JP2931711B2 - Spark detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a spark detecting apparatus for detecting the occurrence of a spark in an electroplating process of a metal material, and for predicting the position and depth of a flaw that may have occurred on the metal material due to the spark. About.

[0002]

2. Description of the Related Art Generally, in an electroplating process, a constant current is applied while maintaining a constant distance between a metal material and a plating cell and a metal material and a plating electrode roll are brought into close contact with each other to perform electroplating in a continuous automatic process. Has been given. FIG.
An apparatus for continuously performing this type of electroplating is schematically shown, and reference numeral 1 indicates a strip-shaped metal material to be electroplated. The metal material 1 is transported by being pressed against the pair of plating electrode rolls 2. Further, two sets of plating cells 3a and 3b are arranged so as to keep a certain distance from the metal material 1. The plating electrode roll pair 2 and the plating cells 3a, 3b are connected via a plating power supply 4, and an electroplating layer is formed on the surface of the metal material 1 by an electrolytic current flowing through the plating cells 3a, 3b and the metal material 1. It has become.

When a seam of the metal material 1 passes between the plating electrode roll pair 2 or when the metal material 1 vibrates, an overcurrent flows, and the connection between the plating electrode roll pair 2 and the metal material 1 occurs. Sparks may be generated in the contact portions, and the heat may cause a part of the metal material to melt, thereby damaging the surface of the metal material 1. Therefore, such electroplating equipment includes
There is a device provided with the following detection device for detecting the spark caused by the overcurrent described above. Generally, when a portion where the metal material 1 has a seam and has a changed thickness or a deformed portion passes through the plating electrode roll pair 2, the distance between the plating cells 3a and 3b and the metal material 1 changes. Also, since the degree of adhesion between the plating electrode roll pair 2 and the metal material 1 becomes unstable, the magnitude of the plating current changes, and an overcurrent flows with the change, often causing a spark. The ammeter 5 detects such a plating current. The detection current i is introduced to the current change detection means 6,
The current change detecting means 6 differentiates the current i to calculate a time change α thereof, and supplies the time change α to the spark detecting means 7.

On the other hand, in the course of the movement of the metal material 1,
A pair of speed detection rolls 8 is provided, and the number of rotations of the rolls of the pair of speed detection rolls 8 is detected by a rotation number detector 9. Is calculated. The moving speed of the metal material 1 is introduced into the position detecting means 11, and the position detecting means 1
Reference numeral 1 denotes a position of a portion in contact with the plating electrode roll pair 2 with respect to the tip of the metal material 1 as a current position p of the metal material 1.
And outputs this to the spark detection means 7.
Therefore, the spark detecting means 7 generates a spark when the time change α of the plating current is equal to or more than a certain value, and the metal material 1
Assuming that a scratch has occurred, the position of the scratch at this time is determined.

[0005]

However, the conditions under which the metal material 1 can be damaged by the generation of a spark can vary depending on various conditions such as the magnitude of the current change, the time, the moving speed of the metal material 1, and the like. Yes, with some ambiguity. Therefore, the damage is not always caused under the condition set in advance, and may not be damaged. In addition, although it is possible to some extent to detect whether a spark has actually occurred and whether it has been scratched as a phenomenon, it is difficult to determine the depth of the scratch, and human inspection is performed exclusively in the subsequent product inspection process. There was a problem that we had to rely on.

Accordingly, the present invention solves the above-mentioned problems of the prior art, and based on data such as the magnitude of a current change and the moving speed of a metal material, determines the depth of a flaw caused by the occurrence of a spark. An object of the present invention is to provide a spark detection device that can be predicted.

[0007]

In order to achieve the above object, the present invention provides a plating current detecting means for detecting a current flowing through an electrode roll, and a time change of a plating current from an output of the plating current detecting means. Current change detecting means
Speed detecting means for detecting the moving speed of the metal material; position detecting means for calculating the position of the metal material based on the moving speed of the metal material; and metal generation by spark generation based on the time change of the plating current and the moving speed of the metal material. Fuzzy inference means for inferring the depth of a flaw likely to have been on the material surface, and spark detection means for detecting occurrence of flaws due to spark from the fuzzy inferred flaw depth and the position of the metal material It is characterized by the following.

[0008]

[Function] The magnitude of the time variation of the detected plating current,
Using the moving speed of the metal material as an input, the depth of the flaw caused by the spark is inferred by the fuzzy inference means. In this case, the fuzzy inference means infers that if the time variation of the plating current is small and the moving speed of the metal material is high, the depth of the scratch due to the spark is shallow, and the time variation of the plating current is slightly small, and If the material moves at a relatively high speed, the inferred depth of the spark is slightly shallower.If the time variation of the plating current is rather large and the metal material moves at a slightly lower speed, the depth of the spark is less. It is inferred that the depth of the spark is deep if the plating current has a large time variation and the moving speed of the metal material is slow. As described above, according to the present invention, it is possible to obtain prediction data on the depth of a flaw caused by a spark and the position of the flaw.

[0009]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a perspective view of a spark detector according to an embodiment of the present invention. Components denoted by the same reference numerals as those in FIG. 4 indicate the same or equivalent components, and a detailed description thereof will be omitted. FIG.
, The plating current i detected by the ammeter 5
Is introduced into the current change detecting means 6, which differentiates the current i with respect to time to calculate the value of the time change α, and calculates the time change α by the fuzzy inference means 12
To give to. This fuzzy inference means 12
Will be described later. On the other hand, the number of rotations of the speed detection roll pair 8 arranged in the middle of the movement path of the metal material 1 is:
The moving speed detecting means 10 calculates the moving speed of the metal material 1 on the basis of the output, which is detected by the rotation speed detector 9. The moving speed v is sent to the position detecting means 11 and the fuzzy inference means 12, among which the position calculating means 7 calculates the current position p of the metal material 1 based on the moving speed v. The fuzzy inference means 12 receives the time change α of the plating current given from the current change detection means 6 and the moving speed v of the metal material 1 according to a fuzzy control rule, which will be described later, used for the fuzzy inference, and generates a metal by spark generation. Board 1
Fuzzy inference of the depth of the flaw estimated to be attached to the surface is output to the spark detection means 7 as the depth S. The spark detecting means 7 detects the flaw by the position and the depth from the fuzzy inferred flaw depth S and the current position p of the metal material given from the position detecting means 11.

Next, a method of fuzzy inference will be described with reference to FIG. Here, the fuzzy inference uses an inference method using the MIN inference method. The control rules for inference are:
In this embodiment, four fuzzy inference rules R1, R2,
It consists of R3 and R4. The fuzzy inference rule is i
Expressed in f-then format, if parts of the antecedent part are memberships A11, A12, A
21, A22, A31, A32, A41, and A42. The then of the consequent part is the membership function B1,
These are represented by B2, B3, and B4, which are collectively shown as fuzzy inference rules R1, R2, R3, and R4, as follows. R1; if α = A11 and v = A12 then
s = B1. R2; if α = A21 and v = A22 then
s = B2. R3; if α = A31 and v = A32 then
s = B3. R4; if α = A41 and v = A42 then
s = B4.

Next, each such fuzzy inference rule R
The membership functions of 1, R2, R3, and R4 will be described.

(1) Fuzzy Inference Rule R1 First, the membership function of the antecedent part will be described.
In the membership function A _11, the input values of the horizontal axis α is the magnitude of the time change of the plating current i, omega ₁₁ of the vertical axis the magnitude of variation of the plating current extent belongs to the fuzzy label "smaller" The degree of conformity. In the membership function A12, the input value v of the horizontal axis at a moving speed of the metal material 1, omega ₁₂ of the vertical axis represents the goodness of fit of whether the moving speed is the extent to which belong to the fuzzy label "fast" . The membership function B1 of the consequent part is
The depth S when the depth of the flaw is “shallow” is set. (2) membership function A21 of the fuzzy inference rules R2 antecedent part, goodness-of-fit the size of the time variation of the plating current i α is referred to as a "slightly smaller" ω ₂₁
give. Similarly, the membership function A22 is moving velocity v of the metal material gives the goodness-of-fit ω ₂₂ that "somewhat faster". The membership function B2 of the consequent sets the depth S when the depth of the flaw is “slightly shallow”.

(3) Fuzzy Inference Rule R3 The membership function A31 of the antecedent part is a fitness ω ₃₁ that the magnitude α of the time change of the plating current i is “slightly large”.
give. Membership function A32 is moving velocity v of the metal material gives the goodness-of-fit ω ₃₂ that "a little slow". The membership function B3 of the consequent sets the depth S when the depth of the flaw is “slightly deep”. (4) membership function A41 of the fuzzy inference rules R4 antecedent part, the magnitude of the time change of the plating current α gives the goodness-of-fit ω ₄₁ referred to as "large". The membership function A42 is calculated as follows:
Gives a fitness ω ₄₂ of “slow”. The membership function B4 of the consequent part sets the depth S when the depth of the flaw is “deep”. According to the above four fuzzy inference rules R1, R2, R3, R4, fuzzy inference is performed as follows.

First, the membership function of the antecedent part for each fuzzy inference rule R1, R2, R3, R4 based on the magnitude of the time change α of the plating current i and the input value of the moving speed v of the metal material. Find the value of In this case, since there are two input values, each fuzzy inference rule A1, A2, A
3, For A4, let ω _i = min ｛ω _i1 , ω _i2 ｝ (i = 1,..., 4) be the result of the antecedent part. Next, from the results obtained in the antecedent part, the membership functions B1, B2, B3, B4 of the consequent part of each fuzzy inference rule R1, R2, R3, R4 are modified to obtain the inference result ω _i C _i . obtain. For example, the fuzzy inference rule R2, omega ₂ is because it is ω _21, ω _i C _i Figure 2
The shaded portion shown in FIG. Then, a synthetic fuzzy set C _* is created using the inference results ω _i C _i by the fuzzy inference rules R1, R2, R3, and R4 using the following equation.

[0015]

(Equation 1) Then, the center of gravity S ₁ of the synthesis ambiguity set C _* is calculated using the following equation.

[0016]

(Equation 2) The center of gravity S1 obtained as a result of this calculation is inferred as the depth of a flaw that would have occurred on the surface of the metal material 1 due to the occurrence of spark.

Here, for example, the detected plating current i
Is the time change α of α ₁ and the moving speed v of the metal material 1 is v
₁ , the fuzzy inference means 12 outputs
Determine the depth S ₁ of the flaw that would have attached to the surface of in the following manner. First, the fuzzy inference rules R1, a fitness omega ₁₁ obtained from the membership function A11 in time variation alpha is alpha ₁ plating current i, the moving velocity v is fit omega ₁₂ obtained from the membership function A12 in v1 The smaller one (in this case, ω ₁₂ = 0) is set as the result of the antecedent part.
Then, when the membership function B1 in the consequent part is corrected, the membership function becomes zero. Similarly, in the fuzzy inference rule R2 and the fuzzy inference rule R3, the inference results are the membership function B2 and the membership function B, respectively.
3 is a hatched portion. Note that the inference result based on the fuzzy inference rule R4 indicates that the fitness ω42 of the membership function B42 is 0, so that the membership function B1
Becomes 0. Therefore, the sparking of the metal material 1
Wound depth that would have attached to the surface is inferred centroid s coordinates of the pattern shown by oblique lines obtained by combining the inference result by the fuzzy inference rule R4 as shown in FIG. 3, i.e. s ₁ and.
The inference result of this depth is given to the spark detection means 7 together with the current position p of the metal material 1. The spark detection means 7 outputs data specified by the position of the flaw and its depth to a printer or the like. Therefore, by using the data of this flaw for product inspection, the flaw can be easily found and can be used as a measure for repair.

Note that the fuzzy inference is not limited to the above-described embodiment, and the form of the membership function,
Of course, the number can be set according to the actual electroplating conditions.

[0019]

As is apparent from the above description, according to the present invention, a spark is generated at the contact portion between the metal material and the electrode roll when the joint portion of the metal material passes through the electrode roll. Since the depth and location of the flaw can be predicted with considerable certainty, this data can be used in product inspection, and the flaw can be reliably detected and repaired easily.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram showing an embodiment of a spark detection device according to the present invention.

FIG. 2 is an explanatory diagram of a membership function used for fuzzy inference.

FIG. 3 is an explanatory diagram showing a result of fuzzy inference for estimating a depth of a flaw.

FIG. 4 is a block diagram showing an example of a conventional spark detection device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Metal material 2 Plating electrode roll pair 3 Plating cell 4 Plating power supply 5 Ammeter 6 Current change detection means 7 Spark detection means 8 Speed detection roll pair 9 Speed detector 10 Moving speed detection means 11 Position detection means 12 Fuzzy inference means

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) C25D 21/12 C25D 7/06

Claims (1)

(57) [Claims] An apparatus for detecting the occurrence of a spark when applying a constant current while moving a metal material in close contact with a pair of electrode rolls for electroplating, wherein the plating detects the current flowing through the electrode rolls. Current detecting means, current change detecting means for detecting a time change of the plating current from the output of the plating current detecting means, speed detecting means for detecting a moving speed of the metal material, and a metal material based on the moving speed of the metal material. Position detecting means for calculating a position,
Fuzzy inference means for inferring the depth of a flaw that would have been on the metal material surface due to spark generation based on the time change of the plating current and the moving speed of the metal material, and the fuzzy inferred flaw depth and the position of the metal material. And a spark detecting means for detecting the occurrence of scratches due to sparks.