Design and Implementation of a Smart Meter with Demand Response Capabilities

Transcription

1 Available online at ScienceDirect Energy Procedia 103 (2016 ) Applied Energy Symposium and Forum, REM2016: Renewable Energy Integration with Mini/Microgrid, April 2016, Maldives Design and Implementation of a Smart Meter with Demand Response Capabilities Luis I. Minchala-Avila a, Jairo Armijos a, Daniel Pesántez a, and Youmin Zhang b,c,* a Universidad de Cuenca, Av. 12 de Abril y Agustín Cueva, , Cuenca, Azuay, Ecuador b Concordia University, 1455 de Maisonneuve Blvd., Montreal, Quebec H3G 1M8, Canada c Xi'an University of Technology, Xi'an, Shaanxi, , China Abstract This paper presents the design of a smart meter (SM) with demand response (DR) capabilities. The SM design is tested in a simulation that implements an advanced measurement infrastructure (AMI), which allows a bidirectional communication between the household smart meters and the distribution management system (DMS). The DMS deploys an energy management system (EMS) that runs a simple demand response program (DRP) based on time of use (TOU), consisting in peak and off-peak rates. Results from the simulation and the data collected from the SM show significant improvements in energy consumption during peak hours thanks to the load curtailment strategies The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( The Authors. Published Elsevier Ltd. Peer-review Selection and/or under responsibility peer-review of under the scientific responsibility committee of REM2016 of the Applied Energy Symposium and Forum, REM2016: Renewable Energy Integration with Mini/Microgrid. Keywords: demand response; fuzzy control; smart meters 1. Introduction Smart meters (SMs) are gradually replacing the traditional meters, and are also being installed in new microgrids. SMs transmit information to different information clients via SCADA systems and other networks. One of the main features that SMs offer to consumers is the possibility to read in real time rates and pricing policies, allowing the implementation of DRP. These features are being exploited by utilities in order to achieve energy efficiency. Peak load reduction through an interactive reaction of the loads installed at the customer premises, e.g. turn on schedulable loads when cheap generation is available, increases network reliability and produce significant economic savings to the utility and the customers [1]. * Corresponding Author (youmin.zhang@concordia.ca) The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the scientific committee of the Applied Energy Symposium and Forum, REM2016: Renewable Energy Integration with Mini/Microgrid. doi: /j.egypro

2 196 Luis I. Minchala-Avila et al. / Energy Procedia 103 ( 2016 ) DR may limit, in certain degree, the comfort of the consumers. Therefore, it is desirable to limit the time during they may be exposed to such discomfort. Automation, monitoring and control techniques are fundamental to correctly manage the energy-use process, making DR less hindering for the customer [2]. A report with successful results on the implementation of DRP is presented in [3]. Significant economic savings are reported, since demand peak flattening avoids the need for more generation capacity. In [4], a load scheduling algorithm is implemented through a combinatorial optimization problem constrained by the maximum demand limit on a microgrid. The utility establishes TOU rates. In [5], a DRP is proposed that enables households to participate in DR services. Half-hour-ahead rolling optimization and a real-time control strategy are combined to achieve household economic benefits. A fuzzy logic controller is utilized to determine battery charging/discharging power; proper rules are proposed to ensure benefits from operating the battery storage system under the real-time electricity price. Reference [6] presents a demand side management strategy for load shifting based on heuristic optimization. The proposed optimization algorithm aims to shape the final load curve close to the desired load curve. The restriction of this strategy is compliant in the number of shiftable loads in the system, which users are willing to use at a different time. The results of this strategy are positive and show 5-10% operational cost reduction and % peak demand reduction. Additional DR methods are overviewed in [7]. This paper presents the design of a household SM with DR capabilities. An AMI simulation scenario is developed in order to test the DR capabilities of the SM. The TOU rates method is implemented for the simulation events. Additionally, an EMS processes real time data coming from the SMs, consisting of: 1) power measurements; 2) time and date; 3) load priority; and 4) prices of the electricity. The SM in conjunction with the EMS takes decisions on loads curtailment strategies. A knowledge based fuzzy controller (KBFC) calculates dimming percentages of loads that allow these commands, and shifts schedulable loads from peak consumption hours to low price energy generation hours. The results show improvements in energy savings and successful application of the DR commands in the user premises through the SM. 2. SM design SMs allow traditional meters capabilities as well as new features that aggregate intelligence to the grid. SMs integrate the ability to remotely manage loads at the end-user premises by monitoring and controlling the customer s devices and appliances. A dedicated communication infrastructure, sensors and control devices are required by SMs in order to effectively provide the interaction between the utility and the customer facility [8]. There are a variety of experimental setups of SMs, as detailed in [9] [12], which are designed to provide the following features: To measure customer s power consumption and generation; To support control function commands; To enable demand response capabilities; and To enable a variety of communication capabilities. Fig. 1 shows the block diagram of the SM design proposed in this research. The central processing unit of the SM is based on open source hardware (Raspberry Pi) due to the flexibility of programming and configuration this platform offers. The power measurement is performed through the dedicated integrated circuit AD7753, which needs an analog front-end (AFE) interface composed by a current sensing stage (Hall effect current transducer FHS-40P/SP600) and a voltage sensing stage (voltage divider). The measurements are sent via a serial peripheral interface (SPI). Fig. 2 shows the printed circuit boards (PCBs) of the implemented hardware design.

3 Luis I. Minchala-Avila et al. / Energy Procedia 103 ( 2016 ) Fig. 1. Block diagram of the SM Fig. 2. (a) ADE7753 board design. (b) FHS-40P/SP600 board design The Ethernet link of the control unit (Fig. 1) allows the SM identification through an IP and MAC addresses. The SM data is the combination of the following: meter identifier, timestamp, energy consumption values, control commands acknowledgment, and loads status (on/off, dimming level, and priority). The SM establishes a bidirectional communication with the EMS in order to send information regarding the energy consumption and to execute DR commands. Reference [13] shows a typical EMS architecture to apply the AMI concept. The DRP that is to be tested in this research is the TOU program. A simple pricing scheme composed of low, medium, and high rates is implemented. The energy price depends on the generation availability and load consumption. The DRP is implemented through a KBFC, which processes the energy consumption, pricing information, date/time, and load priorities. Fig. 3 shows the KBFC overview for implementing the DRP in the households through the SMs. Fig. 3. KBFC overview The KBFC processes four inputs and calculates the dimming levels of the loads and the shifting times for schedulable loads. Fig. 4 shows the membership functions of the inputs and outputs of the KBFC. Fuzzy rules are designed to identify non-priority loads during peak hours and to shift the consumption to the off-peak hours, as well as to reduce the load consumption of loads of the type HVAC (heating, ventilation and air conditioning), through dimming commands. This KBFC does not consider consumer comfort and is only devoted to the energy saving strategy. For consumer comfort consideration, see for instance the DRP proposed in [14], which includes a consumer comfort index in the decision variables prior deciding load shifting and switching.

4 198 Luis I. Minchala-Avila et al. / Energy Procedia 103 ( 2016 ) Fig. 4. (a) MF of the energy consumption. (b) MF of the energy price. (c) MF of the load priority. (d) MF of the consumption time. (e) MF of the dimming command. (f) MF of the shifting load command 3. Simulation setup The simulation scenario developed to test the SM design is shown in Fig. 5. The real-time pricing is simulated with information obtained in [15]. The AMI developed consists of a telecommunications infrastructure, five simulated SMs and one SM implemented in a household prototype. Every simulated household and the prototype contain three types of loads: 1) HVAC loads (dimming loads); 2) programmable loads; and 3) interruptible loads. Fig. 5. Simulated AMI scenario for testing DR capabilities of the SM The benchmark model is developed in the SimPowerSystems of Simulink, which also simulates the EMS that runs the FKBC for every household of the model. There are some limitations in the testing platform as well as the controller design: Delays of the telecommunications network are not considered; There are not any load prediction scheme included in the KBFC (future work); and The climate conditions are not considered for dimming commands calculation.

5 Luis I. Minchala-Avila et al. / Energy Procedia 103 ( 2016 ) Simulation results Figures 6 to 8 show the results of applying the DRP through the designed SM meter in a household prototype (Fig. 6), in a simulated household that includes three types of loads (Fig. 7), and the neighborhood energy consumption that consists of the simulated load plus the load installed in the prototype (Fig. 8). The results show improvement in the energy consumption, since the peak load is reduced and the valleys of the demand curve without the DRP have a better usage due to the shifting of the schedulable loads to these hours of consumption, which have lower rates. The curtailment strategy calculated by the KBFC shows an important improvement in the load consumption of the entire system Normal Consumption Consumption applying DR 5000 Demand (Watts) :00 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time (hours) Fig. 6. Results of energy consumption in the household prototype when the SM implements DR 9000 House Normal consumption Consumption applying DR Demand (Watts) :00 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time (hours) Fig. 7. Results of the response of simulated programmable loads under DR commands x Normal Consumption Consumption applying DR 5 Demand (Watts) :00 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time (hours) Fig. 8. Load demand of the simulated neighbourhood The dimming percentages calculated by the KBFC are in the range of 70% to 95%, but no weather conditions are considered in the calculations, which could have an important impact in the user comfort. On the other hand, the shifting commands applied on the programmable loads allow a better energy usage of the entire system, as it is shown in Fig. 8. The total peak demand reduction is about 10% and the total percentage reduction of the operational energy consumption is in the order of 15%. 5. Conclusions An SM prototype that implements DR commands is presented in this paper. A significant improvement in the energy usage is achieved when the SM implements the DRP designed in the EMS. The peak demand is shifted to the valleys of the consumption curve, improving the system reliability and

6 200 Luis I. Minchala-Avila et al. / Energy Procedia 103 ( 2016 ) generating significant costs savings for the utility and the consumer. The proposed KBFC helps to mitigate excessive consumptions, prioritizing the energy savings. One of the drawbacks of the proposed system is that it does not consider the user comfort, which is important in order to develop a less aggressive load curtailment strategy. The results of the energy savings in the household prototype show an effective implementation of the DR commands by the SM in a domestic house. References [1] P. Palensky and D. Dietrich, Demand side management: Demand response, intelligent energy systems, and smart loads, IEEE Trans. Ind. Informatics, vol. 7, no. 3, pp , [2] P. Siano, Demand response and smart grids A survey, Renew. Sustain. Energy Rev., vol. 30, pp , [3] Smart Grid Legislative and Regulatory Policies and Case Studies, [4] S. Q. Ali, S. D. Maqbool, T. P. I. Ahamed, and N. H. Malik, Load scheduling with maximum demand and time of use pricing for microgrids, in 2013 IEEE Global Humanitarian Technology Conference: South Asia Satellite, GHTC-SAS 2013, 2013, pp [5] S. Zhou, Z. Wu, J. Li, and X. Zhang, Real-time energy control approach for smart home energy management system, Electr. Power Components Syst., vol. 42, no. 3 4, pp , [6] T. Logenthiran, D. Srinivasan, and T. Z. Shun, Demand side management in smart grid using heuristic optimization, IEEE Trans. Smart Grid, vol. 3, no. 3, pp , [7] L. Gelazanskas and K. A. A. Gamage, Demand side management in smart grid: A review and proposals for future direction, Sustain. Cities Soc., vol. 11, pp , [8] J. Zheng, D. W. Gao, and L. Lin, Smart meters in smart grid: An overview, in IEEE Green Technologies Conference, 2013, pp [9] D.-M. Han and J.-H. Lim, Smart home energy management system using IEEE and ZigBee, IEEE Trans. Consum. Electron., vol. 56, no. 3, pp , [10] X. Hao, Y. Wang, C. Wu, S. Member, A. Y. Wang, L. Song, C. Hu, and L. Yu, Smart meter deployment optimization for efficient electrical appliance state monitoring, pp , [11] A. Arif, M. Al-Hussain, N. Al-Mutairi, E. Al-Ammar, Y. Khan, and N. Malik, Experimental study and design of smart energy meter for the smart grid, in 2013 International Renewable and Sustainable Energy Conference (IRSEC), 2013, pp [12] J. Wang and C. Qi, The design of control system for smart meter, in 2012 International Conference on Computer Science and Service System, 2012, pp [13] L. I. Minchala-Avila, L. E. Garza-Castañón, A. Vargas-Martínez, and Y. M. Zhang, A review of optimal control techniques applied to the energy management and control of microgrids, Procedia Comput. Sci., vol. 52, pp , [14] M. M. Rahman, S. Hettiwatte, and S. Gyamfi, An intelligent approach of achieving demand response by fuzzy logic based domestic load management, in 2014 Australasian Universities Power Engineering Conference (AUPEC), 2014, pp [15] Live Prices ComEd s Hourly Pricing Program. [Online]. Available: [Accessed: 14-Mar-2016].