Inflow and Infiltration Analysis
Inflow and infiltration (I&I) is calculated for each pumping station by analysing incoming wastewater volumes based on available measurement data — such as flow measurements, level measurements in the wet well, or pump run times combined with known pump capacity. The model first establishes a reference level based on dry periods, where the inflow is mainly assumed to be sanitary wastewater. Deviations above this level are interpreted as inflow and infiltration, typically from infiltration, ingress, or cross-connections.
Gross I&I expresses the total amount of inflow and infiltration recorded at a pump, regardless of what has already been added from upstream areas.
To avoid double-counting, net I&I is also calculated, where I&I from upstream pumping stations is subtracted based on the hierarchical structure of the pipe network. The net value therefore gives a better picture of where in the network the I&I actually originates and where measures should be prioritised.
Real-Time Calculation of Key Performance Indicators
InfoTiles SewerIntelligence calculates the following approximately in real time:
- Total volume of pumped water at each pumping station
- Expected wastewater volume (baseflow) at each pumping station
- Volume and proportion of I&I in near real time, per hour and per pumping station, including:
- Total volume of I&I transported
- Net proportion of I&I originating within the pumping station's catchment area, after accounting for upstream measured stations
Data Sources
InfoTiles' SewerIntelligence module calculates I&I volumes using historical and real-time SCADA data (flow and level measurements) as well as rainfall data.
Method for Identifying Inflow and Infiltration
InfoTiles' method for identifying I&I is based on existing measurement points, such as:
- Flow meters installed in wastewater pumping stations, as well as level meters in wet wells and at overflow weirs
- Level meters at overflow weirs
- Level meters in flumes in the pipe network
- Flow meters at treatment plant inlets
The method is developed to function with only level measurements, as this is the minimum variable that is always available. If flow measurements exist — either in pumps or in the network — these can also be used, but they are not required. In addition, the solution uses rainfall data and some static variables, such as the wet well surface area and nominal pump capacity, in the calculations or during calibration.
Cost and CO₂ footprint are based on industry standards for electricity consumption per cubic metre moved. It is known that this varies with lift height and friction; values between 150 and 500 Wh per cubic metre are common in the literature, and 200 is used as a baseline to avoid overestimation. This is scaled by a continuously updated spot price or a static electricity price. A fixed charge for treatment costs can also be added where this is known.
Calculation of Inflow and Infiltration
The model calculates total pumped volume (Vt) and expected baseflow (Vn) per hour.
The total pumped volume is a proprietary statistical and deterministic function of:
- Pump geometry / surface area
- Rate of change in pump level
- Nominal pump capacity
This ensures that it is possible to use high-resolution level measurements in the wet well, which are almost always available, while the solution also supports:
- Pump run times (on/off)
- Available flow measurements
The flexibility of using multiple different input types makes it possible to deliver across systems with different monitoring regimes and time resolutions.
Expected baseflow per hour, per weekday (Vn), is determined using a combination of manual labelling and statistical analysis of raw data. To avoid bias from abnormal years (dry or wet), at least 2 years of data are desired so that each season is represented in the training data. Rainfall is shown in the results dashboard but is not included mathematically in the calculation; river level/tidal data and other factors can also be added to understand the cause where increased volume is seen in the absence of rain.
The most common sources of good dry-weather data are persistently cold winters or persistently dry summers. Training data with as much information as possible gives the most accurate results — i.e. training data that contains very dry, very wet, and «normal» measured values.
In the absence of long training data, the driest scenario available will become the baseline for I&I calculations.
The Calculation Formula
Vf = Vt – Vn
Where: Vf = volume of I&I, Vt = total volume, Vn = expected dry-weather inflow (baseflow)
Total volume using the most common method (change in wet well level):
Vt = As · δh
Where: Vt = total volume, As = surface area of the wet well, δh = change in level
Identifying Likely Sources of Inflow and Infiltration
Starting from a pumping station where you want to analyse sources of I&I, you can go directly to an overview of all upstream components connected to the selected pump. Work is underway to expand the PipeFusion library to include stream inlets in a future version.
The analysis shows, among other things, the distribution of ownership, estimated rehabilitation costs for selected elements, risk calculations per pipe component (based on component properties), age, material types, and other relevant parameters.
Components with the highest probability of failure are highlighted in red in the map and summarised in an associated table. In the work to reduce I&I, these calculations can be used as a basis for:
- Planning and prioritising inspections, with a focus on components with high risk
- Identifying new measurement points to detect I&I (e.g. at network branches to confirm or rule out I&I in delimited areas)
- Planning rehabilitation with cost-benefit assessment
The solution also takes into account the connections between pumping stations in the network, and distinguishes between:
- Gross I&I: total volume handled by a pump.
- Net I&I: volume originating within the pump's own catchment area, after subtracting water measured at upstream stations.
Comments
0 comments
Please sign in to leave a comment.