Energy Systems Operational Optimisation Emmanouil (Manolis) - - PowerPoint PPT Presentation

Energy Systems Operational Optimisation

Emmanouil (Manolis) Loukarakis Pierluigi Mancarella

Workshop on Mathematics of Energy Management University of Leeds, 14 June 2016

Overview

2/18 (Overview)

1: Perspective 2: Electrical Distribution Networks Management

3: Other Problems

What’s this presentation about?

What’s the Problem?

3/18 (Perspective) Conventional Generation Inflexible Demand Flexible Demand Large-Scale Renewables Distributed Generation Distributed Storage Large-Scale Storage

Towards increased energy efficiency & reduced emissions…

Distribution medium/low voltage radial networks Transmission high voltage meshed networks

Electrical Distribution Networks Management Resources Coordination Over Time

Heating Hot Water / Other Processes Heat Network CHP Gas Network

Optimising Heat Networks Operation Optimising Gas (or

• ther fuel) Usage

Current State of Play

4/18 (Perspective)

Unit Commitment (every 24h to <1h) Local Device Controls (instant)

Operating status Transmission Network Optimise! Detailed model / bids Large Generators Distribution System

Forecasts

Aggregate model/bids Simplified models Operating state / Control mode / Power set-points Transmission Network Large Generators Large Generators Distribution System Transmission Network Detailed models Detailed models Aggregate measurements Optimise!

Economic Dispatch (every 15min)

real-time

• significant uncertainty
• integer variables
• typically coupled with

reliability requirements

Is this the right time to optimise… … devices at the end-user level? Not really! … other energy vectors? Probably not in detail!

• limited number of discrete controls
• contingency considerations
• limited look-ahead

Is this the right time to optimise… … devices at the end-user level? … other energy vectors? If not now…when?!

Extending Dispatch

5/18 (Perspective)

1 2 3 4 5 6

Area 1 Area 3 Area 2

Transmission Distribution

Large-scale generation (conventional & renewable) Transmission system (multiple areas) Bus Aggregate Demand

IGs TSOs Users Transmission Distribution Microgrid

Large-scale generation (conventional & renewable) Transmission system (multiple areas) Distribution system (high/medium voltage feeders) Distribution system (medium/low voltage feeders) Individual Users (inflexible & flexible demand / small scale renewables)

2 4 6 8 10 12 energy time-step 2 4 6 8 10 12 power time-step infeasible infeasible curtailment curtailment

• Very large scale!
• Uncertainty!
• Peculiarities of individual

devices. Need for one more

• ptimisation step!

A Step Further

6/18 (Perspective)

Unit Commitment (every 24h to <1h) Economic Dispatch (every 15min) Local Device Controls (instant)

Operating status Transmission Network Optimise! Detailed model / bids Large Generators Distribution System

Forecasts

Aggregate model/bids Simplified models Transmission Network Distribution System Large Generators Detailed model / bids OPF Operating state / Control mode / Power set-points Transmission Network Large Generators Optimise! Aggregating function / OPF Microgrids Aggregate models

Forecasts Microgrid (Local) Dispatch (every 1min)

Microgrids Network constraints Users Flexible & inflexible energy

• ffers / requests

Users Optimise! Operating state / Control mode / Power set-points

Disaggregating the network

• perators schedule…

Microgrid Dispatch

7/18 (Distribution Networks Management)

… or in other words: close-to-real-time distribution network management IEEE-123… the good old days IEEE-123… in a test case with lots of EVs if left uncontrolled Controls

• many discrete: tap

changers, capacitor banks, loads …

• some continuous: small-

scale generation, storage, some EVs… Objectives

• follow a given power output (market signal)
• serve customers!
• alleviate constraints violations

Requirements

• solution time …

up to a few minutes

Modelling Considerations (part 1)

8/18 (Distribution Networks Management)

Point 2 Symmetrical components → no advantage in 1p/2p loads Point 1 Return currents not

• f interest → Kron’s

reduction! Point 3 Constant power models → not good enough → go ZIP + VI formulation

Non-linear! Non-convex!

Modelling Considerations (part 2)

9/18 (Distribution Networks Management)

Point 4 If V in polar coordinates the energy balance (right part) is non-linear → use rectangular coordinates! Point 5

• Voltage

constraints → non-convex

Still non-linear!

Modelling Considerations (part 3)

10/18 (Distribution Networks Management)

0.8 0.9 1 1.1 1.2 0.6 0.8 1 voltage (p.u.) real{I} (p.u. cP) non-linear exact curve linear approximation feasibility region

Approximation 2

• Approximate P-part, as

a ZI-part

• 2
• 1

1 2

• 1
• 0.5

0.5 1 real{I} (p.u. Imax) imag{I} (p.u. Imax)

• uter

approximation

Approximation 3

• Imbalance / capacity

bounds → linearize!

Linear (assuming Z part is fixed)!

Modelling Considerations (part 4)

11/18 (Distribution Networks Management)

Approximation 4

• Modified utility function

to prioritise demand

• Follow the market

power reference

Formulation Multi-time-step? stochastic?

650 632 632A 632B 632C 632D 632E 645 646 633 634 671 675 680 684 652 611

Formulation Single-time-step? deterministic?

Point 6

• Do we really need

tight voltage bounds?

Does It Work?

12/18 (Distribution Networks Management)

Time (sec)

IEEE-13

0.0176 0.0031 0.0161 0.0024 0.9945 0.9999 0.17 (0.25)

IEEE-34

0.0056 0.0003 0.0168 0.0006 0.9929 1.0000 0.24 (0.50)

IEEE-37

0.0002 0.0001 0.0005 0.0001 0.9998 1.0000 0.23 (0.43)

IEEE-123

0.0079 0.0012 0.0099 0.0013 0.9964 1.0000 0.33 (1.80)

Approximate problem at a given voltage reference frame Collect info from smart meters Needs adjustment? NO YES Send energy schedules to devices

Algorithm 1

149 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 300 610 250 450

Algorithm 2

Tap-Changers

13/18 (Distribution Networks Management)

Approximate demand & taps at a given voltage reference frame Solve for state and taps (approximate) Collect info from smart meters Needs adjustment? NO YES Send energy schedules to devices Update trust-region Solution time (sec) Iterations

• Max. tap

rounding errors Power change

IEEE-13

0.32 5 0.0029

• 3.78%

IEEE-34

0.89 4 0.0038

• 5.16%

IEEE-37

0.49 4 0.0030

• 8.03%

IEEE-123

2.62 16 0.0031

• 4.42%

Approximation 5

• Taps are

continuous

800 802 806 808 810 812 814 850 816 818 820 822 824 826 828 830 854 856 852 832 888 890 858 864 834 860 836 840 862 838 842 844 846 848

Algorithm 3

Discrete Controls

14/18 (Distribution Networks Management)

Approximate demand & taps at a given voltage reference frame Solve for state and taps (approx. continuous relaxation) Collect info from smart meters Needs adjustment? Is integral? NO YES YES NO Send energy schedules to devices Update trust-region Adjust penalty Tap controls number Added EVs number Solution time (sec)

IEEE-13

3 592 7.7

IEEE-34

9 316 4.3

IEEE-37

3 409 4.9

IEEE-123

9 623 14.9 Approximation 5

• Solve continuous

relaxation → restricting deviations from nearest integral solution

Mixed integer programming… An feasible integral solution was recovered… Due to high number of small controls… no significant difference between the continuous relaxation objective value…

Summing Up

15/18 (Distribution Networks Management)

months/ years ahead minutes / hours ahead min. ahead Model detail Uncertainty Model detail Uncertainty Model detail Uncertainty sec. ahead real-time Model detail Not now!

There are more problems out there!

How should we solve them? Important!

• Problem characteristics!
• Solver characteristics!

Another Problem : energy district management (1)

16/18 (Other Challenges)

from gas supply network renewables building heating CHP

• ther

electrical demand hot water heat exchanger from / to electricity supply network pump storage

BUILDING HEAT / POWER GENERATION INSTALLATION OTHER BUILDINGS / INSTALLATIONS

boiler gas network heat network electrical network

• ther gas

demand

Computational difficulties

• thermal network storage

capacity

• thermal network dynamics

The problem… …optimising over time subject to network constraints and detailed device and building models

Another Problem : energy district management (2)

17/18 (Other Challenges)

The Manchester University test case…

electricity gas heat

Solution Fast enough? Reliable enough?

Another Problem : distributed optimisation applications

18/18 (Other Challenges)

Solving very large scale problems… Getting closer to control…

IG … IG 1 TSO 2 TSO 1 TSO 3 Users 2 Users 4 Users 1 Users 3 Users 5 Users 6

1 2 3 4 5 6

Area 1 Area 3 Area 2 IG … IG 1

Power System Decomposition Optimization Problem Structure

Transmission Distribution

Large-scale generation (conventional & renewable) Transmission system (multiple areas) Bus Aggregate Demand

IGs TSOs Users

Agent/subproblem representing users Large Generator subproblem Network Operator subproblem

200 400 600 800 20 40 60 iteration marginal price (m.u./MWh)

<10-3 <10-4 <10-2 <10-1

Thank you for your attention… …Questions?