JPWO2019207718A1 - Condition monitoring device and asynchronous data adjustment method - Google Patents

Abstract

A state monitoring device that monitors the state of production equipment (2), which is expected to operate at a constant rate for each cycle, by waveform band monitoring based on the measured values from the measuring device (3) that does not respond to time information. In (1), when the number of measurement points (N) of the raw waveform data (Wr) does not match the reference number of measurement points (X), the measurement points (Mi) of the raw waveform data (Wr) or between the measurement points Measurement point adjustment unit that adjusts the raw waveform data (Wr) so that it matches the reference measurement points (X) by deleting the measurement points (Mi) set by a predetermined reference or adding measurement points between measurement points. It was configured to have (13).

Description

The present invention relates to a condition monitoring device that monitors the condition of an equipment using data (asynchronous data) that does not have time information from the measuring equipment, and a method of adjusting the asynchronous data so that it can be applied to the condition monitoring.

In condition monitoring for monitoring the presence or absence of equipment abnormalities, the current waveform during operation of production equipment is acquired in advance a predetermined number of times, and the upper and lower thresholds are set for each measurement point that composes the waveform using a statistical method. , A method for detecting an abnormality in a production facility is known (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2002-341909 (paragraphs 0024 to 0031, FIGS. 1, 5, and 6)

However, in the above method, time information for each measurement point is required when generating waveform data, but most general-purpose measuring instruments do not respond to time information. Further, a general-purpose measuring device does not measure data in synchronization with a data collecting device that monitors a state, and even a higher-level device that collects data does not always be able to collect data at regular intervals.

Therefore, for example, even when targeting equipment such as MC processing machines and press machines that are expected to operate at regular intervals and operations, the number of data points that can be acquired from the start to the stop of operation may vary. It was. That is, when data is collected from a general-purpose measuring device, there is a possibility that the measurement points for comparing the upper and lower limits may shift, making accurate abnormality detection difficult.

The present invention has been made to solve the above problems, and is collected from a condition monitoring device or a general-purpose measuring device capable of accurately detecting an abnormality even if a general-purpose measuring device is used. The purpose is to obtain an adjustment method for asynchronous data that adjusts asynchronous data so that status monitoring is possible.

The state monitoring device of the present invention is a state monitoring device that monitors the state of a device that is expected to operate in a constant operation for each cycle by determining a threshold value for each measured value constituting waveform data for one cycle. Then, from the measuring device that does not respond to the time information, the measured value collecting unit that collects the measured value indicating the state of the device and the measured value collected by the measured value collecting unit are arranged in chronological order for one cycle of the device. The waveform processing unit that generates the waveform data of the above, the reference waveform that is the waveform data for one cycle when the device is normal, the threshold value for each measured value of the reference waveform, and the number of measurement points of the reference waveform are determined. Wave data adjustment that adjusts and outputs the reference waveform data storage unit that stores the reference waveform data including the reference measurement points, and the raw waveform data generated by the waveformization processing unit to compare with the reference waveform data. The waveform data adjusting unit includes a unit and an abnormality presence / absence determination unit that compares the waveform data output by the waveform data adjusting unit with the reference waveform data and determines whether or not there is an abnormality in the apparatus. When the number of measurement points of the raw waveform data does not match the reference measurement points, the measurement points set by a predetermined reference among the measurement points or measurement points of the raw waveform data are deleted, or the measurement points between the measurement points are deleted. By adding the above, the raw waveform data is adjusted so as to match the reference measurement points.

The asynchronous data adjustment method of the present invention is assumed to operate in a constant operation for each cycle using waveform data in which asynchronous measured values output from a measuring device that does not respond to time information are arranged in chronological order. This is a method of adjusting waveform data for use in monitoring the state of a device, which is a measurement value collection step for collecting measurement values indicating the state of the device from the measurement device, and a measurement collected in the measurement value collection step. The values are arranged in chronological order, a waveformization step that generates waveform data for one cycle of the device, a reference waveform that is waveform data for one cycle when the device is normal, and each measured value of the reference waveform. The reference waveform data storage step for storing the reference waveform data including the threshold value of and the reference measurement points determined as the measurement points of the reference waveform, and the raw waveform data generated in the waveformization step are compared with the reference waveform data. An abnormality presence / absence determination step of comparing the waveform data adjustment step adjusted and output in order to perform the adjustment and the waveform data output in the waveform data adjustment step with the reference waveform data and determining whether or not there is an abnormality in the apparatus. In the waveform data adjustment step, when the number of measurement points of the raw waveform data does not match the reference measurement points, the measurement points of the raw waveform data or between the measurement points are set by a predetermined reference. It is characterized in that the raw waveform data is adjusted so as to match the reference measurement points by deleting the measurement points or adding the measurement points between the measurement points.

According to the condition monitoring device of the present invention or the asynchronous data adjustment method, the number of data can be adjusted appropriately even if the number of data points varies, so that accurate measurement equipment that does not respond to time information can be used. Status monitoring can be performed.

It is a functional block diagram which shows the structure of the condition monitoring apparatus which concerns on Embodiment 1 of this invention. This is a timing chart for explaining the difference between the acquisition time of data transmitted by a measuring device having no synchronization function and the request time of data to be collected by the condition monitoring device. This is a timing chart for explaining the difference between the time when the production equipment outputs the equipment operation information and the time when the condition monitoring device confirms the equipment operation status. It is a waveform diagram for demonstrating the method of setting the reference waveform data including the upper and lower limit values for each measurement point from a plurality of collected waveform data. It is a flowchart for demonstrating the operation which collects measurement data and generates raw waveform data in the condition monitoring apparatus and the adjustment method of asynchronous data which concerns on Embodiment 1 of this invention. In the flow chart for explaining the operation of processing the raw waveform data and generating the reference waveform data as a criterion for determining the presence or absence of abnormality in the condition monitoring device and the asynchronous data adjustment method according to the first embodiment of the present invention. is there. Example of waveform data acquired when there is an excess or deficiency of one point in the number of measurement points of data transmitted from a measurement device that does not have a synchronization function with respect to the number of measurement points of data to be collected by the condition monitoring device. It is a figure which shows. FIG. 5 is a waveform diagram for explaining the processing of measurement points for waveform data having an excess or deficiency of one measurement point in the condition monitoring device and the asynchronous data adjustment method according to the first embodiment of the present invention. It is a figure for demonstrating the process of additionally inserting a measurement point in the condition monitoring apparatus and the adjustment method of asynchronous data which concerns on Embodiment 1 of this invention. It is a figure for demonstrating the process of deleting a measurement point in the condition monitoring apparatus and the adjustment method of asynchronous data which concerns on Embodiment 1 of this invention. Example of waveform data acquired when there are two excesses or deficiencies in the number of measurement points of data collected from a measurement device that does not have a synchronization function with respect to the number of measurement points of data to be collected by the condition monitoring device. It is a figure which shows. FIG. 5 is a waveform diagram for explaining the processing of measurement points for waveform data having an excess or deficiency of two measurement points in the condition monitoring device and the asynchronous data adjustment method according to the first embodiment of the present invention. FIG. 5 is a flowchart for explaining an operation of processing raw waveform data to determine the presence or absence of an abnormality in the condition monitoring device and the asynchronous data adjustment method according to the first embodiment of the present invention. In the state monitoring device and the asynchronous data adjustment method according to the second embodiment of the present invention, a flowchart for explaining an operation of specifying a data deletion point or an addition point for waveform data having excess or deficiency in the number of measurement points. is there. FIG. 5 is a waveform diagram for explaining the processing of measurement points for waveform data having excess or deficiency in the number of measurement points in the condition monitoring device and the asynchronous data adjustment method according to the second embodiment of the present invention.

Embodiment 1.
1 to 12 are for explaining the configuration and operation of the state monitoring device according to the first embodiment of the present invention, or a method for adjusting asynchronous data, and FIG. 1 is a functional block showing the configuration of the state monitoring device. FIG. 2 and FIG. 2 show the acquisition time of data transmitted by a measuring device having no synchronization function and the request for data to be collected by the state monitoring device as explanations of the problems to be solved in each embodiment of the present invention. A timing chart for explaining the deviation from the time point, FIG. 3 explains the time difference between the time when the production equipment outputs the equipment operation information indicating its own operation status and the time when the state monitoring device confirms the equipment operation status. It is a timing chart for. Further, FIGS. 4A to 4E show a method of setting a reference waveform and an upper / lower limit value for each measurement point from a plurality of collected waveform data as an explanation of the reference waveform data for waveform band monitoring. It is a waveform diagram for demonstrating.

FIG. 5 is a flowchart for explaining an operation of collecting measurement data and generating raw waveform data in a state monitoring device and an asynchronous data adjustment method, and FIG. 6 is a process of processing the raw waveform data to determine the presence or absence of an abnormality. A flowchart for explaining the operation of generating the reference reference waveform data as a reference, FIGS. 7 (a) to 7 (c) are measurements that do not have a synchronization function with respect to the number of measurement points of the data to be collected by the state monitoring device. Figures 8 (a) and 8 (b) show examples of waveform data acquired when there is an excess or deficiency of 1 point in the number of measurement points of the data transmitted from the device. A waveform diagram for explaining the process of adding and deleting measurement points to a certain waveform data, FIGS. 9A and 9B are diagrams for explaining a backward shift process of measurement points when additional measurement points are inserted. 10 (a) and 10 (b) are diagrams for explaining the forward shift process of the measurement points when the measurement points are deleted.

11 (a) and 11 (b) are diagrams and diagrams showing an example of waveform data acquired when there are excess or deficiency of 2 points in the number of measurement points of data collected from a measuring device having no synchronization function. 12 (a) and 12 (b) are waveform diagrams for explaining the process of adding and deleting measurement points to waveform data having excess or deficiency of two measurement points. FIG. 13 is a flowchart for explaining an operation of processing the raw waveform data to determine the presence or absence of an abnormality.

As shown in FIG. 1, the condition monitoring device 1 according to the first embodiment of the present invention is a device (production equipment 2) that is expected to operate at a constant cycle / operation of, for example, an MC processing machine, a press machine, or the like. It is for monitoring the status of. Then, the state of the production equipment 2 is determined based on the data from the measuring device 3 which outputs the measured values for detecting the equipment state such as the current supplied to the production equipment 2 and the equipment state such as pressure and temperature by communication without time information. It is to be monitored. Specifically, the monitoring waveform data measured for each operation cycle of the production equipment 2 is compared with the preset reference waveform data for each measurement point, and whether or not the upper and lower limit values are deviated for each measurement point. The state of the production equipment 2 is monitored by determining the presence or absence of an abnormality.

The reference waveform data refers to a combination of data of each measurement point forming a reference waveform (reference waveform Wa) and an upper limit value Tu and a lower limit value Thl provided for each measurement point. Further, the range in which the reference waveform Wa is widened by the upper limit value Tu and the lower limit value Thl is referred to as a waveform band T (see FIG. 4), and whether the acquired waveform data for monitoring deviates from the waveform band T. Determining the presence or absence of an abnormality based on whether or not it is present and monitoring it is called waveform band monitoring.

Therefore, the state monitoring device 1 for waveform band monitoring is a transmission / reception unit 16 that transmits / receives data such as reference waveform data and measured values for generating monitoring waveform data by communication with the measuring device 3, and a reference waveform. The reference waveform data generation unit 10 that generates data, the reference waveform data storage unit 11 that stores the generated reference waveform data, and the waveform data (raw waveform data Wr) for each cycle are arranged in chronological order. An abnormality presence / absence determination unit that compares the generated waveform processing unit 14 with the reference waveform data stored in the reference waveform data storage unit and the waveform data for monitoring, and determines the presence / absence of an abnormality in the production equipment 2 to be monitored. 15, the operation unit 17 that performs input / output operations, the display unit 18 that outputs the operating status of the production equipment 2 and the occurrence of an abnormality, and the operating state of the production equipment 2 in response to signals from the production equipment 2. It is provided with a state monitoring control unit 12 for integratedly controlling each of the above-mentioned units.

The feature of the condition monitoring device 1 according to each embodiment of the present invention is that the number of measurement points of waveform data having excess or deficiency in the measurement points described later is adjusted to be suitable for reference waveform data and waveform data for monitoring. It is provided with the measurement point adjusting unit 13 for modifying the form. Here, before the explanation of each part including the measurement point number adjusting unit 13 which is a characteristic part, the factors causing excess or deficiency of measurement points in the waveform data of the device which is supposed to operate at a constant cycle / operation explain.

As explained in the background technology, general-purpose measuring instruments output only measured values without time information. Therefore, the condition monitoring device arranges the data acquired from the measuring device in chronological order, and generates waveform data in which the measured values are plotted on the XY plane, such as the measured values (Y-axis) and the measurement point order (X-axis). To do. When waveform data for one cycle from the start to stop of operation is acquired for a production facility that operates at a fixed cycle and operation, the operating time does not change, so ideally the same number of measurement points can be acquired. Here, the measuring device updates the measured value at a cycle peculiar to the device, and the condition monitoring device similarly requests data from the measuring device at the cycle peculiar to the device. However, when a general-purpose measuring device is used, there is no means for synchronizing the measurement value update of the measuring device and the timing at which the condition monitoring device makes a data request, and the acquired data becomes asynchronous data.

Therefore, the waveform data generated by the condition monitoring device acquiring the measured values in chronological order does not match the measurement sampling cycle. For example, as shown in FIG. 2, data 3 and data 5 are missing. As it occurs, there is a discrepancy between the actual measurement points and the collected measurement points. In other words, as long as a general-purpose measuring device that cannot link time information to the measured value is used, even if the production equipment operates at a fixed cycle / operation, the waveform data of one cycle from the start to the stop of the operation can be used for each cycle. There is a possibility that there may be a slight excess or deficiency (variation) in the number of measurement points.

Further, the timing of switching ON / OFF of the actual operating state in the production equipment and the delay time required for the condition monitoring device to detect ON / OFF of the production equipment are also considered to be factors of variation in the number of measurement points. For example, as shown in FIG. 3, the time until the condition monitoring device requests a signal indicating the operating state from the production equipment at the timing when the production equipment actually turns ON / OFF and detects ON / OFF. It is estimated that a delay time (ΔTi, ΔTe) may occur. In this case, there is a deviation of ΔTe−ΔTi from the 1-cycle operating time TcD that the condition monitoring device detects (recognizes) as ON with respect to the actual 1-cycle operating time TcR, and the number of measurement points varies. ..

On the other hand, as shown in FIGS. 4A to 4C, the waveform band T used for the waveform band monitoring has a plurality of waveform data W1, W2, ... Wn for each cycle in a normal state of the production equipment. (In the figure, it is expressed as n = 3 for simplification). The values (measured values) of the acquired plurality of waveform data W1 to W3 for each measurement point usually fluctuate as shown in FIG. 4 (d). Therefore, statistical processing can be performed for each measurement point, and for example, the average value for each measurement point can be calculated as the reference waveform Wa. Similarly, if the standard deviation for each measurement point is added to or subtracted from the average value to calculate the upper limit value Tu and the lower limit value Thl, it is possible to generate reference waveform data in which the above-mentioned waveform band T can be defined.

However, if the number of measurement points varies from cycle to cycle, the target of statistical processing shifts, and accurate reference waveform data cannot be calculated. Further, even if the reference waveform data is calculated only from the cycle data in which the measurement points are aligned, accurate waveform band monitoring cannot be performed if the measurement points of the monitoring waveform data are deviated. That is, the waveform band cannot be monitored if the number of measurement points varies.

However, since the condition monitoring device 1 according to each embodiment of the present invention is provided with the measurement point adjusting unit 13 for appropriately adjusting the number of measurement points of the waveform data for each cycle, the general-purpose measuring device 3 is sequentially started. Accurate waveform band monitoring can be performed based on the collected data. The details will be described below.

<Reference waveform data generator>
The reference waveform data generation unit 10 includes a reference waveform data storage unit 10a for storing a plurality of raw waveform data Wr generated by the waveform processing unit 14 for generating reference waveform data, and a predetermined waveform acquired a plurality of times. From the data, the reference measurement point derivation unit 10b for deriving the reference measurement point number (reference measurement point number X) and the adjusted waveform data W in which the measurement points of the raw waveform data Wr are adjusted by the measurement point adjustment unit 13 described later Statistical value calculation unit that derives statistical values such as average value and standard deviation for each measurement point from the adjusted waveform data storage unit 10c to be stored and the adjusted waveform data W stored in the adjusted waveform data storage unit 10c. The reference waveform Wa is calculated from the average value of 10d and each measurement point calculated by the statistical value calculation unit 10d, and the upper limit value Tu and the lower limit value Thl are set for each measurement point of the reference waveform Wa based on the standard deviation. A waveform band generation unit 10e that generates a waveform band T is provided.

In each embodiment, an example in which the average value is used to generate the reference waveform Wa, the upper limit value Tu for generating the waveform band T, and the standard deviation is used to generate the lower limit value Thl is shown, but the present invention is limited to this. There is no. Depending on the characteristics of the monitoring target, for example, the mode value or the median value may be used as the reference waveform, or the upper limit value Tu and the lower limit value Thl may be generated using other statistical values. Alternatively, a smoothing process using the values of nearby measurement points may be added. Further, adjustments such as excluding data having a large deviation from other data may be made.

<Measurement point adjustment unit>
The measurement point adjustment unit 13 determines whether or not the number of measurement points of the raw waveform data Wr output from the state monitoring control unit 12 matches the reference measurement point X stored in the reference waveform data storage unit 11. If there is an excess or deficiency between the shortage determination unit 13a and the number of measurement points of the raw waveform data Wr, the adjustment target of the raw waveform data Wr (the position between the measurement points where the measurement points should be inserted or the position of the measurement points to be deleted). The adjustment point determination unit 13b is provided, and the waveform data adjustment unit 13c that adjusts the raw waveform data Wr based on the adjustment target determined by the adjustment point determination unit 13b to generate the adjusted waveform data W.

The measurement point adjustment unit 13 outputs the generated adjusted waveform data W to the reference waveform data generation unit 10 or the condition monitoring control unit 12 under the control of the condition monitoring control unit 12.

As described above, the condition monitoring control unit 12 controls the operation of the entire condition monitoring device 1 in an integrated manner, and requests a signal indicating the operating state of the production equipment 2 by the method described with reference to FIG. It has a function to judge the operating status. In each embodiment, an example in which the operating status of the production equipment 2 is determined by the output signal from the operating state output unit 20 of the production equipment 2 is shown, but the present invention is not limited to this. For example, the determination may be made in conjunction with the start switch and the stop switch provided in the production equipment 2, and the operating status of the production equipment 2 may be determined by the signal from the measuring device 3.

In each embodiment, each part of the condition monitoring device 1 is described as if there is individual hardware, but it does not necessarily have to be composed of physically separated members. For example, it may be realized by operating a computer or the like by a program composed of modules corresponding to each part of the condition monitoring device 1 or each step in the asynchronous data adjustment method described in the subsequent operations. Further, it may be realized by having a computer read a storage medium containing the above-mentioned program.

Next, the operation will be described.
When the condition monitoring control unit 12 determines that the production equipment 2 is in the ON state, the waveform processing unit 14 uses the measured values collected via the transmission / reception unit 16 until it is determined that the production equipment 2 is in the OFF state. The conversion processing operation is performed, and the raw waveform data Wr for one cycle is generated. Specifically, as shown in the flowchart of FIG. 5, the state monitoring control unit 12 acquires a signal indicating the operating state from the operating state output unit 20 (step S100), and whether it is in the ON state or the OFF state. Is determined (step S110). The ON state here does not mean that the power supply is ON, but means a state in which it is operating as a production facility.

When it is determined that the state is ON (“Y” in step S110), the condition monitoring control unit 12 sends and receives the transmission / reception unit 16 from the sensor unit 30 and the measurement calculation unit 31 as described in FIG. The measured value of the production equipment 2 to be obtained is requested, and the measured value Mi is acquired by the response of the measured value from the communication unit 32 (step S120). This operation is continued until the condition monitoring control unit 12 acquires the signal of the operating state (step S130) and determines that it is OFF.

If it is determined that the ON has been switched to OFF, the waveform processing unit 14 arranges a series of measured values Mi received by the transmission / reception unit 16 in chronological order to generate raw waveform data Wr, and outputs the raw waveform data Wr to the condition monitoring control unit 12 ( Step S140).

When the initial state of the condition monitoring device 1 or the condition of the monitoring target such as switching of the production equipment 2 changes and it is necessary to generate the reference waveform data, the raw waveform data Wr generated and output by the waveform processing unit 14 is , Is output to the reference waveform data generation unit 10. Then, as shown in the flowchart of FIG. 6, reference waveform data is formed.

The reference waveform data generation unit 10 determines whether or not a predetermined number (n) of waveform data, which is the source of the reference waveform data, can be acquired (whether or not it is stored in the reference waveform data storage unit 10a). (Step S200). When it is determined that the predetermined number (n) has been acquired (“Y” in step S200), the reference measurement point number derivation unit 10b obtains each raw waveform data from the raw waveform data Wr stored in the reference waveform data storage unit 10a. Based on the average of the constituent measurement points, the reference measurement point X of an integer value is derived (step S210). Needless to say, in deriving the reference measurement point X, not only the average value but also other statistical values may be used as described in the generation of the waveform band T. However, in waveform band monitoring, comparison is performed for each measurement point, and it is necessary to adjust to an integer.

When the reference measurement point X is determined, the raw waveform data Wr is output to the measurement point adjustment unit 13, and the waveform data W after the measurement point adjustment is stored in the adjusted waveform data storage unit 10c (steps S220 to S230: measurement point adjustment). The operation of the part 13 will be described in detail). When a predetermined number (n) of waveform data W is stored in the adjusted waveform data storage unit 10c, the statistical value calculation unit 10d sets the average value Ma and the standard value Ma for each measurement point arranged in the time series order of each waveform data W. The deviation Mu is derived (steps S240 to S250).

When the calculation of the statistical value is completed, the waveform band generation unit 10e has a waveform having upper and lower thresholds (upper limit value Tu, lower limit value Thl) for each measurement point arranged in chronological order based on the calculated statistical value. A band T is generated (step S260). When the waveform band T is generated, the reference measurement point number X calculated by each part of the reference waveform data generation unit 10, the average value of each measurement point representing the reference waveform Wa, the standard deviation, and the waveform band T are used as the reference waveform data. It is stored in the reference waveform data storage unit 11. As a result, the data that will be the standard for subsequent status monitoring have been prepared.

Here, the operation of the measurement point adjusting unit 13 including steps S220 to S230 (the same applies to steps S320 to S330 in the state monitoring flow (FIG. 13) described later) will be described. The excess / deficiency determination unit 13a counts the number of measurement points of the raw waveform data Wr output from the reference waveform data generation unit 10 or the condition monitoring control unit 12, and whether or not the counted measurement points match the reference measurement point X. Is determined. For example, as shown in FIG. 7 (c), when the measurement points N of the raw waveform data Wr ₃ match the reference measurement points X, the coefficient points of the raw waveform data Wr ₃ are not adjusted and are adjusted. as already waveform data W _3, as it is returned to the output source.

On the other hand, when the number of measurement points N does not match the reference number of measurement points X, the raw waveform data Wr is output to the adjustment point determination unit 13b to determine the adjustment point. When the number of measurement points N of the raw waveform data Wr is less than the reference number of measurement points X, the point where the number of measurement points N is equally divided by the value obtained by adding 1 to the insufficient number (the part between the nearest measurement points) is the measurement point. Determine as an addition target. On the other hand, when the number of measurement points N of the raw waveform data Wr is larger than the reference number of measurement points X, the point (the latest measurement point) obtained by equally dividing the number of measurement points N by the value obtained by adding 1 to the excess number is used as the measurement point. Determine as a deletion target.

Specifically, the i-th measurement point in one waveform data will be described as Mi. As shown in FIG. 7 (a), when the number of measurement points N of the raw waveform data Wr ₁ is one less than the reference number of measurement points X (N = X-1), the measurement is performed as shown in FIG. 8 (a). The portion between the nearest measurement points of the point _{MN / 2} is determined as an addition target of the measurement points. Alternatively, as shown in FIG. 7 (b), when the number of measurement points N of the raw waveform data Wr ₂ is one more than the reference number of measurement points X (N = X + 1), the measurement is performed as shown in FIG. 8 (b). The nearest measurement point of the point _{MN / 2} is determined as the deletion target.

In this way, when the addition target or the deletion target is determined, the waveform data adjusting unit 13c receives the waveform data Wr1 from the measurement points _MN among the measurement points as described in FIG. 8A. Add a new measurement point to the part between the nearest measurement points of _{/ 2} . As the value of the measurement point to be added, for example, the average value of the measurement points on both sides is used. The order of the measurement points after the measurement point _{MN / 2} is moved down one by one (backward shift). Similarly, for the waveform data Wr2, as described with reference to FIG. 8B, the nearest measurement point of the measurement point _{MN / 2} among the measurement points is deleted. As a result, the order of the measurement points after the measurement point _{MN / 2} is advanced one by one (forward shift).

More specifically, for example, FIG. 9 (a), the (b), the case _{where M 3} is added between Mr ₂ and Mr _3, Mr ₂ is no different previous order, are Mr ₃ The order of the inserted measurement points after M ₃ shifts backward to M ₄ and Mr ₄ to M ₅ . Similarly, as shown in FIGS. 10A and 10B, when Mr ₃ is deleted, the order before Mr ₂ does not change, and Mr ₄ is inserted into M ₃ and Mr ₅ is inserted into M _4. The order of the measurement points after Mr ₃ is shifted forward.

Alternatively, as shown in FIG. 11 (a), when the number of measurement points N of the raw waveform data Wr ₁ is two less than the reference number of measurement points X (N = X-2), it is shown in FIG. 12 (a). As described above, the portion between the nearest measurement points of the measurement points _{MN / 3} and the measurement points M _{2N / 3} is determined as the target for adding the measurement points. Similarly, as shown in FIG. 11 (b), when the number of measurement points N of the raw waveform data Wr ₂ is two more than the reference number of measurement points X (N = X + 2), as shown in FIG. 12 (b). The measurement points of the nearest measurement points of the measurement points _{MN / 3} and the measurement points M _{2N / 3} are determined as deletion targets.

In that case as well, the waveform data adjusting unit 13c has a measurement point _{MN / 3} and a measurement point M _{2N / 3} among the measurement points for the raw waveform data Wr1 as described in FIG. 12 (a). A new measurement point is added between the nearest measurement points. As a result, the order of the measurement points after the measurement point _{MN / 3} is shifted backward by one, and the measurement points after the measurement point M _{2N / 3} are further shifted backward by one (two in total). Similarly, for the raw waveform data Wr2, as described in FIG. 12B, the nearest measurement points of the measurement points _{MN / 3} and the measurement points M _{2N / 3} are deleted from the measurement points. To do. As a result, the measurement points after the measurement point _{MN / 3} are shifted forward by one, and the measurement points after the measurement point M _{2N / 3} are further shifted backward by one (two in total).

In this way, the excess or deficiency of the raw waveform data Wr is returned to the output source as the adjusted waveform data W after adjusting the number of measurement points by the waveform data adjusting unit 13c. The same applies when the number of excess or deficiency is 3 or more.

As a result, the reference waveform data generation unit 10 can generate the reference waveform data by using the waveform data having the same number of measurement points. Similarly, in the abnormality presence / absence determination unit 15, the upper and lower limits of each measurement point can be compared by using the waveform data in which the measurement point number N matches the reference measurement point number X of the reference waveform Wa, and the equipment abnormality monitoring becomes possible. Hereinafter, it will be described with reference to the flowchart of FIG.

When the state monitoring control unit 12 acquires the raw waveform data Wr (FIG. 5: steps S100 to S140), the state monitoring control unit 12 confirms whether or not the reference waveform data is stored in the reference waveform data storage unit 11 (step S300). If it is not stored (“N” in step S300), the raw waveform data Wr is output to the reference waveform data generation unit 10 to execute the reference waveform data generation step (S200 to). On the other hand, when it is stored (“Y” in step S300), the acquired raw waveform data Wr is output to the measurement point adjustment unit 13 in order to use it for state monitoring.

In the measurement point adjustment unit 13, as described above, the excess / deficiency determination unit 13a first determines whether or not the measurement point N matches the reference measurement point X (step S310). When the number of measurement points N of the raw waveform data Wr does not match the reference number of measurement points X (“N” in step S310), the adjustment point determination unit 13b and the waveform data adjustment unit 13c adjust the number of measurement points to monitor the waveform. Data W is generated and returned to the state monitoring control unit 12 (steps S320 to S330). When the number of measurement points N of the raw waveform data Wr matches the reference number of measurement points X (“Y” in step S310), the output raw waveform data Wr is used as it is as the waveform data W for monitoring, and the condition monitoring control unit 12 Reply to.

The condition monitoring control unit 12 outputs the returned waveform data W for monitoring to the abnormality presence / absence determination unit 15 to determine the presence / absence of an abnormality. The abnormality presence / absence determination unit 15 compares the output waveform data W with the reference waveform data, and determines whether or not there is an abnormality based on whether or not each measurement point of the waveform data W is within the waveform band T. (Step S340). When it is within the waveform band T (“Y” in step S350), it is determined that there is an abnormality, and deterioration information is obtained based on, for example, the degree of deviation and the past history (continuous or sudden, etc.). Display or output.

As described above, by providing the measurement point number adjusting unit 13 or the step of adjusting the number of measurement points (for example, steps S310 to S330), it is possible to detect an abnormality by combining asynchronous embedded devices. In particular, since the addition or deletion target of the measurement points is set evenly according to the excess or deficient number, the position of the measurement points is not locally biased and shifted. In other words, even if there is a misalignment between the part where the defect actually occurred or the part that was acquired excessively and the part to be added or deleted, the influence can be minimized, and the influence of the measurement point can be minimized. Additions or deletions can be performed appropriately with simple operations.

In this embodiment and the next embodiment, an example of monitoring one monitoring target will be described, but the present invention is not limited to this. For example, multifaceted abnormality detection may be performed by using a plurality of condition monitoring devices to which different types of measuring devices with communication functions are connected to one monitoring target.

Embodiment 2.
In the configuration and operation of the condition monitoring device or the asynchronous data adjustment method according to the second embodiment of the present invention, there is a method of deleting or setting additional points when there is an excess or deficiency in the number of measurement points as compared with the first embodiment. Is different. Other than that, it is the same as that of the first embodiment. Therefore, the drawings and explanations of the first embodiment are used for the same parts. 14 and 15 are for explaining the operation of the state monitoring device according to the second embodiment of the present invention or the method of adjusting asynchronous data, and FIG. 14 shows a raw waveform when determining the presence or absence of an abnormality. Flow charts including an operation of setting addition or deletion points when adjusting the number of measurement points of data using the correlation with the reference waveform, FIGS. 15 (a) and 15 (b) show an excess or deficiency of one point in the number of measurement points. It is a waveform diagram for demonstrating the setting process of the addition point or the deletion point of the measurement point with respect to the waveform data.

In the first embodiment, by setting the addition / deletion points by simple even allocation according to the excess / deficient number, the influence of the misalignment is minimized without increasing the load of the arithmetic processing. , An example of adjusting the number of measurement points is shown. In the second embodiment, the load of arithmetic processing is heavier than that in the first embodiment, but the deviation itself from the part where the defect actually occurs or the part where the excessive acquisition is obtained is minimized and accurate. It enables the addition or deletion of various measurement points.

In the state monitoring device or the asynchronous data adjustment method according to the second embodiment, as shown in FIG. 14, when the state monitoring control unit 12 acquires the raw waveform data Wr (FIG. 5: steps S100 to S140), It is confirmed whether or not the reference waveform data is stored in the reference waveform data storage unit 11 (step S400). If it is not stored (“N” in step S400), the raw waveform data Wr is output to the reference waveform data generation unit 10 to execute the reference waveform data generation step (S200 to). On the other hand, when it is stored (“Y” in step S400), the acquired raw waveform data Wr is output to the measurement point adjustment unit 13 in order to use it for state monitoring.

In the measurement point adjustment unit 13, as described above, the excess / deficiency determination unit 13a first determines whether or not the measurement point N matches the reference measurement point X (step S410). When the number of measurement points N of the raw waveform data Wr matches the reference number of measurement points X (“Y” in step S410), the output raw waveform data is directly used as the waveform data W for monitoring and is sent to the condition monitoring control unit 12. Reply. On the other hand, when the number of measurement points N of the raw waveform data Wr does not match the reference number of measurement points X (“N” in step S410), the adjustment point determination unit 13b sets the addition / deletion target (step S420). The waveform data adjusting unit 13c adjusts the number of measurement points of the raw waveform data Wr according to the set addition / deletion target, generates the waveform data W for monitoring, and returns it to the state monitoring control unit 12 (step S420). ~ S430).

Here, a method of determining an adjustment point when the number of measurement points N is one point excess or deficiency with respect to the reference number of measurement points X will be described with reference to FIG. When the number of shortages is 1, as shown in FIG. 15A, the adjustment point determination unit 13b inserts the measurement points sequentially between all the measurement points and at positions outside the measurement points, and the provisional waveform obtained by inserting the measurement points. The correlation between and the reference waveform Wa is calculated and stored. Then, the insertion point where the temporary waveform having the highest correlation can be formed is set as the insertion point for score adjustment (step S420). The operation of the waveform data adjusting unit 13c after the setting is the same as that of the first embodiment. The average value may be used between the measurement points as in the first embodiment, but in the case of the outer position, extrapolation may be performed from the inclination of the nearby measurement points. For example, when adding a measurement point in front of Mr ₁ may be the difference between the Mr ₂ and Mr ₁ to the value obtained by dividing the value of Mr _1. Similarly, if you add a measurement point behind Mr _N may be the difference between the Mr _N and _{Mr N-1} to a value obtained by adding the Mr _N.

Alternatively, when the excess number is 1, the adjustment point determination unit 13b sequentially deletes all the measurement points, and the correlation between the provisional waveform obtained by deleting and the reference waveform Wa, as shown in FIG. 15 (b). Is calculated and stored. Then, the deleted portion where the temporary waveform having the highest correlation can be formed is set as the deleted portion for score adjustment (step S420).

The operation after step S440 is the same as that after step S340 of FIG. 13 described in the first embodiment. As a result, the number of measurement points can be adjusted by adding or deleting measurement points at accurate positions that minimize the deviation from the points where defects actually occur or the points that have been excessively acquired.

It is necessary to create temporary waveform data by multiplying the number of deletion or addition targets by the number of excess or deficiency, but even if the number of excess or deficiency is 2 or more, this method of setting the deletion or addition target Is valid.

As described above, according to the state monitoring device 1 according to the first or second embodiment of the present invention, the state of the device (production equipment 2) that is expected to operate at a constant operation for each cycle is set to one cycle. A state monitoring device 1 that monitors by upper / lower limit judgment (or threshold judgment) for each measured value constituting minute waveform data, and a measured value indicating the state of the production equipment 2 from the measuring device 3 that does not respond to time information. The measurement value collection unit (transmission / reception unit 16) to be collected, the waveform processing unit 14 that arranges the measurement values collected by the transmission / reception unit 16 in chronological order and generates the waveform data for one cycle of the production equipment 2, and the production equipment 2 As the reference waveform Wa, which is the waveform data for one cycle when is normal, the upper limit value Tu and the lower limit value Thl (or either threshold) for each measured value of the reference waveform Wa, and the number of measurement points of the reference waveform Wa. The reference waveform data storage unit 11 that stores the reference waveform data including the determined reference measurement points X and the raw waveform data Wr generated by the waveform processing unit 14 are adjusted and output for comparison with the reference waveform data. Abnormality determination unit 15 that compares the waveform data adjustment unit (measurement point adjustment unit 13) and the waveform data W output by the measurement point adjustment unit 13 with the reference waveform data, and determines whether or not there is an abnormality in the production equipment 2. When the measurement point N of the raw waveform data Wr does not match the reference measurement point X, the measurement point adjustment unit 13 sets the measurement point Mi of the raw waveform data Wr or between the measurement points with a predetermined reference. By deleting the measured measurement points Mi or adding measurement points between the measurement points, the raw waveform data Wr is adjusted so as to match the reference measurement point number X. As a result, even if a general-purpose measuring device that does not respond to time information is used, condition monitoring using the waveform band T can be accurately executed.

Further, as in the condition monitoring device 1 according to the first embodiment, the waveform data adjustment unit (measurement point adjustment unit 13) responds to the excess or deficiency number from the reference measurement point number X of the measurement point number N of the raw waveform data Wr. , Since the position where the number of measurement points N of the raw waveform data Wr is evenly divided is set as the target for deleting or adding the measurement points, the place where the defect actually occurs or the excess is not increased. Even if there is a misalignment between the acquired part and the addition or deletion target, the effect can be minimized and accurate status monitoring can be performed.

Alternatively, as in the state monitoring device 1 according to the second embodiment, the waveform data adjusting unit (measurement point adjustment unit 13) has an excess or deficiency from the reference measurement point X for all the measurement points of the raw waveform data Wr. Since the temporary waveform data generated each time the measurement points of the above are sequentially deleted or added are compared with the reference waveform Wa, the temporary waveform data having the highest correlation with the reference waveform Wa is output as the adjusted waveform data W. , It is possible to adjust the number of measurement points by adding or deleting measurement points at accurate positions that minimize the deviation from the points where defects actually occur or the points that have been acquired excessively.

Further, the reference measurement point X is calculated from the waveform data for a plurality of times collected when the device (production equipment 2) is normal, and the waveform data adjustment unit (measurement point adjustment unit 13) is based on the calculated reference measurement point X. If the waveform data W obtained by adjusting the waveform data (raw waveform data Wr) for a plurality of times is used and the reference waveform data generation unit 10 for generating the reference waveform data Wa is provided, a location where a defect actually occurs or a location where a defect actually occurs or The number of measurement points can be adjusted by adding or deleting measurement points at accurate positions that minimize the deviation from the excessively acquired points.

It has an operating state determination unit (a part of the function of the condition monitoring control unit 12) that determines whether the device (production equipment 2) is in the ON state or the OFF state based on the signal received from the outside, and has a measurement value collection unit ( If the transmission / reception unit 16) starts or stops collecting the measured values from the measuring device 3 based on the determination result of the operating state determination unit, the waveform data for one cycle can be obtained without human intervention. It can be acquired and status monitoring is possible.

As described above, according to the method for adjusting asynchronous data according to the first or second embodiment of the present invention, waveform data in which asynchronous measurement values output from the measuring device 3 that does not respond to time information are arranged in chronological order is obtained. It is a method of adjusting waveform data for use in monitoring the state of an apparatus (production equipment 2) that is expected to operate at a constant operation every cycle, and is a method of adjusting the waveform data from the measuring device 3 to the state of the production equipment 2. A measurement value collection step (steps S120 to S130) for collecting measurement values indicating the above, and a waveformization step for arranging the measurement values collected in the measurement value collection step in chronological order to generate waveform data for one cycle of production equipment 2. (Step S140), the reference waveform Wa which is the waveform data for one cycle when the production equipment 2 is normal, and the upper limit value Tu and the lower limit value Thl (or either threshold value) for each measured value of the reference waveform Wa. , And the reference waveform data storage step for storing the reference waveform data including the reference measurement point X defined as the measurement points of the reference waveform Wa, and the raw waveform data Wr generated in the waveformization step are compared with the reference waveform data Wa. The waveform data adjustment step (step S330 or S430)) adjusted and output for the purpose and the waveform data W output in the waveform data adjustment step are compared with the reference waveform data, and whether or not there is an abnormality in the production equipment 2. When the presence / absence of abnormality determination step (steps S340 to S360 or S440 to S460) is provided, and the measurement points N of the raw waveform data do not match the reference measurement points X in the waveform data adjustment step (step S310, Or S410), among the measurement points Mi of the raw waveform data Wr or between the measurement points, the measurement points set by the predetermined reference are deleted, or the measurement points are added between the measurement points so that the measurement points match the reference measurement points. Since the raw waveform data is adjusted (steps S320 to S330 or S420 to S430), the asynchronous data collected from general-purpose measuring equipment can be adjusted so that the state can be easily monitored. it can.

Further, as in the method of adjusting the asynchronous data according to the first embodiment, in the waveform data adjustment step, the raw waveform data Wr is converted according to the excess or deficient number from the reference measurement point X of the measurement point number N of the raw waveform data Wr. If the position where the number of measurement points is evenly divided is set as a target for deleting or adding measurement points (steps S320 to S330), the place where the defect actually occurs or excessively without increasing the load of arithmetic processing. Even if there is a misalignment between the acquired part and the addition or deletion target, the data can be adjusted with the effect minimized.

Alternatively, as in the asynchronous data adjustment method according to the second embodiment, in the waveform data adjustment step, the excess or deficiency number of measurement points from the reference measurement point X is sequentially deleted from all the measurement points of the raw waveform data Wr. Alternatively, the temporary waveform data generated each time it is added is compared with the reference waveform Wa, and the temporary waveform data having the highest correlation with the reference waveform Wa is output as adjusted waveform data (steps S420 to S430). Data can be adjusted by adding or deleting measurement points at accurate positions that minimize deviations from locations where defects actually occur or where they have been acquired excessively.

In each of the above embodiments, an example of upper / lower limit determination for determining the presence / absence of an abnormality based on both the upper limit value Thu and the lower limit value Thl has been shown, but the present invention is not limited to this. The same effect can be obtained by the threshold value judgment which is judged by at least one threshold value.

1: Condition monitoring device, 2: Production equipment (equipment), 3: Measuring equipment,
10: Reference waveform data generation unit, 11: Reference waveform data storage unit, 12: Condition monitoring control unit, 13: Measurement point adjustment unit (waveform data adjustment unit), 14: Waveform processing unit, 15: Abnormality presence / absence determination unit, 16: Transmission / reception unit (measured value collection unit),
Mi: Measured value, N: Number of measurement points, T: Waveform band, Thl: Lower limit value (threshold value), Tu: Upper limit value (threshold value), W: Waveform data, Wa: Reference waveform, Wr: Raw waveform data, X: Reference Number of measurement points.

Claims (8)

It is a condition monitoring device that monitors the state of the device that is expected to operate with a constant operation for each cycle by determining the threshold value for each measured value that constitutes the waveform data for one cycle.
A measurement value collection unit that collects measurement values indicating the state of the device from a measurement device that does not respond to time information.
A waveform processing unit that arranges the measured values collected by the measured value collecting unit in chronological order and generates waveform data for one cycle of the apparatus.
Reference waveform data including a reference waveform which is waveform data for one cycle when the device is normal, a threshold value for each measured value of the reference waveform, and a reference measurement point determined as the measurement point of the reference waveform. Reference waveform data storage unit to be stored and
A waveform data adjusting unit that adjusts and outputs the raw waveform data generated by the waveform processing unit in order to compare it with the reference waveform data.
It is provided with an abnormality presence / absence determination unit that compares the waveform data output by the waveform data adjustment unit with the reference waveform data and determines whether or not there is an abnormality in the apparatus.
When the number of measurement points of the raw waveform data does not match the reference measurement points, the waveform data adjusting unit deletes the measurement points of the raw waveform data or the measurement points set by a predetermined reference among the measurement points of the raw waveform data. A condition monitoring device characterized in that the raw waveform data is adjusted so as to match the reference measurement points by adding measurement points between measurement points. The waveform data adjusting unit deletes or adds the measurement points to the positions where the measurement points of the raw waveform data are evenly divided according to the excess or deficiency of the measurement points of the raw waveform data from the reference measurement points. The condition monitoring device according to claim 1, wherein the condition monitoring device is set. The waveform data adjusting unit compares the provisional waveform data generated each time the excess or deficiency number of measurement points from the reference measurement points is sequentially deleted or added to all the measurement points of the raw waveform data with the reference waveform. The state monitoring device according to claim 1, wherein the provisional waveform data having the highest correlation with the reference waveform is output as adjusted waveform data. The reference measurement points are calculated from the waveform data of a plurality of times collected when the apparatus is normal, and the waveform data adjusting unit adjusts the waveform data of the plurality of times based on the calculated reference measurement points. The state monitoring device according to any one of claims 1 to 3, further comprising a reference waveform data generation unit that generates the reference waveform data. It has an operating state determination unit that determines whether the device is in the ON state or the OFF state based on a signal received from the outside.
According to any one of claims 1 to 4, the measured value collecting unit starts or stops collecting measured values from the measuring device based on the determination result of the operating state determination unit. The described condition monitoring device. Waveforms for monitoring the status of equipment that is expected to operate at a constant rate for each cycle, using waveform data in which asynchronous measurement values output from measuring equipment that does not respond to time information are arranged in chronological order. It ’s a way to adjust the data,
A measurement value collection step for collecting measurement values indicating the state of the device from the measurement device, and
A waveform-forming step in which the measured values collected in the measured value collecting step are arranged in chronological order and waveform data for one cycle of the apparatus is generated.
Reference waveform data including a reference waveform which is waveform data for one cycle when the device is normal, a threshold value for each measured value of the reference waveform, and a reference measurement point determined as the measurement point of the reference waveform. Reference waveform data storage step to be stored and
A waveform data adjustment step that adjusts and outputs the raw waveform data generated in the waveformization step in order to compare it with the reference waveform data, and
It has an abnormality presence / absence determination step of comparing the waveform data output in the waveform data adjustment step with the reference waveform data and determining whether or not there is an abnormality in the apparatus.
In the waveform data adjustment step, when the number of measurement points of the raw waveform data does not match the reference measurement points, the measurement points set by a predetermined reference among the measurement points or measurement points of the raw waveform data are deleted, or A method for adjusting asynchronous data, which comprises adjusting the raw waveform data so as to match the reference measurement points by adding measurement points between measurement points. In the waveform data adjustment step, the position where the number of measurement points of the raw waveform data is evenly divided according to the excess or deficiency of the number of measurement points of the raw waveform data from the reference measurement point is set as the target for deleting or adding the measurement points. The method for adjusting asynchronous data according to claim 6, wherein the data is set. In the waveform data adjustment step, the provisional waveform data generated each time the excess or deficiency number of measurement points from the reference measurement points is sequentially deleted or added to all the measurement points of the raw waveform data is compared with the reference waveform. The method for adjusting asynchronous data according to claim 6, wherein the provisional waveform data having the highest correlation with the reference waveform is output as adjusted waveform data.

Images