THE technological evolution of microcontrollers and their

Transcription

1 Design and experimental characterization of a low-cost, real-time, wireless AC monitoring system based on ATmega 38P-PU microcontroller M. Caruso, A. O. Di Tommaso, R. Miceli, Member, IEEE, G. Ricco Galluzzo, P. Romano, G. Schettino and F. Viola Department of Energy, Information Engineering and Mathematical Models University of Palermo, Viale delle Scienze, Building nr Palermo, Italy Abstract This paper describes the development and experimental characterization of a real-time, low-cost, wireless AC monitoring system based on ATmega 38P-PU microcontroller. The proposed system is composed by a main brain for the acquisition and the processing of the data measured by smart meters. From the experimental tests discussed in this work it is demonstrated that the proposed system can monitor, with adequate accuracy, the working cycles of electrical loads in order to evaluate its energy consumption. Index Terms Monitoring system, energy consumption, low cost. I. INTRODUCTION THE technological evolution of microcontrollers and their massive diffusion in several fields of the industrial world has determined a significant increase of high performance systems capable to manage simultaneously several activities in real-time. This progress has allowed the implementation of functions for signal processing with reduced costs and consumption and with high reconfigurability. In addition, the adoption of such systems allows the digital signal processing, bringing countless advantages to traditional analog processing [], []. Moreover, the reduction of energy consumption has reached great interest because of its connection with several environmental problems (e.g.co emissions). As a matter of fact, the European Directives /3/CE, 9/8/CE and /3/CE promote the energy saving and, for this purpose, intelligent strategies for energy management and control are widely used. As example, the application of the wireless sensor network is used for monitoring and real-time diagnosis of energy consumption [3] [5]. Based on these considerations, the purpose of this work is the development of a low-cost monitoring system, whose high flexibility is achieved by using open-source technologies. The proposed system has been properly designed in both its hardware and software components in order to reduce costs and to maintain an adequate accuracy, if compared with commercial devices. The experimental tests discussed in section III demonstrate that the proposed system can monitor, with good accuracy, the working cycles of electrical loads in order to evaluate its energy consumption. The proposed system is the starting point for a future development of a model that allows the realtime detection of fault conditions in electrical drives and the analysis of the related harmonic content. II. DESCRIPTION OF THE PROPOSED SYSTEM The proposed monitoring system allows the AC real-time monitoring of voltage and current of electrical loads. In particular, an automatic measuring system (SAM) has been designed and developed. This system is mainly composed by the interconnection of several smart meters with a central main brain. Such interconnection is achieved by using a wireless network with radiofrequency technology operating at 35 MHz. The principle of operation of the system can be summarized into four main steps: SAM activation, consisting in the transmission of an identification code from the main brain to the related SAM Data measurement, where the voltage and current waveforms are acquired by the activated SAM and, then, processed Data transmission, where the activated SAM transmits the voltage and current values to the main brain Data processing, where the data are processed by the main brain and saved into its storage system. A. The Automatic Measuring System (SAM) The Automatic Measuring System (SAM) described in this paper has been developed for the real-time detection of voltage and current of an electrical load. The SAM is mainly composed by: A control system, realized with the adoption of a AT- MEL microcontroller (model ATmega 38P-PU) and an external clock circuit with operating frequency at MHz. A -bit, SAR- type, ADC converter is integrated in the microcontroller [] and the Arduino UNO IDE (Integrated Development Environment) is used for the related programming. An electrical circuit for the voltage measurement, shown in fig.. The output voltage signal from the TV is filtered by a low-pass filter, composed by the resistance R and

2 the capacitance C (see fig. ), which allows the reduction of high-frequency noise and aliasing. The signal is, then, reduced in amplitude by the use of the voltage divider composed by the resistances R and R 3, in order to obtain a peak value of V. Finally, the voltage follower (integrated circuit LM9) allows the decoupling of the voltage signal. An electrical circuit for the current measurement, shown in fig.. The current probe is a split-core type autotransformer (model YHDC SCT-3-). The output current signal from the TA (see fig. ) is converted into a voltage signal through the load resistancer burden. In dependence of the expected current range, the value of R burden is determined in order to obtain the maximum resolution and to prevent the core saturation. In addition, a low-pass filter and a voltage follower (integrated circuit LM9) provide the adequate operation of the sample and hold circuit. A transmission system based on the RF-RX-35 and RF- TX-35 modules, which is described more in detail in section II-C. Table II TECHNICAL FEATURES OF THE SAM SYSTEM Features SAM Supply voltage [V] 9 Input voltage range a.c. [V] 3 Input current range [A] Sampling frequency [KS/s]. Bit resolution Input connection single-ended ADC SAR Communication [MHz RF-35 Launch distance [m] max length transmission [bytes] 7 bit rate [bps] antenna length [cm].5 subsequent data processing in order to determine voltage, current, active power and power factor. More in detail, the rms values of both voltage and current are determined by the following equations: V {rms} = N s N S {i=} V i () I {rms} = N s N S {i=} Furthermore, the active power P is given by: I i () Figure. Electric circuit for the voltage measurement. P = N s N S {i=} Finally, the power factor is given by: cos ϕ = where N S is the number of samples. V i I i (3) P V rms I rms () Figure. Electric circuit for the current measurement. Table I summarizes the values of R burden evaluated in the current range intended to be measured, whereas Table II summarizes the technical features of the SAM system. B. Data Processing Table I LOAD RESISTANCE VALUES Current range [A] R burden [Ω] 33 ± % 5 8 ± % 33 ± % 5 8 ± % 33 ± % The control software implemented in the system provides the sampling of both voltage and current waveforms and the C. The main brain The main brain provides the management of all the information received from the smart meters. It is mainly composed by: A control system, which is realized by using a ATMEL microcontroller (model ATmega 38P-PU) and an external clock circuit with operating frequency at MHz. A data storage system, composed by a SD card (LC STUDIO), which uses an SPI (Serial Peripheral Interface) communication, managed with the SD.h software library. A wireless communication network based on a radiofrequency technology operating at 35 MHz, used for the communication between main brain and smart meters. In particular, the modules RF-RX-35 (receiver) and RF-TX-35 (transmitter), whose features are reported in Tables III and IV, respectively, have been used [7], [8]. The wireless technology is chosen for its simpler and cheaper use, if compared the traditional wired networks. For the management of the communication between the components, the open-source VirtualWire.h software library is used [9].

3 Table III TECHNICAL FEATURES OF THE RECEIVER MODULE Features RF-RX-35 Operating voltage [V] 5 ±.5 Operating current [A] 5.5@ 5 V Receiving frequency [MHz] 35 Bandwidth [MHz] Data output TTL Dimensions [mm] 3 x x 7 Antenna length [cm].5 those measured with the reference device previously described. The input voltage has been varied from to 3 V and the proposed monitoring system has detected 8 current and voltage measurements. A photograph of the test bench during the related measurements is shown in fig.. Table IV TECHNICAL FEATURES OF THE TRANSMITTER MODULE Features RF-TX-35 Operating voltage [V] 3.5 Operating current [ma] max ( V) min 9 (3 V) Transmitting frequency [MHz] 35 Launch distance [m] Frequency error [khz]] ±5 (max) Transmitting power [mw] Dimensions [mm] 9 x 9 x 7 Antenna length [cm].5 III. EXPERIMENTAL TESTS AND RESULTS In order to verify the proper operation of the proposed system, several experimental tests have been carried out through the assembly of a test bench at the LEPRE and SDESLab laboratories, University of Palermo, Italy. More in detail, the bench, whose schematic representation is shown in fig. 3, is composed by: A variable, single-phase autotransformer, whose main rated values are reported in Table V. An electrical load, consisting of a variable rheostat. A digital multimeter, whose rated values and are summarized in Table VI, used as reference measuring device. A LeCroy LT3L scope, used for the visualization of the acquired signals. Table V RATED VALUES OF THE AUTOTRANSFORMER Producer Producer A Belotti Input voltage [V] 3 Output voltage [V] 3 Power [VA] Current [A] 5. Frequency [Hz] Table VI FLUKE Model 79 AC voltge range [V] 3 ( khz) DC voltage range [V]. AC current range [A].3 DC current range [A]. The first test has been carried out in order to experimentally verify the accuracy of the proposed system. More in detail, the data acquired with the device have been compared with Figure 3. Schematic representation of the test bench. Figure. A photograph of the test bench. Figure 5shows the comparison between the trend as function of the number of readings of the current detected by the proposed system (red) and the current measured by the reference device (blue). Moreover, fig. shows the comparison between the voltage/readings trends detected by the proposed system (red) and the voltage/readings trends measured by the reference device (blue). For both trends it is possible to notice a difference between the values measured by the proposed system and those measured with the reference device. This error can be compensated via software, obtaining the trends reported in figures 7and 8. The second test has involved the monitoring of a working cycle of an electrical load in order to estimate its energy consumption. For this purpose, a common washing machine, whose technical features are summarized in Table VII, has been used as load. It is mainly composed by: A control system for the management of the washing programs 3

4 Current [A] 5 Current [A] Figure 5. Comparison between the current measured by the reference device (blue) and the current measured by the proposed monitoring system (red). Figure 7. Compensation of the current measurement. 5 5 Voltage [V] 5 Voltage [V] Figure. Comparison between the voltage measured by the reference device (blue) and the voltage measured by the proposed monitoring system (red). An electrical motor for the washing machine drum An electrical resistance for the water heating A drain pump A solenoid valve for the water charging. Table VII MAIN FEATURES OF THE ELECTRICAL LOAD Producer SAMSUNG Model WF7F5ENW Efficiency class A++ Voltage supply [V] 3 Energy consumption f.l. [kwh].98 Energy consumption h.l. [kwh].79 maximum load [kg] 7 During the working cycle of the machine, 9 samples of both voltage and current have been acquired and the trends over time of voltage, absorbed current and active power are shown in figures 9, and, respectively Figure 8. Compensation of the voltage measurement. As shown in fig., the power absorption changes over time in dependence of the working activity. As example, in the range 5-9 min a high power absorption, equal to 3 W, is detected. This fact is mainly due to the water heating, with an overall value of absorbed energy of.9 kwh. Then, the power absorption highly decreases in the time range 3-7 min. In conclusion, from the carried out tests it is demonstrated that the proposed SAM system is suitable for the real-time detection and analysis of electrical consumption occurring in specified working cycles of electrical loads. This analysis is needed when energy saving policies are required. IV. ECONOMIC ANALYSIS The economic analysis has been a very important aspect for the design of the proposed system. In fact, this work is finalized to relevantly reduce the realization costs of the previously described prototype, but without decreasing its performances with respect to commercial devices. The costs for the realization of the main brain and the SAM are reported in Tables VIII and IX, respectively. As it is possible to notice,

5 5 5 Voltage, rms [V] 5 Power [W] time [min] 3 5 time [min] Figure 9. Trend over time of the network voltage. Figure. system. Trend over time of the active power detected by the proposed Current, rms [A] 8 Table IX SAM COST Components Cost [ɛ] Control system 5. Voltage probe. Current probe 7. Wireless modules (RX and TX).5 Other components. Overall cost time [min] Figure. Trend over time of the absorbed current. in comparison to similar systems available in the electrical market, the overall cost has been significantly reduced. Table VIII MAIN BRAIN COST Components Cost [ɛ] Control system 5. Wireless modules (RX and TX).5 SD card 3. Other components. Overall cost.5 V. CONCLUSIONS This work has presented the design and the characterization of a low-cost, wireless SAM system. The hardware components have been chosen in order to minimize the overall cost and the wireless technology provides high flexibility to the system. Several experimental tests have been carried out for the validation of the device and the obtained results have demonstrated the adequate accuracy of the proposed system. Furthermore, this system is suitable for monitoring the trend over time of the main electrical quantities of an electrical load, in order to carry out evaluations in terms of its energy consumption. ACKNOWLEDGMENTS This work was financially supported by MIUR - Ministero dell Istruzione dell Università e della Ricerca (Italian Ministry of Education, University and Research), by PONa_H "i- NEXT" - Italian research program, and by SDESLab (Sustainable Development and Energy Saving Laboratory) of the University of Palermo. The Authors wish to thank all the SDESLab and LEPRE staff. REFERENCES [] J. Xiao, P. Liu, L. Jiao, H. Zhu, and Y. Du, Design of pv power station remote monitoring system data acquisition device, in Advanced Mechatronic Systems (ICAMechS), International Conference on, pp , Aug. [] U. Mujumdar and D. Tutkane, Development of integrated hardware set up for solar photovoltaic system monitoring, in India Conference (INDICON), 3 Annual IEEE, pp., Dec 3. [3] W. Naruephiphat, C. Charnsripinyo, S. Satienpaisarn, and R. PromYa, Applying wireless sensor network for power consumption monitoring, in Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), 9th International Conference on, pp., May. [] C. Yongpan, M. Xianmin, Z. Jili, and L. Zhen, Development of monitoring system of building energy consumption, in Computer Science- Technology and Applications, 9. IFCSTA 9. International Forum on, vol., pp. 33 3, Dec 9. 5

6 [5] S. Shaw, S. Leeb, L. Norford, and R. Cox, Nonintrusive load monitoring and diagnostics in power systems, Instrumentation and Measurement, IEEE Transactions on, vol. 57, pp. 5 5, July 8. [] Atmel, Atmega 38p-pu datasheet, in 3. [7] C. Technologies, Rf rx 35 receiver module,product user manual, November 8. [8] C. Technologies, Rf rx 35 receiver module,product user manual, November 8. [9] M. McCauley, Virtualwire, documentation for the virtualwire communications library for arduino, March 3.