In industrial processes, accurate quality prediction is essential for optimizing operations and ensuring product consistency. However, the inherent complexity of industrial processes typically involves a large number of monitoring variables, accompanied by significant challenges such as high nonlinearity and data redundancy. To address this issue, this article proposes the broad learning system combined with feature-weighted stacked target-related Laplacian autoencoder (BLS-FW-STLapAE). The proposed model innovatively incorporates hidden-layer features extracted by the stacked target-related Laplacian autoencoder (STLapAE) into the feature nodes of the broad learning system (BLS). This approach addresses the issue of insufficient feature extraction caused by random mapping in BLS. This integration enables the model to capture both local geometric structures and key target-related features, significantly enhancing predictive performance in complex industrial environments. In addition, mutual information (MI) is introduced in STLapAE to evaluate the correlation between features and the target variable, further improving the model’s ability to capture relevant information about the target variable and ensuring that all relevant information is effectively utilized. Finally, the feasibility and practical value of the proposed algorithm were validated on benchmark public datasets and two real-world industrial applications: penicillin fermentation and solid propellant manufacturing. The superior predictive performance demonstrated across these tests enables operators to proactively adjust process parameters and prevent quality deviations, thereby significantly improving production safety and reducing operational costs in such high-risk chemical engineering domains.

The core quality data, such as interior ballistic performance, are seriously unbalanced in the plasticizing and molding process, which makes it difficult for traditional supervised learning methods to accurately predict this kind of index. A Tri-training weighted ensemble JITL-RVM model based on an integrated confidence evaluation strategy is proposed to solve the above problem. The method is based on Tri-training semi-supervised regression architecture and uses both labeled and unlabeled data for modeling. First of all, the traditional single similarity measure method is difficult to use to evaluate the real similarity between data samples reliably and stably. This method realizes diversity enhancement and data expansion of the data set for modelling through ensemble just-in-time modelling based on three homologous and heterogeneous mixed similarity measures. Secondly, a new integrated confidence evaluation strategy is used to select the unlabeled samples, and the pseudo-labeled data, which can improve the prediction performance of the model, can be selected. To improve the prediction effect of the model, the pseudo-label value of the data is revised continuously. The integrated confidence evaluation strategy can overcome many shortcomings of the traditional confidence evaluation method based on Co-training regression (Coreg). Finally, the final quality prediction value is obtained through weighted integration fusion, which reflects the difference between different models and further improves the prediction accuracy. The experimental results of interior ballistic performance prediction of single-base gun propellant show the effectiveness and superiority of the proposed method, and it can improve the RMSE, R², and PHR to 0.8074, 0.9644, and 93.3%, respectively.