Interpretation of Dynamic Models Based on Neural Networks in the Form of Integral-Power Series

Fomin, Oleksandr; Polozhaenko, Sergii; Krykun, Valentyn; Orlov, Andrii; Lys, Daria

doi:10.1007/978-3-031-20141-7_24

Cited by 3 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the example of a test object, we investigated the effectiveness of interpreting non-network models. The simulation model of the test object with a first-order dynamic block and a nonlinear feedback block [7] is shown in Fig. 1.…”

Section: Simulation Model Of the Test Objectmentioning

confidence: 99%

“…Among these types of neural network models, TDNNs are the most general structure consisting of several layers with direct signal propagation [6][7][8]. Such models are able to learn from the input-output data of nonlinear dynamic objects and have excellent convergence properties [7], which are advantages over the aforementioned Dynamic Neuro-SM and Wienertype DNN methods. Therefore, TDNN models are an effective tool for modeling nonlinear dynamic objects with continuous characteristics.…”

Section: Fig 2 Transient Characteristics Y03(t) Y06(t) Y09(t) Of the ...mentioning

confidence: 99%

“…where M is the memory length of the object model, K is the number of neurons of the received layer with a nonlinear activation function, b 0 , b i is the bias of the neurons of the output and hidden layers, respectively; S 0 , S i are the activation functions of the neurons of the output and input layers, respectively; w i , w i, j are the weighting coefficients of the neurons of the output and hidden layers, respectively. A TDNN model can be trained to behave dynamically with the incorporation of nonlinear characteristics [7] on the input-output data.…”

Section: Fig 2 Transient Characteristics Y03(t) Y06(t) Y09(t) Of the ...mentioning

confidence: 99%

“…Analytical model. For a wide class of nonlinear dynamic objects with continuous characteristics, the relationship between the input x(t) and output y(t) signals can be written in the form of an integral-power series based on multidimensional weight functions [6,7]. Thus, for an object with one input and one output in the time domain, the model takes the following form:…”

Section: Fig 5 Graph Of Importance Of Test Model Featuresmentioning

confidence: 99%

“…To interpret the neural network model of a test object in the form of an analytical expression of an integral-power series, expressions that establish an analytical relationship between these models are used [7]. These expressions are used to determine the multidimensional weight functions of the first w 1 (τ 1 ) and second w 2 (τ 1 , τ 2 ) orders.…”

Section: Fig 5 Graph Of Importance Of Test Model Featuresmentioning

confidence: 99%

See 4 more Smart Citations

Improving the Accuracy of the Neural Network Models Interpretation of Nonlinear Dynamic Objects

Krykun

2023

mcm-tech

View full text Add to dashboard Cite

The paper is devoted to the problem of neural network interpretation in the tasks of modeling nonlinear dynamic objects. The purpose of the work is to improve the accuracy of the neural network models interpretation of nonlinear dynamic objects and to determine the scope of their effective application. This goal is achieved by applying analytical models in the form of integral-power series based on multidimensional weight functions. The scientific novelty of the work lies in the use of nonlinear dynamic models in the form of integral-power series based on multidimensional weight functions instead of linear surrogate models. It allows to improve modeling accuracy. The practical usefulness of the work is determination of the effective application area of analytical interpretive models. The practical significance of the obtained results lies in the application of the proposed models for the interpretation of neural network models of nonlinear dynamic objects, which allows to increase the accuracy of interpretation models compared to linear surrogate models.

show abstract

Section: Simulation Model Of the Test Objectmentioning

confidence: 99%

Section: Fig 2 Transient Characteristics Y03(t) Y06(t) Y09(t) Of the ...mentioning

confidence: 99%

Section: Fig 2 Transient Characteristics Y03(t) Y06(t) Y09(t) Of the ...mentioning

confidence: 99%

Section: Fig 5 Graph Of Importance Of Test Model Featuresmentioning

confidence: 99%

Section: Fig 5 Graph Of Importance Of Test Model Featuresmentioning

confidence: 99%

See 3 more Smart Citations

Improving the Accuracy of the Neural Network Models Interpretation of Nonlinear Dynamic Objects

Krykun

2023

mcm-tech

View full text Add to dashboard Cite

show abstract

Modeling nonlinear dynamic objects using pre-trained time delay neural networks

Fomin,

Orlov

2024

AAIT

View full text Add to dashboard Cite

The work is devoted to resolving the contradiction between the accuracy of modelling nonlinear dynamics and the speed of model construction under conditions of limited computing resources. The purposeof the work is to reduce the time for building time delay neural networks while ensuring a given accuracy in the tasks of identifying nonlinear dynamic objects with continuous characteristics. This goal is achieved by developing a method for pre-training neural networks that reflect the basic characteristics of the subject area. The scientific novelty of the work is the development of a method for identifying nonlinear dynamic objectsin the form of time delay neural networks based on the use of a set of basic pre-trained neural networks that reflect the typical properties of the subject area. In contrast to the traditional approach to pre-training, the developed method allows building models of lower complexity.A formal criterion is proposed for determining the moment of termination of the neural network pre-training, the use of which allows avoiding retraining of the base model and ensuring a significant reduction in the model training time on the target data set. The practical utility of the work lies in the development of an algorithm for the method of pre-training time delay neural networks in the tasks of identifying nonlinear dynamic objects with continuous characteristics, which allows to significantlyreduce the training time of neural networks without losing model accuracy. The value of this study is todetermine the area of effective use of the proposed method, namely, when the general and target datasets do not have significant differences and the target dataset is of sufficient size to reflect the properties of the research object.

show abstract

Use of Dynamic Neural Networks for Modeling Nonlinear Objects with Significant Nonlinearity

Fomin,

Polozhaenko,

Krykun

et al. 2023

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

The work is devoted to the problem of nonlinear modeling of objects based on dynamic neural networks. The aim of the work is to improve the accuracy of modeling dynamic objects with significant nonlinearities using neural network models, and identify the scope of their effective application. This aim is achieved by applying the dynamic nonlinear models in the form of time delay neural networks. The scientific novelty of the work lies in the determination of the dependences between the accuracy of suggested models and the types of model input signals, as well as the amplitudes of model input signals. Practical usefulness of the research lies in the determination of the area of effective use of suggested models of dynamic objects with significantly nonlinear features, such us saturation. Significance of the obtained results: the application of the proposed models for identification dynamic objects with significantly nonlinear characteristics allows to improve the accuracy of the modelling process in comparison with models based on deterministic identification methods, such as integro-power series based on multidimensional weight functions.

show abstract

Interpretation of Dynamic Models Based on Neural Networks in the Form of Integral-Power Series

Cited by 3 publications

References 25 publications

Improving the Accuracy of the Neural Network Models Interpretation of Nonlinear Dynamic Objects

Improving the Accuracy of the Neural Network Models Interpretation of Nonlinear Dynamic Objects

Modeling nonlinear dynamic objects using pre-trained time delay neural networks

Use of Dynamic Neural Networks for Modeling Nonlinear Objects with Significant Nonlinearity

Contact Info

Product

Resources

About