JP2011122853A - Apparatus failure evaluation system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To analyze a frequency of an audio signal generated from a consumer apparatus, and estimate a failure and an abnormality as a precursor of the consumer apparatus. <P>SOLUTION: An apparatus failure evaluation system is equipped with: an audio measurement apparatus 8 for measuring the audio signal generated from a load apparatus (including a facility, a device and a component) 7 supplied with power; a frequency component calculation means 11 for analyzing a waveform of the measured audio signal, and calculating an actually measured frequency component; a data storage means 12 for storing a frequency-abnormal pattern having a frequency-abnormal frequency component estimated as a frequency abnormality for inducing the failure and the precursor in each load apparatus; a frequency component coincidence calculation means 13 for calculating a frequency component coincidence between an actually-measured frequency component calculation result and a plurality of frequency-abnormal frequency components; and a frequency abnormality output means 14 for estimating the frequency-abnormal pattern as the failure and its precursor in the load apparatus from each frequency component coincidence, and outputting or displaying it. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention, for example, calculates a frequency component of an acoustic signal generated from a consumer device such as an uninterruptible power supply system or a power supply and demand system (for example, a microgrid supply and demand management system), and evaluates the state of the consumer device. About the system.

  Conventionally, several techniques have been proposed as a method of analyzing the frequency of a physical quantity measured from a power consumer device and evaluating the possibility of deterioration or failure of the consumer device from the frequency analysis result.

  As a conventional technique, a waveform for each failure factor is sampled in advance using a simulated high-voltage distribution line, and harmonic components up to a predetermined order of each sample waveform are analyzed. And after calculating | requiring the sum total of the average value for every order of a harmonic content rate according to a failure factor, the maximum value and minimum value of the sum total data according to a failure factor are calculated and memorize | stored. Estimate the cause of a ground fault in the high-voltage distribution line based on whether the current waveform flowing in the zero phase of the actual high-voltage distribution line is in the range between the maximum and minimum values for each failure cause (Patent Document 1).

  Another prior art detects a fluctuation active part and a fluctuation reactive part before and after the transient state of the fundamental wave active power and the fundamental wave reactive power in the consumer, and loads the consumer based on the magnitude of the fluctuation part. , And specify the operational status such as turning on / off power factor correction capacitors and transformers.

  Also, with this technology, the current at the time of customer equipment input is frequency-analyzed, and the change pattern in which the harmonic component ratio varies with the equipment, which is the waveform pattern of the harmonic component, and the temporal harmonic component ratio changes, is extracted. The operation state of the facility and the facility where the failure occurs are specified (Patent Document 2).

Japanese Patent Laid-Open No. 06-217451 Japanese Patent No. 3952355

  However, the former technique only identifies the cause of the ground fault in the high-voltage distribution line, and the latter technique exclusively identifies the operational state of the customer equipment on / off, It is not intended to estimate a failure of the customer device or an abnormality that is a precursor thereof from the frequency component of the sound generated from the customer device.

  The present invention has been made in view of the above circumstances, and measures an acoustic signal generated from a consumer device, analyzes the frequency of the measured acoustic signal, and estimates a failure or a precursor abnormality of the consumer device. An object is to provide an apparatus failure evaluation system.

  In order to solve the above-described problems, the present invention provides an acoustic measurement unit that measures an acoustic signal generated from a consumer device (including equipment, equipment, and components) that is supplied with power from an electric power system, and the acoustic measurement unit. A frequency component calculating means for analyzing the waveform of the acoustic signal measured in step (b) and calculating an actual frequency component; and for each consumer device, a frequency abnormality frequency component estimated to be a malfunction or a precursor to the abnormal frequency for each consumer device. A data recording means for storing a frequency abnormality pattern having, and a frequency component coincidence between a measured frequency component calculation result obtained by the frequency component calculation means and a plurality of frequency abnormal frequency components stored in the data recording means; The frequency component coincidence calculating means, and the frequency components coincidence calculated by the frequency component coincidence calculating means from the frequency that coincides with the failure of the consumer device or a precursor thereof. It estimated that several abnormal pattern, a power quality evaluation system and a frequency error output means for outputting or displaying.

  In addition, the present invention provides a frequency normal pattern having a frequency normal frequency component estimated in advance as a normal frequency of each consumer device, or a frequency abnormal frequency estimated as a frequency abnormality that is a failure or a precursor of each consumer device. Frequency abnormal pattern having a component and a frequency normal pattern having a normal frequency frequency component estimated to be a normal frequency of each home device, or a frequency abnormality presuming a frequency abnormality to be a failure or a precursor of each consumer device A frequency abnormal pattern having a frequency component, a frequency normal pattern having a frequency normal frequency component estimated to be a normal frequency of each home device, and a frequency unique to another customer device to which each customer device is connected in parallel The frequency load pattern of the estimated load frequency component is stored in the data recording means, and the actual measurement obtained by the frequency component calculating means Calculate the frequency component coincidence with the wave number component calculation result, and estimate or output or display the frequency normal pattern, the frequency abnormal pattern, or the frequency load pattern that is a precursor or failure of the consumer device from the frequency component coincidence. It may be a configuration.

  ADVANTAGE OF THE INVENTION According to this invention, the apparatus failure evaluation system which can measure the acoustic signal which generate | occur | produces from a consumer apparatus, carries out frequency analysis of the measured acoustic signal, and can estimate the malfunction and precursor abnormality of a consumer apparatus can be provided.

The system diagram which shows the example of application to the electric power system of the apparatus failure evaluation system which concerns on this invention. BRIEF DESCRIPTION OF THE DRAWINGS The schematic block diagram which shows Embodiment 1 of the apparatus failure evaluation system which concerns on this invention. The block diagram using the FFT process part which is a specific example of a frequency component calculation means. The figure explaining an example of a frequency normal pattern and a frequency abnormal pattern. The figure explaining the example of several frequency abnormality patterns previously stored in a database. The figure explaining the functional block and processing content of a frequency component coincidence calculation means. The figure explaining normalization of a frequency component. The figure explaining the example with which the normalized actual frequency component result and the normalized frequency abnormal frequency component correspond completely. The figure explaining the example in which the normalized actual frequency component result and the normalized frequency abnormal frequency component do not correspond completely. The figure explaining the example in which the standardized measurement frequency component result and the standardized frequency abnormal frequency component partially correspond. The figure showing the relationship between the degree of coincidence of a standardization measurement frequency component calculation result and a standardization frequency abnormal frequency component, and a standardization difference total value. The figure showing the relationship between a frequency component coincidence degree and a normalization difference total value from another viewpoint. The block diagram which shows one specific example of the frequency abnormality estimation means shown in FIG. The figure explaining the relationship between a frequency abnormal frequency component and a frequency component coincidence degree. The figure which shows an example which displayed the frequency abnormality estimation result on the display part, without providing a frequency abnormality extraction threshold value. The block diagram which shows the modification 1 of Embodiment 1 of the apparatus failure evaluation system shown in FIG. The figure explaining the relationship between a frequency normal frequency component and a frequency component coincidence degree. The block diagram which shows the modification 2 of Embodiment 1 of the apparatus failure evaluation system shown in FIG. The block diagram which shows the modification 3 of Embodiment 1 of the apparatus failure evaluation system shown in FIG. The figure explaining the example of a display of a plurality of different kinds of frequency patterns. The block diagram which shows Embodiment 2 of the apparatus failure evaluation system which concerns on this invention. The figure which shows the example of a display of the time-sequential change of the frequency component coincidence for every some frequency abnormality pattern acquired for every fixed time. The block diagram which shows Embodiment 3 of the apparatus failure evaluation system which concerns on this invention. The figure which shows the alarm transmission system of the alarm which generate | occur | produces from the alarm generation means of FIG. The figure explaining the generating condition of an alarm. The block diagram which shows Embodiment 4 of the apparatus failure evaluation system which newly provided the control command generation | occurrence | production means in the structure shown in FIG. The figure which shows the control command transmission system of the control command which is generated from the control command generation means of FIG. The figure explaining the generation conditions of a control command.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 is a diagram showing an example in which an apparatus failure evaluation system 1 according to the present invention is applied to a power system.
The apparatus failure evaluation system 1 is a customer apparatus via a distribution line (which can also be a transmission line) 6 for a bus 5 that receives predetermined power from a transmission line 3 of a higher power system 2 via a transformer 4, for example. A load device 7 is connected. Here, the load device 7 is a device installed for a power consumer such as an uninterruptible power supply system or a power supply and demand system (for example, a microgrid demand and supply management system), and is not only a simple device but also a host device. This includes equipment, devices, parts, and the like that receive power from the power system 2. In FIG. 1, one load device 7 is illustrated, but all load devices 7 that emit some sound in the event of a failure or an abnormality that is a precursor thereof are targeted.

  An acoustic measurement unit 8a composed of a sound collecting microphone, an amplifier, and the like is provided inside the casing of the load device 7 that can be a sound generation source in the event of a failure or an anomaly that is a sign of the failure, or in the vicinity of the casing of the casing. An acoustic measurement device 8 including an AD conversion unit 8b for converting into a digital signal is disposed. If the acoustic measurement unit 8a directly outputs an acoustic signal as a digital signal, the AD conversion unit 8b is not necessary. The acoustic measurement device 8 measures the acoustic signal 10 from the load device 7 and transmits it to the device failure evaluation system 1 through the transmission system 9.

  The device failure evaluation system 1 collects the acoustic signal 10 generated from the load device 7 (sound generation source) measured by the acoustic measurement device 8, analyzes the frequency component of the acoustic signal 10, and detects the failure of the load device 7 or its Estimate an anomaly that is a precursor.

  Therefore, in FIG. 1, the device failure evaluation system 1 and the acoustic measurement device 8 are described as separate components, but in the device failure evaluation system according to the present invention, an acoustic measurement device that measures the acoustic signal 10. 8 is also included as an essential component.

FIG. 2 is a block diagram showing Embodiment 1 of the apparatus failure evaluation system according to the present invention.
The apparatus failure evaluation system 1 uses a computer to perform software processing according to a predetermined processing procedure. Functionally, the apparatus failure evaluation system 1 has a frequency component calculation means 11, a data recording means 12, and a frequency component coincidence degree. The calculation means 13 and the frequency abnormality estimation means (corresponding to the frequency abnormality output means in a broad sense) 14 are configured.

  The frequency component calculation means 11 analyzes the waveform of the acoustic signal 10 generated from the load device 7 measured by the acoustic measurement device 8, acquires the actual frequency component calculation result 15 composed of the frequency components constituting the acoustic signal 10, The data is stored in the data recording means 12 and sent to the frequency component coincidence calculation means 13.

The data recording unit 12 records various data necessary for estimating a failure of a device (including equipment, equipment, and parts) and an abnormality that is a precursor thereof, and includes a database 16. In the database 16, the actual frequency component calculation result 15 calculated by the frequency component calculation means 11 is stored, and for example, frequency components (hereinafter referred to as frequency abnormal frequencies) corresponding to the load devices 7. The frequency component coincidence calculation means 13 is stored in the measured frequency component calculation result 15 and the frequency abnormal frequency component 17 stored in the database 16. It has a function of calculating the degree of coincidence with frequency components (hereinafter referred to as frequency component coincidence) 19.

  When the frequency component coincidence degree 19 calculated by the frequency component coincidence degree calculating unit 13 is high, the frequency abnormality estimating unit 14 determines that the frequency abnormality pattern 18 corresponding to the corresponding frequency abnormality frequency component 17 has a high possibility of frequency abnormality. The frequency abnormality pattern 18 is estimated, and the estimated frequency abnormality pattern 18 is output as the frequency abnormality estimation result 20.

  The output format of the frequency abnormality estimation result 20 is, for example, displayed on a display device, printed out from a printer, stored in the data recording means 12 or a separate storage means, or externally via a dedicated transmission line or the like. There are formats to output to the output device.

Next, the operation of the apparatus failure evaluation system 1 as described above will be described.
First, the frequency component calculation means 11 calculates a component for each order of the frequency included in the acoustic signal 10 sent from the acoustic measurement device 8.

  As a method for calculating a component for each order of frequency, fast Fourier transform (FFT) is most commonly used.

  The frequency is expressed as a collection of sine waves having a constant period and different periods. Now, assuming that the fundamental wave of frequency is primary, a sine wave whose period is 1/2 of the fundamental wave (frequency is twice), and a sine whose period is 1/3 of the fundamental wave (frequency is 3 times) A sine wave whose period is 1 / n (frequency is n times) of the fundamental wave is called the nth order.

  For example, if the fundamental wave is 1 Hz, the second order is 2 Hz, the third order is 3 Hz, and the order and Hz match. Hz is a unit of frequency. The frequency of one cycle per second is 1 Hz. Similarly, if the fundamental wave is 10 Hz, the second order is 20 Hz, the third order is 30 Hz, and the order is 10 times the Hz.

  Generally, the frequency range of the acoustic signal 10 that can be perceived by humans is 20 Hz to 20 kHz, but the frequency range of conversation between humans is 200 Hz to 8 kHz.

  By the way, the observation data (measurement data) may be discrete sampling data due to the operating characteristics of the apparatus. When obtaining frequency components from such sampling data, it is necessary to obtain them using a technique called discrete Fourier transform.

  On the other hand, the fast Fourier transform (FFT) is an improved method so that the discrete Fourier transform can be solved at high speed, and can cope with either continuous or discrete sampling data, and is the most versatile in frequency analysis. This is a technique.

  Therefore, the frequency component calculation means 11 in the present embodiment performs a frequency component calculation process using an FFT processing unit 11a having a high-speed Fourier transform function shown in FIG.

  That is, the FFT processing unit 11 a performs waveform analysis on the input acoustic signal 10 and calculates a component for each order of the frequency included in the acoustic signal 10.

  Here, if the magnitude of the frequency of each order is the magnitude of the frequency spectrum, the frequency content is calculated for each order of the frequency, and is expressed by Expression (1).

Frequency content (order) = frequency spectrum size (per order) ÷ total sum of frequency spectrum size at sample frequency (1)
For example, if the sample frequency is 1 Hz to 1000 Hz, the sum of the sample frequencies is the sum of frequency vectors (for each order) from 1 Hz to 1000 Hz.

  On the other hand, the database 16 stores the frequency abnormal frequency component 17 of the frequency abnormal pattern 18 as described above. The frequency abnormality pattern 18 is an abnormality pattern of a frequency that becomes a failure or a sign of failure of equipment, devices, equipment, or parts.

The normal / abnormal frequency pattern will be described with reference to FIG.
FIG. 4A is a diagram showing an example of the normal frequency pattern 21 and shows a frequency spectrum distribution having a peak from 400 Hz to 500 Hz. Here, the normal frequency pattern 21 is a frequency pattern that does not become a sign of device failure or a sign thereof, and is a frequency pattern during normal operation. This is different for each facility, each device, each device, and each part.

  FIG. 4B is a diagram showing an example of the frequency abnormality pattern 18, and there is a frequency spectrum distribution showing an abnormality in the vicinity of 1500 Hz, apart from a frequency spectrum distribution having a peak from 400 Hz to 500 Hz.

  5A to 5C are examples of frequency abnormality patterns 18-A, 18-B, and 18-C stored in the database 16, and the vertical axis indicates the magnitude of the frequency (frequency content). The horizontal axis represents the frequency (Hz). As described above, the frequency content is calculated by the equation (1) for each order of frequency.

  The frequency abnormality patterns 18-A, 18-B, and 18-C have an abnormal frequency distribution (frequency abnormal frequency component) separately from the normal frequency distribution. Incidentally, the frequency abnormality pattern 18-A has an abnormal frequency distribution in the vicinity of 1500 Hz, the frequency abnormality pattern 18-B has an abnormal frequency distribution in the vicinity of 1200 to 1300 Hz, and the frequency abnormality pattern 18-C has a frequency near 1300 Hz. , 1700 Hz to 1800 Hz is an example having an abnormal frequency distribution.

  The frequency abnormality pattern 18 is not a frequency distribution in which the normal frequency distribution and the abnormal frequency distribution are completely separated. For example, the frequency abnormality pattern 18 has an abnormal frequency that shows a distribution that partially protrudes from the normal frequency distribution. It may be a distribution.

  By the way, the acoustic signal 10 (abnormal sound) including an abnormal frequency is not only a failure of the load device (equipment, device, equipment, component) 7 constituting the power supply and demand system, but often serves as a precursor of the failure. . Therefore, in the database 16, in addition to the failure of each load device 7 constituting the power supply and demand system, etc., there are a number of combinations of the components for each order of frequency and the content ratios of components for each order of frequency. The frequency abnormality pattern 18 is stored.

  FIG. 6 is a diagram for explaining the functional blocks and processing contents of the frequency component coincidence calculation means 13.

  The frequency component coincidence calculation means 13 includes a normalization calculation processing unit 13a that executes a normalization process so that the total of the actual measurement frequency component calculation results 15 is 1, and an actual measurement calculated by the normalization calculation processing unit 13a. The degree of coincidence between the standardized result (standardized value) of the frequency component calculation result 15 and the standardized value so that the sum of the frequency components becomes 1 for each frequency abnormal frequency component 17 stored in the database 16 is calculated. A coincidence calculation processing unit 13b is provided.

  When receiving the measured frequency component calculation result 15 from the frequency component calculation means 11 in the sample frequency region, the normalization calculation processing unit 13 a executes normalization processing and includes frequency components of respective orders included in the measured frequency component calculation result 15. Is normalized so that the sum of 1 becomes 1.

  In the normalization process, a j-th order component (j is a natural number of 2 or more) after normalization is calculated using the following equation (2).

J-order component after normalization = j-th frequency component before normalization / total value of frequency components before normalization
(2)
Assuming that the sample frequencies of Expression (1) and Expression (2) are the same, the j-th order component after normalization of Expression (2) is the frequency content rate (j-th order) in the measured frequency component calculation result 15. Become.

  However, as shown in FIG. 5, when the normal frequency distribution is larger than the abnormal frequency distribution, the difference between the frequency normal pattern 21 and the frequency abnormal pattern 18 is relatively small. It is necessary to sample from a region including a relatively large frequency region.

  For example, the sample frequency in the frequency abnormality pattern 18-A is in the vicinity of 1500 Hz, the sample frequency in the frequency abnormality pattern 18-B is in the vicinity of 1200 Hz to 1300 Hz, and the sample frequency in the frequency abnormality pattern 18-C includes many regions in the vicinity of 1300 Hz to 1800 Hz. Sampling.

FIG. 7 is a diagram for explaining sample frequencies in normalization of frequency components.
FIG. 7A shows a case where the sample frequency region of the frequency abnormality pattern 18-A is 1000 Hz to 2000 Hz, and the sum of the integrated values of the abnormal frequency distribution is 1. Similarly, in FIG. 7B, when the region of the sample frequency of the frequency abnormality pattern 18-B is 1000 Hz to 1500 Hz, FIG. 7C shows the region of the sample frequency of the frequency abnormality pattern 18-C is 1000 Hz to 2000 Hz. This is the case.

  The coincidence calculation processing unit 13b stores a value after the standardization of the actually measured frequency component calculation result 15 (hereinafter referred to as a standardized actually measured frequency component calculation result 22; see FIG. 8A) in the sample frequency domain, and the database 16. An absolute value of a difference from a value after normalization of a certain frequency abnormal frequency component 17 (hereinafter, referred to as a normalized frequency abnormal frequency component 23; see FIG. 8B) is taken for each order, as shown in Expression (3) To the total value (hereinafter referred to as the normalized difference total value 24).

Total normalized difference value 24 = | component of frequency iHz after normalization (measured frequency component calculation result 15) −component of frequency iHz after normalization (frequency abnormal frequency component 17) | + | frequency of frequency jHz after normalization Component (measurement frequency component calculation result 15)-component of frequency jHz after normalization (frequency abnormal frequency component 17) | + ... + | component of frequency nHz after normalization (measurement frequency component calculation result 15)-normalization Later component of frequency nHz (frequency abnormal frequency component 17) |
= | Frequency iHz component (standardized actual frequency component calculation result 22) −frequency iHz component (normalized frequency abnormal frequency component 23) | + | frequency jHz component (standardized actual frequency component calculation result 22) −frequency jHz (Normalized frequency abnormal frequency component 23) | +... ++ frequency nHz component (standardized actual frequency component calculation result 22) −frequency nHz component (normalized frequency abnormal frequency component 23) | The frequency range is iHz to nHz). ...... (3)
That is, when the coincidence calculation processing unit 13b completely matches the frequency components and magnitudes of the normalized actual frequency component calculation result 22 and the normalized frequency abnormal frequency component 23 as shown in FIG. From (3), the normalized difference total value 24 = 0 (however, the sample frequency region is in the range of 1500 Hz to 1510 Hz).

On the other hand, FIG. 9 is a diagram showing an example in which the normalized measurement frequency component calculation result 22 and the normalized frequency abnormal frequency component 23 do not match at all.
Based on the equation (3), the coincidence calculation processing unit 13b calculates the sum of the normalized measured frequency component calculation results 22 when the normalized measured frequency component calculation result 22 and the normalized frequency abnormal frequency component 23 do not completely match. Since the total value of the value and the normalized frequency abnormal frequency component 23 is “1”, the normalized difference total value 24 is “2”. However, the sample frequency region is in the range of 1500 Hz to 1510 Hz.

  FIG. 10 is a diagram illustrating an example in which the normalized actual measurement frequency component calculation result 22 and the normalized frequency abnormal frequency component 23 partially match. The standardized difference total value 24 in the case of partial match is the total value of the non-matched part, from the value “0” in the case of complete match to the value “2” in the case of no match at all. The intermediate value of.

  Incidentally, in the example of FIG. 10, the sum of the inconsistent portions is the value 0.6 of the frequency region 1503 Hz of the normalized actual frequency component calculation result 22 and the value of the frequency region 1507 Hz of the normalized frequency abnormal frequency component 23. It is “1.2” which is the sum of 0.6. However, the sample frequency region is in the range of 1500 Hz to 1510 Hz.

  FIG. 11 is a diagram showing the relationship between the degree of coincidence between the normalized actual measurement frequency component calculation result 22 and the normalized frequency abnormal frequency component 23 and the normalized difference total value 24. As is apparent from this figure, the standardized difference total value 24 for perfect match is “0”, and the standardized difference total value 24 for perfect mismatch is “2”. It represents that becomes.

Here, the definition of the frequency component coincidence 19 is considered.
Now, the standardized actual frequency component calculation result 22 and the standardized frequency abnormal frequency component 23 are 100% when they are completely coincident, −100% when they are not completely coincident, and 0% between them. In this way, it can be expressed easily as shown in FIG.

Therefore, the conversion equation between the frequency component coincidence 19 and the normalized difference total value 24 shown in FIG. 12 can be expressed by equation (4). As a result, the result calculated by Expression (4) can be defined as the frequency component coincidence degree 19.
Frequency component coincidence 19 = (1−normalized difference total value 24) × 100 (%) (4)
That is, the coincidence calculation processing unit 13 b calculates the frequency component coincidence 19 with the actually measured frequency component calculation result 15 for each frequency abnormal frequency component 17 by the equation (4), and passes it to the frequency abnormality estimating means 14.

  The frequency abnormality estimator 14 estimates the frequency abnormality pattern 18 that causes a frequency abnormality based on the frequency component coincidence 19 for each frequency abnormal frequency component 17 received from the frequency component coincidence calculator 13.

  FIG. 13 is a diagram for explaining the functional blocks and processing contents of the frequency abnormality estimation means 14.

  The frequency abnormality estimation means 14 includes a matching degree rearrangement processing unit 14a and a frequency abnormality extraction unit 14b.

  The matching degree rearrangement processing unit 14a performs processing for rearranging the frequency abnormal frequency component 17 based on the frequency component matching degree 19 of the frequency abnormal frequency component 17, and has a high frequency component matching degree 19, for example. The process of rearranging in order is executed.

  The frequency abnormality extraction unit 14b extracts a pattern 18 that is presumed to be frequency abnormality from the frequency abnormality patterns 18 having the frequency abnormality frequency components 17 that are rearranged in order from the one having the highest frequency component coincidence 19, and the frequency abnormality estimation The result 20 is output.

  FIG. 14 is a diagram for explaining the relationship between the frequency abnormal frequency component 17 and the frequency component coincidence degree 19.

  As shown in FIG. 14, when the frequency abnormality estimation means 14 arranges the frequency abnormality patterns 18 corresponding to the frequency abnormal frequency component 17 in order from the highest frequency component matching degree 19 by the matching degree rearrangement processing unit 14 a from the left side. It can be considered that the frequency component coincidence 19 falls between 100% and −100%, the frequency abnormality is higher as it is on the left side, and the frequency abnormality is lower as it is on the right side.

  Therefore, the frequency abnormality extraction unit 14b of the frequency abnormality estimation means 14 sets the frequency abnormality extraction threshold 25 in consideration of the possibility of affecting the failure of the apparatus and the abnormality that is a precursor thereof through the accumulation and experience of experiments, It is possible to extract a frequency abnormality pattern 18 having a frequency abnormality frequency component 17 having a frequency component matching degree 19 higher than the frequency abnormality extraction threshold 25 as a frequency abnormality. That is, the frequency abnormality extraction unit 14b outputs a frequency abnormality pattern 18 having a frequency abnormality frequency component 17 existing above the frequency abnormality extraction threshold 25 as a frequency abnormality estimation result 20 as shown in FIG.

  The frequency abnormality estimation means 14 may output the frequency abnormality estimation result 20 by rearranging the frequency component matching degree 19 in descending order without providing the frequency factor extraction threshold 25.

  FIG. 15 is a diagram showing an example in which the frequency abnormality estimation result 20 is displayed on the display device without providing the frequency abnormality extraction threshold 25.

  In the figure, the vertical axis represents the frequency component coincidence 19 and the horizontal axis represents the frequency abnormality pattern 18, which are displayed from the left to the right in order of the frequency component coincidence 19.

  Therefore, according to the embodiment as described above, the acoustic measuring device 8 is arranged in the vicinity of each of the load devices 7 that are customer devices or in the vicinity of the load device 7, and the acoustic measuring device 8 uses the load devices 7. The acoustic signal 10 including the generated abnormal sound is measured, the component for each order of the frequency included in the acoustic signal 10 is calculated, and the actually measured frequency component calculation result 15 is extracted. Then, the frequency component coincidence between the actually measured frequency component calculation result 15 and a plurality of abnormal frequency components 17 representing a failure or an anomaly that is a precursor thereof stored in the database 16 in advance is calculated. Since the frequency abnormality pattern 18 having a high degree of coincidence is extracted, it is possible to reliably and accurately estimate a failure of a customer device (including equipment, equipment, and parts) and a frequency abnormality pattern that is a precursor.

  The frequency component coincidence calculation unit 13 includes a normalization calculation processing unit 13a that normalizes the total of the actual frequency component calculation results 15 obtained by the frequency component calculation unit 11, and a coincidence calculation processing unit 13b. The coincidence calculation processing unit 13b calculates the total of the normalized actual measurement frequency component calculation result 22 normalized by the normalization calculation processing unit 13a and the frequency component for each frequency abnormal frequency component 23 stored in the database 16. Since the frequency component coincidence is calculated from the difference from each normalized frequency abnormal frequency component 23 standardized to be 1, and arranged in order of high coincidence, the frequency abnormal pattern 18 can be displayed in an easy-to-view manner, It can be easily ascertained which frequency abnormality pattern 18 has the most influence on the failure or its predecessor abnormality.

(Modification of Embodiment 1)
(Modification 1)
FIG. 16 is a block diagram showing a first modification of the first embodiment of the apparatus failure evaluation system according to the present invention.

  As in the first embodiment, the apparatus failure evaluation system 1 according to the first modification includes a frequency component calculation unit 11, a data recording unit 12, a frequency component coincidence calculation unit 13, and a frequency abnormality estimation unit 14. The

  In the first embodiment, the frequency abnormality coincidence calculation is performed after the frequency abnormality patterns 18,... Having the frequency abnormal frequency components 17,... The means 13 calculates the degree of coincidence 19 between the actually measured frequency component calculation result 15 obtained by the frequency component calculating means 11 and the frequency component of the frequency abnormal frequency component 17 stored in the database 16. 16 instead of a plurality of frequency abnormality patterns 18 having frequency abnormal frequency components 17 as shown in FIG. 16, normal frequency patterns having normal frequency components 17b,... 18b, and so on, and the frequency component coincidence calculation means 13 calculates the coincidence with the actually measured frequency component calculation result 15. It is.

  Therefore, in the first modification, the acoustic measurement device 8 is installed inside the housing of the load device 7, outside the housing, and in the vicinity of the housing, the acoustic signal 10 generated from the load device 7 is measured, and the frequency component calculation unit 11. The point that the waveform analysis of the acoustic signal 10 is performed and the frequency component is calculated is the same as in the first embodiment.

  Based on the configuration shown in FIG. 6, the frequency component coincidence calculation means 13 performs normalization processing on the actual measurement frequency component calculation result 15 and then stores the normalized actual frequency component calculation result 15 and the database 16. The frequency component coincidence 19b with the frequency component of the normalized normal frequency component 17b is calculated. In addition, the normalization process of the frequency component coincidence degree calculation means 13, the normalization difference process, and the process of the frequency component coincidence degree 19b are performed according to the above-described equations (2) to (4).

  As shown in FIG. 13, the frequency abnormality estimation means 14 includes a matching degree rearrangement processing unit 14a and a frequency abnormality extraction unit 14b. The matching degree rearrangement processing unit 14a performs a process of rearranging the normal frequency component 17b based on the frequency component matching degree 19b of the normal frequency component 17b. For example, the components are rearranged in descending order of the frequency component coincidence 19b.

  The frequency abnormality extraction unit 14b includes the normal frequency component 17b of the normal frequency component 17b having the normal frequency component 17b rearranged in order from the highest frequency component coincidence 19b by the coincidence rearrangement processing unit 14a. When the frequency component coincidence 19b of the normal frequency pattern 18b having a low value is low, the normal frequency pattern 18b having the possibility of frequency abnormality is estimated and output as a frequency abnormality estimation result 20.

  Incidentally, when the frequency component coincidence 19b between the actually measured frequency component calculation result 15 and the normal frequency component 17b is high, for example, when both frequency components completely coincide with each other, the normal frequency pattern 18b itself is obtained and the frequency is normal. Means.

  FIG. 17 is a diagram showing the relationship between the frequency component coincidence 19b and the normal frequency component 17b. That is, FIG. 17 shows the normal frequency patterns 18b corresponding to the normal frequency components 17b arranged in order from the left to the right in descending order of the frequency component coincidence 19b, and the frequency component coincidence 19b is from 100%. It is considered that the frequency is within the range of −100%, and the higher the frequency is on the left side.

  Therefore, if the frequency abnormality threshold 26 is set in advance, when the frequency component coincidence 19b is lower than the frequency abnormality threshold 26, it can be estimated that the frequency is abnormal.

(Modification 2)
FIG. 18 is a block diagram showing a second modification of the first embodiment of the apparatus failure evaluation system according to the present invention.

  As in the first embodiment, the apparatus failure evaluation system 1 according to the second modification includes a frequency component calculation unit 11, a data recording unit 12, a frequency component coincidence calculation unit 13, and a frequency abnormality estimation unit 14. The

  A frequency abnormality pattern 18 having a frequency abnormality frequency component 17,..., Which is a failure or a precursor to each load device 7 in the database 16 constituting the data recording means 12, and a frequency estimated to be normal for each load device 7. The normal frequency patterns 18b having the normal frequency components 17b,... Are stored, and the frequency component coincidence calculating means 13 calculates the coincidence with the actually measured frequency component calculation result 15.

  The frequency component coincidence calculation means 13 is stored in the measured frequency component calculation result 15 and the database 16 calculated by the frequency component calculation means 11 based on the configuration shown in FIG. 6 and the expressions (2) and (3). The frequency component coincidence 19, 19b of the frequency component of the abnormal frequency component 17 and the frequency component of the actually measured frequency component calculation result 15 and the frequency component of the normal frequency component 17b is calculated.

  Based on the frequency component coincidence degrees 19 and 19b calculated by the frequency component coincidence degree calculating unit 13, the frequency abnormality estimation unit 14 converts the frequency abnormality pattern 18 having a high frequency component coincidence 19 into a frequency abnormality pattern having a high possibility of frequency abnormality. 18 and the frequency normal pattern 18b having a low frequency component matching degree 19b is estimated as the frequency normal pattern 18b indicating the possibility of frequency abnormality. Then, the frequency abnormality estimation means 14 determines a frequency abnormality from both the frequency component coincidence 19 of the frequency abnormality pattern 18 and the frequency component coincidence 19b of the frequency normal pattern 18b, and the determination result is used as the frequency abnormality estimation result 20. Output as.

  Therefore, according to this modification, not only the harmonic coincidence 19 of the frequency abnormal frequency component of the frequency abnormal pattern 18 having a high possibility of frequency abnormality but also the normal frequency of the frequency normal pattern 18b indicating the possibility of frequency abnormality. It is possible to estimate a failure of the load device 7 or an anomaly that is a precursor while taking into account the harmonic matching degree 19b of the component.

(Modification 3)
FIG. 19 is a block diagram showing a third modification of the first embodiment of the apparatus failure evaluation system according to the present invention.

  In the power grid bus 5 of the customer connected to the upper commercial power grid 2, for example, in a company factory, a lot of facilities, devices, and equipment are installed in each building. Even if there are, different facilities, devices, and equipment are installed on each floor. As a result, it is difficult to estimate a failure of the load device 7 or an abnormality that is a precursor thereof only from the harmonic matching degree of the frequency normal frequency component of the frequency abnormality pattern 18 that has a high possibility of frequency abnormality. For example, not only the frequency abnormality pattern 18 having a high possibility of frequency abnormality, but also the frequency indicating the possibility of frequency abnormality due to the influence of other equipment, device, and equipment even if the equipment, device, and equipment are normal. When the frequency normal pattern 18b of the normal frequency component is generated, or when a plurality of facilities, apparatuses, and devices are connected in parallel to the power system bus 5, the frequency load of the load frequency component that is highly influenced by the load-specific frequency A pattern may be generated. It is desirable to comprehensively grasp the frequency component coincidence 19 and estimate the abnormality of the frequency component.

  Therefore, the apparatus failure evaluation system 1 according to the modified example 3 includes a frequency component calculation unit 11, a data recording unit 12, a frequency component coincidence calculation unit 13, and a frequency abnormality estimation unit 14 as in the first embodiment. Composed.

  The database 16 constituting the data recording means 12 is presumed to have a frequency abnormality pattern 18,... Having a frequency abnormality frequency component 17,... .. Of the normal frequency components 17b,..., And the frequency load patterns 18c of the load frequency components 17c,. , ... are stored.

  In this state, the frequency component coincidence calculation means 13 calculates the actual frequency component result 15 and the frequency abnormal frequency stored by the frequency component calculation means 11 based on the configuration of FIG. 6 and the expressions (2) to (4). The frequency component coincidence 19 with the frequency component of the component 17, the frequency component coincidence 19b between the measured frequency component result 15 and the frequency component of the normal frequency component 17b, the measured frequency component result 15 and the load frequency component 17c. The frequency component coincidence 19c with the frequency component is calculated.

  The frequency abnormality estimation means 14 has a high possibility of frequency abnormality in the frequency abnormality pattern 18 having a high frequency component coincidence 19 out of the frequency component coincidences 19, 19b, 19c calculated by the frequency component coincidence calculation means 13. A frequency normal pattern 18b having a low frequency component coincidence 19b is estimated as a frequency normal pattern 18b indicating the possibility of frequency abnormality, and a frequency load pattern 18c having a high frequency component coincidence 19c is loaded. The frequency load pattern 18c having a high frequency influence is estimated.

  Then, the frequency abnormality estimation means 14 determines the frequency abnormality from the frequency component coincidence 19 of the frequency abnormality pattern 18, the frequency component coincidence 19b of the frequency normal pattern 18b, and the frequency component coincidence 19c of the frequency load pattern 18c, The determination result is output as the frequency abnormality estimation result 20.

  FIG. 20 is a diagram showing a display example of the frequency component matching degree and the frequency pattern. In the drawing, the vertical axis represents the harmonic component coincidence 19, 19b, 19c, and the horizontal axis represents the relationship between the harmonic component coincidence 18 and the patterns 21a to 21c.

  The frequency abnormality estimation means 14 has a function of outputting the frequency abnormality estimation result 20. The function to be output includes a function to display on the display unit in addition to a function to print and output, a function to transmit to other monitoring center devices, a function to accumulate in the database 16 of the system.

  Therefore, as shown in FIG. 20, the frequency abnormality estimation means 14 can display the frequency abnormality estimation result 20 on the display unit.

  In FIG. 20A, the harmonic normal pattern 18b exists above the frequency abnormality extraction threshold 25, the harmonic load pattern 18c is below the frequency normal pattern 18b, and the frequency abnormality pattern 18 is the frequency abnormality extraction threshold. Although it is below 25, since the frequency of the frequency abnormality pattern 18 is affected by the load frequency, it can be determined that the frequency is normal and there is no failure. In other words, it is considered that the frequency of the component that becomes abnormal due to the influence of the load device 7 connected to the bus 5 is affected by other nearby load devices such as a personal computer or a rotating machine.

  On the other hand, in FIG. 20B, the frequency abnormality pattern 18 exists above the frequency abnormality extraction threshold 25 and the harmonic normal pattern 18b is below the frequency abnormality extraction threshold 25, but the frequency abnormality pattern 18 is normal. Since it is above the pattern 18b and the frequency load pattern 18c, it can be determined that the frequency of the frequency abnormality pattern 18 is abnormal and there is a possibility of failure.

  Therefore, according to Modification 3 as described above, the harmonic component coincidence calculation means 13 calculates the coincidence, and can display them in the order of high coincidence, so that each frequency pattern 18, 18b, 18c. The frequency components coincidence 19, 19b, 19c can be comprehensively grasped from the plurality of patterns 18, 18b, 18c, and the abnormality of the frequency components can be estimated.

(Embodiment 2)
FIG. 21 is a block diagram showing Embodiment 2 of the apparatus failure evaluation system according to the present invention.
In addition to the frequency component calculation unit 11, the data recording unit 12, and the frequency component coincidence calculation unit 13 shown in FIG. 2, the apparatus failure evaluation system 1 includes a frequency abnormality estimation unit 14 on the output side of the frequency component coincidence calculation unit 13. Instead of this, a result display control means 31 is provided. Note that the frequency abnormality estimation means 14 and the result display control means 31 all have a role corresponding to the frequency abnormality output means in a broad sense.

  The frequency component coincidence calculation means 13 calculates the frequency component coincidence 19 between the actually measured frequency component calculation result 15 stored in the database 16 and the frequency component of the frequency abnormal frequency component 17, for example, a predetermined storage in the data recording means 12. Store in area and output.

  The result display control means 31 displays the frequency component coincidence 19 for each frequency abnormal frequency component 17 output from the frequency component coincidence calculating means 13 on the display device 32 and returns to the frequency component calculating means 11 at regular intervals. The processes by the constituent units 11, 13, and 31 are repeatedly executed. As a result, in the predetermined recording area of the data recording means 12, the frequency component coincidence degree 19 for each frequency abnormality pattern 18 (18-A to 18-D) for each predetermined time over a required period (for example, 24 hours). Recorded in time series.

  FIG. 22 is a diagram illustrating a display example of the frequency component coincidence degree 19 for each of the frequency abnormality patterns 18 -A to 18 -D by the result display control means 31. The vertical axis represents frequency component coincidence 19, and the horizontal axis represents time.

  The example of FIG. 22 is a display example for 24 hours a day. That is, the frequency component coincidence calculation means 13 accumulates the frequency component coincidence 19 of the frequency abnormality patterns 18-A, 18-B, 18-C, and 18-D in a time series for 24 hours a day. This is an example displayed by the display control means 31.

  According to this embodiment, in addition to the same effects as the first embodiment, each frequency abnormality pattern 18-A, 18-B, 18-C, 18-D for each time over a required period. Since the change of the frequency component coincidence 19 for each time can be displayed, the transition of the temporal change of each frequency abnormality pattern 18 can be grasped, and it is easy for the specific frequency abnormality pattern 18 to become greatly abnormal in any time zone. I can judge.

(Modification of Embodiment 2)
In the second embodiment, the frequency component coincidence 19 of the plurality of frequency abnormality patterns 18 is calculated every predetermined time, accumulated in the database 16 in time series, and stored in the display device 32 in a time series for 24 hours a day, for example. Although the frequency component coincidence degree 19 that changes to is displayed, the combination frequency patterns of the first to third modifications of the first embodiment described above can be applied to the second embodiment.

  That is, the frequency component coincidence calculation means 13 performs the combination frequency pattern of the first to third modifications of the first embodiment, that is, the frequency component coincidence 19b of the frequency normal pattern 18b, the frequency abnormal pattern 18 and the frequency normal pattern every predetermined time. Frequency component coincidence 19, 19b of 18b, frequency abnormal pattern 18, frequency normal pattern 18b, and frequency load pattern 18c of frequency component coincidence 19, 19b, 19c are calculated, stored in database 16 in time series, and For example, the result display control means 31 may display the frequency component coincidence degree 19 that changes in time series over 24 hours a day on the display device 32.

(Embodiment 3)
FIG. 23 is a block diagram showing Embodiment 3 of the apparatus failure evaluation system 1 according to the present invention.
The apparatus failure evaluation system 1 according to the third embodiment has a configuration in which an alarm generation unit 40 is newly provided in addition to the frequency component calculation unit 11, the data recording unit 12, the frequency component coincidence calculation unit 13, and the frequency abnormality estimation unit 14. It is.

  In the alarm generation means 40, the frequency abnormality estimation result 20 or the frequency abnormality generation source 41 is specified by the frequency abnormality estimation means 14. When the degree of frequency abnormality is large at that time, the load having the frequency abnormality (abnormal sound) generation source An alarm 43 is sent to the central monitoring control center 42 that centrally manages the device 7 and the power energy system.

  FIG. 24 is a view for explaining an alarm transmission system of the alarm 43 by the alarm generating means 40. In FIG.

  In the apparatus failure evaluation system 1, an alarm transmission system 44A for transmitting an alarm 43 is connected to each load apparatus 7,... (One load apparatus in FIG. 24), and is connected to the central monitoring control center 42. Similarly, when an alarm transmission system 44B is connected in between, and the degree of frequency abnormality is large, the load transmission device 7 and the central monitoring control center 42 that are the sources of frequency abnormality (abnormal sound) are transmitted via the alarm transmission systems 44A and 44B. On the other hand, an alarm 43 is sent out. The degree of frequency abnormality is a frequency component coincidence degree 19 of the frequency abnormality pattern 18.

  As shown in FIG. 25, the alarm 43 is generated when the frequency warning level threshold value 45 indicating a predetermined frequency warning level exceeds a predetermined time (also called an alarm generation time limit value) 46 as shown in FIG. appear. The frequency warning level threshold 45 may be set separately or the frequency abnormality extraction threshold 26 shown in FIG. 7 may be used.

  Therefore, according to the embodiment as described above, when the frequency abnormality degree is large from the frequency component coincidence 19 of the frequency abnormality pattern 18 and the frequency warning level threshold 45 exceeds the alarm occurrence time limit value 46, the abnormal frequency ( Since the alarm 43 is sent to the load device 7 and the central monitoring and control center 42 that are the source of abnormal sound), it is possible to estimate the failure of the load device (including equipment, equipment, and parts) 7 and the abnormality that is a precursor thereof. The necessary measures can be taken promptly.

(Modification of Embodiment 3)
In the third embodiment, the alarm 43 is transmitted when the frequency component coincidence 19 of the frequency abnormality pattern 18 exceeds the frequency warning level threshold 45 and reaches the alarm occurrence time limit 46. 3 can be applied to the combination frequency patterns of the first to third modifications of the first embodiment described above. That is, the alarm generation means 40 is provided on the output side of the frequency abnormality estimation means 14 of the apparatus failure evaluation system 1 shown in FIGS. 16, 18 and 19, and an abnormal frequency (abnormal sound) is obtained under the conditions shown in FIG. The alarm 43 may be notified to the load device 7 having the generation source and the central monitoring control center 42 that centrally manages the power energy system.

(Embodiment 4)
FIG. 26 is a block diagram showing Embodiment 4 of the apparatus failure evaluation system 1 according to the present invention.
In the fourth embodiment, a control command generating means 47 is newly provided in the configuration shown in FIG.

  That is, the apparatus failure evaluation system 1 includes an alarm generation unit 40 and a control command generation unit 47 on the output side of the frequency abnormality estimation unit 14, and the alarm generation unit 40 generates an alarm 42 in the manner described above.

  On the other hand, when it is determined that a certain load device 7 is a frequency abnormality generation source, the control command generation means 47 sends a control command 48 such as a stop to the load device 7.

  FIG. 27 is a view for explaining a transmission system of the control command 48 by the control command generating means 47.

  In this example, a control command transmission system 49A,... Is newly connected between the operation failure evaluation system 1 and each of the load devices 7,. When it is determined that there is an alarm, the alarm generation means 40 generates an alarm 42 indicating that the negative device 7 is in an abnormal frequency state to the load device 7 via the alarm transmission system 44A.

  On the other hand, when a certain load device 7 is identified as a frequency abnormality generation source by the frequency abnormality estimation unit 14 and the degree of the frequency abnormality is large, the control command generation unit 47 transmits a control command to the specified load device 7. By sending a control command 48 relating to load reduction including a stop via the system 49A, a measure is taken so as not to cause a failure when the load device 7 is abnormal, or a failure of the load device 7 Take immediate recovery measures.

  Furthermore, the alarm generation means 40 sends the control command to stop the load device 7 to the corresponding load device 7 that the corresponding load device 7 is in an abnormal frequency state to the central monitoring control center 42 via the alarm transmission system 44B. Alarm 43 is generated.

  As the generation timing of the control command 48, a frequency risk level threshold value 50 and a fixed time (control command generation time limit value) 51 are newly set as shown in FIG. When the frequency abnormality level exceeds the frequency warning level threshold value 45 and the alarm occurrence time limit value 46, and further exceeds the frequency risk level threshold value 50 indicating the frequency danger level for a certain time (control command occurrence time limit value 51), the operation stops. The control command 48 is sent to the load device 7.

  Therefore, according to the embodiment as described above, the frequency component coincidence between the analysis frequency component result of the acoustic signal 10 generated from the load device 7 and the frequency abnormal frequency component, and an alarm indicating that the frequency abnormality is at a warning level. Therefore, the on-site monitoring personnel can grasp that the load device 7 is deteriorating, and when the influence of the frequency abnormality is in a dangerous state, control that should suppress the frequency abnormality to the source of the frequency abnormality. Since the command is transmitted, it is possible to avoid a disaster due to frequency abnormality.

(Modification of Embodiment 4)
In the fourth embodiment, the alarm that the frequency component coincidence 19 of the frequency abnormality pattern 18 is at a warning level is generated and the operation control of the load device 7 is stopped. The same can be applied to the combination frequency patterns of the first to third modifications. That is, the alarm generating means 40 and the control command generating means 47 are provided on the output side of the frequency abnormality estimating means 14 of the apparatus failure evaluation system 1 shown in FIGS. 16, 18 and 19, and under the conditions shown in FIG. A control command 48 relating to load reduction including a stop can be sent to the load device 7 in a dangerous state via the control command transmission system 49A.

  In addition, the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the scope of the invention.

  DESCRIPTION OF SYMBOLS 1 ... Device failure evaluation system, 2 ... Upper power system, 5 ... Bus, 7 ... Load device, 8 ... Acoustic measuring device, 9 ... Transmission system, 10 ... Acoustic signal, 11 ... Frequency component calculation means, 11a ... FFT processing part , 12 ... Data recording means, 13 ... Frequency component coincidence calculating means, 13a ... Normalization calculation processing section, 13b ... Concordance calculation processing section, 14 ... Frequency abnormality estimation means, 14a ... Concordance rearrangement processing section, 14b ... Frequency anomaly extraction unit 15 ... calculated frequency component calculation result 16 ... database 17 ... frequency anomalous frequency component 17b ... frequency normal frequency component 17c ... load frequency component 18 ... frequency abnormal pattern 18b ... frequency normal pattern 18c ... frequency load pattern, 31 ... result display control means, 32 ... display device, 40 ... alarm generation means, 42 ... central monitoring control center, 47 ... control Decree generating means.

Claims (11)

An acoustic measurement means for measuring an acoustic signal generated from a consumer device (including equipment, equipment, and parts; hereinafter the same) that receives power supply from the power system;
A frequency component calculating means for analyzing a waveform of an acoustic signal measured by the acoustic measuring means and calculating an actually measured frequency component;
Data recording means for storing a frequency abnormality pattern having a frequency abnormal frequency component estimated to be a frequency abnormality that is a failure or a precursor of each customer device for each customer device;
A frequency component coincidence calculating means for calculating a frequency component coincidence between a measured frequency component calculation result obtained by the frequency component calculating means and a plurality of frequency abnormal frequency components stored in the data recording means;
A frequency abnormality output means for estimating and outputting or displaying the frequency abnormality pattern as a precursor or a failure of the consumer device from each frequency component coincidence calculated by the frequency component coincidence degree calculation means; An equipment failure evaluation system. An acoustic measurement means for measuring an acoustic signal generated from a consumer device receiving power supply from the power system;
A frequency component calculating means for analyzing a waveform of an acoustic signal measured by the acoustic measuring means and calculating an actually measured frequency component;
Data recording means for storing a frequency normal pattern having a frequency normal frequency component estimated as a normal frequency of each consumer device for each consumer device;
A frequency component coincidence calculating means for calculating a frequency component coincidence between a measured frequency component calculation result obtained by the frequency component calculating means and a plurality of normal frequency components stored in the data recording means;
Of the frequency component coincidence calculated by the frequency component coincidence calculating means, a frequency normal pattern corresponding to the frequency normal frequency component having a low frequency component coincidence is defined as a frequency normal pattern indicating the possibility of frequency abnormality. An apparatus failure evaluation system comprising frequency abnormality output means for estimating, outputting or displaying. An acoustic measurement means for measuring an acoustic signal generated from a consumer device receiving power supply from the power system;
A frequency component calculating means for analyzing a waveform of an acoustic signal measured by the acoustic measuring means and calculating an actually measured frequency component;
For each of the customer devices, a frequency abnormality pattern having a frequency abnormal frequency component that is estimated as a frequency abnormality that is a failure or a precursor of each customer device, and a normal frequency that is estimated as a normal frequency of each of the customer devices. Data recording means for storing a normal frequency pattern having components;
Frequency component for calculating the frequency component coincidence between the actual frequency component calculation result obtained by the frequency component calculating means, the plurality of frequency abnormal frequency components stored in the data recording means, and the plurality of normal frequency frequency components. A degree-of-match calculation means;
Of the frequency component coincidence calculated by the frequency component coincidence calculating means, a frequency normal pattern corresponding to the frequency abnormal frequency component having a high frequency component coincidence is a frequency normal pattern having a high possibility of frequency abnormality. And a frequency abnormality output means for estimating and outputting or displaying a frequency normal pattern corresponding to the frequency normal frequency component having a low frequency component coincidence as a frequency normal pattern indicating the possibility of frequency abnormality. An apparatus failure evaluation system characterized by that. In the apparatus failure evaluation system according to claim 3,
The data recording means stores a frequency load pattern of a load frequency component estimated to be a frequency unique to another consumer device to which each consumer device is connected in parallel for each consumer device,
The frequency component coincidence calculation means includes an actual frequency component calculation result obtained by the frequency component calculation means and a plurality of frequency abnormal frequency components stored in the data recording means, the plurality of frequency normal frequency components, and the plurality of frequency components. Calculate the frequency component coincidence with the load frequency component,
The frequency abnormality output means has a frequency normality pattern corresponding to the frequency abnormality frequency component having a high frequency component coincidence among the frequency component coincidence calculated by the frequency component coincidence calculation means as a possibility of frequency abnormality. The frequency normal pattern is estimated to be a normal frequency pattern with a high frequency, the frequency normal pattern corresponding to the frequency normal frequency component with a low frequency component matching degree is estimated as a frequency normal pattern indicating the possibility of frequency abnormality, and the frequency component matching A device failure evaluation system, wherein a frequency load pattern corresponding to a load frequency component having a high degree is estimated as the frequency load pattern having a high frequency influence by the other consumer device, and is output or displayed. In the apparatus failure evaluation system according to claim 1,
The apparatus for evaluating apparatus failure, wherein the frequency abnormality output means displays in order from a frequency abnormality pattern having a frequency abnormality frequency component having a high degree of coincidence of the frequency components. In the apparatus failure evaluation system according to claim 2,
The apparatus for evaluating failure of a device, wherein the frequency abnormality output means displays in order from a frequency normal pattern having a frequency normal frequency component having a high frequency component coincidence. In the apparatus failure evaluation system according to claim 3,
The apparatus for evaluating an apparatus failure, wherein the frequency abnormality output means displays a frequency abnormality pattern having the frequency abnormality frequency component and a frequency normal pattern having a frequency normal frequency component in descending order of the frequency component coincidence. In the apparatus failure evaluation system according to claim 4,
The frequency abnormality output means displays a frequency abnormality pattern having the frequency abnormality frequency component, a frequency normal pattern having a frequency normal frequency component, and a frequency load pattern having the load frequency component in descending order of the frequency component coincidence. Equipment failure evaluation system characterized by In the apparatus failure evaluation system according to any one of claims 1 to 4,
The frequency abnormality output means repeatedly executes the frequency component calculation means and the frequency component coincidence calculation means at predetermined times over a required period, thereby changing the frequency coincidence in each pattern in time series. An apparatus failure evaluation system characterized by displaying. In the apparatus failure evaluation system according to any one of claims 1 to 4,
Generation of an alarm for notifying alarm information for calling attention to the consumer device that is a frequency abnormality generation source when the frequency abnormality level exceeds a predetermined frequency warning level threshold on the output side of the frequency abnormality output means An apparatus failure evaluation system comprising means. In the apparatus failure evaluation system according to any one of claims 1 to 4 and claim 10,
Control that sends a control command for operation suppression or stop to the consumer device that is a frequency abnormality generation source when the degree of frequency abnormality exceeds a predetermined frequency risk level threshold value on the output side of the frequency abnormality output means An apparatus failure evaluation system, comprising command generation means.

Images