Discussion
Despite lacking sufficient data and understanding of the Case study site to model this concept, it will still be discussed in more detail here.
First, the implementation of this system will be considered. Before being stored wastewater entering the plant would need to go through mechanical pre-treatment and primary clarification to remove grit and solids to prevent a build-up of settled sludge within the tank and the water becoming sceptic. Although this will require energy on arrival and additional energy will be needed to pump the wastewater in and out of the flow equalisation basin the timing of the aeration process acts as the primary mechanism for load shifting with this process being the most energy intensive. Exactly how long this water could safely be held before being sent for aeration requires further study, but from discussions with Scottish Water it is thought there should be no major issues with holding this wastewater for up to 6 hours provided solids have been removed.
The WWTW in the California case study made use of two pre-existing flow equalisation basins. (2) These basins, placed between mechanical pre-treatment and primary treatment, are used to control the flow of wastewater into WWTWs by dampening variations in flow to the site with the aim of achieving a more consistent flow through the WWTW. (17) Considering Fig.12 and the discussion in the “modelling problems” section something similar may already be employed at our case study site resulting in the relatively consistent energy demand at the site. If this is the case the viability of this option changes significantly as the potential benefits may be achieved without the disadvantages discussed below. Considering WWTWs are designed to deal with the increased demand during wetter months, if our case study WWTW site employs equalisation flow basins to dampen the increased flow from storm events during this time, and these tanks are the reason behind the fairly consistent energy demand, by rethinking how these operate within the plant they could be used to implement this idea of storing wastewater as a means of matching demand with supply. Rather than being used to create a consistent flowrate and energy demand they could operate in the opposite way and create periods of increased demand at desired at times.
Again, considering the California case study, the water utility was paid to remove demand on the grid at peak times by making use of these basins to limit flow through the plant. (2) This occurred at a known set time and so could be scheduled into the plant’s operations. Attempting to use this flow equalisation basin method to match demand with supply adds an extra difficulty due to the stochastic nature of wind generation and the time needed to ramp up operations. Accurate weather predicting and estimates of wind power would be required although systems could be made that automate this process diverting wastewater to and from the basins as needed to maximise use of the generated power. Another consideration is a period of increased wind and renewable generation may not last for the full 6-hour aeration process. Factors such as this and their influence on the effectiveness in employing this method could have been examined through a model, however given the data available that was not possible.
Even without a full model an idea of the potential load shifting available from implementing this system can be estimated along with the required basin size by making use of the earlier assumptions and work. Assuming the capacity of the plant in terms of energy found in the “Case study site: WWTW capacity” section holds, the potential available flexibility in operations for a given hour can be found by subtracting the energy consumption for that hour from the 54kWh assumed maximum capacity. Following this the maximum sum of available capacity for any given 6-hour consecutive period, the assumed time wastewater can be stored, was found to be 144.8kWh. This increases to 180.8kWh if the capacity is taken to be 60kWh with this upper capacity considered here to allow a better understanding of the potential of this system. These values are the maximum flexibility during any 6-hour period considering the two capacities discussed and the equalisation flow basin should be sized accordingly to hold the corresponding volume of wastewater.
Although the maximum flexibility in operations taking the capacity to be 54kWh is 144.8kWh this will vary throughout the year particularly with the seasons. Fig.14 shows a histogram of the available capacity for all 6-hour consecutive periods of a year. It can be seen that the typical available capacity in any 6-hour period is between 80kWh and 120kWh with a median value of 92kWh. Note the negative values correspond to extreme weather conditions causing the plant to operate consistently above 54kWh. An average value of 86.4kWh is found by taking any negative values to be zero.
Histogram of the available capacity at the case study WWTW site for every 6-hour consecutive period in a year.
Assuming the site does not already have flow equalisation basins installed, from the maximum possible flexibility found above a rough estimate of the required tank size can be made. Considering the energy time profile in Fig.8, the first 6 hours occur before storage. Therefore, the energy required to process a metre cubed of stored wastewater is given by the last 8 hours of the profile, 0.3623kWh. This gives a volume of wastewater of 399.7mᶟ or 499.0mᶟ depending on if the 54kWh or 60kWh capacity is used. The minimum area for an open top cylinder is found when the radius and height of the tank are equal. (18) Therefore, the tank required to meet the case study sites needs and minimise materials would have a radius and height of either 5.03m or 5.42m. Fig.15 gives an idea of the scale of these tanks in context, showing the WWTW case study site and what a 10.06m diameter tank would look like amongst the existing infrastructure. (19)
WWTW case study site, as seen on Google Maps with a 10.06m diameter white circle to show the scale of an appropriately sized flow equalisation basin.
A smaller tank may be more desirable and beneficial when all factors are considered but looking at the ability to load shift in isolation, to maximise this benefit the tank at the case study site would be required to hold between 273,400 and 343,900 litres of wastewater again depending on the value assumed for the capacity. This method requires large flow equalisation basins, and this introduces issues with sustainability and cost when compared to more conventional energy storage options. The exact construction of such tanks requires further study, but they are likely to be either metal or concrete and would likely require civil works and incur significant costs. This civil works and materials will also incur a significant carbon cost. However, as lithium-ion batteries are not currently a sustainable solution, with a limited amount of lithium available to mine, no economical way of recycling, (20) and a solution that may only last 10 to 20 years before being replaced, an argument could be made for the use of these tanks as they could last 60-100 years, (21) particularly if they can also be used or are already in place to handle storm surges. However, for small sites such as the case study site, installing these flow equalisation flow basins to store wastewater does not seem like a viable energy storage solution when compared to conventional lithium-ion batteries that can be installed quickly, easily, and cheaply, whilst being capable of storing more energy and being significantly simpler to operate effectively.
Although likely not a viable option when looking to install energy storage solutions at existing sites, the financial as well as carbon costs from construction could be reduced if such a system were designed into the WWTW at inception. Given Fig.12 these basins may in fact already be installed and could be one of the many tanks seen in Fig.15 in which case the operation of the plant could be changed to make use of the available capacity. Given the ongoing transition to a zero-carbon energy system, which is generally more stochastic in nature, there could perhaps be value in designing a plant with these systems which would give the operator significantly more flexibility in their operations.
Although a full model could not be constructed, given the data provided a rough estimate of the flexibility implementing this system across the full network could give Scottish Water as well as its role within a future energy system can be considered. Looking at the summer and springs months where the plant requires less energy it can be seen from Fig.3 that on average the plant operates for 6 hours at around 35kWh giving 19kWh of available capacity which could be shifted away from peak times. The case study site was estimated to process 1.35m litres of wastewater per day out of the 996m per day across the entire network. (6) Scaling up the 19kW demand that could be shifted if this method was implemented across the whole network gives a total of 14MW which could be moved for up to 6 hours. Given the peak demand in Scotland during summer for electricity is 3GW (22) a 14MW shift equates to a 0.47% reduction in demand on the grid at peak times. Note this would not be the case in winter where the plant would have significantly less flexibility and peak national demand is higher. The peak demand in electricity is also likely to become significantly higher in the coming years with the electrification of heating and transport. This just gives an idea of the potential application of such a system in a future energy system, although it should also be noted that this is not unique to this flow equalisation flow basin approach and other energy storage options are just as capable of moving demand on the grid away from peak times with the same outcome.