WO2017212809A1 - Terminal crimp quality evaluation device and quality evaluation method - Google Patents

Abstract

This quality evaluation device (100) comprises: a pressure waveform acquisition unit (40) for acquiring the pressure waveform during terminal crimping; an area surface size calculation unit (50) dividing the pressure waveform into a plurality of areas to calculate an area surface size for each of the areas; an evaluation area extraction unit (60) using one or more areas contained in the plurality of areas to extract an evaluation area; a threshold value setting unit (70) setting a threshold value on the basis of a mean value and a standard deviation of the area surface sizes from a plurality of quality product samples and defective product samples in the evaluation area; and an evaluation unit (80) for comparing, when the terminal crimping of a product being examined has been performed by a terminal crimping device, the threshold value and the area surface size in the evaluation area of the product being examined, thereby determining the quality of the terminal crimping on the product being examined.

Description

Terminal crimping quality determination device and quality determination method

The present invention relates to a terminal crimping pass / fail judgment device and a pass / fail judgment method for judging whether or not a terminal is satisfactorily crimped to an end of an electric wire.

Conventionally, a terminal crimping device for crimping a terminal to an end of an electric wire is used in the manufacture of a wire harness or the like. An electric wire with a terminal whose terminal is not crimped well is a defective product. Therefore, it is desired to inspect whether or not the terminal is favorably crimped during terminal crimping. Therefore, conventionally, a terminal crimping quality determination device that determines whether or not a terminal is crimped to the end of an electric wire has been used. Patent Document 1 describes a pass / fail judgment device that detects a pressure received by a base plate of a terminal crimping device with a pressure sensor and judges pass / fail of terminal crimping based on the pressure. In the pass / fail judgment apparatus, first, a non-defective product generates a pressure waveform representing a change in pressure with respect to the elapsed time of the terminal crimping process. Hereinafter, this pressure waveform is referred to as a reference waveform. Then, the pressure waveform of the inspection product (hereinafter referred to as inspection waveform) is compared with the reference waveform. Whether the inspection product is a non-defective product or a defective product is determined based on whether or not the difference is larger than a predetermined value.

In the above pass / fail judgment device, the reference waveform is set as follows. That is, first, a pressure waveform is prepared when good terminal crimping is performed using a plurality of electric wires and terminals. Hereinafter, the wire and the terminal in which the terminal crimping is performed well are referred to as a non-defective sample. Then, an average of the pressure waveforms of the plurality of non-defective samples is used as a reference waveform. When the standard deviation is σ, ± 3σ is set as a threshold from the reference waveform. At the time of inspection, the pressure waveform is sampled at predetermined time intervals. When the ratio of the number of times exceeding the threshold to the total number of samplings exceeds the allowable value, it is determined as defective.

Patent Document 2 describes a quality determination device that takes into account a plurality of modes of defective crimping. In the quality determination apparatus described in Patent Document 2, one point in time at which the difference between the inspection waveform and the reference waveform is the largest is specified in advance for each crimping failure mode. Then, at each of a plurality of time points corresponding to a plurality of modes, it is determined whether or not the difference between the pressure value of the inspection waveform and the pressure value of the reference waveform exceeds a predetermined value (tolerance). Patent Document 2 lists 6σ (X) / AVE (X) and (good product average−defective product average) / 2 as specific examples of tolerances. Note that σ (X) and AVE (X) are the standard deviation and the average value of the non-defective samples, respectively.

JP 2005-135820 A JP 2014-56796 A

In the pass / fail judgment devices of Patent Documents 1 and 2, the value (threshold value) that serves as a reference for pass / fail judgment is set based on how far the average value of the non-defective samples is deviated. In setting the threshold, variations in defective products are not taken into consideration. However, even if the average value of defective products is the same, if the degree of variation of defective products is different, the possibility that a defective product is regarded as a non-defective product will be different.

FIG. 27A and FIG. 27B are diagrams showing an example in which the average value of good products, the standard deviation of good products, and the average value of defective products are the same, but the standard deviation of defective products is different. 27B has a larger variation of defective products and a larger standard deviation than FIG. 27A. Since the threshold value is set based on the average value of the non-defective products, or the average value and the standard deviation, the threshold values are the same in the examples of FIGS. 27A and 27B. Here, the tolerance D = (average of good products−average of defective products) / 2. However, the same applies to the tolerance D = 6σ (X) / AVE (X). In this example, it can be seen that the ratio of the defective product in FIG. 27B exceeds the threshold is higher than that of the defective product in FIG. 27A. That is, it can be seen that FIG. 27B has a higher possibility that a defective product is mistaken for a non-defective product. The conventional quality determination device cannot take into account such a difference in variation.

The present invention has been made in view of such a point, and an object of the present invention is to provide a terminal crimping pass / fail judgment device and a pass / fail judgment method capable of performing non-defective product judgment with higher accuracy.

A terminal crimping quality determination device according to the present invention includes a pressure waveform acquisition unit that acquires a pressure waveform representing a relationship between a degree of progress of terminal crimping by a terminal crimping device and a pressure generated in the terminal crimping device, and the pressure waveform. An area area calculation unit that calculates an area area, which is an area of a portion surrounded by the pressure waveform, divided into a plurality of areas according to the progress of terminal crimping, and one or two included in the plurality of areas A determination area extraction unit that extracts a determination area using the above areas, an average value and a standard deviation of a plurality of non-defective samples in the determination area, and an average of the area areas of a plurality of defective samples in the determination area When the terminal crimping of the inspection product is performed by the terminal crimping device, the threshold setting unit that sets the threshold based on the value and the standard deviation, By comparing the area area in the judgment area of the serial test article and the said threshold value, and a, and a judging section that judges acceptability of terminal crimping of the article to be inspected.

According to the pass / fail determination apparatus, the threshold value used as a criterion for determining whether or not terminal crimping is good is the average value and standard deviation of the area areas of the non-defective samples in the determination area, and the area area of the multiple defective samples in the determination area. It is set based on the average value and the standard deviation. Therefore, the threshold value can be set in consideration of the non-defective product variation in addition to the non-defective product average value, non-defective product variation, and defective product average value. Therefore, it is possible to reduce the possibility of misidentifying a defective product as a non-defective product, and to perform quality determination with higher accuracy.

In addition, the above pass / fail judgment device performs calculation for each area having a certain width, not at one point in time of the pressure waveform. Therefore, it is not necessary to perform calculation for every enormous number of time points, and the calculation time can be shortened. In addition, since it is difficult to be affected by sudden disturbance (noise), it is possible to suppress a decrease in determination accuracy due to the disturbance.

According to a preferred aspect of the present invention, the determination area extraction unit includes an average value and standard deviation of area areas of non-defective samples in each area, and an average value and standard deviation of area areas of defective samples in each area, And an effective area selection unit that selects one or more areas from the plurality of areas as an effective area, and a determination area determination unit that determines the determination area using the effective area. ing.

According to the above aspect, based on the average value and the standard deviation of the area areas of the non-defective samples and the defective samples, it is possible to select an area where it is easy to distinguish between non-defective products and defective products as an effective area. The determination area can be determined using the selected effective area. Therefore, a good product and a defective product can be distinguished relatively clearly, and a quality determination with high accuracy can be performed.

According to a preferred aspect of the present invention, the determination area extraction unit includes an average value μ _{OK of} the non-defective sample area area in each area, a standard deviation σ _{OK of} the non-defective sample area area, an area area of the defective sample, It has an area-specific calculation unit that calculates an average value μ _NG and a standard deviation σ _NG of the area area of the defective product sample, and the effective area selection unit is μ _NG when m and n are one or more real numbers. _{> μ OK B = (μ NG} -n × σ NG) when the _{- (μ OK + m × σ} OK), _{or, μ OK> μ B = (} μ OK -n × σ OK) when the _NG - ( One or more areas are selected as effective areas from the plurality of areas based on a first variable whose value increases as (μ _NG + m × σ _NG ) increases.

According to the above aspect, an area where it is easy to distinguish good products from defective products can be selected as an effective area. A non-defective product and a defective product can be distinguished relatively clearly, and high-accuracy determination can be performed.

According to a preferred aspect of the present invention, the first variable is the degree of separation when μ _NG > μ _OK = [(μ _NG −n × σ _NG ) − (μ _OK + m × σ _OK )] / (μ _NG− μ _OK ) or separation degree when μ _OK > μ _NG = [(μ _OK −n × σ _OK ) − (μ _NG + m × σ _NG )] / (μ _OK −μ _NG ).

According to the above aspect, an area where it is easy to distinguish good products from defective products can be selected as an effective area. A non-defective product and a defective product can be distinguished relatively clearly, and high-accuracy determination can be performed.

According to a preferred aspect of the present invention, m = n = 3.

According to the above aspect, an area where it is easy to distinguish good products from defective products can be selected as an effective area. A non-defective product and a defective product can be distinguished relatively clearly, and high-accuracy determination can be performed.

According to a preferred aspect of the present invention, the effective area selection unit is configured to select one area having the largest value of the first variable as the effective area among the plurality of areas. The determination area determination unit is configured to use the effective area as a determination area.

According to the above aspect, the determination area can be determined quickly and easily.

According to a preferred aspect of the present invention, the effective area selection unit is configured to select two or more areas having a larger value of the first variable as the effective area among the plurality of areas. . The determination area determination unit is configured to determine a determination area using the two or more effective areas.

According to the above aspect, since the determination area is determined using a plurality of effective areas, a more advanced determination is possible.

According to a preferred aspect of the present invention, the effective area selection unit, in the plurality of areas, at least in the order of the value of the first variable, or in the order of small standard deviation of the area area of the defective product sample, It is configured to select two or more areas as effective areas. The determination area determination unit weights the selected effective area, and then determines a determination area from the combination area and a combination area generation unit that generates a combination area formed by combining the weighted effective areas. A combination area determination unit.

According to the above aspect, it is possible to perform better quality determination.

According to a preferred aspect of the present invention, the effective area selection unit includes at least a first area having a value of the first variable, a second area having the second largest value, and a third value among the plurality of areas. The third area larger than the second area is selected as the effective area. The determination area determination unit includes a combination area obtained by adding the first area and the second area, a combination area obtained by adding the first area and the third area, the second area, and the third area. A combination area creating unit for creating a plurality of combination areas including a combination area obtained by adding together the first area, the second area, and the third area, and a non-defective product in each combination area Based on the average value and standard deviation of the area area of the sample and the average value and standard deviation of the area area of the defective sample in each combination area, any one of the plurality of combination areas is used as a determination area And an area determination unit.

According to the above aspect, by appropriately adding a plurality of effective areas, it is possible to create an area in which a good product and a defective product can be more easily distinguished than each effective area. It can be. Therefore, the good product and the defective product can be more clearly distinguished, and the quality determination can be performed with higher accuracy.

According to a preferred aspect of the present invention, the determination area determination unit includes an average value μ _{OK of} the non-defective sample area area in each combination area, a standard deviation σ _{OK of} the non-defective sample area, and an area area of the defective sample. It has an average value mu _NG, and a different combination area calculation unit that calculates a standard deviation sigma _NG with an area of defective samples. The combination area determination unit sets B = (μ _NG −r when μ _NG > μ _OK among a plurality of combination areas created by the combination area creation unit when r and s are real numbers of 1 or more. × σ _NG ) − (μ _OK + s × σ _OK ) or B = (μ _OK −r × σ _OK ) − (μ _NG + s × σ _NG ) when μ _OK > μ _NG The combination area having the largest value of the two variables is configured as the determination area.

According to the above aspect, among the plurality of combination areas, a combination area that is more easily distinguished from a non-defective product and a defective product can be set as the determination area. Therefore, a good product and a defective product can be clearly distinguished, and a higher-accuracy determination can be made.

According to a preferred aspect of the present invention, the second variable is the degree of separation when μ _NG > μ _OK = [(μ _NG −r × σ _NG ) − (μ _OK + s × σ _OK )] / (μ _NG− μ _OK ), or separation degree when μ _OK > μ _NG = [(μ _OK −r × σ _OK ) − (μ _NG + s × σ _NG )] / (μ _OK −μ _NG ).

According to the above aspect, the non-defective product and the defective product can be clearly distinguished, and the quality determination can be performed with higher accuracy.

According to a preferred aspect of the present invention, r = s = 3.

According to the above aspect, the non-defective product and the defective product can be clearly distinguished, and the quality determination can be performed with higher accuracy.

According to a preferred aspect of the present invention, the combination area creation unit weights at least the first area, the second area, and the third area when creating the plurality of combination areas. It is configured.

According to the above aspect, when creating a combination area, it is possible to create a combination area that is more easily distinguished from non-defective products and defective products. Therefore, a good product and a defective product can be clearly distinguished, and a higher-accuracy determination can be made.

According to a preferred aspect of the present invention, the combination area creation unit is configured to perform weighting such that the weighting increases as the standard deviation decreases.

According to the above aspect, it is possible to create a combination area where a good product and a defective product are more easily distinguished.

According to a preferred aspect of the present invention, the combination area creating unit is configured to perform weighting such that the weighting increases as the degree of separation increases.

According to the above aspect, it is possible to create a combination area where a good product and a defective product are more easily distinguished.

According to a preferred aspect of the present invention, the determination area extraction unit sets one or more areas as the effective area in order of increasing standard deviation value of the area area of the defective product among the plurality of areas. An effective area selection unit to select; and a determination area determination unit that determines the determination area using the effective area.

According to the above aspect, an area with less variation in the distribution of defective samples can be selected as an effective area. Thereby, a favorable quality determination can be performed.

According to a preferred aspect of the present invention, the threshold setting unit includes an average value area μd _OK of non-defective samples in the determination area, an area area standard deviation σd _OK of non-defective samples, and an average area area of defective samples. A determination area calculation unit for calculating the value μd _NG and the standard deviation σd _NG of the area area of the defective product sample, and when p and q are real numbers of 1 or more, μd _NG > μd _OK and (μd _NG −p × A predetermined value between (μd _NG −p × σd _NG ) and (μd _OK + q × σd _OK ) when σd _NG )> (μd _OK + q × σd _OK ), or μd _OK > μd _NG and (μd Threshold value that determines a predetermined value as a threshold value between (μd _OK −p × σd _OK ) and (μd _NG + q × σd _NG ) when _OK− p × σd _OK )> (μ _NG + q × σd _NG ) Decision Part.

According to the above aspect, it is possible to set a threshold value that is easy to distinguish good products from defective products. By using such a threshold value, it is possible to clearly distinguish a non-defective product from a defective product, and it is possible to perform a quality determination with high accuracy.

According to a preferred aspect of the present invention, the terminal crimping apparatus includes an anvil on which an end of a wire and a terminal are placed, a crimper that can approach and separate from the anvil, and the crimper in the anvil. And an electric actuator for approaching and separating from the actuator. The progress of the terminal crimping is any one of elapsed time, the position of the crimper, the travel distance of the crimper, the speed of the crimper, the acceleration of the crimper, the current supplied to the actuator, and the operating position of the actuator. It is.

According to the above aspect, a good pressure waveform can be obtained, and whether or not the terminal crimping is good can be suitably determined. In addition, each pressure waveform when a good product sample, a defective product sample, and a test product are crimped to terminals can be accurately aligned and compared according to the degree of progress of terminal crimping. The determination can be suitably performed.

In the terminal crimping quality determination method according to the present invention, a terminal crimping apparatus performs terminal crimping of a plurality of non-defective samples and a plurality of defective samples, and the degree of progress of terminal crimping of each sample and the pressure generated in the terminal crimping apparatus. A pressure waveform representing the relationship between the pressure waveform and the pressure waveform of each sample is divided into a plurality of areas according to the degree of terminal crimping, and an area of an area surrounded by the pressure waveform is determined for each area. Calculating the average value and standard deviation of the area areas of the non-defective samples in each area, calculating the average value and standard deviation of the area areas of the defective samples in each area, and the non-defective samples and Based on the average value and standard deviation of the area area of the defective sample, one or two or more from the plurality of areas A step of selecting the area as an effective area, a step of extracting a determination area using the effective area, and a threshold based on an average value and a standard deviation of the area areas of the good sample and the defective sample in the determination area And determining the quality of the terminal crimping of the inspection product by comparing the area of the inspection product in the determination area with the threshold value. And steps.

According to the present invention, it is possible to provide a terminal crimping quality determination device and a quality determination method capable of performing a quality determination with higher accuracy.

FIG. 1 is a front view of the terminal crimping apparatus. FIG. 2 is a side view of the terminal crimping apparatus. FIG. 3 is a block diagram of the quality determination device. FIG. 4 is a functional block diagram of the CPU. FIG. 5 is a flowchart of a threshold setting method. FIG. 6A is a plan view of a non-defective sample. FIG. 6B is a plan view of a defective product sample. FIG. 6C is a plan view of another defective sample. FIG. 6D is a plan view of another defective sample. It is a figure showing a pressure waveform and a divided area. It is a figure showing the normal distribution data of each sample in each area. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A1. It is a figure showing the normal distribution data of the non-defective product sample and the defective product sample in area A2. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A3. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A4. It is a figure showing the normal distribution data of the non-defective product sample and the defective product sample in area A5. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A6. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A7. It is a figure showing the normal distribution data of the non-defective product sample and the defective product sample in area A8. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A9. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in area A10. It is explanatory drawing of A and B of separation degree C = B / A. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in combination area A5 + A6. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in combination area A6 + A7. It is a figure showing the normal distribution data of the non-defective product sample and defective product sample in combination area A5 + A7. It is a figure showing the normal distribution data of the non-defective product sample and the defective product sample in combination area A5 + A6 + A7. It is a figure showing the threshold value in a determination area. It is a flowchart of the quality determination method of the terminal crimping which concerns on embodiment. It is a flowchart of the quality determination method of the terminal crimping which concerns on other embodiment. It is a figure showing the relationship between the probability distribution of a good article and defective goods, and a threshold value. It is a figure showing the relationship between the probability distribution of a good article and inferior goods, and a threshold value.

Hereinafter, embodiments of the present invention will be described. The terminal crimping quality determination device according to the present embodiment (hereinafter simply referred to as “quality judgment device”) indicates whether or not the terminal is crimped well when the terminal crimping device crimps the terminal to the end of the electric wire. It is a device for judging.

(Configuration of terminal crimping device)
First, the configuration of the terminal crimping device 1 will be described with reference to FIGS. 1 and 2. FIG. 1 is a front view of the terminal crimping device 1, and FIG. 2 is a side view of the terminal crimping device 1. The terminal crimping device 1 includes an anvil 2, a crimper 3 that can be raised and lowered, and a motor 4 that raises and lowers the crimper 3. The motor 4 is configured by a servo motor that can detect the rotational position of the motor 4. However, the type of the motor 4 is not particularly limited, and may be a motor other than the servo motor. The motor 4 is an example of an electric actuator that moves the crimper 3 toward and away from the anvil 2. However, the actuator is not limited to the motor 4. Any actuator that can move the crimper 3 up and down can be used as the actuator.

As shown in FIG. 1, a terminal supply device 5 is arranged on the side of the anvil 2. The terminal supply device 5 is configured to sequentially supply a terminal group 11 formed by connecting a plurality of terminals 10 toward the anvil 2. When the crimper 3 is lowered with the electric wire 12 and the terminal 10 disposed between the crimper 3 and the anvil 2, the terminal 10 is separated from the adjacent terminal 10 by the crimper 3 and is crimped to the electric wire 12. In the present embodiment, the motor 4, the crimper 3, and the anvil 2 constitute a crimping portion that crimps the terminal 10 to the electric wire 12 by applying pressure to the terminal 10.

When the terminal 10 is crimped to the end of the electric wire 12, pressure is generated in the crimped portion. In the present embodiment, a pressure sensor 21 that detects the pressure generated in the crimping portion is provided. Here, the pressure sensor 21 is configured by a piezoelectric element connected to the anvil 2. However, the specific configuration of the pressure sensor 21 is not limited at all. Not only a piezoelectric element but any sensor capable of detecting pressure can be suitably used. The pressure sensor 21 may be directly connected to the anvil 2, but is indirectly connected in the present embodiment. That is, another member is interposed between the pressure sensor 21 and the anvil 2. Unless specifically limited in the present specification, the “connection” includes both a case of being directly connected and a case of being indirectly connected via another member. Moreover, the pressure sensor 21 should just be arrange | positioned so that the pressure which generate | occur | produces in any location of a crimping | compression-bonding part may be detected, and does not necessarily need to be connected to the anvil 2. The pressure sensor 21 may be connected to the crimper 3, for example. Moreover, as long as the pressure which generate | occur | produces with terminal crimping is detectable, you may connect to places other than a crimping | compression-bonding part.

The pressure sensor 21 is electrically connected to the controller 20. Note that “electrically connected” means connected so as to be communicable by wire or wirelessly. The controller 20 may be a microcomputer built in the terminal crimping apparatus 1 or a computer (for example, a personal computer) disposed outside the terminal crimping apparatus 1. In the present embodiment, the pressure sensor 21 and the controller 20 are connected by an electric wire. When pressure is received, the pressure sensor 21 is configured to output a voltage value or a current value corresponding to the magnitude of the pressure. The controller 20 detects the pressure generated in the crimping part based on the voltage or current value received from the pressure sensor 21.

As mentioned above, although the structure of the terminal crimping apparatus 1 was demonstrated, said terminal crimping apparatus 1 is only an example. The terminal crimping device to which the quality determination device according to the present invention is applied is not limited at all. The quality determination device according to the present invention is applicable to various known terminal crimping devices.

(Configuration of pass / fail judgment device)
Next, the configuration of the pass / fail determination apparatus 100 will be described. As shown in FIG. 3, the quality determination device 100 according to the present embodiment includes the pressure sensor 21, the A / D converter 22, the controller 20, and the display 26. The pass / fail determination apparatus 100 may further include an external memory 27 such as a hard disk. The controller 20 includes a CPU 30, a ROM 24, and a RAM 25. A servo amplifier 8 is connected to the motor 4, and the servo amplifier 8 is connected to a controller 20.

FIG. 4 is a functional block diagram of the CPU 30. Although details will be described later, the CPU 30 functions as a pressure waveform acquisition unit 40, an area area calculation unit 50, a determination area extraction unit 60, a threshold setting unit 70, and a determination unit 80 when executing processing to be described later. The determination area extraction unit 60 includes an effective area selection unit 61, an area calculation unit 62, and a determination area determination unit 63. The determination area determination unit 63 includes a combination area creation unit 64, a combination area calculation unit 65, and a combination area determination unit 66. The threshold setting unit 70 includes a determination area calculation unit 71 and a threshold determination unit 72. Since each of the above units is realized by the CPU 30, the pressure waveform acquisition unit 40, the area area calculation unit 50, the determination area extraction unit 60, the threshold setting unit 70, and the determination unit 80 are respectively a pressure waveform acquisition processor, an area area calculation processor, and a determination. It can be read as an area extraction processor, a threshold setting processor, or a determination processor. The effective area selection unit 61, the area calculation unit 62, and the determination area determination unit 63 can be read as an effective area selection processor, an area calculation processor, and a determination area determination processor, respectively. The combination area creation unit 64, the combination area calculation unit 65, and the combination area determination unit 66 can be read as a combination area generation processor, a combination area calculation processor, and a combination area determination processor, respectively. The determination area calculation unit 71 and the threshold value determination unit 72 can be read as a determination area calculation processor and a threshold value determination processor, respectively.

(Pass / fail judgment method)
Next, a quality determination method performed by the quality determination apparatus 100 will be described. The pass / fail determination apparatus 100 determines whether the terminal crimping is good or not based on whether or not a predetermined detection value (area area of a determination area described later) exceeds a predetermined threshold when the terminal crimping of the inspection product is performed. judge. First, a threshold setting method will be described.

FIG. 5 is a flowchart of a threshold setting method. First, in step S1, a plurality of sets of electric wires 12 and terminals 10 are prepared. In this specification, the wire 12 used for setting the threshold and the terminal 10 crimped to the end of the wire 12 are referred to as “sample”. In step S1, a plurality of non-defective samples and a plurality of defective samples are prepared.

A non-defective sample refers to a pair of electric wires 12 and terminals 10 that are well crimped. A defective product sample refers to a set of electric wires 12 and terminals 10 that are not crimped well. FIG. 6A represents a good sample. In the non-defective sample, the insulation barrel 10 a of the terminal 10 is crimped to the coating 12 a of the electric wire 12, and the wire barrel 10 b of the terminal 10 is crimped to the core 12 b of the electric wire 12.

6B to 6D show examples of defective samples. For example, as shown in FIGS. 6B to 6D, there are various modes of the crimping failure of the terminal 10. FIG. 6B shows a defective sample (so-called core wire spillage) in which some core wires 12b of the electric wires 12 are not crimped to the wire barrel 10b. FIG. 6C shows a defective sample (so-called core wire drop) in which the core wire 12b of the electric wire 12 is not crimped to the wire barrel 10b but is crimped to the insulation barrel 10a. FIG. 6D shows a defective sample (so-called coating bite) in which the coating 12a of the electric wire 12 is crimped to the wire barrel 10b. In this embodiment, a plurality of defective product samples are prepared for each aspect of defective crimping.

Next, in step S2, the terminal crimping process is performed on the non-defective product sample and the defective product sample using the terminal crimping device 1. And a pressure waveform is acquired about each sample (refer FIG. 7). Here, the pressure waveform is a waveform representing a change in the pressure P detected by the pressure sensor 21 with respect to the progress of the terminal crimping process by the terminal crimping apparatus 1. In the present embodiment, the elapsed time t of the terminal crimping process is used as a variable representing the progress of the terminal crimping process. However, the variable is not limited to the elapsed time t of the terminal crimping process as long as it can represent the progress of the terminal crimping process. In addition, when performing the process of step S2, CPU30 functions as the pressure waveform acquisition part 40. FIG. The acquired pressure waveform is stored in the RAM 25 as a sample waveform. The RAM 25 is an example of a sample waveform storage unit that stores pressure waveforms of a good product sample and a defective product sample. However, the sample waveform storage unit is not limited to the RAM 25 but may be the external memory 27 or the like.

In step S3, as shown in FIG. 7, the pressure waveform of each sample is divided into a plurality of areas for each elapsed time. Here, the pressure waveform is divided into ten areas A1, A2, A3, A4,. The number of area divisions is not particularly limited. The width of each area (time width in the present embodiment) may or may not be constant. The CPU 30 functions as a pressure waveform dividing unit (not shown) when dividing the pressure waveform. Next, CPU30 calculates the area (henceforth an area area) of the part enclosed by a pressure waveform for every area. For example, in area A4, the area of the region surrounded by point P1, point P2, point P3, and point P4 is the area area. The CPU 30 functions as the area area calculation unit 50 when calculating the area area of each area.

Next, it progresses to step S4 and the area area of the non-defective sample and the area area of the defective sample of each aspect are totaled for every area. In step S2, a plurality of non-defective sample pressure waveforms and a plurality of non-defective sample pressure waveforms are acquired. Therefore, data of probability distribution (normal distribution) as shown in FIG. 8 is obtained for each area for each of the non-defective product sample and the defective product sample of each aspect. For each area, the CPU 30 calculates the average value μ _{OK of} the non-defective samples, the standard deviation σ _OK of the non-defective samples, the average value μ _NG of the defective samples of each aspect, and the standard deviation σ _NG of the defective samples of each aspect. At this time, the CPU 30 functions as the area-specific calculation unit 62.

FIGS. 9 to 18 are diagrams comparing the normal distribution of the non-defective samples in the areas A1 to A10 with the normal distribution of the defective samples in a predetermined mode, respectively. When the difference between the area area of the non-defective sample and the area area of the defective sample is small, for example, as shown in FIG. 10, the normal distribution of the non-defective sample and the normal distribution of the defective sample tend to approach each other. On the other hand, when the difference between the area area of the non-defective sample and the area area of the defective sample is large, for example, as shown in FIG. 14, the normal distribution of the non-defective sample and the normal distribution of the defective sample tend to be separated. The larger the difference between the area area of the non-defective sample and the area area of the defective sample, the more the normal distribution of the defective sample is separated from the normal distribution of the non-defective sample. The portion where the normal distribution of the non-defective sample and the normal distribution of the defective sample overlap is a portion where it is difficult to distinguish whether it is a good product or a defective product. In other words, even a defective product may be misidentified as a non-defective product. On the other hand, when the normal distribution of non-defective samples and the normal distribution of defective samples are separated, it is easy to distinguish good products from defective products. Therefore, in the pass / fail determination apparatus 100 according to the present embodiment, an area where the normal distribution of the non-defective product sample and the normal distribution of the defective product sample are small is selected, and the pass / fail determination is performed using the selected area. Hereinafter, the area selected in this way is referred to as an effective area.

In step S5, the CPU 30 selects an effective area. An area where the normal distribution of the non-defective samples and the normal distribution of the defective samples are separated is suitable as an effective area. Therefore, an area where the normal distribution of the non-defective samples and the normal distribution of the defective samples are most separated may be selected as the effective area. However, in all areas, the normal distribution of non-defective samples and the normal distribution of defective samples may overlap. In that case, it is difficult to distinguish good products from defective products using only one area. However, by combining a plurality of areas, the overlap between the normal distribution of the non-defective samples and the normal distribution of the defective samples can be reduced or the overlap can be eliminated. Therefore, in the present embodiment, a plurality of areas are selected as effective areas so that there is no overlap between the normal distribution of the non-defective samples and the normal distribution of the defective samples, or the overlap is as small as possible. Consider the added area (hereinafter referred to as combination area).

In the present embodiment, of all the areas, the area having the highest degree of separation, the second largest area, and the third largest area are selected as effective areas. However, the method for selecting the effective area is not particularly limited. The degree of separation C is such that m = n is a real number (eg, a natural number) of 1 or more, and when μ _NG > μ _OK , C = B / A = [(μ _NG −m × σ _NG ) − ( μ _OK + n × σ _OK )] / (μ _NG −μ _OK ) (see FIG. 19), and when μ _OK > μ _NG , C = [(μ _OK −m × σ _OK ) − (μ _NG + n × σ _NG )] / (μ _OK −μ _NG ). In the present embodiment, m = n = 3. Here, areas A5, A6, and A7 are selected as effective areas. The CPU 30 functions as the effective area selection unit 61 when selecting an effective area.

Of course, when creating the combination area from the effective areas A5 to A7, it is possible to use the data of the effective areas A5 to A7 as they are. However, as can be seen from FIG. 13 to FIG. 15, among the effective areas A5 to A7, there are relatively large and small variations in the normal distribution of the non-defective samples or defective samples. Since data with less variation is less likely to cause errors, it is preferable for quality determination. Therefore, in the present embodiment, in step S6, when the effective areas A5 to A7 are combined, weighting is performed such that the smaller the data variation is, the larger the weight is. Assuming that the weighting coefficients of the effective areas A5, A6, and A7 are K5, K6, and K7, respectively, in this embodiment, K5 = k / σ5, K6 = k / σ6, and K7 = k / σ7 are set. Here, k is a constant value, and σ5, σ6, and σ7 are standard deviations of non-defective samples in the effective areas A5, A6, and A7, respectively. Note that σ5, σ6, and σ7 may be standard deviations of defective samples in the effective areas A5, A6, and A7, respectively. In the present embodiment, K5 <K6 <K7. However, the above weighting coefficient is an example, and the specific method of weighting is not particularly limited. For example, the effective area may be weighted so that the weighting increases as the degree of separation increases.

In step S7, a combination area is created by combining effective areas after weighting. Here, combination areas A5 + A6, A6 + A7, A7 + A5, A5 + A6 + A7 are created for the effective areas A5, A6, A7 after weighting. The CPU 30 functions as the combination area creation unit 64 when weighting the effective area and creating the combination area. Next, the CPU 30 calculates, for each combination area, the average value μ _{OK of} the non-defective samples, the standard deviation σ _OK of the non-defective samples, the average value μ _NG of the defective samples of each aspect, and the standard deviation σ _NG of the defective samples of each aspect. calculate. At this time, the CPU 30 functions as the combination area calculation unit 65. 20, FIG. 21, FIG. 22, and FIG. 23 are diagrams showing normal distributions of non-defective samples and defective samples in combination areas A5 + A6, A6 + A7, A5 + A7, and A5 + A6 + A7, respectively.

Next, the process proceeds to step S8, and a combination area that is most suitable for quality determination is determined as a determination area. In the present embodiment, the degree of separation of each of the combination areas A5 + A6, A6 + A7, A5 + A7, A5 + A6 + A7 created in step S7 is calculated, and the one with the largest degree of separation is set as the determination area. Here, the combination area with the highest degree of separation is A6 + A7. Therefore, the combination area A6 + A7 is a determination area. The CPU 30 functions as a combination area determination unit 66 when determining a determination area from a plurality of combination areas.

Next, proceeding to step S9, a threshold E is set from the normal distribution of the non-defective samples and the defective samples in the determination area (see FIG. 24). At this time, the CPU 30 functions as the threshold setting unit 70. The threshold value E can be a boundary value that separates a good product sample and a defective product sample. The specific setting method of the threshold value E is not limited at all. For example, the average value of the non-defective sample area area in the determination area is μd _OK , the standard deviation of the non-defective sample area is σd _OK , and the defective sample area area is the average value _{[mu] d NG,} a standard deviation with an area of defective samples is taken as .sigma.d _NG (Note that in this case CPU30 functions as a determination area calculation unit _{71.), (μd NG -p} × σd NG) And a predetermined value between (μd _OK + q × σd _OK ) can be set as the threshold value E. Note that p and q are one or more real numbers (for example, natural numbers). For example, a predetermined value between (μd _NG −3σd _NG ) and (μd _OK + 3σd _OK ) may be used as the threshold value E. Further, (μd _NG −p × σd _NG ) may be set as the threshold value E. For example, the threshold value E = μd _NG −3σd _NG may be set, or the threshold value E = μd _NG −6σd _NG may be set. The threshold E may be (μd _OK + q × σd _OK ). For example, the threshold value E = μd _OK + 3σd _OK may be set, or the threshold value E = μd _OK + 6σd _OK may be set. The CPU 30 functions as the threshold value determination unit 72 when determining the threshold value.

As described above, the threshold value E, which is a criterion for pass / fail judgment, is set. The set threshold value E is stored in the RAM 25 or the external memory 27. In the above step, the pressure waveform, the normal distribution of the non-defective product and the defective product in each area, the average value, the standard deviation, and the like may be displayed on the display 26.

The pass / fail judgment by the pass / fail judgment apparatus 100 is performed as follows (see FIG. 25). Here, the set of the electric wires 12 and the terminals 10 that are the objects of the pass / fail judgment are referred to as inspection products. The pass / fail determination apparatus 100 receives a signal from the pressure sensor 21 when the terminal crimping apparatus 1 performs terminal crimping of the inspection product, and acquires a pressure waveform of the inspection product (step S11). Then, this pressure waveform is divided into areas A1 to A10 (step S12). Next, the effective areas A5 to A7 are selected (step S13), and the effective areas A5 to A7 are weighted in the same manner as described above (step S14). Then, the determination area A6 + A7 is extracted, and the area area of the determination area A6 + A7 is set. Calculate (step S15). Then, the size of the area of the determination area A6 + A7 for the inspection product and the threshold value E are compared (step S16). If the area of the inspection product determination area A6 + A7 exceeds the threshold value E, the inspection product is determined to be defective (step S17). On the other hand, if the area of the inspection product determination area A6 + A7 is equal to or less than the threshold value E, the inspection product is determined to be a non-defective product (step S18). The CPU 30 functions as the determination unit 80 when performing the processes of steps S11 to S18.

Here, in the determination area A6 + A7, since the average value of the area area of the non-defective sample is smaller than the average value of the area area of the defective sample (see FIG. 24), the area area of the determination area A6 + A7 of the inspection product is the threshold value E. The inspection product is determined to be a non-defective product in the following cases. However, if the average value of the area area of the non-defective sample is larger than the average value of the area area of the defective sample in the judgment area, the inspection product is good if the area area of the judgment area of the inspection product is greater than or equal to the threshold value. If it is smaller than the threshold value, it is determined as a defective product.

The determination result is displayed on the display 26, for example. Thereby, the user can easily recognize whether or not the terminal crimping of the inspection product has been performed satisfactorily. The determination result notification method is not limited to display on the display 26. For example, the pass / fail determination device 100 may include a speaker and output a sound from the speaker to notify the pass / fail determination result of terminal crimping.

(Effect of embodiment)
As described above, according to the quality determination device 100 according to the present embodiment, the terminal crimping device 1 performs terminal crimping of a plurality of non-defective samples and a plurality of defective samples, and the threshold value is automatically set based on the data. Set. When a person manually sets a threshold based on intuition or experience, the accuracy of pass / fail judgment depends on the intuition and experience of the person, and human error may occur. However, according to the quality determination apparatus 100 according to the present embodiment, such a human error does not occur because the threshold is automatically set based on objective data. In addition, if a person manually sets a threshold based on intuition or experience, if a new singularity (a point where it is easy to distinguish the difference between a good product and a defective product) occurs in the data, use that singularity. I can't. However, according to the pass / fail determination apparatus 100 according to the present embodiment, the threshold value is automatically set based on the data of the non-defective product sample and the defective product sample, so that it is possible to use a newly generated singularity.

According to the pass / fail judgment apparatus 100 according to the present embodiment, the threshold value E, which is a criterion for pass / fail judgment of terminal crimping, is the average value and standard deviation of the area areas of a plurality of non-defective samples in the judgment area, and a plurality of faults in the judgment area. It is set based on the average value and standard deviation of the area area of the non-defective samples. Therefore, the threshold value E can be set in consideration of the variation of defective products in addition to the average value of non-defective products, the variation of non-defective products, and the average value of defective products. Therefore, it is possible to reduce the possibility of misidentifying a defective product as a non-defective product, and to perform quality determination with higher accuracy.

Although a specific value of the threshold value E is not particularly limited, a predetermined value between (μd _OK −p × σd _OK ) and (μd _NG + q × σd _NG ) is set as the threshold value E in the present embodiment. Thereby, the quality determination with high accuracy can be performed.

Moreover, the quality determination apparatus 100 according to the present embodiment calculates the average value and the standard deviation of the non-defective product samples and the defective product samples for each area having a certain width. Therefore, the prior art (for example, see the above-mentioned Patent Document 2) that specifies a time (in other words, a single time point) where the difference between the non-defective product sample and the defective product sample is the largest, and uses a section around that time as a determination area. ) Has the following advantages. First, in this embodiment, an area area is calculated for each predetermined number of predetermined areas, and a difference between a pass / fail sample and a defective product sample is calculated based on the area area. Therefore, it is not necessary to perform calculation for every enormous number of time points as in the above-described conventional technique for calculating a difference between a good product sample and a defective product sample every elapsed time. Therefore, calculation time can be shortened. Next, in the above prior art, for example, in a defective product sample, a pressure change occurs suddenly due to disturbance (noise), and at a time different from the original time when the difference between the non-defective sample and the defective product sample is the largest, The difference between the good product sample and the defective product sample may be the largest. In that case, the accuracy of the pass / fail determination may be reduced due to the disturbance. However, in the present embodiment, since the difference between the non-defective product sample and the defective product sample is calculated based on the area area, such disturbance can be reduced. Therefore, a decrease in determination accuracy due to disturbance is unlikely to occur.

According to the pass / fail judgment apparatus 100 according to the present embodiment, the pressure waveform is divided into a plurality of areas A1 to A10, and the areas A5 to A7 in which the good and defective products can be easily distinguished among the areas A1 to A10 are effective areas. Choose as. Then, the determination area is determined using the effective areas A5 to A7. Therefore, a good product and a defective product can be distinguished relatively clearly, and a quality determination with high accuracy can be performed.

In addition, when there is only one area for the determination area for one type of defect, it is not possible to effectively use a singular point of a defective product that appears in another area (a point where the difference from a good product can be easily distinguished). In this embodiment, since the plurality of areas A5 to A7 are selected as effective areas, a plurality of singular points appearing in the plurality of areas A5 to A7 can be used effectively.

In addition, according to the quality determination device 100 according to the present embodiment, the areas A5 to A7 having a high degree of separation are selected as the effective areas from the plurality of areas A1 to A10. Therefore, a good product and a defective product can be clearly distinguished, and a high / defective judgment can be performed with high accuracy.

Further, according to the quality determination device 100 according to the present embodiment, a plurality of combination areas are created by appropriately adding the effective areas A5 to A7, and the average value and standard deviation of the area areas of the non-defective samples in each combination area are calculated. The determination area is determined based on the average value and standard deviation of the area areas of the defective product samples in each combination area. Since it is possible to use the judgment area as an area where it is easier to distinguish between non-defective products and defective products than the effective areas A5 to A7, it is possible to more clearly distinguish between non-defective products and defective products, and pass / fail with higher accuracy. Judgment can be made.

According to the pass / fail determination apparatus 100 according to the present embodiment, an area having a high degree of separation is selected as a determination area from among a plurality of combination areas. Therefore, a good product and a defective product can be clearly distinguished, and a high / defective judgment can be performed with high accuracy.

Further, according to the pass / fail determination apparatus 100 according to the present embodiment, when combining the effective areas A5 to A7, weighting is performed so as to create a combination area that can more clearly distinguish non-defective products from defective products. . According to the quality determination apparatus 100 according to the present embodiment, weighting is performed such that the weighting increases as the standard deviation decreases, or the weighting increases as the separation degree increases. Therefore, a more suitable combination area can be created, and thus a more suitable determination area can be obtained. Therefore, it is possible to perform pass / fail judgment with higher accuracy.

Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and the present invention can be implemented in various other forms. Next, examples of other embodiments will be briefly described.

(Other embodiments)
In the embodiment, in step S5, three effective areas A5 to A7 are selected from all the areas A1 to A10, and a determination area is extracted from a plurality of combination areas obtained by appropriately combining these effective areas A5 to A7. It was. However, as described above, of all the areas A1 to A10, one area where the normal distribution of the non-defective samples and the normal distribution of the defective samples are most separated is selected as the effective area, and the effective area is determined as the determination area. May be extracted as Further, a variable (hereinafter referred to as a third variable) for indicating the degree of separation between the normal distribution of the non-defective samples and the normal distribution of the defective samples is determined, and an area where the value of the third variable is equal to or greater than a predetermined value is selected as an effective area. May be. An area having the largest value of the third variable (in the case where there is one effective area where the value of the third variable is equal to or greater than a predetermined value) may be extracted as the determination area.

Also, as shown in FIG. 26, after step S4, it is determined whether there is an area where the value of the third variable is equal to or greater than a predetermined value (step S4A). If the determination result is YES, the value of the third variable May be selected as an effective area (step S4B), and the effective area may be set as a determination area (step S4C). If the determination result of step S4A is NO, the processes after step S5 may be performed. The above process can be performed by the CPU 30.

In the above embodiment, three areas are selected as effective areas. However, the number of effective areas to be selected is not limited to three. The number of effective areas may be four or more. Even in this case, an area obtained by appropriately adding two or more effective areas can be used as a combination area.

In the above-described embodiment, the elapsed time t of the terminal crimping process is used as a variable representing the progress of terminal crimping. However, the variable representing the progress of the terminal crimping process is not limited to the elapsed time t. Any variable that can uniquely identify the progress of the terminal crimping process can be used. Terminal crimping is performed by moving the crimper 3. The crimper 3 is driven by a motor 4. For this reason, for example, based on the position or moving distance of the crimper 3, the speed of the crimper 3, the acceleration of the crimper 3, the current supplied to the motor 4, or the rotational position of the motor 4 (actuation position of the actuator), Can be uniquely identified.

In the embodiment, the first variable used when selecting the effective area is the degree of separation. However, the first variable is not limited to the degree of separation. As the first variable, an arbitrary variable indicating the degree of separation of the normal distribution of the area area between the non-defective product sample and the defective product sample can be used. For example, as the first variable, when m and n are real numbers of 1 or more, B = (μ _NG −n × σ _NG ) − (μ _OK + m × σ _OK ) when μ _NG > μ _OK , or A variable whose value increases as B = (μ _OK −n × σ _OK ) − (μ _NG + m × σ _NG ) when μ _OK > μ _NG can be used.

In the embodiment, the second variable used when determining the determination area is the degree of separation. However, the second variable is not limited to the degree of separation. As the second variable, an arbitrary variable indicating the degree of separation of the normal distribution of the area area between the non-defective product sample and the defective product sample can be used. For example, when r and s are real numbers of 1 or more as the second variable, B = (μ _NG −r × σ _NG ) − (μ _OK + s × σ _OK ) when μ _NG > μ _OK , or A variable whose value increases as B = (μ _OK −r × σ _OK ) − (μ _NG + s × σ _NG ) when μ _OK > μ _NG can be used.

The method for selecting an effective area is not limited to the method of the above embodiment. The effective area selection unit may be configured to select one or two or more areas as the effective area from the plurality of areas in ascending order of the standard deviation value of the area area of the defective product sample. The determination area determination unit may be configured to determine the determination area using these effective areas. As a result, an area with less variation in the distribution of defective product samples can be selected as an effective area, and a good / bad determination can be made.

When selecting an effective area, select a plurality of areas in descending order of the value of the first variable, and further select one or more areas from the plurality of areas in ascending order of σ _NG. These may be effective areas. In other words, when a plurality of areas are selected in descending order of the value of the first variable and n is a real number of 1 or more, | (μ _NG + n × σ _NG ) − (μ _NG One or two or more areas may be selected in order from the smallest value of −n × σ _NG ) |, and these may be used as effective areas. As a result, an area in which the probability distribution between the non-defective product sample and the defective product sample is large and the variation in the distribution of the defective product sample is smaller can be selected as an effective area, and a good / non-defective determination can be performed. .

As described above, weighting may or may not be performed when combining effective areas. When performing weighting, the specific method is not particularly limited. The effective area selection unit selects at least two areas as the effective area from the plurality of areas in order of at least the first variable value or in ascending order of standard deviation of the area area of the defective product sample. It may be configured. The combination area creation unit is configured to create a combination area formed by combining the weighted effective areas after weighting the selected effective area, and the combination area determination unit determines the determination area from the combination areas. May be configured to determine.

Note that the above-described embodiments can be appropriately combined.

DESCRIPTION OF SYMBOLS 1 Terminal crimping apparatus 2 Anvil 3 Crimper 4 Motor 10 Terminal 12 Electric wire 20 Controller 21 Pressure sensor 40 Pressure waveform acquisition part 50 Area area calculation part 60 Judgment area extraction part 70 Threshold setting part 80 Judgment part 100 Pass / fail judgment apparatus

Claims (19)

1. A pressure waveform acquisition unit that acquires a pressure waveform representing a relationship between a progress of terminal crimping by the terminal crimping device and a pressure generated in the terminal crimping device;
   An area area calculation unit that divides the pressure waveform into a plurality of areas according to the degree of progress of terminal crimping, and calculates an area area that is an area surrounded by the pressure waveform for each area;
   A determination area extraction unit that extracts a determination area using one or more areas included in the plurality of areas;
   A threshold setting unit that sets a threshold based on an average value and standard deviation of area areas of a plurality of non-defective samples in the determination area, and an average value and standard deviation of area areas of a plurality of defective samples in the determination area; ,
   When the terminal crimping of the inspection product is performed by the terminal crimping device, the determination unit that determines whether the inspection product is crimped by comparing the area with the threshold value in the determination area of the inspection product. When,
   A terminal crimping quality determination device comprising:
2. The determination area extraction unit
   One or more areas from the plurality of areas based on the average value and standard deviation of the area areas of the non-defective samples in each area and the average value and standard deviation of the area areas of the defective samples in each area An effective area selection section for selecting as an effective area,
   A determination area determining unit that determines the determination area using the effective area;
   The terminal crimping quality determination device according to claim 1, comprising:
3. The determination area extraction unit includes an average value μ _OK of non-defective samples in each area, a standard deviation σ _OK of non-defective sample areas, an average value of area areas of defective samples μ _NG , and An area-specific calculation unit that calculates the standard deviation σ _NG of the area area;
   The effective area selection unit is B = (μ _NG −n × σ _NG ) − (μ _OK + m × σ _OK ) when μ _NG > μ _OK , where m and n are real numbers of 1 or more, or _{, μ OK> μ B = (} μ OK -n × σ OK) when the _NG - based on _{(μ NG + m × σ NG} ) becomes large as the first variable value increases, one of said plurality of areas Or the quality determination apparatus of the terminal crimping of Claim 2 comprised so that two or more areas may be selected as an effective area.
4. The first variable is the degree of separation when μ _NG > μ _OK = [(μ _NG −n × σ _NG ) − (μ _OK + m × σ _OK )] / (μ _NG −μ _OK ), or μ _OK The degree of separation of terminal crimping according to claim 3, wherein the degree of separation when> μ _NG = [(μ _OK −n × σ _OK ) − (μ _NG + m × σ _NG )] / (μ _OK −μ _NG ) Judgment device.
5. The terminal crimping quality determination device according to claim 4, wherein m = n = 3.
6. The effective area selection unit is configured to select, as an effective area, one area having the largest value of the first variable among the plurality of areas.
   The terminal crimping quality determination device according to any one of claims 3 to 5, wherein the determination area determination unit is configured to use the effective area as a determination area.
7. The effective area selection unit is configured to select, as an effective area, two or more areas having a larger value of the first variable among the plurality of areas.
   The terminal crimping quality determination device according to any one of claims 3 to 5, wherein the determination area determination unit is configured to determine a determination area using the two or more effective areas.
8. The effective area selection unit selects two or more areas as an effective area from the plurality of areas in order of at least the value of the first variable or in ascending order of standard deviation of the area area of defective samples. Configured to
   The determination area determination unit
   A combination area creating unit for creating a combination area formed by combining the weighted effective areas after weighting the selected effective area;
   A combination area determination unit for determining a determination area from the combination area;
   The terminal crimping quality determination device according to any one of claims 3 to 5, wherein
9. The effective area selection unit selects at least a first area, a second largest second area, and a third largest third area, in which the value of the first variable is the first largest among the plurality of areas. Is configured to select as
   The determination area determination unit
   A combination area obtained by adding the first area and the second area, a combination area obtained by adding the first area and the third area, and a combination area obtained by adding the second area and the third area. A combination area creating unit that creates a plurality of combination areas including a combination area obtained by adding the first area, the second area, and the third area;
   Any one of the plurality of combination areas based on the average value and standard deviation of the area areas of the non-defective samples in each combination area and the average value and standard deviation of the area areas of the defective samples in each combination area A combination area determination unit with a determination area,
   The terminal crimping quality determination device according to any one of claims 3 to 5, wherein
10. The determination area determination unit includes an average value μ _OK of non-defective sample area areas in each combination area, a standard deviation σ _OK of non-defective sample area areas, an average value of defective area area μ _NG , and a defective sample. A calculation unit for each combination area that calculates the standard deviation σ _NG of the area of
    The combination area determination unit sets B = (μ _NG −r when μ _NG > μ _OK among a plurality of combination areas created by the combination area creation unit when r and s are real numbers of 1 or more. × σ _NG ) − (μ _OK + s × σ _OK ) or B = (μ _OK −r × σ _OK ) − (μ _NG + s × σ _NG ) when μ _OK > μ _NG The terminal crimping quality determination device according to claim 9, wherein a combination area having the largest value of two variables is set as the determination area.
11. The second variable is the degree of separation when μ _NG > μ _OK = [(μ _NG −r × σ _NG ) − (μ _OK + s × σ _OK )] / (μ _NG −μ _OK ), or μ _OK The quality of terminal crimping according to claim 10, wherein the degree of separation when> μ _NG = [(μ _OK −r × σ _OK ) − (μ _NG + s × σ _NG )] / (μ _OK −μ _NG ) Judgment device.
12. The terminal crimping quality determination device according to claim 11, wherein r = s = 3.
13. The combination area creating unit is configured to weight at least the first area, the second area, and the third area when creating the plurality of combination areas. The terminal crimping quality determination device according to any one of 12.
14. The terminal crimping quality determination device according to claim 8 or 13, wherein the combination area creation unit is configured to perform weighting such that weighting increases as the standard deviation decreases.
15. 15. The terminal crimping quality determination device according to claim 8, 13, or 14, wherein the combination area creation unit is configured to perform weighting such that weighting increases as the degree of separation increases.
16. The determination area extraction unit
    Among the plurality of areas, an effective area selection unit that selects one or two or more areas as an effective area in ascending order of the value of the standard deviation of the area area of the defective product sample,
    A determination area determining unit that determines the determination area using the effective area;
    The terminal crimping quality determination device according to claim 1, comprising:
17. The threshold setting unit includes:
    The average value [mu] d _OK with an area of good sample in the determination _area, the standard deviation .sigma.d _OK with an area of good _sample, the mean value [mu] d _NG with an area of defective _samples, and standard deviation .sigma.d _NG with an area of defective samples A determination area calculation unit for calculating
    p, when with one or more of the real number q, μd _NG> μd _OK and _{(μd NG -p × σd NG)} > at the time of the _{(μd OK + q × σd OK} ) and (μd _NG -p × σd _NG) predetermined value between the _{(μd OK + q × σd OK} ), or, μd _OK> μd _NG and _{(μd OK -p × σd OK)} > (μ NG + q × σd NG) when the ([mu] d _OK -p × a threshold value determination unit that determines a predetermined value between (σd _OK ) and (μd _NG + q × σd _NG ) as a threshold value;
    The terminal crimping quality determination device according to any one of claims 1 to 16, further comprising:
18. The terminal crimping apparatus includes an anvil on which an end of a wire and a terminal are placed, a crimper that can approach and separate from the anvil, and an electric actuator that causes the crimper to approach and separate from the anvil. And having
    The progress of the terminal crimping is the elapsed time, the position of the crimper, the travel distance of the crimper, the speed of the crimper, the acceleration of the crimper, the current supplied to the actuator, and the operating position of the actuator. The terminal crimping quality determination device according to any one of claims 1 to 17, which is any one.
19. Performing terminal crimping of a plurality of non-defective samples and a plurality of defective samples by a terminal crimping device, and obtaining a pressure waveform representing a relationship between the degree of progress of terminal crimping of each sample and the pressure generated in the terminal crimping device; ,
    Dividing the pressure waveform of each sample into a plurality of areas according to the degree of progress of terminal crimping, and calculating an area area that is an area of a portion surrounded by the pressure waveform for each area;
    Calculating the average value and standard deviation of the area area of the non-defective samples in each area, and the average value and standard deviation of the area area of the defective sample in each area;
    Selecting one or more areas as an effective area from the plurality of areas based on an average value and standard deviation of area areas of the non-defective samples and the defective samples;
    Extracting a determination area using the effective area;
    Setting a threshold value based on an average value and standard deviation of the area area of the non-defective product sample and the defective product sample in the determination area;
    Performing the terminal crimping of the inspection product by the terminal crimping device, and comparing the area in the determination area of the inspection product with the threshold value, and determining the quality of the terminal crimping of the inspection product;
    For determining whether or not a terminal is crimped.

Images