· · · Institutional Repository

Introduction In order to improve people’s health worldwide, many efforts have been made in order to meet Millennium Development Goal 7c: reducing by half the proportion of people without sustainable access to safe drinking water and sanitation. Kenya is in the top ten of countries with the largest population without access to safe drinking water (UNICEF and WHO, 2012). Because most of these people live in rural areas, large investments are done in the Kenyan rural water supply. But recent studies show that many of the new water supplies stop functioning within a few years after implementation (MWI, 2007 and RWSN, 2007). Causes for this low ‘post-construction sustainability’ can be technical, institutional, financial, social or environmental. One of the most critical factors which is mentioned in literature is an adequate financing of operation and maintenance. This Master Thesis is about the post-construction sustainability of rural water supplies in Western Kenya, with a specific focus on the financial part of it (or financial sustainability). In the rural water supply practice in Western Kenya, several water supply technologies exist. Some of these technologies require hardly any operation and maintenance (O&M), like springs, surface water catchment, rain water catchment and a well without a pump. These technologies are not included in the current research. Remaining technologies are a handpump and a motorized pump, both used for groundwater pumping. Apart from the differences in technologies, several management models for rural water supply exist within Western Kenya: community management, government management, private management and institutional management. The latter one is not included in the current research because at these locations serving the community is in general not the main purpose. As the access to clean and safe water in adequate quantities is recognised as a human rights issue in Kenya (Constitution of Kenya, 2010), mechanisms for finding sustainable service delivery is a key national priority. As different management models are likely to result into different levels of sustainability, government of Kenya is in search of a most sustainable model for Kenyan context. Objective The objective of this study is to compare the financial sustainability of rural water supplies in Western Kenya. Within this comparison the aim is to compare different technology types, different management models and different combinations of these two. The final goal is that this comparison can be used by the Government of Kenya and other supporting entities in the development of policies and projects for the rural water supplies. Methodology Data for this study is collected during interviews with the responsible persons for the water supplies. Data is collected about service level, O&M, financial management, cost recovery and finances. Service level includes system functioning, water quantity, walking distances and water quality. O&M includes who is responsible for the daily operation and pump check and for the maintenance arrangements and the days it takes between a breakdown and a repair. Financial management includes: responsibility for the finances, water tariff, tariff structure, bank account, bookkeeping and service cut-off for non- payment. Cost recovery includes the practice of the payments, the extent in which the income covers the O&M costs and whether replacement is expected to be a problem on the long term. The finances include the yearly income, costs and costs per user. To all above mentioned factors scores are assigned depending on the output per criteria. The scores are also given a weighing factor. In this way, for every water supply a weighted score can be determined for all the four sustainability categories. In total 27 handpumps and 25 motorized pumps were evaluated. Conclusions and recommendations Out of all handpumps, the locations with community management and the locations with combined community and government management scored low. The communities were not able to collect enough money to keep the system functioning on the long term. The private managed handpumps scored good, especially in terms of cost recovery and quick response to breakdowns. The motorized pumps scored low at the locations with combined community and government management and at the locations with government management. At the combined managed motorized pumps the responsibilities for O&M and financial management were not clear. At the government managed motorized pumps the payments were not good enough to cover the costs. At the community managed motorized pumps, the committees were well organized but they did not manage to make all users pay. At the privately managed motorized pumps, the responsibilities for O&M and financial management were not clearly defined but the financial situation was good. There was enough money for the O&M and for replacement on the long term. Comparing the two technologies, the handpumps score higher on cost recovery and the motorized pumps score higher on O&M and financial management. At the handpumps it happens more often that the regular money collection is neglected. The responsible entities at the motorized pumps have more need to be organized because of the daily need for staff and money for e.g. fuel refilling. A negative side of the motorized pumps are the high costs per user per year, about nine times higher than at the handpumps. Comparing the four management models, the differences were not big. The community managed locations have difficulties with making people pay. At the combined managed systems the responsibilities for O&M and financial management are not clear. At the government systems the costs (including high salaries of government staff) are too high for the amount of users. The privately managed systems score slightly higher, especially in terms of cost recovery. Because of the fact that the water quantity is not sufficient at many locations and the walking distances are large, more water supplies are needed in the research area. It is recommended to focus more on handpumps than on motorized pumps for new water supplies. The reason for that are the high yearly costs at the motorized pumps. These costs make the motorized pumps less suitable for the rural areas of Western Kenya, where domestic income is low and people are not open to pay for their water. In some situations, with low water tables or high population density, a handpump is not feasible and than a motorized pump can be a good option for rural water supply. It is also recommended that action is taken in order to improve users’ willingness to pay. Four recommendations are: - Activities for economic development like job creation and microfinance projects. When people get to spend more, they become more open to pay for water. - Training in communities about the importance of clean water, which is not free. This includes basic insight in costs of water supply technologies. - Training for responsible entities about dealing with sanctions against non-payment and about making finances more transparent. In order to improve community management it is recommended that costs and responsibilities are shared within communities, local authorities and the central government. In the current situation, especially the tasks of the local authorities are not fully recognized: financing a part of major repairs of water supplies, monitoring the performance of individual facilities, conflict and problem resolution and retraining of mechanics and communities. For their monitoring task, the current study and other studies are used to constitute a basic half yearly performance monitoring. Since the private management model scores high on financial sustainability, it is recommended that the Government of Kenya and development partners pay more attention to this option. In order to create a situation where private management is a serious option, several aspects need to be considered: - The government need to contribute in investment costs. For e.g. the handpumps, they can contribute in the same way as in the current programmes with the community managed handpumps where the government pays 65% of the investment costs. - Community sensitization is required about the option of a private handpump. People need to know about this option. And they need insight in the costs and possible revenue. - Training is required for private owners of a water supply about water supply technologies, maintenance and dealing with financial management. - Formal recognition and regulation of such private investors is necessary as they will be running water systems as businesses.

The objective of this research is to limit the risks of fully automated operation of drinking water treatment plants and to improve their operation by using an integrated system of process models and emulated process automation software. This thesis contains the design of such an integrated system. The use of the system is investigated in the three identified applications, i) optimization of process control, ii) training of operation supervisors and iii) virtual commissioning of process automation software.

Phosphorus depletion is an emerging and serious global environmental issue. At this moment, research and policy discussion on phosphorus scarcity is still limited. This research investigates the possibilities of phosphorus recovery by controlled precipitation of struvite from digested sludge. At Waternet’s wastewater treatment plant Amsterdam West, plans for a struvite reactor are scheduled. The advantage of phosphorous recovery through struvite precipitation from digested sludge is three-fold. First, struvite can be directly used as fertilizer. Second, undesirable struvite precipitation in the wastewater treatment plant is prevented by reducing the phosphorus concentration in the dewatering reject stream which is fed back to the inlet of the treatment plant. Third, sludge dewaterability improves due to the addition of MgCl2. This thesis investigates the influence of mixing speed, aeration rate, magnesium dosing method and crystal recycle method on struvite growth and phosphorus removal, as well as separation of struvite from sludge. For that purpose, experiments have been performed in a crystallization reactor and a counter-current washing column at lab scale at wastewater treatment plant Amsterdam West. MgCl2 was added under varying reactor conditions, struvite constituent concentrations were measured and struvite growth was assessed. First, it is demonstrated that struvite recovery is well possible in a stirred sludge environment at neutral pH commonly applied in sludge digesters (7.0 - 7.1). Phosphorous removal under these circumstances is at least 85%. More complete mixing by stirring at a higher speed further improves struvite recovery by keeping supersaturation low. Secondly, a significant difference in struvite recovery was observed between experiments in which MgCl2 is dosed instantly versus experiments in which MgCl2 is dosed gradually. Gradual MgCl2 dosage, and therefore rapid mixing, improves recovery compared to instant dosage. Mixing at a higher stirring speed further improves recovery. Thirdly, it is found that struvite recovery under given circumstances is poor in a combined aerated and stirred sludge environment. In such environment higher aeration rates deteriorate struvite recovery further, while struvite recovery improves with decreasing aeration rates at a higher stirring speed. Fourthly, struvite separation experiments have verified that separation is well possible in a counter-current washing set-up, separating 86% of detectable struvite within 15 minutes at an upflow velocity of 1.3 mm/s.

The drinking water in the Netherlands is of high quality and the production cost is low. This is the result of extensive research in the past decades to innovate and optimise the treatment processes. The processes are monitored and operated by motivated and skilled operators and process technologists, which leads to an operator-dependent, subjective, variable and possibly suboptimal operation of the treatment plants. Furthermore, the extensive automation of the treatment plants reduces the possible operator attention to the individual process units. The use of mathematical process models might solve these problems. This thesis focuses on the application of models in model-based monitoring, optimisation and control of drinking-water treatment plants, with the Weesperkarspel treatment plant of Waternet as a case study. To shift the operation of drinking water treatment plants from experience driven to knowledge based, a model-based approach is shown to be effective. Models are successfully used in plant analysis and basic control design, resulting in the successful implementation of new basic control for the softening reactors at the Weesperkarspel plant. Model-based monitoring schemes abstract relevant information from the large amount of data and the schemes estimate the current state of the processes. Model-based control uses the monitored process state to dynamically optimise the treatment without introducing new disturbances in the treatment plant. Model-based optimisation gives the process technologist the possibility to improve treatment operation without disrupting the full-scale plant.

gReverse Osmosis systems are wildly used for sea water desalination and water reclamation but they meet problems regarding bio-fouling. Bio-fouling increases their work pressure and operational costs and decreases their removal efficiencies. Phosphorus limitation is one strategy to control bio-fouling. This report focuses on phosphorus removal by ceramic tight ultra-filtration (CTUF) membranes (1kD and 3kD MWCO) as a pretreatment before RO. In this research we investigated different factors affecting phosphorus rejection by CTUF such as flux, cross flow velocity, ion strength, zeta potential and pH. The results show that increasing the flux, the cross flow velocity and the zeta potential increased the removal rate of phosphate. Increasing ion strength decreased the double layer thickness and decreased the removal rate of phosphate. All these results are in agreement with the theory of membrane filtration. The pH affected both the zeta potential of the membrane and the charge of the phosphate ion. Increasing pH increased the removal rate of phosphate but after pH 8.3, the removal rate of phosphate began to decrease. There is no explanation for this decrease in rejection at higher pH at this moment.

Drinking water production from anaerobic groundwater is usually achieved by so called conventional techniques such as aeration and sand filtration. The notion conventional implies a long history and general acceptation of the application, but doesn’t necessarily mean a thorough understanding of the processes involved. This is certainly the case for groundwater filtration, with groundwater being the major source for drinking water production in the world. During infiltration and soil passage groundwater may become deeply anaerobic and be loaded with methane, iron, ammonium and manganese that have to be removed during drinking water production. The removal processes for these compounds may be physical-chemical or biological. Since the important PhD-researchers Lerk and Graveland used a chemical approach in the 1960th, the general perception is that only methane and ammonium removal is biological under environmental conditions. Biological iron and manganese removal would be more exceptional as a result of the specific conditions required. The origin of this PhD lies in a persistent nitrification problem in the filters of Oasen Drinking Water Company. When no strong chemical oxidizers are used, like in the Netherlands, nitrification is the only applicable process to fully remove ammonium. It comprises the two-step biological conversion of ammonia via nitrite to nitrate. The first step, the microbial oxidation of ammonia to nitrite, becomes incomplete during the aging of the filter. The relapse typically becomes visible three to six months after the startup of a filter with new filter material. Oasen has only one technique to counteract these nitrification problems, namely subsurface aeration. In that technique, a limited amount of aerated water is periodically injected into the groundwater aquifer, resulting in in situ iron oxidation. The iron colloids that are also formed in the aquifer stimulate the nitrification in the filters, but the working mechanism is unknown. As this “Wonder van Nieuw Lekkerland” is not understood and restricted by licenses, Oasen looks for alternative techniques to maintain a sound nitrification. The general hypothesis for this problem was that the nitrification problem resulted from the interaction with the other removal processes. The problem was studied in full-scale filters and lab-scale setups. Molecular techniques, such as DGGE and clone libraries, were used to identify the major groups of microorganisms present in the groundwater and filters. Ammonia oxidation is performed by Nitrosomonas and archaea, nitrite oxidation by Nitrospira, while Nitrobacter bacteria are not found in drinking water filters. Notable was the presence of the iron-oxidizing Gallionella bacteria in the subsurface aerated groundwater. Another molecular technique, quantitative PCR, was used to quantify Gallionella and ammonia-oxidizing bacteria and archaea in all incoming and outgoing water flows and attached to the filter material. From these numbers balances were made for the filters. The activity of the ammonia-oxidizers was assessed in standardized batch experiments. One of the major findings was that Gallionella grew extensively in a groundwater filter with nitrification problems. In a filter fed with subsurface aerated water, however, Gallionella pumped up with the subsurface aerated water did not continue to grow in the filter. In fact, clone libraries showed, that Gallionella growing in situ deviated from the ones growing in the filter. The growth of Gallionella in well-ventilated trickling filters is remarkable, because these organisms are supposed to be micro-aerophilic. Trickling filters are used for their efficient gas transfer, resulting in effective aeration and stripping of methane and carbon dioxide. The effluent water has a pH of 7.5 to 8 and is almost saturated with oxygen. To verify the growth of Gallionella under these conditions Gallionella bacteria were cultured in continuously operated oxidation and filtration columns (Figure 1). These experiments confirmed the growth of Gallionella under oxygen saturated conditions and at a pH up to 7.7 (Figure 2). The balance approach of ammonia-oxidizing bacteria (AOB) showed that the nitrification problem was not caused by the excessive washout of these microorganisms. In fact, the number of AOB was higher in a filter with nitrification problems, but their activity was much lower. This low cell-specific activity was caused by limitation of the essential nutrient phosphate that could be corrected by addition of phosphate (see Figure 3). So what is the relation between the growth of Gallionella and nitrification problems? While phosphate in groundwater is readily removed for the greater part by co-precipitation with chemically formed iron oxyhydroxides, the biogenic iron precipitates of Gallionella have a higher adsorption capacity for phosphate and further lower the phosphate concentration to limiting levels for the growth of AOB. The outcome of this PhD research provides solutions for the groundwater nitrification problem (such as phosphate dosage and suppression of Gallionella growth) and perspective to further optimize trickling filtration, the most efficient process to remove methane and iron from groundwater.

Advanced Oxidation Processes (AOPs) are innovative, cost-effective, catalyzed chemical oxidation processes for treating pollutants in low or high concentration from contaminated soil, sludge and water. The common used AOPs in drinking water treatment include UV/H2O2 process, UV/Ozone Process, UV/Titanium Dioxide and Fenton’s Reagent. AOPs are ultraviolet driven, which share predominance from photochemical technology, and often, give the clients dual benefit of both environmental contaminant treatment and disinfection. Endocrine Disrupting Chemicals (EDCs) are very disturbing contaminants measured in natural waters. Phenols and their tert-butyl derivatives are important contaminants belonging to EDCs. After a successful workshop for developing alternative drinking water treatments at Shanghai, AOPs with UV/H2O2 technology are chosen to remove 4-tert-butylphenol (4TBP) from Shanghai water. The kinetics of reaction was studied in the first part of this thesis. The results show that UV/H2O2 process can effectively decrease 4TBP concentration than hydrogen peroxide alone. Good free oxidation radical production can be achieved within UV dose range from 0 to 200mJ/cm2 by a low pressure mercury lamp. The 4TBP degradation process fits with pseudo first order equation for UV-dose and H2O2-dose. However, at very high H2O2-doses, the scavenging of hydroxyl-radicals needs to be taken into account. Computational Fluid Dynamics (CFD) modeling of UV reactor and validation of CFD model were studied in the second part of this thesis work in order to provide an applicable UV reactor design for the 4TBP treatment and also give possible reactor improvement suggestions. The CFD model used in this study is a 2-D model developed using software Comsol, V3.3a, based on a current UV reactor design at Kiwa Water Research, the Netherlands. The developed UV dose model includes three parts, a k-ε flow model, a UV intensity model and a random walk model. Different feed flow rates and different lamp configurations were studied by the model. The calculation results show that a higher feed flow rate contributes to a relative narrow UV dose distribution than the lower flow rate. With three lamp configurations, position 0 is the best among the three with the highest average UV dose as well as the narrowest dose distribution pattern. Model also predicted low pressure lamps have about 8% higher power output to UV dose efficiency than medium pressure lamps. Validation of the flow model was helped by flow measurements at Delft University of Technology. Experimental studies of velocity measurements by Laser Doppler Velocity Meter were conducted together with salt and dye dose experiments. After comparisons of model predictions and experimental measurements, it was found that the k-ε CFD flow model demonstrated generally good qualitative prediction of flow inside the reactor but failed to give correct prediction of recirculation zones behind the quartz tubes. There are dead zones of water at the top and bottom near the inlet of the reactor. Bigger areas exist behind the quartz tubes that have water recirculation than the model predicted, which may result 25% of more UV dose prediction by the model. And differences caused by 2-D model and 3-D measurements may result about 20% less UV dose model prediction. Current UV reactor design at Kiwa Water Research, position 0 and low pressure mercury lamps applied at a feed flow rate of 4.1m3/h appears to be an applicable design for advanced oxidation treatment of 4TBP by UV/H2O2 in Shanghai. High roughness quartz tubes walls and relative smaller ratio of reactor to feed pipe diameters are recommended to improve reactor performance in the recirculation and dead zones with current design. Further investigations of the dose model and UV-sensitive dyed microspheres particle tracking experiments are recommended.

WRK III zuivert IJsselmeer water tot een halffabrikaat (coagulatie, sedimentatie, filtratie en gedeeltelijk actievekoolfiltratie) en transporteert dit water naar PWN, voor infiltratie in de duinen, en naar Hoogovens. Bij het transport en zuivering van het water wordt steeds meer gebruik gemaakt van online monitoring en automatische sturing. In dit project is gekeken of met behulp van de beschikbare gegevens het transportsysteem en de zuivering geoptimaliseerd kan worden.

Voor de uitbreiding van de oppervlaktewaterzuivering zijn de alternatieven diepinfiltratie en ultrafiltratie onderzocht.

Powdered Activated Carbon (PAC) has been successfully used in conjunction with membrane microfiltration (MF) to reduce taste, odour, colour and other concerns caused by organic material present in raw drinking water sources. PAC addition also significantly reduces the extent of fouling in membranes. And the operation of continuously dosed PAC before membrane filtration is widely applied in drinking water treatment while the operation of PAC precoated on the membrane is a new approach. This research, therefore, aims at evaluating the membrane fouling on these two different PAC operations. Water from the Schie canal in Delft was coagulated with Iron (Ⅲ) Chloride and enhanced with PAC, then filtered through a ceramic membrane. The trans-membrane pressure was monitored during filtration. Samples of the raw water and the permeate water were collected and analysed for DOC and UV254, while LC-OCD tests were carried out on the collected samples as well. In addition, the the backwash water and chemical cleaning solutions after soaking membrane were analysed. The PAC in the continuous dosing channel was dosed before flocculation and the membrane was frequently backwashed every 19 minutes for 1minute. However, the PAC in the precoating channel was precoated on the membrane before the experiment and operated without frequent backwashes. The results show that the membrane fouling in the continuous channel was more severe than that in the precoating channel which was mainly concluded from the recovery of TMP. That is because the different PAC addition modes lead to a different cake structure on the membrane surface. The cake layer in the continuous channel was mostly non-backwashable while it was easy to be backwashed in precoating channel. Recovery in the precoating channel was quite good with a recovery of 99.2% while the continuous channel presented a normal recovery of 85.5%. The DOC and UV254 removal for these two operations was similar. Also, the LC-OCD results of the backwash water and the chemical cleaning solution reveal that, if ignoring the cake fouling, the total fouling for the two operations were not too much different while big difference existed in compositions of NOM. Specially, about 40% of the neutrals fraction and 35% of humic substances fraction were found in sodium hydroxide solution after the membrane had been soaked 24 hours in continuous channel. Whereas, the first two important foulants in precoating channel were humic substances and bio-polymers, with approximate 33% and 19% fraction, respectively.

Ion exchange resins have found an increasing application in the drinking water treatment sector over the last few decades. The ion exchange resins have a positive ability to remove the charged natural organic matter (NOM). To apply this process in full-scale treatment, the most suitable resin has to be selected and the hydraulic behaviour of the treatment process must be known. This study has the purpose to study the NOM fractions removal with different resins and select the most suitable resin for NOM removal of the Weesperkarspel water. Another purpose is to study the hydraulic behaviours of fluidized ion exchange and testing of the fluidization models of Ergun and Richardson-Zaki. The thesis study is characterizing NOM into the specific fraction and later observing the removal of each fraction with the different ion exchange resins. The kinetic and the equilibrium of NOM removal of each resin are also studied. Four ion exchange resins are tested (Lewatit VP OC 1071, Purolite Macronet 200, Duolite A7 and MIEX DOC). The ultraviolet absorbance method (UV), dissolved organic carbon detection method (DOC), specific ultraviolet absorbance (SUVA), fluorescence excitation emission matrix (fluorescence EEM) and liquid chromatography-organic carbon detector method (LC-OCD) were applied. Weesperkarspel water contains a high degree of aromatic NOM fractions, mostly in the form of humic substances. The Lewatit VP OC 1071 is the most suitable resin for removal of NOM in general view, especially aromatic NOM and humic substances fractions. It appears to be removed as high as 65% and 94% respectively. The MIEX DOC removed 57% aromatic NOM fraction and 70% of humic substances. The pH was found as the dominant parameter for the NOM removal by the weak base Duolite A7. In normal raw water (pH = 7.8), this resin is almost ineffective while it can remove the aromatic NOM up to 35% and humic substances fraction of 45% in raw water pH adjusted to 5. The sorbent Purolite Macronet 200 can remove only biopolymers and neutral NOM fractions. Adsorption is an important mechanism for the removal process of high molecular weight NOM fractions and the neutrals fractions. By using the linear driving force model (LDF) to describe the ion exchange rate of the resin, it was found that MIEX DOC resin can remove NOM faster than other resins. The LDF- k constant of MIEX DOC is also higher than other resins. MIEX DOC has the smallest resin bead size. This can be the reason for the fast removal of NOM. The resin exchange capacity is related with the Freundlich constant K and n . Increase of K and n values lead to increase of exchange capacity. The Lewatit VP OC 1071 has the highest K and n values and thus the highest exchange capacity. The exchange rate and exchange capacity is a specific property of each resin. With help from the LDF model and Freundlich isotherm the breakthrough curve can be determined. The Lewatit VP OC 1071 has longest running time compared with the MIEX DOC and Duolite A7 due to the highest exchange capacity. The hydraulics behaviour of fluidized bed ion exchange has been investigated with the strong base gel resin Lewatit VP OC 1071. The temperature and the feed velocity influence the expanded behaviour. The wet density and wet porosity are important parameters for the model prediction. Combination of mathematical modelling of ion exchange and the treatment conditions of Weesperkarspel drinking water treatment plant, the fluidized ion exchange process can be designed. For Weesperkarspel drinking water treatment plant, the aim is to decrease the DOC concentration of 7.2 mg C/l to 3.0 mg C/l with the fluidized ion exchange process. The 20 fluidized ion exchange reactors with a height of 9 m, 10 m2 of resin bed surface area, a bed height of 2 m and a feed velocity of 20 m/h are designed. The reactors consist of 4 groups, each group is started with delays of 15 days. With this operational step, the running time of each reactor is 60 days. The cost is estimated 11-euro cent per m3 treated water.

Iron is the primary source for discolouration problems in the drinking water distribution system. The removal of iron from groundwater is a common treatment step in the production of drinking water. Even when clear water meets the drinking water standards, the water quality in the distribution system can deteriorate due to settling of iron (hydroxide) particles or posttreatment flocculation of dissolved iron. Therefore it is important to remove dissolved and particulate iron to very low levels. The objective of this study was to reduce the particle load towards the distribution system and to improve the iron removal at the groundwater pumping station Harderbroek, consisting of aeration, rapid sand filtration and tower aeration. A maximum flow of 1800 m3/h can be treated and the average production is 800 m3/h. Although previous research showed the clear water meets the drinking water regulations, the drinking water company Vitens is not satisfied with the turbidity and iron concentration of the clear water. The mean iron concentration is 0.04 mg/l.

Het kabinet is van plan met het in werking treden van de nieuwe waterleidingwet het monopolie op het gebied van drinkwaterlevering voor zakelijk grootverbruik op te heffen. Door het aanbieden van pasklare oplossingen in de vorm van ‘Water op maat’ of ‘ander water’ wil de drinkwatersector grootverbruikers echter als klant behouden en voorziet zij tevens in de concrete behoefte van bedrijven om tot outsourcing van de wateractiviteiten over te gaan; watervoorziening is voor bedrijven immers zelden een kernactiviteit. Door deze vorm van klantenbinding is een bepaald minimum aan inkomsten gegarandeerd en is de kans op grote fluctuaties in de drinkwaterprijs kleiner.

De zorg voor de waterketen is in Nederland in vergelijking met het buitenland sterk versnipperd. De schaal van de betrokken organisaties is klein (regionaal of lokaal), en verschilt bovendien sterk per deeltaak van de waterketen. Alle onderdelen van de waterketen voor huishoudelijk afvalwater zijn in publieke handen. De drinkwatervoorziening middels NV’s in eigendom van de overheid (waterleidingbedrijven), de afvalwaterketen wordt rechtstreeks verzorgd door overheden, te weten gemeenten (riolering) en water- en zuiveringsschappen (afvalwaterzuivering). Voor de gemeenten en de waterschappen zijn de waterketentaken slechts een deel van hun activiteiten, die daarnaast ook beleidsmatige en regulerende taken omvatten. Grote industriële afnemers hebben de keuze zowel hun drinkwatervoorziening als hun afvalwaterzuivering in eigen beheer laten uitvoeren door particuliere bedrijven in plaats van gebruik te maken van de publieke voorzieningen.

Bacteriën van de Nitroso-groep kunnen ammonium oxideren tot nitriet. Bacteriën van de Nitro-groep kunnen nitriet verder oxideren tot nitraat. De keten van beide processen wordt nitrificatie genoemd. Ruwwaterbronnen in Nederland bevatten vaak enkele mg/l ammonium. In deze concentraties is het ammonium op zich niet schadelijk.

Gemeente Amsterdam Waterleidingbedrijf (WLB) heeft twee productielocaties, Leiduin en Weesperkarspel. In Weesperkarspel is één van de zuiveringsstappen desinfectie door middel van ozondosering. Het water dat in Weesperkarspel gezuiverd wordt, wordt gewonnen in de Bethunepolder en voorgezuiverd in Loenderveen. Hierbij wordt ozon geproduceerd uit droge lucht en wordt dan gedoseerd in een bellenkolom. Echter, door de hoge concentratie aan organisch materiaal (DOC) in het water, wordt een groot deel van de gedoseerde ozon niet gebruikt voor het beoogde doel, desinfectie, maar reageert met dit organisch materiaal. Hierbij ontstaat makkelijk afbreekbaar materiaal (AOC), dat geconsumeerd wordt door de bacteriën op de actieve koolfilters. Als dit AOC niet verwijderd wordt is het een belangrijke voedingsbron voor bacteriën in het distributienetwerk. In de zomermaanden wordt er zoveel AOC gevormd dat de actieve koolfilters te vaak verstopt raken door de biomassa en dat er dus te vaak teruggespoeld moet worden. Daarom is besloten om de ozondosering in de zomerperiode te verminderen. Echter, door deze verlaging van de ozondosis is het niet aantoonbaar dat de desinfectie op Weersperkarspel goed werkt en zal waarschijnlijk de Ct-waarde verhoogd moeten worden. In dit onderzoek is de invloed van DOC-verwijdering op de ozonisatie onderzocht. Ook is onderzocht of er een invloed is van de soort DOC die verwijderd wordt op de ozonisatie en of het mogelijk is om met een lagere ozondosering te werken wanneer de DOC-concentratie verlaagd is. Hiertoe zijn er experimenten gedaan in de proefinstallatie zonder DOC-verwijdering en met DOCverwijdering door actieve koolfiltratie en ionenwisseling. Uit het onderzoek is gebleken dat actieve koolfiltratie alle fracties van het DOC in dezelfde mate afvangt, terwijl de gebruikte ionenwisselaar vooral de humuszuren verwijderde. Dit was terug te zien in de lagere SUVA-waarden, maar ook in de DOC-karakteriseringen was dit heel duidelijk waarneembaar. In Ct, AOC- en bromaatvorming was er echter vrijwel geen verschil te zien tussen water dat was voorbehandeld met ionenwisseling of met actieve koolfiltratie. Wel was de invloed van DOC-concentratie zichtbaar op de Ct, bromaat- en AOC-vorming. Over het algemeen kon gezegd worden dat zowel Ct als bromaatvorming stijgt, naarmate er minder DOC in het water aanwezig is bij eenzelfde ozondosering. Er wordt meer AOC gevormd bij hogere ozondoseringen. Het verschil is groter bij hoge DOC-concentraties. Verder heeft een DOC-concentratie een grote invloed op de standtijd van actieve koolfilters. Bij een verlaging van de DOC-concentratie van 6 naar 4 mg/l, de wordt de standtijd van de actief koolfilters ruim 4 maal zo lang kan zijn. Dit zal nog wel met behulp van experimenten onderzocht moeten worden.

Door wijzigingen in de inrichting van de drinkwatersector en veranderde eisen van de consument is de wens ontstaan de drinkwaterproductie verder te optimaliseren. Anderzijds zijn onder invloed van de ontwikkelingen van de computer, modellen ontwikkeld die de verschillende facetten van de productie beter beschrijven. Gebruik van deze modellen in de sturing van de productie is één van de mogelijkheden om optimalisatie te realiseren. Ten aanzien van de hoeveelheid water (kwantiteit) zijn dergelijke modellen inmiddels met succes opgenomen in de sturing van de productie. Zo wordt de werkelijke productie bepaald op basis van een gemodelleerde voorspelling van de vraag. Inmiddels zijn voor de beschrijving van het waterkwaliteitsverloop in de verschillende zuiveringsstappen modellen aanwezig of in ontwikkeling. Deze modellen geven een beter inzicht in de werking van de installatie. Gebruik hiervan in de sturing (als Model Predictive Control) kan leiden tot kwalitatief beter drinkwater en minimalisatie van de kosten. Daarnaast ontstaat door middel van continue monitoring een betere informatievoorziening met betrekking tot de noodzaak van onderhoud. Invoering van een dergelijke Model Predictive Control houdt een ingrijpende wijziging in de bedrijfsvoering in. In dit onderzoek zijn de mogelijkheden voor optimalisatie van de snelfiltratie van pompstation [PS] Loosdrecht beschreven. Daartoe zijn in het pompstation metingen verricht om de zuiveringsprestaties en de variaties daarin in kaart te brengen en is het programma “Stimela” gebruikt om het gevonden gedrag te simuleren. Aan de hand hiervan worden de mogelijkheden tot verbetering en de beperkingen nader uitgewerkt. PS Loosdrecht zuivert grondwater door middel van achtereenvolgens versproeiing, droogfiltratie, torenbeluchting, vlokmiddeldosering en natfiltratie. Hoge ijzergehaltes in het rein water en een daarbij behorende hoge vuillast worden als de belangrijkste problemen van deze zuivering gezien.