JP2006155200A - Load operation status monitoring apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To correctly monitor a load operation status without erroneous determination even when time variation exists in the load operation status. <P>SOLUTION: A load operation status monitoring apparatus comprises a measurement means of change in the amount of consumed electricity to measure the amount of the consumed electricity of a power supply system in order to measure the change in the amount of the consumed electricity per predetermined period, a change pattern learning means to learn a change pattern of the change in the amount per predetermined period measured by the measurement means of the change in the amount of consumed electricity, and an abnormality detection means to detect abnormalities of a load operation status by comparing the change pattern learned by the change pattern learning means with the amount of change per predetermined period in an amount of actually consumed electricity measured by the measurement means of the amount of consumed electricity. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a load operation status monitoring device that monitors an operation status of a load in a facility in which a plurality of loads are connected to a power supply system.

As a conventional load operation status monitoring device, for example, the monitoring target is a machine tool, the load current of the descent motor is fine, and sampling points on the machining time axis are finely divided for complex machining points in units of one cycle from the start to the end of machining. In addition, the sampling points on the processing time axis are roughly taken at simple machining points, and the sampling data is stored and numerically processed at each sampling point, and the measured load current is determined by the sampling data at the sampling points. A machine tool monitoring method is known in which a work process is monitored by comparing whether the standard deviation value is within an upper or lower limit range (see, for example, Patent Document 1).
JP 2002-341909 (first page, FIG. 1)

However, in the conventional example described in Patent Document 1 above, when the upper and lower limit waveforms are created, the time variation of the load variation is not taken into consideration, so that when the load varies with time, it is normal. However, there is an unsolved problem that a phenomenon deviating from the upper and lower limit waveforms occurs and accurate monitoring cannot be performed.
In addition, since the method of determining the sampling point on the time axis regardless of the workpiece shape data determines the sampling time interval in the first cycle, measurement is performed at the determined time interval from the second cycle. There is an unsolved problem that there is no factor for shortening the time interval by the correction after the second cycle, and the time interval does not become finer than the value determined in the first cycle.

  In addition, in the sampling process, if the specified number of samples is not acquired, the next step will not proceed, so if an external trigger signal or internal trigger signal is generated before the specified number of samples is reached due to missing measurements, etc., the trigger signal There is also an unsolved problem that the routine for changing the next sampling rate is not reached using the key as a key, and the measurement data of the next sampling location is originally recorded as the data of the previous sampling location.

Furthermore, since the work processing work of the machine tool is monitored, the standby current is not monitored, so there is no sign of abnormality or abnormality in the equipment such as the standby power is larger than usual or the equipment is stopped. There is also an unsolved problem that it cannot be detected.
Therefore, the present invention has been made paying attention to the unsolved problems of the conventional example described above, and by patterning and learning the change in the amount of electricity consumed every predetermined time, there is a temporal variation in the operating status of the load. Even if there is a load, it is possible to accurately monitor the load operation status without making a misjudgment, and to monitor the overall load operation status including standby time without specifying the number of samples The object is to provide an operational status monitoring device.

  In order to achieve the above object, a load operation status monitoring device according to claim 1 is a load operation status monitoring device for monitoring a load operation status of a facility in which a plurality of loads are connected to a power supply system, wherein the power supply system An electricity consumption change measuring means for measuring the amount of electricity consumed and measuring the amount of change of the electricity consumption per predetermined time, and a change pattern of the amount of change per predetermined time measured by the electricity consumption change measuring means. The change pattern learning means for learning, the change pattern learned by the change pattern learning means, and the amount of change in the actual electricity consumption measured by the electricity consumption measuring means for each predetermined time are compared, and the load operation status abnormality And an abnormality detecting means for detecting.

  The load operation status monitoring device according to claim 2 is a load operation status monitoring device that monitors the load operation status of a facility in which a plurality of loads are connected to the power supply system, and measures the amount of electricity consumed by the power supply system. And a change pattern learning means for learning a change pattern of a change amount per predetermined time measured by the consumption amount change measurement means. And an abnormality detection means for detecting an abnormality in the load operation status by comparing a change pattern learned by the change pattern learning means and a change amount of the actual electricity consumption measured by the electricity consumption measuring means for every predetermined time And a power supply stopping means for stopping power supply to each load of the power supply system when the abnormality detection means detects an abnormality in the load operation status.

Furthermore, the load operation status monitoring device according to claim 3 is the invention according to claim 1 or 2, wherein the electricity consumption change measuring means uses any one of current, electric power and electric energy as the electricity consumption. It is characterized in that it is configured to measure and measure the amount of change per predetermined time.
Furthermore, according to a fourth aspect of the present invention, in the load operation status monitoring device according to any one of the first to third aspects, the change pattern learning means changes the consumed electricity amount measured by the consumed electricity amount change measuring means. It is characterized in that the pattern is formed by dividing the quantity into three states of increasing tendency, decreasing tendency, and no change.

  Still further, according to a fifth aspect of the present invention, in the load operation status monitoring device according to any one of the first to fourth aspects, the change pattern learning means is a change in the consumed electricity amount measured by the consumed electricity amount change measuring means. The pattern is formed by dividing the quantity into three states of increasing tendency, decreasing tendency, and no change, and when the same pattern continues, it is compressed to form a compressed pattern string. If it is different from the compressed pattern sequence, the compressed pattern sequence is registered as a new compressed pattern sequence.

  According to the first aspect of the present invention, the electricity consumption change measuring means measures a change in electricity consumption every predetermined time of the power supply system, and the change pattern learning means based on the measurement result changes the change amount change pattern. The load change status is different from the learned change pattern because the load change status is detected by comparing the learned change pattern with the change amount of the actual electricity consumption every predetermined time. When the state is reached, the amount of change in the actual amount of electricity consumed every predetermined time deviates from the learned change pattern, and the effect that the abnormality in the operation status can be detected accurately is obtained. At this time, the change pattern learning means constantly learns the change in the amount of electricity consumed, so that it is possible to accurately detect an abnormality in the operation status regardless of the standby time, operating time, and the like.

According to the invention of claim 2, in addition to the configuration of the invention of claim 1, when an abnormality in the load operation status is detected by the abnormality detection means, power supply to each load is performed by the power supply stop means. Therefore, if an abnormality such as an overload occurs at any load, the power supply from the power supply system should be stopped immediately to prevent the abnormal state from continuing. The effect of being able to be obtained.
Furthermore, according to the invention of claim 3, the electricity consumption change measuring means measures any one of current, power, and power amount as the consumed electricity amount. Since it changes according to the operation status, the effect of being able to accurately detect the operation status of the load is obtained.

  Furthermore, according to the invention according to claim 4, the change pattern learning means sets the amount of change in the electricity consumption measured by the electricity consumption change measuring means into three states of increasing tendency, decreasing tendency, and no change. Since it is configured to form a pattern row by dividing the pattern, for example, as in a factory, it keeps a constant value after increasing the amount of electricity consumed in the morning, and then when it is lunch break, the amount of electricity consumed is When the power consumption decreases and the power consumption increases in the afternoon, the electricity consumption increases again to a constant value.At the end of the day, the electricity consumption decreases, and this trend increases, decreases, and changes. By representing the three states without, the effect that the load operation status can be accurately patterned is obtained.

  Still further, according to the invention of claim 5, the detected pattern string is compressed to form a compressed pattern string, and a new compression is performed when the formed compressed pattern string is different from the existing compressed pattern string registered. Because it is registered as a pattern string, different pattern strings are registered in order, and the abnormality determination is accurately performed by comparing all the registered compressed pattern strings with the actually measured change amount per predetermined time. The effect of being able to be obtained.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a system configuration diagram showing a schematic configuration of the present invention. In the figure, 1 is a power supply line as a power supply system arranged in a factory, and this power supply line 1 is connected via a breaker 2. Connected to the distribution board 3. A plurality of, for example, five loads (for example, electric lights) L1 to L5 are connected to the power supply line 1.

A current detector 4 for detecting the load current as the amount of electricity consumed is disposed on the output side of the breaker 2 of the power supply line 1, and the detected current detected by the current detector 4 is different from that of the power supply line 1. The power is supplied to a load operation status monitoring device 5 supplied with power from a power supply line (not shown).
In this load operation state monitoring device 5, an A / D converter 6 that converts a detected current detected by the current detector 3 into a digital value, and a digital detection value of the A / D converter 6 is input, for example, a microcomputer. Connected to the arithmetic processing unit 7 via the display control circuit 9, the non-volatile memory 8 for storing the detected current value and the change pattern sequence connected to the arithmetic processing unit 7, and the arithmetic processing unit 7. Display device 10.

The arithmetic processing unit 7 includes at least a CPU 7a, a ROM 7b, and a RAM 7c, and performs a load management process by executing a load monitoring process program stored in the ROM 7b in advance by the CPU 7a.
As shown in FIG. 2, in the load monitoring process, first, in step S1, it is determined whether or not the preset learning period has ended. If the learning period has not ended, the process proceeds to step S2. After the learning process shown in FIG. 3 is executed, the process returns to step S1, and when the learning period ends, the process proceeds to step S3, the abnormality determination process shown in FIG. 7 is executed, and then the process returns to step S1.

  As shown in FIG. 3, in the learning process, first, it is determined whether or not the learning period preset in step S11 has ended. If the learning period has not ended, the process proceeds to step S12 to perform learning. Therefore, it is determined whether or not the current measurement time is set in advance, and if it is not the current measurement time, the learning process is ended as it is, and the process returns to step S1 in FIG. 2, and if it is the current measurement time, the process proceeds to step S13. After starting the current measurement process shown in FIG. 4, the process returns to step S1 in FIG.

On the other hand, if the result of determination in step S11 is that the learning period has ended, the process proceeds to step S14 to execute the code compression process shown in FIG. 5, and then the process proceeds to step S15 to determine the compression pattern sequence shown in FIG. After executing the processing, the learning processing is terminated.
In the current measurement process shown in FIG. 4, first, in step S21, the detected current value I (i) detected by the current detector 3 is read from the A / D converter 6, and then the process proceeds to step S22. After the current detection current value I (i) is stored in the detection current storage area at the corresponding time formed in the nonvolatile memory 8, the process proceeds to step S23.

In this step S23, the detected current value I (i-1) at the previous time stored in the non-volatile memory 8 is read, and then the process proceeds to step S24 where the calculation of the following equation (1) is performed and the current change amount ΔI. (I) is calculated, and the calculated current change amount ΔI (i) is temporarily stored in the current change amount storage area at the corresponding time formed in the nonvolatile memory 8.
ΔI (i) = {I (i) −I (i−1)} / T (1)
Here, T is a timer interruption period.

  Next, the process proceeds to step S25, and it is determined whether or not the current change amount ΔI (i) is increasing. In this determination, it is determined whether or not the current change amount ΔI (i) exceeds a preset positive threshold + α larger than zero. If ΔI (i)> + α, it is determined that the current change amount ΔI (i) is increasing. The process proceeds to step S26, and the current measurement process is terminated after the code “a” representing the increasing tendency is written in the state storage area of the corresponding time formed in the nonvolatile memory 8.

  When the determination result in step S25 is ΔI (i) ≦ + α, the process proceeds to step S27 to determine whether or not the current change amount ΔI (i) is not changed. In this determination, it is determined whether or not the current change amount ΔI (i) is equal to or less than a negative threshold −α that is less than or equal to + α, and when + α ≧ ΔI (i) ≧ −α, the current change amount ΔI ( i) It is determined that there is no change with no change, the process proceeds to step S28, and the code “b” indicating the no change state is written in the state storage area at the corresponding time formed in the nonvolatile memory 8, and then the current End the measurement process.

Furthermore, when the determination result in step S27 is ΔI (i) <− α, it is determined that the change in the current change amount ΔI (i) has a decreasing tendency, the process proceeds to step S29, and the symbol “c” indicating the decreasing tendency is displayed. "" Is written in the state storage area at the corresponding time formed in the nonvolatile memory 8, and the current measurement process is terminated.
In the code compression processing of FIG. 5, first, in step S111, the variable i is set to “0”, and then the process proceeds to step S112 where the code of the time T (i) stored in the nonvolatile memory 7 is set. C (i) is read, and then the process proceeds to step S113, where it is determined whether or not the read code C (i) is the first code C (1). The process jumps to step S117, and if it is not the first code C (1), the process proceeds to step S114.

  In this step S114, the code C (i-1) at the time T (i-1) is read, and then the process proceeds to step S115 to check whether the codes C (i) and C (i-1) match. If both codes are matched, the process proceeds to step S116, and the number of consecutive codes t stored in the data writing area of the nonvolatile memory 8 and the minimum / maximum current values Imin · Imax, The minimum / maximum change amount ΔImin · ΔImax is stored in the data storage area of the nonvolatile memory 8 in association with the current compression code CC (i), and then the process proceeds to step S117.

In step S117, the detected current I at time T (i) is set to the detected current minimum value Imin, and the current change amount ΔI is set to the minimum change amount ΔImin, and these are temporarily stored in the nonvolatile memory 8.
Next, the process proceeds to step S118, where the detected current I (i) is set to the detected current maximum value Imax and the current change amount ΔI (i) is set to the maximum change amount ΔImax, and these are temporarily stored in the nonvolatile memory 8. After temporarily storing in the area, the process proceeds to step S119.
In step S119, the number of code continuations t is set to “1”, and then the process proceeds to step S120, and the compression pattern formed in the nonvolatile memory 8 with the current code C (i) as the compression code CC (j). The data is stored at the end of the column storage area, and the process proceeds to step S121. The variable i is incremented by “1”, and then the process proceeds to step S122.

  In step S122, it is determined whether or not the code C (i) corresponding to the variable i exists. If C (i) exists, the process returns to step S112. If not, the process returns to step S123. As in step S116 described above, the current number of code continuations t, minimum / maximum current values Imin / Imax, and minimum / maximum changes ΔImin / ΔImax stored in the range data writing area of the RAM 7c are compressed. The code compression process is terminated after the data is stored in the data storage area of the nonvolatile memory 8 in association with the code CC (i), and the process proceeds to step S15 in FIG.

  When the determination result in step S115 is that the codes C (i) and C (i-1) are different, the process proceeds to step S124, and the detected current I (i) at time i is detected current minimum value Imin. If I (i) ≧ Imin, the process jumps directly to step S126. If I (i) <Imin, the process proceeds to step S125, and the detected current I (i) is reduced to the minimum detected current. After updating and storing the value Imin in the range data storage area of the nonvolatile memory 8, the process proceeds to step S126.

  In step S126, it is determined whether or not the detected current I (i) at time i is larger than the minimum detected current value Imax. If I (i) ≦ Imax, the process jumps directly to step S128, where I (i)> Imax If it is, the process proceeds to step S127, the detected current I (i) is updated and stored in the range data storage area of the nonvolatile memory 8 as the detected current maximum value Imax, and then the process proceeds to step S128.

  In step S128, it is determined whether or not the current change amount ΔI (i) is smaller than the minimum change amount ΔImin. If ΔI (i) ≧ ΔImin, the process jumps directly to step S130 and ΔI (i) < When it is ΔImin, the process proceeds to step S129, the current change amount ΔI (i) is updated and stored in the range data storage area of the nonvolatile memory 8 as the minimum change amount ΔImin, and then the process proceeds to step S130.

In step S130, it is determined whether or not the current change amount ΔI (i) is larger than the maximum change amount ΔImax. If ΔI (i) ≦ ΔImax, the process jumps directly to step S132 and ΔI (i)> When it is ΔImax, the process proceeds to step S131, the current change amount ΔI (i) is updated and stored in the range data storage area of the nonvolatile memory 8 as the maximum change amount ΔImax, and then the process proceeds to step S132.
In step S132, the code sequence count t is incremented by "1", and then the process proceeds to step S121.

  Further, as shown in FIG. 6, in the compression pattern sequence determination process in step S15 in FIG. 3, first, in step S200, the variable j is set to “1”, and then the process proceeds to step S201, where the nonvolatile memory 8 The compression code CC (j) is read from the compression pattern sequence stored in the, and then the process proceeds to step S202 to determine whether or not the read compression code CC (j) is the code b. If there is, the process proceeds to step S203.

  In this step S203, it is determined whether or not the maximum detected current value Imax that is the range data of the compression code CC (i) is less than a set value, for example, the minimum current consumption value ia of the loads L1 to L5, and Imax <ia. Sometimes, the process proceeds to step S204, where it is determined whether or not the subsequent compression code CC (i + 1) is “a”, and when it is “a”, that is, when the continuous code is “ba”, the compression code CC (i ) Is determined to be the head of the compression pattern sequence, the process proceeds to step S205, and the compression code CC (i) = b is stored in the compression pattern sequence storage area of the nonvolatile memory 8 and the compression pattern b is stored. To step S206.

  In this step S206, the variable j is incremented by “1”, and then the process proceeds to step S207, where it is determined whether or not the compression code CC (j) of the new variable exists in the non-volatile memory 8. If the compression code CC (j) is not present, the compression pattern sequence determination process is terminated.

  On the other hand, if the determination result in step S202 is that the compression code CC (j) is not the code “b”, the process proceeds to step S208 to determine whether or not the compression code CC (j) is the code “c”. When the code is “c”, the process proceeds to step S209, where it is determined whether or not the compression code CC (j + 1) next to the compression code CC (j) is the code “b”. The process proceeds to step S210, where it is determined whether or not the maximum detected current value Imax of the compression code CC (j + 1) is less than the set value ia. If Imax <ia, it is the last code of the compression pattern sequence. Determination is made and the process proceeds to step S212.

  In this step S212, the compression code string is determined by adding the compression code CC (j) and the compression code CC (j + 1) to the end of the compression pattern string storage area formed in the nonvolatile memory 8. Next, the process proceeds to step S213, and it is determined whether or not an existing compressed pattern sequence that matches the compressed pattern sequence stored in the compressed pattern sequence storage area of the nonvolatile memory 8 exists in the nonvolatile memory 8. If there is no matching existing compressed pattern string, it is determined as a new compressed pattern string, and the process proceeds to step S214. The compressed pattern string and its range data are stored in the existing compressed pattern string storage area of the nonvolatile memory 8. The process proceeds to step S206, and if there is a matching existing compressed pattern string, the process proceeds to step S215, and the compressed pattern string stored in the compressed pattern string storage area of the nonvolatile memory 8 is stored. Are compared with the range data stored in the existing compressed pattern string storage area, and the minimum value tmin and maximum of the code continuation number t are compared. tmax, shifts the minimum amount of change DerutaImin, maximum variation ΔImax and detected current minimum value Imin, after performing the process of updating the detected current maximum value Imax to the step S206.

  On the other hand, when the determination result in step S203 indicates that the maximum detected current value Imax of the compression code CC (j) is greater than or equal to the set value ia, the determination result in step S204 indicates that the compression code CC (j + 1) is other than the code “a”. When the determination result in step S208 is that the compression code CC (j) is other than the code “c”, the determination result in step S209 is that the code CC (j + 1) is other than the code “b”. The process proceeds to S211 and the compression code CC (j) is stored at the end of the compression pattern string storage area of the nonvolatile memory 8, and then the process proceeds to Step S206.

  Further, as shown in FIG. 7, the abnormality determination process is executed as a timer interruption process for the main program every predetermined time (for example, 1 hour) between predetermined times of the day as in the learning process described above. In step S31, the detected current value I (i) is read from the A / D converter 5, then the process proceeds to step S32, the previous detected current value I (i-1) is read, and then the process proceeds to step S33. The current change amount ΔI (i) is calculated based on the equation (1) in the same manner as in step S14 of the learning process described above.

  Next, the process proceeds to step S34, and it is determined whether or not the current change amount ΔI (i) is increasing. In this determination, it is determined whether or not the current change amount ΔI (i) exceeds a preset positive threshold + α larger than zero. If ΔI (i)> + α, it is determined that the current change amount ΔI (i) is increasing. The process proceeds to step S35, and the code “a” indicating the increasing tendency is written in the state storage area at the corresponding time formed in the nonvolatile memory 8, and then the process jumps to step S39.

When the determination result in step S34 is ΔI (i) ≦ + α, the process proceeds to step S36 to determine whether or not the current change amount ΔI (i) is not changed. In this determination, it is determined whether or not the current change amount ΔI (i) is equal to or less than a negative threshold −α that is less than or equal to + α, and when + α ≧ ΔI (i) ≧ −α, the current change amount ΔI ( i) It is determined that there is no change without change, and the process proceeds to step S37, where the code “b” indicating the no change state is written in the state storage area at the corresponding time formed in the nonvolatile memory 8 and then the step. Jump to S39.
Furthermore, when the determination result in step S36 is ΔI (i) <− α, it is determined that the change in the current change amount ΔI (i) has a decreasing tendency, the process proceeds to step S38, and the code “c” indicating the decreasing tendency is displayed. "" Is written in the state storage area of the corresponding time formed in the nonvolatile memory 8, and then the process proceeds to step S39.

  In this step S39, it is determined whether or not it is the measurement start time T (1). If it is the measurement start time T (1), the process proceeds to step S40, and the number of consecutive codes t is set to “1”. From step S41, it is determined whether or not the non-volatile memory 7 has a compressed pattern string having, as the first code C (1), a code that matches the obtained code C (i) (= ac). If they match, the process proceeds to step S42, and the detected current value I (i) is within the range of the detected current minimum value Imin and the detected current maximum value Imax corresponding to the first code C (1). If Imin ≦ I (i) ≦ Imax, it is determined that the detected current is normal, and the process proceeds to step S43, where the current change amount ΔI (i) is the minimum change amount ΔImin and the maximum change amount. Within the range of ΔImax Whether or not, and when ΔImin ≦ ΔI (i) ≦ ΔImax, it is determined to be normal, the process proceeds to step S44, and the compressed pattern sequence having the code obtained in step S35, S37 or S38 as the head code Is stored in the RAM 7c, and the timer interrupt process is terminated.

  On the other hand, when the determination result of step S41 does not match the first code C (1) of each compression pattern sequence, and when the determination result of step S42 is I (i) <Imin or I (i)> Imax, When the determination result in step S43 is ΔI (i) <ΔImin or ΔI (i)> ΔImax, it is determined that the current consumption is abnormal, the process proceeds to step S45, and a breaker command for shutting off the breaker 2 is issued. After the output to 2, the timer interrupt process is terminated.

  If the determination result in step S39 is not the measurement start time T (1), the process proceeds to step S46, and the code C (i) obtained in any of steps S35, S37, and S38 is the previous code C. It is determined whether or not there is a code change that does not match (i-1). When there is no code change, the process proceeds to step S47, and when there is a code change, the process proceeds to step S51.

  In step S47, the code continuation number t is incremented by “1”, and then the process proceeds to step S48 to determine whether or not the code continuation number t exceeds the maximum number tmax of the corresponding compressed pattern sequence, and t> tmax. If it is, it is determined that the load operating status is abnormal, and the process proceeds to step S45. If t ≦ tmax, it is determined that the load operating status is normal and the process proceeds to step S49.

  In step S49, the detected current value I (i) corresponding to the current code C (i) of the compressed pattern sequence stored in the nonvolatile memory 7 and the detected current value I (i) are detected as in step S42 described above. It is determined whether or not the current is within the range of the maximum current value Imax. If it is out of the range, it is determined that the operation state of the load is abnormal, and the process proceeds to step S45, where Imin ≦ I (i) ≦ Imax. Sometimes, it is determined that the operation status of the load is normal, and the process proceeds to step S50.

  In step S50, the current change amount ΔI (i) is within the range of the minimum change amount ΔImin and the maximum change amount ΔImax corresponding to the current code C (i) of the compressed pattern sequence stored in the nonvolatile memory 7. When it is out of the range, it is determined that the load operation status is abnormal, and the process proceeds to step S45. When ΔImin ≦ ΔI (i) ≦ ΔImax, the load operation status is normal. The timer interrupt process is terminated.

  Further, if the determination result in step S46 shows that a code change has occurred, the process proceeds to step S51, where the current code continuation number t is less than the minimum value tmin corresponding to the current code C (i). If t <tmin, it is determined that the load operating condition is abnormal, and the process proceeds to step S45. If t ≧ tmin, the process proceeds to step S52, and the code continuation number t is set. Set to “1”, and then the process proceeds to step S53 to form a compressed pattern sequence in which the current code is added to the end of the compressed pattern sequence, which is stored in the nonvolatile memory 8, and then proceeds to step S54. To do.

In this step S54, it is determined whether or not a compression pattern string that matches the compression pattern string stored in the RAM 7c exists in the nonvolatile memory 7, and if it exists, it is determined that the operation status of the load is normal. Then, the process proceeds to step S49, and if it does not exist, it is determined that the operation status of the load is abnormal, and the process proceeds to step S45.
2 to 7, the processes in steps S12 and S13 in FIG. 3, the processes in FIG. 4 and the processes in steps S31 to S38 in FIG. 7 correspond to the electricity consumption change measuring unit, and in FIGS. The processing corresponds to the change pattern learning means, among the processing of FIG. 7, the processing in steps S39 to S44 and S46 to S50 corresponds to the abnormality detection means, and the processing in step S45 and the breaker 2 correspond to the power supply stop means. ing.

Next, the operation of the above embodiment will be described.
Now, in order to monitor the operation status of the loads L1 to L5 in a state where, for example, loads L1 to L5 having substantially equal power consumption such as lamps are connected to the existing power supply line 1, a current is supplied to the output side of the breaker 2. A detector 4 is connected, and this current detector 4 is connected to a load operation status monitoring device 5 driven by a power supply of another system.

In this state, when the power of the load operation status monitoring device 5 is turned on, execution of the load monitoring process shown in FIG. 2 is started. In this load monitoring process, the process proceeds to step S2 until a predetermined learning period such as one week or one month set in advance elapses, and the learning process of FIG. 3 is executed.
In this learning process, for example, the operation status of the loads L1 to L5 is learned by detecting a change pattern of current consumption within a predetermined time, for example, in the range of 5 to 20:00.

  That is, for example, the learning process of FIG. 3 is executed every hour from 5:00 am to 20:00, and the detected current value I (i) detected by the current detector 4 at that time is read and stored in the nonvolatile memory 8. Is stored in the detection current storage area at the corresponding time formed in step S21, S22. Then, the previous detected current value I (i−1) stored in the detected current storage area at the previous time is read into the nonvolatile memory 8, and the previous detected current value I (i−1) and the current detected current value are read. Based on the detected current value I (i), the calculation of the equation (1) is performed to calculate the current change amount ΔI (i), and the current change amount ΔI (i) is calculated at the corresponding time in the nonvolatile memory 8. It memorize | stores in an electric current variation storage area (step S24).

Then, it is determined whether the current change amount ΔI (i) has an increasing tendency, a decreasing tendency, or no change, and codes a to c corresponding to each situation are formed in the nonvolatile memory 8 based on the determination result. The state is stored in the state storage area (steps S25 to S29).
By repeating this process every hour, as shown in FIG. 8A, the detected current value I (i) and codes a to c representing the change state are stored in the nonvolatile memory 8 as shown in FIG. . At this time, the loads L1 to L5 are extinguished at 5 o'clock, and the current consumption continues to be zero, so the sign is also set to “b” indicating no state change. Thereafter, when one of the loads L1 to L5 is turned on at 6 o'clock, the detected current value I (6) increases by the consumption current ia of the loads L1 to L5, so that the current change amount ΔI (6) is greater than the threshold value + α. Large + ia, which changes the sign to “a”.

  Next, at 7 o'clock, one more load is turned on, so that the detected current value I (7) becomes 2 × ia, the current change amount ΔI (7) becomes + ia, and the increasing trend continues. Continue a ″. Next, at 8 o'clock, the detected current value I (8) becomes 4 × ia, the current change amount ΔI (8) becomes + 2 × ia, and the increasing trend continues. Therefore, the sign continues to “a”, and 9 When the time comes, all the loads L1 to L5 are turned on, so that the detected current value I (9) becomes 5 × ia, the current change amount ΔI (9) becomes + ia, and the increasing trend continues. Continue a ″.

Thereafter, the lighting states of all the loads L1 to L5 are continued at 10 o'clock and 11 and the detected current values I (10) and I (11) continue to be in the state of 5 × ia. Therefore, the current change amount ΔI (10) And ΔI (11) are “0”, and there is no state change, so the sign is set to “b”.
Thereafter, at the twelve o'clock stage, one load is turned off, so that the detected current value I (12) becomes 4 × ia, and the current change amount ΔI (12) becomes −ia smaller than the threshold −α. Therefore, the sign is “c”, and at the 13:00 stage, only one load is lit, so that the detected current value I (13) becomes ia and the current change amount ΔI (13) becomes −3 × ia. Since the decreasing trend is continued, the sign continues “c”.

  Thereafter, when all the loads L1 to L5 are turned on between 14:00 and 16:00, the detected current values I (14) to I (16) become 5 × ia, and the signs are “a”, “b”, “b” Since the detected current values I (17), I (18), and I (19) are 3 × ia, ia, and 0 from 17:00 to 19:00, the current change amounts ΔI (17), ΔI (18 ) And ΔI (19) become −2 × ia, −2 × ia, and −ia, respectively, and the signs are “c”, “c”, and “c”, and at 20 o'clock, all the loads L1 to L5 Since the extinguishing state continues, the detected current values I (20) to I (24) and the current change amounts ΔI (20) to ΔI (24) become zero, and the sign is “b”.

For this reason, the code pattern of the first day stored in the state storage area of the nonvolatile memory 8 is “baaaaabbccbbcccb”.
Similarly, the code pattern for the second day is “baaaabbbbcabbccccb” as shown in FIG. 8B, and the code pattern for the third day is “bbaaaabcccbcccb” as shown in FIG. 8C. As shown in FIG. 8D, the pattern is “baabaabbbcbabcccb”. For example, the code pattern for the fifth day, which is a holiday, is “bbbaabbbbccbabbcb”.

Thus, for example, when a learning period of 5 days elapses, in the process of FIG. 3, the process proceeds from step S11 to step S14, and the compression process of FIG. 5 is started.
In this compression processing, first, since the variable i is set to “1” in step S111, the first code C (1) = b stored in the state storage area of the nonvolatile memory 8 is read (step S112). Since it is a code C (1), the process proceeds to step S117, and the detected current value I (1) = 0 at 1 o'clock is set as the minimum minimum current value Imin, and the current change amount ΔI (1) = 0 is minimized. In step S118, the detected current value I (1) at 1 o'clock is set as the maximum current value Imax, and the current change amount ΔI (1) = 0 is set as the maximum change amount ΔImax. In step S119, the number of consecutive codes t is set to “1”.

Next, in step S120, the code C (1) = b is stored as the compression code CC (1) at the head of the compression pattern string storage area of the RAM 7c.
Then, the variable i is incremented to “2”, and there is a code C (2) corresponding to the variable i = 2. Therefore, the process returns to step S112, and the code C (2) = a of the variable i = 2 is read. Since this is not the code C (1), the previous code (C1) = b is read (step S114), and the codes C (2) and C (1) do not match, so the process proceeds from step S115 to step S116. The minimum detection current value Imin = 0, the maximum detection current value Imax = 0, the minimum change amount ΔImin, the maximum change amount ΔImax, and the code continuation number t = 1 stored in the range data storage area of the nonvolatile memory 8 are the range data. And stored in the range data storage area of the nonvolatile memory 8 in association with the compression code CC (1) = b.

The current detection value I (2) = ia of the code C (2) is set as the minimum detection current value Imin and the maximum detection current value Imax, and the current change amount ΔI (2) = 1 / T is the minimum change amount. ΔImin and the maximum change amount ΔImax are set, and the code continuation number t is set to “1” (steps S117 to S119).
Next, the code C (2) = a is formed in the nonvolatile memory 8 as the compression code CC (2) = a, and is stored next to the compression code CC (1) in the compression pattern string storage area.

Next, since the variable i is incremented by “1” to become i = 3 (step S121), the code C (3) = a is read out, and the code C (2) = a is read out so that both codes match. Therefore, the process proceeds from step S115 to step S124.
In this step S124, since the detected current value I (3) = 2 · ia, it is larger than the minimum detected current value Imin = ia. Therefore, the process proceeds directly to step S126, and since it is larger than the maximum detected current value Imax = ia, The process proceeds to S127, the detected current value I (3) = 2 · ia is set to the maximum current value Imax, and then the process proceeds to step S128.

  In this step S128, the current change amount ΔI (3) = 1 / T, which is the same as the minimum change amount ΔImin = 1 / T and the maximum change amount ΔImax = 1 / T. In S133, the number of consecutive codes t is incremented and set to "2", the variable i is incremented to "4", and the process returns to Step S112.

  For this reason, the code C (4) = a is read out, and the code C (3) = a is read out, and the two match, so that the process proceeds from step S115 to step S124 as in the previous time. At this time, since the detected current value I (4) = 4 · ia and the current change amount ΔI (4) = 2 / T, the maximum detected current value Imax is updated to 4 · ia and the maximum change amount ΔImax. Is updated to 2 / T, the number of consecutive codes t is set to “3”, the variable i is incremented to “5”, and the process returns to step S112.

  For this reason, the code C (5) = a is read out, and the code C (4) = a is read out, and the two match, so that the process proceeds from step S115 to step S124 as in the previous time. At this time, since the detected current value I (4) = 4 · ia and the current change amount ΔI (4) = 2 / T, the maximum detected current value Imax is updated to 4 · ia and the maximum change amount ΔImax. Is updated to 2 / T, the number of consecutive codes t is set to “4”, the variable i is incremented to “6”, and the process returns to step S112.

  At this time, since the code C (6) = b and the code C (5) = a are inconsistent, the process proceeds from step S115 to step S116. Therefore, as the range data of the compression code CC (2) stored in the nonvolatile memory 8, the minimum detection current value Imin = ia, the maximum detection current value Imax = 5 · ia, the minimum change amount ΔImin = 1 / T, the maximum The change amount ΔImax = 2 / T and the number of code continuations t = 4 are stored in association with the compression code CC (2) = a as shown in FIG. Next, the detected current value I (6) = 5 · ia is set as the minimum detected current value Imin and the maximum detected current value Imax, and the current change amount ΔI (6) = 0 is the minimum change amount ΔImin and the maximum change amount ΔImax. The number of consecutive codes t is set to “1”. Further, the code C (6) = b is additionally stored at the end of the compression pattern string storage area of the nonvolatile memory 8 as the compression code CC (3).

In this manner, the code string is sequentially compression-encoded and sequentially stored in the non-volatile memory 8 together with the range data as shown in FIG. 9, so that a compression pattern string of “babacabcb” is finally formed. .
Also, as shown in FIG. 8B, the pattern pattern on the second day becomes the code pattern “bbbbbabaabbbbbcabbccccbbbb”, and when this is compressed to form a compressed pattern string, it becomes “babacabcb”, and the compressed pattern string on the third day As shown in FIG. 8 (c), “bacabcabcb” is obtained, and the compression pattern string on the fourth day is “babcbabcb” as shown in FIG. 8 (d), and the compression pattern string on the fifth day is FIG. 8 (e). As shown in the figure, it is “babcbabcb”.

When the compressed pattern sequence from the first day to the fifth day is calculated in this way, a continuous compressed pattern sequence is formed in the compressed pattern sequence storage area of the nonvolatile memory 8 as shown in FIG.
By executing the compression pattern sequence determination process of FIG. 6 for the continuous compression pattern sequence, the compression code “b” starts with the continuous compression code “ba”, ends with the continuous compression code “cb”, and is included in these compression codes “b”. ”Is determined as a learning pattern sequence, in which the maximum detected current value Imax is less than the minimum current consumption ia of the loads L1 to L5.

  That is, in the continuous compression pattern sequence of FIG. 10, the compression pattern sequences CC (1) and CC (2) corresponding to the first day are “ba” and match the start condition, and then CC (7) and CC (8) becomes “cb”, which matches the end condition, and from FIG. 8A, the maximum detected current value Imax of the compression code “b” included therein is “0”. (1) to CC (8) are selected as learning pattern sequence candidates, and since the compression pattern sequence is not yet stored in the existing compression pattern sequence in the nonvolatile memory 8, the compression pattern sequence “babcabcb” is used as the learning pattern sequence. This is confirmed and stored in the existing compressed pattern sequence.

  In addition, the compression pattern string CC (8) to CC (15) corresponding to the second day also becomes the compression pattern string “bacabcabb”, which is equal to the compression pattern string of the first day. For this reason, in the compression pattern determination process in FIG. 6, the process proceeds from step S213 to step S215, and the range data of the compression pattern sequence on the first day is compared with the range data of the compression pattern sequence on the second day. Calculates the minimum continuous number tmin and the maximum continuous number tmax of the continuous number t, and corrects the minimum detected current value Imin, the maximum detected current value Imax, the minimum change amount ΔImin, and the maximum change amount ΔImax based on the range data on the second day. Then, as shown in FIG. 9C, the range data is updated.

Further, the compression pattern strings CC (15) to CC (22) corresponding to the third day are also “babcabcabb”. In this case as well, the minimum number of consecutive times tmin and the maximum number of consecutive times tmax are calculated. At the same time, the minimum detected current value Imin, the maximum detected current value Imax, the minimum change amount ΔImin, and the maximum change amount ΔImax are corrected based on the range data on the third day.
However, in the compression pattern sequence CC (22) to CC (30) corresponding to the fourth day, “babcbabcb” is obtained, the fifth compression code CC (26) is “b”, and the compression patterns of the first to third days are obtained. Since it is different from the fifth compression code CC (5) = “a” in the sequence, it is determined that the existing compression pattern sequence does not exist, and it is non-volatile together with its range data as a new second compression pattern sequence It is stored in the existing compressed pattern string storage area of the memory 8.

  Furthermore, in the compression pattern sequence CC (30) to CC (38) corresponding to the fifth day, “babcbabcb” is obtained, which is the same compression pattern sequence as that on the fourth day, but intermediate compression codes CC (33) and CC (34) becomes “cb”. Since the maximum detected current value Imax of the compression code b at this time is “0” as shown in FIG. 8E, it is determined as the final code, so that “babcb” Is stored in the existing compressed pattern sequence storage area of the nonvolatile memory 8 together with the range data as a third compressed pattern sequence.

Further, the remaining compression pattern strings CC (34) to CC (38) are also “babcb”, but are equal to the third compression pattern string, and thus the range data is updated.
Therefore, as the learning result, “bacabcabcb”, “babcbabcb”, and “babcb” are set as the learning compression pattern string.
In this way, when the learned compressed pattern sequence is stored in the existing compressed pattern sequence storage area of the nonvolatile memory 8, the abnormality determination process of FIG. 7 is executed from the sixth day by the end of the learning period.

  In this abnormality determination process, the compression pattern string and its range data are stored in the existing compressed pattern string storage area of the nonvolatile memory 8 by learning a plurality of types of compression pattern strings by the learning process described above. For example, if the range data of the compression pattern string “babcabcabb” is as shown in FIG. 9C except for the third day, the range data of the first compression code CC (1) = b is tmin = 1, tmax = 2, Imin = 0, Imax = 0, ΔImin = 0, ΔImax = 0, and the range data of the next compression code CC (2) = a is tmin = 3, tmax = 4, Imin = 1 · ia, Imax = 5 · ia, ΔImin = 1 / T, ΔImax = 3 / T, and in the next compression code CC (3) = b, tmin = 2, tmax = 4, Im n = 5, Imax = 5, ΔImin = 0, and has a ΔImax = 0.

  For this reason, if it is a normal load operation situation, it will be within the range data of the learned compression pattern and will not be determined to be abnormal, but if overload etc. occur and abnormal current flows The current exceeding the range data is detected by the compression current CC (1) or CC (8) where the detection current value I (i) is zero by the detection current value I (1), I (8) or the like. Therefore, in step S42 to S43 in FIG. 7 or in S49 to S50, it is determined that there is an abnormality, and a shut-off command for shutting off the breaker 2 is output, and the breaker 2 is automatically shut off.

Therefore, when an operation situation of a load different from the learned compression pattern sequence occurs, the breaker 2 is automatically shut off, and safety when an overload occurs can be ensured.
Thus, according to the above embodiment, the load operation status monitoring device 5 is installed in the power supply system, and the operation status of the loads L1 to L5 is learned by patterning by performing learning for a predetermined period in which the current consumption is detected. Since the normal load operation status is monitored based on the learning result, the load operation status can be accurately monitored.

  In addition, the learning is encoded by storing the increasing trend, decreasing trend, and no change state of the detected current to form a compressed pattern string and storing it, and for each compressed code of the compressed pattern string, the minimum / maximum current and the minimum / maximum change amount are stored. Since range data such as these are stored in association with each other, it is possible to accurately learn a change in the operating status of the loads L1 to L5 without making a false determination even when there is a temporal variation in the operating status of the load. In addition, even if a missing detection current occurs, it can be determined to be normal if it falls within the range data, and the overall load operation status including standby can be determined without specifying the number of samples. Can be monitored accurately.

  Further, in the abnormality determination process, the compression pattern at the time of daily load operation is stored, the range data thereof is stored, and this is displayed on the display device 10 together with the corresponding existing compression pattern sequence and the range data. The operation status of L1 to L5 can be confirmed, and further, the compressed pattern sequence and its range data can be output to a printing apparatus such as a printer, and can be used as a study material for the load operation status. In this case, for example, it is possible to easily grasp whether or not the lights are turned off to save power during the lunch break and the usage pattern of the load.

In the above-described embodiment, the case where the breaker 2 is shut off when the load operation status monitoring device 5 detects a load operation abnormality is described. However, the present invention is not limited to this. A display or sound warning may be displayed on the display device 10.
In the above embodiment, a case has been described in which the compression pattern string is automatically formed and stored in the learning process, and abnormality detection is performed by comparing the storage compression pattern string with the actual load operation status. However, the present invention is not limited to this, and at the end of the learning period, the stored compressed pattern sequence and range data may be temporarily displayed on the display device 10, and the operator may confirm and move to the abnormality determination process.

Furthermore, although the case where the learning process period is automatically set has been described in the above embodiment, the present invention is not limited to this, and the operator manually sets the start and end of the learning period, and the abnormality determination process The transition time to may be set manually.
Furthermore, in the said embodiment, although the case where a lighting fixture was applied as load L1-L5 was demonstrated, it is not limited to this, such as an air conditioner connected to a power supply system, a personal computer, a machine tool It is possible to detect the operating status of various loads.
Furthermore, in the above-described embodiment, the case where the current consumption is detected as the amount of electricity consumed has been described. However, the present invention is not limited to this, and the power consumption or the amount of power consumption may be detected.

It is a block diagram which shows the system configuration | structure of this invention. It is a flowchart which shows an example of the load monitoring process procedure performed with CPU of a load operation condition monitoring apparatus. It is a flowchart which shows the learning process procedure of FIG. It is a flowchart which shows an example of an electric current measurement process sequence. It is a flowchart which shows an example of the code compression process sequence of FIG. It is a flowchart which shows an example of the compression pattern sequence decision processing procedure of FIG. It is a flowchart which shows an example of the abnormality determination processing procedure of FIG. It is a time chart which shows the change of detection current. It is explanatory drawing which shows a compression pattern row | line and its range data. It is explanatory drawing of the process which determines a learning compression pattern sequence from the continuous compression pattern sequence.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Electric power supply line, 2 ... Breaker, 3 ... Distribution board, 4 ... Current detector, 5 ... Load operation condition monitoring apparatus, 6 ... A / D converter, 7 ... Arithmetic processing apparatus, 7a ... CPU, 7b ... ROM, 7c ... RAM, 8 ... nonvolatile memory, 10 ... display device

Claims (5)

A load operation status monitoring device for monitoring the load operation status of a facility in which a plurality of loads are connected to a power system,
The electricity consumption change measuring means for measuring the electricity consumption of the power supply system and measuring the amount of change of the electricity consumption per predetermined time, and the amount of change per predetermined time measured by the electricity consumption change measuring means. Load pattern operation by comparing a change pattern learning means for learning a change pattern, a change pattern learned by the change pattern learning means, and a change amount of the actual electricity consumption measured by the electricity consumption measuring means for each predetermined time A load operation status monitoring device comprising: an abnormality detection means for detecting an abnormality of a situation. A load operation status monitoring device for monitoring the load operation status of a facility in which a plurality of loads are connected to a power system,
The electricity consumption change measuring means for measuring the electricity consumption of the power supply system and measuring the amount of change of the electricity consumption per predetermined time, and the amount of change per predetermined time measured by the electricity consumption change measuring means. Load pattern operation by comparing a change pattern learning means for learning a change pattern, a change pattern learned by the change pattern learning means, and a change amount of the actual electricity consumption measured by the electricity consumption measuring means for each predetermined time An abnormality detecting means for detecting an abnormality of the situation, and a power supply stopping means for stopping the power supply to each load of the power supply system when the abnormality detecting means detects an abnormality of the load operation situation. Load operation status monitoring device.   The electricity consumption change measuring means is configured to measure any one of current, electric power, and electric energy as the electricity consumption, and measure the amount of change per predetermined time. The load operation status monitoring apparatus according to claim 1 or 2.   The change pattern learning unit is configured to form a pattern sequence by dividing the change amount of the consumed electricity amount measured by the consumed electricity amount change measuring unit into three states of increasing tendency, decreasing tendency, and no change. The load operation status monitoring apparatus according to any one of claims 1 to 3, wherein the load operation status monitoring apparatus is provided.   The change pattern learning means divides the change amount of the electricity consumption measured by the electricity consumption change measurement means into three states of increasing tendency, decreasing tendency, and no change to form a pattern sequence, and the same pattern The compressed pattern sequence is formed by compressing when continuing, and the formed compressed pattern sequence is registered as a new compressed pattern sequence when the compressed pattern sequence is different from the existing compressed pattern sequence. 4. The load operation status monitoring device according to any one of 1 to 3.

Images