project title: improvement of energy efficiency in ... · groups, four sand filters as well as the...

WaterWatt case studies report WATERWATT 280761 30 NOBVEMBER 2017

WaterWatt 1st Periodic report Part B – Annex 1

H2020-EE-2015-3-MarketUptake

Project Title: Improvement of energy efficiency in industrial water circuits using gamification for online self-assessment, benchmarking and economic decision support

Acronym: WaterWatt

Grant Agreement No: 695820

Title Comprehensive Report of all case studies examined

Prepared by 3 - BFI

Dissemination Level Public



Table of Content Introduction to WaterWatt case studies ............................................................................... 3

Case study 1: Open cooling circuit of a hot rolling mill ...................................................... 5

Case Study 2: Closed cooling circuit of an inductive furnace ...................................... 10

Case Study 3: Closed cooling circuit of a blast furnace ................................................ 13

Case Study 4: Open gas washing circuit of a basic oxygen furnace ........................... 16

Case Study 5: Open cooling circuit for rebar rods and wire coils ................................ 18

Case Study 6: Closed cooling circuit at Manganese plant ............................................ 21

Case Study 7: Open gas washing circuit at Manganese plant ...................................... 23

Case Study 8: Open cooling circuit of a pharmaceutical company .............................. 25

Case Study 9: Raw material transportation ..................................................................... 27

Case Study 10: Water Treatment Circuit ......................................................................... 31

Case Study 11: Open cooling circuit of a barometric condenser ................................. 36



Introduction to WaterWatt case studies In order to provide a benchmark for energy consumption in industrial water circuits (IWC), sev-eral steps have been performed:

1. Identification and selection of representative circuits in various industrial branches,

2. Characterisation of representative IWCs and their units,

3. Evaluation of specific energy consumption in IWCs

The studies have been performed in four countries: Germany, Portugal, Norway, United King-dom; and 4 industrial sectors: metal and steel, chemical, paper, food and beverage. The follow-ing representative circuits with a flow >50 m3/h were identified:

• Open cooling circuits with contact to product and/or atmosphere up to 5000 m3/h,

• Closed cooling circuits up to 5000 m3/h,

• Flow through cooling with surface water up to 75 m3/h,

• Gas washing circuits up to 4500 m3/h and

• Process water circuits up to 2000 m3/h.

Case studies were performed with selected representative IWCs (Table 1). Their units were characterised, specific energy consumption was measured and measures for the improvement of energy efficiency were developed. For reasons of comparability, specific energy consumption per pumped m³ was taken as the main benchmark. Furthermore, regional factors and sociologi-cal boundary conditions for circuit operation were studied and transferability of the specific en-ergy consumption and improvement measures to other circuits was evaluated.



Table 1: Overview of studied circuits

Industry Case study Origin Representative circuits Flow

[m³/h]

Installed power [kW]

Nr*

Metal Stainless wire processing

DE Open cooling circuit (rolling mill) with sand filtration

2400 1220 1

Closed cooling circuit (inductive furnace)

63 37 2

Carbon steel production

UK Closed cooling circuit (blast fur-nace)

5700 901 3


DE Open gas washing circuit (basic oxygen furnace)

3200 800 4


NO Open cooling/quenching of rebar rods and wire coils

780 315 5

Manganese production

NO Closed cooling circuit (furnace) 350 171 6

Open gas washing circuit 250 111 7

Chemical Pharmaceuticals DE Open cooling circuit 3600 1254 8

Paper Paper factory PT Fiber transportation circuit 2400 240 9

Food and beverage

Sugar factory PT Water treatment (filtration) 145 135 10

Open cooling circuit 1600 627 11

* Case Study Number



Case study 1: Open cooling circuit of a hot rolling mill

1.1 Description of the plant, production process and related circuit

At the stainless steel wire processing plant steel billets are transformed to wire in several pro-duction steps. One of these steps is hot rolling. The billets are reheated in an inductive furnace and sent through a series of rolls. The rolls and the rolled material are continuously cooled with water to prevent equipment damage and to reach required material quality. Furthermore, water is used to break the oxide scale and wash it off the surface the hot material. Water demand is not constant because of the batch character of the production (1 billet at a time) as well as the variabilities in the material and operational conditions of the rolling mill (rolling speed and cool-ing speed required).

In order to provide cooling water for the rolling mill and separate oxide scale as well as oils, the following water circuit was constructed (Figure 1). It contains four cooling towers, three pump groups, four sand filters as well as the oil separator and a cyclone for the separation of sus-pended solids. All the pumps are KWP 200-400 with P = 110 kW, Q = 600 m3/h, H = 45 m and n = 1450 min-1. The cooling towers of the type 3/70 Z XL are equipped with four fans of 22 kW.

Figure 1: Open cooling circuit of a hot rolling mill

The pump group 1 (Figure 2a) transports cooled water from the cooling tower tank (Figure 2b) to the rolling mill. Both pumps (+ 1 spare pump) are equipped with variable speed drive (VSD). The frequency and operation mode (1 or 2 pumps) are adjusted automatically according to the required pressure at the rolling mill. From the rolling mill the water flows down the channel to the oil separator and cyclone. The pre-treated water is directed to the sand filters by the pump group 2. The 4 pumps are operated without VSD. They are just turned on and off automatically depending on the water level in the cyclone tank. From the 4 sand filters the water flows to the cooling tower tank. Should the temperature in the cooling tower tank increase above the target



value by 0,5 K, the pump group 3 as well as the cooling tower fans turn on, providing efficient demand oriented cooling. As soon as the temperature drops to the target value, the operation of the cooling towers stops.

a) Pump group 1

b) Cooling tower

Figure 2: Pump group 1 (a) and the cooling tower (b) of the hot rolling mill circuit

1.2 Regional factors and boundary conditions

The production site is located in Germany and is not a subject to water or energy shortage. The IWCs have benefited from a recent modernisation drive that has focused on technological solu-tions to minimise energy use as far as possible. For example, operation of cooling towers was separated from the operation of the circuit. Additional pumps were installed for that (pump group 3). This way the cooling towers can work at optimal pressure and fan speed providing energy efficient cooling. The additional pumps are of the same type that in the rest of the circuit to re-duce maintenance and logistic effort. The use of numerous sensors constantly measuring pres-sure and temperature of the water as part of a digitalised control infrastructure in conjunction with frequency-regulated pumps has led to an increased optimisation of energy use in the main IWC.

Due to the high automation of the circuit there is only very limited room for human actions to influence the way energy is used. This relates to the continuous maintenance of IWCs as insuf-ficient maintenance can increase energy use: leaking pipes require more energy as pressure



has to be kept up, blocked pipes and blocked valves as well as worn or defect pumps increase the energy demand in the circuits. At this location the maintenance is organised in a pro-active way. Visual inspections as well as exchange of gaskets and bearings are performed regularly. The teams register their observations and repairs in SAP-system so that by the repeated fail-ures the reason can be quickly identified and eliminated. Installation of vibration sensors for the pumps can be considered to reduce reaction times for failures.

1.3 Specific energy consumption

The energy demand of the circuit units has not been measured before the project start. In the frame of WaterWatt under the support of the plant management the circuit was equipped with permanent flow and energy meters. Due to the separate meters on each pump and fan, the en-ergy demand could be broken down to the demand for pumping and for cooling. An overview of the specific energy consumption values is presented in Table 2. The average annual demands for pumping and cooling are comparable. The variation of specific energy demand based on the flow rate of the main circuit is presented in Figure 3.

Figure 3: Specific energy consumption in a cooling water circuit of a hot rolling mill



Table 2: Specific energy consumption in a cooling water circuit of a hot rolling mill

Parameter Unit Pumps Cooling tower

Total Remark

Flow nominal m3/h 1750 1200 8 Pumps: KWP 200-400 à 600 m³/h

Flow real m3/h 900 900 900

Pressure nominal bar 4,5 4,5 4,5

Pressure real bar 3,3 3,3 3,3

Δ T nominal 1

Δ T real K 1

Production rate t/h 14,4 14,4 14,4

Operational time h/y 6600 6600 6600

Power installed kW 660 308 968 8 Pumps à 110 kW; 4 Fans à 22 kW

Specific power installed W×h/m3 377 257

Specific power installed W×h/(m3×bar) 114

Specific power installed W×h/(m3×K) 257

Specific power installed W×h/t 46 21 67

Power consumed MWh/y 1650 1497 3147 Estimated

Specific power consumed W×h/m3 278 252 530

Specific power consumed W×h/(m3×bar) 84 161

Specific power consumed W×h/(m3×K) 252,0

Specific power consumed W×h/t 17,4 15,8 33,1

Transferred heat power kWth 1047 Q = C × q × ∆T real; C = 1.163 kWh /(m3 × K)

Specific power consumed for heat transfer

Whel/kWhth 217

Installed power use % 49 Power used/power installed



The energy demand for cooling includes the energy demand of the fans as well as the energy demand of the pump group 3. The specific energy demand for pumping is with 0,17 ± 0,05 kWh/m3 fairly constant. The energy demand for cooling fluctuates stronger (Figure 3). This fluctuation correlates with the average outdoor temperatures (Figure 4). Higher energy consumption by the higher outdoor temperatures and constant cooling water tempera-ture takes place due to the reduced heat transfer rate. Thus, by the transferability of energy effi-ciency benchmarks for cooling circuits outdoor temperatures have to be considered.

Figure 4: Specific energy consumption for cooling correlates with the out-door temperature

1.4 Improvement of energy efficiency

As the circuit has been completely refurbished in 2016, very few improvement possibilities are still open. The circuit model shows that energy efficiency could be slightly improved by the opti-misation of the switch points of the pumps equipped with variable speed drives (pump group 1). Furthermore, introduction of fills into the cooling towers will improve its efficiency and reduce energy consumption.

The economic evaluation (Table 3) shows that both measures are feasible and their imple-mentation can contribute to savings and improvement of energy efficiency. The validation pos-sibility for the simulation results is being discussed with the plant operator.

The proposed measures as well as the circuit automation and maintenance practice are trans-ferable to other circuits with variable water demand. The savings and payback times may vary depending on the local energy costs.



Table 3: Economic evaluation of selected improvement measures

Measure Investment and installation

costs

Energy savings

Operating costs

savings

Payback time

Comment

[€] [€/kW] [€/y] [%] [€/y] [%] [y]

Introduction of fills into cool-ing tower

45000 1560 19000 6 0 0 2.4 Considerable energy savings (> 3 %) and economically efficient (I < 3 y)

Optimisation of the pump switch points

0 0 1700 1 0 0 0 Minor energy savings but without any invest-ment costs

Case Study 2: Closed cooling circuit of an inductive furnace


The cooling circuit is installed at the same location as case study 1. At the stainless steel wire processing plant steel billets are transformed to wire in several production steps. One of these steps is hot rolling. The billets are reheated in an inductive furnace and sent through a series of rolls. The inductive furnace is cooled by a closed circuit (Figure 5). The cooling demand is con-stant during the plant operation. During production breaks the cooling circuit is automatically turned off.

Figure 5: Closed cooling circuit of an inductive furnace



The circuit is operated by two pumps RITZ 40-200 with P = 18.5 kW, Q = 31.5 m3/h, H = 61.2 m and n = 2940 min-1. The hybrid cooling tower is equipped with one fan of 4.1 kW. In summer the city water is used for additional cooling.


The circuit is located in Germany. Regular maintenance is performed. Due to constant operation conditions and automasation there are limited possibilities to further improve the energy effi-ciency.


The specific energy demand is presented in Table 4. It is higher than that of the circuit from case study 1 due to the higher pressure losses in the system and lower efficiency of the smaller pump.


Due to the constant operation conditions no improvement possibilities is seen for this studied circuit. Exchange of the pump for a more efficient one can provide 2 % improvement of the en-ergy efficiency. Due to very low absolute savings expected (ca. 160 €/y per pump) this ex-change will be performed only in case the pump breaks.



Table 4: Specific energy consumption in a closed cooling circuit of an inductive furnace


Total Remark

Flow nominal m3/h 63 63 63

Flow real m3/h 60 60 60

Pressure nominal bar 6.1 6.1 6.1

Pressure real bar 5 5 5

Δ T nominal 10

Δ T real K 3

Production rate t/h 14.4 14.4 14.4


Power installed kW 37 4.1 41.1 2 Pumps à 18.5 kW; 1 fan à 4.1 kW

Specific power installed W×h/m3 587 65 652


Specific power installed W×h/(m3×K) 6,5


Power consumed MWh/y 212 27 240 Estimated



Specific power consumed W×h/(m3×K) 22.8

Specific power consumed W×h/t 2236 285 2521



Whel/kWhth 173




Case Study 3: Closed cooling circuit of a blast furnace


In a UK metallurgical plant pig iron is produced in a blast furnace. In order to avoid damage from the high temperatures (up to 2,200 °C), its stack is cooled by a closed cooling circuit. The circuit consisting of two pumps Uniglide-e SDN 400/600B Clyde (P = 350 kW; Q = 2850 m3/h; H = 45 m, n = 1800 min-1) and two GEA plate heat exchangers at 25 °C (Figure 6). The heat is transferred into the environment by means of a secondary cooling circuit with three cooling tow-er units (3 fans à 67 kW). The closed circuit is operated continuously at the same flow because the furnace has to be cooled continuously. It is operated at the overpressure of 4.2 bar to pre-vent local water boiling. The cooling tower fans are equipped with variable speed drives and are automatically adjusted to the heat load and the weather conditions.

Figure 6: Closed cooling circuit of a blast furnace


The safety is a very important aspect of this circuit. The failure results in irreparable damage to the furnace. Thus energy efficiency is a secondary aspect of its operation.


The specific energy demand is presented in Table 5. It is fairly low comparing to the other cir-cuits because the pumps are constantly operated close to the optimal point. Furthermore, the larger pumps are more efficient than the smaller ones.




The conditions in the blast furnace are not always constant. Thus, automatic flow adjustment of the circuit by means of variable speed drives or additional pumps could lead to improvement of energy efficiency. At the current conditions it is not realistic due to the high safety requirements.



Table 5: Specific energy consumption in a closed cooling circuit of a blast furnace


Total Remark

Flow nominal m3/h 5700 8000

Flow real m3/h 5000 5000 5000

Pressure nominal bar

Pressure real bar 3.3 3.3

Δ T nominal 7

Δ T real K 3

Production rate t/h 275 275 275 Pig iron production


Power installed kW 700 201 901 2 Pumps à 350 kW; 3 fans à 67 kW



Specific power installed W×h/(m3×K) 3.6


Power consumed MWh/y 5510

Specific power consumed W×h/m3 98 28 126 Distribution between pumps and fans was estimated on basis of power installed



Specific power consumed W×h/t 1777 510 2287



Whel/kWhth 36




Case Study 4: Open gas washing circuit of a basic oxygen furnace


In a German metallurgical plant various steel types are produced. The pig iron from the blast furnace is treated in basic oxygen furnace to reduce carbon content. The gas from the basic oxygen furnace is cleaned with water. The gas washing circuit (Figure 7) is operated by five pumps RDL 250-400A with P = 200 kW, Q = 800 m3/h, H = 62 m and n = 1480 min-1. The water from the storage tank is filtered by three Dango filters and directed to two Venturi scrubbers. The suspended solids from the gas are separated in the sedimentation tank.

Water demand is variable because the scrubbers do not operate continuously: in basic oxygen furnace pig iron is treated in batches. The pumps are equipped with frequency converters and are automated by means of pressure sensors.

Figure 7: Gas washing circuit of a basic oxygen furnace


At some operational conditions the required pressure cannot be reached even when the pumps are operated at the maximal power. Increase of the maximal frequency from 50 Hz to 53 Hz is being discussed as an option to meet this challenge.


The specific energy demand is presented in Table 6. The energy demand of the cooling tower was not studied due to accessibility issues. During the measurement period only 4 of 5 pumps were operated.



Table 6: Specific energy consumption in a closed cooling circuit of a blast furnace

Parameter Unit Pumps Total Remark

Flow nominal m3/h 4000 4000 5 pumps à 800 m3/h

Flow real m3/h 2850 2850

Pressure nominal bar 6.2 6.2

Pressure real bar 5.5 5.5

Production rate t/h

Operational time h/y 8750 8750

Power installed kW 1000 1000 5 pumps à 200 kW (4 pumps in operation)

Specific power installed W×h/m3 250


Specific power installed W×h/(m3×K)

Specific power installed W×h/t

Power consumed MWh/y 4638 4638

Specific power consumed W×h/m3 186 186




Due to the current automation algorithm the pumps are running under optimal conditions. One of the pumps showed higher vibrations than normal. It has to be maintained to prevent damage and energy losses.



Case Study 5: Open cooling circuit for rebar rods and wire coils


The plant is part of an industrial park. It is connected to the central industrial park water supply. The plant produces a range of steel products, mostly wire, and has an electric arc furnace as well as a rolling mill. The water is used to provide cooling for rebar rods and wire coils.

There are three pumps in total operating the flow-through water circuit (Figure 8). One of them is from the 1950's and is rarely used, while the other two are frequency regulated pumps. The pumps are used to increase the pressure level of the water coming from the industrial park wa-ter circuit. The water demand and required pressure level is varying according to what being produced at the plant. After the water has cooled the products, it flows out of the plant.

Figure 8: Scheme of the open cooling circuit for rebar rods and wire coils


The water being used is rain and melt water from the mountains. It possesses hydrostatic pres-sure thus reducing energy demand in the flow-through circuit in comparison to the plants locat-ed elsewhere. The water use does not have a negative impact on the environment. Even though energy efficiency is a company objective, the cost of water and electricity is very low and does not encourage energy efficient investment that saves water or reduces the energy con-sumption. The energy use related to the water circuit is very low compared to the overall energy consumption for the plant.




The water flow through each pump is monitored at the plant, but the energy demand of the pumps is not measured. During the WaterWatt project, the pumps were equipped with energy meters measuring the energy demand of the pumps for nine days.

During these days, various product types were manufactured, resulting in different demands for water flow and pressure level (see Figure 9). Based on the measured energy used by the pumps, data for the water flow and products being produced and answers from questionnaires, the average specific energy consumption was calculated as shown in Table 7.

Figure 9: Power used by the two pumps operated during the measuring period

0"

20"

40"

60"

80"

100"

120"

140"

160"

180"

200"

0" 2" 4" 6" 8" 10"

Power"[kW]"

Time"[Day]"

Pump"1"Pump"2"



Table 7: Specific energy consumption in a cooling water circuit for case study 5


Total Remark

Flow nominal m3/h 650 780 780

Flow real m3/h Depends on what is being produced

Pressure nominal bar 16 7

Pressure real bar 10 4

Production rate t/h 75 Rebar rods, rebar coils, wire coils


Power installed kW 630 225 855 2 pumps à 315 kW 1 pump à 225 kW

Specific power installed W×h/m3 969 288 1096

Specific power installed W×h/(m3×bar) 97 72


Specific power installed W×h/t 11400

Power consumed MWh/y 2000 2200 Estimated

Specific power consumed W×h/m3 380

Specific power consumed W×h/(m3×bar) 38

Specific power consumed W×h/(m3×K)

Specific power consumed W×h/t 3500



Since the pumps in use already have installed frequency converters, the main improvement measure related to energy use is to install new, more efficient pumps. Table 8 shows an eco-nomic evaluation of the improvement measure and indicates that the payback time would be a bit higher than usually acceptable in the industry.



Table 8: Economic evaluation of improvement measures


costs

Energy savings

Operating costs

savings

Payback time

Comment

[€] [€/kW] [€/y] [%] [€/y] [%] [y]

Installation of new pumps

90000 1685 25000 20 0 0 3.6 Longer payback time than usually accepted in the industry, but shorter than the pump lifetime. High absolute savings.

Case Study 6: Closed cooling circuit at Manganese plant


The plant is part of an industrial park, and connected to the central industrial park water circuit, which delivers water to the plant at a certain pressure level. The plant is producing manganese in two furnaces, and each furnace has a closed cooling circuit. Each of these circuits uses two pumps in order to be operated, and have two pumps for back-up.

Even though the plant is operated continuously all year round on full capacity with constant de-mand for water, frequency-regulated pumps are installed. The reason is, that these pumps were of the cheapest pump type available.


Since the process is so stable and is operated continuously, investments become very expen-sive due to the necessary production stops. Since the water demand is constant, neither fre-quency-regulated pumps nor automated start-stop technology would provide energy savings.

As for case study 5, the plant is located in an industrial park. The topography in the area is mountainous landscape with sufficient natural precipitation which ensures viability of using mountain top rain and melt water to supply water.


The process at the plant is operated continuously at constant capacity, which makes it easy to calculate the specific energy consumption. The values are presented in Table 9.



Table 9: Specific energy consumption in a cooling water circuit


Flow nominal m3/h 350 350 2 pumps à 350 m3/h (1 in standby for backup)



Pressure real bar

Δ T nominal

Δ T real K

Production rate t/h


Power installed kW 171 171 2 pumps à 160 kW (1 in standby for backup) 2 pumps à 11 kW pressure holding; (1 in standby for backup)


Specific power installed W×h/(m3×bar)





Specific power consumed W×h/(m3×bar)


Specific power consumed W×h/t



Whel/kWhth





The water circuits are newly built, and no energy efficiency measures could be found.

Case Study 7: Open gas washing circuit at Manganese plant


In addition to the two cooling circuits, described in case study 6, the plant has one open gas washing circuit. In this circuit, there are three pumps that are used during operation, and two pumps for back-up.


(see case study 6)


The process at the plant is operated continuously at constant capacity, which makes it easy to calculate the specific energy consumption. The values are presented in Table 10.


These water circuits are newly built, and no energy efficiency measures could be found.



Table 10: Specific energy consumption in an open gas washing circuit


Flow nominal m3/h 250 250 5 pumps à 85 m3/h (2 in standby for backup)



Pressure real bar

Δ T nominal

Δ T real K

Production rate t/h


Power installed kW 111 111 5 pumps à 37 kW (2 in standby for backup)


Specific power installed W×h/(m3×bar)





Specific power consumed W×h/(m3×bar)



Transferred heat power kWth Q = C × q × ∆T real; C = 1.163 kWh /(m3 × K)


Whel/kWhth




Case Study 8: Open cooling circuit of a pharmaceutical company


At a pharmaceutical plant located in Germany there is a centralised open cooling system with three identical cooling units providing many small consumers with cooling water. Each unit con-sists of one cooling tower and four pumps (Figure 10). The cooling demand is variable. The pumps are turned on/off automatically on basis of the monitored system pressure. The fans are equipped with variable speed drives and automated by means of temperature sensors.

Figure 10: One of the three identical units of the open cooling circuit of a pharmaceutical company


Each cooling unit is equipped with flow and electricity meters. This was the only circuit in the study where no additional measurements had to be performed. In the last two years the circuit was completely refurbished. Pumps were exchanged against smaller ones, automation was optimised. These measures resulted in the energy savings of approx. 30 %.


Specific energy consumption is calculated in Table 11. It is comparable to the system with simi-lar pump size (case study 1). It is higher than in the circuits operated with constant flow because due to the flow variation the system does not always run under optimal conditions.



Table 11: Specific energy consumption in an open cooling circuit of a pharmaceutical company


Total Remark

Flow nominal m3/h 6000 6000

Flow real m3/h 3600 3600 3600

Pressure nominal bar 6 6

Pressure real bar 4,3 4,3

Δ T nominal 10

Δ T real K 10

Production rate t/h Various pharmaceuticals


Power installed kW 1164 90 1254 12 pumps à 97 kW; 3 fans à 30 kW



Specific power installed W×h/(m3×K) 1.5


Power consumed MWh/y 3300 180 3480







Whel/kWhth 28





Due to the recent circuit refurbishment and improvement of energy efficiency not further measures could be identified.

Case Study 9: Raw material transportation


The selected circuit at a paper production factory represents the dilution and pressing process. This circuit starts with the dilution process in order to feed the pressing machine. Overall, there are two types of paper: F1 and F3 and they are mixed with water in two separate tanks. The resulting pulp is then transported to the pressing machine. The pressing machine consists of four press rolls, in which the dry solids content of the bulk increases step by step. The presses are usually fitted with press felts, which distribute the pressing pressure on the pulp web and remove water. Afterwards the paper pulp is transported to the drying machine.

The dilution process is represented in the schematics displayed in Figure 11 by the dilution tank. The pulp is supplied from this tank to the feed tank by two pumps named as P_F1 and P_F3. Each pump supplies one type of pulp: F1 and F3. The feed tank will feed the pressing machine by gravity and also feed small amount back to the dilution tank to keep the same lev-el in this tank. The pulp enters the pressing machine by four sections. Every two sections corre-spond to one type of pulp: F1 and F3. The pulp is pressed between the rolls and along the pro-cess, water and bulk are falling down. This wasted water and bulk are dissociated and collected into four tanks (two for F1 and F3 each), named as: WaterwasteF1, BulkwasteF1, Water-wasteF3, BulkwasteF3. Furthermore, WaterwasteF1 and WaterwasteF3 are pumped by P_WaterwasteF1 and P_WaterwasteF3 respectively to the dilution tank. BulkwasteF1 and BulkwasteF3 are pumped by P_BulkwasteF1 and P_BulkwasteF3 respectively to the feed tank. The remaining water and pulp that are not collected into the aforementioned tanks, inte-grate other processes not under analysis in this study. Hence, these “massflows” are aggregat-ed into a sink named as water for other processes. Figure 12 shows some pictures of the in-stallments.



Figure 11: Paper transportation schematics (OpenModelica design)

Figure 12: Top left: dilution tank; top right: pumps (P_F1 and P_F3); bottom: pressing machine




The raw material transportation has special requirements on the pumps. They are usually not as efficient as process water pumps. Furthermore, energy efficiency has not been in the focus of the company. The first priority was to provide smooth process operation with limited investment funds. No systematic upgrade of the facility is planned because of the same reason.


The energy demand of the circuit units has not been measured before the project start. In the frame of WaterWatt under the support of the plant management the circuit was equipped with permanent flow and energy meters. Due to the separate meters on each pump, the energy de-mand could be broken down to the demand for pumping. The energy consumption per pump of selected branch of main circuit was constant during the measurements. An overview of the spe-cific energy consumption values are presented in Table 12. From this table, the average annual demands for transport of raw material are comparable.


As the circuit has not been refurbished, hence few improvement possibilities are open. The cir-cuit model shows that energy efficiency could be slightly improved by refurbishment of all old pumps. The economic evaluation (Table 13) shows that this measure is feasible and its imple-mentation can contribute to savings and improvement of energy efficiency. The validation pos-sibility for the simulation results is being discussed with the plant operator.

The proposed measures as well as the circuit automation and maintenance practice are trans-ferable to other circuits with variable water demand. The savings and payback times may vary depending on the local energy costs.



Tabl

e 12

: Spe

cific

ene

rgy

cons

umpt

ion

of th

e ra

w m

ater

ial t

rans

porta

tion

circ

uit

Par

amet

er

Uni

t P

_F3

P_F

1 P

_Bul

k-w

aste

F3

P_W

ater

-w

aste

F3

P_

Bul

k-w

aste

F1

P_W

ater

-w

aste

F1

Tota

l R

emar

k

Flow

nom

inal

m

3 /h

300

300

100

250

100

250

1300

Inst

ant f

low

m

3 /h

226

226

57

200

57

200

966

Pre

ssur

e no

min

al

bar

2.65

(2

6.5m

) 2.

65

(26.

5m)

1.52

(1

5.2m

) 3.

75

(37.

5m)

1.52

(1

5.2m

) 3.

75

(37.

5m)

15.8

4 (1

58.4

m)

Pre

ssur

e lif

t rea

l ba

r 2.

85

(28.

5m)

2.85

(2

8.5m

) 1.

86

(18.

6m)

3.95

(3

9.5m

) 1.

86

(18.

6m)

3.95

(3

9.5m

) 17

.32

(173

.2m

)

Ope

ratio

nal t

ime

h/y

6000

60

00

6000

60

00

6000

60

00

6000

Pow

er m

easu

red

kW

38.1

36

.8

4.7

31.3

3.

9 31

.6

146.

4

Pow

er in

stal

led

kW

45

45

15

30

15

30

180

Spe

cific

pow

er u

sed

W×h

/m3

160

160

80

150

60

160

770

Spe

cific

pow

er in

stal

led

W×h

/(m3 ×

bar)

60

60

40

40

30

40

27

0

Pow

er c

onsu

med

M

Wh/

y 22

8 22

0.8

28.2

18

7.8

23.4

18

9.6

877.

8 es

timat

ed

Inst

alle

d po

wer

use

%

84

.6

81.8

31

10

4.3

24

105.

3 –

Pow

er u

sed/

po

wer

inst

alle

d





costs

Energy savings

Operating costs

savings

Payback time

Comment

[€] [€/kW] [€/y] [%] [€/y] [%] [y]

Refurbishment of all 6 old pumps

960 0.13 5705.5 5 0 0 0.16 Considerable energy savings (> 3%) and economically efficient (I < 3 y)

Case Study 10: Water Treatment Circuit


The main objective of this circuit is to provide fresh water to the sugar refinery. Part of the water is provided by ground water sources. It is stored inside a deposit (ground tank, Figure 13) and filtered by two pumps (pump1 and pump2) through a sand filter. Then it is sent to the drinking water tank which also receives water from the water supplier company. Then the water is trans-ported by three pumps (pump3, pump4 and pump5) into the factory water tank, which is located 14 m above the ground. Figure 14a-c show some pictures of the installments.

Figure 13: Water treatment schematics (OpenModelica design)



Figure 14a: Left: ground water pumps (Pump1, Pump2); right: water tank

Figure 14b: Left: ground water tank Pump3; right: Pump4

Figure 14c: Left: Pump5; right: factory water tank




The water scarcity and high awareness of energy costs in this company are the boundary condi-tions that make the implementation of improvement measures likely.


The energy demand of the circuit units has not been measured before the project start. In the frame of WaterWatt under the support of the plant management the circuit was equipped with permanent flow and energy meters. Due to the separate meters on each pump, the energy de-mand could be broken down to the demand for pumping. The energy consumption per pump of selected branch of main circuit is presented in Figure 15 to Figure 19. An overview of the specif-ic energy consumption values are presented in Table 14.

Figure 15: Energy consumption of Pump1




Tabl

e 14

: Spe

cific

ene

rgy

cons

umpt

ion

of th

e w

ater

trea

tmen

t circ

uit

Para

met

er

Uni

t Pu

mp1

Pu

mp2

Pu

mp3

Pu

mp4

Pu

mp5

R

emar

k

Flow

nom

inal

m

3 /h

45

45

80

97.2

97

.2

Inst

ant f

low

m

3 /h

58

58

– 35

.8

28.2

Pre

ssur

e no

min

al

bar

3.12

3.

12

3.2

5.7

5.7

Pre

ssur

e lif

t rea

l ba

r 2

2 –

5.6

4.8

Ope

ratio

nal t

ime

h/y

595

555

42.5

46

95

4222

.5

Pow

er m

easu

red

kW

5.3

5.

6 4.

8 12

.8

10.1

38

.6

Pow

er in

stal

led

kW

(5.5

) (5

.5)

(14.

7)

(15)

(1

5)

Spe

cific

pow

er u

sed

W×h

/m3

90

90

– 36

0 36

0 90

0

Spe

cific

pow

er in

stal

led

W×h

/(m3 ×

bar)

45

45

–

60

80

230

Pow

er c

onsu

med

M

Wh/

y 3.

2 3.

1 –

60.1

42

.6

Est

imat

ed

Inst

alle

d po

wer

use

%

96

1

01

33

85

67

Pow

er u

sed/

pow

er in

stal

led




As the circuit has not been refurbished, hence few improvement possibilities are open. The cir-cuit model shows that energy efficiency could be slightly improved by:

• Pump of ground water to operate during a longer period of time with a lower flow thus achieving the same water volume

• The refurbishment of old pumps requires changing rolling bearing, sealing rings, and applying an internal ceramic coating. This refurbishment will improve the pump energy efficiency by 5 %.

The economic evaluation (Table 15) shows that both measures are feasible and their implemen-tation can contribute to savings and improvement of energy efficiency. The validation possibility for the simulation results is being discussed with the plant operator.



costs

Energy savings

Operating costs

savings

Payback time

Comment

[€] [€/kW] [€/y] [%] [€/y] [%] [y]

Adapt variable speed device – ground water pump (single pump)

414 380 518 42 0 0 0.8 Considerable energy savings (> 3 %) and economically efficient (I < 3 y). Low absolute savings

Refurbish-ment of old pumps (all pumps)

640 708.5 5 0 0 0.9 Considerable energy savings (> 3 %) and economically efficient (I < 3 y). Low absolute savings

Case Study 11: Open cooling circuit of a barometric condenser


This circuit is an auxiliary circuit of the process of a sugar factory. Sugar must be cooked at low temperature (70 ºC), and to boil water at this temperature it is necessary to produce vacuum. This vacuum is produced in seven barometric condensers denominated by TV (one for each (7) cooking recipient) and one evaporator. To produce vacuum, it is necessary to deliver cold water (30 ºC) into the barometric condensers, to condensate the steam into hot water (40 ºC). This hot water is collected in a tank and then is sent, by four pumps to the top of a cooling tower. Then it is cooled down to 30 ºC and sent through dedicated pipes, one per each of the seven baromet-



ric condensers and to the evaporator by eight pumps (Figure 20). Figure 21 shows some pic-tures of the installments.

Figure 20: Water cooling schematics (OpenModelica design)

Figure 21: Top left and right: cold water pumps; bottom: cooling tower




The cooling tower is old, made of wood and does not provide the best cooling efficiency. A con-struction of a new cooling tower is planned.


The energy demand of the circuit units has not been measured before the project start. In the frame of WaterWatt under the support of the plant management the circuit was equipped with permanent flow and energy meters. Due to the separate meters on each pump, the energy de-mand could be broken down to the demand for pumping. The energy consumption per pump of selected branch of main circuit are presented in Figure 22 to Figure 33. An overview of the spe-cific energy consumption values are presented in Table 16. From this table, the average annual demands for transport of raw material are comparable.







Figure 26: Energy consumption of Pump_TV1 (barometric condenser 1)



Figure 33: Energy consumption of Pump_evaporator



Tabl

e 16

: Spe

cific

ene

rgy

cons

umpt

ion

of th

e w

ater

trea

tmen

t circ

uit

Para

met

er

Uni

t TV

1 TV

2 TV

3 TV

4 TV

5 TV

6 TV

7 Ev

ap.

P1

P4

P5

P6

Rem

arks

Flow

nom

inal

m

3 /h

200

200

200

200

250

200

250

76

350

350

350

586

Inst

ant f

low

m

3 /h

188

228

175

324

338

188

219

93

280

280

280

280

Pre

ssur

e no

min

al

bar

2.44

2.

44

2.44

2.

44

3.10

2.

44

3.09

3.

18

1.49

1.

49

1.49

2.

31

Pre

ssur

e lif

t rea

l ba

r 2.

04

2.20

2.

17

2.10

1.

95

2.45

2.

15

2.38

1.

4 1.

3 1.

45

1.4

Ope

ratio

nal t

ime

h/y

1187

27

10

2452

29

08

2892

12

90

2864

32

11

3344

33

44

3345

31

79

Pow

er m

easu

red

kW

17.7

19

.7

20.8

32

.4

31.5

23

.5

21.8

18

.5

38.2

37

.6

20.2

31

.3

Pow

er in

stal

led

kW

37

37

37

37

55

37

55

25

37

37

37

55

Spe

cific

pow

er u

sed

W×h

/m3

190

160

210

110

160

190

250

270

130

130

130

190

Not

app

licab

le fo

r thi

s ci

rcui

t

Spe

cific

pow

er in

stal

led

W×h

/(m3 ×

bar)

90

70

90

50

80

70

11

0 11

0 90

10

0 90

14

0 N

ot a

pplic

able

for t

his

circ

uit

Pow

er c

onsu

med

M

Wh/

y 21

,0

53,3

51

,1

94,7

91

,2

30,0

62

,4

59,3

12

7,5

125,

9 67

,4

99,6

Inst

alle

d po

wer

use

%

48

53

56

88

57

64

40

–

103

102

55

57

Pow

er u

sed/

pow

er

inst

alle

d




As the circuit has not been refurbished, hence few improvement possibilities are open. The cir-cuit model shows that energy efficiency could be slightly improved by:

• The refurbishment of old pumps requires changing rolling bearing, sealing rings, and applying an internal ceramic coating. This refurbishment will improve the pump energy efficiency by 5 %.

The economic evaluation (Table xx) shows that both measures are feasible and their implemen-tation can contribute to savings and improvement of energy efficiency. The validation possibility for the simulation results is being discussed with the plant operator.


costs

Energy savings

Operating costs

savings

Payback time

Comment

[€] [€/kW] [€/y] [%] [€/y] [%] [y]

Refurbishment of all pumps

1920 69 15967 15 0 0 0.12 Considerable energy savings (> 3 %) and economically efficient (I < 3 y). Low absolute savings

Adjustment of the existent variable speed device of the fan in the cool-ing tower

0 0 170 1 0 0 0 Considerable energy savings (> 3 %) and economically efficient (I < 3 y). Low absolute savings

project title: improvement of energy efficiency in ... · groups, four sand filters as well as the...

Documents