grid a -1) te (geo -t...2016/09/18 · slide 2 2016 tion, (geo -t (t6a -1) i will discuss the...

Update (Geo-) Thermal Smart

Grid Mijnwater Heerlen (T6a-1)Strasbourg, 22th September 2016

MSc. René Verhoeven ([email protected])

Cluster and Conceptual Design Manager

Mijnwater B.V. Heerlen, the Netherlands

Thank you chairman for the nice introduction. (And indeed) As winner of the European Geothermal Innovation Award 2015, we are very honoured to have the opportunity to give you an update of the Minewater developments in Heerlen.

September 2016

Status quo, New developments, Operation,

Performance and Lessons learned

Update (Geo-) Thermal Smart

Grid Mijnwater Heerlen (T6a-1)

I will discuss the current status, new developments and some cases concerning operation, performance and lessons learned.

HLN1HLN3

HH1

Mijnwater 2.0Clusters of buildings

September 2016

CLUSTER B

CBS-Maankwartier

CLUSTER D

Componenta-Otterveurdt

CLUSTER C

Weller HHC

CLUSTER A

Arcus-APG

CO2-reduction

65 %

Hydraulic cloud grid

Instant heat/cold exchange

Minewater as storage

Fully demand driven

Bidirectional wells (2017)

Multiple sources

All electric (100% HP)

HLN2

HH2

First, a brief summary of the innovative Minewater 2.0 thermal smart grid as presented the first time at the EGC 2013 in Pisa and shown in this geographical overview: • A low temperature cloud structured exchange grid with:

• Decentralize all-electric energystations in the connected

buildings and • Instant heat and cold exchange between buildings

through local cluster grids and between these cluster grids through the existing mine water grid;

• The minewater is used as a storage instead of a source to prevent depletion and improve capacity;

• The system is: • Fully automatic and demand driven; • With bidirectional minewater wells, to be ready in 2017; • And it is suited for the application of multiple sources.

The system is in operation now for more than 3 years. It works very well as designed, some parts even better than expected, others initially failed and needed additional tuning, improvement or mitigation.

HLN1HLN3

HH1

Mijnwater 2.0Current status

CLUSTER B

CBS-Maankwartier

CLUSTER D

Componenta-Otterveurdt

CLUSTER C

Weller HHC

CLUSTER A

Arcus-APG

HLN2

HH2

Ltd since November 2013

Mijnwater owner/operator

grid & energy stations

Part of PALET (Carbon

neutral 2040)

175.000 m2 connected

30 M€ invested

Heat 4,4 MW; Cold 4,2 MW

September 2016

Some facts about the current status: • Minewater became a private company in November 2013,

fully owned by the municipality of Heerlen. • Minewater has made a shift form just being owner and

operator of the grid to additional ownership and operation of

the decentralized energy stations; • Minewater became part of the Parkstad Limburg Energy

Transition, targeting to become carbon neutral in 2040, Minewater is the core initiative to achieve this.

• Today 175.000 m2 of buildings are connected to the grid, aiming for 500.000 m2 in 2017.

• 30 Million Euros are invested. • The installed heating and cooling capacity is 4 to 4.5 MW

Mijnwater 2.0New connections

Maankwartier

50.000 m2

All electric

CO2: - 65%

MAB

4.500 m2

All electric

CO2: - 65%

Rabobank

3.200 m2

All electric

CO2: - 65%

Acquisition HHC

Transition to all electric

30.000 m2

CO2: - 65%

September 2016

Here impression of the some new built connections. • All energy stations owned and operated by Mijnwater except

the Rabobank; • Most remarkable is Moon quarter with 50.000 m2. A thermal

buffer of 70 m3 is installed for peak shaving and passive

reuse; Due to this, the connection to the grid could be reduced to 50% of peak demand;

• (Mijnwater also acquired the existing energy station of the Heerlerheide Center Complex from the Weller housing company including all energy services and transformed the energy station into all-electric.)

Mijnwater 2.0Cluster D – Heat recovery - Industry

September 2016

Here a view of the Cluster D with a high waste heat recovery potential with high and low temperature waste heat that can be used in cascade between the industries and by the nearby low temperature end-users like the swimming pool and existing dwelling areas.

Mijnwater 2.0Smart energy stations: low-exergy & hybrid

Energy exchange between building

and cluster grid

Energy station

all-electric (HP)

Booster heat pump for

domestic hot water

25 ˚C

40 ˚C

65 ˚C

September 2016

Minewater developed a blueprint for the energy stations designed in accordance with the low-exergy principle. The heat from the grid with a temperature of about 25 ˚C is lifted by

small heat pump in cascade up to the required heating temperature of the end-users, 40 ˚C max for new buildings, 50

˚C max for renovated buildings. For domestic hot water the temperature is lifted up additionally to 65 ˚C by individual

booster heat pumps at the end-users. Needless circulation of hot water with corresponding high heat losses are prevented. These energy stations are also hybrid, able to adjust to different temperature in the cluster grid that can occur due to the application of multiple sources. In this way the available energy can be used in most passive way.

Mijnwater 2.0Local area grids for new and existing renovated dwellings

Cluster grid(2-pipe)

Local area grid(4-pipe)

Energy Transfer Stationwith booster heat pump for DHW

September 2016

A new building block in our system are local area grids for supply of heat and cold to new and/or renovated dwellings, provided by an energy station in an underground basement, connected to the cluster grid. Existing dwellings will be equiped with an energy transfer station in a box out-side the dwelling

with an individual booster heat pump and buffer for domestic hot water.

Mijnwater 3.0Self-learning Thermal Operational Resource Management

(STORM – Horizon 2020)

Heat Cold

BioCHPS

PV HP

S

S

HPPV

INTELLIGENTTOP LEVEL CONTROL

FRAMEWORK

Self-learning Adaptive

season

week/month

hour/day

Cell/clusterbalancing

Peak shavingValley filling

CO2-reduction

80-100 %

MarketInteraction

September 2016

The next innovation stage called Minewater 3.0 focusses on mastering demand and supply with intelligent control and smart storage. Within the Horizon 2020 project called STORM an intelligent top

level control frame work is in development. A self-learning, predictive and adaptive intelligent controller for optimal operation of the total energy infrastructure at each level (buildings, clusters, minewater backbone) with multiple control strategies like cell/cluster balancing; peak shaving and interaction with the electricity market. Most effective when combined with advanced storage, like seasonal long-term storage in the minewater reservoir, mid-term storage within the clusters and short-term storage in the buildings.

Cluster gridconnection

Heatpumpinstallation

Temperature

buffer ˚C

PV

Mijnwater 3.0Smart storage: Energy carrousel

HT-Source (Solar, biomass, waste heat

or surplus green electricty E grid)

Multifunctional:

Peak shaving

Passive reuse

Seasonal/HT storage

Autonomous operation

September 2016

For the cluster level a new smart storage concept called the Energy carrousel is developed, based on the Ecovat application. A big multi-functional subsurface stratified thermal buffer for peak shaving, passive reuse and seasonal or HT storage of solar

thermal, biomass, waste-heat and surplus of green electricity. With these buffers it is possible to reduce the connection to the grid with a factor 3 to 5. Also autonomous operation of the connection and/or the compleet cluster grid is possible by using the buffer as a temporary source.

Mijnwater OperationExperiences and lessons learned

Energy exchange by pumps in series (yes it can);

Behaviour custer grids as buffers (balancing/back-up);

Bio-fouling cluster grids (pH control; bio-shots);

Depletion cold production well;

Leakage incident cold production well;

Performance production wells.

September 2016

Let’s discuss now some of our experiences and lessons learned

with operation. • An energy exchange grid like Mijnwater means putting pumps

in series. Operation proofs that this is possible if designed and

tuned well; • We discovered that cluster grids behave like buffers and can

be used for short-term balancing and back-up; • We also experienced that bio-fouling can occur in cluster

grids, which is typical for closed low temperature grids. It can easily be prevented by right pH control on 9 – 10 and

frequent bio shots; The next 3 items: depletion of the cold production well, the leakage incident at the cold production well and the performance of the production wells will be discussed more in

detail.

Mijnwater OperationDepletion cold production well

September 2016

Temperatures Minewater wells in 2014 [˚C] Temperatures Minewater wells in 2015 [˚C]

Monitoring indicated that the cold production well was depleting. The yearly average weighted temperature rose in 2015 by 0.5 degrees (see red circles).

Reason: until February 2016 we had to use the former infiltration well HLN3 for the most part due teething problems of the new injection wells (HH2 and HLN2) and the fact that the connections of initial two buildings to the grid still had to be changed into minewater 2.0 exchange connections. It means that the minewater reservoir was still used as a source for the most part. Nice thing about this abuse is the fact that this definitely

confirms that geothermal capacity of the minewater reservoir is limited and that use as a storage is key.

Mijnwater OperationDepletion cold production well

HLN1

HLN3

Flow path HLN3 – HLN1 Mijnwater EPANET model (Vito)

September 2016

Depletion is also confirmed by the analysis of Vito with the Minewater EPANET model. This figure shows the flow path between infiltration well HLN3 and the cold production well HLN1. Based on the monitoring

data and a fixed conductivity factor the model indicates that the temperatures of the galleries between wells have risen compared to the initial natural temperatures.

Mijnwater OperationLeakage incident cold production well

Gasket blown outPressure peaks

Inflow of waterLack of saveguards

September 2016

The leakage incident of the cold production well occurred in April 2014. It proofs that Murphy is always lurking. It was the result of a cascade of failures: • Pressure peaks due defects in the system; • A blown out soft gasket; not suited for the application;

• Inflow of water through untied wall penetrations; • And a lack of safeguards. The basic reason for this incident was a lack of communication and poor selection, execution, tuning and maintenance of the buffer and boosting system by the contractor and suppliers. (All short comings are mitigated by upgrading and recommissioning of the buffer and boosting system, including a double leakage detection and overpressure protection that shut down the installation.)

Mijnwater OperationPerformance production wells

Average COP’s production wells and Cluster A

September 2016

Performance monitoring of the production wells, tells us that the COP (Coefficient of Performance) of the hot production and cold production stays far behind expectation, as shown in this graph.

The yearly average is around 14. The expected COP was 25. This has led to a profound investigation of the causes.


Booster pumps

Total

Well

pump

Start/stop

Well pump

Continuous

Well pump

September 2016

It is important to know that the production wells consists of a well pump that brings the mine water to the surface in series with a boosting system that takes care of the distribution to the clusters and that the well pump at demands below min flow is in start/stop operation.

The graph shows the theoretical and practical COP of the well pump, booster pumps and total COP of the system versus flow. The COP of the well pump and total COP are displayed and on the left axis and the COP of the booster pumps on the right axis. The green line versus the orange line show that the bad performance is mainly caused by the booster pumps.

The dark blue line confirms that a total COP of 20 to 25 must be possible. So what is happening? For this we have to look at the pump and operation curve of the booster pumps.


Pump and actual operating curve booster pumps

Start/stop

Well pump

Continuous

Well pump

1th booster pump

(speed control)

2th booster pump

(no speed control)

September 2016

• The boosting system is a black-box with 4 equal booster pumps. Only the first booster pump has speed control.

• Important to know is that the graph shows the pump curve of one booster pump and operation curve of the total boosting system.

• During start/stop operation of the well pump pressure and flow are fluctuating continuously and the first booster pump is operating in this area with very low frequencies and

corresponding low efficiencies. The booster pump is simply too big for this point of operation.

• But even at higher flow when the well pump is in continuous operation the performance is bad due to the fact that all other booster pumps have no speed control.

• When the second booster pump is activated it will run at full load and run of the pump curve on the right side with corresponding bad efficiencies.

• At the same time the speed controlled booster pump is drawn back to the left side of the pump curve also with corresponding bad efficiencies.

• When the pressure set point is reached the second pump shuts-off and the first speed controlled pump has to speed up again. It means that the booster pumps are commuting all the time, resulting in an overall inefficient operation.

That leads us to the following necessary mitigations:


Mitigations:

One small booster (jockey) pump for operation during

start/stop well pump;

All booster pumps provided with speed control;

Advanced process control on maximum pump efficiency;

More steady operation with less fluctuations of demand by

improved tuning cluster installations.

September 2016

Mitigations: One small booster (jockey) pump for operation during

start/stop operation of the well pump; All booster pumps provided with speed control; Advanced process control on maximum pump efficiency;

More steady operation with less fluctuations of demand by improved tuning of the cluster installations.

Mijnwater HeerlenGeneral lesson learned

Innovative concepts like Minewater need Master of

Concept assessing consultants, contractors and suppliers

during the entire development process (not only design)

by sufficient witness points and a solid commissioning

protocol to achieve the pre-set targets

September 2016

I like to end my presentation with one general lesson learned. Innovative concepts like Minewater need a Master of Concept assessing consultants, contractors and suppliers during the entire development process (not only design) by sufficient

witness points and a solid commissioning protocol to achieve the pre-set targets (at last) (In other words: don’t take anything for granted, make sure

that all parties involved understand the new concept completely, take responsibility (not only on paper), think along and overcome their conventional thinking and habits.)

September 2016

Questions?

Thank you for your attention!

Thank you four your attention! Questions?

September 2016

Back-up slides

Mijnwater 2.0Minewater from source to storage

September 2016

Mijnwater 2.0Smart grid: Exchange!

September 2016

grid a -1) te (geo -t...2016/09/18 · slide 2 2016 tion, (geo -t (t6a -1) i will discuss the...

Documents