INTELLIGENT DC MICROGRID WITH SMART GRID COMMUNICATIONS: CONTROL STRATEGY CONSIDERATION AND DESIGN Presented by: Amit Kumar Tamang, PhD Student Smart Grid Research Group-BBCR aktamang@uwaterloo.ca 1 Supervisor : Prof. Weihua Zhuang 9 January, 2013
MAIN REFERENCE Wang, B. C.; Sechilariu, M.; Locment, F.;, "Intelligent DC Microgrid With Smart Grid Communications: Control Strategy Consideration and Design," Smart Grid, IEEE Transactions on, vol.3, no.4, pp.2148-2156, Dec. 2012 2
OUTLINE Introduction System Overview Power Subsystem Behavior Operation Layer Control Strategy Supervision Upper Layer Design Discussion Conclusion 3
INTRODUCTION (1/5) Smart Grid: Modern Electricity grid capable of bidirectional power and information flow. ( Power+Information+Communication) Complex Network with Randomness and Nonlinearity. Issues: Distributed Generation, Demand Response and Load Control, Energy Storage, Anticipated Massive Amount of Energy Transaction 4
INTRODUCTION (2/5) MicroGrid Localized Grouping of Electricity Sources (Wind, Photovoltaic etc) and Loads. Can Work with/without Traditional Centralized Grid. DC MicroGrid Avoids DC to AC & AC to DC conversion preventing energy loss due to conversions. Smart Grid + Microgrid = Provision of Injecting energy to or getting energy from Utility Grid. Undesirable Injection = Fluctuation in grid power. 5
INTRODUCTION (3/5):ISSUES Renewable Power Generation: Intermittent and Random nature. Uncontrolled injection increases the power mismatching in utility grid and fluctuation in voltage and frequency. Storage system to combat intermittent energy production. Lead Acid batteries for storage. Limited storage capacity, energy management to optimize the use of renewable energy for high penetration. Grid Need (Injection) & Availability(Peak load shaving) 6
INTRODUCTION (4/5) Fig: Possible Smart Grid Topology Small Scale, Middle Scale & Large Scale = traditional Grid to Smart Grid Microgrid Controller: Power Balancing & Load Management. 7
INTRODUCTION (5/5) : OBJECTIVE Power Balance With Load shedding & PV Constrained Production Focus Design Control Strategies Better DC Microgrid Integration USING Intelligent Multi-layer Supervision Interaction with Smart Grid, End user Predictions & Energy Management. Avoid Undesirable Injection Mitigates Fluctuation in Grid Power Reduces Grid Peak Consumption 8
DC MICROGRID SYSTEM :OVERVIEW Power Balancing with Load Shedding, PV Constrained control, w.r.t power limits(utility Grid) Improve Energy Efficiency And reducing energy cost. Predicts load consumption And renewable energy production End user sets some criteria Fig: DC Microgrid System Overview 9
DC MICROGRID SYSTEM POWER SUBSYSTEM BEHAVIOR (1/6): Elements: Grid, PV, Storage, Load Each element modeled by MATLAB Stateflow Simulating event-driven systems based on finite state machine theory. Symbols: PG = Grid Power; PS = Storage Power P_G_S_lim, P_G_I_lim = Grid power supply and injection limits P L = Load Power; Ppv = PV Array Power 10
DC Microgrid System POWER SUBSYSTEM BEHAVIOR (2/6): P G & P S controlled by corresponding reference current ig* and is*. Power reference p* = output of controller for stabilizing dc bus voltage. C I Integral Gain Cp Proportional Gain = distribution Coefficient [0,1] i.e energy storage not injected into Grid 11
DC MICROGRID SYSTEM POWER SUBSYSTEM BEHAVIOR (3/6) : Fig : Grid Behavior State flow model Maximum Load Power PV A peak power production 12
DC MICROGRID SYSTEM POWER SUBSYSTEM BEHAVIOR (4/6): Fig: Storage Behavior Stateflow model SOC = State of Charge in storage. 13
DC MICROGRID SYSTEM POWER SUBSYSTEM BEHAVIOR(5/6) : Fig: PV source Behavior Stateflow model MPPT = Maximum Power g = Solar irradiation (W/m 2 ) Point Tracker. GMIN = Minimum irradiation threshold 14
DC MICROGRID SYSTEM POWER SUBSYSTEM BEHAVIOR (6/6): Fig: Load Behavior Stateflow Model KL = load power limit controlling coefficient (Load shedding) [0,1] 15
DC Microgrid System :OPERATION LAYER CONTROL STRATEGY Energy Flow in power subsystem Controlled by variables: K D, P _G_S_lim & P _G_I_lim (Smart Grid Messages), P* PV_lim, K L Algorithm calculates: P* G, P* S, P* PV_LIM w.r.t limitations and gives Value of K L Fig: Power Control Algorithm 16
DC Microgrid System :SUPERVISION UPPER LAYERS DESIGN Interface variable: upper layer controls Lower layer Physical Parameters from Different fields Relates Different Time scales Fig: Supervision Hierarchical Structure 17
DISCUSSION This work mainly focuses on Power control Algorithm. (Operation Layer) Provides Basic Idea for other layers. Like Optimized value of K D from Energy Management layer is untouched. Mechanism of Prediction of load and PV power is untouched. (Considered as future work) K D needs low speed communication (in range of Minutes). But what Communication infrastructure would be appropriate?? 18
CONCLUSION An intelligent comprehensive DC Micro grid with multi layer supervision was suggested. Supervision exchange data with smart grid, interact with End user, predicts load & PV production and manages energy cost. DC micro grid control design avoiding undesired power injection, Mitigating fluctuation in grid power and reducing grid peak consumption Was proposed. Apparently, supervision interface reduced the negative impact of renewable Sources to grid with better seamless integration to grid. 19
Thank you! 20