Investigation of the Possibility of Applying a Genetic Algorithm for Electroacoustics Problems
Article Sidebar
Main Article Content
Abstract
This paper considers the adaptation and application of a genetic algorithm to find the parameters of the electrodynamic transducer model. The advantages and disadvantages of this method in comparison with the classical method of identification using added mass are considered. The derivation of the suitability function for estimating the identified parameters is presented, which can also be used to identify other types of electroacoustic transducers. The theory underlying genetic algorithms has been examined and shown how genetic algorithms work by assembling the best solutions from small structural elements with excellent qualities. Next, the differences between genetic and traditional algorithms were analyzed, including population population support and the use of genetic representation of solutions.
After that, the strengths of genetic algorithms were described, including the possibility of global optimization and applicability to problems with complex mathematical representation or without representation at all, and noise resistance. Disadvantages were also highlighted: the need for special definitions and settings of hyperparameters, the danger of premature convergence. In conclusion, the situations when the use of genetic algorithms are listed
This algorithm is not tied to a specific engineering or scientific field, which makes it universal, it is equally used in genetics and computer science. The parameters were determined using a genetic algorithm and compared with the more classical method of added mass for acoustics. The comparative table in the work illustrates the high accuracy of the genetic algorithm in comparison with the method of added mass. During the work on the practical part, also to improve the behavior of the model at frequencies higher than the resonant, it was decided to complicate the model of the electrical subsystem of the tranducer and introduce additional parameters: parallel resistance and parallel inductance. As a result, the complicated model began to correspond better to the measured values in the entire frequency domain, and is therefore more accurate. This is an example of the convenience of using a genetic algorithm in the transition from identification of one model with specific parameters to another. The results of this work prove that the use of a genetic algorithm is appropriate for solving electroacoustic problems because its application allows to quickly experiment and identify more complex models for which the added mass method can not be applied.
Also, in the future, genetic algorithm can be used to identify transducer models of in time domain, for example, nonlinear models of electrodynamic transducers or models in a state space, which is the subject of future research.This paper considers the adaptation and application of a genetic algorithm to find the parameters of the electrodynamic transducer model. The advantages and disadvantages of this method in comparison with the classical method of identification using added mass and the method of parameter selection BL are considered. The derivation of the fitness function for assessing the quality of the identified parameters is presented, which can also be used to identify other types of electroacoustic transducers. The directly measured values for the application of the algorithm are the voltage at the terminals of the converter, the current through the coil of the converter and the displacement of the moving part of the converter. The undoubted advantage of the genetic algorithm compared to classical identification methods is its versatility and the ability to quickly adapt and configure for research and experimentation with different models and different types of transducers used in acoustics.
This article describes the adaptation and application of a genetic algorithm to find parameters of an electrodynamic transducer model. The advantages and disadvantages of this method in comparison with the classical identification method using added mass are considered. The derivation of the fitness function for assessing quality of the identified parameters is presented, which can also be used to identify other types of electroacoustic transducer models.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
R. H. Small, “Direct-Radiator Loudspeaker System Analysis,” J. Audio Eng. Soc., vol. 20, no. 5, pp. 383–395, 1972.
D. D. Volkov, “Finding the physical parameters of the electrodynamic converter by the method of using the parameter BL and the method of mass added,” Microsystems, Electron. Acoust., vol. 24, no. 6, pp. 65–68, Dec. 2019, DOI: https://doi.org/10.20535/2523-4455.2019.24.6.195547.
S. N. Sivanandam and S. N. Deepa, “Applications of Genetic Algorithms,” in Introduction to Genetic Algorithms, Berlin, Heidelberg: Springer Berlin Heidelberg, 2008, pp. 317–402, URL: http://link.springer.com/10.1007/978-3-540-73190-0_10.
S. Slimane and M. Benbouziane, “Portfolio Selection Using Genetic Algorithm,” J. Appl. Financ. Bank., vol. 2, no. 4, pp. 143–154, 2012, URL: https://www.scienpress.com/download.asp?ID=347.
Y. LI, K. C. NG, D. J. MURRAY-SMITH, G. J. GRAY, and K. C. SHARMAN, “Genetic algorithm automated approach to the design of sliding mode control systems,” Int. J. Control, vol. 63, no. 4, pp. 721–739, Mar. 1996, DOI: https://doi.org/10.1080/00207179608921865.
S. Wang et al., “A multi-approaches-guided genetic algorithm with application to operon prediction,” Artif. Intell. Med., vol. 41, no. 2, pp. 151–159, Oct. 2007, DOI: https://doi.org/10.1016/j.artmed.2007.07.010.
E. Wirsansky, Hands-On Genetic Algorithms with Python. Packt Publishing, 2020, ISBN: 9781838557744.
“DEAP documentation — DEAP 1.3.1 documentation.” [Online]. Available: https://deap.readthedocs.io/en/master/.
A. Novak, “Measurement of Loudspeaker Parameters: A Pedagogical Approach,” in 23rd International Congress on Acoustics : integrating 4th EAA Euroregio 2019, 2019, DOI: https://doi.org/10.18154/RWTH-CONV-239247.
P. Brunet, “Nonlinear System Modeling and Identification of Loudspeakers,” Northeastern University, 2014, URL: https://repository.library.northeastern.edu/files/neu:336724/fulltext.pdf.
