For batch processes with partial actuator failures and unknown system dynamics, an innovative two-dimensional (2D) model-free Q-learning algorithm is proposed to obtain the optimal controller's gains, achieving output feedback fault-tolerant control. First, a 2D linear model is constructed to describe batch processes with partial actuator failures. Then, the state increments in the batch direction and the output errors in the time direction are used as novel state variables to construct a multi-degree-of-freedom model. Second, a 2D Bellman equation is proposed through a connection between a 2D value and a 2D Q functions. Next, a 2D off-policy model-free Q-learning algorithm is highlighted, which incorporates target policies into a multi-degree-of-freedom model and focuses on using policy iteration to solve the fault-tolerant tracking control problem. The robustness analysis rigorously proves the stability of the closed-loop system. Lastly, the simulation results of the holding stage prove the feasibility and effectiveness of the presented algorithm. 

Two-dimensional model-free Q-learning-based output feedback fault-tolerant control for batch processes

In the present work, MgO–TiO2–P2O5–Al2O3–SiO2 glasses with different Ti/P ratios were prepared by the melt-quenching method, and Mg0.5Ti2(PO4)3 glass-ceramics were obtained by controllable crystallization. The effects of the Ti/P ratio on the structure, crystallization behaviour, and properties of the glasses and glass-ceramics were studied by FTIR, DSC, XRD, EDS, UV–Vis DRS, and so on. The results show that with the increase of the Ti/P ratio, the content of the [TiO6] group and Ti–O–P bond in the glass network increases, which strengthens the cross-linking of the glass network. However, the decrease of glass former P2O5 reduces the thermal stability of the glass. The appropriate Ti/P ratio is beneficial to the formation of single-phase glass-ceramics, and the band gap width of glass-ceramics containing single Mg0.5Ti2(PO4)3 crystal phase is 3.0 eV. With the irradiation of a high-pressure mercury lamp for 120 min, the glass-ceramics after crystallization at 850 °C for 4 h can effectively degrade MB dye solution (41%∼75%). The potential application prospect of the prepared glass-ceramics in the field of water purification of organic dyes was confirmed. 

Effect of Ti/P ratio on structure, crystallization behaviour and photocatalytic activity of MgO–TiO2–P2O5–Al2O3–SiO2 glass-ceramics

Injection molding accounts for a huge share of the energy consumption of material processing. It has received considerable attention for its potential to enhance production efficiency, which relies on both product consistency and energy efficiency. Far-infrared heating, which has the advantages of high energy efficiency and low dissipation, has emerged as a promising heating technology in injection molding. However, the issue of low heat dissipation can lead to excess heat accumulation and cause substantial temperature errors. In this paper, a new barrel heating control method based on generalized predictive control (GPC) with real-time excess heat prediction model is proposed, in order to improve control accuracy and heating efficiency of far-infrared heating. A constrained multistep optimization combined with excess heat compensation is stated and an experimental platform is designed to validate the superiority of the proposed method under different operation scenarios. Experimental results show that the proposed method outperforms traditional proportional-integral-differential (PID) and GPC methods. Specifically, the proposed method can improve temperature control accuracy by 85.3 and 78.9 %, compared to PID and GPC, respectively. Such high temperature accuracy goes a long way in improving product consistency. Moreover, energy efficiency is improved, e.g., heating time is reduced by 42.7 and 16.5 %, energy consumption is reduced by 27.9 and 25.4 %, compared to PID and GPC, respectively. Control robustness is also verified under different operating conditions. In second-heating experiments, the temperature accuracy is improved by 84.4 % compared to PID. Therefore, the proposed method has the potential to enhance the production efficiency of injection molding. 

High efficiency far-infrared barrel heating control with excess heat prediction based on generalized predictive control in injection molding

With developments in global economic integration and the increase in future economic uncertainty, it is imperative to have the ability to predict future capital in relation to financial capital inflow and outflow predictions to ensure capital optimization is within a controllable range within the current macroeconomic environment and situation. This paper proposes an automated capital prediction strategy for the capital supply chain using time series analysis artificial intelligence methods. Firstly, to analyze the fluctuation and tail risk of the financial characteristics, the paper explores the financial characteristics for measuring the dynamic VaR from the perspectives of volatility, tail, and peak with the Bayesian peaks over threshold (POT) model. Following this, in order to make the modeling more refined, the forecast targets are split before modeling with seasonal Autoregressive Integrated Moving Average (ARIMA) models and Prophet models. Finally, the time series modeling of the wavelet Long Short-Term Memory (LSTM) model is carried out using a two-part analysis method to determine the linear separated wavelet and non-linear embedded wavelet parts to predict strong volatility in financial capital. Taking the user capital flow of the Yu’e Bao platform, the results prove the feasibility and prediction accuracy of the innovative model proposed.

Improved Financial Predicting Method Based on Time Series Long Short-Term Memory Algorithm

In this article, an iterative cost-learning model predictive control (ICLMPC) is proposed for nonlinear networked control systems (NCSs) in the presence of aperiodic sampling. The proposed ICLMPC is useful not only to guarantee asymptotic stability of the closed-loop system with aperiodic sampling but also to improve control performance in the case of performing an iterative task. In the proposed method, the nonlinear system of NCSs is mathematically represented as an aperiodic sampled-data Takagi–Sugeno (T–S) fuzzy system. Based on this representation, the ICLMPC design is formulated in terms of a finite-horizon optimal control problem in which a new terminal cost function is considered. The terminal cost function is constructed by a Lyapunov function with a looped-functional and an iteratively minimized function (IMF). From the Lyapunov function with the looped-functional, it is possible to guarantee that the ICLMPC asymptotically stabilizes the aperiodic sampled-data T–S fuzzy system. To obtain an iteratively improved control performance, the IMF takes the minimized value among the integrals of the collected data at each iteration. The validity and effectiveness of the proposed method are illustrated by two practical examples in the simulation section.

Sampled-Data-Based Iterative Cost-Learning Model Predictive Control for T–S Fuzzy Systems

Iterative learning model predictive control (ILMPC) has been considered as potential control strategy for batch processes. ILMPC can converge to the desired reference trajectory with high precision along batches and ensure system stability within batches. However, as a model-based control method, the control performance of the ILMPC algorithm deteriorates when exists model parameter uncertainty. Therefore, guaranteeing system tracking performance in the case of model parameter uncertainty is a challenging task in the framework designing of ILMPC method. To this end, we develop a two-stage robust ILMPC strategy for batch processes, which integrates the robust iterative learning control (ILC) in the domain of batch-axis and robust model predictive control (MPC) in the domain of time-axis into one comprehensive control scheme. The integrated control law of the developed two-stage robust ILMPC algorithm is obtained by solving two convex optimization problems. As a result, the developed control method obtains faster convergence speed and better tracking performance in the case of model parameter uncertainty. Moreover, the convergence analysis of the system is presented. Finally, comparative simulations are provided to verify the superiority of the developed control algorithm. 

A two-stage robust iterative learning model predictive control for batch processes.

Transparent zinc silicate/ zinc oxide crystallized glass-ceramics for water remediation application under visible light

Multi-mode characteristics of industrial processes are prominent in the area of chemical production due to a diversified market demand. Despite mounting interests in predictive modelling for the optimization of operating conditions in chemical production processes, particularly in the petrochemical industry with multiple feeds and a range of cracking furnaces, targeted solutions that could hold wider applicability are typically hindered by the lack of available data. To overcome the limitation posed by data scarcity, an inductive approach based on transfer learning for fault detection is proposed utilizing copula subspace division (CSD), named TrAdaBoost CSD (TCSD). The proposed TCSD method is based on the probability view to transfer different most similar source samples to target samples. To select the optimal number of source samples, two adaptive indices were proposed and designed to adaptively assign the optimal model training iterations and sample number per iteration. Imbalance in data samples between the target and source datasets was addressed via the adaptive active vine copula-based probabilistic method. The effectiveness and superiority of the proposed TCSD approaches are validated via a numerical example (with the ranges of normalized fault detection and false alarm rates [FDR and FAR] between 0.82–0.94 and 0.01–0.04, respectively), the Tennessee Eastman process (~10% and 2.24% improvement for FDR and FAR, respectively), and the ethylene cracking furnace process (~15% and 8% improvement for FDR and FAR, respectively). 

A transfer learning approach using improved copula subspace division for multi‐mode fault detection

This paper proposes a new method of monotonicity, which is used to solve the overfitting problem of the Long-Short-Term Memory (LSTM) model. The main contribution of this paper is applying the monotonicity as priori knowledge to the modeling process. This study uses scatter plots to describe bivariate variables and the Spearman coefficient to extract the monotonicity of data. To exclude most noise point, the scatter diagram is filtered by a binary 0–1 liner program. Base on the monotonicity of data have known, an optimization problem with constraint is proposed to obtain the LSTM neural network model. An experiment of ethylene cracking show that the proposed method can achieve a good predicting performance and less overfitting effects.

Long Short-term Memory modeling method with monotonicity analysis as constraints base on Spearman coefficient

Multi-dimensional climate factors like irradiation and humidity will lead to strong randomness of photovoltaic power generation, and it is difficult to consider all climate factors in traditional photovoltaic power generation point forecasting. Vine copula can accurately and flexibly describe the dependencies between multi-dimensional variables, establish conditional distribution expressions between photovoltaic power generation and multi-dimensional climate factors, and improve prediction accuracy. This paper proposes a D-vine copula model under typical climatic conditions. The model uses fuzzy C-means (FCM) to complete typical climate clustering, and then proposes a D-vine copula model to achieve point predictions and probability interval predictions. The results of the example show that the model has a good performance in point predictions and provide corresponding probability interval predictions, which prove the accuracy and applicability of the model.

Prediction of Photovoltaic Power Generation Based on D-vine Copula Model in Typical Climates

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Yang Zhou

Author Tools

Chat about Author