 2 and an input device 3, and is assigned a multi-item input section 1-11 for performing multi-target input. A control method determination unit 1-2 that generates a program, a control method determination knowledge base 1-3 used in the determination unit 1-2, and a simulator 1.
-4, an environment used in the simulator 1-4, and a traffic flow database 1-5.
-1, car number 4-2, hall call button 4-3, and a control method database 4-4 that stores the assignment program passed from the individualization support device 1.

In the present invention, normal group management control is as follows.

Performed only by Group I control system [4].

First, the general operation of the same-sex assimilation support device 1 according to the embodiment of the present invention will be explained with reference to the processing flow shown in FIG. 2.

(1) When the system is first started, this process is started in response to a request for goal setting from the user.

In the multi-target input routine of the multi-target input unit 1-1.

Input from the user is accepted (step 5100).

(2) Next, the control method is determined by the control method determination routine of the control method determination unit 1-2, and the operation result of the elevator is predicted by the simulation routine of the simulator 1-4 (steps 5200, 5400).

(3) Display the predictive control results obtained by the simulation routine of step 5400 and ask the user whether the results are satisfactory (steps 5OOI, 5
OO2).

(4) In step 5002, if the user indicates dissatisfaction, the process returns to step 5100, changes the control target 1 weight, etc., and performs the processing up to 5OOI again. This process is
This is repeated interactively until the user is satisfied.

(5) If the user indicates satisfaction in step 5Oo2, an assignment program that is a control method corresponding to the predictive control result obtained in step 5400 is set in the group management control device 4 and processing is performed. The process ends (step 5003).

Next, each routine of the processing of the embodiment of the present invention will be explained in order using actual numerical values. The embodiment of the present invention applies knowledge processing techniques to the program, but since the knowledge used here is extremely large, the following explanation will be limited to the extent that it does not interfere with the operation. ,
We will cover only the minimum amount of knowledge.

As shown in FIG. 3, the control method knowledge base is composed of three types of knowledge: classification knowledge, extraction knowledge, and correction knowledge. Each of these pieces of knowledge is hierarchized for each control target item to which the knowledge is related, and is described in the form of If-condition and Then-decision production rules.

The classification knowledge allows the control method determining unit 1-2 to classify the input plurality of control targets into a direct control target group to be directly controlled and an indirect control target group that is incidentally controlled by the result of control by this direct control target group. This is for classifying into control target groups.

Then, as knowledge for classifying control targets into direct control targets and indirect control targets, two targets a and a' of the multiple targets input by the user are classified according to their correlation. The following knowledge will be used:

(1) If there is a positive correlation between targets a and 81, the one with the greater weight or improvement effect is set as the direct control target, and the other is set as the indirect control target.

(2) If target a is improved, target a' will also be improved, but if target a is not improved even if target a is improved, target a is set as a direct control target, and target a' is set as an indirect control target. .

(3) When there is a negative correlation between targets a and a', the one with the larger weight or improvement effect is set as the direct control target,
With the other being an indirect control target, the trade-off between targets a and a' is performed by a modifying means.

The extraction knowledge is knowledge that holds control methods and control parameters for achieving one individual goal for the Raku goal items classified into the direct control goal group. In addition, the correction knowledge is for each control goal classified into the indirect control goal group.
This is the knowledge used to modify the control method and control parameters of a group of direct control targets that individually affect each other.

Furthermore, control target items that can be input by the user through the input device 3 include items related to efficiency improvement of the entire elevator system such as shortening average waiting time and reduction of long waiting rate, and items related to building management such as energy saving and low noise. , items to improve customer comfort, such as reducing congestion in the car and improving reservation changes, etc. The unit of measurement for these target items is 1 hour (second).

It may be something that is incommensurable and can be measured as a proportion (%), amount (person), amount per unit (person/minute), etc.

The operation of the embodiment of the present invention will be described below.

The building type is a building exclusively owned by one company, and the traffic demand is normal.

First, the contents of the processing in the multi-target input routine 5100 described above will be explained with reference to the flow shown in FIG.

(1) Display control target items on the display device 2. Here, we will take up the following eight items as control objectives.

1. Waiting time 2.Ride time 3. Car congestion level 4. Reservation change rate 5. Longevity rate 6. Amount of information guidance 7. Energy saving 8. Transportation capacity (Step S101).

(2) Let the user select a target item (step S1
02).

(3) If the user's selections in step 102 are: 1. Waiting time; 20. Riding time; 3. Car congestion degree, 4. Reservation change rate, 5. If there are 5 items for longevity rate, 1. Store the selected target number of items 5 in Tn (step 5103).
).

(4) Sequentially set the target values for each of the selected targets, and set the priority order and weighting coefficient of 0 for each target item (steps 104 to 107).

Through the above processing, a plurality of targets for group management control, their target values, priorities, and weights are determined.

Can be set. A setting example is shown below.

Target item Target value Rank Weighted waiting time 18 seconds or less 1 0.41 Riding time
35 seconds or less 2 0.27 Car congestion level
40% or more or 3% or less 4 0.12 Reservation change rate
5% or less 5 0.07 long waiting rate
3% or less 3 0.13 The processing of the multi-target input routine 5100 described above has already been proposed by the same applicant, such as Japanese Patent Application No. 63-47480 and Japanese Patent Application No. 63-10.
It is also possible to use the method described in No. 2544 and the like.

Next, the processing of the control method determination routine 5200 will be explained using the flow shown in FIG.

(1) First, target items are classified into direct control targets and indirect control targets using classification knowledge 2 201 as illustrated in FIG. 201 for the input goal and classification knowledge
By matching with the contents of , the target items are classified as follows.

Direct control goals: Waiting time (unconditional) Riding time (knowledge 3) Car congestion degree (unconditional) Indirect control goals: reservation change rate (knowledge 1, 2) longevity rate (knowledge m1) (step 5201).

(2) The number of target items classified as direct control targets in step 5201, 3, is set to Dn (step 520
2).

(3) Extract the target-specific evaluation formula φl based on the control method and control parameters for each direct control target using the extraction knowledge 204 illustrated in FIG.

The extraction result is as follows.

Waiting time: Min-Max (Knowledge 1) Boarding time
: Riding time Min-Max (Unconditional) Car congestion level:
Threshold control (knowledge 2) set value 40% (knowledge 4) (steps 5203 to 5205).

(4) Regarding the target classified as an indirect control target in step 5201, the correction knowledge illustrated in FIG.
7 to modify the previously extracted control method and control parameters. The correction result is as follows.

Reservation change rate: Car congestion setting value 40% (Knowledge 1.3-
2) Longevity rate: weight of duration 0.54 (knowledge 1a1) (steps 8206 to 5208).

(5) According to the results obtained above, a comprehensive allocation program is generated in the form of a one-weighted sum by a comprehensive control method generation routine (step 5300).

Next, details of the comprehensive control method generation routine 5300 will be explained with reference to the flow shown in FIG. In general, group management control methods include continuous methods such as the Min-Max method, which performs control based on continuous numerical values, and threshold control, which excludes one point from assignment candidates, such as the threshold control method. There are also types. In addition, in multi-goal control, since the units for each goal are different, evaluation of control methods for each goal (
0, 1) and take a weighted sum, and when the value is "1", control is performed to prohibit allocation. To this end, the comprehensive control method generation routine performs the following processing.

(1) First, the value of the sum ωatt of weights only for continuous type weights is calculated. In this example, ωa*t=0.54+o, 27:0.81 (steps 8301 to 304).

(2) Next, the continuous type numerical values are normalized based on the target value (in this case, twice the target value), and the weighting is the sum and
Find the sum of the threshold type and set the Min
Generates an assignment program (evaluation formula) Φ that takes , and returns. The evaluation formula Φ is expressed as follows.

However, Φ: assignment evaluation formula φ□: Waiting time evaluation formula φct: Riding time evaluation formula φcd: In-car congestion evaluation formula τt: target value ωI: Weight (Step 5305).

Next, the processing in the simulation routine 5400 will be explained using the flow shown in FIG.

(1) The comprehensive allocation program generated in the control method determination routine 5200 described above is transferred to the simulator 1-4.
, and other simulation conditions and the like are initialized (steps 5401 and 5402).

(2) Generating passengers according to the settings, performing group management processing, moving the aircraft, and collecting result data regarding each Meifuku are repeatedly executed for a predetermined period of time (Step 34
03-409).

(3) Statistically processes the result data collected in step 407 and ends the simulator processing (step 541
0).

(4) Finally, step 5, already explained in FIG.
Return to 001 and display the results. The displayed results are
As shown in FIG. 11, the target value and the predicted value based on the generated allocation program are presented. As already explained, with the user's consent, this assignment program is transferred to the control method database 4-4 using an IC card or the like, and the process is completed.

According to the embodiment of the present invention described above, a plurality of control targets are
By classifying targets into direct control targets and indirect control targets, the optimal control method can be adopted for a single direct control target, and targets that are difficult to control directly can also be incorporated as indirect control targets. becomes. According to one embodiment of the present invention, for generating the allocation program,
Since weights based on the user's wishes are used and prediction of results through simulation is incorporated, it is possible to provide a group management control device that the user can understand.

〔Effect of the invention〕

As explained above, according to the present invention, by using means for classifying multiple targets into direct targets and indirect targets, it is possible to use the optimal control method for each direct target, This has the effect that targets that have been difficult to control can now be controlled by incorporating them as indirect targets. Furthermore, according to the present invention, by performing the generation of the one-allocation program off-line independently of the group management control device, it is possible to perform group management control that requires real-time control and reassignment programs that use knowledge processing methods that cannot be executed in real time. This has the effect of being able to achieve both the generation of Further, according to the present invention, the elevator group management control device is capable of responding to various user requests, including performing simulation-based prediction and user approval steps on the generated allocation program. can be provided.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the overall structure of an embodiment of the present invention, FIG. 2 is a schematic flowchart explaining the operation, FIG. 3 is a diagram explaining the configuration of the control method determination knowledge base, and FIG. Flowchart 1 for explaining multi-target input processing, FIG. 5 is a flowchart for explaining control method determination processing, and FIG. 6 is a flowchart for explaining comprehensive control method generation processing.
Figure 7 is a flowchart explaining simulation processing, Figure 8 is a diagram explaining an example of classification knowledge, Figure 9 is a diagram showing an example of extraction knowledge, and Figure 10 is a diagram showing an example of correction knowledge. , FIG. 11 is a diagram illustrating an example of a prediction result display screen. 1...Individuation support device, 1-1...Multi-target input unit,
1-2...Control method determination unit, 1-3...Control method determination knowledge base, 1-4...Simulator, 1-5...
・Environment, traffic flow database, 2...display device, 3.
...Input device, 4...Group management control device, 4-1...
Group management control section, 4-2... No. machine, 4-3... Hall call button, 4-4... Control method database. Bird 5 diagram φ Allotment 7 tsubo 1 [Tl 1 formula % formula % Deyo Target 11 Yu'V mouth spring 90 τ Yuji, Tagusa high school 8 ward Zui 10 diagram

Claims (1)

[Claims] 1. In an elevator group management control device that allocates a plurality of elevators in service between multiple floors to a newly generated hall call according to a plurality of control objectives for group management control, multi-objective input means for inputting a plurality of control objectives, their target values, and priorities; a control method determining means for determining an allocation program for allocating elevators in accordance with the input plurality of control objectives; and a determined allocation. A group management control device for elevators, comprising a group management control means for allocating elevators according to a program. 2. The elevator group management control device according to claim 1, wherein the control method determining means operates off-line separately from the group management control means. 3. The control method determining means is a classification means for classifying the set multiple targets into a direct control target group to be directly controlled and an indirect control target group to be controlled incidentally based on the control result of the direct control target group. and a target control method extracting means for extracting control methods and control parameters for achieving the goals individually for each target classified into the direct control target group; a modifying means for modifying the control method and control parameters of the direct control target group that affect
3. The elevator group management control device according to claim 1, further comprising control method generating means for generating a comprehensive allocation program. 4. The plurality of control objectives are items related to efficiency improvement of the elevator system as a whole such as shortening average waiting time and reduction of long waiting rate, items related to building management such as energy saving and low noise, and degree of congestion inside the car. Claims 1 and 2 consist of a group of goals, such as items for improving customer comfort, such as reducing reservation change rate and improving reservation change rate.
The elevator group management control device according to item 1 or 3. 5. Incommensurable goals in which the plurality of control target items are measured in time (seconds), proportion (%), amount (person), amount per unit (person/minute), etc. An elevator group management control device according to any one of claims 1 to 4, characterized in that the elevator group management control device comprises a group. 6. The classification means classifies the control targets into direct control targets and indirect control targets based on the correlation in goal achievement between two targets a and a' among the plurality of control targets input by the user. The knowledge is that when there is a positive correlation between goals a and a', the one with the larger weight or improvement effect is set as the direct control goal, and the other is set as the indirect control goal.If goal a is improved, goal a' will also be improved. However, if target a is not improved even if target a' is improved, target a is set as a direct control target, and target a' is set as an indirect control target.When there is a negative correlation between targets a and a', , the one with a larger weight or improvement effect is set as a direct control target, the other is set as an indirect control target, and a trade-off between goals a and a' is performed by a correction means. The elevator group management control device according to any one of items 3 to 5. 7. The classification means, target-specific control method extraction means, modification means, and control method generation means are composed of software such as a computer program based on expert knowledge of elevator group management, and are capable of classifying, extracting, modifying, A patent claim characterized in that the target value, weight, priority of each item of a plurality of control targets, the type of building in which the elevator is installed, traffic flow, etc. are used as data used for judgment for generation. The elevator group management control device according to one of the items 3 to 6. 8. The elevator group management control device according to claim 1, wherein the direct control target group includes at least one target.