JPH07144842A - Method of starting a plurality of elevator car in building - Google Patents

Abstract

PURPOSE: To accurately fuzzy-indicate the number of passengers to the measured weight by obtaining a passenger counting fuzzy set on the basis of a basic fuzzy set on the weight of a single passenger and a composite fuzzy set on a plurality of passengers to start a plurality of elevator cars. CONSTITUTION: In the case of starting a plurality of elevator cars in a building, signals indicating a basic fuzzy set including a basic single passenger weight element and a corresponding membership value, and a factor fuzzy set including a basis element indicating additional weight, are outputted. A plurality of signals indicating a plurality of composite fuzzy sets on a plurality of passengers are then outputted. A signal indicating the weight of passengers is outputted, and a plurality of signals indicating passenger counting fuzzy sets corresponding to one related term of the composite fuzzy set with the basic element equal to the weight are outputted. Elevators are then started according to the passenger fuzzy set signals.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for evaluating the number of passengers in an elevator car on the basis of a weight that takes into account the obtainable changes in the passenger weight.

[0002]

2. Description of the Related Art In a method of leaving an elevator car, the number of passengers in the elevator car is often adopted as a main factor in order to provide the most effective service to passengers. A typical example is how to determine if a large number of passengers are leaving the lobby floor during the morning rush hour. In that case, the up-peak mode will be started for departure. Similarly, for a down-passing passenger arriving at the lobby floor during the evening rush hour, the departure down-peak mode is sought. S entitled "How to Use Fuzzy Logic to Determine Transportation Modes for Elevator Systems"
U.S. patent application Ser. No. 07 / 879,558, filed May 4, 1992 by irag and Weisser.
An example is given in the issue. In that application, passenger numbers are described herein, and "How to use fuzzy logic to determine the number of passengers in an elevator car."
Determined by using fuzzy logic in a manner also described in US patent application Ser. No. 07 / 879,558, filed May 4, 1992 by Sirag.

The method is (for example, 2, 8, 28
A large amount of data is used to indicate the change in weight that different people (such as people) would weigh in a given elevator system. These data are obtained empirically by simply observing the number of passengers in the elevator car and recording the number of passengers along with the weight indicated by the elevator load weighing system at that time. After collecting a large amount of data, different weights for passenger counts are combined together to form a fuzzy set. In that fuzzy set, each primitive is a perceptible weight to the elevator, and the membership of each primitive is
The number of times such weight is observed (normalized) is dependent. A fuzzy set is then established which is a membership of that weight primitive in each of the fuzzy sets associated with all possible passenger numbers for that given weight. Therefore, the slices (partial) through these fuzzy sets representing the weight versus the number of passengers result in a single fuzzy set of passengers for a given weight. There, each primitive is a number of passengers, and the membership of that primitive is the relative likelihood that that number of passengers will result in a given weight of the set.

[0004]

There are other methods of determining the number of people in an elevator car, especially the use of discrete people counters such as infrared and video techniques, most of which are known. However, they are extremely expensive and do not have the absolute accuracy to give an absolutely accurate count of the number of people entering or leaving an elevator car. In the methodology of applying Sirag described above,
If appropriate information is empirically available, it must give a very accurate prediction of the average number of people at a given weight.
Discrete people counters, on the other hand, give more actual data than average data. However, they have also been found to be inaccurate (eg missing some people, counting luggage as passengers).

In the aforesaid application, the fact that the weight difference between male and female passengers has a significant effect on the accuracy of the fuzzy set, which represents the average number of passengers, is due to the fact that male passengers cause a given elevator car load. A fuzzy set of numbers and a fuzzy set of female passengers causing a given elevator car load are given separately, and the fuzzy sets are each proportioned by the proportional distribution of the male and female population within the building. It was adopted by The combination of these two fuzzy sets more accurately determines the approximate mixed number of male and female passengers in the building than the fuzzy sets of average passenger numbers used in the prior art.

It is an object of the present invention to improve the accuracy and with its accuracy the use of a fuzzy set relating to the number of passengers and the weight, to fuzzy indication of the number of passengers in an elevator for a given measured weight. Especially.

Other objects, features and advantages of the present invention will become more apparent by the detailed embodiments shown in the drawings.

[0008]

SUMMARY OF THE INVENTION The present invention provides that the approximate weight of an individual passenger is the passenger's general national heredity (ie, birthplace), the amount of clothing the passenger has, and the weight of the luggage carried by the passenger. It stands on the recognition of the fact that it changes as a function of.

In accordance with the invention, in a method of leaving an elevator car in a building, a passenger count fuzzy set showing the approximate number of passengers resulting in a given elevator car load is the difference between the average weights of people in different countries, It was factored into the usual amount of clothes worn in a given temperature climate and its factors indicating the transport of luggage in and out of the building during various hours of the day. According to one aspect of the invention, the aforementioned changes are accommodated by adjusting a given passenger counting fuzzy set for a given elevator car load. In accordance with another aspect of the present invention, the above-described changes include an average male passenger fuzzy set and an average female passenger fuzzy, prior to accepting according to the relative mix of gender in population within the building, as described in the aforementioned application. It is applied separately to the set.

[0010]

In one method of practicing the present invention, there is a fuzzy set indicative of the approximate number of passengers for a given elevator car load, determined empirically or demographically for each passenger count. Prior to drawing from, multiple fuzzy sets of weights have the aforementioned changes applied to them as weights. According to another implementation of the invention, a fuzzy set indicating either the weight of a given number of passengers or a given elevator car load occurs when the weight difference (so-called each passenger is at a higher weight, Each passenger has a basic element modified in a way to reflect a greater or lesser respective passenger weight).

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention provides a method for varying the number of individual passengers given a minimum international weight for a country (i.e., in the possible elevator population) whose populations have the minimum unit weight of so-called people on earth. Basics that represent weight, etc.
Using single-passenger fuzzy sets, for countries where people have heavier weights, for seasonal clothing in different temperature climates, and for packages carried in either the up or down direction at a given time of day, A plurality of fuzzy sets indicating the increased weight are added to the above basic fuzzy set.

In FIG. 5, the dotted line indicates a single passenger (N =
1) 5 passengers (N = 5) and 10 passengers (N = 1)
It shows a fuzzy set for 0). In the example of FIG. 5, the basic fuzzy set for a single passenger is. 00│
10 ,. 05│20 ,. 50│40 ,. 80│50,
1.0│70 ,. 70│100 ,. 50│110 ,. Two
0│120 ,. It is expressed as 001│130. ． 0
0│100 ,. 05│200 ,. A fuzzy set of 10 passengers is also identical, except that each basic element weighs ten times, such as 50│400. Therefore, the curve shape is the same for any number of passengers, with the curve for 10 passengers being best shown. Figure 5
Is given in kilograms, but all weights (basic elements) can be expressed as pounds or as a percentage of total load. The basic fuzzy set for a single passenger can be determined empirically by keeping a record of the number of passengers in each elevator traveling from floor to floor and the weight of passengers in that group. This method requires a large number of observations to give meaningful data. Moreover, the method is typically performed by making observations on a given type of elevator in a similar building environment. Fuzzy sets of single passengers, on the other hand, can be determined from the demographics of the weight of inhabitants in various regions of the world as elevator passengers within the promise. Further, the example presented here begins with a single fuzzy set for a single passenger. However, as mentioned in the aforementioned application, the single passenger fuzzy set is a composite function of male and female fuzzy sets based on the sex demographics of individual buildings. Thus, in providing the original single-passenger fuzzy set, the invention of the aforementioned application takes into account the demographics of men and women in the building. Of course, using the previous invention can make the overall result more accurate,
The selection as described above is not relevant to the present invention.

Single passenger fuzzy set (FIG. 5, N = 1)
Are, for example, people from Vietnam, Thailand, or other South Pacific coast countries. It is chosen to be a fuzzy set for the country on earth, where the lightest people in the elevator countries live. In FIG.
The added weights found in Japan range from 0 membership values for 0 added kilograms to 1 membership values for 1-4 kilogram primitives, with fuzzy sets returning to 0 memberships at 5 kilograms. Represented by On the other hand, the U.S. added weight ranges from a membership value of 0 for 12 kilograms to a membership value of 1 for 16 kilograms to 22 kilograms and returns to a membership value of 0 at 28 kilograms. This reflects the possibility that, in the United States, the average body weight is 20 additional kilograms more than the weight of people in Japan.
Similarly, fuzzy sets for the UK range from 0 membership for a 4 kilogram weight add-on to 1 membership for a 6-13 kilogram weight add-on, with an average of essentially 10 weight add-ons thereafter. It reflects the decrease in kilograms. Gradually graded to 0 membership for 19 kilograms of added weight. Similarly, demographics are used to extend the fuzzy set of added weights to other relevant countries. Of course, when deciding on demographics, not only should the indigenous people of the country be included, but also the demographics of publicly used buildings, regardless of where each passenger was born. In the United States, public buildings include many lightweight ethnic groups, including Asians.

In FIG. 2, the winter fuzzy set represents the weight gain for maximum winter clothing in various temperature climates such as Europe, the United States and Japan. The winter fuzzy set ranges from a membership value of 0 for 2 kilograms of added weight to a membership value of 1 for 3-5 kilograms of added weight,
Reduce to a membership value of 0 for 6 kilograms of added weight. The fuzzy sets for spring and autumn range from a membership value of 0 for a basic element of 0 additional weight to a membership value of 1 for 1-3 kilograms of additional weight, reflecting intermediate additional clothing. , Inclined to a membership value of 0 for 4 kilograms of added weight. The fuzzy set in the summer is calculated from the membership value of 1 for 0 kilogram of added weight
It is in the range of up to 0 membership value for 2 kilograms of added weight, reflecting the fact that it is lightweight even with clothes on.

FIGS. 3 and 4 represent preferred fuzzy sets for different times of the day for both the up and down directions, reflecting the possibility of passengers carrying parcels in and out of the building. In the up direction, parcel shipments are most likely during early morning rush hours, run low late in the morning, then slightly resume at noon, 2 pm
After time, it inclines to 0. In the down direction, it reflects the great potential for transporting the parcel downwards during the mid lunch time. Also, in the early hours of the evening there is a reduced chance around daytime and next to the minimum chance during the afternoon. The parcel fuzzy set shown in FIGS. 3 and 4 is made from statically priced data from samples in a suitable building, or in any other suitable manner. The factors of Figures 2-4 are additionally adjusted for different parts of the world. That is, the actual values of the fuzzy sets of FIGS. 2-4 are determined differently from other countries if it is determined that a given country does not meet international standards.

In the present invention, the basic single passenger fuzzy set (N = N) shown in FIG. 5 is generated so that the elements shown in FIGS. 2-4 generate a single passenger composite fuzzy set.
1) is added. An example of a US national factor is shown for 10 passengers by a dotted line. When the factor of 10 passengers, as shown by the dotted line, is fuzzy added to the basic passenger fuzzy set of 10 passengers shown by the dotted line N = 10, the result is the composite 10 represented by the solid line for 10 passengers. It is a passenger fuzzy set. Due to the fact that the algebraic property of multiplying and adding terms to a fuzzy set is commutative, associative and distributive, the factor fuzzy set is multiplied by the number of passengers, and the passenger set also depends on the number of passengers. It is multiplied.
Then the product of the two is added. Alternatively, a single passenger factor fuzzy set is added to the basic single passenger fuzzy set and then the number of passengers is multiplied to give a composite fuzzy set for a given number of passengers. The latter is chosen in the preferred embodiment of the present application to save processing.

Next, referring to FIG. FIG. 6 shows a logic flow diagram for determining and recording vehicle loads and minimum weights. From there, the number of passengers coming in and out and up to the next stop is determined, and via input point 20 proceed to test 21. The first test 21 determines if the door is fully closed (DFC). If the door is not closed, this means the elevator is landing,
Indicate that passengers are getting on and off. When the negative result of the test 21 is reached, the process proceeds to the test 21, and the “minimum initialization (M
IN. It is determined whether the "INIT)" flag has been set. This is the minimum weight of an elevator (MIN to indicate passengers whose weight comparison has begun and is not leaving the elevator).
WGT) is a local flag used to keep track of what is being asked for. When the door first begins to open, the first pass through test 21 reaches a negative result of test 22 and a pair of steps 24 causes the current actual weight of the elevator (before the passengers get off) (ACTL).
Initialize the minimum weight in WGT (SET MIN
INIT) and set the minimum initialization flag for use in test 22. Test 25 then tests the actual weight (determined by the elevator load sensor) (ACT
L WGT) is equal to or less than the currently recorded minimum weight (MIN INIT) (initialized in step 24). Test 25
In the first pass through, it was initialized in step 24 to equal the actual weight, so the result is negative. Therefore, the minimum weight remains set.

On the next pass through the routine, the negative result of test 21 and the positive result of test 22 result in:
Go to test 25 again. Then, when some passengers leave the elevator, the actual weight (ACTL WGT) is reduced, less than the minimum weight (MIN WGT).
This yields a positive result for test 25, step 2
6 is reached, where the minimum weight is adjusted to the actual weight at the present time and the "new minimum weight (NEW MIN WG
The "T)" flag is set (for use as described with respect to FIG. 8 described below).

On the other hand, if the door is fully closed, then step 21 has a positive result, bypassing steps or tests 22-26. In either case, test 29 shows that the elevator is in the direction of travel (DIR).
Determine whether or not to have. If the elevator is stopped for landing and the door is open or open, close the door,
The elevator has no direction until it is ready to move. Therefore, the presence of the direction indicates that the door is closing rapidly for the next run. A positive result of test 29 advances to test 30 to determine if the door is fully closed (DFC). Initially, as soon as the direction was assigned, the door tried to close, and thus did not close enough, resulting in a negative test 30,
Proceed to test 31 and the door is closing (as opposed to opening to indicate that the car is approaching landing) (DR C
See if it is LSNG). When the car stopped with the door fully open and passengers got on and off, there was a command and the door was not fully closed, but the door was about to close,
With the positive result of the test 31, the process proceeds to the test 32, and the closing flag is set (SET CLSNG). The closure flag is set when the door eventually becomes closed following a stop on the floor. If the test 30 is a positive result, the process goes to the test 33 to check whether the closing flag (CLSNG FLG) is set. If not set, this indicates that the car is simply going up or down with the door fully closed. And
It does not need to perform any function. On the other hand, if the close flag was set, this indicates that the door was just closed after a stop on the floor. Therefore,
The positive result of the test 33 proceeds to a series of steps 34.
There, the load (indicating the weight of the passenger traveling to the next floor) (LD WGT) is the actual load of the elevator (AC
TL WGT). This is done just before the elevator is moved. Therefore, the increase in passengers does not affect the measured weight signal. Step 34 also sets a "new load" flag to indicate that a new weight has been measured so that a new determination of the number of passengers in progress can be made (SET NEW LD WG.
T). In preparation for the next stop, the closing flag is reset (RST CLSNG FLG). Further, in preparation for the next stop, the minimum initialization flag is reset (R
ST MIN INIT). Then, through return point 35, the rest of the program is restored.

In FIG. 7, an update routine updates the fuzzy set used to determine the number of passengers remaining from the minimum weight signal and the number of loaded passengers traveling from the load signal to the next floor. As described above in connection with FIGS. 1-5, the fuzzy set used is a synthetic fuzzy set containing a possible sum of at least three factors, which is an approximate average of the passengers in the building. It includes a country factor that indicates weight, a seasonal factor that indicates the weight of additional clothing in different temperature climates, and a time factor that indicates the likelihood of a carrying passenger when traveling in either the up or down direction. The update routine of FIG. 7 is entered via input point 40 to form a composite fuzzy set. Whether the elevator was newly assigned to a country by the first test 41 (NEW CNT
RY) otherwise determines if the assigned fuzzy set among the fuzzy sets shown in FIG. 1 has been modified. Otherwise, test 42 determines whether the season has changed from summer to autumn, autumn to winter, winter to spring, or spring to summer (NEW SESN).
When any of these situations occur, the country and season fuzzy sets become one of the working fuzzy sets F to be added, which becomes the final fuzzy set as seen in FIG.
(W) is set equal (from FIG. 1) to the selected fuzzy set for that country. Also, another operating fuzzy set F
(W) is set equal to the seasonal fuzzy set selected from FIG. The two fuzzy sets are then added together with the fuzzy addition rules, as described below in connection with FIG. And as a result, F (W)
Are stored as F (TOTL) for later use in a series of steps 45. Whenever a new sum of these two fuzzy sets is prepared, a new synthetic fuzzy set needs to be prepared, and therefore step 45 requires the "newup" and "newdown" flags for the purposes described below. To set.

For the up direction (FIG. 2), no change occurs between 2:00 pm and 7:00 am. Therefore, for the up direction, there is no need to change the composite fuzzy set during that time as a result of that time. Similarly, for the down direction (FIG. 3), there is no change in the time factor between 6 pm and 10 am. Therefore, it is not necessary to create a new synthetic fuzzy set. However, for the up direction, new-up signals are generated at 7 am, 9 am, 10 am, 12 pm, 1 pm and 2 pm. In addition, a new down signal is generated for the down direction at 11:00 am, 12:00 pm, 1 pm, 2 pm, 3 pm, 4 pm, 5 pm and 6 pm. Thus, the new-up flag is set when the fuzzy set changes in the up direction, and independently, the new-down flag is set whenever the fuzzy set for the down direction changes. As mentioned above, whenever a country or seasonal fuzzy set is changed, to form a new composite set,
Both flags are set.

Test 48 is up direction (DIR
= UP). If so, proceed to test 49 to determine if the NEW UP flag is set. If so, a positive result is reached and the operation proceeds to step 50, where the first working fuzzy set F (W) is the country and seasonal fuzzy set F (TO
TL) is added. Then, the second working fuzzy set F (w) is set equal to the current fuzzy set F (TIM, UP) in the up direction. This means that any one of the fuzzy sets in Figure 3 is related to the current time. The two are then added together in routine 51, as described below in connection with FIG. The upstream fuzzy set F (UP) (FIG. 3) to be used during the remainder of the current time increment is
The result is set equal to F (W). And
In step 52, the new up flag is reset. Then, via return point 53, the rest of the program is executed.

If the direction is up, a positive result at test 48 leads to test 49. However, if the new up flag has been reset (at step 52), then test 49 is negative and proceeds to test 56 to determine if the new down (NEW DN) flag is set. The reason for considering the direction in test 48 is simply to make the most recent synthetic fuzzy set needed for the direction of travel to which the elevator is currently assigned, and later the other if necessary.

On the other hand, if the result of test 48 is negative,
A test 57 is run to determine if the newdown flag is set. But test 56 or 57
If any of the above are affirmative, then a series of steps 60 are performed to set the first working fuzzy set F (W) equal to the addition result of the country and seasonal fuzzy set F (TOTL). Then, another working fuzzy set F (W) is a fuzzy set F (TIM, D) for the current time in the down direction.
N). Thereafter, the routine 61 is entered and fuzzy addition is performed as described below in connection with FIG. A pair of steps 62 saves the result as a fuzzy set F (DN) for the down direction and resets the newdown flag. It is then returned to the rest of the program via return point 53. The real effect of those steps and tests 48-62 is that if there is one new fuzzy set for the up and down directions whenever the time for each country, season, or up or down direction changes. The first is to simply generate for the current direction of movement.

While the determination of the minimum weight and load in the routine of FIG. 6 is done on a time basis which depends on the operation of the elevator, the occurrence of the new fuzzy set is (in the maximum number of cases) totally different including time. Is based. The use of the determined minimum load and upper and lower fuzzy sets is also carried out with different criteria. Typically this is the case as the elevator progresses from one floor to the next. It determined the minimum load before proceeding between floors. This may be done at a time selected by the programmer in connection with other functions to be performed in activating the elevator and controlling its travel profile, doors, call buttons and lighting and the like.

In FIG. 8, membership values for various numbers of passenger primitives in the fuzzy set associated with a given weight (minimum weight or load) are obtained via input point 70. The first test 71 determines whether a new minimum weight flag (FIG. 6) has been set. If so, a pair of steps 72 set the weight running value (WGT) equal to the last determined minimum value. As a result, a fuzzy set of passenger numbers associated with that minimum weight may be generated. Then, the new minimum weight flag is reset. The fuzzy set is then selected by a test 73 that determines if the direction is up.
If so, the fuzzy set F (W) in operation
Is set equal to the upstream fuzzy set F (UP) in step 74. If the direction is not up, then in step 75 the fuzzy set is set equal to the down fuzzy set F (DN). On the other hand, if the new minimum weight test 71 is negative, test 76 determines whether a new load flag is set. If so, in a pair of steps 77, the working weight factor (WGT) is set equal to the load and the new load flag is reset. However, if neither the new minimum weight flag nor the new load flag is set, the routine of FIG.
Return to the rest of the program via 8. In other words, once the weight fuzzy set has been created (as described in connection with the rest of FIGS. 8 and 9 below), the diagram, as indicated by resetting the new weight flag, Routine 8 includes tests 71 and 7
It is simply bypassed by a negative result of 6 and it is point 78
Let go back through.

After selecting the weight (MIN, LD) and the synthetic fuzzy set (F (UP), F (DN)) to be used, the routine of FIG. 8 proceeds to step 81, where the value N indicating the number of passengers is indicated. Is set equal to 1. Then, to initialize the process, set the number W used for steps through the fuzzy set equal to one. Step 83
Is the initial element [b (N, W)] of the Wth term for one passenger (initially) is the Wth term of the single passenger composite fuzzy set (selected in either step 74 or 75). It is set equal to 1 times the basic element [N * b (W)] of. Then test 84 determines if the primitive is for the current term, ie, step 8
It is tested whether it is the Wth term multiplied in 3. Initially, this is the Wth term for a single passenger that is taken directly from the fuzzy set selected in one of steps 74,75. Test 84
Determine whether it is greater than or equal to the weight (selected in either step 72, 77 as either minimum weight or load). Initially, and typically, for a few cycles it is not. Therefore, a negative result proceeds to a series of steps 85 where the factor Δ1 is set equal to the difference between its weight (WGT) and the primitive [b (N, W)], and the membership value m1 is ( For the applications described in
W is also incremented (INCR W) set equal to the membership of the Wth term of the N passenger fuzzy set.

Typically, as in the 600 kilogram example of FIG. 5, the actual weight tested (whether minimum weight or load) is (although it may be different).
It becomes larger than the basic element of the first term, and it becomes a negative result,
Proceed to step 85 to increment W. Then, a test 87 determines whether W = 12 is still present.
To illustrate that the present invention can be effectively implemented in a basic single passenger fuzzy set for very small numbers, such as twelve, the number twelve is chosen in the preferred embodiment herein. . The number may be even smaller (N = 10, FIG. 5 has 9 terms), and is more so if the included data processor has sufficient time and output to process more terms of the fuzzy set. Can be made bigger. In one embodiment, term 300 is used. Initially, the result of test 87 is negative because only one term was processed. This returns the routine to step 83. In step 83, the second primitive of the fuzzy set is multiplied by N (in this case N is still equal to 1). Again, test 84 is negative. Therefore, the process proceeds to step 85. This process is performed until the final weight processed (either minimum weight or load) is greater than the basic element of the Wth term (in which case a series of steps 88 is reached), or , W is 12
(In that case, the loop is exited by the positive result of test 87). Suppose for N = 1 the weight does not exceed any term in the single passenger's chosen fuzzy set. This is certainly true for the case shown in FIG. In FIG. 5, the weight being processed is 600 kilograms as shown. 1
When all primitives for the passenger composite fuzzy set have been tested negatively, a positive result of test 87 leads to step 89, where N is incremented to 2 passengers (INCR
N) Allow. Test 90 determines that 32 passengers (the maximum possible number of elevators) have been processed, and a negative result advances to step 82. There, W is stored again at 1 and the process is repeated for the two passenger composite fuzzy set. In fact, there are no primitives in any of the fuzzy sets for N = 1, 2 or 3 passengers (not shown in FIG. 5). But,
When the single passenger fuzzy set is multiplied by 4 in the example of FIG. 5, the twelfth term (W = 12) is actually
Having a weight greater than the preferred weight of kilograms. Therefore, a positive result of test 84 results in a series of steps 8
Proceed to 8, where the other factor Δ2 is set equal to the difference between the basic elements of the 12th term and the weight to be processed and the membership value m2 is set equal to the membership of the 12th term. The weight fuzzy set terms for a particular weight being processed (600 kilograms in the example of FIG. 5) and having a four-passenger primitive is, in a known manner, the addition of Δ1 and Δ2. By using the membership value of the 11th term, which is determined as the ratio of Δ1, the 12th and 11th terms in this example are used.
Given a membership value M equal to a linear interpolation between the membership values of the th term. At this point, the first term of the 600 kilogram fuzzy set is established with the slight possibility (probably 0.055% in the example of FIG. 5) of four passengers causing a weight of 600 kilograms. The terms for the four passengers were determined in step 88, so no further processing of the fuzzy set of four passengers is required. Therefore, step 88 is followed by step 89 where N is incremented to 5 passengers. Test 90
Determines whether the fuzzy set for 32 passengers has been processed. At this point, if they have not been done, a negative result of test 90 advances to step 82 where W = 1 is set again and processing for five passengers begins.

A positive result in test 84 means that the weight is either in the last element of the last term or between the last two elements. However, as will be more clearly understood in connection with the five-passenger composite set below, if no term is greater than the weight being processed, then it is of two terms depending on that weight. Since it is the lower term of these, the characteristics of this term are Δ1 and m1.
Test 84 following step 85 to set
Protected by the negative result of. In the case of the five-passenger composite set (N = 5 in FIG. 5), it is clear that there are some primitives below 600 kilograms being processed before there is a larger primitive. is there. Thus, a continuous pass through steps and tests 82-87 is, for example, W = 1, where the primitives of the sixth term for a five-passenger composite fuzzy set have been processed.
Step 8 until weight is greater than 00 kilograms
At 5 consecutively update Δ1 and m1. This yields a positive result of test 84 and proceeds to step 88. There, Δ2 is taken to be the difference between the weight of the 6th term primitive and weight. Also, m2 is taken to be equal to the sixth membership for the five-passenger composite fuzzy set. That is, in the example of FIG. 8 or so. The last one in step 92 is linear interpolation to give membership values for the five-passenger term of the 600 kilogram fuzzy set. Thereafter, step 89 increments N and proceeds to the 6-passenger fuzzy set, since it is not necessary to examine the 5-passenger fuzzy set again (the term was established in step 88). Test 90 determines if the fuzzy set for up to 32 passengers has been processed. If not, finally the positive result of test 90
The additional fuzzy sets are processed for 6-, 7-, and 8-passengers, etc. until they are advanced to the rest of the program via return point 91.

Next, referring to FIG. Via input point 100, we enter an exemplary routine that adds two fuzzy sets according to the rules of fuzzy arithmetic. First step 1
01 is the factor W used to record the continuous term of one of the two fuzzy sets being added.
Is set to 1. Then, step 102 sets the factor W used to record the other term of the fuzzy set to be added to one. As is known, the addition of two fuzzy sets according to the rules of fuzzy arithmetic adds the basic elements of each term of one fuzzy set to the basic elements of all terms of the other fuzzy set, and its The sum of the two, so added
Combining into a membership value consisting of the smallest membership value in one term. Therefore,
All terms with the same primitive are simply combined by keeping the primitive with the maximum membership value among all similar terms in the result. Thus, the first step of the addition process 103 is the sum fuzzy set b (W, b) equal to the sum of the basic elements of the first term (in the beginning) of both fuzzy sets b (W), b (w).
w) is produced. Then, the test 104 shows that the membership value m (W) of the first term of one of the fuzzy sets F (W) is similar to the fuzzy set F (W) of the other fuzzy set F (W).
(W) Determine if it is less than or equal to If so, the membership value for that sum of terms is taken to be the membership value m (W) in step 105. Otherwise, the negative result of test 104 is the first in the total.
The membership value of the term is m in step 106.
(W). The second term of F (W) is then added to the first term of F (W) as step 107 increments w. Then, the test 108 determines that w is still the total number of terms in the fuzzy set, 12 in this example. Initially, though not, the negative result of test 108 returns the routine to step 103, where the primitive of the first term of one of the fuzzy sets is added to the primitive of the second term of the other fuzzy set. . Then, tests and steps 104-106 determine the minimum membership value between these two terms. Then w is incremented in step 107 and tested in step 108 to see if all 12 terms of the second fuzzy set have been added to the first term of the first fuzzy set. After all, it will be. Therefore, a positive result of the test 108 advances to step 112. Where W
Are incremented to point to the second term of the first fuzzy set.
Test 113 determines whether all terms in the second fuzzy set have been added to the terms in the first fuzzy set. Initially, it is not, so the negative result of test 113 returns the routine to step 102. There, because of addition with the second term of the first fuzzy set F (w), the first term of the second fuzzy set is pointed out as follows:
w is reinitialized to 1. After all terms in each set have been added to each of the other set's terms, the positive result of test 113 advances a pair of steps 115. There, both working fuzzy sets F (W) and F (w) are set equal to the total F (W, w) just obtained.
Then both W and w are initialized to 1 in a pair of steps 116 and 117, and the terms are tested to see which terms are similar. This begins by testing one first term b (W) of the working fuzzy set with the first term b (w) of the other working fuzzy set. Of course, in this case they are identical.
Therefore, the positive result of test 121 results in test 1
Proceeding to 22, it is determined which terms have a large membership value. In this case, they are
Equal, and therefore the positive result of test 122
Bypass step 123. There, the terms of the other fuzzy set are substituted. Therefore, in this protocol, the final result of the addition of two fuzzy sets, where similar terms are combined with the maximum membership value, is the first of the working fuzzy sets F (W). After the first terms are compared together, W is incremented in step 124 to determine whether all terms in the second working fuzzy set F (w) have been compared to the first term in the first fuzzy set F (W). Tested to see. Initially, it isn't. Therefore, the negative result of test 125
The routine returns to test 121 and compares the first term b (W) of the first working fuzzy set with the second term b (w) of the other working fuzzy set. In this process, the first term of the first operating fuzzy set F (W) is the other operating fuzzy set F.
Continue until all 12 terms in (w) have been compared. At this point, a positive result of test 125 results in step 130.
And W is incremented to point to the second term of the first working fuzzy set F (W). Therefore, it will be compared with all terms of the other working fuzzy set F (w). Then test 131 determines if all terms in the second fuzzy set have been compared. beginning,
Since they are not, the negative result of test 131 returns the routine to step 117, where the second
The term identifier w of the fuzzy set is initialized to 1 again.
Similarly, all terms in the first working fuzzy set F (W) identified by W are compared against all terms in the other working fuzzy set F (w) identified by w. When all comparisons are complete, a positive result from test 131 causes the rest of the program to return via return point 132. The result is the first working fuzzy set F (W)
Note that it is stored as

Here, the processed fuzzy sets were not normalized (ratio max membership to equal 1). If desired, any of them will do so. Perhaps only the resulting fuzzy set generated in FIG. 8 needs to be normalized.

The exemplary routine described with respect to FIG. 9 can be used to add a fuzzy set everywhere found in FIGS. 6-8. On the other hand, fuzzy addition can also be performed in other ways. For example, a fuzzy set represented as a pair of stored values containing a primitive element and a membership value as described above is, instead, a primitive element or breakpoint value that the expression for the membership value holds internally. It may be stored as a series of expressions having a range of. For example, the basic single passenger fuzzy set F (PASS), or N = 1 in FIG. 5, is expressed as:

B ≦ 0 m = 0 20 <b <40 m = 2b-30 40 <b <50 m = 3b-70 50 <b <70 m = b + 30 70 <b <100 m = 170-b 100 <b <110 m = 270-b 110 <b <120 m = 380-3b / 2 120 <b <130 m = 260-b 130 ≦ b m = 0 where m is in the range of 0 to 100, b is in kilograms.

When a fuzzy set is so expressed, fuzzy addition is described by Kaufman and Gupta in "Introduction to Fuzzy Mathematics: Theory and Applications," von Nostrand Rheinfold, 1985, Section 1.
4 is performed. Also, the multiplication (FIG. 8, step 83) is performed in the manner of its section 1.5. Such a process is less repetitive than that described in connection with FIG. 9 above. However, it requires a large amount of storage capacity and more complex calculations. Let me say that it has nothing to do with the present invention. Any addition method is within the scope of the present invention.

The resulting passenger counting fuzzy set can be used accordingly and can be converted to a concise number if desired.

The present invention can be implemented in a very basic form by simply adding the weight to the single passenger base fuzzy set. Its weight is a succinct value equal to the sum of the average amount of demographic weight gains for clothes, parcels and specific buildings, season and time also matters. this is,
It avoids the need to add all the factor fuzzy sets with respect to the single-passenger basic fuzzy set, thus avoiding a large amount of processing. Selecting the membership value of each passenger primitive of the fuzzy set corresponding to the rest of the process, ie the actual weight involved, remains the same. A simpler way of implementing the invention in its basic form is as in the embodiment of FIG.
And a simple basic single passenger set (FIG. 5, N = 1) in the processing of steps 81 and 91 would generate a fuzzy set for weight. Passengers reduced by the average proportion of average passengers, which represents the sum of average clothing, parcel and demographic weights in the building, after which the basic elements of the passenger fuzzy set for the included weight (eg 600 kilograms in this example) It has each of the basic elements. This would divide the passengers, but if desired, the fuzzy set could be rewritten to the nearest whole number, or many of the departure processes could use a fuzzy set to show the number of passengers for a given weight. It can be used in a split state that is compatible with. However, the methodology disclosed herein in FIGS. 1-9 is clearly preferred because of its accuracy. If accuracy is desired, it is preferable to use sex demographics as in the aforementioned patents.

FIGS. 10 and 11 show that a passenger fuzzy set with multiple terms is used for the departure of an elevator. Each of these terms has a basic element that indicates the number of passengers. The set also has a membership value that indicates that the number of passengers may have contributed the weight in question. For example, such a fuzzy set is stored for future use in order to have fuzzy values for predicting departures, number of passengers, number of passengers getting off and movement for an elevator. Alternatively, these brief numbers are determined from the fuzzy set and stored for future use to predict traffic patterns. Fuzzy sets, on the other hand, can be used in real time for driving processes, such as in hall call assignments, or for peak period determination and driving. These are not all related to the present invention.

The season (clothes) and time of day (parcel) fuzzy sets represent the likelihood of passengers carrying extra weight. On the other hand, the national (demographic) fuzzy set represents the possibility of even heavier passengers. 1 to 5 of the present application
The fuzzy set represented by the above is merely an example, and its details are not relevant to the present invention. As used herein in connection with the fuzzy set of Figure 1, the term "country" is
Or area, or minimum international (or other base weight)
It is meant as another criterion for selecting a particular fuzzy set to represent the potential added weight of passengers beyond the potential. The fuzzy sets are added and then multiplied as described. Or each fuzzy set is
It is multiplied by the number of passengers N and then added to the other multiplied fuzzy sets. That is, any other method can be used to provide an N times equivalent of the base fuzzy set and one or all factor fuzzy sets. A fuzzy set is a binary number arranged to associate each primitive with a corresponding membership value to form a signal, typically a term, and to associate those terms with each set. Is. The given weight, eg 613 kilograms, does not apply to the primitives obtained as the Wth term of the fuzzy set used here, but the term “primitives” was derived by interpolation as explained. It is a thing.

[0039]

According to the invention, the use of a fuzzy set relating to the number of passengers and the weight with which the accuracy is improved is utilized to fuzzy the number of passengers in an elevator for a given measured weight. You can give instructions.

[Brief description of drawings]

FIG. 1 depicts an exemplary fuzzy set showing weight change as a function of the country in which the elevator is operating.

FIG. 2 represents a fuzzy set showing the weight change occurring in different seasons as a function of the clothes worn in a given temperature climate.

FIG. 3 represents a fuzzy set showing the resulting weight difference of packages carried at various times of the day in the up direction.

FIG. 4 represents a fuzzy set showing the resulting weight difference of packages carried at various times of the day in the down direction.

FIG. 5 represents a fuzzy set for one passenger, five passengers, ten passengers, a factor of the type represented in FIGS. 1 to 4 for ten passengers, and a composite fuzzy set for ten passengers. It is a figure which shows the synthetic function which exists.

FIG. 6 records the minimum weight in the elevator during landing, showing the number of passengers remaining after all passengers have boarded, and the elevator load showing the number of passengers moving between floors. FIG. 6 is a logic flow diagram of an exemplary routine for recording.

FIG. 7 is a logic flow diagram of an update routine that is added when needed with a fuzzy set of the type depicted in FIGS.

FIG. 8 is a logic flow diagram of an exemplary routine for performing fuzzy addition.

9 is a logic flow diagram of a routine for determining passenger primitive element membership values as a function of load utilizing a synthetic fuzzy set of the type depicted in FIG.

FIG. 10 is a simple logic flow diagram representing the recording of passenger information derived in accordance with FIGS. 6-8.

FIG. 11 is a simple logic flow diagram when the passenger information of FIG. 10 is used for departure of an elevator.

[Explanation of symbols]

 N ... Number of passengers

Continued Front Page (72) Inventor David Jay. Shirag, Junia United States, Connecticut, South Windsor, Fon Sheen Rain 4-5

Claims (20)

[Claims] 1. A method for departing multiple elevator cars in a building, the responding member indicating a basic element of the possible weight a passenger can add to an elevator car and the probability that the passenger has caused the associated weight. The steps of providing multiple signals representing the basic fuzzy set of a single passenger containing the ship value, the basic components of the possible added passenger weight and the corresponding membership value indicating the probability that one passenger has caused the relevant weight. Providing a plurality of signals representative of a single passenger factor fuzzy set, each including a basic element indicating an added weight as compared to a weight probability expressed in the single passenger fuzzy set, Represent a plurality of composite fuzzy sets, one for each possible number N of passengers that may be in the elevator, each of which is said basic fuzzy set and And providing a plurality of signals that are N times the fuzzy sum of the factor fuzzy set, providing a weight signal indicative of the weight of a passenger in the elevator, each of the terms being associated with the weight indicated by the weight signal. A passenger counting fuzzy set corresponding to one related term of the composite fuzzy set having equal basic elements, each term of the passenger counting fuzzy set being a number of passengers corresponding to the combined fuzzy set of related terms. Providing a plurality of signals representative of a passenger counting fuzzy set having equal membership and the same membership degree as the related term, and a corresponding representation of the number of passengers in the elevator. To give
Responsive to the process of using the passenger fuzzy set signal, departing an elevator for servicing in the building. 2. The step of providing a signal representative of a factor fuzzy set comprises: a basic element of possible additional passenger weight and a corresponding membership value indicating the likelihood of a passenger having an associated additional weight, each passenger being the single passenger. Carrying a plurality of signals representing the factor fuzzy set as a fuzzy addition of a fuzzy set containing a basic element indicating an added weight compared to the probability of weight expressed in the fuzzy set, carrying weight fuzzy 2. A set according to claim 1, characterized in that the set comprises basic elements indicating the different weights carried by each passenger and corresponding membership values indicating the likelihood of passengers carrying an associated weight. Method. 3. A method according to claim 2, wherein the basic elements of the haul weight fuzzy set indicate the weight of different clothing quantities. 4. The method of claim 3, wherein the haulage fuzzy set is seasonally different. 5. The method according to claim 2, wherein the basic element of the haulage fuzzy set indicates the weight of the cargo being carried by a passenger. 6. The method of claim 5, wherein the haulage fuzzy set is different for each hour of the day. 7. The method of claim 5, wherein the haulage fuzzy set is different for the up direction rather than the down direction. 8. The method of claim 7, wherein the haul weight fuzzy set is different for each hour of the day. 9. A method for departing multiple elevator cars in a building, the responding member indicating a basic element of the possible weight one passenger can add to the elevator car and the probability that one passenger has caused the associated weight. Providing multiple signals representing a basic fuzzy set of a single passenger, including ship values, basic elements indicating the different weights each passenger is carrying and the likelihood of the passenger carrying an associated weight. Providing a plurality of signals representing a factor fuzzy set of a single passenger containing corresponding membership values indicating, a plurality of composites, one for each possible number N of passengers that may be in the elevator Presenting a plurality of signals representing a fuzzy set, each signal being N times the fuzzy sum of the basic fuzzy set and the factor fuzzy set, Providing a weight signal indicative of the weight of passengers in the beta, each term corresponding to one relevant term of the synthetic fuzzy set having a primitive element equal to the weight indicated by the weight signal. A fuzzy set, wherein each term of the passenger counting fuzzy set has a basic element equal to the number of passengers corresponding to the composite fuzzy set of related terms, and a membership degree equal to the membership degree of the related term. , Providing a plurality of signals representative of a passenger counting fuzzy set, and to provide a corresponding representation of the number of passengers in the elevator,
Responsive to the process of using the passenger fuzzy set signal, departing an elevator for servicing in the building. 10. A method according to claim 9, wherein the basic elements of the haulage fuzzy set represent the weight of different clothing quantities. 11. The method of claim 10, wherein the haulage fuzzy set is seasonally different. 12. The method of claim 9 wherein the haulage fuzzy set primitives indicate the weight of a load being carried by a passenger. 13. The method of claim 12, wherein the haul weight fuzzy sets are different for each hour of the day. 14. The method of claim 12, wherein the haulage fuzzy set is different for the up direction rather than the down direction. 15. The method of claim 14, wherein the haulage fuzzy set is different for each hour of the day. 16. The step of providing a signal representative of a factor fuzzy set includes: a basic element of possible added passenger weight and a corresponding membership value indicating the likelihood of a passenger having an associated added weight, each passenger being the single passenger. Carrying a plurality of signals representing the factor fuzzy set as a fuzzy addition of a fuzzy set containing a basic element indicating an added weight compared to the probability of weight expressed in the fuzzy set, carrying weight fuzzy 10. A set according to claim 9, characterized in that the set comprises basic elements indicating the different weights carried by each passenger and corresponding membership values indicating the likelihood of passengers carrying an associated weight. Method. 17. A method of departing an elevator utilizing fuzzy logic to determine the number of passengers in an elevator in response to a weight signal indicative of the weight of the elevator and a predetermined fuzzy set indicative of a passenger's base weight. At
A method comprising: generating a passenger fuzzy set that considers the added weight for each passenger as a function of a variable weight factor. 18. The variable weight factor comprises a passenger's regional weight change.
The method described in. 19. The method of claim 17, wherein the variable weight factor comprises a time- and direction-related change in the weight of luggage carried by a passenger. 20. The method of claim 17, wherein the variable weight factor comprises a seasonally related weight change in passenger clothing.