To optimize machine allocation and task dispatching in smart manufacturing factories, this paper proposes a manufacturing resource scheduling framework based on reinforcement learning (RL). The framework formulates the entire scheduling process as a multi-stage sequential decision problem, and further obtains the scheduling order by the combination of deep convolutional neural network (CNN) and improved deep Q-network (DQN). Specifically, with respect to the representation of the Markov decision process (MDP), the feature matrix is considered as the state space and a set of heuristic dispatching rules are denoted as the action space. In addition, the deep CNN is employed to approximate the state-action values, and the double dueling deep Q-network with prioritized experience replay and noisy network (D3QPN2) is adopted to determine the appropriate action according to the current state. In the experiments, compared with the traditional heuristic method, the proposed method is able to learn high-quality scheduling policy and achieve shorter makespan on the standard public datasets.

No abstract

A method to one-pot synthesize solid fluorescent materials by coating carbon dots via the crystallization of phthalic acid precursor as matrix is reported. The crystalline CDs， emitting blue and green fluorescence respec-tively， are synthesized by changing the solvent in microwave method. After detailed structural and spectral character-izations， it is found that a crystalline matrix is grown around the CDs by the phthalic acid precursor when the CDs is formed， and the dispersion effect of the matrix on the CDs effectively blocked the aggregation of the CDs， thus preventing the occurrence of fluorescence quenching of the CDs. Furthermore， the change of crystalline matrix structure in G-CDs leads to the increase of pyridine nitrogenous groups at the interface between core and matrix in the CDs， resulting in a change of fluorescent color of the CDs with different crystalline structures. The resultant crystalline CDs are also used to fabricate white-light emitting diode devices (WLED) in view of its excellent luminescence perfor-mance， which achieves a warm-white light with the correlated color temperature (CCT) of 4 061 K， color rendering index (CRI) of 88. 4 at the chromaticity coordinates of (0. 37， 0. 36) using B-CDs combined with the commercial phosphors and another warm-white light with the CCT of 4 478 K， CRI of 85 at the chromaticity coordinates of (0. 36， 0. 34) using G-CDs combined with the commercial phosphors. The excellent photometric parameters give

Time series data has attached extensive attention as multi-domain data, but it is difficult to analyze due to its high dimension and few labels. Self-supervised representation learning provides an effective way for processing such data. Considering the frequency domain features of the time series data itself and the contextual feature in the classification task, this paper proposes an unsupervised Long Short-Term Memory (LSTM) and contrastive transformer-based time series representation model using contrastive learning. Firstly, transforming data with frequency domain-based augmentation increases the ability to represent features in the frequency domain. Secondly, the encoder module with three layers of LSTM and convolution maps the augmented data to the latent space and calculates the temporal loss with a contrastive transformer module and contextual loss. Finally, after self-supervised training, the representation vector of the original data can be got from the pre-trained encoder. Our model achieves satisfied performances on Human Activity Recognition (HAR) and sleepEDF real-life datasets.

No abstract