The Maximum Minimal Cut Problem (MMCP), a NP-hard combinatorial optimization (CO) problem, has not received much attention due to the demanding and challenging bi-connectivity constraint. Moreover, as a CO problem, it is also a daunting task for machine learning, especially without labeled instances. To deal with these problems, this work proposes an unsupervised learning framework combined with heuristics for MMCP that can provide valid and high-quality solutions. As far as we know, this is the first work that explores machine learning and heuristics to solve MMCP. The unsupervised solver is inspired by a relaxation-plus-rounding approach, the relaxed solution is parameterized by graph neural networks, and the cost and penalty of MMCP are explicitly written out, which can train the model end-to-end. A crucial observation is that each solution corresponds to at least one spanning tree. Based on this finding, a heuristic solver that implements tree transformations by adding vertices is utilized to repair and improve the solution quality of the unsupervised solver. Alternatively, the graph is simplified while guaranteeing solution consistency, which reduces the running time. We conduct extensive experiments to evaluate our framework and give a specific application. The results demonstrate the superiority of our method against two techniques designed.

In recent years, the demand for computing resources of AI industry is urgent because of the data explosion, which promoted the construction of computing power networks in the new era for operators. From the cloud network era to today's computing power network, stricter requirements are proposed to ensure the efficiency and security of computing services. Despite computing power scheduling technologies such as on-demand edge computing and efficient compute first network, knowledge graph techniques for graphs are less explored. As a new technology that can express the relationship between nodes in the graph extremely easily, knowledge graph has a natural advantage in expressing feature information of computing nodes in computing power network. Therefore, a novel knowledge graph representation for the architecture of computing power networks is proposed. And the knowledge graph of the computing power network is constructed by using the knowledge representation method. The scheduling tasks of computing power network is automatically executed by the proposed knowledge driven method based on the constructed knowledge graph. Different with the current scheduling technology of computing power network, the model will theoretically become more and more efficient and accurate with continuously addition of knowledge.

Grid-connected converters have been widely used in distributed generations. However, the resonance brought by converters in weak grid has been a great threat to system stability. Impedance-based method provides a powerful tool for stability analysis. Active damping is also widely applied to improve system stability. This paper derives the impedance model of grid-connected converter system and analyzes the harmonic resonance mechanism. The effects of passive damping and active damping are also compared, especially their influence on non-passive region. Impact of capacitor current feedback gain on the output admittance is also analyzed. High-pass filter is used to compensate the digital control delay and reduce the non-passive region of output admittance. An improved active damping is proposed to eliminate second harmonic introduced by active damping. The performance of proposed method is verified in Matlab/Simulink simulations. The simulation result shows the effectiveness of proposed method.

Automatic wafer prealignment is an important process in wafer manufacturing, whereby its flat edge is set at a predefined angle and its center is set in a predefined position; translational and rotational movement compensate for deviation of the wafer. However, as the traditional wafer prealignment algorithm depends on marked templates at the training stage when the type of wafer is changed, the templates need to be retrained. This paper proposes a new wafer prealignment algorithm, based on Fourier transform for orientation prealignment and least square regression for position prealignment, which will automatically adapt to different types of wafers in 2-D space. Results from experiments with the proposed algorithm on a laser-scribing machine using two types of wafer show that the orientation prealignment achieved an accuracy of 0.05°, the position prealignment achieved an accuracy of 5 pixels, and the average operation time was approximately 1.5 s. As prealignment algorithm therefore meets real-time efficiency and precision requirements, it is suitable for use with semiconductor devices.