Search

Fluorescence microscopic images of cells contain a large number of morphological features that are used as an unbiased source of quantitative information about cell status, through which researchers can extract quantitative information about cells and study the biological phenomena of cells through statistical and analytical analysis. As an important research object of phenotypic analysis, images have a great influence on the research results. Saturation artifacts present in the image result in a loss of grayscale information that does not reveal the true value of fluorescence intensity. From the perspective of data post-processing, we propose a two-stage cell image recovery model based on a generative adversarial network to solve the problem of phenotypic feature loss caused by saturation artifacts. The model is capable of restoring large areas of missing phenotypic features. In the experiment, we adopt the strategy of progressive restoration to improve the robustness of the training effect and add the contextual attention structure to enhance the stability of the restoration effect. We hope to use deep learning methods to mitigate the effects of saturation artifacts to reveal how chemical, genetic, and environmental factors affect cell state, providing an effective tool for studying the field of biological variability and improving image quality in analysis. [ABSTRACT FROM AUTHOR]

To identify the location of the strongest vibration of skeletal membrane structures under pulsating wind and predict the response and damage of the structures. To guide engineering reinforcement, base on the platform of ANSYS Workbench as a platform, the complex skeleton membrane structure is first simplified using the modal fitting method, and the error of the 30th-order frequency of the membrane structure before and after the simplification is calculated to be less than 5% compared to the unsimplified model. Based on the simplified model, the unidirectional coupling and bidirectional coupling methods were compared using the surface averaging wind pressure distribution parameters of the membrane structure. The results show that the extreme value and extreme difference of the averaged wind pressure of the bidirectional coupling are larger than those of the unidirectional coupling under the same wind angle, which verifies the reliability of the vibration analysis of the large-span membrane structure under the bidirectional coupling. By analyzing the wind pressure coefficients at the key measure points, it can be seen that the structure is in unfavorable working condition at the wind angle of 0°–45°. Based on the bidirectional coupling method, the maximum displacement time variation curve, maximum acceleration time variation curve, and maximum equivalent force time variation curve for different wind direction angles are basically the same, and the membrane structure gradually equilibrates with time, and the maximum instantaneous displacement, maximum instantaneous acceleration, and maximum equivalent force all reach the maximum at wind direction angle 30° and the minimum at 90°. [ABSTRACT FROM AUTHOR]

Rule-based energy management strategies not only make little use of the efficient area of engines and generators but also need to perform better planning in the time domain. This paper proposed a multi-signal vehicle speed prediction model based on the long short-term memory (LSTM) network, improving the accuracy of vehicle speed prediction by considering multiple signals. First, various signals were collected by simulating the vehicle model, and a Pearson correlation analysis was performed on the collected multiple signals in order to improve the model's prediction accurate, and the appropriate signal was selected as the input to the prediction model. The experimental results indicate that the prediction method greatly improves the predictive effect compared with the support vector machine (SVM) vehicle speed prediction method. Secondly, the method was combined with the model predictive control-equivalent consumption strategy (MPC-ECMS) to form a control strategy suitable for power maintenance conditions enabling the equivalent factor to be adjusted adaptively in real-time and the target state of charge (SoC) value to be set. Pontryagin minimum principle (PMP) enables the battery to calculate the range extender output power at each moment. PMP, as the core algorithm of ECMS, is a common real-time optimal control algorithm. Then, taking into account the engine's operating characteristics, the calculated range extender power was filtered to make the engine run smoothly. Finally, hardware-in-the-loop simulation (HIL) was used to verify the model. The simulation results demonstrate that this method uses less fuel than the equivalent fuel consumption minimum strategy (ECMS) by 1.32%, 9.47% when compared to the power-following control strategy, 15.66% when compared to the SVM-MPC-ECMS, and only 3.58% different from the fuel consumption of the dynamic programming (DP) control algorithm. This shows that this energy management approach can significantly improve the overall vehicle fuel economy. [ABSTRACT FROM AUTHOR]