JP2003009430A - Method and apparatus for remotely monitoring electrical device and method and apparatus for estimating power consumption utilizing the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow estimating individual currents consumed by a plurality of devices based only on a total current. SOLUTION: An apparatus for estimating power consumption is provided with a total current meter 11 for measuring a total load current to an electric supply line 2 of a customer 1, a fast fourier transform apparatus 12 for transforming the total load current into a fundamental wave and higher harmonic waves of the total load current, a temporal difference apparatus 13 for finding changes of the currents of the fundamental wave and the higher harmonic waves in the converted current, an independent component analyzing apparatus 14 for separating the changes of the currents into components estimated as groups of the devices respectively having the same intensity ratio of higher harmonic wave based on independent component analysis, and an apparatus 15 for separating signals per device for estimating an operating status (the change of the current) per device 3 to be monitored based on a waveform of the change of the current per component of the same intensity ratio of higher harmonic wave. The changes of the currents of the fundamental wave and the higher harmonic waves are separated into the components estimated as the groups of the devices respectively having the same intensity ratio of higher harmonic wave based on the independent component analysis. The change of the consumed current per device to be monitored is estimated based on the waveform of the change of the current per the same intensity ratio of higher harmonic wave.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the operating status of individual appliances based on the total current consumption obtained by adding up the current consumption of a plurality of electric appliances used by electric power consumers (users of electricity). The present invention relates to a remote electric device monitoring method and device capable of estimating consumption current, and a power consumption estimation method and device using the same. More specifically, the present invention provides a remote electric device monitoring method and device suitable for estimating the operating status (current consumption) and power consumption of each of a plurality of electric devices by a non-intrusive method, and the consumption of the electric device. The present invention relates to a power estimation method and device.

[0002]

[Technical terms] In this specification, "non-invasive" means that a measurement sensor is installed at one place near the feed line inlet, and a measurement sensor is attached to each branch circuit downstream of the feed line or connected to the circuit. It refers to a state in which a measurement sensor is not attached to each of the electric devices. Further, an inverter device is a device equipped with an inverter and capable of continuously changing the operation of the device from low output to high output. The power consumption of this inverter device continuously changes from a small value to a large value according to the output. Further, the non-inverter device refers to a device which does not have an inverter and whose operation is limited to ON and OFF. The power consumption of the non-inverter device has a limited value corresponding to the on / off operation.

[0003]

2. Description of the Related Art Monitoring the operating status of each device in a factory or a home is essential for promoting efficient use of power devices.
A reliable method to understand the operating status of multiple devices is
It is to collect information by installing a measuring device for operating conditions and a transmitting device for the measured information in each device. However, the installation of the measurement device and the transmission device increases the cost when the number of target devices increases. In addition, the measuring device and the transmission device may not be installed depending on environmental conditions. In particular,
It is difficult for electric power consumers such as general households to install sensors inside the house. Therefore, there is a demand for a monitoring technology that estimates the operating status of a device such as the current consumption of an individual device and its change from the total current consumption of the device without installing such an individual measuring device or information collecting device. There is.

Conventionally, as a monitoring system for non-intrusively estimating the operating state of such an electric device, MI has been used.
T (Massachusetts Institute of Technology; USA)
EPRI (Electric) using the algorithm developed in
Power Research Institute (USA) has been instrumentalized. This monitoring system captures the on / off operation of electric equipment as a stepwise time change of the total power load curve of an electric power consumer, and the electric equipment turned on or off based on the rated power consumption and power factor of the electric equipment. Is specified and the operating state is estimated. Therefore,
It is possible to identify and estimate the operating state of electric devices that perform simple on / off operations.

[0005]

However, recently, inverter equipment such as cooling and heating devices have become widespread in general households, and non-inverter equipment and inverter equipment are used in a mixed state. Is increasing. Since the inverter device controls the output according to the state of the load, the power consumption also changes accordingly. Therefore, the temporal change of the power consumption is not necessarily step-like, and it may fluctuate gently or irregularly.

Therefore, in a situation in which inverter equipment and non-inverter equipment are mixed, the above-mentioned conventional EPR is used.
Depending on the monitoring system developed by I, it is difficult not only to estimate the power consumption of each individual electric device but also to estimate the operating state of the electric device.

In order to solve this problem, the applicant of the present application changed the operating state of each electric device such as on / off as a technique for estimating the operating condition of a plurality of electric devices including an inverter device and a non-inverter device. Based on the harmonics of the total current, we proposed a system that estimates the operating status of each device in advance from the effective current and phase of each harmonic of the total current using an estimation algorithm such as a large margin classifier ( Japanese Patent Application No. 2000-111271).

However, in this method, it is necessary to prepare in advance a large amount of harmonic data of the total current when the assumed monitored device is operated in each operating state, which is an unknown unknown monitored device. Can not be targeted. Therefore, if the measured total current includes an unknown monitored device, the unknown electric device or electric device group should be operating and its power consumption. Since it is only possible to make a comprehensive estimate, it becomes virtually difficult to monitor the operating status of the electric devices as the number of unknown electric devices increases. Therefore, if there are various types of monitoring target devices that are unique to each electric appliance or power consumer that is not assumed in advance, or if the target devices are replaced, the It is necessary to prepare the data in advance and recreate the operation state determination system by learning the estimation algorithm.

The present invention is a remote electric device which enables non-intrusive estimation of individual operating states of a plurality of electric devices used by electric power consumers in a situation where non-inverter devices and inverter devices are mixed. An object of the present invention is to provide a monitoring method and device, and a power consumption estimation method and device using the same. Furthermore, the present invention makes it possible to estimate individual consumption currents of a plurality of devices from only total currents without having to collect data in advance to configure an operation state determination system for each device to be monitored. It is intended to provide a monitoring system.

[0010]

In order to achieve such an object, the present inventor has found that a change in the consumption current of an electric device installed in a power consumer is independent of a change in the consumption current of another electric device. Focusing on the point that changes to, by applying the signal separation technology by the estimation means such as independent component analysis,
From the measurement result of the total load current at one point of the power supply line of the power consumer, for example, the power supply line inlet to which the monitored device is connected,
We considered estimating individual operating conditions and current consumption of each connected electric device.

That is, according to the first aspect of the invention, in a remote monitoring system for estimating individual current consumption of a plurality of electric appliances used by an electric power consumer, a measuring sensor installed on a power supply line of the electric power consumer is used. With the effective current of each harmonic of the total current obtained as input, based on the signal separation algorithm, the consumption current for each device of the multiple electric devices used by the power consumer is provided as an estimation means. There is. That is, the remote electric device monitoring method of the present invention is a remote electric device monitoring method for estimating the operating status of a plurality of electric devices used by a power consumer, and measures the total load current from the power supply line of the power consumer. Then, the total load current is converted into a current for each of the fundamental wave and the harmonics, and current change data is created by taking a time difference between the currents for the fundamental wave and the harmonics. The current change of each component is estimated by independent component analysis as a component estimated as a device group having the same harmonic intensity ratio, and the current change waveform for each component of the same harmonic intensity ratio is used to calculate the consumption of the monitored device by device. It is characterized by estimating a current change.

Further, this remote electric equipment monitoring method is realized by, for example, a remote electric equipment monitoring apparatus for estimating the operating condition of a plurality of electric equipment used by an electric power consumer according to the invention of claim 4. This remote electrical equipment monitoring device includes a total current sensor that measures a total load current from a power supply line of an electric power consumer, and a frequency component conversion device that converts the total load current into a fundamental wave and a harmonic current of the total load current. , A time difference device that obtains a current change by obtaining a time difference between currents of fundamental waves and harmonics of total load current, and the current change is estimated as a device group having the same harmonic intensity ratio by independent component analysis. An independent component analyzer that separates each component, and a device-separated device that estimates the operating status (current change) of the monitored device by device from the waveform of the current change for each component of the same harmonic intensity ratio are provided. ing.

A device incorporating the inverter circuit and the rectifier circuit emits a unique harmonic. The harmonic effective current itself of each device at each time point may behave in a similar manner as a whole, for example, between the devices having the same start time, the harmonic currents increase with time. However, the change in the effective current of each harmonic (time derivative) is highly statistically independent even if the devices activated at the same time do not operate a special synchronization mechanism. Called statistical independence). Further, the intensity ratio of the effective current (change) of each harmonic of each device is almost stable with time. In addition, the intensity ratio is unique to an individual device or a device group having similar harmonic characteristics (for example, a device group that does not emit harmonics). Therefore, it is possible to estimate the current change of each device by the weighted sum of the effective current change of each harmonic of the total current (this is called linearity of individual device current change). Note that the harmonic ratio unique to each device group and the weight of each harmonic of the total current have an inverse matrix relationship.

On the basis of the linearity and independence of the individual device current change as described above, the weighting that maximizes the statistical independence between the individual current changes of each device estimated from the harmonics of the total current is estimated. As a result, it is possible to estimate the current change of each device group having a unique harmonic intensity ratio.

Further, in the invention according to claim 1 or 4, when the current change of the device group having a unique harmonic intensity ratio is separated into one and is separated, the operating characteristic relating to the intensity level of the current change of the individual device. Based on the model, it is possible to separate individual current changes for each device with the same harmonic intensity ratio. That is, according to the invention of claim 2, in the remote electric device monitoring method according to claim 1, among the current changes for each component of the same harmonic intensity ratio, the component of the device exhibiting the same harmonic intensity ratio is monitored. The current change of each individual device is estimated by further separating based on the information on the current change intensity. Further, the invention according to claim 5 is the remote electric device monitoring device according to claim 4, wherein among the current changes for each component of the same harmonic intensity ratio, the component of the device exhibiting the same harmonic intensity ratio is monitored. The current change of each individual device is estimated by further separating based on the information on the current change intensity.

Further, each component estimated as a device group showing the same harmonic intensity ratio output in the remote electric device monitoring method according to claim 1 or 2 and the remote electric device monitoring device according to claim 4 or 5. Since the change in the current separated by 2 is device-specific, the power consumption can be estimated by obtaining the current consumption from these current changes as in the invention according to claims 3 to 6.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION The structure of the present invention will be described below in detail based on an example of an embodiment shown in the drawings.

FIG. 1 shows an outline of one embodiment of the remote electric equipment monitoring method of the present invention. This remote electric equipment monitoring method is for estimating the operating status of a plurality of electric equipment used by an electric power consumer, measuring the total load current from the power supply line of the electric power consumer, and measuring the total load current. While converting to the current for each of the fundamental wave and harmonics, current change data is created by taking the time difference of the current for each of the fundamental wave and harmonics,
These fundamental and harmonic current changes are separated by the independent component analysis for each component estimated as a group of devices with the same harmonic intensity ratio, and the current change waveform for each component with the same harmonic intensity ratio is monitored. The current change of each device can be output to estimate the operating state of each device.

Here, the separate estimation of the current consumption of each individual device from only the total current consumption is more generally performed by adding signals from a plurality of signal sources and measuring them as one signal at one location. It is considered to be one of the methods of "signal separation tracking" by separating the signal of each signal source from the measured signal and tracking the change during the measurement. There, the "signal source" is the device and the "signal measured at one place".
Corresponds to the total current, and the "signal of each signal source" corresponds to the "current (current change) of each device".

Therefore, first, the total current, which is the sum of the power consumption of each device, is measured at a high sample rate, and the measured total current is subjected to Fourier transform to obtain each frequency (the fundamental wave of the total load current and its harmonics). ) Is converted into a time series of the effective current value. Then, signal separation based on independence is performed. The current of each device changes autonomously unique to each device. In other words, it fluctuates independently of current changes of other devices. Further, the operation of each device changes in a plurality of frequency bands at the same time, but the frequency band in which the change is strong varies depending on the device.
Therefore, by analyzing independent current changes that occur simultaneously in a plurality of frequency bands, it is possible in principle to restore the current consumption of each device. Due to this property, it is possible to restore the current consumption of each device without a device-specific operation model.

However, when the rated current or the like of the monitored device is known in advance, or when the total current is continuously monitored and the device current is appropriately restored, the operating characteristic model peculiar to the device is obtained. Can be configured. The operating characteristic model can be represented by the probability distribution of the value that can be taken by the amount of change in the effective current I (t) of the device (I (t + 1) -I (t); hereinafter called current change). .

In the present invention, when an operating characteristic model peculiar to a device is available, it is used to improve the estimation accuracy. When using the operation characteristic model of each device, the separation accuracy is improved by using the operation characteristic model that matches the behavior characteristic of the operating device. Therefore, in order to continuously track the operation of the device, it is necessary to select the device operation characteristic model used in step 2 of FIG. In this embodiment, under the assumption that the operations of many devices do not change at the same time, the device operation characteristic model to be used is selected, and the process returns to step 2.

The above is an outline of the signal tracking approach.
The following outlines the algorithm used in each step.

[Signal Separation Based on Independence] The signal separation method is an extension of the standard independent component analysis method so that the separation considering the operation characteristics of the device can be performed.

The independent component analysis method is a method of estimating the signal waveform of the signal source from the observed signal, such as separating the sound emitted from each sound source from the simultaneous recording at a plurality of points. In applying it, the following two assumptions of "linearity" and "independence" must be satisfied.

"Linearity Hypothesis 1" (1) The kth harmonic current variation dIi (t) of the total current is expressed as the sum of the kth harmonic current variation dSk, i (t) of each device. . dIk (t) = Σj dIk, j (t) (strictly established when the current value and current change amount are displayed in complex numbers) (2) kth harmonic current change amount dSk, i (t of device i ) Is the device i
Normalized harmonic current variation dSi * (t) constant times Ak, i dSk, i (t) = Ak, i ・ dS * i (t) That is, dIk (t) = Σj Ak, i ・ dS * In the i (t) matrix representation, dI = A · dS * where dI = (dIk (t)), A = (Ak, i), dS * = (dS * i (t)). That is, the "linearity hypothesis 1" has the same value as the following "linearity hypothesis 2" when the inverse matrix of A exists.

"Linearity Hypothesis 2" (1) The normalized harmonic current variation dS * i (t) of device i is
Constant multiple of each harmonic current variation dIk (t) of total current Bk, i
Is the sum of dSi * (t) = Σk Bi, k ・ dIk (t) In the matrix representation, dS * ＝ B ・ dI where dI = (dIk (t)), B = (B, i, k), dS * = ( dS * i (t)) (2) The kth harmonic current variation dSk, i (t) of equipment i is a constant multiple of the normalized harmonic current variation dSi * (t) of equipment i Ak,
i dSk, i (t) = Ak, i * dS * i (t) where Ak, i is the (k, i) component of the inverse matrix of B. The matrix A is called a “mixing matrix” and the matrix B is called a “separation matrix”.

Under the linearity hypothesis, if the separation matrix B is known, the specified harmonic current change amount of the device can be estimated from the harmonic current change amount of the total current. However, frequency strength A
Devices with the same ratio of k and i cannot be distinguished because there is no inverse matrix. Therefore, strictly speaking, a device group having the same frequency intensity ratio is virtually treated as one device.

"Independence" The current waveform of each signal source device is independent of each other. That is, the independence of the amount of change in the current of a device is derived from the fact that the operation of each device (switch on / off or change of operation mode) is performed independently of the operation of other devices.

Mathematically, the current change amount dS ^* _{i of the} devices i and j
The independence of (t), dS ^* _j (t) is as follows: Probability that the current change amount dS ^* _i (t) = x of device i at time t Pr (dS ^* _i = x), Probability P of current change amount dS ^* _j (t) = y of _j
r (dS ^* _j = y), ・ Current change amount of device i at time t dS ^* _i (t) = x and device
When Pr (dS ^* _i = x, dS ^* _j = y) is the probability that the current change amount of _j is dS ^* _j (t) = y, Pr (dS ^* _i = x, dS ^* j = y) = Pr (dS ^* _i = x) Pr
(dS ^* _j = y) holds. At this time, the estimated signal of each signal source is called an independent component.

This standard independent component analysis has no source-specific characteristic model, and only assumes that the observed signal is a linear superposition of signals from independent sources, The signal source S is estimated by obtaining the separation matrix B that maximizes the index (independence index) indicating the independence of.
That is, according to the present invention, even if the separation matrix B in the linearity hypothesis 2 is unknown, by focusing on the independence of the current change amount of the device and applying the independent component analysis method, it is possible to estimate the harmonics of the device. The separation matrix B that maximizes the independence index between the current amounts is obtained. Further, according to the linearity hypothesis 2 (2), a separation matrix, which is a matrix of frequency intensity,
By obtaining the inverse matrix A of B, the fundamental current variation for each device group with the same frequency intensity ratio can be estimated.

As a result, it is possible to estimate the change in the basic current amount for each device group having the same frequency intensity ratio from the change in the harmonic current of the total current. [Improvement of accuracy by device operating characteristic model] The standard independent component analysis described above does not have a characteristic model peculiar to a signal source, and the characteristic of the signal is completely unknown, so signal sources with similar characteristics may interfere. There is a case. Therefore, if the ratio of the current change amount of each device and the rating of the current change of the device are known, the basic current change amount obtained by the standard independent component analysis described above is used as the current of each device based on that information. By separating into changes, it is possible to estimate the amount of change in current for each device.

In the proposed method, an operating characteristic model peculiar to each signal source (device) is given as a probability distribution Pr _i (s) of possible signal levels S _i (t), and this is used to improve separation accuracy. .

The current variation of the current consumption of the device tends to be extremely concentrated around a certain value. For example, in an on / off type device, it is either 0 or the rated value. A similar tendency occurs in inverter equipment.

Two measures are taken to improve the separation accuracy. First
First, the independence index maximized by independent component analysis is optimized based on the behavioral characteristic model. By maximizing the obtained independence index, signal separation suitable for device characteristics is performed.

Second, the device-specific operating characteristic model is used for the separated independent components to separate the signals of the corresponding devices. This is effective for signal separation when interference occurs because the operation characteristics are similar except for the signal level.

FIG. 2 shows an embodiment of an apparatus for implementing the remote electric equipment monitoring method of the present invention. This remote electric device monitoring device is basically a total current sensor 11 that measures a total load current from a power supply line 2 of an electric power consumer 1, and a total current and a fundamental wave and a harmonic current of the total load current. A frequency component conversion device 12 for converting into a frequency component, a time difference device 13 for obtaining a current change by taking a time difference between the fundamental wave of the total load current and the current of each harmonic, and the same harmonic by the independent component analysis of the current change. The independent component analysis device 14 that separates each component estimated as a device group having an intensity ratio, and the movement status (current change) of each device of the monitored device 3 from the waveform of the current change for each same harmonic intensity ratio component. And a device-specific signal separation device 15 for estimating.

The measuring sensor 11 is installed only at one place near the service entrance of the service line 2 of the power consumer 1 in order to make the system non-invasive. The measurement sensor 11 obtains an electric current and is composed of, for example, a current transformer. In the present embodiment, the case where it is carried out for a general electric power consumer in Japan that uses a single-phase three-wire service wire is taken as an example, and therefore, for a current transformer for an A-phase instrument and for a B-phase instrument. It is composed of a current transformer. For example, if a through-type current transformer is used for the A-phase and B-phase measuring instruments, the measuring current transformer measures the current flowing in the A-phase on the primary side and measures the current flowing from the secondary side to the A-phase. Output a current similar to the current of
The current transformer for measuring instrument measures the current flowing in the B-phase on the primary side and outputs a current similar to the B-phase current from the secondary side. These currents are input to the fast Fourier transform device 12, which is a frequency component converter.

The fast Fourier transform device 12 extracts data on the fundamental wave of the total load current and the current for each harmonic from the total load current detected by the measurement sensor 11. Although not specifically shown, for example, analog /
It is composed of a digital (A / D) converter and a fast Fourier transformer, and converts the A-phase and B-phase currents I _A and I _B input from the measurement sensor 11 into digital data by the A / D converter. Then, the fast Fourier transformer is used to obtain the harmonic current data IA _(1-13) and _{IB (1-13)} . Here, the current data I _A1 and I _B1 respectively indicate the current of the fundamental wave of the total load current, and the current data I _A1 and I _{B1
A (2-13)} and _{IB (2-13)} are subscripts (2-13)
Denote the harmonic orders, that is, the currents of the harmonics representing the 2nd to 13th orders, respectively, and the frequency of the harmonics is expressed by multiplying the fundamental frequency of the AC power supplied to the feeder line by the numerical value of the order. For example, if the fundamental frequency is 50Hz,
The third harmonic current refers to a current component having only a frequency component of 150 Hz. In general, odd-numbered harmonics are predominant, and even-numbered harmonics are small, so that the fundamental wave and odd-numbered harmonic data are input to the time difference device 13 here.

The time difference device 13 includes a Fourier transformer 12
The time difference between the fundamental wave and the harmonic wave of the total load current converted in step S1 is taken to obtain and output the amount of change in current and current.

The independent component analysis device 14 separates the current change into each component estimated as a device group having the same harmonic intensity ratio by the independent component analysis, and executes the algorithm of FIGS. 3 to 5. Independent component analysis is performed by a computer. The device-based signal separation device 15 estimates the movable condition (current change) of the monitored device 3 for each device from the waveform of the current change for each identical harmonic intensity ratio component, and executes the algorithm of FIG. 6. Independent component analysis is performed by a computer.

3 to 6 show an example of a standard algorithm and an algorithm using an operating characteristic model for estimating the fundamental wave current variation for each device group having the same frequency intensity ratio.

First, the current value dIk (t) of odd-numbered harmonics from the first (fundamental wave) to the thirteenth (harmonic) of the total current =
Input Ik (t) -Ik (t-1) (step S
1).

Then _, the initial value of the separation matrix B = (B _{i, k} ) is set (step S2). The settings are as follows. U is a random n-th rotation matrix, and the eigenvalue decomposition of the covariance matrix of the first to nth harmonic current variation dIk (t) of the total current is performed.
When V ・ D ・ Vt, B = U ・ D-1 / 2 ・ V. The covariance matrix is the covariance Σ _t (mean of dI _k (t) -dI _k ) ・ (dI _l (t) -dI _l
Is the average) / T.

Next, the normalized current change amount dS for each device group
* _i (t) = Σ _i B _{i, k} · dI _k (t) is estimated (step S
3).

Then, the separation matrix B is improved by the natural gradient method or the fixed point independent component analysis so that the independence index IND (dS ^* ) of the standardized current change amount of each device group is improved (step S4). ). Here, various independent component analysis methods can be used as the independent component analysis method used in the present invention. In general, Σ _i (Σ _t dS ^* _i (t) ⁴ / T-3) ² : each as the independence index IND (dS ^* ) that is maximum when the current variation dS ^* i, dS ^* j is independent. The sum of squares of the instrument's Kurtosis. The sum of squares of Σ _i (Σ _t log (cosh (πdS ^* _i )) / T) ² is used.

Here, it is effective to use the sum of Kurtosis as an index in order to improve the estimation accuracy of a large amount of current change of the device such as on / off operation. Further, according to the present invention, when the current record of the monitored device is already present, the independence IND (dS) with high separation accuracy for these waveforms is set as shown in FIG.
The accuracy of separation of monitored devices is increased by using the index determined by the algorithm shown in.

The algorithm for maximizing the independence index is as follows: Fixed point independent component analysis for finding the fixed point of IND (dS)
(Fixed Point ICA) ・ High speed when IND (dS) is Kurtosis sum of squares.
There is a hill climbing method (especially suitable for online update) in the natural gradient direction that is an improvement of the hill climbing method of JADE algorithm / IND (dS).

In the present invention, when the independence index IND (dS) is obtained by the algorithm X, the update by the natural gradient method (separation matrix update algorithm of FIG. 5) is used.

Next, if the number of updates of the separation matrix B is equal to or greater than a set value or the amount of update is equal to or greater than a fixed value, the process jumps to step 4, otherwise, jumps to step 6 (step S).
5).

Next, the estimation of the mixing matrix A is made an inverse matrix of A = B (step S6). A _{k, i} is the harmonic intensity of each device group.

Then, the fundamental wave current variation dS _{i of} each device group
The estimation of (t), that is, dS _i (t) = A _{1, i} · dS ^* _i (t) is executed (step S7).

Then, the fundamental current variation d of each device group i
S _i (t), the fundamental wave and each harmonic intensity A _{k, i} are output (step S8).

The estimation of the device-specific current change amount from the device group basic current change amount obtained in step 8 described above can be performed by, for example, referring to FIG.
The algorithm is used. That is, the amount of change in current by device group dSi (t) k-th frequency intensity of device group Aki Frequency intensity Ckj of target device j during operation (In the case of on-off type device, Ckj of the fundamental wave is the rated current value or high Next, input Ckj = 0) (step 9), and perform initial setting dS '(t) = 0, t = 1..T for estimating the current change amount of device j (step 1).
0).

Next, the coincidence of the frequency intensity ratios is determined (step 11). This is the frequency intensity vector {Aki, k = 1..n} of device group i and the frequency intensity vector {Ckj, of device j.
Cosine of angle k = 1, .. n} (cos) = | Σ _k A _ki・ C _kj | / √ ((Σ _k
If A _ki 2) (Σ _k C _kj 2)) is less than or equal to the constant value 1−ε ₀ , the process ends as nonconformity.

Then, the coincidence of the current change amounts is judged (step 12). This is because at time t when dSi (t) of device group i is within the fundamental frequency intensity C1j ± ε of device j, it is determined that the current change is due to device j dS'j (t) = dSi ( t) and dSi (t) = 0.

Then, the current change dS'j (t), t = of the device j
1. Current change dSi that cannot be explained by the operation of T. and device j
(t), t = 1..T is output (step 13).

The optimum independent index is estimated according to the algorithm of FIG. 4, and the separation matrix B in step 4 is updated according to the separation matrix updating algorithm when the optimum independence index G of FIG. 5 is used. .

According to the remote electric equipment monitoring apparatus configured as described above, unknown measurement data (currents I _A , I _A) from the measurement sensor 11 installed near the power feed line entrance is provided.
_B ) is taken out from the fast Fourier transform device 12, separated by the time difference device 13, the independent component analysis device 14 and the device-specific signal separation device 15 into current changes for each electric device group having the same frequency characteristic, that is, the same ratio, and output. To be done. Therefore, from the waveform of this current change, it is possible to estimate the individual operating status and the consumed current (and thus the power) of the electric device 3 (non-inverter device, inverter device).

The following experiments were conducted to confirm the usefulness of the present invention. It is assumed that an ammeter is connected to the power line at the service entrance and the total current used is measured in seconds. The effectiveness of the proposed method is verified by estimating the used current per second of each device, which is a breakdown of the measured signal.
For this purpose, data on the current used and the total current of each device every second are required. The data includes an ammeter for measuring a current used for each electric device as a load, an ammeter for measuring a total load current, and a fast Fourier transform device for converting the total load current into a fundamental wave and a harmonic. It was obtained by a measuring device (not shown). Here, the measurement device includes a sine wave power supply device (50 Hz, 100 V, 2 KVA) and a switch that switches the operating state for each individual load, and the current of the inverter device (for example, in the case of an inverter air conditioner, the indoor set temperature or the set wind speed). Can be changed to change the current) or the current of the non-inverter device (for example, the current can be changed by increasing or decreasing the number of incandescent lamps to be lit). We are trying to obtain various combinations of load usage.

The current value per second of each device is measured by using an ammeter connected to each device. The measured value of each ammeter is
It is the effective current value of the fundamental wave (50Hz) component. on the other hand,
The total current is not strictly the total current per second. 1 per second
The total current is measured only for 5 cycles of the fundamental wave (5/50 seconds), and the fundamental wave (50Hz) to the 13th order (650Hz) of the high frequency component of the odd order obtained by the fast Fourier transform (FFT) is measured. The effective current value, effective voltage, and phase angle between voltage and current are recorded. In the analysis, the voltage and phase angle are not used, and only the effective current is focused.

Since the measuring methods are different, the total current value does not always match the total current value of the ammeters of the equipment. The average error of the total current value and the total current value of each device is -0.1 A, standard deviation.
It was 0.05A. The rate of error of -0.1 ± 0.4 A or more is 0.
It was less than 01%.

In the experiment, 15 devices shown in Table 1 were connected and measured.

[Table 1] Among them, for 8 devices 256 changed the switch on / off (Table 1 trailing column reference) (= 2 ⁸⁾ cases, it is carried out measuring about 5 minutes (900 seconds). Of the seven other models, two devices, a pot and a refrigerator, entered the heat retention / cooling mode autonomously by automatic operation, but the five devices such as a microwave oven were connected in the off state, and therefore consume almost no current. In addition,
There are 5 incandescent lamps of the same type, but in the case of turning this on, one, two, ..., 5 were turned on sequentially as time passed. Similar operations are performed for fluorescent lamps and inverter fluorescent lamps. [Independence hypothesis of electric equipment waveform characteristics and current change value] In order to perform signal separation based on the assumption of independence, it is necessary for each waveform resulting from the separation to be independent of each other, that is, to change without any association. There is. However, this hypothesis does not always hold for the current waveform of electrical equipment.

As an example, FIG. 7 shows a case 197 (11000101) in which two air conditioners (A1, A2), an inverter type fluorescent lamp (F1), and one TV (T3) are turned on. The upper part of FIG. 7 is a graph of the effective total current value of each device for 5 minutes (900 seconds).

It can be seen that the air conditioners A1 and A2 are changing in a similar manner. This is the result of starting both air conditioners at about the same time. Therefore, when the components are separated based on the independence hypothesis, the component of the common behavior of both air conditioners and the difference component become the independent components. Since it is often found that multiple devices are switched at almost the same time, it is difficult to directly assume independence between electric current values of electric devices.

On the other hand, the lower part of FIG. 7 shows the amount of change in the effective current value of each device at each time (difference in current value between time t and time t + 1, hereinafter, current change (amount)). It can be seen that there is little similarity in the time change of the current change amount of each device and the independence is high. Even if the overall behavior of the equipment is similar,
Unless a special synchronization mechanism is provided, the timing at which each device enters a specific operation mode, such as when the air conditioner enters a cooling operation (rising part of the effective current value), is autonomously controlled by each device. Therefore, it can be inferred that the independence of the current change is high.

The independence of the current change of each device reflects the autonomy of the operation of each device and can be expected to be a generally valid and valid hypothesis. The separation matrix obtained from the current change can be used as it is as the separation matrix for the effective current.

In order to verify the independence hypothesis, the standard independent component analysis algorithm JADE for the effective current of the operating equipment is used.
(JF Cardoso et. Al .: “Blind beamforcing for n
on-Gaussian signals ”, IEE Proceedings -F, 140 (6):
362-370, 1993.) and a case where the same algorithm is applied to the current change of the operating equipment are compared for all 256 cases. The result is shown in FIG. In this experiment, the effective current and current change of each operating device are normalized to variance 1 and given as an input signal.
Ideally, the input signal itself is an independent component. That is, the separation matrix should be the identity matrix. FIG. 8 shows the deviation between the obtained separation matrix and the unit matrix (difference between the minimum absolute value of the diagonal components and 1). 0 means that the separation and reproduction are complete, and the closer to 1, the more difficult the separation is and the more the interference occurs.

In the case of a change in current, the degree of interference is 0.57 at maximum, and in 217 cases out of 256 cases, the degree of interference is 0.05 or less, which realizes highly accurate separation, while the effective current According to the above, interference of 0.9 or more occurs in more than half of 132 cases.

It was confirmed that the independence hypothesis of the current change of each device is effective for separating the effective current of the device.

[Equipment-Specific Effective Current and Linearity Hypothesis] Next, the relationship between the effective current of each frequency of the measured total current and the effective current of each device to be estimated will be examined.

FIG. 9 is a time change of each frequency component and current change (amount) of the air conditioners (A1, A2) in the case 197. Focusing on the time point when the air conditioners A1 and A2 are switched from the cooling mode to the steady operation (the dotted line portion around the time 150 seconds in FIG. 9), the third-order wave is generated in the same way, while the fifth wave is generated. It can be observed that only the air conditioner A2 has a strong influence in the opposite direction on the next wave and on the 13th wave. As described above, the change in the operation mode of each device makes a different contribution to the current change value of each frequency component. Therefore, there is a possibility that the current change of each device can be estimated from the current change of each frequency component. The problem here is the assumption of "linearity" required for independent component analysis, that is, whether the property that the waveform of each signal source is expressed by the linear combination of the waveforms of the observed signals is satisfied. Is.

Since the sum of the currents from each device is the total current, the present hypothesis is that the ratio of the effective current (Ij) at each frequency of the device i is unique to the device and is constant regardless of the output level and time. That is, it means that (I ₁ (t), I ₃ (t), ..., I ₁₃ (t)) = (a ₁ , ..., A ₁₃ ) · I ₀ (t). This hypothesis that "the ratio of the effective current at each frequency of the device does not change over time and has a value unique to the device" does not hold exactly like the independence hypothesis.
However, if it holds approximately, it is possible to estimate the current change of the device current from the effective current (current change) of each frequency component of the total current.

In the experiment, the effective current at the fundamental frequency of each device alone was measured, but the effective current for each frequency was not measured, so it is difficult to directly verify the above hypothesis.
Here, by verifying the prediction accuracy of the effective current at the fundamental frequency of the device when the above-mentioned hypothesis is assumed to be established using the effective current for each frequency of the total current, the status of establishment of the above-mentioned hypothesis is verified.

FIG. 10 shows the result of obtaining the separation matrix B from the effective current value of each frequency component and the effective current value of the device by minimizing the mean square of the error (least square method). Figure 10
The upper row is the approximate effective current value of each device obtained by the separation matrix B from the effective current value at each frequency. The lower part of FIG. 10 shows the difference value of the approximate effective current value of each device (hereinafter, approximate difference current). Compared with Fig. 7, the air conditioner (A1, A2)
For example, the approximate effective current is close to the actual effective current, and it can be seen that the linearity is approximately established. On the other hand, for the inverter fluorescent lamp (F1) and the pot (J), the approximation waveforms differ greatly from the actual effective current, and the linearity assumption does not give a good approximation. However, regarding the total value (F1 + J) of both, the approximation accuracy is improved compared to the case of estimating each independently, and it is considered that linearity is approximately established. This is because the frequency characteristics of the inverter fluorescent lamp and the pot are similar (the ratio of the effective current at each frequency (a1 /
a1, .., a13 / a1)).

Therefore, the assumption of linearity is not always established for individual devices, but is approximately established if a specific device group having similar frequency characteristics is regarded as one device. In order to further separate devices that show similar frequency characteristics such as inverter fluorescent lamps and pots, a hypothesis regarding the fundamental wave intensity level I ₁ = a1 · I ₀ of each device, that is, the value of the current change peculiar to the device (Example: Pot J: 0.6A)
Need to be introduced. (As described below, this includes
A device operating characteristic model may be used. )

In order to study the device group based on the frequency characteristics, the correlation coefficient of the approximate difference current between the operating devices in each case and the approximation error of the approximate difference current (approximate difference current-actual current change) is obtained. FIG. 12 shows the average of the absolute values. As discussed in the previous section, the actual current changes themselves are highly independent and do not show a strong correlation. A large correlation coefficient of the approximate difference current means that the devices have similar frequency intensity ratios, and it is predicted that the current changes of both devices are likely to be separated as one independent signal component.
On the other hand, the correlation coefficient of the approximate current change error indicates the similarity in behavior of the part of the current change that cannot be captured linearly. Therefore, although the devices with high correlation are separated as separate independent signal components, it is predicted that there is a tendency for interference between them to occur at a specific time.

From FIG. 12, the device group {incandescent lamp (4), fluorescent lamp
(5), the inverter type fluorescent lamp (6), and the pot (9)} have a high correlation coefficient (linear portion) of the approximate difference current, and can be predicted to be a device group that is difficult to distinguish by independent component analysis. The device group {air conditioner A2 (2), fan (3), television 1 (7)} also has a high degree of similarity in linearity. Air conditioners (A1 (1), 2 (2))
It can be seen that there is a non-linear similarity between the fan (3) and all equipment except the television 1 (7). In particular, air conditioners are strongly related to each other, and partial interference is likely to occur.

[Estimation of Current Change by Device] [Experiment of Application of Standard Algorithm] It was confirmed that independence on current change by device and linearity with respect to an appropriate device group can be assumed. In this section, we describe an experiment to estimate the current change per second of an unknown connected device based on the above assumption of independence and linearity from only the effective current value per second of each frequency component.

In the experiment, first, a standard independent component analysis method was used.
JADE was applied to estimate the current change of individual equipment. In estimating device operation, spike-like current changes such as on / off, which can be called an exceptional value, are important.
Efficiently maximize the sum of squares of the independence index Kurtosis, which is strongly influenced by the exceptional value. For this reason, the JADE algorithm was chosen.

In FIG. 12, the waveform of the signal source (current change of the fundamental frequency) in the case 197 is estimated by inputting 7 frequencies (1st to 7th stages in FIG. 9) from the fundamental wave to the 13th order (odd order). The results are shown. current change X _j of j-th harmonic, mixing matrix A,
When the separation matrix is B, the current change of the fundamental wave of the i-th estimated component is
S _i = Σ _j A (1, i) · B (i, j) · X _j 1st component (maximum amplitude 3A), 2nd component (2.8A), 3rd component (0.8A), 4th component
(0.4A) is the main component. FIG. 13 shows the waveforms of the first to third components and the corresponding operating equipment (current changes of the fundamental wave).

The first component reproduces the air conditioner A1 relatively well, and the second component reproduces the air conditioner A2 relatively well. However, looking closely, the first component is partially interfering with the component when the air conditioner A2 is started (in the vicinity of 300 seconds and 600 seconds). In addition,
When the mixing matrix A is examined, the frequency intensity of the first and second components is mainly the fundamental wave, the third order wave, and the fifth order wave, and the first component is about 10: 6: 2.
In addition, the second component has an intensity ratio of 10: 5: -1.

The third component does not correspond to a single device. Spike components corresponding to the lighting of the inverter fluorescent lamp (4 times, about 0.2A), and the start / stop of the pot (5 times, about ± 0.7A), and part of the air conditioner A2 (small spikes when stopped ( 3 times, -0.5A) and a part of the start-up) are mixed. Although not shown, from the experimental results, the fourth component (spike twice, 0.4A) is twice the air conditioner A2, three times.
It corresponds to the peak at the start of the second time. The fifth component has a strong correlation with TV.

In the third component, the intensity ratio of the fundamental wave and the tertiary wave is 10: 1, and the intensities of other frequencies can be ignored. That is,
It is a device component with almost no harmonics. The fact that the pot and the inverter fluorescent lamp are included in the third component is consistent with the above-mentioned discussion of the similarity between the inverter fluorescent lamp F1 and the pot J. It is considered that the air conditioner temporarily interferes with the third component because the harmonics are temporarily weakened by the inverter control of the air conditioner. As for the air conditioner A2, especially the waveform at startup interferes with other components than the corresponding second component,
Also, some spikes are separated as the fourth component. This is probably due to the non-linearity of the air conditioner inverter control. This point is also consistent with the above discussion (Fig. 11).
(B)).

From the above experiment, the same frequency intensity characteristic and
There is some interference due to linearity, but the odd current order of the total current
From the effective current value of the wave number component, the operating status of each major device
It has been shown that can be almost reproduced. [Accuracy improvement by device-specific model] General independent component analysis
Then, assuming that the characteristics of the equipment to be separated are completely unknown,
Signal source s_iThe same independence index H (s) regardless of the nature of
Assuming that sum Σ_iH (s _i) Is minimized
Determine the separation matrix.

However, in the signal separation / tracking problem in which steady monitoring is performed, particularly in the equipment operation signal separation, the type of the connected equipment can be assumed to some extent in advance, and the electrical equipment characteristics can be grasped in advance from the rating and the like. is there. Even if it is unknown in advance, it may be specified as a signal source having the same characteristics during steady progress of signal separation. In such cases,
Separation accuracy can be improved by utilizing the operating characteristics peculiar to the device.

In the proposed method, the device operating characteristic is given by the standard current change value of the current change (fundamental wave) of the existing device.
For example, in the ON / OFF type device, the current change value during the ON operation and the current change value during the OFF operation take a substantially constant value.

In the proposed method, when these standard distributions Qi (x) are given, an appropriate independence index H (s is obtained by independent component analysis based on signal source characteristics, for example, the independent component analysis including the algorithm of FIG. ) = Σ _i H (s _i ) is found and maximized.

FIG. 15 shows an optimal independence index (main factor) obtained when a standard probability distribution of an air conditioner or the like is given and an independence index used in a typical method. It is thought that the reason why JADE shows good separation performance is that the independence index (x ⁴ ) used in JADE is a better approximation of the optimal independence index than other independence indexes.

The proposed algorithm further eliminates interference caused in the separated independent components based on the given device operating characteristic model.

In the example of case 197, some interference of the inverter fluorescent lamp (F1), the pot (J) and the air conditioner (A2) can be seen in the third component. Characteristic model Qi (s) (Fluorescent lamp: Mixing distribution with two peaks at steady state 0A and lighting 0.2A, pot: steady state 0A and start / stop approximately ± 0.65A
Mixed distribution with a peak at) and

FIG. 16 shows a change in the device current finally estimated by using the device-specific model and the extended independence analysis algorithm. It can be seen that the incorporation of the operation model peculiar to the device has improved the separation accuracy of the device with particularly few harmonics.

From the above, it was shown that the proposed method, which pays attention to the independence of the proposed device operation, is effective for the device operation separation / tracking. Furthermore, by applying the proposed method to the data measured by 10 working home appliances, information about operating characteristics of connected devices can be obtained without installing a measuring device or information collecting device in individual devices. It was shown that the operating states of major devices can be estimated without using them.

Further, when a device operation characteristic model is given, it is possible to improve the estimation accuracy by recognizing a current change which is recognized as a current change of one device due to interference and a current change of a different device. It was

In the device operation signal separation problem, attention is paid to the fact that the property of independent operation of the device appears in the current change (current change) of the device, and the component corresponding to the current change (current change) of the main connected device. Could be separated. This property is based on the very general property that devices operate asynchronously and autonomously.

However, the linearity hypothesis, that is, the characteristic that the intensity ratio of the harmonics of each device is device-specific and almost constant in time, is not strictly established. There are two causes for this. One is that there are devices that have similar intensity ratios, such as inverter fluorescent lamps and pots, and that they have non-linearity that the intensity ratio changes from time to time, such as in air conditioners. The first cause causes a plurality of devices having similar intensity ratios to be the same, and the second cause causes signal dispersion such as interference with other components or separation of a single device into a plurality of components. These cause the estimation accuracy to deteriorate.

It has been shown that, for a device having a similar intensity ratio, if a device-specific operating characteristic model is given, it can be recognized as a current change of another device by the proposed method based on intensity information obtained from the model. For the variance of the signal due to the non-linearity, the corresponding signal component (case 197
In the air conditioner 2 described above, the dispersion is not so severe that the third component) becomes unclear, and remains at an effective level for operating state estimation. However, in order to improve the estimation accuracy and estimate the operating state of the small current device, it is necessary to take measures.

The technique of estimating the operation of the connected device while paying attention to the independence is a very useful technique in that the operation can be estimated to some extent without preparing a device-specific model in advance.

The above-described embodiment is an example of a preferred embodiment of the present invention, but the present invention is not limited to this, and various modifications can be made without departing from the gist of the present invention. For example, in this embodiment, the effective current value of the fundamental wave of each device, that is, the time difference δI of the effective current value I ₀ (t).
Although the method of using (t) is described, the current value of the harmonic of each order is I (t) = √2 · X by the phase difference ω (t) between the current and voltage of the harmonic order of each order. _{By using} complex expression (i: imaginary number) as ₀ (t) ・ (cos (ω (t)) + i ・ sin (ω (t))), it is possible to easily expand the current including the reactive current. It is possible.

For the time difference δI (f, t) of the harmonic current value (complex expression) of each order of the total current value, by using the independent component analysis method for the complex matrix, similarly, It is possible to estimate the time difference ΔS (i, t) of the current value (complex expression) of the fundamental wave. This method corresponds to the case where the phase difference ω (t) is approximated to be substantially constant.

Although the present embodiment has mainly described the estimation of individual power consumption of non-intrusive electric power equipment, the method of use is not particularly limited, and it can be used to warn of an abnormal operation of electric equipment. . That is, from the information about the power consumption obtained by the power consumption estimation system of the electric device, for example, when it is determined to be abnormal in comparison with the daily power consumption,
It is possible to judge the safety of the person in the power consumer room, the safety in the power consumer, the presence or absence of an abnormality in the electric device or the electrification system, and transmit the information to the outside. For example, this system will allow people in the room to be turned on and off daily, TVs,
It is possible to determine the "safety of people in the power consumer's room" from the operating state of electric kettles, hot water toilet seats, etc., and also to use electric irons, electric stoves, electrified kitchens, etc. for a long time It is possible to judge the "safety in the electric power consumer" from "Panashi" etc. For "external transmission of these information", please refer to existing telephone lines, PHS, pagers,
You can use the Internet, etc.
Can be assumed to be the occupants themselves, relatives of the occupants, fire departments, welfare medical personnel such as local governments, etc.

[0102]

As is apparent from the above description, according to the remote electric equipment monitoring method and apparatus of the present invention as set forth in claims 1 and 4, the current change of one equipment is independent of the current change of another equipment. Since it is possible to separate the current change for each device of the same harmonic intensity ratio component from the total current consumption by using the independent component analysis method and estimate the operating condition for each device, It is possible to grasp the operating status of multiple devices without constructing a database in. Since it is not necessary to build a database, it is possible to easily estimate the current consumption of individual devices, its change, and the operating state from only the total current consumption values of multiple devices at low cost. Moreover, since it is not necessary to attach the measurement sensor to each electric device to be measured, only by installing the measurement sensor at any one point of the power supply line of the customer, for example, near the power supply line inlet of the outdoors.
When this system is installed in an electric power consumer, it has the advantage that it violates privacy and does not require additional wiring.

Then, such a customer (factory, building,
It is important for both the electric power industry and consumers to understand the actual usage status of electrical equipment in ordinary households. In electric power, it can be used for building a fee system, developing various energy-saving service businesses for customers, and for customers in energy-saving operation control and detection of equipment failure.

According to the second and fifth aspects of the present invention, the current change of the device group having the same harmonic ratio can be further separated based on the information for each device, for example, the current change of the fundamental wave (rating information). Therefore, the estimation accuracy of the current change amount can be improved.

Further, according to the inventions of claims 3 and 6, it is possible to obtain the current consumption and power consumption for each device group having the same harmonic ratio.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing an embodiment of a remote electric device monitoring method of the present invention.

FIG. 2 is a schematic block diagram showing an embodiment of a remote electric device monitoring apparatus of the present invention.

FIG. 3 is a flowchart showing an example of a process of estimating a fundamental wave current variation amount for each device group having the same frequency utilization ratio.

FIG. 4 is a flowchart showing an example of an optimal independence index estimation step.

FIG. 5 is a flowchart showing an example of a separation matrix update algorithm (fourth step of the estimation step of FIG. 3) when the optimum independence index G is used.

FIG. 6 is a flowchart showing an example of a process of estimating a current change amount for each device from a device group basic current change amount.

FIG. 7: Change in effective current and current change of each device (Case 1
97) is a graph.

FIG. 8 is a graph showing the degree of interference between separation due to effective current of equipment and separation based on current change.

[Figure 9] Current change at each frequency and air conditioner (A1, A2)
6 is a graph showing the current change value of

FIG. 10 is a graph showing an approximate effective current and an approximate differential current for each device.

11A and 11B are graphs showing absolute values (corresponding to all cases) of correlation coefficients between an approximate differential current and an approximate differential current error, where (a) is the correlation of the approximate differential current and (b) is the correlation of the approximate differential current error. Are shown respectively (the numbers on each axis are the device numbers).

FIG. 12 is a graph of a signal separation result (Case 197) based on a change in effective current by frequency, which is a result of independent component analysis according to the present invention.

FIG. 13 is a comparison diagram of current changes of the first, second, and third independent components and the operating device.

FIG. 14 is a graph showing an operating characteristic model of the air conditioner A2 (logarithm of the probability that a current change can be obtained).

FIG. 15 is a graph showing a representative independence index and an optimized independence index.

FIG. 16 is a graph showing an estimated current change (in parentheses: corresponding operating equipment) as a result of further independent component analysis using an operation characteristic model.

[Explanation of symbols]

1 electricity consumers 2 power lines 3 electrical equipment 11 measurement sensor 12 Frequency component converter 13 hour difference device 14 Independent component analyzer 15 Equipment-based signal separation device

Claims (6)

[Claims] 1. A remote electric device monitoring method for estimating a movable state of a plurality of electric devices used by an electric power consumer, wherein a total load current is measured from a power supply line of the electric power consumer,
The total load current is converted into a current for each of the fundamental wave and the harmonics, current change data is created by taking a time difference between the currents for the fundamental wave and the harmonics, and the current for each of the fundamental waves and the harmonics is created. The change is separated for each component estimated as a device group having the same harmonic intensity ratio by independent component analysis,
A method for monitoring a remote electric device, characterized in that the operating condition (current change) of each device of the monitored device is estimated from the waveform of the current change for each of the same harmonic intensity ratio components. 2. Among the current changes for each of the same harmonic intensity ratio components, the component of the device exhibiting the same harmonic intensity ratio is further separated based on the information on the current change intensity of the monitored device to separate the components. The remote electric equipment monitoring method according to claim 1, wherein the change in current is estimated. 3. The power consumption is estimated from the current consumption for each device showing the same harmonic intensity ratio obtained based on the remote electric device monitoring method according to claim 1 or 2. Power consumption estimation method. 4. A remote electric device monitoring apparatus for estimating a movable state of a plurality of electric devices used by an electric power consumer, and a total current sensor for measuring a total load current from a power supply line of the electric power consumer; A frequency component converter for converting the total load current into a fundamental wave and a harmonic current of the total load current, and a time difference for obtaining a current change by obtaining a time difference between the fundamental wave and the harmonics of the total load current. From an apparatus, an independent component analyzer that separates the current change for each component estimated as a device group having the same harmonic intensity ratio by independent component analysis, and a waveform of the current change for each of the same harmonic intensity ratio components A remote electric device monitoring device, comprising: a device-based signal separation device that estimates the operating status (current change) of the monitored device for each device. 5. Among the current changes for each of the same harmonic intensity ratio components, the component of the device exhibiting the same harmonic intensity ratio is further separated based on the information regarding the current change intensity of the monitored device to separate the components. The remote electric equipment monitoring apparatus according to claim 4, wherein the change in current is estimated. 6. The power consumption is estimated from the current change of each device showing the same harmonic intensity ratio obtained in the remote electric device monitoring device according to claim 4 or 5, and the power consumption is estimated. Power estimation device.