Further development of the photovoltaic industry is restricted by the productivity of mono-crystalline silicon technology due to its requirements of low cost and high efficient photocells. The heat shield is not only the important part of the thermal field in Czochralski(Cz) mono-crystalline silicon furnace, but also one of the most important factors influencing the silicon crystal growth. Large-diameter Cz-Si crystal growth process is taken as the study object. Based on FEM numerical simulation, different heat shield structures are analyzed to investigate the heater power, the melt-crystal interface shape, the argon flow field, and the oxygen concentration at the melt-crystal interface in the process of large Cz-Si crystal growth. The impact of these factors on the growth efficiency and crystal quality are analyzed. The results show that the oxygen concentration on the melt-crystal interface and the power consumption of the heater stay high due to the lack of a heat shield in the crystal growth system. Argon circumfluence is generated on the external side of the right angle heat shield. By the right-angle heat shield, the speed of gas flow is lowered on the melt free surface, and the temperature gradient of the free surface is increased around the melt-crystal interface. It is not conducive for the stable growth of crystal. The shape of the melt-crystal interface and the argon circulation above the melt free surface are improved by the inclined heat shield. Compared with the others, the system pulling rate is increased and the lowest oxygen concentration is achieved at the melt-crystal interface with the composite heat shield. By the adoption of the optimized composite heat shield in experiment, the real melt-crystal interface shapes and its deformation laws obtained by Quick Pull Separation Method at different pulling rates agree with the simulation results. The results show that the method of simulation is feasible. The proposed research provides the theoretical foundation for the thermal field design of the large diameter Cz-Si monocrystalline growth.

In object detection networks, one widely used and effective approach to address the challenge of detecting small-sized objects in images is to employ multiscale pyramid features for prediction. Based on the fundamental principles of pyramid feature extraction, shallow features with small receptive fields are responsible for predicting small-sized objects, while deep features with large receptive fields handle large-sized objects. However, during the actual network training process using this structure, the loss function only provides the error between all positive samples and labels, treating them equally without considering the relationship between the actual size of the label and the feature layer where the sample resides, which to some extent affects the object detection performance. To address this, this paper proposes a novel method for constructing a loss function, termed Predicted Probability Loss (PP-Loss). It determines the probability of each feature layer predicting the objects labeled by the labels based on the size of the labels and uses this probability to adjust the weights of different sample anchors in the loss function, thereby guiding the network training. The prediction probability values for each feature layer are obtained from a prediction probability function established on a statistical basis. The algorithm has been experimentally validated on different networks with YOLO as the core. The results show that the convergence speed and accuracy of the network during training have been improved to varying degrees.

In the Czochralski silicon single crystal growth process, the tail diameter is a key parameter that cannot be directly measured. In this paper, we propose a real-time soft measurement method that combines a deep belief network (DBN) and a support vector regression (SVR) network based on system identification to accurately predict the crystal diameter. The main steps of the proposed method are as follows: First, we address the delay problem of the effects of the temperature and crystal pulling speed on the tail diameter growth by using a back propagation (BP) neural network based on the mean impact value (MIV) method to determine the optimal delay time. Second, we construct a prediction model of the tail diameter by using the DBN network with the temperature and crystal pulling speed as input variables in the crystal growth process. Third, we improve the DBN network by using the SVR network to enhance its linear regression capability. We also employ the ant colony optimization (ACO) algorithm to obtain the optimal parameters of the SVR network. Finally, we compare the performance of the DBN-ACO-SVR network based on system identification with the DBN and SVR networks, and the results show that our method can effectively deal with the delay problem and achieve the accurate prediction of the tail diameter in the Czochralski silicon single crystal growth process.