Simulation of Secondary Clarifiers

What are the benefits of hydrograv simulations?

  • higher capacity: With an optimized inlet design above the limits of design guide lines, e. g. the German DWA-A 131
  • maximum security against sludge overflow
  • maximum effluent quality (SS, Ptot, COD) by optimized inlet design
  • maximum use of the tank’s depth by optimized effluent design
  • optimization of  operational strategy, e. g.  RAS, sludge displacement
  • optimal function of the scraper system
  • determination of the flocculation potential

What we offer:

  • maximum predictability based on measurements of sludge concentrations on site in the settling tanks
  • extensive data analyses to obtain representative loadings and for evaluation of the simualtion results
  • extensive studies of design variants, operational strategies and optimizations

What are the features?

  • two and threedimensional multiphase simualtios
  • simulation of the sludge displacement between aeration und secondary clarifier
  • realistic modelling of different scraper systems and sludge properties

.

Example 1: Measurement of the sludge concentration and validation of the simulation models

Figure:   Distribution of the sludge concentratin in a circular tank.

 

Figure:   Comparison between measurement (points) and simulation (red line) of flow velocities from Armbruster (2004).

.

Example 2: Determination of representative loadings with proload.data

  • extensive data analysis with hydrograv data analysis software proload.data, e. g. QWWTP, QRAS, SVI, MLSS, SS, Ptot, COD
  • simulation verification for high, average und low loadings and optimization of a best possible inlet design for all loading cases

Figure:   Two dimensional statistical analysis of the sludge volume considering the MLSS and the SVI.

Examle 3: Automatic optimization of the inlet design

  • simulation with representative average sludge volumen for dry weather and stormwater flow
  • determination of a best possible compromise of the inlet height and the opening height of a fixed inlet construction considering the individual operational strategy of the WWTP

Figure:   Sludge concentration and automatic adapted inlet geometry (inlet height and opening height) for representative average sludge volume for dry weather flow  (above) and storm water flow (below).

Figure:   Determination of a best fitting fixed compromise inlet height.

 

Example 4: Determination of the maximum capacity

  • e.g. capacity’s increase of 25 % by optimized inlet design
  • e.g. loading exceeds 40 % the limits of the design guideline DWA-A 131 (see figure)

Figure:   Sludge concentration in a secondary clarifier with height adaptive inlet system.

Figure:   Determination of the maximum SVI for different inlet designs. Comparision between the status quo, a fixed optimization with two inlet heights und a height varialbe inlet design with two diameters.

.
Example 5: Improvements of the effluent quality

  • minimization of unstable sludge bed areas (ineffective Zone) = minimization of sludge flocs in the supernatant

Figure:   Categorisation of the sludge bed into functional zones. The green zone represents diluted unstable areas of the sludge bed. Above: Status quo. Below: Optimization.

.

Example 6: Optimization of geometric details – inlet design

Figure:   3D-simulation of a inlet structure and analysis of the hydraulic distribution at the inlet opening.

.

Example 7: Verification of the flocculation potential

Figure:   3D-simulation of a inlet structure and analysis of the G-value.

.

Example 8: Optimization of a suction scraper

  • in status quo unsufficient hydraulic capacity of the suction scraper system
  • with optimization higher hydraulic capacity by optimization of different components

Figure:   Detailed 3D-simulation of a suction scraper system and determination of hydraulic losses.

Figure:   Detailled 3D-simulation of a lifting pipe.

.

Example 9: Verification of a rotation suction scraper system

  • simulation of a rotating suction scraper system to study the impact of the sludge removal on the effluent quality

Figure:   Detailled 3D-simulation of a rotating suction scraper system. Sludge level during dry weather loading.

Figure:   Detailled 3D-simulation of a rotating suction scraper system. Velocities close to the bottom during dry weather loading.