ITU Kaleidoscope 2014 Living in a converged world -­‐ impossible without standards? Distributed Demand-Side Management with Load Uncertainty Emmanuel Manasseh Hiroshima University, JAPAN manasejc@ieee.org Saint Petersburg, Russian FederaEon OUTLINES v Introduction v Smart Grid & Demand Side Management v Energy Consumption Schedule v System Model & Optimization Problem v Simulation results v Conclusion Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? What is Smart Grid? A smart grid is a modernized electrical grid that uses information and communications technology to gather and act on information in an automated fashion to improve the efficiency, reliability, economics, and sustainability of the production and distribution of electricity. v v Electric grid modernization involves, 1. Improvement of infrastructure 2. Addition of the digital layer 3. Business process transformation Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? How is it different from the existing Electricity Grid ? Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? What is Demand side Management ? v DSM is the modification of consumer’s demand of electricity through various methods such as financial incentives and consumer education v A process by which electrical utilities in collaboration with consumers, achieve predictable and sustainable change in electricity demand v Usually the goal of DSM is to encourage the consumers to use less energy during peak hours or to move the time of energy use to the off-peak hours Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Energy Consumption Schedule (ECS) v The ECS devices are built inside the smart meters v Smart meters are connected to communication networks and power grid. v The aim is to reduce the total energy cost as well as the peakto average-ratio (PAR) v Two approaches for energy consumption management in buildings: Reducing consumption and Shifting Consumption v Deployment of storage and ECS devices for DSM Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? System Model I v We consider N users served by one utility company. v Each user has a smart meter with ECS, that communicates with his/her devices and the utility company through the advanced metering infrastructure (AMI) v Each user n ∈ N, is considered to have non-adjustable loads, adjustable load and a storage device. v Kn→ a set of adjustable load appliances: PHEVs, air conditioner, etc. v For each device k 2 Kn , we define energy consumption Scheduling vector 1 T xn,k , [xn,k , . . . , xn,k ] Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? t x v n,k System Model II → one-hour energy consumption scheduled for device k from subscriber n v Non-adjustable load at slot t is t xn,0 v User n is also having a storage device v Let pnt 0 , t = 1 , . . . , T , be the energy available in the battery at the end of slot t , and pnmax the capacity of the battery. 0 p v The available energy at the beginning of the horizon is n v The battery can be either charged or discharged during slot t Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? System Model III t v bn → the energy drawn from or provided to the battery. v bn t < 0 → discharging & bnt > 0 → charging. v The energy stored in the battery is given by ptn = ptn 1 + btn , t = 1, . . . , T. t b v n is constrained by the maximum charge/ discharge b dis < b t < b ch n n n v The battery-supply energy is no more than the P t t consumption, bnt + k xn,k + pn,0 0 v Battery has efficiency ⌘n 2 (0 , 1 ), the discharge at slot is t limited by t t 1 bn ⌘ n pn Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? System Model IV v The total hourly energy consumption for n ∈ N is lnt , X xtn,k + xtn,0 + btn . k2Kn v For the adjustable loads, each user selects the time interval [↵n,k , n,k ] that the energy consumption for device k is valid to be scheduled v Total energy consumption for device k from user n X is t n,k En,k = xn,k t=↵n,k t xn,k = 0 , 8t 2 / {↵n,k , . . . , n,k }) Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Motivation How to manage energy consumption of adjustable appliances in order to reduce PAR and minimize the cost. Formulation of Optimization Problem v Devices may have some maximum power levels min well as the minimum power level n,k max n,k , as v The upper-bound and lower constraints on the ECS vector xn,k for each device min n,k xtn,k max n,k , 8t 2 {↵n,k , . . . n,k }. v The total load of N residential users at each hour of the X Lt = lnt day is n2N Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Cost of Electricity v C t (·) → the cost of electricity over a slot t v The cost that the utility incurs to provide electricity to the end user v The cost of the same load may be different at different time of the day v The multi-residential load control task amounts to minimizing the total cost of electricity Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? min x1 ,...,xN Optimization Problem !! T X X X Ct btn + xn,k + xtn,0 t=1 subject to btn + X n2N xtn,k + xtn,0 k2K 0, k n,k X xtn,k = En,k , t=↵n,k min t max x 8t 2 {↵n,k , . . . , n,k }, n,k n,k n,k , xtn,k = 0, 8t 2 / {↵n,k , . . . , n,k }, t ch bdis < b < b t = 1, . . . , T, n n n , ptn = ptn 1 + btn , t = 1, . . . , T ⌘n ptn 1 . btn Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Optimization Problem v The centralized fashion can be used to solve the problem v Central unit performs the optimization v It requires each user to provide detailed information about his/her energy storage capabilities as well as the energy consumption of devices. Private issues may discourage users to subscribe to the optimization process if Centralized fashion is adopted v We proposed a distributed fashion. Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Simulation Parameters v N = 20, each having random adjustable and non-adjustable devices between 10 to 20 v Non-adjustable load appliances include electric bulbs, TVs, refrigerators etc. v These are loads whose instantaneous power or starting time cannot be adjusted. v Adjustable load refers to the loads whose instantaneous power, starting time or both can be adjusted. v Adjustable load appliances includes PHEVs, dish washers, washing machines etc. v Users with and without storage devices are considered. Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Energy Consumption (kWh) No Energy consumption Schedule ECS (a) 80 60 40 20 0 0 5 10 15 20 25 15 20 25 Time−slot t (b) Cost(Dollar) 15 10 5 0 0 5 10 Time−slot t Users without ECS, total cost $112.62 Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Energy Consumption (kWh) Energy consumption Schedule ECS 60 40 20 0 0 5 10 15 20 25 15 20 25 Time−slot t Cost(Dollar) 6 4 2 0 0 5 10 Time−slot t The ECS for users without storage devices, total cost $102.72 Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Aggregate load (kWh) (a) 60 40 20 0 1 8 16 24 16 24 16 24 Storage strategy (kWh) Time−slot t (b) 20 10 0 −10 1 8 Time−slot t (c) Cost(Dollar) 6 4 2 0 1 8 Time−slot t Users deploying ECS and storage devices, total cost $92.54 Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? 8 User with No ECS User without Storage dev. User with Storage dev. Daily Cost per User (Dollar) 7 6 5 4 3 2 1 0 0 5 10 User n 15 20 Comparison of cost paid by each user with/without storage devices Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Conclusion v We considered deployment of ECS devices in smart meters for DSM v Difference in pricing mechanisms employed by utility companies gives incentive for users to trade the energy. v Simulation results show significant reduction in cost for the ECS scheme over the system without ECS v Energy Storage devices play an important role in reducing the cost paid by the consumers. Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards? Acknowledgement Special thanks to Tanzania Communication Regulatory Authority (TCRA) for their generous donation of the travel costs. Saint Petersburg, Russian Federation, 3-5 June 2014 ITU Kaleidoscope 2014 - Living in a converged world - impossible without standards?