1 An Introduction and Overview of Technology Damien Coyle

2 Agenda IntroductionTechnology Why Smart Street?

3 Connecting the North West 4.9 million 25 terawatt hours 2.4 million £8 billion of network assets

4 Project overview £8.4m from LCNF, £1.5m from Kelvatek, £1m from ENW Started in Jan 2014 and finishes in Dec 2017 Facilitates quicker cheaper connection of domestic LCTs Trials period Sep 2015 – Aug 2017 £11.5m, 4 year innovation project Smart Street project overview

5 Project partners

6 Smart Street trial areas 6 primary substations 11 HV circuits Wigan & Leigh Manchester Wigton & Egremont 38 distribution substations 163 LV circuits Around 62,000 customers

7 Smart Street trial design Two years One week on One week off Five trial techniques LV voltage control LV network management and interconnection HV voltage control HV network management and interconnection Network configuration and voltage optimisation One year’s worth of data To be designed to avoid placebo affect Five trial regimes to test full effects

8 LV capacitors in street furniture 80 LV capacitors One on each closed ring Tried and tested high spec

9 HV capacitors 4 pole mounted HV capacitors Installed similar to pole mounted transformers 4 ground mounted HV capacitors Housed in containers but not on street

10 Weezap & Lynx 240 LYNX Installed in 80 LV link boxes 489 Weezaps Fitted across 163 LV Circuits

11 Existing radial network Network limitations Diversity between feeders is untapped Fuses unable to cope with cold load pick up Customer impact Customers’ needs invisible to the network Demand and generation levels limited by passive voltage control systems Reliability driven by fix on fail

12 Drift range Voltage profile Historic networks have no active voltage regulation Normal voltage range

13 Problem - LCTs create network issues LCTs rapidly surpass voltage and thermal network capacity Drift range

14 Smart Street – the first intervention Voltage stabilised across the load range Power flows optimised Low cost Quick fit Minimal disruption Low carbon Low loss Invisible to customers W C L W

15 Network reliability improvement Builds on C 2 C and CLASS Storage compatible Transferable solutions C2CC2C Capacity to Customers C Capacitor W WEEZAP L LYNX CLASSC2CC2CLCCCCC2CC2CC2CC2CLWW Spectrum TC On-load tap changer TCW

16 How much could customers save? GB Smart Street benefits Reduced energy consumption, 2013 (from CVR ≈ 3 - 7%) £15 - £30 pa£390 - £780m pa Maximise DG output (from maximising Feed In Tariff income)£70 pa£20m pa Reinforcement savings via DUoS£8.6b over 25 years£330 over 25 years Efficient network solutions Energy savings Carbon benefits Now we can stabilise voltage Reduced demand Reduced customer energy consumption Maximised DG output We can set the voltage level lower This will lead to:

17 Smart Street summary Lower energy bills More reliable supply Reinforcement savings Benefit Faster LCT adoption Less embedded carbon Re-usable technology Optimise energy and losses Carbon Footprint First example of CVR First example of centrally controlled LV network Range of intervention solutions Low Risk Combine into one end- to-end system Network optimisation Challenge

18 & QUESTIONS ANSWERS

19 Want to know more? Thank you for your time and attention linkedin.com/company/electricity-north-west facebook.com/ElectricityNorthWest youtube.com/ElectricityNorthWest e

20 Smart Grids and Community Energy Cara Blockley Low Carbon Projects Manager

21 Our smart grid programme £30 million Deliver value from existing assets Leading work on developing smart solutions Three flagship products Capacity to Customers

22 Agenda Supporting community energy Community projects

23 Community projects

24 24 hours Demand Power Saver Challenge Pilot project to look at ways of reducing ‘peak demand’

25 Community engagement Promote the project Incentivised involvement Benefited community groups Community groups build trust Customers will change behaviour Stronger, cohesive communities Help our engineers and customer facing employees What we’ve doneWhat we’ve learnt What we want to achieve

26 Supporting community energy

27 NEDO smart community project £20m smart community project led by Japan’s New Energy Development Organisation (NEDO) 600 electric and gas hybrid heat pumps installed in social housing properties in Wigan and Greater Manchester, some with PV Working with Electricity North West and Wigan & Leigh social housing project Heat pumps and information and communication technologies (ICT) aim to reduce carbon and help provide a demand response Three-year demonstration phase running from this April 2014 to the end of March 2017

28 Problem - LCTs create network issues LCTs rapidly surpass voltage and thermal network capacity Drift range

29 Smart Street – the first intervention Voltage stabilised across the load range Power flows optimised Low cost Quick fit Minimal disruption Low carbon Low loss Invisible to customers W C L W

30 Now we can stabilise voltage Reduced demand Reduced customer energy consumption Maximised DG output This will lead to: Smart Street benefits Efficient network solutions Energy savings Carbon benefits We can set the voltage level lower

31 Summary Lower energy bills, more reliable supply, connection savings Reduction in greenhouse gas emissions achieved through community energy schemes Community energy schemes best supported with trusted Partners Visibility and automation to provide networks responsive to customers’ needs

32 Want to know more? Thank you for your time and attention linkedin.com/company/electricity-north-west facebook.com/ElectricityNorthWest youtube.com/ElectricityNorthWest e

33 Electricity North West’s Demand Response demonstration Simon Brooke Low Carbon Projects Manager

34 Electricity North West’s innovation strategy Delivering value to customers Maximise use of existing assets Innovative solutions to real problems Proven technology deployable today Generate value for customers now Offer new services and choice for the future

35 Our smart grid development C 2 C, CLASS and Smart Street demonstrate demand response Deliver value from existing assets Leading work on developing smart solutions Capacity to Customers Three flagship products £30 million Customer choice

36 What is Capacity to Customers? Combines proven technology and new commercial contracts Capacity to Customers Utilised capacity Innovative demand side response contracts New commercial contracts Remote control equipment on HV circuit and close the NOP Technical innovation Latent capacity Current demand Releases significant network capacity Facilitates connection of new demand and generation without reinforcement Effectively doubles the available capacity of the circuit Enhanced network management software Allow us to control customer’s consumption on a circuit at the time of fault

37 Customers Active network management Efficiency Demand reduction Key hypotheses Existing or new customers can directly benefit financially by providing the demand response Defers/ optimises reinforcement and reduces carbon intensity Creates a post fault demand response capability Network automation creates self healing capability and facilitates capacity release

38 Contract arrangements NTCDCUSA Managed connection agreement Construction & installation agreement New customers Existing customers Demand and generation Contract

39 Contract arrangements Share contracts early and support customer through discussions Direct relationship with I&C customer for ‘value’ discussion Early lessons Works best with one point of contact - a customer relationship manager for BaU

40 Demand response results (EXISTING) Size, sector and price of DR from existing customers

41 Demand response results (EXISTING) Early lessons Post fault response is attractive to customers and Electricity North West Wide range of trial participants, appears most favourable to small manufacturers Very attractive to multiple site operators

42 Utilities IT Manufacturing Transportation Demand response results (NEW) New connection customers' managed capacity, kVA by sector

43 Demand response results (NEW) Early lessons Good range of enduring post fault DR capacities Post fault DR can operate in with other DR programmes New DR predominantly from small manufacturers again

44 Project benefits summary Releases network capacity for use by customers’ LCTs Creates post fault demand response market which is less intrusive to customers Minimises carbon- intensive infrastructure Can defer reinforcement costs and the time taken to complete the associated works Will better exploit existing assets, thus cost-effective and quickly implemented Full set of results and learning from Capacity to Customers will be included in closedown report with dissemination events planned starting early 2015 Develops new DR market Carbon reduction Cost deferral £ Reinforcement deferral Rapidly deployable solution

45 Want to know more? Thank you for your time and attention linkedin.com/company/electricity-north-west facebook.com/ElectricityNorthWest youtube.com/ElectricityNorthWest e

46 Low Voltage Network Solutions Breakout LV Network Management Dr Rita Shaw

47 LV Network Solutions Our largest Tier 1 LCN Fund project £1.5 million £ Modelling and analysis

48 Project scope Text Model LV networks, identifying LCT impacts and solutions Assess monitored LV network performance Improve LV assessment and policy for all network Monitor 200 LV substations and feeders To understand our LV networks now and in future scenarios

49 LV monitoring deployment Challenge Prepare for data capture Roll out to site - 28 pole mounted and 172 ground Train installation crews Develop installation procedures Determine monitoring requirements without customer interruptions Site selection / surveys Prepare functional specifications Tender and procure equipment £

50 Monitoring equipment 2012 UK Energy Innovation award for the ‘Best Smart Grid Technology’ GridKey monitoring equipment at 100 substations

51 Monitoring equipment Nortech monitoring equipment at 100 substations

52 Communications approach Monitoring unit fitted with SIM card Assigned private, static IP address Time stamped data logs created every 1 – 10 minutes GPRS /3 G 1 set of Rogowski coils fitted per LV way 3 phases and neutral measured DPN3 Protocol between iHost and monitor Unsolicited event reporting transfers data logs in near real time iHost server at Electricity North West consists of communication modules, databases and web user interface Export produces CSV files to be used by the University of Manchester

53 LV monitoring – outcomes 10,000 days of good 10-minute data At transformer and head of each feeder, per phase + neutral Value of monitoring within LVNS Performance evaluation of monitored LV networks’ Review / improve load estimates for whole network Validation of network models Monitoring used in other innovation projects and BAU Challenging but achieved!

54 Also... LV feeder midpoint monitoring 100 midpoints and 100 endpoints outside LVNS project Smart joint technique developed by us

55 University of Manchester’s inputs (1) Challenge: Transforming data from GIS into a power flow engine format Build network models Real LV networks = big step forward Three-phase four-wire power flow (OpenDSS) Using our GIS + MPAN data + impedances

56 University of Manchester’s inputs (2) Create diverse sets of load and generation profiles Created pools of 1000 domestic load / PV /HP / EV/ microCHP profiles as inputs to Monte Carlo analysis Reflect uncertainty of impacts by picking from pool 24 hours – 5 min resolution

57 Random allocation for each customer node Loads Random allocation of sites and sizes PV Time Series Simulation 3 Phase four wire power flow Power Flow This process is repeated 100 times for each feeder and penetration level (% of houses with PV panels). Impact Assessment: 1. Customers with voltage problems 2. Utilization level 3. Energy losses, etc Impact Assessment: 1. Customers with voltage problems 2. Utilization level 3. Energy losses, etc Results Storage Monte Carlo Impact Assessment Method

58 Voltage analysis for one feeder % customers with PV % customers with voltage problems Against BS EN50160 Eg 95% of the 10 min mean rms values within +/- 10% Against BS EN50160 Eg 95% of the 10 min mean rms values within +/- 10%

59 Assuming balance understates issues PV penetration (%) Customers (% )

60 30 min resolution understates problems 1-15 min resolution min resolution PV penetration (%) Customers (% )

61 Multi-feeder analysis - What is the hosting capacity for LCTs? 1/3 of feeders would have no problem with any PV uptake level Detailed Monte Carlo analysis of 128 LV underground feeders Often our feeders can accommodate lots of LCT without thermal or voltage issues Graph shows issues on feeders with >25 customers

62 If there is a problem... Voltage or thermal first? Voltage problems before thermal problems Thermal problems before voltage problems

63 But if there is a problem.... % Customers with PV Number of cases Can we predict for a particular feeder? Utilisation? Length? Impedance? Customer numbers? Can we predict for a particular feeder? Utilisation? Length? Impedance? Customer numbers? First voltage problem occurs at wide variety of % PV uptake

64 Scatter even on the best metrics Combined network metric = (initial utilisation x total path impedance) DNO friendly network metric = (customer numbers x total length) % Customers with PV Many feeders present no problems even at 100% uptake

65 Connect and manage/monitor for PV Our network can often accept lots of LCTs, so let customers connect quickly! But hosting capacity is variable and difficult to predict, so monitor to check voltage etc The analysis can only broadly suggest when problems will occur, so monitor early! Eg on average no problems until around 45 PV on feeder, but we monitor at ~20 PV systems

66 Analysis of solutions Network reinforcements On-load tapchangers in LV networks Loop connection of LV feeders Link to ‘Smart Street’ project % of customers with voltage problems PV penetration (%) Customers (% )

67 Hosting capacity of underground LV networks for LCTs Potential network solutions, with implications for future DNO policy A (rough) future capacity headroom model for whole LV+ HV network In detail for monitored networks Improved load estimates for whole LV+HV network Products + procedures What /when/ where to monitor in future What we have learnt How our LV performs now How to monitor at LV How our LV network will perform with LCTs

68 Why are we doing this Drive value for our customers Leverage learning to support business

69 & QUESTIONS ANSWERS

70 Want to know more? Thank you for your time and attention linkedin.com/company/electricity-north-west facebook.com/ElectricityNorthWest youtube.com/ElectricityNorthWest e e

71 Network Management: Centralised or Distributed? Dr Geraldine Bryson Future Networks Technical Manager

72 What does it mean? Applications layer SCADA layer NMS layer C Centralised Distributed Communications layer

73 Distributed – historic Operate based on setting Limited communications requirements Report when operated Substation Voltage control Protection relays

74 Distributed – BaU today Increased communications requirements Instructions sent from control engineer or NMS Substation Voltage control Protection relays Remote control

75 Distributed – smart LoVIA – Low Voltage Integrated Automation Sensors on network Distribution substation Voltage control Protection relays Remote control

76 Distributed management Not communication reliant Pros Improvement in performance – no latency No integration with NMS No protocol issues Only localised control Network awareness can be expensive High cost to maintain local systems Cons

77 Communications Hard wired to strategic sites Controllable devices at 132kV and 33kV Unreliable to remote sites Reliability and resilience improved Driven by increased use of mobile devices Controllable devices at 11kV Controllable devices at LV Smart meters Increasingly reliable communications HistoricalTodayFuture

78 Centralised - historic Mainly at higher voltages Central system to show operational status Operations commanded by control engineer SCADA Centralised NMS

79 Centralised - today Operations commanded by application Fault restoration algorithms Knowledge of network topology Centralised Utilises remote control at distribution substations SCADA NMS ARS

80 Centralised – smart ARSC2CC2CCLASS Smart Street Others Applies algorithms to wide area of network Data from remote sensors Knowledge of network Centralised Only central logic to be kept updated SCADA NMS

81 Centralised management Offers control over a wider area Pros Optimise across a number of apps Network aware at lower cost Lower cost to maintain central system Heavily reliant on communications Needs local fail safe mechanism Cons

82 Conclusions (1) Increased monitoring of network Need to distinguish data from information Data to app for processing Increasingly more reliable Increased use of controllable devices More forms of communication Centralised with reliable comms – way forward Works in other industries Deterministic or iterative SensorsAlgorithms Communications

83 Conclusions (2) ENWL have both centralised and distributed Both have roles in smart grids - application dependent Centralised solves more at lower cost Vendors have basic building blocks Need exact requirements from Industry ENA active network management group UK need to influence EU standards EU standards influence vendors Need BaU industry standards Distributed require repeated investment to maintain Standards Deployment Availability

84 Want to know more? Thank you for your time and attention linkedin.com/company/electricity-north-west facebook.com/ElectricityNorthWest youtube.com/ElectricityNorthWest e e