Explore the vast range of titles available.

Failure detection methods are of significant interest for photovoltaic (PV) site operators to help reduce gaps between expected and observed energy generation. Current approaches for field-based fault detection, however, rely on multiple data inputs and can suffer from interpretability issues. In contrast, this work offers an unsupervised statistical approach that leverages hidden Markov models (HMM) to identify failures occurring at PV sites. Using performance index data from 104 sites across the United States, individual PV-HMM models are trained and evaluated for failure detection and transition probabilities. This analysis indicates that the trained PV-HMM models have the highest probability of remaining in their current state (87.1% to 93.5%), whereas the transition probability from normal to failure (6.5%) is lower than the transition from failure to normal (12.9%) states. A comparison of these patterns using both threshold levels and operations and maintenance (O&M) tickets indicate high precision rates of PV-HMMs (median = 82.4%) across all of the sites. Although additional work is needed to assess sensitivities, the PV-HMM methodology demonstrates significant potential for real-time failure detection as well as extensions into predictive maintenance capabilities for PV.

Classification machine learning models require high-quality labeled datasets for training. Among the most useful datasets for photovoltaic array fault detection and diagnosis are module or string current-voltage (IV) curves. Unfortunately, such datasets are rarely collected due to the cost of high fidelity monitoring, and the data that is available is generally not ideal, often consisting of unbalanced classes, noisy data due to environmental conditions, and few samples. In this paper, we propose an alternate approach that utilizes physics-based simulations of string-level IV curves as a fully synthetic training corpus that is independent of the test dataset. In our example, the training corpus consists of baseline (no fault), partial soiling, and cell crack system modes. The training corpus is used to train a 1D convolutional neural network (CNN) for failure classification. The approach is validated by comparing the model’s ability to classify failures detected on a real, measured IV curve testing corpus obtained from laboratory and field experiments. Results obtained using a fully synthetic training dataset achieve identical accuracy to those obtained with use of a measured training dataset. When evaluating the measured data’s test split, a 100% accuracy was found both when using simulations or measured data as the training corpus. When evaluating all of the measured data, a 96% accuracy was found when using a fully synthetic training dataset. The use of physics-based modeling results as a training corpus for failure detection and classification has many advantages for implementation as each PV system is configured differently, and it would be nearly impossible to train using labeled measured data.

Although unique expected energy models can be generated for a given photovoltaic (PV) site, a standardized model is also needed to facilitate performance comparisons across fleets. Current standardized expected energy models for PV work well with sparse data, but they have demonstrated significant over-estimations, which impacts accurate diagnoses of field operations and maintenance issues. This research addresses this issue by using machine learning to develop a data-driven expected energy model that can more accurately generate inferences for energy production of PV systems. Irradiance and system capacity information was used from 172 sites across the United States to train a series of models using Lasso linear regression. The trained models generally perform better than the commonly used expected energy model from international standard (IEC 61724-1), with the two highest performing models ranging in model complexity from a third-order polynomial with 10 parameters (Radj2 = 0.994) to a simpler, second-order polynomial with 4 parameters (Radj2=0.993), the latter of which is subject to further evaluation. Subsequently, the trained models provide a more robust basis for identifying potential energy anomalies for operations and maintenance activities as well as informing planning-related financial assessments. We conclude with directions for future research, such as using splines to improve model continuity and better capture systems with low (≤1000 kW DC) capacity.

Understanding the fate of heterogenous herbicide resistant weed populations in response to management practices can help towards overcoming the resistance issues. We selected one pair of susceptible (S) and resistant (R) phenotypes (2B21-R vs 2B21-S and 2B37-R vs 2B37-S) separately from two glyphosate resistant heterogeneous populations (2B21 and 2B37) of Echinochloa colona and their fate and adaptive plasticity were evaluated after glyphosate application. Our study revealed the glyphosate concentration required to cause a 50% plant mortality (LD 50 ) was 1187, 200, 3064, and 192 g a. e. ha −1 for the four phenotypes 2B21-R, 2B21-S, 2B37-R, and 2B37-S respectively. Both S phenotypes accumulated more biomass than the R phenotypes at the lower application rates (34 and 67.5 g a. e. ha −1 ) of glyphosate. However, the R phenotypes generally produced more biomass at rates of glyphosate higher than 100 g a. e. ha −1 throughout the growth period. Plants from the R phenotypes of 2B21 and 2B37 generated 32% and 38% fewer spikes plant −1 than their respective S counterparts in the absence of glyphosate respectively. The spike and seed numbers plant -1 significantly higher in R than S phenotypes at increased rates of glyphosate and these relationships were significant. Our research suggests that glyphosate-resistant E. colona plants will be less fit than susceptible plants (from the same population) in the absence of glyphosate. But in the presence of glyphosate, the R plants may eventually dominate in the field. The use of glyphosate is widespread in field, would favour the selection towards resistant individuals.

An understanding of phenology and reproductive plasticity of a weed species can provide valuable information to manage it precisely. This study evaluated the phenotypic plasticity of feathertop Rhodes grass (Chloris virgata Sw.) where cohorts of four different populations (two from cropping and two from roadside situations) were initiated in early spring (4 September), late spring (4 November), mid-summer (4 January), and early autumn (4 March) in southern New South Wales (NSW), Australia. The team grew individual plants in the absence of competition under natural conditions. Life-history and fitness-related traits of both phenology and morphology were measured, and dry biomass of vegetative and reproductive parts were determined at physiological maturity. Among the four sowing times, the late-spring sowing treatment took the longest time from emergence to the first seed head emergence (70–110 days), while it had the shortest seed maturity period (8–16 days). Length of reproductive and total life period of the four populations differed across the four sowing-time treatments. The plants that emerged in mid-summer had the longest reproductive period (30 days) whereas the early-autumn emerging plants died before the reproductive stage because of the cold temperatures during winter. The mid-summer cohort required slightly longer time (63–85 days) to achieve seed head formation and less time (19–24 days) for seed maturity than those plants that emerged in early or late spring. All the reproductive features were varied by sowing times and population. The number of seed heads (12–15 per plant) and spikelets (12–13 per seed head), as well as the seed head biomass, re-productive biomass allocation pattern, and seed production, generally increased in the mid-summer-emerged cohort. Seed production in the mid-summer (9942 seeds/plant) cohort was 10% and 70% higher than the late spring (8000 seeds/plant) and early spring (3240 seeds/plant) cohorts, respectively. The ratio of reproductive biomass to vegetative biomass increased in the mid-summer sowing times in all populations, and this species displayed true plasticity in reproductive allocation. Additionally, the four populations of feathertop Rhodes grass differed significantly in phenological, vegetative, and reproductive traits, depending on the sowing time. The reproductive fitness of the four populations varied, with the two roadside populations (FELT 04/20 and STURT/16–17) appearing to be better adapted than the two cropping populations (PARK 01/20 and GLEN 03/18). The results from our study could help construct a basic framework for a variety of weed-management tactics to achieve successful control.

Sampling is an important step in the machine learning process because it prioritizes samples that help the model best summarize the important concepts required for the task at hand. The process of determining the best sampling method has been rarely studied in the context of graph neural networks. In this paper, we evaluate multiple sampling methods (i.e., ascending and descending) that sample based off different definitions of centrality (i.e., Voterank, Pagerank, degree) to observe its relation with network topology. We find that no sampling method is superior across all network topologies. Additionally, we find situations where ascending sampling provides better classification scores, showing the strength of weak ties. Two strategies are then created to predict the best sampling method, one that observes the homogeneous connectivity of the nodes, and one that observes the network topology. In both methods, we are able to evaluate the best sampling direction consistently.

Weed seeds can be spread by different vectors, and seed dispersal is an important mechanism for the weed to persist. Weed seeds passaging through the digestive tract of a ruminant animal is expected to result in reduced viability. Two separate experiments were conducted to determine the germinability and viability of the mature seeds of feathertop Rhodes grass (Chloris virgata Sw.) after exposure to four treatments, that is, 3 months in silage, 48 h in the rumen of steers, silage plus digestion, and passing through the digestive tract of sheep. Our results showed that three different treatments (silage, digestion, and silage plus digestion) can inhibit 90%–100% of the seed germination of feathertop Rhodes grass. Both silage and digestion reduced seed viability by 65%–90%, depending on the population. Silage followed by digestion reduced viability by 80%–97%. The sheep feeding study showed that total viable seeds from the daily recovery of feces for 12 consecutive days after ingestion was only 0.084% and 0.022% in the 2020 and 2021 experiments, respectively. In comparison with the untreated control, the seed viability of feathertop Rhodes grass was reduced by more than 99.9% after feeding through sheep, indicating that the spreading of feathertop Rhodes grass seeds via sheep feces is minimal. These results indicate that silage, digestion, silage followed by digestion, and the ingestion of mature seeds are effective non-chemical weed management options for an integrated weed management package for feathertop Rhodes grass.

Digital cameras and mobile phones have given people around the world the ability to take a large number of photos and store them on their computers. As these images serve the purpose of storing memories and bringing them to mind in the potentially far future, it is important to also store the impressions a user may have from them. Annotating these images can be a laborious process and the work here presents an application design and functioning implementation, which is openly available now, to ease the effort of this task. It also draws inspiration from interface developments of previous applications such as the Nokia Lifeblog and the Facebook user interface. A different mode of sentiment entry is provided where users interact with slider widgets rather than select a emoticon from a set to offer a more fine grained value. Special attention is made to avoid cognitive strain by avoiding nested tool selections.

The forward-forward algorithm presents a new method of training neural networks by updating weights during an inference, performing parameter updates for each layer individually. This immediately reduces memory requirements during training and may lead to many more benefits, like seamless online training. This method relies on a loss (\"goodness\") function that can be evaluated on the activations of each layer, of which can have a varied parameter size, depending on the hyperparamaterization of the network. In the seminal paper, a goodness function was proposed to fill this need; however, if placed in a one-class problem context, one need not pioneer a new loss because these functions can innately handle dynamic network sizes. In this paper, we investigate the performance of deep one-class objective functions when trained in a forward-forward fashion. The code is available at \\url{https://github.com/MichaelHopwood/ForwardForwardOneclass}.

Clipping or saturation in audio signals is a very common problem in signal processing, for which, in the severe case, there is still no satisfactory solution. In such case, there is a tremendous loss of information, and traditional methods fail to appropriately recover the signal. We propose a novel approach for this signal restoration problem based on the framework of Iterative Hard Thresholding. This approach, which enforces the consistency of the reconstructed signal with the clipped observations, shows superior performance in comparison to the state-of-the-art declipping algorithms. This is confirmed on synthetic and on actual high-dimensional audio data processing, both on SNR and on subjective user listening evaluations.