Fuzzy Logic Model For Prediction Of Properties Of Exfoliated Vermiculite based Light weight concrete

Concrete is highly sought material in construction industry and is made with cement, Natural Sand and gravel. There is lot of advancements done in the field of concrete. A model based on fuzzy logic is used to arrive at fresh as well as indurate properties of Exfoliated vermiculite related Light weight concrete (EVLWC) containing Exfoliated Vermiculite, Granite Fines and Silica Fume has been developed. Materials investigated experimentally possess 0 %, 5 %, 10 %, 15 % and 20% weight of Exfoliated Vermiculite (EV) and fixed 10% Granite Fines (GF) both replacing Natural Sand and fixed 5% Silica Fume (SF) replacing cement in each concrete. Water/cement ratio and content of super plasticizer were chosen as 0.40 % and 1.0 % of content of used respectively. The slump property of fresh concrete and hardened properties of concrete like Stress-strain relationship, compressive strength and acid resistance are arrived for said concrete.


1.
Introduction: Concrete is a heterogeneous material formed with combination of cement, aggregates, water, and some admixtures.In order to arrive at the strength and durability achieved by concrete members it is a need to ascertain certain properties of concrete matrix over different age periods.The concrete strength is observed from the first day of curing to the 28 th day and fuzzy logic is applied pertaining to the 28-day compressive strength and durability pertaining to acid resistance of cement concrete under water curing.Light Weight concretes (LWC) were developed with seven proportions, each of which had fixed Silica Fume (5%) and Granite Fines (10%) and several Exfoliated Vermiculite (EV) ratios (0-20%).The compressive strength were checked at the end of 3 rd , 7 th , and 28 days.A fuzzy logic (FL) was developed keeping in mind input variables such as contents of Sand, Exfoliated Vermiculite, Granite Fines,, Silica Fume, cement, and age) and the output variable such as slump, compressive strength and durability as shown in figure 1.
After concrete cubes are casted, experimental set up was established, process for measurement of data commenced using signals and their responses were recorded.The response of signals depend on hydration process the concrete material undergone and stiffness of the concrete material.

2.
Literature Review Fuzzy logic was first coined by Loffi Zadeh in 1965, which varies between completely true and completely false condition between integer value 0 and 1 and these fuzzy models represents a mathematical presentation of confusion existing in system and locate the truth in the system to exact degree. 1 As per the studies of Gokmen Tayfur et.al five different binders with varying silica fume between 0 to 15% for age of 3, 7 and 28 days were adopted and total 60 number of concrete cubes were prepared.With input like contents of silica fume, cement and age and with output as compressive strength which were fuzzified and thus fuzzy logic was successfully applied. 2 Uygunoglu et.alIn studied about effect of fly ash content, water/cement ratio and concrete age on the compressive strength was evaluated using fuzzy logic and a model was developed for varying amount of fly ash of 0%,10%,20% and 30% and for age period of 3,7,28,90, 180 and 365 days and optimum content of Fly Ash content, water/cement ratio and properties like compressive strength was arrived for concrete. 3Similar study on High Strength concrete was performed by Paratibha Aggarwal et.al, (2013) who applyied fuzzy logic for arriving at a model for predicting the compressive strength having different cementitious material.They further attempted to optimize mateials by reducing their cost of targeted mixes. 4he properties like temperature during hydration, design mix of materials and Concrete compressive strength, are evaluated using fuzzy logic and were compared with physical testing of concrete cubes and cylinders by developing a fuzzy logic-based model.This model took into consideration data taken from maturity method as well as data obtained from universal testing machine. 5

3.1
Popular Fuzzy System Methods:

Mamdani rule System: Fuzzification
Out of the various Fuzzy systems available, Mamdani rule System involving fuzzifying input values are converting into fuzzy functions and applying rules and arriving at calculate fuzzy output functions which are later later defuzzify the fuzzy output functions resulting with output values for given input values.
While fuzzification two values considered are '0' and '1', The value '0' is not to be considered in the defined set with value '1' is said to be highest value in the set.For example consider Slump, generally it may be Low or high.'Low' is considered to be zero and 'high' condition is said to be having value of '1'.The condition in between low and high condition is termed to be 'medium' and value varies between 0 to 1.The Blue portion represents to low Condition, red triangle corresponds to high and yellow color triangle represents 'medium' condition which is shown in figure 2. For example the results show that it is 1282 (2023) 012005 IOP Publishing doi:10.1088/1757-899X/1282/1/0120053 "slightly medium" then value can vary between 0.1 to 0.3 and moderately medium means value ranges between 0.4 to 0.6 and "higher medium" has values between 0.7 to 0.9.

Fuzzy operators
Fuzzy set represented triangle shaped curve has an increasing slope from 0 to 1 and decreasing slope between 1 to 0. IF-THEN concept is applied to determine compressive strength based on slump (Slump high, compressive strength low and vice versa).

Defuzzification
After fuzzification there is a need for obtaining truth value which are exactly matching with the "intention" shown in the truth value.From obtained truth values of slump, an actual slump is to be caluculated that exactly mentions 'least', 'medium' and 'highest'. [From the obtained values defuzzification is applied.

Takagi-Sugeno-Kang (TSK)
The Takagi-Sugeno-Kang varies from Mamdani technique only that execution of fuzzy rules consists of defuzzification expressed in polynomial function.IF Slump Is very low=2 IF Slump Is very low and Water cement ratio is high=2 x Slump + 1 x (water cement ratio).
The prime benefit of this method is ease in calculations and easy control of usage of algorithms. 6

Sensors
The fuzzy logic can be applied using sensors.The sensor used is of brand xcluma whose specifications are revealed in table 1. EFFICACY Concrete Spacers of 20/25/40MM RCC Cover Block was used in 50 numbers along with Bosch GMS120 Professional Detector for measuring signals are used which are depicted in figure 3.

Results and Analysis:
While inducing compressive loading, the strain is measured using the strain gauge by obtaining the cross head displacement.The minute change in electrical resistance was arrived utilising the four probe method, while utilizing DC current ranging between 0.5 to 3.5 Amperes.The details of compressive stress using sensor and CTM instrument and strain are shown in Table 2, Figure 4-6.

Conclusion:
The sensors Fuzzy Logic can opportunely be used to arrive at the mechanical properties like slump, stress, strain and modulus of elasticity, compressive strength and durability of Light Weight Concrete.The input variables like Sand content, Exfoliated Vermiculite content, Granite Fines content, Silica Fume content, cement binder content, and age were adequate to achieve precise results.The operators utilized in the Fuzzy Logic were identified as suitable for determination of slump.Stress, strain and other mechanical properties like modulus of elasticity, compressive strength and durability prediction.In future other PZT material, and other CNTs, can be used as sensors for developing concrete to meet strength and durability requirements towards alkali, sulphates and chlorides.

Figure 6 : 3 :The
Stress-sensor (vs) strain of various concrete mixesTable Mix proportions and Electrical Resistivity of Various concrete Mixes relating to various Electrical resistivity (ER) values and various ER ratios for slump test, compressive strength test and Acid resistance test are shown in table 3-4, Figure 7-11

Fig 7 : 4 :Fig 8 :Fig 9 :
Fig 7: Electrical Resistivity of Concrete samples before and after soaking in Acid Table 4: Compressive strength (CTM & Sensor) and ΔR/R 0 + of Various concrete Mixes relating to various tests Test Sample Compressive Strength-CTM(MP a)

Table 5 .
[8][9]sults pertaining to compressive strength are in accordance to studies performed by sedat akkurt, Gokmun tayfur, Uygunoglu et.al, Paratibha Aggarwal et.al, and Tareen N. The results achieved are in conformance with studies predicted by Vardhan Nagarkar et.al7.The results pertaining to slump, stress, strain, Acid resistance tests and Modulus of elasticity are in accordance with the codes referred[8][9].1282 (2023) 012005