TY - JOUR
T1 - Inclined planes system optimization
T2 - Theory, literature review, and state-of-the-art versions for IIR system identification
AU - Mohammadi, Ali
AU - Sheikholeslam, Farid
AU - Mirjalili, Seyedali
N1 - Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/8/15
Y1 - 2022/8/15
N2 - The Inclined Planes System Optimization (IPO) algorithm is recent algorithm that uses Newton's second law to perform optimization. After conducting a thorough literature review, this paper proposes an improved version of IPO called IIPO. This improvement is achieved by changing exploratory and exploitative behavior of the standard IPO proportional to the progress of optimization (iteration). The IIPO is employed for optimizing IIR digital filter design, which is a challenging engineering problem. Adaptive IIR modeling as a multimodal optimization problem is designed and developed under system identification structure with an appropriate single-objective function in the frequency domain. Implementations are performed in both modeling cases with same and reduced orders, and under two identification forms with and without environmental additive noise. The results are reported along with various analyzes compared to a wide range of IPO variants. The statistical results on 100 independent trials show a success of more than 90% of cases, the proposed IIPO algorithm substantially outperforms other comparative algorithms in terms of accuracy of estimated coefficients, convergence, fitness, output responses, noise analysis, stability, and reliability.1
AB - The Inclined Planes System Optimization (IPO) algorithm is recent algorithm that uses Newton's second law to perform optimization. After conducting a thorough literature review, this paper proposes an improved version of IPO called IIPO. This improvement is achieved by changing exploratory and exploitative behavior of the standard IPO proportional to the progress of optimization (iteration). The IIPO is employed for optimizing IIR digital filter design, which is a challenging engineering problem. Adaptive IIR modeling as a multimodal optimization problem is designed and developed under system identification structure with an appropriate single-objective function in the frequency domain. Implementations are performed in both modeling cases with same and reduced orders, and under two identification forms with and without environmental additive noise. The results are reported along with various analyzes compared to a wide range of IPO variants. The statistical results on 100 independent trials show a success of more than 90% of cases, the proposed IIPO algorithm substantially outperforms other comparative algorithms in terms of accuracy of estimated coefficients, convergence, fitness, output responses, noise analysis, stability, and reliability.1
KW - Adaptive system identification
KW - IIR filter
KW - Inclined planes system optimization
KW - Multimodal design problems
KW - State-of-the-art versions
UR - http://www.scopus.com/inward/record.url?scp=85128183209&partnerID=8YFLogxK
U2 - 10.1016/j.eswa.2022.117127
DO - 10.1016/j.eswa.2022.117127
M3 - Article
AN - SCOPUS:85128183209
VL - 200
JO - Expert Systems with Applications
JF - Expert Systems with Applications
SN - 0957-4174
M1 - 117127
ER -