JP2948475B2 - Runout sensor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vibration sensor for detecting a vibration amount of a suspended load suspended by, for example, a container crane.

[0002]

2. Description of the Related Art With reference to FIGS. 7 to 11, a description will be given of a detection state of a swing amount of a hanging tool suspended by a crane rope or the like and in which a shake amount (one-way shake) and a skew amount (rotational shake) are mixed. I do. As shown in FIG. 7, a target 7 having a band-shaped marker 8 shown in FIG. 8 is installed on a hanging tool 4 suspended from a trolley 5 of a crane via a wire 6. The movement is photographed every moment by the camera 1 installed on the trolley 5, and the photographed image is subjected to image processing, and then the output signal is used for correlation processing to calculate the maximum shake amount of the hanging tool 4.

[0003] First, the detection state of the one-way shake amount is described.
You. FIG. 10A shows a marker photographed as described above.
8 shows an example of image 8 and marker image C of image A.₀And of image B
Marker image C₁Is ΔX₁Just away. Like time
Different marker images C at different intervals _Two, C_ThreeIs ΔX_Two, Δ
X_ThreeThere is a gap in the interval. FIG. 10B shows this image A,
4 shows an output signal obtained by performing image processing on B. Now the image
A line L_aThe output signal in the y-axis direction is φ_A, Image B
Line L_bThe output signal in the y-axis direction is φ_BThen
Output signal φ_AAnd φ_BIs the interval ΔX₁Just away. Out
Force signal φ_AIs the reference signal, and h = ΔX₁After all,
The cross-correlation function φ (h) is

(Equation 1) Is defined as This cross-correlation function φ (h) is shown in FIG.
(B) As is clearer, in a finite section of x, h =
Since ΔX ₁ is zero except when it is a certain value, the cross-correlation function φ (h) is calculated for each output signal of each image photographed at every minute time interval τ ₀ , and max | φ (h) | When determining the value of becomes h changes as h _i in Figure 11. Thus, the one-way maximum shake amount (shake amplitude) h ₀ of the hanging tool 4 is calculated as h ₀ = h ₀₁ + h ₀₂ .

Next, a case where the amount of one-way shake and the amount of skew (rotational shake) coexist will be described. As shown in FIG. 9, the hanging tool 4 that swings with rotation moves from a position shown by a solid line to a position shown by a two-dot chain line with a gradient of θ during a certain time interval, and the marker 8 attached to the target 7 For example a point in the x-direction moves by h ₁ to aa point, b point to bb point
and it was moved by h _2. Now, assuming that the distances between the reference point O and the point a, point b and point e of the end of the hanger at the original hanger position are L ₁ , L ₂ and L ₃ , respectively, the runout ΔP ₁ and the skew ΔP ₂ Is expressed as follows. ΔP ₁ = h ₁ − ｛(h ₂ −h ₁ ) / (L ₂ −L ₁ )｝ · L ₃ (1) ΔP ₂ = ｛(h ₂ −h ₁ ) / (L ₂ −L ₁ ) ｝ · L ₂ (2) Therefore, similarly to the above, the cross-correlation function φ (h ₁ ) at the points a and b of the marker 8 is obtained by using the output signal photographed at every minute time interval τ ₀ and subjected to image processing. Calculate φ (h ₂ ) and max ｜ φ (h ₁ )
_{|, Max | φ (h 2} ) | and seek h _1, h ₂ when made, the h _1,
The shake amount ΔP ₁ and the skew amount ΔP ₂ are obtained by substituting h ₂ into the above equations (1) and (2). This shake amount Δ
P _1, the maximum deflection amount Pmax occurring at an end portion e of the skew amount [Delta] P from the load block 4 is calculated as Pmax = ΔP ₁ + ΔP _2.

[0005]

In the prior art for detecting the swing amount of a hanging tool in which one-way shake and rotational shake (skew amount) are mixed, the distance between points a and b attached to the target 7 is determined by the marker 8. Due to restrictions on dimensions etc., it can not be too large,
In the case of a runout in which the inclination of the hanger 4 is not so large, the detection accuracy of the skew amount deteriorates, and as a result, the detection accuracy of the entire runout amount of the hanger 4 has a disadvantage.

[0006]

According to a first aspect of the present invention, there is provided a vibration sensor provided in a crane for transferring a load by suspending the load via a hanging tool and moving the trolley. , A pair of targets installed at predetermined intervals on a hanging tool and attached with a band-shaped marker, a pair of cameras installed on a trolley and photographing the movement of the marker on each target, and photographed by the camera every moment An image processing apparatus that performs image processing on the processed video, and one of the output signals that have been subjected to the image processing is used as a reference signal, and the reference signal and the output signal obtained at predetermined time intervals are used to generate a marker for each target. cross-correlation function phi (h _1), to calculate the phi (h _2), the cross-correlation function _{φ (h 1), φ (} h 2) determine the value h _1, h ₂ when is maximized, this value with h _1, h ₂ calculates the shake amount and the skew amount of the load block, An arithmetic unit obtaining a maximum deflection amount of the suspender by the shake amount and a skew amount is calculated, in that it has construction features.

Further, a second invention for solving the above-mentioned problem is provided.
Ming configuration moves the trolley by suspending the load via the hanging tool
The crane that carries the load.
The sensor is installed on the hanging device at a predetermined interval.
A pair of targets each having two strip-shaped markers
And a marker for each marker on each target installed on the trolley
A pair of cameras that capture each movement, and the cameras
Image processing device for image processing of video taken at
One of the processed output signals is used as a reference signal, and this reference signal
And output signals obtained at predetermined time intervals,
The cross-correlation function f (δ for each marker of the target₁), f (
δ_Two) And the cross-correlation function f (δ_Three), f (δ _Four)
Cross-correlation function f (δ₁), f (δ_Two) Is the maximum value
δ₁, δ_TwoFrom δ_m1And the cross-correlation function
f (δ_Three), f (δ_Four) At which δ is maximum_Three, δ_FourKarahira
Average δ_m2And the average value δ_m1, Δ_m2Using of hanging equipment
Calculate the runout and skew, and calculate the runout and
An arithmetic unit that calculates the maximum run-out of the lifting gear based on the amount of skew
And characterized in that:

[0008]

According to the first aspect of the invention, the movement of the band-shaped marker attached to the pair of targets on the hanger is photographed by the pair of cameras on the trolley, and the photographed image is sent to the image processing device for image processing. The output signal of the image processing apparatus is sent to the arithmetic unit. Here, one of the output signals is used as a reference signal, and the cross-correlation function φ is used for each target marker by using the reference signal and the output signal obtained at predetermined intervals. (h ₁ ), φ (h ₂ ) is calculated, and the value when the cross-correlation functions φ (h ₁ ), φ (h ₂ ) is maximized
The shake amount and the skew amount are calculated using h ₁ and h ₂ , and the maximum shake amount of the hanging tool is obtained from the shake amount and the skew amount.

In the second invention, the movement of the two band-shaped markers attached to the pair of targets on the hanger is photographed by the pair of cameras on the trolley, and the photographed image is sent to the image processing device for image processing. Then, the output signal of the image processing device is sent to the arithmetic device. Here, one of the output signals is used as a reference signal, and the reference signal and the output signal obtained at predetermined intervals are used for each marker of each target. Correlation functions f (δ ₁ ), f (δ
₂ ) and the cross-correlation functions f (δ ₃ ), f (δ ₄ ) are calculated, and the average is calculated from the values δ ₁ , δ ₂ when the cross-correlation functions f (δ ₁ ), f (δ ₂ ) are maximized. The value δ _m1 is obtained, and the values δ ₃ , δ ₄ when the cross-correlation functions f (δ ₃ ), f (δ ₄ ) are maximized
An average value [delta] _{m @ 2} from the average value [delta] _m1, calculate the shake amount and the skew amount using the [delta] _{m @ 2,} obtaining the maximum deflection amount of the load block from the shake amount and a skew amount.

[0010]

Next, an embodiment of a shake sensor according to the first invention will be described with reference to FIGS. 1 shows the entire shake sensor, FIG. 2 shows the state of the target, FIG. 3 shows the principle of shake detection, FIG. 4 shows a block diagram of the device, and FIG. 5 shows the shake detection procedure. Note that members that are the same as those in the related art are given the same reference numerals.

[0011] As shown in FIG. 1, at predetermined intervals l ₀ from the trolley 5 crane on hanger 4 hung by a wire 6 pair of targets 7a, 7b are installed,
As shown in FIG. 2, strip-shaped markers 8a and 8b are attached to the targets 7a and 7b, respectively, and a pair of cameras 1a and 8b for photographing the movement of the markers 8a and 8b of the targets 7a and 7b on the trolley 5. 1b is installed. It is preferable that the strip markers 8a and 8b have a white background.

As shown in FIG. 4, the shake sensor includes cameras 1a and 1b, an input unit 9 connected to the cameras 1a and 1b, an operation unit 10 connected to the input unit 9, and an operation unit 10
An image processing device 13 including a storage device 11 connected to the
It comprises an arithmetic unit 12 connected to an image processing unit 13.

Next, the operation of the shake sensor will be specifically described with reference to FIGS. As shown in FIG. 3, during a certain time interval, the hanging tool 4 swinging with rotation moves from a position shown by a solid line to a position shown by a two-dot chain line with an inclination of an angle θ, and is attached to a target 7a. The point a of the marker 8a is a
moved by h ₁ in the x-direction to a point, b point marker 8b as those in the target 7b is similarly has moved in the x direction by h ₂ to bb point. Now, the reference point O and the point a in the original suspender position, and the distance between b and the load block ends e, respectively and L _1, L ₂ and L _3, the shake amount [Delta] P ₁ and skew amount [Delta] P ₂ is above Equations (1) and (2) can be used.

Thus, as described above, the cross-correlation function φ () at the point a of the marker 8a and the point b of the marker 8b is obtained by using the output signals obtained by photographing the images at the minute time intervals τ ₀ with the cameras 1a and 1b and performing image processing. h ₁ ), φ (h ₂ ), and when the cross-correlation function φ (h ₁ ), φ (h ₂ ) is maximized, max | φ (h ₁ ) |, max |
Obtain the values h ₁ and h ₂ of φ (h ₂ ) |, and substitute the values h ₁ and h ₂ into the above-described equations (1) and (2) to obtain the shake amount ΔP ₁
And the skew amount ΔP ₂ . In this way, the movement of the hanging tool 4 arbitrarily swinging can be separated into one-way shaking and rotational shaking, and can be accurately detected. Further, the swing amount ΔP ₁ and the skew amount ΔP ₂ can be used to detect the end of the hanging tool 4. The maximum shake amount h max occurring in the section e can be detected as h max = ΔP ₁ + ΔP ₂ . Note that detection accuracy can be improved by increasing l ₀ = L ₂ −L ₁ .

Next, a shake sensor according to a second invention will be described with reference to FIG. As shown in FIG. 6, the shake sensor according to the second invention has a pair of targets 7a and 7b, _two strip-shaped markers 8a ₁ and 8a ₂ and markers 8b ₁ and 8b ₂ attached thereto. was placed on the load block 4 at intervals l _0, a pair of cameras 1a was placed on the trolley 5, intended to 1b movement of the markers 8a _1, 8a ₂ and the marker 8b _1, 8b ₂ every moment in shooting, and other The device configuration of the shake sensor is the same as that of the first invention (see FIG. 4).

Next, the operation of the shake sensor will be specifically described. As in the first invention, the cameras 1a and 1b take images at minute time intervals τ ₀ , and use the output signals subjected to image processing to obtain the points a and c of the markers 8a ₁ and 8a ₂ and the markers 8b ₁ and 8b ₂ . b
The cross-correlation functions f (δ ₁ ), f (δ ₂ ) and the cross-correlation functions f (δ ₃ ), f (δ ₄ ) at the points d and d are calculated, and the cross-correlation functions f (δ ₁ ), f (δ ₂ ), f (δ ₃ ), f (δ ₄ ) are maximized max | f (δ ₁ ) |, max | f (δ ₂ ) |, max | f (δ
₃ ) |, max | f (δ ₄ ) values δ ₁ , δ ₂ , δ ₃ , δ ₄ are obtained. Then, the respective average values δ _m1 = 1/2 (δ ₁ + δ ₂ ), δ _m2 =
Find 1/2 (δ ₃ + δ ₄ ). On the other hand, as shown in FIG. 6, the hanging tool 4 that swings with rotation moves from a position shown by a solid line to a position shown by a two-dot chain line at an angle θ during a certain time interval, and is attached to the target 7a. The points a and c of the markers 8a ₁ and 8a ₂ moved in the x direction by δ _a1 and δ _a2 , and the points b and d of the markers 8b ₁ and 8b ₂ attached to the target 7b are also δ _{b1 in the} x direction. , Δ _b2 . Now, the reference point O at the original hanging tool position and the points a (c) and b (d)
) And the distances from the end of the lifting gear e are L ₁ , L ₂ and L ₃ respectively, and δ _m1 = 1/2 (δ _a1 + δ _a2 ), δ _m2 = 1/2 (δ _b1 + δ
putting a _b2), shake amounts [Delta] q ₁ and skew amount [Delta] q ₂ is expressed as follows. Δq ₁ = δ _m1 − ｛(δ _m2 −δ _m1 ) / (L ₂ −L ₁ )｝ · L ₃ (3) Δq ₂ = ｛(δ _m1 −δ _m1 ) / (L ₂ −L ₁ ) ｝ · L ₂ …… (4)

Therefore, the shake amount Δq ₁ and the skew amount Δq ₂ are obtained by substituting the average values δ _m1 and δ _m2 obtained by the correlation processing into the above-mentioned equations (3) and (4). In this manner, the movement of the hanging tool 4 arbitrarily swinging can be separated into one-way shaking and rotational shaking, and accurately detected, and the end values of the hanging tool 4 can be detected from these detected values Δq ₁ and Δq _2. Part e
The maximum deflection amount delta] max occurring can be detected as δmax = Δq ₁ + Δq _2. Note that the detection accuracy can be improved by increasing l ₀ = L ₂ −L ₁ , and by distinguishing one target from two white lines by attaching two markers to one target, False detection can be prevented.

[0018]

As described in detail above, according to the shake sensor of the present invention, the movement of the arbitrarily swinging hanging tool is separated into one-way shake (shake amount) and rotational shake (skew amount). Detection can be performed quickly and accurately, and the maximum swing amount of the hanging tool can be accurately detected from these detection values. As a result, it is possible to perform fine anti-sway control in consideration of the one-way anti-sway control and the skew control. According to the second aspect of the invention, by attaching two markers to one target, it is possible to reliably distinguish the target from other white lines and prevent erroneous detection.

[Brief description of the drawings]

FIG. 1 is an overall view of a shake sensor according to the present invention.

FIG. 2 is a marker diagram of a target according to the first invention of the present invention.

FIG. 3 is an explanatory diagram of a shake amount, a skew amount, and a detection principle according to the first invention of the present invention.

FIG. 4 is an apparatus block diagram of the first invention of the present invention.

FIG. 5 is a diagram showing a shake amount detection procedure according to the first invention of the present invention.

FIG. 6 is an explanatory diagram of a principle of detecting a shake amount and a skew amount according to the second invention of the present invention.

FIG. 7 is an overall view of a conventional shake sensor.

FIG. 8 is a marker diagram of a conventional target.

FIG. 9 is an explanatory diagram of a conventional shake amount and skew amount detection principle.

FIG. 10 is an explanatory diagram of a correlation processing procedure applied to the present invention.

FIG. 11 is an explanatory diagram of a correlation processing procedure applied to the present invention.

[Explanation of symbols]

1a, 1b camera 4 suspender 5 Trolley 6 wires 7a, 7b target _{8a, 8b, 8a 1, 8a} 2, 8b 1, 8b 2 band marker 12 calculation unit 13 an image processing apparatus

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Goo Murata 4-6-22 Kannon Shinmachi, Nishi-ku, Hiroshima-shi, Hiroshima Mitsubishi Heavy Industries, Ltd. Hiroshima Research Institute (56) References JP-A-6-208624 (JP, A) JP-A-4-201988 (JP, A) JP-A-3-36190 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) B66C 13/18-13/50

Claims (2)

(57) [Claims] 1. A trolley is moved by suspending a load via a suspender.
Be equipped with a crane that transports loads by moving
In the run-out sensor, a band-shaped marker is installed on the hanging device at a predetermined interval.
And the movement of the marker on each target set on the trolley.
A pair of cameras that shoot each one, and image processing that processes images that are shot by the cameras every moment
The device and one of the image-processed output signals is used as a reference signal.
Signal and an output signal obtained at predetermined time intervals.
And the cross-correlation function φ (h ₁), φ
(h_Two), And this cross-correlation function φ (h₁), φ (h_Two) Is maximum
The value h when_1,h_TwoAnd this value h_1,h_TwoUsing hanging equipment
Calculate the run-out and skew of the
Calculation equipment that calculates the maximum runout of the lifting gear based on the skew amount
And a vibration sensor comprising: 2. A swing sensor provided on a crane that transports a load by suspending a load via a hanging tool and moving a trolley, wherein two strip-shaped markers installed at predetermined intervals on the hanging tool. A pair of targets respectively attached to the trolley, a pair of cameras installed on the trolley for photographing the movement of each marker on each target, an image processing device for image-processing an image captured by the camera, and an image processing device. One of the output signals thus obtained is used as a reference signal, and the cross-correlation function f () for each marker of each target is obtained by using the reference signal and the output signals obtained at predetermined time intervals.
δ ₁ ), f (δ ₂ ) and the cross-correlation function f (δ ₃ ), f (δ ₄ ) are calculated, and the value when the cross-correlation function f (δ ₁ ), f (δ ₂ ) is maximized The average value δ _m1 is obtained from δ ₁ , δ ₂ and the value δ ₃ , at which the cross-correlation function f (δ ₃ ), f (δ ₄ ) is maximized,
an average value [delta] _{m @ 2} from [delta] _4, the average value [delta] _m1, calculate the shake amount and the skew amount of the suspender with [delta] _{m @ 2,} obtaining the maximum deflection amount of the load block by the calculated shake amount and the skew amount A shake sensor, comprising: an arithmetic unit;