skip to main content

Title: The Effects of Gravity on the Response of Centrifugal Pendulum Vibration Absorbers1
Abstract This article describes the effects of gravity on the response of systems of identical, cyclically arranged, centrifugal pendulum vibration absorbers (CPVAs) fitted to a rotor spinning about a vertical axis. CPVAs are passive devices composed of movable masses suspended on a rotor, suspended such that they reduce torsional vibrations at a given engine order. Gravitational effects acting on the absorbers can be important for systems spinning at relatively low rotation speeds, for example, during engine idle conditions. The main goal of this study is to predict the response of a CPVA/rotor system in the presence of gravity. A linearized model that includes the effects of gravity and an order n torque acting on the rotor is analyzed by exploiting the cyclic symmetry of the system. The results show that a system of N absorbers responds in one or more groups, where the absorbers in each group have identical waveforms but shifted phases. The nature of the waveforms can have a limiting effect on the absorber operating envelope. The number of groups is shown to depend on the engine order n and the ratio N/n. It is also shown that there are special resonant effects if the engine order is more » n = 1 or n = 2, the latter of which is particularly important in applications. In these cases, the response of the absorbers has a complicated dependence on the relative levels of the applied torque and gravity. In addition, it is shown that for N > 1, the rotor response is not affected by gravity, at least to leading order, due to the cyclic symmetry of the gravity effects. The linear model and the attendant analytical predictions are verified by numerical simulations of the full nonlinear equations of motion. « less
; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
Journal of Vibration and Acoustics
Sponsoring Org:
National Science Foundation
More Like this
  1. Switched reluctance motors (SRM) have been seen as a potential candidate for automotive, aerospace as well as domestic applications and High-Rotor pole SRM (HR-SRM) present a significant advancement in this area. This machine configuration offers most of the the benefits offered by conventional SRMs and has shown significant benefits in efficiency and torque quality. However, HR-SRM has a narrower inductance profile with a lower saliency ratio as compared to a conventional SRM with an identical stator. This can make it inherently challenging to directly adopt mathematical models and sensorless control approaches currently in use. This paper presents a time-efficient analytical model for the characterization of a 6/10 SRM using an inductance model utilizing truncated Fourier series as well as multi-order polynomial curve-fitting algorithm. The inductance model is extended to accurately predict back-EMF and electromagnetic torque response towards obtaining a comprehensive model for every operating point of the machine during dynamic operation. The effectiveness of the proposed concept has analyzed for a prototype machine and verified using Finite Element Analysis (FEA).
  2. This paper is an extended version of our paper presented at the 2016 TORQUE conference (Shapiro et al., 2016). We investigate the use of wind farms to provide secondary frequency regulation for a power grid using a model-based receding horizon control framework. In order to enable real-time implementation, the control actions are computed based on a time-varying one-dimensional wake model. This model describes wake advection and wake interactions, both of which play an important role in wind farm power production. In order to test the control strategy, it is implemented in a large-eddy simulation (LES) model of an 84-turbine wind farm using the actuator disk turbine representation. Rotor-averaged velocity measurements at each turbine are used to provide feedback for error correction. The importance of including the dynamics of wake advection in the underlying wake model is tested by comparing the performance of this dynamic-model control approach to a comparable static-model control approach that relies on a modified Jensen model. We compare the performance of both control approaches using two types of regulation signals, RegA and RegD, which are used by PJM, an independent system operator in the eastern United States. The poor performance of the static-model control relative to the dynamic-model control demonstratesmore »that modeling the dynamics of wake advection is key to providing the proposed type of model-based coordinated control of large wind farms. We further explore the performance of the dynamic-model control via composite performance scores used by PJM to qualify plants for regulation services or markets. Our results demonstrate that the dynamic-model-controlled wind farm consistently performs well, passing the qualification threshold for all fast-acting RegD signals. For the RegA signal, which changes over slower timescales, the dynamic-model control leads to average performance that surpasses the qualification threshold, but further work is needed to enable this controlled wind farm to achieve qualifying performance for all regulation signals.« less
  3. Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)
    The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should producemore »identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues.« less
  4. Abstract
    Excessive phosphorus (P) applications to croplands can contribute to eutrophication of surface waters through surface runoff and subsurface (leaching) losses. We analyzed leaching losses of total dissolved P (TDP) from no-till corn, hybrid poplar (Populus nigra X P. maximowiczii), switchgrass (Panicum virgatum), miscanthus (Miscanthus giganteus), native grasses, and restored prairie, all planted in 2008 on former cropland in Michigan, USA. All crops except corn (13 kg P ha−1 year−1) were grown without P fertilization. Biomass was harvested at the end of each growing season except for poplar. Soil water at 1.2 m depth was sampled weekly to biweekly for TDP determination during March–November 2009–2016 using tension lysimeters. Soil test P (0–25 cm depth) was measured every autumn. Soil water TDP concentrations were usually below levels where eutrophication of surface waters is frequently observed (> 0.02 mg L−1) but often higher than in deep groundwater or nearby streams and lakes. Rates of P leaching, estimated from measured concentrations and modeled drainage, did not differ statistically among cropping systems across years; 7-year cropping system means ranged from 0.035 to 0.072 kg P ha−1 year−1 with large interannual variation. Leached P was positively related to STP, which decreased over the 7 years in all systems. These results indicate that both P-fertilized and unfertilized cropping systems mayMore>>
  5. Venom systems are key adaptations that have evolved throughout the tree of life and typically facilitate predation or defense. Despite venoms being model systems for studying a variety of evolutionary and physiological processes, many taxonomic groups remain understudied, including venomous mammals. Within the order Eulipotyphla, multiple shrew species and solenodons have oral venom systems. Despite morphological variation of their delivery systems, it remains unclear whether venom represents the ancestral state in this group or is the result of multiple independent origins. We investigated the origin and evolution of venom in eulipotyphlans by characterizing the venom system of the endangered Hispaniolan solenodon ( Solenodon paradoxus ). We constructed a genome to underpin proteomic identifications of solenodon venom toxins, before undertaking evolutionary analyses of those constituents, and functional assessments of the secreted venom. Our findings show that solenodon venom consists of multiple paralogous kallikrein 1 ( KLK1 ) serine proteases, which cause hypotensive effects in vivo, and seem likely to have evolved to facilitate vertebrate prey capture. Comparative analyses provide convincing evidence that the oral venom systems of solenodons and shrews have evolved convergently, with the 4 independent origins of venom in eulipotyphlans outnumbering all other venom origins in mammals. We findmore »that KLK1 s have been independently coopted into the venom of shrews and solenodons following their divergence during the late Cretaceous, suggesting that evolutionary constraints may be acting on these genes. Consequently, our findings represent a striking example of convergent molecular evolution and demonstrate that distinct structural backgrounds can yield equivalent functions.« less