MOVEMENT EVALUATION METHOD FOR AN ELEVATOR CAR

Abstract

The present invention is about a system and method for evaluating the movement of an elevator car within a hoistway. The same is based on gathering movement data of the car by means of a rotation encoder or acceleration sensor. Since these data are counted pulses there is a need to convert them into real movement data. This conversion is calibrated automatically by comparing the movement data with a length distance being passed by the car, wherein said length distance is configured to be unchangeable over the time. Base on the exact movement results gained therewith by means of said calibration one also gets better position results of the car in the hoistway.

Claims

1. Method for evaluating the movement of an elevator car within a hoistway, comprising the steps of gathering movement data of at least one component involved in moving the elevator car by means of a rotation encoder or acceleration sensor, reading values of at least one identification marker that is installed in the hoistway and which the elevator car can pass by or arrive at when moving, transmitting simultaneously the rotational movement data and the values of the identification marker to a controller, picking up two signal values of the at least one identification marker that indicate a defined length of the at least one identification marker, sampling at least two movement data readings responsive to the receipt of the two signal values, and calibrating the movement data by calculating a conversion factor, wherein the conversion factor is based on the two picked up signal values of the at least one identification marker and the at least two movement data readings.

2. Method according to claim 1, wherein the evaluation takes place automatically any time the car passes the identification marker by.

3. Method according to claim 1, wherein the calibrated movement data is converted into position data of the car referenced to a relative position of the car in the hoistway by means of the conversion factor.

4. Method according to claim 1, wherein position data are gathered from further identification markers that are installed in the hoistway.

5. Method according to claim 1, wherein the calibrating step is performed in an electronic safety controller.

6. Movement determination system for an elevator car, comprising an elevator car that moves in a hoistway, the system comprising an online measuring device like a rotation encoder or acceleration sensor gathering movement data of at least one component involved in moving the elevator car, at least one identification marker that is installed in the hoistway and which the elevator car can pass by when moving, wherein the identification marker is designed to indicate a defined length-dimension, and means for calibrating the measured movement data of the online measuring device based on the length data of the identification marker, wherein the system is configured to carry out the method according to claim 1.

7. The movement determination system according to claim 6, wherein the means for calibrating the measured movement data of the online measuring device comprise means for calculating a conversion factor, wherein the conversion factor is based on the length data of the identification marker and the movement data.

Description

[0029] In the following, the invention is elucidated by means of an embodiment as shown in the drawings. In these,

[0030] FIG. 1 is a view of parts of an elevator car with a rope pulley mounted according to the invention

[0031] FIG. 2 shows details of the rope pulley according to the invention;

[0032] one detail is a plan view whereas the other one is a sectional view.

[0033] FIG. 3 visualized a possible calibration algorithm.

[0034] FIG. 1 shows a sling of an elevator car 1. At its bottom there is a pulley beam 3 at which there are mounted two rope pulleys 2 via which a hoisting rope (not shown) for the suspension of the car is guided. Both of these rope pulleys 2 are provided with an encoder. As there are two rope pulleys with encoders, respectively, a reciprocal comparison of the encoder information is performed to increase the reliability of safety level of the arrangement for determining the position of the car within the shaft.

[0035] The encoder is preferably a magnetic encoder, as shown in FIG. 2. It comprises a magnetic band 5 mounted on a shelf of the rope pulley 2. A reader 6 is mounted in a hole of the pulley beam 3.

[0036] Instead of an encoder, an acceleration sensor mounted to the car could be used for speed and position calculation of elevator car.

[0037] While the elevator car is further equipped with an identification marker reader device, there are identification markers installed in the elevator shaft that functionally act together.

[0038] The elevator car is also provided with a safety bus node, which is connected to an electric safety controller via a data bus, i.e. safety bus, which is included in the trailing cable. The reader 6, as well as the identification marker reader device, is connected to the bus node such that movement data of the encoder is transferred to the safety controller.

[0039] According to FIG. 3 there is visualized a possible algorithm for the calibration. As to be seen on left side, there is shown an elevator hoistway in symbolic view, in which a car can move up and down (shown by the left arrow). A floor magnet stripe is installed in the hoistway at the height of a floor, which stripe is divided into ten almost equidistant sample areas S1 to S10. There is an upper scaling area “USA” and a lower scaling area “LSA”, each encompassing five of them, namely the upper scaling area having numbers from S1 to S5 and the lower scaling area having numbers from S6 to S10. These two groups of “USA” and “LSA” can be separated over a specific distance but belong to one and the same identification strip constituting the identification marker. While the upper scaling area “USA” is from 55 mm to 105 mm, the lower scaling area “LSA” is from −55 mm to −105 mm in this example. Each area from S1 to S10 is provided with an identification mark, i.e. a value resolved from the varying magnetic field of the magnets of the identification strip, which value identifies the respective area position as a linear position “LP”.

[0040] So to say, [0041] area “sample 1” is allocated the linear position value “100” retrieved from the identification marker by means of the identification marker reader device; [0042] area “sample 2” is allocated the linear position identifier “91”; [0043] area “sample 3” is allocated the linear position identifier “80”; [0044] area “sample 4” is allocated the linear position identifier “70”; [0045] area “sample 5” is allocated the linear position identifier “60”; [0046] area “sample 6” is allocated the linear position identifier “−60”; [0047] area “sample 7” is allocated the linear position identifier “−70”; [0048] area “sample 8” is allocated the linear position identifier “−81”; [0049] area “sample 9” is allocated the linear position identifier “−90”; [0050] area “sample 10” is allocated the linear position identifier “−100”;

[0051] When the car has passed by the entire identification strip, a linear position change can be calculated for each sample 1 to 10 over the entire range of said sample areas by: [0052] Linear position change.sup.1=“LP of S1” minus “LP of S10”; [0053] Linear position change.sup.2=“LP of S2” minus “LP of S9” value;

[0054] etc.

[0055] A similar listing is accomplished with the movement data coming from the encoder. To each sample S1 to S10 a corresponding encoder pulse count “EPC” is allocated, wherein an encoder pulse count change is reversely calculated by [0056] Encoder pulse count change.sup.1=“EPC of S10” minus “EPC of S1”; [0057] Encoder pulse count change.sup.2=“EPC of S9” minus “EPC of S2”;

[0058] etc.

[0059] In the next step, an encoder resolution value is calculated for all the samples by:

[00001]$\mathrm{Encoder}\mathrm{resolution}\mathrm{value}“\mathrm{ERV}”\left[\mathrm{mm}\mathrm{per}\mathrm{pulse}\right]=\frac{\mathrm{Linear}\mathrm{position}\mathrm{change}}{\mathrm{Encoder}\mathrm{pulse}\mathrm{count}\mathrm{change}}$

[0060] This leads to five results as listed in the table below and titled “ERV”.

[0061] Then, the encoder resolution values are sorted in an ascending order which listing is titled “SERV” for Sorted Encoder Resolution Values.

[0062] In the next step, a median value is stored to an array that includes the Encoder resolution values for all passed area positions, i.e. the magnets allocated therewith. This array-listing is titled “ERVM” for Encoder Resolution Values Median. When having repeated the calculation for the median value four times, an array with five placeholders is filled into which a median-resolution-value is set for all passed magnets. This shows the best-mode, while a minimum of three resolution median values should be calculated for passing at least three magnets to gain a reasonable result. This is a matter of statistical phenomenon: While a more reliable measurement result may be achieved when the number of samples increases, it showed in practice that three magnets would be adequate for a minimum reliable result. A repetition of the same calculation is then made for at least three magnets, and their median values are set in the same way. As one can see in the lowest EVRM table of FIG. 3 there are 5 different values stored, meaning the same calculation that has been repeated for five different magnets, and their median values have been stored in said EVRM table.

[0063] Now, a median resolution is calculated for each successfully sampled magnet and stored into array. When sufficient number (let's say three) of such median values exist, a mean value is calculated and taken as conversion factor. In the end, from the encoder median resolution values for all magnets an encoder resolution value is calculated—that is in the present example 0.2498. This value is now taken as a conversion factor for converting the encoder pulse counts into the distance travelled, what reflects the calibration of the movement data. The shown algorithm has some benefits: First of all, it is easy of being implemented in a computer program of a processor. For example, selecting a median value instead of a mean value, means that a computer program doesn't have to make calculations, but only a comparison of separate values and a selection therefrom, which doesn't require much processing power. Secondly, different lengths between the samples within the same magnet will be covered, including the maximum length as defined with samples 1 and 10. Of course, there will be shorter lengths also, such as that between samples 5 and 6, but nevertheless there is a median value selection, too, which will exclude possible individual errors.