The present invention relates to an intensity modulation
type optical sensor, and in particular, an intensity modulation type optical current
/ voltage sensor capable of reducing noise.
As an optical sensor with high precision, for example,
there is a so-called optical heterodyne sensor (for example, refer to Non-patent
Document 1) in which interference occurring when mixing signal light modulated by
a measured object with a local oscillation signal having a different frequency from
the signal light is received and detected in a photoelectric detector. However,
due to the complicated structure, a current detection sensor that is based on a
principle of modulation of light intensity using the Faraday effect and a voltage
detection sensor that is based on a principle of modulation of light intensity using
the Pockels effect are recently being developed for practical use.
The basic operation of the optical intensity modulation
type sensors is as follows.
- a) The intensity of light transmitted from a light source to a sensor head is
modulated by a measured object.
- b) The light whose intensity has been modulated is transmitted to a light receiving
element and is converted into an electrical signal proportional to the amount of
light in the received signal.
- c) A change in the amount of received light is acquired from a change in a value
of the electrical signal, such that the size of the measured object is known.
In such an optical intensity modulation type sensor, it
is an important issue in securing the measurement accuracy to inhibit a change in
the amount of light in the received signal, which occurs due to factors other than
the measured object, from affecting a sensor output.
Causes of fluctuation in the amount of light may be attributed
to fluctuation in the light source intensity, a fluctuation in the transmission
efficiency of a transmission line used to transmit light to a sensor head, and fluctuation
in the transmission efficiency of a transmission line from the sensor head to a
light receiving element. In addition, each fluctuation in the amount of light may
be divided into a drift fluctuation and a relatively high-speed fluctuation.
In order to inhibit the fluctuation in the amount of light
from affecting the output, the following measures disclosed in Non-patent Document
2 are taken in addition to hardware measures, such as ensuring robustness of an
- 1) Calculation of a degree of modulation: an AC component and a DC component
are acquired from the received electrical signal obtained by converting the received
signal light by using a filter, and a ratio (degree of modulation) of both components
- 2) Stabilization of energy input to a light source: a control is made such that
electrical energy input to the light source is constant.
- 3) Control of the light source intensity: electrical energy input to an element
in order to drive a light source element is controlled such that the average time
of the amount of received signal light is constant.
In principle, both measures described in 2) and 3) cannot
be performed at the same time.
Non-patent Document 1:
'Analyses of basic characteristics of photocurrent transformer applying an
optical heterodyne method', the Institute of Electrical Engineers of Japan, journal
B. Vol. 117, No. 3, 1997 (pp. 354 - 363
Non-patent Document 2:
'Technical data on optical fiber sensor: revised edition', edited by Toshihiko
Yoshino, 1986 (pp. 404 - 405
DISCLOSURE OF INVENTION
By taking measures described in 1) to 3), the drift fluctuation
may be removed. On the other hand, the method described in 1) is useless for reducing
relatively high-speed fluctuation in the amount of light which is not drift fluctuation.
The reason is that AC components of an electric signal according to the fluctuation
in the amount of light pass through a filter that extracts AC components from a
signal, such that it is not possible to distinguish between the AC components and
the signal. In particular, a main cause of fluctuation in the amount of light in
terms of an alternating current includes a signal ripple input to a light source
Therefore, it is an object of the invention to improve
measurement accuracy by preventing influences on a sensor output even in the case
when there is a ripple in a light source.
In order to solve the above problems, according to a first
aspect of the invention, there an intensity modulation type optical sensor is provided
in which light from a light source is guided to a sensor head formed of an optical
component, the intensity of the light is modulated on the basis of an alternating
current (AC) measured object changing with time in the sensor head, the modulated
light is received and converted into an electrical signal, a normalized received
signal indicating the degree of modulation is acquired from a ratio of an AC component
and a DC component of the electrical signal, and a value of the AC measured object
is acquired on the basis of the normalized received signal. The intensity modulation
type optical sensor is characterized in that a reference signal is acquired by separating
and receiving a part of light incident on the sensor head, a normalized reference
signal is acquired from a ratio of an AC component and a DC component of the reference
signal, and the normalized reference signal is subtracted from the normalized received
signal, thereby reducing noise.
An intensity modulation type optical current sensor may
be obtained by applying a principle of the Faraday effect to the first aspect of
the invention (second aspect of the invention). An intensity modulation type optical
voltage sensor may be obtained by applying a principle of the Pockels effect to
the first aspect of the invention (third aspect of the invention) as well.
According to the invention, noise occurring due to, for
example, a ripple in a light source can be reduced. Accordingly, advantages are
obtained in that it is possible not only to improve the measurement accuracy but
also to easily realize an optical current sensor or an optical voltage sensor.
BRIEF DESCRIPTION OF THE DRAWINGS
BEST MODE FOR CARRYING OUT THE INVENTION
- Fig. 1 is a block diagram illustrating an embodiment of the invention;
- Fig. 2 is a block diagram illustrating a modified example;
- Fig. 3 is a view illustrating the configuration for verifying the invention
and conducting experiments thereon;
- Fig. 4 is a view illustrating waveforms according to a first experimental result;
- Fig. 5 is a view illustrating waveforms according to a second experimental result.
Fig. 1 is a block diagram illustrating an embodiment of
In the drawing, a light source (ASE) 1, a fiber coupler
2, an optical path 3 formed of a fiber, an optical component 4 including a ferromagnetic
Faraday rotator and a polarized light separating element, a mirror 5, a sensor fiber
6, and a measured conductor 7, are shown. Reference numerals 81 and 82 denote light
receiving elements, such as photodiodes (PD). Reference numerals 91 A and 92A denote
band pass filters (BPF), reference numerals 91B and 92B denote low pass filters
(LPF), and reference numerals 101 and 102 denote dividers. Reference numeral 11
denotes a subtractor.
Now, noise included in the light source 1 is modeled as
a sinusoidal wave having a frequency &ohgr;n. Light PL output
from the light source 1 is expressed using &agr; as noise content as follows.
In addition, assuming that a Faraday rotation angle is
&thgr;F, a current flowing through a conductor is i (i = IOcos&ohgr;st),
and a Verdet constant is V, the Faraday rotation angle is &thgr;F is
expressed by Expression (2) given below.
Assuming that an output of the light receiving element
81 is VS and a transmission and conversion efficiency is 'a', the output
VS is expressed by Expression (3) given below.
Substituting Expressions (1) and (2) for Expression (3)
and assuming that |2&thgr;F| <<&pgr; / 2, VS is expressed
by Expression (4) given below.
Expression (5) is obtained by expanding Expression (4).
Here, assuming that &ohgr;n ≠ &ohgr;S,
a first term in Expression (5) corresponds to a DC component and terms subsequent
to the first term correspond to AC components.
Then, assuming that aPO → DS
and aPO (1 + 2VlOcos&ohgr;St + &agr;cos&ohgr;nt
+ 2VIO&agr;cos&ohgr;nt · cos&ohgr;St)
→ AS, a ratio XS of DS and AS
is expressed by Expression (6) given below.
Here, the first term corresponds to a signal component
and second and third terms correspond to noise components. Thus, the received signal
Ps is normalized by calculating the ratio of a DC component and an AC component.
Comparing the second and third terms with each other, the
effect caused by the second term is larger because &agr; and 2VIO&agr;
are considered to be a small amount. Accordingly, in order to remove the term, reference
light Pr is used.
First, assuming that 'b' is a transmission and conversion
efficiency, an output Vr of the light receiving element 82 is expressed
by Expression (7) given below.
In the same manner as described above, a first term corresponds
to a DC component and a second term corresponds to an AC component. Accordingly,
assuming that bPO → Dr and bPO&agr;cos&ohgr;nt
→ Ar, a ratio Xr of Dr and Ar
is expressed in by Expression (8) given below.
That is, the reference signal Pr is normalized by calculating
a ratio Xr of both Ar and Dr.
If Expression (8) is subtracted from Expression (6) described
above, Sout is calculated by Expression (9) given below.
Comparing Expression (6) with Expression (9), it can be
seen that a noise component corresponding to the second term of Expression (6) does
not exist in Expression (9). In addition, even though a noise component corresponding
to a second term still remains in Expression (9), the second term is very small.
That is, since the second term of Expression (9) is &agr; times the second term
of Expression (6), which is very small as compared with the second term of Expression
(6), the second term of Expression (9) is negligible.
That is, it is easily considered that a signal (reference
signal Pr) used as reference is extracted and the signal is subtracted from a received
signal Ps when a noise component is included in the received signal. At this time,
the noise component cannot be removed only by simply subtracting the reference signal
Pr from the received signal Ps. The reason is that time average values (signal levels)
of the received signal Ps and the reference signal Pr are different, and accordingly,
levels of noise components included in both the received signal and the reference
signal are different.
Therefore, in the invention, a reference signal Pr is also
normalized as an AC component and a DC component, the degree of modulation (normalized
reference signal) of the reference signal is calculated, and the normalized reference
signal is subtracted from a normalized received signal, such that noise components
can be accurately removed.
Fig. 2 illustrates a modified example of Fig. 1. While
only the reference signal Pr is transmitted from the fiber coupler 2 in Fig. 1,
the received signal Pr is also transmitted from the fiber coupler 2 in Fig. 2 so
that only one optical path 3 is used in Fig. 2. Since the others are the same as
Fig. 1, a detailed explanation thereof will be omitted.
In the above, a current detection sensor that is based
on the principle of modulation of light intensity using the Faraday effect has been
explained. However, the invention can also be applied to a voltage detection sensor,
which is based on the principle of modulation of light intensity using the Pockels
effect, in the same manner as described above.
Fig. 3 illustrates an example of the configuration for
verifying the invention and conducting experiments thereon.
In this example, the same configuration as in Fig. 1 is
used, and a signal waveform of each terminal is observed using an oscilloscope when
an AC current i (equivalent measured current = 0.35 x 760 (times) = 266 A [rms])
having an effective value (rms) of 0.3 A, 50 Hz flows through a conductor 7. In
addition, the wavelength of light from a light source 1 is 1550 nm, the amount of
light attenuated in the optical fiber coupler 2 is 3 dB, a fiber 6 is wound once,
and a conductor 7 is wound around the fiber 760 times.
Figs. 4 and 5 illustrate waveforms observed by using the
above-mentioned oscilloscope. In addition, Fig. 4 illustrates a case in which a
reference signal terminal T4 is connected in Fig. 3, and Fig. 5 illustrates a case
in which the reference signal terminal T4 is not connected in Fig. 5. A measured
current Y from a terminal T0 is introduced into an upper side (Ch1) of an oscilloscope
screen and signals As, Ar, and Sout from terminals T1 to T3 are introduced
into a lower side (Ch2) of the oscilloscope screen, and the amounts of those described
above are expressed as shown in Figs. 4A, 4B, and 4C. In addition, a horizontal
axis of the oscilloscope screen indicates a time axis of 5 msec / div, the upper
side (Ch1) of a vertical axis of the oscilloscope screen indicates a voltage axis
of 20 mV / div, and the lower side 5 (Ch2) of the vertical axis indicates a voltage
axis of 100 mV / div.
As is apparent from comparison of the relationship between
the measured current Y and the signal Sout shown in Fig. 4C and the relationship
between the measured current Y and the signal Sout shown in Fig. 5, a
waveform of the signal Sout is distorted with respect to a waveform of
the measured current Y in the case corresponding to the related art, which is shown
in Fig. 5 and in which the reference signal is not used. However, in Fig. 4C of
the invention in which the reference signal is used, the waveform of the signal
Sout is a sinusoidal wave, which is almost the same as the waveform of
the measured current Y. Accordingly, it can be understood that noise components