HERAKLION 2019 OPTIMIZATION OF WASTE MATERIAL REUSE IN CONCRETE - - PowerPoint PPT Presentation

HERAKLION 2019

• Pr. M. BENZERZOUR, Ec. M. AMAR, Pr. N. ABRIAK

OPTIMIZATION OF WASTE MATERIAL REUSE IN CONCRETE USING INNOVATIVE COMBINATION METHOD BASED ON MATHEMATICAL MODEL AND AI

26 – 29 June IMT Lille Douai, Civil and Environmental Engineering Department, Université Lille Nord de France, LGCgE, Villeneuve d’Ascq, France

“WASTE” Materials, structure Regulation / Policy Cost Acceptability Feasibility

2

Life cycle

Context

Caractérisation Caractérisation Formulation Formulation Modélisation Modélisation Expérimentation Expérimentation Validatio n Validatio n

Sediments Tires Fibres VP Slag Bauxaline Mâchefers Gravael

Complex decision Multi parameters

Traitement Traitement Durability Durability Environment Environment Cost Cost

Valorisation des SPI

3

Context

Immersion Immersion T emporary solution Environmental pollution stocking stocking T

• o important tonnage

Environmental pollution Sources of aggregates in Europe

Dredged sediments + 500 million m3 / year

4

Context

Environnemental Economic / Engineeri ng

Dragage /dredge déshydratation / drying Stockage / storage, immersion Caracterization Traitement Ouvrage / Design Suivi / Monitoring T ransport, cost, CO2, …

5

Operational & IA approach : sediment reused

Mathematical modeling Mathematical modeling IA IA

Mathematical operational approach

Sediment

• Name
• T

ype of sample

• Characteristics

. Chemical (as, zn, ...) . Mechanical (GTR)

• Centre of studies
• dredging date
• dredging location
• GPS coordinates
• T

ransport costs T / km

• Operating costs
• Notes

Treatment center

• Unit name
• T

ype of treatment

• For each type:

. Name of treatment . Cost € / T . Impact on polluting element%

• Address of the

center

• GPS coordinates
• Notes

Storage areas

• Zone Name
• T

ype (Inert, Not dangerous, Dangerous)

• GPS coordinates
• Storage costs € / T
• Notes

Materials

• Name
• GPS coordinates
• T

ransportation cost

• Operating cost (or

purchase)

• Characteristics

. chemical . mechanics

• Notes

Software : Input data 6

Operational & IA approach : sediment reused

Mathematical model : Objective function 7

Operational approach : sediment reused

athematical model : Constraints

ironmental constraints : Heavy metals ironmental constraints : Organic matter Mechanical constraints

8

Operational approach : sediment reused

Mathematical model : Validation

(PhD GPMD 2012) Experimental results : Optimal mixtures : SED 30%, SAND 70%, Modelisation results : Optimal mixtures : SED 31,9%, SAND 68%, (PhD LMCU 2013) Experimental results : Optimal mixtures : SED 40%, SAND 60%, Modelisation results : Optimal mixtures : SED 41,4%, SAND 58,56%, (PhD PRISMA 2015 ) PRISMA Experimental results : Optimal mixtures : SED 27%, SAND 73%, Modelisation results : Optimal mixtures : SED 23,4%, DAND 76,3%,

SED SAND 20 40 60 80 EXP MODEL SED SAND 20 40 60 80 EXP MODEL SED SAND 20 40 60 80 EXP MODEL

10

Results

11

Architecture

❶ Building up a neural network ❷ Integrating 18 inputs, 2 hidden layers, 1 Output based on 1310 experimental data values. Compressive strength Waste materials

IA approach : sediment reused

12

Results

• The fit is highly good for all data sets
• R² values is 0.95 or above in each case.
• There is Good correlation between

predicted and actual values

IA approach : sediment reused

Reusd possibility : Projects SETARMS, PRISMA, ECOSED, GPMD, SEDIMATERIAUX

Urban furniture Reservoir concrete pavement Aggregates Road Dyke self-compacting concrete 13

Operational & IA approach : sediment reused

FOUNDING MEMBERS OF THE CHAIR: CIRCULAR ECONOMY OF SEDIMENTS ECOSED DIGITAL 4.0 SIGNED ON JUNE 21, 2019