This work describes the architecture for developing physics of failure models, derived as a function of machine sensor data, and integrating with data pertaining to other relevant factors like geography, manufacturing, environment, customer and inspection information, that are not easily modeled using physics principles. The mechanics of the system is characterized using surrogate models for stress and metal temperature based on results from multiple non-linear finite element simulations. A cumulative damage index measure has been formulated that quantifies the health of the component. To address deficiencies in the simulation results, a model tuning framework is designed to improve the accuracy of the model. Despite the model tuning, un-modelled sources of variation can lead to insufficient model accuracy. It is required to incorporate these un-modelled effects so as to improve the model performance. A novel machine learning based model fusion approach has been presented that can combine physics model predictions with other data sources that are difficult to incorporate in a physics framework. This approach has been applied to a gas turbine hot section turbine blade failure prediction example.

We describe a rare case of a 57-year-old women diagnosed with Klazomania (compulsive shouting), rarely reported in literature, complicated by Major Depressive Disorder (MDD), refractory to medical management and treated successfully using electroconvulsive therapy (ECT). This is a unique presentation in which compulsions occurring in the context of MDD involved shouting rather than more typical compulsions such as hand washing or counting. Potential neurological disorders were ruled out through an extensive diagnostic workup prior to a course of electroconvulsive therapy (ECT). Her episodic compulsive shouting abated and eventually remitted after a 12-session course of ECT.Keywords: Klazomania; Compulsive screaming; ECT; Complication of major depression; Adult psychiatry; Elderly depression; Elderly screaming; Aging psychiatry Case ReportMs. X, a 57-year-old Ethiopian woman was admitted to the psychiatric service with a 2-year history of increasingly frequent episodic paroxysmal compulsive shouting (from once every month to a quotidian occurrence), complicated by treatment resistant Major Depressive Disorder (MDD). The shouting episodes were characterized by loud screaming and screeching sounds, occasionally in speech that was perseverative and stereotypic with small sentences, and concurrently accompanied by crying. During these episodes, she was completely oriented; aware she was screaming and able to recollect her thoughts. Afterwards, she reported feeling anxious and irritated because of her inability to control her vocalizations. The episodes occurred daily, 3-7 per hour every hour she was awake, variable in duration, and were emotionally consonant with her mood state (precipitated by intrusive thoughts of a recently deceased fiance). They occurred in a variety of settings (work, home, social) that affected her quality of life and interfered with her work performance. She was admitted twice on previous occasions for the same complaint.Her psychiatric history was significant for MDD that was likely triggered by several major life stressors beginning with the death of her fiancé 2-years prior. During this time she developed severe depression characterized by a low mood, decreased energy, poor concentration, sleep disturbances, lack of appetite, anhedonia, and feelings of worthlessness and hopelessness. As mentioned earlier, it was also during this period that her shouting episodes gradually increased in frequency to a quotidian occurrence. She endorsed no history of hallucinations or head trauma. Ms. X's past medical history was significant for Type 2 Diabetes Mellitus, Hypertension, Chronic Back and Neck Pain, Anxiety, and Mood disturbances. Laboratory evaluations, Lumbar Puncture (LP) revealed a slightly elevated protein level of 60mm/dL (otherwise LP results were within normal limits), a Electroencephalogram (EEG) with and without vocalizations was negative, and a Head-CT revealed periventricular white matter hypodensities with global atrophy. Magnetic Resonance Imaging (MRI) was remarkable for num...

No abstract

This paper presents a multistage identification scheme for structural damage detection using modal data. Previous studies of damage assessment using neural networks mostly involved training a backpropagation neural network (BPN) to learn damage patterns that were obtained either experimentally or by simulation for different damage cases. Damage identification for large structures, especially those involving multiple member damage, could result in large training data sets that require a large BPN and consequently greater computational effort. The proposed scheme involves using a counterpropagation neural network (CPN) in the first stage for sorting the training data into clusters and giving an approximate guess of the damage extent within a very short time. After an approximate estimate of the damage is obtained, a new set of training patterns of reduced size is generated using the CPN prediction. In the second stage, a BPN trained with the Levenberg-Marquardt algorithm is used to learn the new training data and predict a more accurate result. A superior convergence and a substantial decrease in central processing unit (CPU) time are observed for three numerical examples.

No abstract