# Interactive Ga Pso Leco Models Biology Essay

**Published:** **Last Edited:**

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This research involves the performance achievement by doing inspection on two s-parameter amplifier circuit that comprises different stages; single-stage (Case 1) and 2-stage (Case 2) s-parameter amplifier circuits.

The comparison of objectives result is done between interactive GA-PSO LECO models with other APLAC built-in optimizers after performing circuits design variable optimization. The results of these comparisons are demonstrated in this chapter. The graph plot by using APLAC software for final optimized output data are shown for these two cases accordingly.

The output data has different weights to indicate the priority and preference of the desired outputs. The optimization has multi-objective outputs whereby dB[S(2,1)] must be bigger than 10dB with ï¿½0.2 tolerance. Meanwhile, S(1,1) and S(2,2) must be maintained at 0 with ï¿½0.2 tolerance.

4.2 Case 1: Single-Stage Amplifier Circuit

4.2.1 Before Optimization (Default Value):

For single stage amplifier circuit (Case 1), the initial default values of six variables before optimization is done are shown in Figure 4.1. Before optimization, these six default input parameters yield the output of S(1,1), S(2,2) and dB[S(2,1)] as plotted by APLAC software in Figure 4.2, Figure 4.3 and Figure 4.4 respectively and the readings are taken at 1.5 GHz frequency.

Figure 4.1: Variables at default value

Figure 4.2: S(1,1) graph plot at default variable value

Figure 4.3: S(2,2) graph plot at default variable value

Figure 4.4: S(2,1) graph plot at default variable value

Figure 4.5: Smith Chart of S(1,1) and S(2,2) at default variable value

The input variables are shown in Table 4.1 and the output readings are shown in Table 4.2. Referring to Table 4.2, both S(1,1) and S(2,2) do not meet the target since the values exceed 0.2. The power gain dB[S(2,1)] equals to 9.790 also fail to fulfill the requirement where it must be bigger than 9.8dB. The overall fitness value of this single-stage amplifier circuit at default design variables equals to 0.2432.

Table 4.1: Initial Default Value of Single-Stage Amplifier

Design Variable

(Initial Default Value)

Table 4.2: Output Parameters of Single-Stage Amplifier at Default Input Variables

4.2.2 Optimization using Interactive GA-PSO LECO Model

Based on the interactive GA-PSO LECO approach, the optimized variables value and the graph plotting of outputs for single-stage amplifier circuit is given in Figure 4.6, Figure 4.7, Figure 4.8, and Figure 4.9 respectively. Meanwhile, the Smith Chart plot for S(1,1) and S(2,2) is given in Figure 4.10. These readings are then recorded in Table 4.3 and Table 4.4.

Figure 4.6: Optimization variables generated by interactive GA-PSO LECO model

Figure 4.7: S(1,1) graph plot produced by interactive GA-PSO LECO model

Figure 4.8: S(2,2) produced by interactive GA-PSO LECO model

Figure 4.9: S(2,1) produced by interactive GA-PSO LECO model

Figure 4.10: Smith chart of S(1,1) and S(2,2) produced by interactive GA-PSO LECO model

In this research, the interactive GA-PSO LECO model for circuit optimization is compared with a number of various APLAC built-in optimizers for performance evaluation. The performance comparison is depicted in Table 4.3 and Table 4.4. Table 4.3 displays the resultant optimized design variable values by using GA-PSO LECO approach. The total fitness value that produced by GA-PSO LECO model and other types of APLAC built-in optimizers are shown in Table 4.4.

Table 4.3: Optimized Design Variables for Single-Stage Amplifier using Interactive GA-PSO LECO approach

Design Variable

C2 (pF)

Table 4.4. Parameter Value of Different Optimizer

Optimizer

Referring to Table 4.4, most of APLAC built-in optimizer methods do not meet the S(1,1) and S(2,2) requirement. They are Genetic, Gradient, Random, Simulated Annealing (Anneal), Exhausted Search, Multidirectional, Min-Max, and Conjugate Gradient Optimizer. Even though some of these optimizers have at least one satisfying specifications, they are still not considerable for circuit design. This is because the condition of instability can be related to the modified S11 and S22 s-parameters associated with input (source) and output (load) reflection coefficients respectively. If S11 and S22 do not meet the specified requirement, more power is reflected from the device than its incident on it, hence the device tend to be more unstable and can lead to oscillation.

In Table 4.4, it can be seen that values of both S(1,1) and S(2,2) of HookeJeeves and NelderMead optimization methods fulfill the requirement which the values are below than 0.2. At the same time, the value of power gain, dB[S(2,1)] of these two optimizers successfully met the goal to be more than 9.8dB with value of 9.818 and 9.879 respectively.

But it is observed that interactive GA-PSO LECO optimization method also meet the requirement value of S(1,1) and S(2,2). But anyway, the interactive GA-PSO LECO model outperforms the other optimizers in significant maximizing device power gain, dB[S(2,1)] by achieving 15.213 dB in value. It should be noted that more weight are given to S(2,1) as compared to S(1,1) and S(2,2). This can be further affirmed by referring to the total fitness value of 0.0015 whereby the interactive GA-PSO LECO model provides the lowest Ftot value compared to others.

4.3 Case 2: 2-stage Amplifier Circuit

4.3.1 Before Optimization (Default Value)

In Case 2, default or initial design variables of 2-stage amplifier circuit is given in Figure 4.10, whereas S(1,1), S(2,2) and dB[S(2,1)] are plotted in the graph as in Figure 4.11, Figure 4.12, Figure 4.13 and Figure 4.14. The readings of these S(1,1), S(2,2) and dB[S(2,1)] are taken at frequency of 1.5GHz.

Figure 4.10: Variables at default value

Figure 4.11: S(1,1) graph plot at default variable value

Figure 4.12: S(2,2) graph plot at default variable value

Figure 4.13: S(2,1) graph plot at default variable value

Figure 4.14: Smith Chart of S(1,1) and S(2,2) at default variable value

The input variables are shown in Table 4.5 and the output readings are shown in Table 4.6. Referring to Table 4.6, both S(1,1) and S(2,2) do not meet the target since the values exceed 0.2. The overall fitness value of this single-stage amplifier circuit at default design variables equals to 0.1248.

Table 4.5: Initial Default Value of 2-Stage Amplifier

Design Variable

(Initial Default Value)

Table 4.6: Output Parameters of 2-Stage Amplifier at Default Input Variables

4.3.2 Optimization by using Interactive GA-PSO LECO Model

Figure 4.15, Figure 4.16, Figure 4.17, Figure 4.18 and Figure 4.19 show the optimized design variables and the output parameters by using interactive GA-PSO method. This optimization results then tabulated in Table 4.7 and Table 4.8 as follows.

Figure 4.15: Optimization variables generated by interactive GA-PSO LECO model

Figure 4.16: S(1,1) graph plot produced by interactive GA-PSO LECO model

Figure 4.17: S(2,2) graph plot produced by interactive GA-PSO LECO model

Figure 4.18: S(2,1) graph plot produced by interactive GA-PSO LECO model

Figure 4.19: Smith chart S(1,1) and S(2,2) produced by interactive GA-PSO LECO model

The performance of interactive GA-PSO LECO approach is then compared to other types of APLAC built-in optimizers. The optimization result of GA-PSO design variables and the comparison of optimization output and extracted in Table 4.7 and Table 4.8 as follows.

Table 4.7: Optimized Design Variables for 2-Stage Amplifier using Interactive GA-PSO LECO approach

Design Variable

Table 4.8: Parameter Value of Different Optimizer

Optimizer

Referring to Table 4.8, all optimizers have S(1,1) and S(2,2) exceed 0.2 except Nelder Mead optimizer are within the acceptable range. Values of S(1,1), S(2,2) and S(2,1) for Nelder Mead optimizer are 0.097, 0.108 and 9.879 respectively.

Besides, the interactive GA-PSO LECO model obtains a very good result where S(1,1) and S(2,2) are much closer to zero. This indicates the reflected power from the device is at minimum level that may contribute to the stability of the device.

Moreover, interactive GA-PSO LECO approach outperforms the other optimizers including Nelder Mead optimizer in a way that power gain S(2,1) generated by interactive GA-PSO LECO model can reach up to 16.469. Meanwhile, Nelder Mead optimizer just capable to give lower power gain, S(2,1) which equals to 9.879.

As the result, the interactive GA-PSO LECO model produces s-parameter amplifier design in overall performance. This can be further proven by referring to the total fitness value of 0.001 whereby the interactive GA-PSO LECO model gives lowest Ftot compared to other types of optimization method.

Hence, in this study, GA-PSO LECO model has been demonstrated to be a flexible tool, capable to handle human intervention, as well as integrating multi-resolution structure.