Content uploaded by Marco Pellegrini
Author content
All content in this area was uploaded by Marco Pellegrini on Mar 03, 2015
Content may be subject to copyright.
I2SM 2014 Ref no: ID 146
ZERO ENVIRONMENTAL IMPACT PLANT FOR SEABED MAINTENANCE
Bianchini Augusto, Pellegrini Marco, Saccani Cesare
Department of Industrial Engineering (DIN) – University of Bologna
ABSTRACT
The paper shows designing, prototyping and testing carried out since 2002 on an
innovative plant for seabed maintenance, characterized by zero environmental impact and
planned especially for harbour areas. The core of the technology is made up by a jet-
pump device called "ejector". The ejector has been sized through both laboratory
experiments and fluid-dynamic simulations. In 2005 the first full-scale experimental plant
was designed and carried out in the port of Riccione (Italy). The results of the experimental
campaigns demonstrated the functionality of the system itself, the cost-effectiveness and
low environmental impact if compared to the use of the dredge. Finally, in 2011 the first
industrial plant has been realized in the Portoverde Marina (Italy). This plant is
characterized by better performances in automation and control higher than the first
experimental plant. By these features it is possible to increase plant reliability and ensure a
further reduction of the management costs.
Keywords: Seabed maintenance, Dredging zero environmental impact, Jet pump, Industrial
dredging plant, Automation and control.
1. INTRODUCTION
Harbours and tidal inlets located on coastal areas have one common characteristic: the
need to avoid littoral materials collected nearby the entrance. Bypassing can occur
naturally, but in the natural process the harbour entrance becomes unusable for
commercial and navigation purposes and, in the worst cases, the inattention results in the
complete closure of the port itself. The dredging equipment represents the most common
solution for sand bypassing and for keeping harbour and inlets right depth. This kind of
equipment is rugged and reliable; it has been proved over and over and appears to be
irreplaceable for many applications and locations. On the other hand, there are locations
and situations for which this equipment is not suitable and may be detrimental or
prohibitively expensive in terms of costs. Moreover, dredging operation obstructs the
normal navigation operations. Furthermore, dredging operation can have high
environmental impact on marine flora and fauna [1,2] and can contribute to mobility of
contaminants and pollutants [3,4] already present on the seabed.
Need for improved operations and maintenance techniques and equipment for sediment
bypassing is a profitable sector for research and development applications [5]. Among
other technologies, jet pump has a great potential as primary component in sand
bypassing system, since it requires limited personnel, is able of great portability and can
be assembled at reasonable cost: moreover, the technology is reliable since has been
applied starting from 1976 for coastal application [6]. A jet pump (Figure 1a) is a device
that transfers momentum from a high speed primary jet flow to a secondary flow. The
primary jet flow contacts the suction fluid at the nozzle exit and drags it into the jet pump,
thus starting up and sustaining the secondary flow of suction fluid from surrounding water
mass. If present, solid particles are entrained in the secondary flow, thus being introduced
in the mixing chamber, where jet stream and suction fluid are further mixed, exchanging
momentum and recovering pressure. The slurry then pass through a diffuser and into a
discharge pipe for delivery to a discharge point (or into a booster).
Figure 1: a) Schematic of a jet pump; b) Schematic of the “ejector” developed by DIN.
Department of Industrial Engineering (DIN) of Bologna University, in collaboration with two
Companies, Elettromeccanica Muccioli Marco Srl (Rimini, Italy) and Plant Engineering Srl
(Bologna, Italy) developed and tested an innovative plant for seabed maintenance
characterized by the fact that the main element, called “ejector” (Figure 1b), is an open jet
pump (i.e. without closed suction chamber and mixing throat) with a converging section
instead of a diffuser. The ejector can be used as a fixed or mobile device [7]: the paper
focuses on fixed ejectors application. When used as a fixed device, one ejector works on a
limited area whose diameter depends on the sediment characteristic as, for example, the
angle of repose. By ejectors integration in series and in parallel it is possible to create a
seaway. The paper shows laboratory, first experimental plant and first industrial plant
activities and main results achieved over the years. A spin-off company of Emilia
Romagna Region (Italy), named Plant Engineering Srl, has been created to promote the
widest technology dissemination and to design and manage new industrial applications.
2. LABORATORY DEVELOPMENTS AND TESTING
The one-dimensional jet pump theory is well known and described by several models,
improved and implemented over the years [6, 8]. The ejector suction effect is due by the
behavior of a fluid jet in free outflow from a hole (nozzle diameter d) towards an open
environment. A jet under these conditions increases its flow, from inlet to outlet section,
due to the flow absorbed within the jet itself from the surrounding environment: the high
velocity of the jet creates a low pressure area out of the nozzle leading the pumping of the
second flow toward this minimum pressure point. Consequently, there is an exchange of
momentum between the two streams resulting in a uniform mixed stream flowing at an
intermediate velocity between the primary and secondary flow ones.
On the other hand, it has been demonstrated that jet pump performance calculation needs
to be validated by experimental test in order to overcome the uncertainty on the
determination of the value of some fundamental parameters such as friction coefficients or
convergent-divergent drag coefficient [9, 10]. Starting from 2002 and up to 2013, several
ejectors (Figure 2a and 2b), characterized by different geometries, have been tested in
DIN laboratory to verify theoretical performance in a real environment (Figure 3a and 3b):
inlet and outlet ejector streams pressure were measured by pressure gauges, while inlet
and outlet volumetric flow were measured by level variation in the water and discharge
tanks, respectively. Experimental test has been integrated with computer fluid-dynamic
simulation. The results of the preliminary tests have led to the design and realization of an
ejector (Figure 2b) characterized by an implemented secondary flow nozzles system. The
primary nozzle jet pressure is the one described above and is responsible for the action of
water-sediment mixture suction from the surrounding environment; this latter has been
integrated with a series of inclined radial secondary nozzles, positioned at the two edges
of the ejector and fed by the same duct of the primary nozzle, with the aim of increasing
the solid concentration in the mixture through an effect "spade-shovel" in which the
secondary nozzles shake the sediment on the backcloth and the primary nozzle removes
it.
Figure 2: a) Picture of two of first series ejector prototypes (2002-2004); b) Picture of second series ejector
prototype (2004-2005, the one used in Riccione harbour), during laboratory test.
Figure 3: a) Picture of experimental set arrangement; b) Schematic of experimental set arrangement.
Starting from 2005 and up to 2013, the ejector has been tested in order to forecast its
performance at different working and boundary conditions. Operational characteristics of
the ejector have been described by two jet pump dimensionless ratios, flow ratio Q and
head ratio H defined as Q=QD/QP and H=HD/HP, respectively, where QP and QD are
primary volumetric flow and delivery volumetric flow and HP and HD are primary flow
pressure and delivery flow pressure. Ejector efficiency η has been defined as η=Q×H.
Examples of laboratory test results are shown in Figure 4a and 4b: results are given for
different ejectors (different ratio between nozzle diameter d and discharge diameter D) at
the same inlet pressure and with water-water environment (no sediment). The results in
Figure 4a show that, being D a constant value, a higher d decreases the ejector suction
capacity (measured by Q). On the other hand, a higher nozzle diameter d allows reaching
higher pressure at the converging outlet, thus increasing sediment transport distance.
Furthermore, Figure 4b shows that ratio d/D does not affect ejector efficiency. The results
are in line with the theoretical forecasts for the behavior of the jet pump [6]. Moreover,
ejector performance (Figure 5a and 5b) has been characterized as a function of inlet
pressure (measured by adimensional inlet pressure coefficient p, that is the ratio between
measured HP and maximum primary pressure HPmax allowed) and as a function of
equivalent discharge pipe length (measured by adimensional factor Lp, that is the ratio
between equivalent discharge pipe length L and equivalent maximum pipe length Lmax
allowed) through the suction efficiency Ψ, defined as the ratio between the secondary
volumetric flow QS and the delivery volumetric flow QD, that is the sum of primary flow QP
and secondary flow QS (Equation 1). Figure 5a shows Ψ values for a d/D=0,24 ejector,
while Figure 5b is for d/D=0,32.
Ψ=QS/QD=QS/(QP+QS) (1)
Figure 4: a) H as a Q function; b) η as a Q function.
Figure 5: a) Ψ as a p and Lp function for a d/D=0,24 ejector; b) Ψ as a p and Lp function for a d/D=0,32 ejector.
Figures 5a and 5b show how, in the same boundary condition, the lower is the ejector
nozzle diameter the higher is the suction efficiency. On the other hand, when the ejector
works in real environment, suction efficiency shall be controlled to avoid discharge
clogging risk due to sediment deposition along the pipe. Figure 5b also shows an
interesting property of ejector: when Lp increases due to pressure drop in the discharge
pipeline (for instance, by the presence of sediment), since the ejector is fed at constant
rate, suction efficiency can reduce up to 0%, that means no secondary flow is present and
only clean water (the one that feeds the ejector) is present in the discharge pipeline. So,
when delivery flow becomes critical due to a high sediment transport, the ejector reduces
itself the secondary flow and, consequently, its suction efficiency, thus realizing a self-
control of secondary flow rate.
3. FULL-SCALE EXPERIMENTAL PLANT RESULTS
In 2005 the first full-scale experimental plant (Figure 6a and 6b) has been realized in the
port of Riccione (Italy). 15 ejectors have been provided, with a variable distance between
one device and the other, to cover the 65 meters of the inlet canal (Figure 6a), in order to
maintain a constant depth in the middle of the canal. The authorization process for plant
installation and operation involved three main actors: (1) Riccione Municipality authorized
State property land use, (2) Rimini Port Authority approved the project with regard to
safety of navigation and (3) Emilia-Romagna Region authorized plant installation and
operation without considering sediment discharge as nourishment but only as a sediment
displacement, since the plant moves the sediment that is transported naturally in its area
of influence.
This last authorization represents a powerful advantage over dredging. The experimental
phase went on for the whole summer season and the functioning of the experimental plant
ensured sufficient water depth (over 3 meters) for navigation without dredge operation
[11], event that has never happened before that time. The pump was a 90 kW centrifugal
pump. At the pump suction there was a grid filter. Downstream of the pumping system
there was a purging manual hydrociclones battery (placed inside the pumping system
cabin). Water feeding filtration is necessary since ejector radial nozzles could be subjected
to clogging. A manifold with 15 pipes for ejectors feeding was present downstream the
pump. Along each of the feeding lines one manual valve VI (to balance the lines) and one
Y-filter were present (see experimental plant P&ID in Figure 6b).
The experimental plant didn’t have the characteristic of a continuous working and
automatized plant, since it was realized mainly to test ejectors performance in real marine
condition. So, the plant was put into operation only when the manual operator was on site
and for time intervals between 60 and 210 minutes. Before and after plant operation
bathymetric surveys were performed in order to evaluate plant impact. Figure 7 shows an
example of bathymetric, made before and after experimental plant functioning and a sea
storm. After 75 minutes of functioning, the experimental plant was able to restore the water
depth (4-4,5 m) preceding the sea storm of August 7th and 8th 2005. Finally, the effect of
ejector operation on water turbidity was monitored too: it was demonstrated that the
ejector produces a very limited turbidity zone, confined in a volume of about 200 litres near
the ejector itself, and no sediment resuspension was observed. So, ejector technology
reduces to zero the environmental impact of seabed maintenance if compared with
dredging technology.
Figure 6: a) 3D image of Riccione port experimental plant lay-out; b) Riccione experimental plant P&ID.
Figure 7: 3D image of Riccione experimental plant bathymetric done between August 5th and 9th 2005.
The experimental campaign was concluded successfully. Nevertheless, ejector and more
in general the plant needed further development to reach higher efficiency and industrial
reliability.
4. INDUSTRIAL PLANT RESULTS
In 2012 the first industrial plant was realized in the Portoverde Marina (Italy). Plant
installation and operation authorizations were achieved by the same procedure described
for the plant in Riccione. This plant represents the natural evolution of Riccione
experimental plant: thus, even if the ejector has been further developed, the re-design
process focused on the whole plant engineering with the final goal of realizing a fully
automated and remotely accessible plant. Figure 8 shows Portoverde plant P&ID: two
ejectors were installed to protect the dock inlet. The plant layout is similar to the one
shown in Figure 6a, except for the manifold that is inside the pumping cabin and not along
the wharf. The three main plant implementations are: automatic control of the water
feeding flow rate of the ejectors on two different values, design level (also called “flushing”)
and maximum level; automatic balancing of water feeding flow rate of the ejectors;
automatized cycle for pump downstream filter purging. An inverter and a PLC integrated
the 30 kW centrifugal pump in order to reach the first plant implementation: on each
ejector water-feeding line has been installed a flow meter consisting of orifice plate and
differential pressure transducer (PT4 and PT5 in Figure 8). The PLC compares the total
measured flow rate with the target one and controls the pump inverter in order to adjust
the total flow rate to the desired value. The second plant implementation is realized by the
integration of two control electro-valves (VR in Figure 8) in the PLC flow control system: in
fact, the PLC controls the total flow by acting on the pump inverter and balances the flow
through the two ejectors by opening or closing the electro-valves. By the way it is possible
to automatically manage equal or different flow rates in the two ejectors for flushing and/or
maximum flow condition. The third plant implementation is able to give both high water
feeding filtration and continuity to plant operation. Water filtration is guaranteed by an auto-
purging disk filter with 400 μm limit. Auto-purging cycle is started by PLC when pressure
drop through the filter (measured by PT3 in Figure 8) overcomes a set value. Auto-purging
is also timed.
Figure 8: Portoverde Marina plant P&ID.
The shift from the maximum flow to the flushing one and vice versa is controlled by a PLC
fully automated cycle. The flushing flow rate is the design flow rate and corresponds to the
flow rate able to guarantee the cleaning of ejectors and discharge pipelines. In this
condition the ejector carries out a delivery of slurry that is very poor in terms of solid
matter: so, this kind of operation is suitable for uncritical sea weather condition in terms of
sediment supplying in the area of influence of the ejectors. Nevertheless, the plant
operates continuously 24/7, thus over time produces anyway a positive effect in terms of
sediment removal. On the other hand, when critical conditions are reached (i.e. sea
storm), the water feeding flow rate needs to be shifted on the maximum value (that is
higher than flushing) and that realizes a denser slurry by increasing ejector suction
efficiency. Critical sea weather is recognized by the PLC through the measurement of wind
speed and direction. Moreover, over the cabin a web-camera is installed, oriented on dock
inlet: so, the operator can remotely access to PLC control panel and, after checking the
weather and sea conditions, can change the operating mode. Furthermore, two control
signals on delivery flows have been added. The first one is a turbidity measurement,
realized by photoelectric sensor transmitter and receiver (PER and PET, respectively, in
Figure 8) installed in the pipeline discharge. In presence of the on-off turbidity signal, the
PLC sets the flow rate on maximum value. The second one is a delivery flow
measurement, realized by a Doppler effect flow meter (FS in Figure 8) installed in the
pipeline discharge. In presence of the on-off flow signal, corresponding to the overcoming
of the minimum delivery flow allowed, once again the PLC sets the flow rate on maximum
value. Both photoelectric sensors and Doppler effect flow meter are placed inside a
metallic box installed near the discharge point and their signals are transmitted to the PLC
by a wireless LAN bridge (distance about 150 meters). Besides, in the plant there are a
series of automated cycles to protect the pump and the auto-purging filter: the pressure
transducer PT1 (Figure 8) measures pump suction pressure to verify if critical condition for
pump cavitation are reached, while the pressure transducer PT2 (Figure 8) is used to
relieve both high discharge pump pressure (if pressure is too high it may cause damage
on filter and pump) and low discharge pump pressure, as a detector of a leak downstream
of the pump. Finally, the PLC acquires data recorded on the plant in continuous and saves
them in the form of spreadsheets, downloadable from remote.
Marina Portoverde plant has been tried out from April 26th to September 19th 2012. In
particular, in this period the automated management of the plant was tested and
developed. As results of this test period, some improvements were introduced: PLC
software implementation, selection of different sensors through installing and testing,
pipeline substitution (from stainless steel to polyethylene) and reliable anchoring of
ejectors and pipes at seabed. A further trying out session was realized from February 20th
to March 20th 2013 to verify the new implemented solutions. After plant optimization
process, the attention was refocused on the ejector: in particular, new laboratory test were
carried out to increase ejector energy efficiency and a new updated version was realized.
Starting from September 21st 2013 and up to May 1st 2014, the plant worked continuously
and maintained between 2,5 and 3,0 meters depth at the dock inlet for all the winter
season. Table 1 summarizes design, installation and commissioning costs of the industrial
plant; moreover, Table 1 shows management cost, divided into operating costs, ordinary
and extraordinary maintenance costs, projected into a one year functioning (8.600 h).
Average electric energy consumption is reduced of 41% from about 5,50 kW per ejector
[12] to 3,25 kW. Electrical energy cost is calculated by considering a 0,20 €/kWh cost.
Ordinary maintenance costs are mainly due to pump suction filter manual cleaning (at
least to be performed every six months) and to little maintenance and monitoring activities
realized by sub-personnel on the underwater installation. Finally, extraordinary
maintenance activities were computed: they were mainly due to the sea storm on 10th, 11th
and 12th November 2013, that damaged pipelines anchors on wharf, and to undersea
pipeline breakdown, made by external cause, that had led to the replacement of the
pipeline. An yearly management cost of about 17.000€ is competitive with dredge, with the
great advantage that ejector plant ensures the navigability throughout the year.
Item Quantity
Design, installation and commissioning costs 70.000€
Electrical energy consumption 55.900 kWh/year
Operating cost (energy consumption) 11.180€/year
Ordinary maintenance costs 1.800€/year
Extraordinary maintenance costs 4.000€/year
Table 1: Design, installation and commissioning costs, plus operating, ordinary and extraordinary yearly
maintenance costs of Marina Portoverde plant.
5. CONCLUSIONS
The ejector technology has been demonstrated to be an efficient and reliable system and
proved to be more competitive in the industrial scale than the dredging technology both in
economic and environmental perspectives. A further ejector redesign is currently in
progress with the aim of obtaining even more efficient performance in terms of energy
consumption and of increasing ejector reliability in the long-term operation.
ACKNOWLEDGEMENTS
The research activities and plants design and realization were co-financed by Emilia-
Romagna Region.
REFERENCES
[1] Paul L.A. Erftemeijer, Bernhard Riegl, Bert W. Hoeksema, Peter A. Todd. Environmental
impacts of dredging and other sediment disturbances on corals: A review. Mar Pollut Bull.
2012; 64: 1737-1765.
[2] Paul L.A. Erftemeijer, Roy R. Robin Lewis III. Environmental impacts of dredging on
seagrasses: A review. Mar Pollut Bull. 2006; 52: 1553-1572.
[3] Carlos Vale, Ana M. Ferreira, Cristina Micaelo, Miguel Caetano, Eduarda Pereira, Maria J.
Madureira, Elsa Ramalhosa. Mobility of contaminants in relation to dredging operations in a
mesotidal estuary (Tagus Estuary, Portugal). Water Sci Technol. 1998; 37: 25-31.
[4] Yuchuan Bai, Zhaoyin Wang, Huanting Shen. Three-dimensional modelling of sediment
transport and the effects of dredging in the Haihe Estuary. Estuar Coast Shelf S. 2003; 56:
175-186.
[5] P.K. Boswood, R.J. Murray. World-wide Sand Bypassing Systems: Data Report. Coastal
Services technical report R20, Conservation technical report No. 15. ISSN 1037-4701. 2001.
[6] E.C. McNair Jr. A sand bypassing system using a jet pump. A sand bypassing system using a
jet pump. In: ASCE. Coastal Engineering. Proceedings of Fifteenth Coastal Engineering
Conference; 1976 July 11-17; Honolulu, Hawaii (USA); 1976.
[7] A. Bianchini, M. Pellegrini, G. Preda, C. Saccani, D. Vanni. Integrated technologies for
sediment treatment. Environ Eng Manage J. 2013; 12:253-256.
[8] S.H. Winoto, H. Li, D.A. Shah. Efficiency of jet pumps. J Hydraul Eng. 2000; 126: 150-156.
[9] A.H. Hammoud. Effect of design and operational parameters on jet pump performance. In:
WSEAS. Proceedings of the 4th International Conference on Fluid mechanics and
aerodynamics; 2006 August 21-23; Elounda (Greece); 2006.
[10] Ibrahim R. Teaima. A central-type jet pump model for wheat grains removing from water
channels. In: IWTC. Proceedings of the 16th International Water Technology Conference; 2012;
Istanbul (Turkey); 2012.
[11] G. Amati, C. Saccani. Experimental plant for sand removal from harbor areas seabed
(Impianto sperimentale per il desabbiamento dei fondali nelle aree portuali, in Italian). In:
ANIMP. Proceedings of the Thirty-second Engineering and Plants National Symposium; 2005;
Rimini (Italy); 2005.