TY - GEN

T1 - Multi-Expert Regression with Integrated Environment Indicators for Forecasting Unstable Dynamic Systems

AU - Мусаев, Александр Азерович

AU - Макшанов, Андрей

AU - Григорьев, Дмитрий Алексеевич

PY - 2026/5/18

Y1 - 2026/5/18

N2 - Unstable multivariate monitoring processes often exhibit stochastic chaos, where strong nonlinearity and timevarying dependencies limit the effectiveness of classical regression. This paper proposes a regression-based forecasting framework that exploits slowly varying cross-correlations between a target variable and selected immersion-environment parameters. Correlated regressors are compressed into a scalar integrated indicator and used in a rolling, shifted-design model. Three integration strategies are considered and combined within a twolevel multi-expert architecture via weighted averaging. Experiments on data from an unstable gas-dynamic technological process show that multi-expert aggregation reduces systematic forecast delay and improves accuracy, yielding an average MSE reduction of approximately 11% relative to the best single expert.

AB - Unstable multivariate monitoring processes often exhibit stochastic chaos, where strong nonlinearity and timevarying dependencies limit the effectiveness of classical regression. This paper proposes a regression-based forecasting framework that exploits slowly varying cross-correlations between a target variable and selected immersion-environment parameters. Correlated regressors are compressed into a scalar integrated indicator and used in a rolling, shifted-design model. Three integration strategies are considered and combined within a twolevel multi-expert architecture via weighted averaging. Experiments on data from an unstable gas-dynamic technological process show that multi-expert aggregation reduces systematic forecast delay and improves accuracy, yielding an average MSE reduction of approximately 11% relative to the best single expert.

KW - integrated regressors

KW - machine learning

KW - multi-expert systems

KW - nonstationary stochastic processes

KW - online time series forecasting

KW - singular spectrum analysis

UR - https://www.mendeley.com/catalogue/dbabe600-66b6-3190-a0f9-d99d39920eb7/

U2 - 10.1109/icieam69213.2026.11549668

DO - 10.1109/icieam69213.2026.11549668

M3 - Conference contribution

SN - 979-8-3315-8057-5

SP - 407

EP - 412

BT - Proceedings - 2026 International Conference on Industrial Engineering, Applications and Manufacturing, ICIEAM 2026

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 17 May 2026 through 23 June 2026

ER -