Fluid Mechanics Aided Computing for Effective Traffic Management

Fluid Mechanics Aided Computing for Effective Traffic

Management Scientific Traffic Management,

A look at the future.

Jaydeep. S. DeshpandeB.E. Mechanical

PVG’s COET, Pune.

Traffic 2020

A peep into future:

•The latest statistical record shows, in 2008, total number of vehicles on the streets of the country were 106,591 with over 10% in Maharashtra.•With statistical modeling and recent trends it is project that the number will reach to 450,000 in next 10 odd years.•Every vehicle manufacturer has realized the scope in India. Every one is aggressively fielding their products which can offer the best technologies in the Indian market.

Traffic 2020

Consequences:

•Roads, bridges, infrastructure can not be altered at the speed of ever increasing number of vehicles.•Traffic conditions will get worse.•Increased duty/force of traffic cops.•Increased recourses in effectively managing ever growing traffic. •Increased associated health hazards due to noise, air pollution.

Traffic 2020What can we do to make the big picture look good

Fluid Dynamics• Science which studies various flow conditions• It explains energy and mass balance in fluid

system• It can be studied by focusing on a particle or

focusing at a point in fluid to understand the nature of the flow.

Traffic flow• Traffic flow is flow of vehicles. • Roads can be considered as channels through

which vehicles can travel.• Can we apply same governing principles to

traffic flow as we apply on liquid/gas flow?

Agenda

• Nozzle Effect • Boundary Layer implementation• Effects of Viscosity• Example • Conclusion

Nozzle effects

• The diagram shows a typical nozzle.

• Velocity of the liquid increases through the nozzle and pressure drops.

• Formation of eddies at the junction of throat and exit part if half divergence angle is too large.

Image: www.google.com

Nozzle effects

(u2 - u1) (w∝ 2-w1)

∴ (u2 - u1) = c (w2-w1)

Where;u2 – Velocity of vehicles emerging from the nozzle junction

u1 – Velocity of vehicles approaching the nozzle junction

w2 – Width of road at nozzle junction

w1 – Width of road after nozzle junction

c – Constant of proportionality

Viscosity in traffic flow

• It is evident that vehicles travelling along the boundaries move slower.

• As in case of non-ideal fluids, there are interactions between moving vehicles, which are necessary to maintain safe distance and velocity.

Viscosity in traffic flow

T=µ*(Δv/Δy)Where,T- Viscous forceµ- Coefficient of ViscosityΔv- difference in velocityΔy- relative position with respect to vehicle

∴ For a vehicle flow rate Q, we can write it as ∴ T=Q*a= µ*(Δv/Δy)

a- instantaneous acceleration of a vehicle at any given road position Δy

Flow along edges

Sharp Edge

Smooth Edge

Traffic flow around obstacles

Line up to which choking occurs.

Dead spots

Dynamic Traffic Monitoring

Dynamic Traffic Monitoring

Dynamic Traffic Monitoring

Steps:

1.Take real time video feed

2.Run image processing algorithms to determine speed, acceleration of vehicles, density of traffic

3.Compute the traffic at junctions

4.Evaluate likelihood of traffic jams

5.Alter signal timings

Active Signal Timing Optimization

Active Signal Timing Optimization

Example

Results

Sr. Time ‘Q’ ‘t’ calculated

‘t’ actual Remark(signal t)

1 8 AM 6-8 198 100 Can be reduced

2 10 AM 12-15 106 100 Perfect3 7 PM 18-20 73 100 Traffic jam

Traffic 2020

To conclude:•Proposed solution offers a scientific method which an be tested, simulated before being actually put into use.•It can yield the most accurate results.•It can thus, handle heavy traffic conditions efficiently. It will also reduce the dependence and load on traffic cops.•Hence it provides a very good “comprehensive” solution to the future demands and problems.

Any Questions??

Thank you.