JP6554024B2 - Monitoring device - Google Patents

Description

  The present invention relates to a noise detection technique of a monitoring apparatus (power supply monitoring unit) used in a power monitoring system.

  Network wiring is required when constructing a power monitoring system for monitoring power meters and electric leakage in factories and buildings. There are various standards in the network, and it is necessary to perform network wiring that the monitoring device supports. After network wiring, if the monitoring system is operated, noise may prevent communication. Since the generation timing, generation interval, and noise level differ depending on the environment in which the noise is used, it takes time and money to specify that the event that prevents communication is caused by the noise.

  As a technique of the detection method of noise, there exists Unexamined-Japanese-Patent No. 2009-260450 (patent document 1). According to Patent Document 1, "a communication telegram transmitted from a transmission line and a waveform signal are stored in one end storage means, and can be synchronized at acquisition time by synchronization determination means 108. Analysis waveform information generation means 109 is synchronized with communication telegrams. The portion corresponding to the ideal waveform signal stored in the ideal waveform information storage unit 112 of the waveform signal is extracted as an analysis waveform signal to be analyzed, and the communication state determination unit 113 is extracted by the analysis waveform information generation unit 110. The communication state of the transmission path is estimated based on the result of comparing the analysis waveform signal and the ideal waveform signal stored in the storage means, and the estimation result is displayed by the analysis result display means 114 (see summary). ) ".

JP 2009-260450 A

  If noise occurs, communication of the monitoring system will stop. When noise occurs is unknown, and when it is noticed, the noise may have already disappeared. In addition, there is a problem in that it is impossible to notice the presence or absence of noise when communication is possible even when installed in an environment with noise. The invention described in Patent Document 1 analyzes communication messages and waveform signals in association with each other. The processing unit to be associated is a comparison process, which requires a processing time and a message comparison process, and there is a problem that it takes time for the processing speed. Noise is fast, and to measure fast waveforms, we need to minimize processing. In addition, the same processing is performed regardless of the noise level, and when the noise level is small, there is a problem in that processing is performed even for noise at a level that does not affect communication, for example.

  It is an object of the present invention to provide a monitoring device that can determine whether or not there is noise when introducing a monitoring system and can detect high-speed noise with high accuracy in a short time.

  In order to solve the above-described problems, the present invention is characterized in that noise detection is performed only by a value derived by Fourier expansion, thereby performing high-speed processing, and noise detection processing is distributed according to noise levels. .

If a typical example of the “monitoring device” of the present invention is given,
A monitoring device connected to a network and used in a monitoring system that collects data from each device with higher-order software , comprising: a first noise detecting unit that detects noise that does not affect a communication message; Noise detection means Fourier-expands the transmission signal, detects the transmission signal with the fundamental wave component of the Fourier expansion, and detects the transmission signal, each harmonic component of the Fourier expansion exceeds the threshold in the communication message range determining whether it has an shall detect the noise.

  In the monitoring apparatus of the present invention, it is preferable that the monitoring apparatus further includes second noise detecting means for detecting noise that rewrites a communication message.

  Moreover, it is preferable that the monitoring apparatus of the present invention further includes third noise detecting means for detecting instantaneous peak noise.

  According to the present invention, it can be determined whether or not there is noise when the monitoring system is introduced, and high-speed noise can be accurately detected in a short time. In addition, it is possible to notify that the communication of the monitoring system has stopped due to the noise generation.

It is the figure which showed the kind of noise. It is the figure which showed the structure which detects the noise which does not affect a message | telegram of Example 1 of this invention. It is the figure which showed the sampling of the message | telegram. It is the figure which Fourier-expanded the transmission signal containing noise. It is the figure which showed the distortion waveform. It is the figure which showed the message | telegram code. FIG. 7 is a diagram showing a flow of detecting noise that does not affect a message according to the first embodiment. FIG. 7 is a block diagram of noise detection means which does not affect the message in the first embodiment. It is the figure which showed the structure which detects the noise which affects a message | telegram of Example 2 of this invention. It is the figure which showed the structure which detects abnormal peak noise of Example 3 of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings for explaining the embodiments, the same components are denoted by the same names and symbols, and the repeated explanation thereof is omitted.

FIG. 1 shows an example of noise generated in the network. In the figure, reference numeral 100 indicates a steady telegram waveform without noise. There are three types of noise levels. The first one is a level indicated by reference numeral 101 which does not affect the communication message, and no communication stoppage occurs. Therefore, the user can not notice the noise. However, the installation environment is obviously bad, and it can not be overwhelmed when communication outages occur.
The second is a noise level indicated by reference numeral 102 that affects the communication message, and communication stops. By stopping the communication, the user can notice abnormality. Reference numeral 104 denotes a threshold value at which a message is rewritten, and is noise exceeding the threshold value.
The third is a noise with an instantaneous large peak indicated by reference numeral 103. This noise is generated only when the equipment starts and stops and disappears in an instant, so neither the user nor the system provider can confirm the presence or absence of noise.
In the present invention, noise is decomposed into these three types to detect noise.

  FIG. 2 shows the configuration of a monitoring apparatus for detecting noise that does not affect communication messages. The monitoring device is connected to the host software on the RS-485 network. A transmission signal from the RS-485 network is converted into a data signal by the communication circuit unit 201 and sent to the central processing unit 203. Further, data from the central processing unit 203 is sent from the RS-485 network to the upper software (not shown) via the communication circuit unit 201. An amplifier circuit 200 such as a differential amplifier is provided in parallel with the communication line of the communication circuit unit 201 of RS-485. The output of the amplifier circuit is converted into a digital signal by the A / D conversion unit 202 and input to the central processing unit (CPU) 203. By providing the amplifier circuit 200 in parallel with the communication circuit unit 201 that transmits the communication message, a noise signal can be transmitted to the central processing unit 203 separately from the communication message signal. In the figure, FG indicates a ground line.

  FIG. 3 shows sampling of communication messages. The sampling is performed by the A / D converter 202 of FIG. The communication message speed is 9,600 bps, 19,200 bps, 38,400 bps on the product, and sampling is performed at a cycle faster than this speed. Sampling is performed by dividing one cycle 300 into 32 divisions 301 at each transmission rate, and the communication message of the monitoring device is 11 bits.

  FIG. 4 shows the relationship between the transmission signal and noise. The first order waveform (fundamental wave) is a transmission waveform 400. Noise is superimposed on this transmission waveform. Noise is a harmonic and is superimposed on a primary waveform in a frequency band 401 of secondary, tertiary, and so on. The effective value of each order can be determined by performing Fourier expansion.

As shown in FIG. 5, first, the sampling divided into 32 is performed. The sampled values are subjected to Fourier expansion according to the following equation.
If the distortion waveform in FIG. 5 is y (t) 500, it can be interpolated with a trigonometric function, and is represented by a sin wave and a cos wave.

Further, assuming that one period is divided into 32 and each sampling value at that time is y _x , each Fourier coefficient in (Equation 1) is expressed as follows.

It is as follows when only a fundamental wave is taken out from the above-mentioned formula.

By obtaining a ₁ from (Expression 5) and b ₁ from (Expression 6), sampling data v of 32 fundamental wave signals is calculated by the following expression.

The value of each order obtained from the above equation is calculated by the following equation to obtain the effective value.

Here, the communication message of the fundamental wave (first order) is unique to the device, and the data shown in FIG. When the above-described Fourier expansion is performed on this message, the value is Vrms = 5 V. That is, it is possible to determine whether it is the range of the telegram by the 1st order voltage = 5V. Noise is detected only at the point where the 1st order voltage = 5V. Since the part other than the primary voltage = 5 V is not related to the communication telegram, noise detection is not performed on this part. This can reduce the analysis process.

  In the telegram signal, all effective values after the second order are effective values due to noise. In the telegram signal, data is rewritten at a threshold of 0.4V. Since the noise waveform of each order can be approximated to a sine wave, if the peak value is 0.4 V or less, that is, the effective value is 0.282 V or less, it can be recognized as normal operation. Stationary noise detection is performed using this value. The threshold value of the other order has a margin of 0.2 V as a threshold value. By using this threshold value, it is not necessary to perform a comparison process with a message, so that the detection process can be reduced.

  When noise exceeding the threshold is applied, the display device such as the LED of the monitoring device is turned on. This LED notifies that the noise is constantly applied, and can notify the user that the noise is superimposed even if the communication is normal. Also, the installer can recognize the superposition of noise at the installation stage.

FIG. 7 shows a flow chart of noise detection by Fourier expansion.
First, the transmission signal is sampled (communication 1 bit is divided into 32) (S701). Next, the sampled value is Fourier-expanded to obtain a fundamental wave (first order) component and a harmonic (high order) component (S702). Then, the effective values of the fundamental wave component and the harmonic component are calculated (S703).
At S704, it is determined whether the effective value of the fundamental (first-order) component is 5 V. If the effective value is not 5 V, it is determined that there is no communication waveform (S710).
If the effective value is 5 V, it is determined in the next S 705 whether the effective value of the secondary component is 0.2 V or more. If it is 0.2 V or more, it is determined that "noise" is present, and the LED is turned on. If it is smaller than 0.2V, it is determined in next S706 whether the effective value of the tertiary component is 0.2V or more, and if it is 0.2V or more, it is determined that “noise is present”. If it is smaller than 0.2 V, the presence or absence of noise up to the 15th order component is similarly determined (S707).
In any case of S705 to S707, if it is not determined that there is noise, it is determined that there is no noise (S709).

FIG. 8 shows a block diagram of noise detecting means that does not affect the message.
The noise detection unit 800 that does not affect the communication message includes a sampling unit 801, a Fourier expansion unit 802, an effective value calculation unit 803, a transmission signal detection unit 804, and a noise detection unit 805.
The sampling unit 801 samples the transmission signal shown in FIG. The Fourier expansion unit 802 Fourier-expands the sampled values to obtain a fundamental (first-order) component and a harmonic (high-order) component shown in FIG. 4. The effective value calculation unit 803 calculates effective values of the fundamental wave (first-order) component and the harmonic (high-order) component. The transmission signal detection unit 804 determines whether the effective value of the fundamental wave (first-order) component is 5 V. If the effective value is 5 V, it determines that the transmission signal is present, and detects the transmission signal. When the transmission signal detection unit 804 detects a transmission signal, the noise detection unit 805 determines whether the effective value of each harmonic (high order) component is equal to or greater than a threshold. If the effective value is equal to or greater than the threshold, noise is present judge.
The noise detection means 800 that does not affect the electronic message can be configured by software by the central processing unit (CPU) 203 loading and executing a predetermined program on the memory. Also, each can be configured by hardware.

  According to the present embodiment, since it is configured to detect noise that does not affect the communication message, it can be determined whether there is a small level of noise that does not affect the communication, communication stop when the monitoring system is introduced The cause can be investigated. In addition, noise is detected by determining whether the harmonic component is equal to or greater than a threshold value using Fourier expansion, so that high-speed noise can be accurately detected in a short time.

  The second embodiment is a monitoring device having a noise detection configuration in which a communication telegram message is rewritten in addition to a configuration for detecting noise that does not affect the communication telegram.

  FIG. 9 shows a configuration of noise detection in which a message of a communication message is rewritten. Since the noise is an alternating current, a direct current removal filter 900 is provided which passes only the alternating current value. By passing through the filter 900, only the switching part of the message and the noise signal are passed. Since the value 104 to be rewritten is 0.4 V, this value is used as a threshold. If the signal passed is a normal value, the threshold value is exceeded for each cycle of the telegram speed. The message speed is recognized by the device itself. If there is noise, the threshold is exceeded at locations other than the message speed, so it is possible to detect whether there is noise that rewrites the message only in a cycle exceeding the threshold. The noise detection unit 901 performs this noise detection. When noise is detected, it is input to the central processing unit 203 via the interrupt circuit 902. This detection does not require Fourier expansion, and there are few detection processes. When noise that rewrites data is superimposed, the LED that notifies that there was an error in rewriting the message is turned on. The lighting of this LED does not turn off unless a specific operation is performed on the device. By doing so, it is possible to notify that abnormal noise has been superimposed in the past even if no abnormality is currently occurring. Also, the lighting state of the LED is notified to the upper software through the network. By doing so, it is possible to grasp the noise superposition state of all devices not only by the device but also by upper software. Also, in the upper software, the noise occurrence time can be stored. By comparing the noise occurrence time with the user's equipment operation time, the noise occurrence location can be easily identified.

  According to the present embodiment, in addition to the configuration for detecting the noise that does not affect the communication message, the configuration of the noise detection for rewriting the message of the communication message is provided, so that the processing can be distributed according to the noise level. , Noise can be detected with high accuracy.

  The third embodiment is a monitoring device including a configuration for detecting noise when noise having an abnormal peak value is superimposed in addition to a configuration for detecting noise that does not affect a communication message.

  FIG. 10 shows the configuration of noise detection when noise with an abnormal peak value is superimposed. A protection circuit 204 is provided to prevent the device from failing at an abnormal peak value. When the protection circuit 204 operates, current flows in the protection component. Although all the current values that flowed are protected by letting them escape to ground, the current detection unit 1001 detects signals using the current values that flowed to ground as a trigger. The detected signals are all abnormal signals regardless of the level. When this signal is detected, it is held by the signal holding unit 1002 to turn on the LED of the device. The lighting of this LED does not turn off unless a specific operation is performed on the device. By doing so, it is possible to notify that abnormal noise has been superimposed in the past even if no abnormality is currently occurring. Also, the lighting state of the LED is notified to the upper software through the network. By doing so, it is possible to grasp the noise superposition state of all devices not only by the device but also by upper software. Also, in the upper software, the noise occurrence time can be stored. By comparing the noise occurrence time with the user's equipment operation time, the noise occurrence location can be easily identified.

  According to the present embodiment, in addition to the configuration for detecting noise that does not affect communication telegrams, the configuration for noise detection when noise with an abnormal peak value is superimposed is provided, so processing is distributed according to the noise level. Noise can be detected with high accuracy.

  Although not described in FIG. 10, it is preferable that the configuration of noise detection described in FIG. 9 that rewrites the telegram of the communication telegram is also provided.

100 Steady-state telegram waveform 101 Noise level 102 not affecting communication telegrams Noise level 103 affecting communication telegrams abnormal peak noise level 104 threshold 200 amplification circuit 201 communication circuit portion 202 A / D conversion portion 203 central processing portion 204 protection Circuit 300 Message speed (1 cycle)
301 Division number of sampling 302 Number of times of sampling acquisition 400 Basic wave (1st order) waveform 401 Harmonic (2nd order or later) waveform 500 Distortion waveform 800 Noise detection means 801 which does not affect communication message message Sampling part 802 Fourier expansion part 803 Effective value calculation Unit 804 transmission signal detection unit 805 noise detection unit 900 DC removal filter 901 noise detection unit 902 interrupt circuit 1001 current detection unit 1002 signal holding unit

Claims (8)

A monitoring device that is connected to a network and used for a monitoring system that collects data from each device using host software,
A first noise detecting means for detecting noise that does not affect the communication message ;
The first noise detection means Fourier-expands the transmission signal, detects the transmission signal by the fundamental wave component of the Fourier expansion, and detects the transmission signal, each harmonic component of the Fourier expansion is within the communication message range. monitoring apparatus characterized that you detecting noise by determining whether or exceeds the threshold. In the monitoring device according to claim 1 ,
The first noise detecting means includes
A sampling unit for sampling the transmission signal;
A Fourier expansion unit for Fourier expansion of the sampled signal;
An effective value calculation unit for calculating an effective value of each component that has been subjected to Fourier expansion;
A transmission signal detector for detecting a transmission signal from the effective value of the fundamental wave component;
A noise detection unit that detects noise by determining whether the effective value of the harmonic component exceeds a threshold when a transmission signal is detected;
Having a monitoring device. In the monitoring device according to claim 1 ,
A monitoring apparatus for displaying a noise state detected by the first noise detecting means. In the monitoring device according to claim 1, further,
A second noise detecting means for detecting noise rewritten by the communication message ;
The second noise detecting means includes
A DC removal filter that passes only the AC component of the transmission signal;
A noise detector that detects noise by comparing the passed AC component with a threshold;
Monitoring device according to claim Rukoto to have a. The monitoring device according to claim 4 ,
A monitoring apparatus for displaying a noise state detected by the second noise detecting means. In the monitoring device according to claim 4 , further,
And third noise detection means for detecting momentary peak noise ,
The third noise detecting means includes
A current detection unit that detects a current flowing in the protection circuit;
A signal holding unit that holds a noise signal detected by the current detection unit;
Monitoring device according to claim Rukoto to have a. The monitoring device according to claim 6 ,
A monitoring apparatus characterized by displaying a noise state detected by the third noise detection means. The monitoring device according to any one of claims 1 to 7 .
A monitoring device characterized by notifying a host software of a noise condition.

Images