KR100736166B1 - Apparatus and method for detecting failure signal of electrical equipment - Google Patents

Abstract

The present invention provides a method for diagnosing a signal abnormality of an electrical equipment, generating input arrays by dividing measured voltage data read from an outlet of an electrical equipment into a preset array size for each preset acquisition cycle, and inputting the input arrays to a memory unit. Performing a fast Fourier transform sequentially on each of the stored input arrays; and a first process of storing the input arrays, and a second process of storing the array acquisition time for each of the input arrays as time arrays. Performing a third process of storing the result arrays in the result array storage of the memory unit, a fourth process of selecting predetermined frequency arrays from the result sample values of the result arrays, and time-energy conversion using the selected frequency arrays. A fifth process of displaying time-energy coordinate values on the coordinate screen of the monitor is performed.

Electrical equipment, signal error, measured voltage, personal computer

Description

Apparatus and method for diagnosing signal abnormalities in electrical equipment {APPARATUS AND METHOD FOR DETECTING FAILURE SIGNAL OF ELECTRICAL EQUIPMENT}

1 is a schematic configuration diagram of an electrical equipment signal abnormality diagnosis apparatus using a personal computer according to an embodiment of the present invention;

2 is a block diagram of a signal processing unit implemented in a PC according to an embodiment of the present invention;

3 is a control flowchart of a controller in a PC according to an embodiment of the present invention;

4 is a memory map for explaining signal processing on measured voltage data applied to a PC;

5 is a memory map diagram and a screen configuration diagram for performing a time-energy conversion from the result arrays of FIG. 4 and showing the result;

6 and 7 are views for explaining the FFT and time-energy conversion according to an embodiment of the present invention;

8a to 8d is a view showing the state of the electrical equipment in a diagnostic method according to an embodiment of the present invention,

9 is a schematic configuration diagram of an electrical equipment signal abnormality diagnosis device implemented as a separate diagnostic jig according to another embodiment of the present invention;

The present invention relates to an apparatus and method for diagnosing abnormalities in electrical equipment, and more particularly, to an apparatus and method for diagnosing electrical equipment abnormalities in which electrical equipment abnormalities can be easily diagnosed using a personal computer.

Existing electrical equipment abnormality diagnosis systems require expensive measuring equipment, and the measuring method was cumbersome. In addition, it was difficult to use such that only an expert such as an electrician could operate an electrical equipment fault diagnosis system.

Accordingly, a method for easily and simply detecting and confirming an abnormality diagnosis of an electrical installation without a specialist was desired.

In addition, distributed power generation systems, such as fuel cells, will be distributed to homes in the next few years. Distributed power generation systems allow people to produce and use electricity on their own, even those with limited knowledge of electricity.

If a distributed power generation system such as a fuel cell is spread in each household, the general public should also be able to detect abnormalities of electrical installations easily, simply, and at a low cost.

Accordingly, it is an object of the present invention to provide an apparatus and method for diagnosing signal failure of an electrical equipment that can easily diagnose electrical equipment abnormality with a personal computer.

Another object of the present invention is to provide an apparatus and method for diagnosing signal abnormalities of electrical equipment that can easily and simply detect abnormalities of electrical equipment.

It is still another object of the present invention to provide an apparatus and method for diagnosing electrical failure signal abnormalities that can be applied to pattern analysis using neural networks to automate and predict diagnosis of electrical equipment signal abnormalities.

According to the above object, the present invention, in the electrical equipment signal abnormality diagnosis method, by generating the input array by dividing the measured voltage data read from the outlet of the electrical equipment by the preset array size for each preset acquisition cycle, A first process of storing the arrays in an input array storage of a memory unit, a second process of storing the array acquisition time for each of the input arrays as time arrays in a time array storage, and sequentially for each of the stored input arrays A fourth process of performing fast Fourier transform and storing the result arrays in the result array storage of the memory unit, a fourth process of selecting predetermined frequency arrays from the result sample values of the result arrays, and using the selected frequency arrays A fifth process of performing time-energy conversion and displaying time-energy coordinate values on the coordinate screen of the monitor Characterized in that made.

The present invention implements to diagnose the signal abnormality of the electrical equipment based on the following matters.

First, use only the voltage that is easiest to measure from the electrical installation.

Second, in order to lower the cost of the system, signal processing can be implemented in software using a personal computer distributed in each household.

Third, by visualizing the analysis results, the general public can easily detect abnormalities in the electrical installation.

Fourth, to reduce the number of data to be applied to the automatic detection system using a neural network.

Fifth, the neural network is implemented to warn the signal abnormality of the electrical equipment early.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the same elements in the figures are denoted by the same reference numerals or the same reference signs wherever possible. In addition, detailed descriptions of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.
Voltage waveforms in normal operation in electrical installations have fundamental wave components as shown in Figure 8a, but voltage waveforms in case of signal abnormality due to an accident are more and more impending as accidents are shown in Figure 8b, 8b, 8c. It will contain more high frequency components. Therefore, in case of FFT voltage waveform in normal operation of electric equipment, most of them show fundamental wave component, but when FFT voltage waveform in case of accident (switch opening / closing, etc.), high frequency component is generated, The higher the range of the generated high frequency region is, the larger the size becomes.
In the embodiment of the present invention, the FFT is performed using the measured voltage data so that the magnitude change of the high frequency components can be known, and the time-energy conversion is performed to select the selected high frequency components (for example, when it is a steady state and an accident). By combining the high frequency components with a large change in magnitude when they occur, the size of the high frequency components is increased to make it easier for the measurer to observe the change in size.

1 is a schematic configuration diagram of an electrical equipment signal abnormality diagnosis apparatus using a personal computer according to an embodiment of the present invention, Figure 9 is an electrical equipment implemented as a separate diagnostic jig (jig) according to another embodiment of the present invention It is a schematic block diagram of the signal abnormality diagnosis apparatus.

Therefore, the electrical equipment signal abnormality diagnosing apparatus 2 of FIG. 1 may be implemented by using a personal computer widely distributed in each home, and thus, it is easy to disseminate the electrical equipment signal abnormality diagnosing apparatus 2 of FIG. The advantage is that it can be implemented separately without a personal computer.

 FIG. 1 schematically shows a configuration in which a voltage value is obtained from a digital oscilloscope 8 of an abnormality diagnosis device 2 of an electrical installation through an outlet 6 of an electrical installation 4 and transmitted to the personal computer 10. .

The electrical equipment abnormality diagnosing apparatus 2 of FIG. 1 comprises a digital oscilloscope 8 and a personal computer 10 (hereinafter referred to as a "PC"), and the digital oscilloscope 8 comprises an outlet 6 of the electrical equipment 4. Is a popular type measuring instrument connected to the PC, and the PC 10 has a software built-in signal processing unit 20 that analyzes the electrical equipment signal abnormality and displays the result.

The electrical equipment abnormality diagnosis device 2 of the diagnostic jig form of FIG. 9 includes a signal acquisition unit 8a and a signal processing unit 20a. The signal acquisition unit 8a functions as the digital oscilloscope 8 of FIG. The signal processor 20a is responsible for the function of the signal processor 20 of FIG. 1.

Therefore, the electrical equipment abnormality diagnosing apparatus 2 of the present invention can be implemented as shown in FIG. 1 or FIG. 9, and the specific configuration and operation of the electrical equipment abnormality diagnosing apparatus 2 in the form shown in FIG. This will be explained.

Referring to FIG. 1, a measurer connects a digital oscilloscope 8 to an outlet 6 in an electrical installation 4, and measures the voltage from the outlet 6 using a digital oscilloscope 8. The digital oscilloscope 8 measures the voltage at each preset sampling time period Ts and applies the measured voltage data to the PC 10 in digital form. The sampling time period Ts depends on the performance of the digital oscilloscope 8, and the better the performance of the digital oscilloscope 8 is, the shorter the sampling time period Ts is, but it must be considered that the price is high instead. Therefore, an embodiment of the present invention will be described using an example of a low-cost digital oscilloscope 8 that is suitable for home use and has no problem in detecting a signal abnormality. As an example, the sampling time period Ts of the digital oscilloscope 8 can be used in a range of at least 0.1 MHz.

The measured voltage data applied from the digital oscilloscope 8 to the PC 10 is composed of a plurality of voltage sample values to be described later with reference to FIG. 4.

When the measurement voltage data is applied from the digital oscilloscope 8 to the PC 10, the signal processing unit 20 of the PC 10 performs signal processing to analyze whether the signal is abnormal, and the user easily knows the signal analysis result. Graphically display for viewing on a monitor.

2 is a signal processor 20 embedded in the PC 10 according to an exemplary embodiment of the present invention, which includes a controller 22, a memory 24, a fast fourier transform unit 34, and a time-energy conversion. The unit 35 includes a key input unit 36, a neural network analysis unit 37, and a monitor 38. The memory unit 24 includes an input array storage 26, a time array storage 28, a result array storage 30, and an output array storage 32. The FFT unit 34, the time-energy converter 35, and the neural network analyzer 37 may be implemented by a software algorithm in the controller 22, or may be implemented in a separate chip or logic form. .

3 is a control flowchart of the control unit 22 of the PC 10 according to an embodiment of the present invention.

FIG. 4 is a memory map diagram of the memory unit 24 for explaining signal processing for measured voltage data applied to the PC 10. FIG. 5 performs time-energy conversion from the result arrays of FIG. 6 and 7 are diagrams for explaining the FFT and time-to-energy conversion according to an embodiment of the present invention.

8A to 8D are diagrams showing the state of electrical equipment by the diagnostic method according to the embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

When the measured voltage data is applied to the controller 22 of the PC 10, the controller 22 of FIG. 2 checks this in step 100 of FIG. 3 and proceeds to step 102 of FIG. 3. In step 102 of FIG. 3, the controller 22 divides the measured voltage data into a preset array size S for each preset array acquisition period Ta to generate respective input arrays. That is, as shown in Figure 4, the input array obtained from the timing of the acquisition time t0 (timing) is 'input array-1', the input array obtained from the acquisition time t1 is 'input array-1'. .., The input array obtained from the acquisition time t (n-1) is generated as 'input array-n'. Each input array consists of a plurality of voltage sample values.

The array acquisition period Ta according to an embodiment of the present invention is shorter than the array size S, which is because the performance of the digital oscilloscope 8 can be sufficiently operated even at a low specification that can be purchased inexpensively at home. . That is, in the exemplary embodiment of the present invention, some voltage sample values of the measured voltage data are overlapped with each other in adjacent input arrays.

Since the array acquisition period Ta in the present invention may vary depending on the performance of the digital oscilloscope 8, as in the example shown in FIG. 4, the array acquisition period Ta may be shorter than the array size S as in the case where the digital oscilloscope 8 has a low specification. It is to be understood that the digital oscilloscope 8 may be similar to the array size S and may be longer than the array size S when the digital oscilloscope 8 is of high specification. The array size S is defined as a power of two.

After performing step 102 of FIG. 3, the controller 22 sequentially stores each of the input arrays in the row structure of the input array storage unit 26 of the memory unit 24 in step 104 of FIG. 3. That is, as shown in the input array storage unit 26 of the memory unit 24 of FIG. 4, the input array-1, the input array-2, and the input array are stored at respective storage positions in the row direction of the input array storage unit 26. -3, ... Store the input array-n sequentially. In addition, in step 106, the control unit 22 sets a time acquisition sequence t0, t1, t2 .., t (n-1) to the time array storage unit 28, that is, time array sample-1 for each of the input arrays. , Time sequence sample-2, time sequence sample-3, ..., time sequence sample-n. The time sequence may be array acquisition time + alpha (α = constant). That is, the time sequence may be time samples (t0 + α), (t1 + α), (t2 + α) .. (t (n-1) + α).

5B illustrates a memory map of the time array storage 28.

The controller 22 checks whether the storage of the input array-n and the time-array sample-n is completed in step 108 for sequential input array storage and time array storage, and proceeds to step 110 in FIG. do.

In step 110 of FIG. 3, the controller 22 reads the input array values in the order in which they are stored among the input arrays stored in the raw direction in the input array storage unit 26 in the raw direction. The resultant array values are obtained by controlling the FFT processing at the multiplier), and each of the result array values is sequentially stored in the column array in the result array storage unit 30.

The values before and after the FFT process are represented in the memory map of the input array storage 26 of FIG. 4 and the memory map of the result array storage 30.

Referring to the memory map diagram of the result array storage unit 30 of FIG. 4, when the result arrays, that is, the result array-1 through the result array-n, are stored in the column direction of the memory map, Since the frequencies are the same in the raw direction in the sample values, the array in the raw direction of the memory map may be defined as a frequency array from frequency array-0 to frequency array-k.

After performing step 110 of FIG. 3, the controller 22 selects predetermined frequency arrays from the result sample values in each result array value in step 112 of FIG. 3. The selected frequency arrays are obtained through the experience of the observer considering the object to be observed and the measurement conditions. For example, the number of frequency arrays to be selected is preferably set to one to forty when frequency array -k = " 256 ". The present inventors have obtained through experiments that it is effective to select several in the low frequency band among the frequency arrays in the case of electrical equipment.

Selecting a few frequency arrays is for the time-to-energy conversion to be performed later. Since only a few are selected, the amount of data to be processed in the system can be sufficiently reduced and the data processing time can be reduced.

Then, the control unit 22 performs steps 114 to 118 of FIG. 3 to perform time-to-energy conversion using the selected frequency array and the time array to obtain energy by the sum of corresponding samples of the selected frequency arrays for each array acquisition time. The time-energy coordinate value is displayed as a graph on the coordinate screen of the monitor 38. The display of the time-energy coordinate values of the energies by the sum of the corresponding samples of the selected frequency arrays makes it easier to observe the size change of the measurer by making the size change of the FFT high frequency components larger in the case of a fault condition.

In detail, as the control unit 22 controls the time-energy conversion unit 35 in step 114 of FIG. 3, the sample components of the same column between the frequency arrays selected by the time-energy conversion unit 35 are described. If the calculation is to add to each other or if the selected frequency array is one, the result is provided to the controller 22 without calculation.

For example, if the selected frequency arrays are frequency array-0 and frequency array-1 shown in FIG. 4, the time-energy converter 35 corresponds to the same time sample of the time array in frequency array-0 and frequency array-1. Frequency sample components are added to each other, that is, "0.1 + 50", "0.2 + 54", "0.1 + 51", ..., "0.2 + 51.3", the resulting sample component value (energy) "50.1" , "54.2", "51.1", ..., "51.5" are provided to the control unit 22.

As another example, if the selected frequency array is only one frequency array and it is the frequency array-k shown in FIG. 4, the time-energy conversion unit 35 is a corresponding sample of the frequency array-k, that is, "0.01", " 0.01 "," 0.01 ", ...," 0.01 "are provided to the control unit 22 as a result.

Accordingly, the controller 22 stores the result in the output array storage 32 as an output array (step 114 in FIG. 3). 5A illustrates a memory map of the output array storage unit 32.

The controller 22 determines whether the storage of the output array is completed in the output array storage unit 32 in step 116 of FIG. 3, and proceeds to step 117 of FIG. 3 when complete. In step 117 of FIG. 3, the controller 22 determines the state of the electrical equipment through pattern analysis using the neural network analyzer 37. The neural network analysis unit 37 performs a pattern analysis by inputting the values of the samples of the output array, and classifies the state of the electrical equipment into a plurality of stages as a result of the pattern analysis. Level 1), accident imminent status (2 levels), accident status (level 3), etc. are classified and provided to the controller 22.
In the neural network analyzing unit 37, the sample values of the output array are input values, and three displayed in the electrical equipment state 210 of FIGS. 8A to 8D are given as output values of the neural network, and are shown in FIGS. 8A to 8D. When the values of the samples of the output array of the result graph are input, a rating step ("000" in FIG. 8A, "001" in FIG. 8B, "001" in FIG. 8C) corresponding to the electrical equipment state 210 of FIGS. 011 "In FIG. 8D, the state of" 111 ") is learned.

Employment of the neural network analysis unit 37 in the embodiment of the present invention enables the rapid state automatic detection and early warning function of the electrical equipment.

After performing step 117 of FIG. 3, the controller 22 proceeds to step 118 of FIG. 3 to configure a coordinate screen in which the time arrangement is the x-axis (time) and the output arrangement is the y-axis (energy). On the screen of 38), the result graph 200 of the time-energy coordinate value and the state 210 of the electrical equipment as shown in (c) of FIG. 5 are displayed.

Referring to FIGS. 8A to 8D, a specific example will be described. The controller 22 displays the result graph 200 and the electrical equipment state 210 on the monitor 38 as shown in FIG. 8A when the electrical equipment is in a normal state. When the electrical equipment is in an accident-prone state (step 1), the result graph 200 and the electrical equipment state 210 are displayed on the monitor 38 as shown in FIG. 8B, and the electrical equipment is in an accident imminence state (step 2). In FIG. 8C, the result graph 200 and the electrical equipment state 210 are displayed on the monitor 38 as shown in FIG. 8C. When the electrical equipment is in the accident stage state, the result graph 200 and the electrical equipment state 210 as shown in FIG. 8D are displayed on the monitor 38.

When the display is displayed on the monitor 38, the general public can easily find out whether the electrical equipment is abnormal (eg, a normal state or an electrical equipment failure).

6 and 7 are diagrams for explaining an FFT transformation and a time-energy transformation according to an embodiment of the present invention.

First, referring to FIG. 6, if the FFT is performed on each section, for example, section A in the measured voltage data of the time domain as shown in FIG. 6A (corresponding to step 110 of FIG. 3), FIG. It is converted to the measured value of the frequency domain as shown in FIG. 3, where time-energy conversion is performed again (corresponding to step 114 of FIG. 3), as represented by (t ₀ , C) as shown in FIG. Lose.

At this time, since the energy is conserved by arranging the parser feet, the area A hatched in FIG. 6A and the hatched B in FIG. 6B are the same. C is the value of the area of hatched B expressed as an energy value at the start time t ₀ of the interval.

In FIG. 7, which extends the concept of FIG. 6, if each of the intervals (section 1, section 2, .., section n) has the same size, each transformation (result array 1, result array 2, .., result array n) In Equation 2), since the frequencies are the same, Equations 1 and 2 are equal to each other. In FIG. 7, an example of measured voltage data x (t) is measured voltage data of FIG. 4.

The time-energy conversion used in the embodiment of the present invention uses the start time of each section (t ₀ , t ₁ , ..., t _n-1 in FIG. 7) as the time axis, and requires a measurer in Equation 2 This is achieved by finding the energy in Equation 3, which adds only the corresponding sample components of the frequency array. In addition, when the measured voltage data are classified, the number of data may be changed by adjusting the array acquisition period Ta.

In the above description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be defined by the described embodiments, but should be determined by the equivalent of claims and claims.

As described above, the electrical equipment abnormality diagnosis method of the present invention has an advantage that anyone can easily measure by measuring only the voltage of the electrical equipment. In addition, by analyzing the measured voltage signal through a personal computer, it is possible to reduce the cost of diagnosing abnormality of the electrical equipment, and visualize the analysis result. In addition, by having an automatic detection and early warning function using the neural network analysis unit has the advantage that the general public can easily and quickly determine whether the signal of the electrical equipment.

Claims (4)

In the electrical equipment signal abnormality diagnosis method, A first process of generating input arrays by dividing the measured voltage data read from the outlet of the electrical equipment into a preset array size for each preset acquisition cycle and storing each input array in a row direction in the input array storage unit of the memory unit; , A second process of storing an array acquisition time for each of the input arrays as a time array in a time array storage; A third process of sequentially performing fast Fourier transform on each of the stored input arrays and storing the resultant arrays in the columnar storage unit of the result array storage unit of the memory unit; A fourth process of selecting predetermined frequency arrays from a plurality of frequency arrays defined in a row direction in the result array storage unit; The time-energy conversion using the selected frequency arrays and the time array stored in the time array storage unit performs time-energy conversion of the corresponding samples of the selected frequency arrays for each array acquisition time. Electrical equipment signal abnormality diagnosis method comprising the fifth step of displaying the energy coordinate value. The method of claim 1, wherein the output arrangement after performing the time-energy conversion is determined through a pattern analysis using a neural network to determine the state of the electrical equipment as one of a plurality of preset steps, and the determined electrical equipment state is displayed on the monitor screen. And a sixth step of displaying the electrical equipment signal abnormality diagnosis. In electrical equipment signal abnormality diagnosis device: A measuring device connected to an outlet of an electrical facility and measuring voltage data at a predetermined sample time period and outputting measured voltage data; A signal processor configured to be mounted on a personal computer, to perform signal processing for analyzing an abnormality of an electrical equipment signal using the measured voltage data, and to graphically process a signal analysis result and display the same on a monitor; The signal processing unit; A memory unit including an input array storage unit, a time array storage unit, a result array storage unit, and an output array storage unit; The input voltage arrays are generated by dividing the measured voltage data from the measurement equipment into a preset array size for each preset acquisition cycle, and storing each of the generated input arrays in a row direction in the input array storage unit, respectively. The array acquisition time for each of the arrays is stored as a time array in the time array storage, and a fast Fourier transform is sequentially performed on each of the stored input arrays, and the result arrays are stored in the column array in the result array storage of the memory. And selecting predetermined frequency arrays from a plurality of frequency arrays defined in the row direction in the result array storage unit, and performing time-energy conversion using the selected frequency array and time array to select each array acquisition time of the time array. The energy by the sum of the corresponding samples of the frequency arrays Stored in the unit time and the coordinates on the screen of the monitor-Electric signal over diagnostic apparatus characterized in that it consists of a control unit for displaying the coordinate value energy. In electrical equipment signal abnormality diagnosis device: A signal acquisition unit connected to an outlet of an electrical facility and measuring voltage data at a predetermined sample time period and outputting measured voltage data; A signal processor configured to perform signal processing for analyzing an abnormality of an electrical equipment signal using the measured voltage data and to graphically process a signal analysis result and display the same on a monitor; The signal processing unit; A memory unit including an input array storage unit, a time array storage unit, a result array storage unit, and an output array storage unit; The input voltage arrays are generated by dividing the measured voltage data from the measurement equipment into a preset array size for each preset acquisition cycle, and storing each of the generated input arrays in a row direction in the input array storage unit. An array acquisition time for each is stored as a time array in a time array storage, and a fast Fourier transform is sequentially performed on each of the stored input arrays, and the result arrays are stored in a column direction in the result array storage of the memory. And selecting predetermined frequency arrays among the plurality of frequency arrays defined in the row direction in the result array storage unit to perform time-energy conversion using the selected frequency array and the time array stored in the time array storage unit. By the sum of the corresponding samples of the selected frequency arrays for each acquisition time of And a control unit for storing energies in the output array storage unit and displaying time-energy coordinate values on a coordinate screen of a monitor.

Images