JP4441735B2 - Process monitoring method for cycle operation processing machine - Google Patents

Description

Conventionally, increasing the machine accuracy is a way to increase the product yield of machine tools. To that end, to monitor the operating state of the machine, it is necessary to directly measure the mechanical vibration of the drive shaft. Alternatively, the abnormality is detected by measuring the drive current. The present invention relates to a machining process monitoring method for continuously grasping the state of equipment by measuring the drive current. The present invention is applied to equipment that performs a cycle operation such as operation, standby, operation, standby, for example, an injection molding machine, a cutting device, a press machine, a conveyance device, and the like.

Conventionally, when a workpiece is machined by a machine tool, a copying operation is performed according to a design dimension, and the magnitude of a driving current at that time is measured to monitor whether it is within an abnormal range. (See Patent Document 1). As an example, when a cylindrical workpiece (work) shown in FIG. 11 is manufactured, rotation of a drive motor of a machine tool is started by a machining start signal, and a machining table provided with a tool starts to move. At this time, the drive motor of the machine tool takes a large value for the starting current, and then maintains a no-load state until the tool reaches the machining position of the workpiece, and the current flows constant.

As shown in FIG. 11, the processed product has processing portions with different tools A, B, and C, and the current value becomes high when processing with these tools. The replacement frequency of the tools varies depending on the wear state. And tool exchange is performed in a processing no load state. At the end of machining, the rotation of the motor is stopped using both the brake and the current value becomes high.

The machining status can be managed by monitoring the current of the drive motor because the machining operation affects the current of the drive motor by using a drive motor with an output suitable for the workpiece. Thereby, data for improving the accuracy of the processed product or improving the yield can be obtained.

One example of a method for monitoring one cycle (all intervals) of a processed product is to set the measurement point on this scale with the time axis as an equally divided scale, and store the measured values in time series. The digital data of one cycle is used as temporary reference data, and the digital data of one cycle measured repeatedly next is compared for each sampling point, and the maximum value and the minimum value are stored. Thereafter, the measurement is performed a plurality of times, and a replacement operation is performed as necessary to obtain the maximum value and the minimum value of each sampling point.
Similarly, after the data for each sampling point is accumulated and measured a plurality of times, the average value for each sampling point is stored by dividing the data by a plurality of times.
The average value waveform makes it easy to see the maximum value waveform and the minimum value waveform visually.
Look at the sampling points where there is a large variation in the gap between the maximum value waveform and the minimum value waveform, and the sampling points with the maximum value or the minimum value that are far from the average value. Specify one point, measure multiple times again, calculate the average value and standard deviation value of that sampling point, and use it as a reference for setting the maximum value waveform (upper limit) M and minimum value waveform (lower limit) .
Then, using these, as shown in FIG. 12, the maximum value data, the minimum value data, and the average value data of all sampling points are continuously connected, and the maximum value waveform (upper limit) M and the minimum value waveform (lower limit) m. The pattern is set so that processing abnormalities can be monitored visually.

That is, the maximum value waveform M and the minimum value waveform m are displayed on the display unit of the machining apparatus, and the actual measurement value R during work work is traced on the pattern so that the actual measurement value waveform R becomes the maximum value waveform M or the minimum value waveform. When crossing m, an abnormality detection check can be performed.
Japanese Patent Application Laid-Open No. 5-116056

For example, it is effective to repeatedly measure as in Patent Document 1 and monitor, for example, the machining state by obtaining 1000 cycles of data and drawing the maximum value waveform and the minimum value waveform.

However, for example, in machining a machine part in an automobile factory, it is easy to perform a machining operation that sufficiently exceeds 1000 cycles per day. That is, if one cycle of the machining time for one part is 60 seconds, 60 (cycles / hour) × 24 (hours) = 1440 cycles in 24-hour continuous operation. In such a processing number, the number of 1000 samples is insufficient to grasp the state and accuracy of the machine. This is because the number of samples measured is too small, when there is a chance a large (small) value at 1000 measurements were called Matama was very stable during the measurement, enters element of chance. Therefore, it is only necessary to take a very large number of sample measurements. In that case, on the contrary, in the overlay method, if a large number of sample measurements is taken, data when unstable or in a bad state is also mixed, so monitoring is performed accurately. It becomes difficult to do.

In addition, since the conventional processing allowable range is set as continuous upper limit data and continuous lower limit data, it is impossible to respond quickly when an abnormality is detected. Therefore, when a large number of sample points are taken and dealt with mainly at complex machining points in one cycle, data of other complicated machining points and data considering the whole are insufficient. It was necessary to add monitoring devices with different settings.

The present invention relates to a cycle operation machining in which, in a work work process, monitoring data is created by a sampling point that is measured according to a complicated machining location and a simple machining location, and the workpiece machining is monitored by the monitoring data. It aims at providing the monitoring method of the processing process of a machine.

In order to achieve the above object, the invention of claim 1 of the present application detects the presence or absence of abnormality in a machining process by detecting the load current of a machining motor of a cycle operation machining machine that performs a plurality of different machining operations from operation to standby. In the process monitoring method for monitoring
by machine to perform the cycle operation to standby from a) operating, in the working process including a plurality of processing, for a given number of sampling points in one cycle from start to finish machining, the load current of the working motor The step of converting the amount of change into a signal that can be read for each processing by repeating a plurality of cycles,
b) storing the data of the signal of the sampling point in the storage unit of the CPU according to a change in the signal from the input unit to get process data,
c) After the step of b), the load on each part of the workpiece when operating multiple cycles among the data of all sampling point signals obtained by measuring multiple cycles from the start to the end of the work process. storing the plurality of measured values of current, the mean value of the values were assembled in each of the same sampling point in the storage unit,
step of storing in the storage unit the standard deviation of the values were assembled for each sampling point of d) then the c) are determined by the CPU,
e) In the CPU, a new signal value different from the signal stored in the sampling point b) is compared using the average value or the standard deviation value to determine whether there is an abnormality. And a step of notifying the occurrence of an abnormality by an externally connected device.
As an example of the above configuration, the important part of the signal change in one cycle is sampled at a short time interval (example: 1 ms), and the unimportant part is sampled at a long time interval (example: 100 ms) to store the data. To do.

The invention of claim 2 of the present application is a machining process monitoring method for monitoring the presence or absence of an abnormality in a machining process by detecting a load current of a machining motor of a cycle operation machine that performs a plurality of different machining operations from operation to standby. In
by machine to perform the cycle operation to standby from a) operating, in the working process including a plurality of processing, for a given number of sampling points in one cycle from start to finish machining, the load current of the working motor The step of converting the amount of change into a signal that can be read for each processing by repeating a plurality of cycles,
Step you save a data signal of a sampling point in the storage unit of the CPU according to a change in the signal from the input unit for obtaining and b) data processing,
c) After the step of b), the load on each part of the workpiece when operating multiple cycles among the data of all sampling point signals obtained by measuring multiple cycles from the start to the end of the work process. A step of storing an average value of values obtained by collecting a plurality of actual measured values of current at the same sampling point in a storage unit,
step of storing in the storage unit the standard deviation of the values were assembled for each sampling point of d) then the c) are determined by the CPU,
e) In the CPU, an upper limit value and a lower limit value are set by multiplying the standard deviation value by a coefficient, and an actual measured value for one cycle of the new load current for each part of the workpiece is the upper limit value and the lower limit value . step you comparing whether within the allowable range,
f) Based on a new value of the signal different from the signal stored in b) of the sampling point, the CPU compares the average value or the standard deviation value to determine whether there is an abnormality. And a step of notifying the occurrence of an abnormality by an externally connected device.

The invention according to claim 3 of the present application is a machining process monitoring method for monitoring the presence or absence of an abnormality in a machining process by detecting a load current of a machining motor of a cycle operation machine that performs a plurality of different machining operations from operation to standby. In
by machine to perform the cycle operation to standby from a) operating, in the working process including a plurality of processing, for a given number of sampling points in one cycle to the end from the machining start, the load current of the working motor A step of repeating the step of converting the amount of change into a signal that can be read for each processing for a plurality of cycles,
The sampling points of the machining time axis is set according to the change in the b) the signal takes fine for complex machining spot due to the shape of the word over click, and then to take rough simple machining spot, at each sampling point A step of saving sampling point signal data in a CPU storage unit;
c) after said step of b), among the data of all the sampling points of the signal up to termination was obtained by a plurality of cycles measured from the start of the working process, the data of complex sampling points of the signal processing portion, sampling points maximum, minimum, an average value and standard deviation value, stores the value obtained by arithmetic processing in the storage unit step each,
d) collating the standard deviation value obtained by the arithmetic processing with the value of the new signal different from the signal stored at b) of the sampling point, determining the presence or absence of abnormality, and external connection And a step of notifying the occurrence of an abnormality by the device.

In the method invention according to claim 1 of the present invention, data of one cycle of the machining load current from the machining start to the machining end of the workpiece is obtained, and based on this data, an average value and a standard deviation value are obtained for each sampling point. Since the measured value of the load current is compared and monitored by the monitoring range of the standard deviation value, the monitoring can be performed with high accuracy along the workpiece shape. Further, since an abnormality can be detected for each sampling point , the abnormality can be found quickly.
According to a second aspect of the invention, in the method of the first aspect of the invention, an upper limit value and a lower limit value are set to integer multiples of the standard deviation value, and the actual workpiece measurement value corresponds to the actual measurement allowable range of the upper limit value and the lower limit value. Since the monitoring is performed, the yield of the workpiece can be increased as compared with the case of monitoring with the conventional maximum / minimum value waveform.
Invention according to claim 3, in one cycle of the load current of the working motor of the work, for complex machining spot taking finer sampling point machining time axis to create a sampled data, program processing at each sampling point In addition, since it is stored in the storage unit, an accurate data waveform that can be easily monitored can be created.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the present invention, when monitoring a machine tool, the sampled data to be compared is reduced in sampling time intervals to increase the amount of information in a complicated machining portion of the workpiece, and the change in the shape of the workpiece is gradual. Since the change in sampling data is small at the location, the sampling time interval is lengthened, and the minimum sampling is measured and stored. If there is no information on the workpiece shape, the amount of change at each sampling point is detected in the first cycle of the first measurement to determine the sampling time interval. In the next cycle, the data is recorded while being corrected based on the data of the first cycle.

The workpiece in the example shown in FIG. 3 has machined parts at three locations, and the sampling time intervals are different in each machined part. In addition, the whole machining process of the workpiece is divided into 7 from “1” to “7”, and as shown in FIG. 2 corresponding to FIG. Sampling points are set at 1900 points where data of a part with little or no data is thinned out. As shown in FIGS. 2 and 3, the number of sampling points is 50 in the “1” portion, 350 in the “2” portion, 550 in the “4” portion, and 880 in the “6” portion. There are 70 places in the place “7”. Note that “3” and “5” are not processed and are not sampled. In addition, as shown in FIG. 4 (corresponding to the sampling points and the number of measurement cycles), the “1” and “7” portions are sampled once every five cycles at the start and end portions, In the portion “4” where the shape of the workpiece is gentle, sampling is performed once every three cycles. The parts “2” and “6” where the shape of the workpiece is complicated are parts where sampling is performed in all the cycle numbers.

Next, each division from “1” to “7” will be described. In the category “1”, measurement is started by a passage trigger (trigger for changing the current from the bottom to the top or from the top to the bottom with respect to the set value 1) or a start signal input from the outside. In this category, if the sampling interval from sampling points 1 to 50 is 100 ms, the measurement time is 5 seconds (100 ms × 50 = 5 s), and the measured value is stored in the storage unit. In category “2”, the tool is installed and the measurement start instruction is received and measurement is started. This part is a part in which the shape of the workpiece is complicated, and the sampling point is increased because the tool is changed every 200 pieces. Accordingly, the sampling points 51 to 400 (350 locations) are measured every cycle with a sampling interval of 5 ms and stored in the storage unit. The measurement time is 1.75 seconds (5 ms × 350 = 1.75 s).

In "3" category, tool change is performed by ATS (automatic tool changer). Since the required tool is not in a certain place, the time for moving the tool storage unit of the ATS is not constant. Therefore, when a tool change completion signal (command) is input, the process proceeds to “4”. In the category “4”, the tool is installed and the machining start instruction is received and measurement is started. At this location, the shape of the workpiece is gradual, the load is not large, and abnormalities such as tool breakage do not occur so much, so the tool is changed every 600 pieces. Therefore, 550 points from sampling points 401 to 950 are measured with a sampling time interval of 10 ms and stored in the storage unit every three cycles. The measurement time is 5.5 seconds (10 ms × 550 = 5.5 s).

The “5” section performs the same operation as the “3” section. In category “6”, the tool is installed and the measurement start instruction is received and measurement is started. This part is a part in which the shape of the workpiece is complicated, and the tool change is performed every 100 pieces, so that the machining fluctuation increases. Therefore, 880 points from sampling points 951 to 1830 are measured at a sampling time interval of 3 ms, and the measured values are stored in the storage unit every cycle. The required time is 2.64 seconds (3 ms × 880 = 2.64 s). In “7”, measurement is started by a pass trigger or external start signal input. Sampling points 1831 to 1900 (70 locations) are measured with a sampling time interval of 100 ms and stored in the storage unit every five cycles. The required time is 7 seconds (100 ms × 70 = 7 s).

When data is collected in this way, in the connected storage device, as shown in FIG. 4, the amount of sampling points in categories “1”, “2”, “4”, “6” and “7” In response, a plurality of data is stored within a predetermined sampling point. As shown in FIG. 2, when the data of the categories “1”, “2”, “4”, “6” and “7” are converted into waveforms, a continuous waveform of only the processed portion is obtained.

Next, the sampling process will be described with reference to FIG. First, in the setting of the conditions in step 1, the sampling speed and the number of times are set in the order of steps. In step 2, sampling is started. In step 3, the input of a pass trigger or another trigger is checked, and if there is no input, a standby state is entered here.

In step 4, the sampling rate is set to 3 ms, 5 ms, 10 ms, or 100 ms according to the order of the sampling sections. In step 5, it is determined whether or not the sampling point has been reached in the set section. If it has reached (if YES), the measurement is performed in step 6 and the sampling data is stored. Here, step 4 to step 7 are repeatedly measured and stored according to the set value n. The setting of n is as follows. n = 1 stores data for all measurement cycles. n = 2 skips 2 cycles and saves the data. For n = 5, data is stored once in 5 cycles. For n = X, data is stored once in X cycles.
Next, in step 7, the number of set data is checked. If the quantity is determined, the process proceeds to the next stage. If the quantity is not satisfied, the process returns to step 4 to repeatedly collect data.

When the sampling amount at the determined sampling rate is satisfied, the sampling rate is changed to the next rate in step 8. In step 9, it is determined whether or not the work has been able to collect one cycle of data by a series of samplings. If not yet, the process returns to step 3 to enter a waiting state for a passage trigger. If data for a predetermined cycle can be collected, it is determined in step 10 whether data for a set cycle has been obtained.

When all the data are measured and stored in step 10, the measurement of one section is completed (step 11). If the data for the set cycle has not yet been acquired, the sampling rate is set as the value of the next section. ) Return to step 3 and wait for a passage trigger.

When the measurement of all sections is completed (step 11), the data is calculated in step 12, and the command is sent together with the processed data in accordance with the receiving side capability in step 13. As shown in FIG. 6, the main unit is provided with a control unit 1 for controlling each unit in the main unit of the monitoring device, and the data calculation processing is performed by the CPU 2 so that the communication unit 3 can be connected to the external personal computer 4. It makes monitoring and operation easy. The main body is provided with an input unit 5, a display unit 6, and an output unit 7, and an operation unit 8 and a storage unit 9 for data such as actually measured values and calculation results.

Next, creation of a reference waveform to be compared with actual measurement values will be described. As shown in FIG. 7, when all the cycle data measured and stored in a fractional manner or arbitrary cycle data is represented by a histogram, the average value obtained at each sampling point and the maximum and minimum values obtained by a well-known method are shown. Can be compared with the actually measured values at that time. Also, standard deviation processing is performed at each sampling point to calculate standard deviation values (σ1 to σk, 1900 in the case of paragraph 0016), and 1 × (σ1 to σk) values at each sampling point By creating a continuous 1 (σ1 to σk) curve (making an upper limit value on the positive side and a lower limit value on the negative side with respect to the average waveform), A 1 (σ1 to σk) waveform (upper and lower limit waveform) representing a more accurate accuracy and state than the conventional maximum value waveform and minimum value waveform can be obtained. At the sampling points X1, X2, and X3 shown in FIG. 7, the standard deviation value (σ) is different at each sampling point. The upper and lower limit values are narrower at the good points.

Therefore, the statistical defect rate can be set without measuring all the number of cycles by a statistical method. That is, in the 1σ waveform, about 68 to 70% of the whole is within the determination range and is regarded as a non-defective product. Further, it is possible to create a continuous 2σ curve in which 2 × σ values are connected along points, and a continuous 3σ curve in which 3 × σ values are connected along points. The monitoring width may be 1σ, 2σ, 3σ, or a width of σ multiplied by another coefficient. As shown in FIG. 7, when the deviation value waveform centered on the average value waveform is greater than or equal to the deviation value of plus or minus 3σ, the waveform is more suitable for the processed product than the maximum value waveform obtained by connecting the conventional maximum values. In this monitoring range, 99.8% of the total is non-defective.

Therefore, a value close to the target defect rate can be multiplied by the standard deviation value to create an upper limit waveform and a lower limit waveform as shown in FIG. 1, and whether the processing is performed with accuracy within this waveform. Depending on how, the machining process of the workpiece can be monitored.

Next, functions at each sampling point will be described. As shown in FIG. 8, in the work monitoring areas “1” to “7”, corresponding storage spaces are provided in the main body storage unit. In the section corresponding to each sampling point, various elements are incorporated such as a place where a calculation formula for performing program processing at this place is provided and a place only for storing. The areas “3” and “5” are not processed and are not sampled, but a program such as tool change by an ATS (automatic tool changer) is executed during that time.

As an example of processing at each sampling point, as shown in FIG. 8 and FIG. 9 (P449, P450, P451, P452, and P453 of “4” in FIG. It corresponds to the processing of C-1), (C-2), (C-3), and (C-4), and the processing at other points in FIG. 9 is described as “program processing”. This is the same in FIG. 10), in the region “4”, for the sampling points 449 to 453 (as indicated by “4” in FIG. 4), this is 5 of the sampling points 401 to 950. (Sampling points), storage, retrieval and calculation sheet (C-1), calculation sheet (C-2), actual value input sheet (C-3) and command (calculation for monitoring program execution) Results transfer / communication) sheet (C 4 processing) is performed. Here, “sheet” refers to processing for each sampling point , and a series of executions is referred to as “program processing”.

Since the standard deviation value of each sampling point is different as shown in FIG. 7, a program process is provided for each sampling point, and the calculation sheet (C-2), storage sheet (C-1) and command of FIG. A sheet (C-4) is executed (program execution transfers and communicates calculation results).
In the input sheet (C-3), the actually measured data is sampled for each set sampling point, stored in the storage sheet (C-1), and the maximum value data is retrieved from the stored sampling data or the average value And a standard deviation value are calculated and calculated in the calculation sheet (C-2), then stored in the storage unit, and each sheet (C-1) at the sampling point described in the command sheet (C-4) , (C-2) and (C-3) are compared, and the value of the command sheet (C-4 is used to transfer and communicate the calculation result) is compared to determine whether there is an abnormality. judge.

In FIG. 9, with respect to the sampling point 449, the data of only the portion corresponding to the importance of processing among all the sampling data obtained by measuring all cycles from the start to the end of the work process on the sheet (C-1). The maximum value is retrieved from 500 samples at the sampling point 449 stored in the storage unit. In the sheet (C-2), the maximum value is substituted into the correction calculation formula of Y = AX + B (X: stored maximum value, A, B: coefficient), and the maximum calculation value is calculated. In (C-3), a new actual measurement value at the sampling point 449 is input, and in the sheet (C-4), the new actual measurement value is subtracted from the calculated maximum value in the sheet (C-2). The value is transferred in binary format to the necessary components according to the communication standard RS-232C.

In the following, processing at four consecutive sampling points will be described. As for the sampling point 450, in the sheet (C-1), out of all sampling data obtained by measuring all cycles from the start to the end of the work process, only data corresponding to the importance of processing is stored in the storage unit. The 500 minimum values at the stored sampling point 450 are searched, and the correction value of Y = AX + B is calculated as the minimum value in the sheet (C-2) (X: stored minimum value, A, B: Calculate the minimum value in the coefficient, and input a new measured value at the sampling point 450 in the sheet (C-3), and the sheet (C-4) and the sheet (C-2) When the minimum value of the sheet (C-3) is compared and the minimum value of (C-2) is larger, it is determined that there is an abnormality, and Ethernet (registered trademark) communicates with the lower limit abnormality.

As for the sampling point 451, in the sheet (C-1), out of all sampling data obtained by measuring all cycles from the start to the end of the work process, only data corresponding to the importance of processing is stored in the storage unit. The average value of 500 stored sampling points 451 is calculated, a new actual measurement value at the sampling point 451 is input in the sheet (C-3), and the sheet (C−) is input in the sheet (C-4). 1) and the value of sheet (C-3) are compared, and when the value of (C-2) is larger, the correction amount is communicated.

As for the sampling point 452, in the sheet (C-1), out of all the sampling data obtained by measuring all cycles from the start to the end of the work process, only the data corresponding to the importance of processing is stored in the storage unit. 700 standard deviation values and average values at the stored sampling points 452 are calculated, and the upper limit value is calculated from the sum of the average value and 3σ using the calculated standard deviation value as σ in the sheet (C-2). obtain. A new actual measurement value at the sampling point 452 is input in the sheet (C-3), and the values of the sheet (C-2) and the sheet (C-3) are compared with the sheet (C-4). When the value of (C-3) is larger, the difference is transferred in binary format by Ethernet (registered trademark).

As for the sampling point 453, only data corresponding to the importance of processing is stored in the storage unit among all sampling data obtained by measuring all cycles from the start to the end of the work process in the sheet (C-1). 700 standard deviation values and average values at the stored sampling points 453 are calculated, and the lower limit value is calculated from the difference between the average value and 3σ using the calculated standard deviation value as σ in the sheet (C-2). obtain. A new actual measurement value at the sampling point 453 is input in the sheet (C-3), and the values of the sheet (C-2) and the sheet (C-3) are compared with the sheet (C-4). When the value of (C-3) is larger, the difference is transferred in binary format by Ethernet (registered trademark).

Thus, for each sampling point, a data storage part, a calculation part, and a transfer / communication part are provided. As shown in FIG. 10, at each sampling point, an actual measurement value is stored and calculated as a standard deviation value, and the difference is transferred and communicated together with determination of the presence or absence of abnormality. Therefore, since abnormality can be monitored at each point, the number of samples can be reduced, and abnormality can be found quickly.

Further, if the actual measurement value, the upper limit value, and the lower limit value are color-coded by RGB processing rather than comparing the numerical data on the monitoring screen, abnormalities in the actually measured workpiece accuracy can be easily recognized, thereby reducing the monitoring work. As for the monitoring width, the 1σ waveform is displayed in blue, the 2σ waveform is displayed in green, the 3σ waveform is displayed in red, and the actually measured waveform is colored in orange. The stage of abnormality at 1σ, 2σ, and 3σ can be determined. Note that the standard deviation value can be a relative comparison of the current to be measured, current, temperature, and anything else, and if the change in the object to be monitored can be converted into a signal, the accuracy can be managed by this monitoring method.

Next, determination and processing for occurrence of abnormality will be described. First, an abnormality occurs when the upper and lower limit values are continuously exceeded for 1 to N numbers set in advance. For example, when the consecutive number N is 3, it is determined that an abnormality occurs when two consecutive points are below the lower limit value and the third point is above the upper limit value. Also, the first point exceeded the upper limit, the next point entered the upper and lower limit monitoring range, the next two points fell below the lower limit value, but the next fifth point entered the upper and lower limit monitoring range If it is not abnormal.

When normal mode is set (always connected to PC or server)
When an error occurs, the external output that reports the error, the error notification command, and the waveform data for one cycle at the time of the error are sent to a personal computer or server using Ethernet (registered trademark) or other means of communication and immediately monitored again Return to. The waveform data for one cycle at the time of abnormality is stored at each sampling point, and the measured value at the time of abnormality is always transmitted. Therefore, in this case, it is quick only by determining the abnormality. Can be notified.

When setting delay delay (Periodically connected to PC or server)
When an abnormality occurs, an external output for notifying the abnormality and a communication command for the occurrence of the abnormality are sent to a personal computer or server using communication means such as Ethernet (registered trademark), and the monitoring state is immediately returned again. Even in this method, since the data is monitored at each sampling point, the abnormality can be notified promptly.

Further, the waveform data for one cycle at the time of occurrence of abnormality is stored in the internal memory for a preset number of waveforms M (limited by the capacity of the internal memory). When abnormal waveform data is collected exceeding the number of waveforms M, the old waveform data is discarded and new data is stored. As a result, M latest abnormal waveform data can always be stored. This can be easily transferred as a document to a personal computer or server.

Moreover, the following may be sufficient as this invention as another embodiment.
(A) The change amount of one cycle from the start to the end of the work process is converted into a readable signal, sampling points are set according to the signal change, and sampling data measured for a plurality of cycles are stored, and the sampling points Each time, a standard deviation value is obtained and programmed, and the measured value of the signal is checked against the standard deviation value. In this embodiment, when one cycle from the start to the end of a work process is monitored, a standard deviation value is obtained for each sampling point of one cycle, and upper and lower limit data is created from this, and actual values are monitored for upper and lower limit values. Since the presence / absence of abnormality is monitored by the width, it is possible to monitor the accuracy in the work process.

(B) As a method of monitoring the load current of the machining motor provided in the workpiece machining apparatus in units of one cycle from the start to the end of machining, the machining time axis is used for complicated machining locations due to the shape of the workpiece. take the sampling points of the fine, and, by a simple machining spot to take roughly the sampling point of the machining time axis, as well as individually storing sampling data to the CPU memory unit at each sampling point, along with a program tailored to the application Numerical processing may be performed. In this case, an accurate data waveform that can be easily monitored can be created.

(C) In (b) above, regardless of the shape of the workpiece, a complex machining point or a simple machining point is determined from the sampling point data before and after the machining point, and based on this, the sampling point on the time axis is determined. May be determined. In that case, even if there is no design drawing in the shape of the workpiece, sampling is performed by determining the complicated machining location or simple machining location from the sampling point data before and after the machining location, so that the indeterminate workpiece Suitable for sampling.

(D) Also, in (b) above, based on the shape data of the workpiece, the number of sampling points is determined for each complex machining location or simple machining location for a complicated machining location or a simple machining location. May be. In that case, since the number of sampling points at each complicated machining location or simple machining location is determined based on the shape data of the workpiece, accurate data can be obtained.

It is a graph of the upper and lower limit value waveform of the load current of the workpiece | work of embodiment by this invention, and an actual value waveform. 4 is a continuous graph of the number of sampled data and a processed portion according to the present invention. It is a graph which shows the sampling point of the workpiece | work by this invention. It is a figure explaining the memory | storage part of this invention, Comprising: It is a figure which shows the correspondence of the sampling point memorize | stored and the measurement data for every cycle. 4 is a flowchart of a sampling setting cycle operation according to the present invention. It is a block diagram of the main body used for the monitoring method of this invention. In this invention, it is the figure which showed the relationship between the average value, the maximum value, the minimum value, and the upper and lower limit values based on the standard deviation with a waveform and a histogram. In FIG. 1, it is a figure which shows a response | compatibility with a sampling point and a memory | storage part location. In FIG. 1, it is a block diagram of the memory | storage part which shows some content of a sampling point . In FIG. 1, it is a schematic diagram which shows performing and monitoring the program process in each sampling point . It is explanatory drawing which shows the work process of the conventional workpiece | work. It is a schematic diagram explaining the monitoring method by the maximum value waveform and the minimum value waveform of the conventional workpiece | work.

“1” to “7” Monitoring area

Claims (3)

In the process monitoring method that monitors the presence or absence of abnormalities in the machining process by detecting the load current of the machining motor of a cycle operation machine that performs multiple different machining operations from operation to standby, a) from operation to standby by cycle machine to perform the operation, in the working process including a plurality of processing, for a given number of sampling points in one cycle from start to finish machining, the amount of change a load current of the working motor, working Repeating the step of converting to a readable signal every multiple cycles,
b) storing the data of the signal of the sampling point in the storage unit of the CPU according to a change in the signal from the input unit to get process data,
c) After the step of b), the load on each part of the workpiece when operating multiple cycles among the data of all sampling point signals obtained by measuring multiple cycles from the start to the end of the work process. storing the plurality of measured values of current, the mean value of the values were assembled in each of the same sampling point in the storage unit,
step of storing in the storage unit the standard deviation of the values were assembled for each sampling point of d) then the c) are determined by the CPU,
e) In the CPU, a new signal value different from the signal stored in the sampling point b) is compared using the average value or the standard deviation value to determine whether there is an abnormality. And a step of notifying of the occurrence of an abnormality by an externally connected device. In the process monitoring method that monitors the presence or absence of abnormalities in the machining process by detecting the load current of the machining motor of a cycle operation machine that performs multiple different machining operations from operation to standby, a) from operation to standby by cycle machine to perform the operation, in the working process including a plurality of processing, for a given number of sampling points in one cycle from start to finish machining, the amount of change load current of the working motor, working Repeating the step of converting to a readable signal every multiple cycles,
Step you save a data signal of a sampling point in the storage unit of the CPU according to a change in the signal from the input unit for obtaining and b) data processing,
c) After the step of b), the load on each part of the workpiece when operating multiple cycles among the data of all sampling point signals obtained by measuring multiple cycles from the start to the end of the work process. A step of storing an average value of values obtained by collecting a plurality of actual measured values of current at the same sampling point in a storage unit,
step of storing in the storage unit the standard deviation of the values were assembled for each sampling point of d) then the c) are determined by the CPU,
e) In the CPU, an upper limit value and a lower limit value are set by multiplying the standard deviation value by a coefficient, and an actual measured value for one cycle of the new load current for each part of the workpiece is the upper limit value and the lower limit value . step you comparing whether within the allowable range,
f) Based on a new value of the signal different from the signal stored in b) of the sampling point, the CPU compares the average value or the standard deviation value to determine whether there is an abnormality. And a step of notifying of the occurrence of an abnormality by an externally connected device. In the process monitoring method that monitors the presence or absence of abnormalities in the machining process by detecting the load current of the machining motor of a cycle operation machine that performs multiple different machining operations from operation to standby, a) from operation to standby by cycle machine to perform the operation, in the working process including a plurality of processing, for a given number of sampling points in one cycle to the end from the machining start, the amount of change load current of the working motor, machining each Repeating the step of converting the signal into a readable signal for a plurality of cycles,
The sampling points of the machining time axis is set according to the change in the b) the signal takes fine for complex machining spot due to the shape of the word over click, and then to take rough simple machining spot, at each sampling point A step of saving sampling point signal data in a CPU storage unit;
c) after said step of b), among the data of all the sampling points of the signal up to termination was obtained by a plurality of cycles measured from the start of the working process, the data of complex sampling points of the signal processing portion, sampling points maximum, minimum, an average value and standard deviation value, stores the value obtained by arithmetic processing in the storage unit step each,
d) collating the standard deviation value obtained by the arithmetic processing with the value of the new signal different from the signal stored at b) of the sampling point, determining the presence or absence of abnormality, and external connection A method for monitoring a machining process of a cycle operation processing machine, characterized by comprising a step of notifying occurrence of an abnormality by an apparatus.

Images