Yao Jia

Due to the influence of fluctuations in the particle size and throughput of ore from the upstream crushing stage, variations in the return flow of unqualified particles during grinding, and the frequent changes of unknown nonlinearities in the complex ore feeding process, it is difficult to apply existing disturbance-rejection control methods to control the feed rate within the target range. In this article, by utilizing the characteristic that the feed rate fluctuates around the target interval, a dynamic model of the feeding process is established. Based on the obtainable previous-time-step value of the unknown nonlinear dynamic system and the closed-loop tracking error generated by the unknown rate of change of this nonlinear system, and the PI closed-loop equation, a data-driven signal-compensator is designed by combining feedforward control with a one-step optimal control law. A signal-compensated PI controller is proposed. The performance index that minimizes the tracking error at each operating instant is introduced, and a concurrent-learning-driven optimal tuning method for parameters of both the PI controller and the compensator is proposed. Simulation and industrial application experiments demonstrate that the proposed control method is capable of controlling the tracking error within the target interval during the entire operational period.

Optimizing industrial thickening systems requires accurate supernatant interface height prediction. Existing methods struggle with complex dynamics and noise, limiting accuracy. This article proposes a novel hybrid deep learning framework that synergistically combines regression and classification methodologies to improve predictive performance. Specifically, a differential attention-enhanced long short-term memory network provides high-precision point prediction by capturing salient features and addressing uncertain time delays. Concurrently, a multiscale convolutional neural network with a convolutional block attention module extracts features and classifies trend direction. To integrate numerical and trend predictions, a trend-aware dynamic fusion mechanism is introduced, which explicitly ensures trend consistency and adaptively optimizes fusion weights using a deep deterministic policy gradient algorithm, yielding predictions accurate in value and reliable in direction. Real-world data from a large coal preparation plant demonstrates the superior performance of our prediction method compared to baseline methods. In addition, industrial experiments validate that the proposed model supports operator decision-making for the optimization and control of thickening systems.

Accurate real-time prediction in dynamic industrial systems is crucial for optimization and efficiency. This article introduces a novel intelligent monitoring framework leveraging proportional–integral–derivative (PID)-optimized deep learning for adaptive time–frequency forecasting to address the challenges posed by nonstationary industrial data. The proposed method uniquely integrates a channel-independent separable dynamic filter (CSDF) that adapts to real-time multivariate process variables, minimizing cross-channel interference. Furthermore, a closed-loop PID optimization strategy enhances the convergence and prediction accuracy of the deep learning model. The effectiveness of this framework is demonstrated through a case study in the prediction of cleaned coal calorific value, a vital parameter for optimizing coal preparation processes and improving energy efficiency. Industrial experiments in this application show that the proposed method achieves a significant 5.36% increase in forecast hit rate (FHR) compared to existing techniques, highlighting its potential for advanced monitoring and control in dynamic systems.

Multirate industrial processes pose significant challenges for accurate forecasting due to varying sampling frequencies and missing data. This article proposes a novel hybrid deep learning framework that effectively addresses these issues. Our approach uses a combination of time series decomposition, inverted transformer (iTransformer)-based feature extraction, and a modified minimal gated unit (MGU) network. To handle missing quality variables, we introduce a robust adaptive parameter update algorithm based on dead-zone Kalman filtering. Through extensive experiments conducted on real-world industrial datasets, our method achieves a mean absolute error (MAE) reduction of 61.42%, a root-mean-square error (RMSE) reduction of 64.11%, and a high qualification rate improvement of 14.73% compared to the average performance of state-of-the-art technologies, thereby outperforming existing state-of-the-art techniques in terms of both forecasting accuracy and robustness.

To address the challenge of implementing 5G wireless networks in industrial process control, this paper integrates deep learning and reinforcement learning in AI with anti-disturbance industrial control to propose an anti-latency optimized control method for 5G wireless networks. The proposed method includes: an anti-latency controller based on data-driven signal compensators with PID controllers and the controller parameters self-optimization algorithm combining digital twin and reinforcement learning. The proposed anti-latency control method is integrated with end-edge-cloud collaboration technologies of Industrial Internet to create an intelligent control system architecture that achieves anti-latency optimized control. A control experimental system based on commercial 5G and Industrial Ethernet was established, and anti-latency control experiments are conducted on an actual heat exchange system. The experimental results indicate that the proposed anti-latency intelligent control method significantly eliminates the impact of 5G random latency on the dynamic performance of the control system.

The current control of alumina digestion process (ALDP) relies on infrequent manual assays for caustic ratio, leading to open-loop operation between assays and suboptimal performance. This paper proposes a novel cloud-edge collaboration (CEC) architecture utilizing deep learning for real-time caustic ratio prediction, enabling closed-loop operation at all times. Our method combines slow-sampled assays with fast-sampled process measurements. An autoregressive moving average with extra inputs (ARMAX) model is used for feature extraction, while an improved bidirectional gated recurrent unit (I-BiGRU) that incorporates a multi-head attention (MHA) mechanism bridges data gaps through prediction. A soft-collaboration mechanism ensures smooth model updates at the edge, enabling operators to make informed control decisions. Theoretical analysis guarantees the convergence of prediction errors. Real-world data from a large-scale alumina plant demonstrates the superior performance of our prediction method compared to the baseline methods. Additionally, industrial experiments validate the effectiveness of the CEC architecture in supporting human operators for ALDP closed-loop operation.

Deep learning offers promising solutions for quality prediction in industrial processes, improving decision-making and performance monitoring. In this article, we propose a novel deep learning framework that incorporates variable time delay (VTD) estimation and uncertainty quantification into quality prediction. The framework employs a collaborative method that integrates deep Q-network with random forest to estimate VTD values. It then utilizes a hybrid BMCR model, consisting of parallel bidirectional minimal gated unit and 1-D convolutional layers, along with a residual connection, specifically designed to capture both long-term and short-term features in industrial data. The framework produces prediction intervals directly to quantify the uncertainty in the prediction results. This combined method offers high-precision point predictions alongside uncertainty quantification, providing valuable insights for industrial decision-making. The effectiveness of the proposed method is validated through two numerical examples, a benchmark, and a real-world industrial case from the alumina digestion process.

The slurry concentration is a crucial factor in mineral processing production, affected by both upstream and downstream systems, including closed-loop control systems. The variability in slurry concentration presents a challenge due to its complex, nonlinear nature and the difficulty in accurately modeling this dynamic system for real-time monitoring. This article introduces an innovative approach for the real-time prediction of slurry concentration. Our method comprises two key components: a mechanistic model and a nonlinear dynamic system. The mechanism model is developed using the closed-loop input error method, effectively mitigating the influence of control systems. We have rigorously demonstrated the convergence of parameters and the stability of the identification process. Further, a hybrid deep neural network, called convolutional neural network-gated recurrent unit (GRU), is proposed to tackle challenges, such as intervariable dependencies and the inherent nonlinear dynamics of the process. The network integrates an autoregressive integrated moving average (ARIMA) model for laboratory delays. A parallel GRU network captures nonlinear dynamic characteristics after the ARIMA model. This architecture ensures real-time online concentration forecasting. The effectiveness of our approach has been substantiated with actual production data from a large-scale mineral processing facility.

In mineral processing, the dynamic nature of industrial data poses challenges for decision-makers in accurately assessing current production statuses. To enhance the decision-making process, it is crucial to predict comprehensive production indices (CPIs), which are influenced by both human operators and industrial processes, and demonstrate a strong dual-scale property. To improve the accuracy of CPIs’ prediction, we introduce the high-frequency (HF) unit and low-frequency (LF) unit within our proposed dual-scale deep learning (DL) network. This architecture enables the exploration of nonlinear dynamic mapping in dual-scale industrial data. By integrating the Cloud-Edge collaboration mechanism with DL, our training strategy mitigates the dominance of HF data and guides networks to prioritize different frequency information. Through self-tuning training via Cloud-Edge collaboration, the optimal model structure and parameters on the cloud server are adjusted, with the edge model self-updating accordingly. Validated through online industrial experiments, our method significantly enhances CPIs’ prediction accuracy compared to the baseline approaches.

In this article, the method of dynamic performance monitoring and adaptive self-tuning of parameters for actual PID control systems of industrial processes in virtual reality scenes is proposed. This method combines the digital twin model of the PID control process based on system identification and adaptive deep learning and the PID tuning intelligent algorithm based on reinforcement learning with virtual reality and immersive interaction of industrial metaverse. An industrial metaverse-based intelligent PID tuning system is proposed by combining the above method with the end–edge–cloud collaboration technology of Industrial Internet. The challenging problem that the actual operating PID control system in complex industrial processes cannot be optimized online is solved. Using the energy-intensive equipment, the fused magnesium furnace, as an industrial object, we conducted comparative simulation experiments between the proposed control method and several advanced control methods, as well as industrial experiments for the proposed intelligent system. Simulation experiments demonstrate the effectiveness of the proposed control method. The industrial experimental results indicate that the performance monitoring and adaptive self-tuning of parameters for actual PID control systems of industrial processes in virtual reality scenes can be realized, which achieves excellent control effects.