In addition, especially for pendent even of load conditions. In fact, either overloading or distributed solutions, the mechanical characteristics of fibers local overheating can cause high temperatures in transformers. In the case of low load conditions, overheating can occur due On the basis of their previous experience with fiber-optic sen- to reduced cooling efficiency, which in the springtime can be sors [16]—[19], the authors implemented a temperature measure- caused by pollution of the cooling unit due to seeds from trees ment principle [20] based on a sensing probe integrated with a and shrubs and in the winter by the unit freezing over.
The sensing The maximum temperature or so-called hot-spot tempera- element was obtained by replacing a little portion of the plastic ture is of great importance. A temperature change in the it is possible to define an aging rate: this doubles with every reference liquid RL results in a modulation of the refractive 6 C temperature rise of the hot-spot temperature above a index that modifies the propagation regime along the fiber.
This temperature of 98 C. There are a number of thermal models effect allowed the temperature of the fluid, in which the probe is for determining the hot-spot temperature currently under de- immersed, to be evaluated simply by monitoring the fiber output velopment by research institutes in cooperation with industries, using low-cost, dual-slope analog-to-digital ADC processing hardware.
The same measurement principle was then used to set up a temperature transducer based on a low-cost, multifiber Manuscript received May 4, ; revised June 13, FR , Italy e-mail: betta ing.
This second prototype was tested in the 40— C range and A. Pietrosanto and A. Scaglione are with the Department of Information En- had the following metrological characteristics: gineering and Electrical Engineering, University of Salerno, Fisciano SA , Italy e-mail: pietrosa diiie.
Sensor output power versus temperature characteristics for three — measurement time: 1 s; different RLs: a mineral oil; b soya oil, and c cedar oil. Despite these interesting characteristics, it should be pointed out that there are some limits linked mainly to the oils used as RLs.
In fact, the fiber output power versus temperature charac- teristics in the temperature range of interest see Fig. In addition, oils need a preliminary temperature treat- ment in order to improve the long-term stability that in any case is never lower than 0.
In this paper, the previously proposed sensor is redesigned in order to improve its performance and render it suitable for the application considered. The physical principle behind the op- eration of the prototype is briefly reviewed, and then the new prototype is described in detail. Both the laboratory metrolog- ical characterization and the field tests carried out on a power transformer are reported. Hardware layout. Measurement Principle Light propagation along a large-core optical fiber can be an- acts as a localized refractive index change that can be detected at alyzed by the classical ray method [16].
A ray undergoes total the fiber end by a power measurement. To be able to determine reflection at the core boundary if the angle between the ray the temperature value from such measurements, the relationship path and the fiber axis does not exceed the complementary crit- between the fiber output power and the temperature of the liquid ical angle of the core-cladding interface that, for a step-index should be known.
Although this relationship can be numerically fiber, reads: determined under particular assumptions propagation of merid- ional rays only, no polarization-dependent effects, etc. Obviously, this calibration should be car- where and are the core and cladding refractive index, ried out for every liquid whose temperature is to be measured. When total reflection occurs, the ray This problem was overcome by inserting the unclad zone in propagates along the fiber without any power loss bound-ray.
This coupling results in modulation of the index versus temperature characteristic. This implies that only propagating bound-ray power that can be detected at the fiber end. In acteristics of the fluid being measured. The Prototype value, than that of the fiber silica core. Thus, if a little portion of The hardware structure of each of the temperature probes pro- the cladding along an optical fiber is replaced by a liquid with duced is shown in Fig.
Although the LD stability proves to be very high better than 0. In order to min- imize the effect of PS characteristic fluctuations, mainly due to environmental conditions, the LD monitor current was mea- sured and used to normalize the measured sensor output. Receiving Section RS : The fiber output light power is con- verted into an electrical signal by a photodiode P ; then the con- ditioning circuit CC filters the noise and conditions the signal in the range 0—5 V, to be fed into the acquisition section.
Acquisition and Elaboration Section: This section is based Fig. Sensing element in the fixing structure. It runs the measurement software, 3M were used instead of m ones. The optical characteris- whose tasks consist mainly of 1 averaging ten subsequent mea- tics of this type of fiber are almost the same refractive indexes: surements in order to increase the resolution; 2 normalizing the core 1.
Finally, it sends the results to ticeable improvement in both spatial resolution and response a remote control unit via its serial port. Although the aim of the time. The fiber is bent into the tank with a curvature radius previous multifiber probe was to achieve increased temperature mm that does not cause bending losses. Two unclad range and improved sensor reliability and accuracy, in the pro- zones of 10 mm were now produced, but removing the plastic posed new sensor system a multi-fiber approach was still used cladding only on one side see Fig.
Two semi-unclad zones but with the aim of carrying out a larger quantity of measure- were preferred to a single longer unclad zone to improve the ments within a large-sized device.
The optimum length value was obtained through the simulation analysis described in [20]. Laboratory Metrological Characterization movements within the tank. A little hole, not visible in Fig. A number of tests were carried out to obtain the nominal char- Reference Liquid RL : A new RL, automotive antifreeze acteristics of the temperature sensor.
To this end, the calibration liquid, was characterized and used. The characteristics were obtained using a Lab- factor equal to 2. The prototype metrological characterization was carried out Tank T : A small cylindrical brass tank 12 mm diameter, by performing 40 calibrations to both increase and decrease 33 mm height was made by screwing a cover to the previously temperature values. In particular, the following characteristics described fixing structure see Fig.
Differential Pressure. Particle Counts. Light monitoring. Air Changes per Hour. Temperature Monitoring. Read More. Continuous Automated Solutions for. Results and Analysis 3. Static Monitoring and Tracking The results of static monitoring and tracking measurements, known information time interval data delivery to the webserver, delay between the time data is sent and time data received.
Graph of time interval of data transmission 25 Delay seconds 20 15 10 5 0 1 3 5 7 9 ith data received Figure 5. Graph of delay between sensor data transmission and reception According to Fig. Dynamic Monitoring and Tracking 3. It shows the system can monitor temperature and humidity work well. The average interval value of 6. Known average of delay time is 2.
Figure 9. When the middle track process begins, the distance between the car's position coordinates is greater in distance than before and after. This is due to the considerable data reception time of the delay, which is most likely due to the quality of the internet connection in that area which is less stable. Graph of temperature and humidity values resulting from data processing of the result monitoring and tracking test is shown in Figure From the test result of data processing of temperature and humidity values, it can be seen that the average temperature value is The largest interval value on this test is 24 seconds or when sending the 13th data.
While the smallest interval value is 4 seconds. From the total value of data transmission time interval in this test, it can be seen that the average time interval is 5. From the total time delay of each data transmission in this test, it can be seen that the average value of time delay is 1. Figure This can be seen from the distance between the spaced coordinate markers than at the end of the testing process, and it occurs because the greater delay of coordinate data transmission.
Based on Figure 14, the temperature value in the monitored car cabin is increasing progressively, while the humidity value is decreasing. From the data processing of temperature and humidity value of the test results, it can be seen that the average temperature value is The largest interval value in this test is 23 seconds, while the lowest interval value is 4 seconds.
From the overall values of data transmission intervals, it can be seen that the average interval result in this test is 7. The largest delay time in this test is 20 seconds, and the smallest one is 0 second. Based on the total delay time, it can be seen that the average delay time is 2. In this case, it is also affected by the value of data reception time delay.
Overall Results Analysis The measurement results with several speed variations are shown in Table 1. Table 1. This shows the performance of monitoring and tracking systems better when static vehicle conditions than when the conditions are dynamic. The car position is represented using the coordinate point markers displayed in the Googlemaps API.
Each coordinate point marker represents a car position coordinate data received from the GPS receiver, so that the value of data transmission interval and time delay between sending and receiving data determine the distance between the coordinate point marker displayed in the Googlemaps API. Conclusion Prototype temperature and humidity monitoring on IoT based Shipment tracking has been developed. The prototype has been exeperienced by two types of the measurement tested.
One is static either is mobile, and two is dinamic. The results shows that static measurements, the average interval value of the data transmission time is 5,75 seconds from the designed interval value of 5 seconds, the average value of the deviation from the interval of data transmission time is 0,75 seconds, and the average delay time between data transmission and reception is 0,3 seconds.
Based on the average value of the interval, the average value of the deviation and the average value of delay on static testing smaller than the average value of the interval, the average value of the deviation and the average value of delay in the dynamic test, then the performance of the monitoring and tracking system is better when the condition of the vehicle is static than when the conditions are dynamic.
0コメント