Predicting weld responses is a very important but difficult area in the field of welding which can greatly reduce the overall cost of try and error method for any fabrication industry. Fuzzy logic expert tool was used to predict the weld penetration size factor of a weld. The aim of this study is to predict the weld penetration size factor (WPSF) of TIG mild steel welds using fuzzy logic. In this study, the weld specimens were produced using the TIG welding process guided by the central composite experimental design, and thereafter the weld penetration size factor (WPSF) was measured. The process parameters include the voltage, current, gas flow rate and welding speed. The model’s significance, strength and adequacy were checked; for fuzzy logic, fuzzification was done using fuzzy linguistic variable, fuzzy linguistic terms and membership function after which an inference was made based on a set of rules and the output result was defuzzified to a crisp output. Fuzzy logic predicted beyond the boundaries of the given range of parameters. The model developed has proven to be very effective in predicting responses even before actual weld is initiated.
Low weld quality can result from poor combination of the welding input parameters. Therefore, the need for optimization of input parameters in order to obtain the best quality of weld is unending. In Nigeria, the need for quality weld is gradually attracting the interest of Manufacturing and Industrial Engineers, as the importance of a good quality welded joint, which will result in good service life of weld, cannot be over emphasized. Injuries/damages result from catastrophic weld failures in fabrication yards; construction sites, such as church building collapse, have remained uninvestigated in most cases in Nigeria and therefore there is a need to educate local welders on the relevance to improve the weld quality. The mechanical strength of a weld can give good insights about the quality of the weld bead geometry, which can be described by the Bead Width, Height, Reinforcement, Weld Reinforcement Form Factor (WRFF) and Weld Penetration Size Factor (WPSF). WRFF and WPSF fall under the umbrella of the weld bead shape. [
To have real control over the strength of the eventual weld product, precise relationships between the process parameters and the bead parameters controlling the bead shape need to be established. [
In this study, fuzzy logic inferential capability was applied to predict the response. This investigation is geared towards improving the quality and strength properties of weld bead shape factors and geometry.
The Tungsten Inert Gas (TIG) machine was used to weld 10 mm mild steel plates measuring 60 mm in length, 40 mm in width. One hundred and Fifty (150) pieces of plate were cut with the edges bevelled, machined and etched with a 2% NaCl. This experiment was repeated 30 times with each experiment having five specimens, thereby producing a total of one hundred and fifty welded joints. The input parameters used for this study were welding speed, current, arc voltage, and gas flow rate as shown in
In this study, thirty experimental runs were carried out, each experimental run comprising the current, voltage, welding speed and gas flow rate, used to join two pieces of mild steel plates measuring 60 mm × 40 mm ×10 mm. The weld penetration size factor was measured. The results are shown in
A fuzzy logic system (FLS) can be defined as the nonlinear mapping of an input data set to a scalar output data.
The process of fuzzy logic is explained as follows: Firstly, a crisp set of input data are gathered and converted to a fuzzy set using fuzzy linguistic variables,
Parameters | Unit | Symbol | Coded value | Coded value |
---|---|---|---|---|
Low (−1) | High (+1) | |||
Current | Amp | A | 140 | 160 |
Gas flow rate | Lit/min | F | 12 | 14 |
Voltage | Volt | V | 20 | 24 |
Welding speed | cm/min | S | 150 | 170 |
Std | Run | Voltage (Volt) | Current (Amp) | Welding speed (mm/min) | Gas flow rate (L/min) | WPSF (mm) |
---|---|---|---|---|---|---|
26 | 1 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0324 |
29 | 2 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0326 |
30 | 3 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0323 |
25 | 4 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0324 |
27 | 5 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0325 |
28 | 6 | 22.00 | 150.00 | 160.00 | 13.00 | 1.0323 |
18 | 7 | 26.00 | 150.00 | 160.00 | 13.00 | 1.0021 |
23 | 8 | 22.00 | 150.00 | 160.00 | 11.00 | 2.9899 |
21 | 9 | 22.00 | 150.00 | 140.00 | 13.00 | 2.2015 |
20 | 10 | 22.00 | 170.00 | 160.00 | 13.00 | 2.0005 |
19 | 11 | 22.00 | 130.00 | 160.00 | 13.00 | 2.8765 |
24 | 12 | 22.00 | 150.00 | 160.00 | 13.00 | 2.1325 |
17 | 13 | 18.00 | 150.00 | 160.00 | 13.00 | 1.6534 |
22 | 14 | 22.00 | 150.00 | 160.00 | 13.00 | 2.0873 |
5 | 15 | 20.00 | 140.00 | 170.00 | 12.00 | 2.7276 |
4 | 16 | 24.00 | 160.00 | 150.00 | 12.00 | 1.5454 |
7 | 17 | 20.00 | 160.00 | 170.00 | 12.00 | 2.3843 |
14 | 18 | 24.00 | 140.00 | 170.00 | 14.00 | 2.3438 |
10 | 19 | 24.00 | 140.00 | 150.00 | 14.00 | 2.1037 |
6 | 20 | 24.00 | 140.00 | 170.00 | 12.00 | 1.6943 |
16 | 21 | 24.00 | 160.00 | 170.00 | 14.00 | 1.6371 |
2 | 22 | 24.00 | 140.00 | 150.00 | 12.00 | 1.9965 |
8 | 23 | 24.00 | 160.00 | 170.00 | 12.00 | 2.6262 |
3 | 24 | 20.00 | 160.00 | 150.00 | 12.00 | 2.5862 |
9 | 25 | 20.00 | 140.00 | 150.00 | 14.00 | 2.2322 |
13 | 26 | 20.00 | 140.00 | 170.00 | 14.00 | 2.4315 |
1 | 27 | 20.00 | 140.00 | 150.00 | 12.00 | 2.3981 |
11 | 28 | 20.00 | 160.00 | 150.00 | 14.00 | 1.8693 |
12 | 29 | 24.00 | 160.00 | 150.00 | 14.00 | 1.0677 |
15 | 30 | 20.00 | 160.00 | 170.00 | 14.00 | 1.0621 |
fuzzy linguistic terms and membership functions. This step is known as fuzzification. Afterwards, an inference is made based on a set of rules. Lastly, the resulting fuzzy output is mapped to a crisp output using the membership functions, in the defuzzification step.
In this study we developed a fuzzy logic system to predict the weld penetration size factor (WPSF), based on four input variables, namely, voltage, current, welding speed and gas flow rate.
Defining the Linguistic Variables and Terms
Linguistic variables are the input or output variables of the system whose values are words or sentences from a natural language, instead of numerical values. A linguistic variable is generally decomposed into a set of linguistic terms. Consider a welding process aimed at predicting the weld penetration size factor (WPSF). Let voltage (v), current (c), welding speed (ws) and gas flow rate (gfr) be the linguistic variables which represents the weld factors. To qualify the voltage, current, welding speed and gas flow rate, terms such as (very low, low, moderate, high and very high) are used in real life.
For this problem, the linguistic variables and their range of values include:
1) Voltage; this range from 20 to 24 volts;
2) Current; this range from 140 to 160 amps;
3) Welding speed; this range from 150 to 170 mm/min;
4) Gas flow rate; this range from 12 to 14 L/min;
5) Weld penetration size factor; this range from 1.002 to 2.990 mm.
The range of the input and output variable were extracted from the experimental design summary presented in
1) Defining the Inputs and Output Membership Function
Membership functions are used in the fuzzification and defuzzification steps of a Fuzzy Logic Systems (FLS), to map the non-fuzzy input values to fuzzy linguistic terms and vice versa. A membership function is used in most cases to quantify a linguistic term. An important characteristic of fuzzy logic is that a numerical value does not have to be fuzzified using only one membership function. In other words, a value can belong to multiple sets at the same time.
As mentioned earlier, five membership functions were selected for each input and output variable namely; very low, low, moderate, high and very high.
We applied same method to set up the membership function current, welding speed and gas flow rate. Our result summary is presented in
2) Fuzzy Rules
A simple fuzzy logic code is written to control the output variable. Fuzzy rule
is simply an “IF-THEN” rule with a condition and a conclusion. Five critical rules were constructed to predict the weld penetration size factor based on fuzzy logic.
constructed for this problem.
The simplified form of
1) If voltage is low and current is high and welding speed is high and gas flow rate is high, weld penetration size factor is very low;
2) If voltage is high and current is high and welding speed is low and gas flow rate is low, weld penetration size factor is low;
Membership function | Membership sets | ||||
---|---|---|---|---|---|
Voltage | Current | Welding speed | GFR | WPSF | |
Very Low | (16 18 20) | (120 130 140) | (130 140 150) | (10 11 12) | (1.002) |
Low | (18 20 22) | (130 140 150) | (140 150 160) | (11 12 13) | (1.499) |
Moderate | (20 22 24) | (140 150 160) | (150 160 170) | (12 13 14) | (1.996) |
High | (22 24 26) | (150 160 170) | (160 170 180) | (13 14 15) | (2.493) |
Very high | (4 26 28) | (160 170 180) | (170 180 190) | (14 15 16) | (2.990) |
3) If voltage is high and current is low and welding speed is low and gas flow rate is low, weld penetration size factor is moderate;
4) If voltage is low and current is high and welding speed is low and gas flow rate is low, weld penetration size factor is high;
5) If voltage is moderate and current is moderate and welding speed is moderate and gas flow rate is very low, weld penetration size factor is very high.
From the result of
The surface plot which shows the relationship between the input and the output variable is presented in
Result of
To check the prediction accuracy of the fuzzy logic tool, values form
The randomized selected results obtained from fuzzy logic for maximized weld penetration size factor, is presented in
This research work has thrived in developing a prediction of welds of extremely high quality of TIG welding process using fuzzy logic through which the effects of their various process parameters and their interactions were determined and predictions made on expected quality of the weld at known process
Variable combinations dilution | ||||||
---|---|---|---|---|---|---|
R/N | Voltage | Current | W S | GFR | Exp (WPSF) | Fuzzy (WPSF) |
7 | 26 | 150 | 160 | 13 | 1.996 | 2 |
12 | 22 | 150 | 160 | 15 | 1.69 | 2 |
16 | 24 | 160 | 150 | 12 | 2.99 | 2.99 |
18 | 24 | 140 | 170 | 14 | 2.58 | 2.49 |
24 | 20 | 160 | 150 | 12 | 1.545 | 1.5 |
26 | 20 | 140 | 170 | 14 | 2.384 | 2 |
27 | 20 | 140 | 150 | 12 | 1.002 | 1 |
30 | 20 | 160 | 170 | 14 | 2.2 | 2 |
parameters.
Additionally, the following have been ascertained in this study:
1) Welding speed and gas flow rate are found to have great influence on weld penetration size factor as compared to current and voltage at a moderate level;
2) Fuzzy logic was able to predict the expected responses accurately even beyond the boundaries of the given parameters.
A novel concept of an intelligent model has been developed to predict welding process parameters (current, voltage, welding speed and gas flow rate) and bead parameters (WPSF) for improved quality welds using fuzzy logic. The results of this study will help reduce the cost of expensive analytical methods employed during welding operation and it will help fabrication industries to maximize the quality of their products with minimal stress and eliminate time used for trial and error experiment during welding.
The authors declare no conflicts of interest regarding the publication of this paper.
Stephanie, N. and Ifeanyi, A.J. (2019) Prediction of Weld Penetration Size Factor (WPSF) of TIG Mild Steel Weldment Using Fuzzy Logic. Engineering, 11, 119-130. https://doi.org/10.4236/eng.2019.112010