Development and application of raman OTDR for system of optical fibers distributed temperature control of overhead power transmission lines
However, there are situations when the temperature control of OF itself is required. In fiber optic communication lines at overhead power transmission lines (FOCL-OL) the optical power ground wire (OPGW) is often used. In a number of climatic regions the ice formation on such wires is possible. One of the methods of ice removal consists in transmission of electric current through OPGW, which results in its heating and ice removal. During this process OF is also heated, and it is required to avoid its temperature increase above the specified values.
FOCL-OL are quite long. Ice accretion on the cable is inhomogeneous along its length due to the difference of environmental conditions: temperature, humidity, wind velocity and direction. The conditions of OF arrangement relative to cable sheath, from which the heat is transferred to OF, are different too. Therefore, the efficient control of OF state is possible only in case of direct measurement of its temperature distribution along OPGW. Any point sensors installed on the outer surface of cable cannot be the source of reliable data of the whole OF temperature.
METHOD OF RAMAN REFLECTOMETRY
FOR OPTICAL FIBER
OF temperature measurement is usually based on the analysis of characteristics of Brillouin or Raman scattering. Each of these methods has its advantages; in every case, it is required to overcome certain technical difficulties for the method implementation.
During the propagation of optical radiation along OF the scattered signal consists of:
• Components with the same wavelength as the wavelength of incident radiation – Rayleigh scattering;
• Components with different wavelengths – Raman and Brillouin scattering, the spectrums of which, in turn, contain not one but two lines; the spectral lines with larger wavelength in comparison with Rayleigh wavelength is called Stokes line, and line with smaller wavelength is called anti-Stokes line.
Parameter of Brillouin scattering, on the basis of which OF temperature is determined, is its components frequency variation. However, the mechanical OF tension also has impact on this frequency; moreover, the sensitivity to it is significantly greater than in relation to temperature variation. In case of ice formation, both factors have impact, and it is very difficult to distinguish their influence, if not possible to perform this differentiation under the conditions of actual FOCL-OL.
Ability to measure the temperature of OF by the characteristics of Raman scattering is based on the ration between the intensities of anti-Stokes and Stokes components :
IАS and IS – intensity of anti-Stokes and Stokes components;
λАS and λS – wavelength of anti-Stokes and Stokes components;
ν – difference of wave numbers of incident radiation and Stokes (anti-Stokes) component;
Т – OF absolute temperature;
h – Planck’s constant;
k – Boltzmann’s constant;
с – light speed in vacuum.
The expression (1) shows that the ratio between the intensities of components of Raman scattering increases with OF temperature growth; it occurs mainly due to the growth of anti-Stokes component. For the standard single-mode OF and wavelength range used in fiber optic communications IАS/IS change is approximately 0.8%/K.
One of the most common methods of OF state diagnostics is optical pulse domain reflectometry: probing optical pulse is introduced into OF, continuous signal of backward scattering is recorded and dependence of its intensity on time is analyzed. As it was mentioned above, Rayleigh and Raman components are always present in the backscattering signal spectrum. Power of Rayleigh scattering does not depend on OF temperature. In order to determine the temperature of OF, it is required to separate these spectral components.
Thus, in order to solve the task of temperature control of FOCL-OL OF the method of Raman pulse reflectometry was selected.
DISTRIBUTED TEMPERATURE SENSOR АТР-111
As a result of large FOCL-OL lengths, it is reasonable to use for measurements the wavelengths range of 1550–1625 nm, at which OF attenuation is minimal. But then the wavelength of Stokes component becomes larger than 1650 nm, and losses in OF significantly increase, the sensitivity of receiving equipment decreases, and component recording is possible only at short distances. As a result, in order to measure OF temperature, only anti-Stokes component  can be used in accordance with the expression:
The value IАS depends not only on OF temperature but also on the common attenuation, and therefore it is necessary to distinguish their impact. It can be done if the signals of Rayleigh scattering at the wavelengths of probing pulse and anti-Stokes component will be measured . Alteration of their intensity is determined only by of common OF attenuation, and this fact grants the opportunity to distinguish from anti-Stokes signal the contribution, which is made by the temperature variation.
As a result of the analysis of potential variants of construction of Raman OTDR, the device – distributed temperature sensor АТР-111 was developed; the structure chart of this device is shown in Fig. 1 and its appearance is given in Fig. 2. The laser diode (LD) with the wavelength of 1625 nm is used as a of probing pulse source. Its radiation is introduced into the measured OF through the unit for optical signal integration and separation containing the passive optical components – multiplexors and circulators. Upon the propagation in OF the pulse with the wavelength of 1625 nm generates Rayleigh backscattering signal scattering with the same wavelength and anti-Stokes component of Raman scattering, the central wavelength of which is about 1520 nm. These signals go back to OF, they separate from each other in the unit for optical signal integration and separation and get into two-channel receiver. Each channel consists of photodiode (PD), amplifier (A) and analog-to-digital converter (ADC).
One more signal measured by the device is Rayleigh scattering at the wavelength of anti-Stokes component of Raman scattering. For this purpose, LD with the nominal wavelength of 1550 nm is used. When making the device АТР-111 such LD is selected from the batch, the wavelength of which is the closest to 1520 nm; its additional adjustment is performed on the basis of regulation of laser current and temperature.
In the АТР-111, the backscattering signals also pass through the reference optical fiber OF-T, which is located in heat-insulated casing; its temperature is measured by built-in electronic sensor. The mutual disposition of signal levels from this OF and sensor reading are used for binding of measurement results to temperature scale.
The pulse former shown in Fig. 1 controls the operation of LD, and in the same manner as in usual OTDR it allows performing measurements at quite broad set of pulse durations – from 8 ns to 10 µs.
Device processor and personal computer perform the its control and signals processing.
OF reflectograms obtained during the measurements are shown in Fig. 3. There are following designations:
• "1625 nm" – reflectogram of Rayleigh scattering at the wavelength of probing pulse of 1625 nm;
• "1550 nm" – reflectogram of Rayleigh scattering at the nominal wavelength of probing pulse of 1550 nm;
• "AS" – reflectogram of anti-Stokes component of Raman scattering.
The first section of each reflectorgam (to the left from the mark 0 km) is the signal from internal OF-T, and the remaining part – signals of measured OF.
SIGNAL PROCESSING UPON OF TEMPERATURE MEASUREMENT
As it is well-known, the power of Rayleigh backscattering signal in each OF point is in direct proportion to the power of probing pulse in this point, backscattering coefficient and pulse duration . Power of the signal, which comes to OF input, is determined on the basis of the following expression:
Pbs, R (L) – power of the signal of backward Rayleigh scattering, which was supplied to OF entry from the point located at the distance L from OF beginning;
Ppluse – power of probing optical pulse at OF input;
BR – backscattering coefficient, which determines which part of the power of incident pulse is scattered and directed backwards to OF input;
τpulse – duration of probing optical pulse;
аpulse – attenuation coefficient of probing radiation; when considering Rayleigh scattering it is assumed that its value is identical for forward and backward signals.
In order to process the signal of anti-Stokes component of Raman scattering the analogous model is used:
Pbs,АS (L) – power of signal of anti-Stokes component, which was supplied to OF entry from the point located at the distance L from OF beginning;
BАS (T (L)) – coefficient of backward scattering, which determines which part of the power of incident pulse is transformed into anti-Stokes component of Raman scattering and directed backwards to OF entry; its value varies with the change of fiber temperature Т;
aАS – coefficient of decay of anti-Stokes component.
Parameters of the expression (4) are determined on the basis of the measurement results of the relevant reflectograms. Recording anti-Stokes component the device determines the sum of coefficients apulse + aАS. On the basis of Rayleigh scattering at each of two wavelengths the coefficient apulse and component of the coefficient aАS, which does not depend on the temperature, are measured. OF temperature is determined in accordance with the deviation of measured anti-Stokes signal from the values estimated taking into account only Rayleigh coefficients.
As a result of processing of reflectograms shown in Fig. 3, the OF temperature graph is plotted (Fig. 4). In particular, at this graph the temperature of internal OF-T (section to left of the mark 0 km) is 18 °C, and the temperature of external OF is 26 °C.
The actual FOCL consists of several sequentially connected sections with factory length. At splices the intensity of probing signal decreases, and inhomogeneities ("steps") occur on reflectogram. However, the value of variation of bacscattering signal depends not only on the splice loss junction but also on the ratio between their backscattering coefficients BR . The formulas (3) and (4) were generalized for the purpose of analysis of such lines.
Reflectograms of the line, which consists of two different OF – G.652 (fiber А) and G.655 (fiber B) and is measured from two sides, are given in Fig. 5. Although, it can be believed that connection loss does not depend on measurement direction but the "steps’ caused by it turn out to be different due to various coefficients BR. It can be shown that the actual loss value is equal to the average of the values measured from each side. The specific values of backscattering coefficients BR of the fibers, which form FOCL, are practically never known, and therefore for FOCL certification the measurements are performed from both sides and results are averaged.
Different values of the backscattering coefficients have impact on the OF temperature measurement as well. The graphs of temperature of line A-B after both OF holding at the same temperature within 24 hours are shown in Fig. 6. When measuring in the direction A → B, the temperature of the first OF (close to OTDR) is by 3.7°C higher than the temperature of the second OF. When measuring in the backward direction, the situation is opposite. Averaging of the results gives practically identical values of the temperature of both OF.
Thus, in order to obtain the correct values of temperature distribution of the line consisting of several spliced OF it is necessary to perform the measurements from two sides. However, in actual practice of FOCL-OL temperature control it is impossible therefore the difference of coefficients of backward scattering is the source of additional error of temperature measurement.
SYSTEM OF DISTRIBUTED TEMPERATURE CONTROL AND MEASUREMENT RESULTS AT OPERATING FOCL-OL
On the basis of distributed temperature sensor АТР-111, the system of distributed temperature control (SDTC) of FOCL-OL was created. Several such systems were installed in ice-dangerous regions. The structure of one such network is shown in Fig.7. Ice melting can be performed at any line using two ice melting equipments located at the points B and D. The temperature control of long lines is performed by the АТР-111 from two sides in different OF. The АТР-111 operates together with control unit (it is not shown in Fig. 7), which can have built-in optical switch. It allows measuring several FOCL by one АТР-111 as, for example, is organized at the point B.
All АТР-111 control is performed from the remote PC of operator workplace through Ethernet network. For the period of several SDTC operation great number of OF temperature measurements results in different seasons and time of day have been accumulated. They indicate the complex character of influence of environmental conditions on the OF temperature in OPGW. The temperature graphs of one FOCL are given in Fig.8. For convenience of comparison they are shifted in vertical direction in relation to each other. The marks of the splice locations, at which OF factory lengths are connected, are shown in the lower graph.
Graph 3 is measured in the evening of the 3rd of January, in other words during hours of darkness, when the difference of environmental conditions due to solar radiation is absent. OF temperature is practically identical along the whole line. At the same time, at other graphs the sections with temperature deviations occur. The section a–b is located between two adjacent splices, and therefore the difference of its temperature can be explained by OF parameters. On the other hand, the section c–d represents the part of continuous section of OPGW; its properties are probably connected with geographical alteration of FOCL direction and relevant alteration in relation to wind direction and sun rays.
In some points of FOCL-OL, for which ice melting is provided, the ice formation sensors were installed on the outer surface of OPGW. These sensors also measured the cable temperature. When the measurements were performed in the hours of darkness with supposedly steady temperature of fiber and cable, the data of these sensors corresponded to the reading of the АТР-111 with the deviations of not more ±2 °C. In other time of day or in case of ice melting the differences were significantly greater. It proves the assumption that OF temperature can significantly differ from the temperature of OPGW sheathing in case of unsteady conditions.
Variations of temperature distribution of the same OF during ice melting in November 2014 and January 2015 can be observed in Fig. 9 and 10. OF is heated after ice melting equipment is switched on and it is cooled after it is switched off. The moment of ice melting equipment switching off was determined by the operator on the basis of reading of ice detectors. The graphs of heating and cooling are shifted in relation to each other in vertical direction for the convenience of comparison. Graph 1 in both figures is measured in the beginning of ice melting. Numeration of graphs corresponds to the measurement sequence.
It is seen from Fig. 9 and 10 that OF heating takes place in very irregular manner along its length. There are sections with great temperature growth, and there are sections, in which the temperature changed insignificantly. Maximum recorded values of temperature were 76 °C and 37 °C. It is connected with the fact that ice melting in January 2015 lasted significantly less time. Comparison of the graphs, in which maximum OF temperature was reached, shows the presence of concurrent sections with intense and poor heating. It can indicate the similar conditions of ice formation in these cases.
Data of temperature distribution along OF during ice melting grant opportunity to the services which control this process, to regulate the melting mode in order to avoid exceeding the set temperature thresholds. Use of information concerning the climatic conditions, in which different sections of FOCL-OL are located, will make it possible to understand better the dynamics of OF temperature variation and enhance the efficiency of these activities.
Used principle of SDTC organization (remote control of measurements from single center, automatic analysis of obtained results) grant opportunities to perform its integration with other parts of ice melting system and automate the specified processes.
Ice formation takes place within rather limited time intervals and in certain season, therefore in other time the system of distributed temperature control is used as monitoring system with respect to the physical state of FOCL. The АТР-111 operate as usual OTDR, and in continuous automatic mode are performed the measurements of main line parameters: length, complete attenuation, distances to splices, attenuation of OF sections and connections. In case of occurrence of damages and breaks of FOCL, the system sends relevant messages to dispatcher services. The measurement statistical data, which allows tracking the slow degradations of OF and planning preventive measures, is stored on the server. In addition to automatic monitoring, the remote control in manual mode is possible, for example, for FOCL certification.