Xinyi Liu’s research while affiliated with Central South University and other places

Modern industrial processes often exhibit complex and uncertain operating state fluctuations due to the diversification of production materials, the complexity of production processes, and the harsh production environment. To address the real-time control challenge in multimode processes, this article proposes a learning framework for fuzzy neural network explicit control for full operation conditions. It includes: the fuzzy neural network-based explicit control law (FNNECL) and the neural network-based predictive model (NNPM). By adjusting the structure and parameters of FNNECL, precise control of full operation conditions is achieved. Specifically, a novel truncated radial basis activation function is first presented, and the operation condition change measurement index of data coverage is proposed to identify the mismatch degree between current and historical operation conditions. Then, for small-scale operation condition changes, an elastic weight consolidation (EWC) mechanism is introduced to ensure that FNNECL can learn control strategies for new operation conditions while maintaining control performance for historical conditions. Finally, to address large-scale operation condition changes, a radial basis function (RBF) neuron growth mechanism based on data coverage is proposed. This mechanism calculates the data coverage of fuzzy rules and selectively adds new neurons to the FNNECL. This enables the FNNECL to fit large-scale changes and effectively learn control strategies for new operation conditions. It is worth noting that without knowing the current condition, accurate control sequences can be rapidly obtained based on a single FNNECL. Extensive experiments including numerical simulation and industrial roasting process verified the feasibility and effectiveness of the proposed method.

Modern industrial process control suffers from various difficulties, such as multivariable, multiconstrained, multiobjective, and strong nonlinearity. Model predictive control (MPC) is an effective solution and is widely used in industrial processes. However, one limitation of MPC is that sufficient data are required to build accurate predictive models. To this end, this article proposes a knowledge-informed neural network MPC solution. First, a Hammerstein system structure knowledge extraction method based on sparse representation is proposed, which is able to extract system structure knowledge from a small amount of system operation data. Then, a knowledge-informed neural network model is designed, which combines the system structure knowledge to construct a neural network with a special structure, thus overcoming the problem of insufficient data during the model training. Finally, the knowledge-informed neural network model is embedded in the MPC framework, which can reduce the computational cost of rolling optimization while ensuring prediction performance. A numerical simulation and a pH neutralization process experiment are conducted to verify the feasibility and effectiveness of the proposed method.

Edge computing is an indispensable technology in the Industry 4.0 era. Data is stored at the edge and transmission is prohibited due to concerns about data privacy. Thus, the traditional centralized data-driven modeling methods will face many difficulties and can hardly be used in practice. Federated learning aims to let local nodes learn a global model cooperatively while keeping their data localized. Motivated by the pioneering framework of federated learning, a federated dictionary learning method is proposed for process monitoring. In detail, local nodes train local data by K-SVD method and learn local dictionaries. Then, local dictionaries, instead of local data, are transferred to the fusion center for calculation of the global dictionary. In order to guarantee an optimal global dictionary, a novel federated dictionary average strategy is introduced. Finally, the reconstruction errors of local data and the control limit are calculated based on the global dictionary for process monitoring. The performance of the proposed method is verified on a numerical simulation, the CSTH benchmark process, and the industrial aluminum electrolysis process.