TY - JOUR
T1 - On the dynamic neural network toolbox design for identification, estimation and control
AU - Chairez, Isaac
AU - Guarneros-Sandoval, Israel Alejandro
AU - Prud, Vlad
AU - Andrianova, Olga
AU - Ernest, Sleptsov
AU - Chertopolokhov, Viktor
AU - Bugriy, Grigory
AU - Mukhamedov, Arthur
N1 - Publisher Copyright:
© 2022, Emerald Publishing Limited.
PY - 2023/9/25
Y1 - 2023/9/25
N2 - Purpose: There are common problems in the identification of uncertain nonlinear systems, nonparametric approximation, state estimation, and automatic control. Dynamic neural network (DNN) approximation can simplify the development of all the aforementioned problems in either continuous or discrete systems. A DNN is represented by a system of differential or recurrent equations defined in the space of vector activation functions with weights and offsets that are functionally associated with the input data. Design/methodology/approach: This study describes the version of the toolbox, that can be used to identify the dynamics of the black box and restore the laws underlying the system using known inputs and outputs. Depending on the completeness of the information, the toolbox allows users to change the DNN structure to suit specific tasks. Findings: The toolbox consists of three main components: user layer, network manager, and network instance. The user layer provides high-level control and monitoring of system performance. The network manager serves as an intermediary between the user layer and the network instance, and allows the user layer to start and stop learning, providing an interface to indirectly access the internal data of the DNN. Research limitations/implications: Control capability is limited to adjusting a small number of numerical parameters and selecting functional parameters from a predefined list. Originality/value: The key feature of the toolbox is the possibility of developing an algorithmic semi-automatic selection of activation function parameters based on optimization problem solutions.
AB - Purpose: There are common problems in the identification of uncertain nonlinear systems, nonparametric approximation, state estimation, and automatic control. Dynamic neural network (DNN) approximation can simplify the development of all the aforementioned problems in either continuous or discrete systems. A DNN is represented by a system of differential or recurrent equations defined in the space of vector activation functions with weights and offsets that are functionally associated with the input data. Design/methodology/approach: This study describes the version of the toolbox, that can be used to identify the dynamics of the black box and restore the laws underlying the system using known inputs and outputs. Depending on the completeness of the information, the toolbox allows users to change the DNN structure to suit specific tasks. Findings: The toolbox consists of three main components: user layer, network manager, and network instance. The user layer provides high-level control and monitoring of system performance. The network manager serves as an intermediary between the user layer and the network instance, and allows the user layer to start and stop learning, providing an interface to indirectly access the internal data of the DNN. Research limitations/implications: Control capability is limited to adjusting a small number of numerical parameters and selecting functional parameters from a predefined list. Originality/value: The key feature of the toolbox is the possibility of developing an algorithmic semi-automatic selection of activation function parameters based on optimization problem solutions.
KW - Artificial neural networks
KW - Dynamic neural networks
KW - Learning laws
KW - Nonparametric model
KW - Toolbox
UR - http://www.scopus.com/inward/record.url?scp=85139947209&partnerID=8YFLogxK
U2 - 10.1108/K-04-2022-0487
DO - 10.1108/K-04-2022-0487
M3 - Artículo
AN - SCOPUS:85139947209
SN - 0368-492X
VL - 52
SP - 2943
EP - 2957
JO - Kybernetes
JF - Kybernetes
IS - 9
ER -