Yong Zhou 2012-03-08

1. Management and Control of Domestic Smart Grid TechnologyIEEE Transactions on Smart Grid, Sep. 2010Albert Molderink, Vincent Bakker Yong Zhou 2012-03-08

2. Problems • Electricity usage is increasing and becoming fluctuating; • Decreasing the generation efficiency • Renewable resources are fluctuating and uncontrollable; • Keeping demand and supply balance causes higher fluctuation • Charging of electrical cars are uncontrollable; • Resulting in higher peak demands

3. Solution • Transforming domestic customers from static consumers into active participants in the production process • Consider the appliances with controllable load, micro-generation, domestic energy storage of both heat and electricity • Shift electricity consumption in time without harming the comfort of the residents • Optimization potential comes from the scheduling freedom, which is limited by the comfort and technical constraints.

4. Goal and Functionality • Goal • Optimize efficiency of current power plants; • Support the introduction of a large penetration level of renewable sources; • Optimize the usage of the current grid capacity; • Functionality • Control the domestic generation and buffering to reshape the energy profile for efficient energy utilization; • The required heat and electricity supply and the comfort for the residents should be guaranteed;

5. Model • Micro-generator • Heat/electricity buffers • Appliances • Local controller

6. Model (Cont’d) • Multiple houses are combined into a microgrid, exchanging electricity and information between the houses; • Electricity can be imported from and exported into the grid; • Heat is produced, stored, and used only within the house; • The planning horizon is discretisized resulting in a set of consecutive time intervals • 6-min time interval is often used since such in interval length is a good tradeoff between accuracy and amount of data.

7. Methodology • Step 1: predict the production and consumption pattern locally for all appliances for the upcoming day; • Step 2: central planner uses optimization potentials to exploit the potential to reach a global objective; • Step 3: real-time control algorithm.

8. Step 1: Local Prediction • Prediction model: • Neural network technique • Predict the heat profile for the next day as accuracy as possible • Predict the heat demand per house based on recent observations • Input: historical heat demand, whether (wind speeds, outdoor temperatures), and house characteristics.

9. Step2: Global Planning • The planning focuses on a large fleet of houses combined into a VPP, all equipped with a microCHP and heat buffer. • Based on the heat demand prediction for a single house, this paper plans the runs of the corresponding microCHP. • The planning should consider the constraints of heat demand of the house and the technical constraints. • This paper considers the generator in a group of houses as a VPP, so the global constraints on the total electricity production should be considered.

10. Step2: Global Planning (Cont’d) • The planning horizon of a single day is divided into Nt intervals for which a decision must be made for each microCHP in each house. • Production plan: • is the produced electricity in house n during time period j.

11. Step2: Global Planning (Cont’d) • Min • s.t. (1) (2) (3) (4)

12. Step2: Global Planning (Cont’d) • This optimization problem is NP-complete, so the authors proposed a heuristic method to separate the two elements: • Finding a local plan satisfying local constraints • Dynamic programming approach • Minimizing the squared mismatch from the global production plan P. • By collecting all local production plans, the squared mismatch can be obtained. • Minimizing the mismatch by iteratively steering the local production plans in a mismatch-reduction direction.

13. Step3: Local Scheduling • Goal: optimize the electricity import/export while guarantee the comfort for the residents and proper usage of devices. • The demand is given as an input parameter and can be matched with: • Import from the grid • Production by generators • Buffers • Switching off consumers • When the sum of the four possibilities gives more heat and electricity than the demand, the corresponding energy flows to a buffer or gird.

14. Step3: Local Scheduling (Cont’d) • The algorithm is executed for each time interval and the matching cost for each device is determined at the beginning of the time interval, based on the current status of the device. • Decision variable Xi means the amount of matching of device • The value of Xi is the amount of electricity can be imported/exported • Hi and Ei are the multiplication factors of heat and electricity. • For each device i, a set Si of intervals is specified and the variable Xi is allowed to take only values from one of these disjoint intervals.

15. Step3: Local Scheduling (Cont’d) • Each interval specifies a uniform area for the variable Xi, in the sense that the costs associated with can be expressed by . expresses the matching costs and is the startup costs if Xi is chosen from the interval . • and are heat and electricity demand, respectively.

16. Step3: Local Scheduling (Cont’d)

17. Case Studies • The first case is a simulation of 39 houses using real heat demand data and real prediction to verify whether it is possible to make a planning based on prediction. It is also verified how well the actual scheduler follows the planning.

18. Case Studies (Cont’d) • The second case study is a test with a single house prototype to verify whether the methodology is also applicable in a real-world situation.

19. Conclusions • The three-step methodology proposed in this paper using a hierarchical plan is a scalable solution with limited communication requirements. • The local prediction and scheduler result in a generic solution supporting different technologies and houses with different optimization potential.