Wind Turbine Condition Monitoring Using the SSA-Optimized Self-Attention BiLSTM Network and Changepoint Detection Algorithm.

Condition-monitoring and anomaly-detection methods used for the assessment of wind turbines are key to reducing operation and maintenance (O&M) cost and improving their reliability. In this study, based on the sparrow search algorithm (SSA), bidirectional long short-term memory networks with a self-attention mechanism (SABiLSTM), and a binary segmentation changepoint detection algorithm (BinSegCPD), a condition-monitoring method (SSA-SABiLSTM-BinSegCPD, SSD) used for wind turbines is proposed. Specifically, the self-attention mechanism, which can mine the nonlinear dynamic characteristics and spatial-temporal features inherent in the SCADA time series, was introduced into a two-layer BiLSTM network to establish a normal-behavior model for wind turbine key components. Then, as a result of the advantages of searching precision and convergence rate methods, the sparrow search algorithm was employed to optimize the constructed SABiLSTM model. Moreover, the BinSegCPD algorithm was applied to the predicted residual sequence to achieve the automatic identification of deterioration conditions for wind turbines. Case studies conducted on multiple wind turbines located in south China showed that the established SSA-SABiLSTM model was superior to other contrast models, achieving a better prediction precision in terms of RMSE, MAE, MAPE, and R2. The MAE, RMSE, and MAPE of SSA-SABiLSTM were 0.2543 °C, 0.3412 °C, and 0.0069, which were 47.23%, 42.19%, and 53.38% lower than those of SABiLSTM, respectively. The R2 of SABiLSTM was 0.9731, which was 4.6% higher than that of SABiLSTM. The proposed SSD method can detect deterioration conditions 47-120 h in advance and trigger fault alarm signals approximately 36 h ahead of the actual failure time.

Monitoring the extracellular matrix remodeling of high-grade serous ovarian cancer with nonlinear optical microscopy.

The mortality of high-grade serous ovarian cancer (HGSOC) accounts for 70% to 80% of all ovarian cancer deaths and overall mortality rate has not declined in the last decade. Recently, many studies have demonstrated that HGSOC originates from the fallopian tubes. The extracellular matrix (ECM) is present in all tissues, its remodeling and interaction with cells are crucial for regulating cell proliferation, migration, and differentiation. In this paper, we used label-free nonlinear optical microscopy to image tissues of the fallopian tube and ovary. Combining a set of image processing algorithms, we monitored the remodeling of ECM in the fallopian tube and ovary during the invasion of primary serous fallopian tube tumor into the ovary in microscopic dimension. With this approach, we can obtain physiological information of HGSOC at the early stage, which provided useful data for auxiliary clinical diagnosis.

Human serum albumin gradient in serous ovarian cancer cryosections measured by fluorescence lifetime.

Human serum albumin (HSA) is a depot and carrier for many endogenous and exogenous molecules in blood. Many studies have demonstrated that the transport of HSA in tumor microenvironments contributes to tumor development and progression. In this report, we set up a multimodal nonlinear optical microscope system, combining two-photon excitation fluorescence, second harmonic generation, and two-photon fluorescence lifetime imaging microscopy. The fluorescence lifetime of a small squaraine dye (SD) is used to evaluate HSA concentrations in tumor tissue based on specific binding between SD and HSA. We used SD to stain the cryosections from serous ovarian cancer patients in high-grade (HGSOC) and low-grade (LGSOC), respectively, and found a gradient descent of HSA concentration from normal connective tissue to extracellular matrix to tumor masses from 13 to 2 µM for LGSOC patients and from 36 to 12 µM for HGSOC patients. We demonstrated that multimodal nonlinear optical microscopy can obtain similar results as those from traditional histologic staining, thus it is expected to move to clinical applications.

Label-free whole-colony imaging and metabolic analysis of metastatic pancreatic cancer by an autoregulating flexible optical system.

Cancer metastasis is a Gordian knot for tumor diagnosis and therapy. Many studies have demonstrated that metastatic processes are inevitably affected by the tumor microenvironment. Histopathology is used universally as the gold standard for cancer diagnosis despite the lengthy preparation process and invasiveness. Methods: Here, we introduced a supercontinuum and super-wide-tuning integrated multimodal platform, which combines the confocal, nonlinear and fluorescence lifetime microscopy with autoregulations, for label-free evaluation of fresh tissue and pathological sections. Based on various automated tunable lasers, synchronized and self-adjusting components and eight fast switching detection channels, the system features fast, large-field and subcellular-scale imaging of exogenous and endogenous fluorophores, nonlinear coherent scattering and lifetime contrast. Results: With such an integrated multi-dimensional system, we searched the metastatic region by two-photon and three-photon excited autofluorescence, analyzed the cancer invasion by second harmonic generation and revealed the affected cellular metabolism by phasor-lifetime. We demonstrated the flexible measurement of multiple nonlinear modalities at NIR I and II excitation with a pre-compensation for group delay dispersion of ~7,000 fs2 and low power of <40 mW, and of dual autofluorescence lifetime decays for phasor approach to decompose cancer-associated and disassociated components. This significantly revealed the metastatic and metabolic optical signatures of the whole colony of pancreatic cancers. Conclusion: The synergistic effect of the system demonstrates the great potential to translate this technique into routine clinical applications, particularly for large-scale and quantitative studies of metastatic colonization.