Application of regularization techniques to improve forecast stability in noisy data for industrial automation

The article explores modern approaches to the application of regularization methods — Ridge and LASSO — in problems of forecasting technological process parameters under industrial automation conditions. Special attention is given to addressing challenges associated with the high-dimensional feature spaces and the presence of noise in input data, which are typical in industrial environments. The theoretical foundations of these methods are presented, along with their specific characteristics and mechanisms that reduce model overfitting and enhance robustness under varying input data. An experimental evaluation of the effectiveness of regularized regression models is conducted using real industrial datasets, including time series with missing and distorted values. The results demonstrate improved forecasting accuracy, model stability, and, consequently, the reliability of automated monitoring and control systems. These methods help cope with data noise, avoid retraining, and highlight key parameters, which is especially important in conditions of limited computational resources and complex production systems.