Temporal Transfer Research Articles

Reinforcement Aided Latent Temporal Feature Transfer Learning: Time Series Prediction With Insufficient Labeled Data for Industrial Chemical Process

Reinforcement Aided Latent Temporal Feature Transfer Learning: Time Series Prediction With Insufficient Labeled Data for Industrial Chemical Process

Vehicle Crashworthiness Performance Prediction Through Fusion of Multiple Data Sources

Abstract This study aims to improve the prediction accuracy of the computer-aided engineering (CAE) model for crashworthiness performance evaluation at speeds beyond those defined by current regulations and public domain testing protocols. One way of achieving this is by integrating data from a few physical crash tests with the CAE data using machine learning models. In this study, two scenarios are investigated: (1) improving CAE model prediction accuracy using test data of a vehicle type that is the same as that of the CAE model; (2) improving CAE model prediction accuracy using test data from two different types of vehicles (e.g., two different sizes of SUVs). In the first scenario, a novel approach is proposed in the displacement domain (deceleration versus displacement) to enable data fusion to help recover the unmodeled physics in the CAE model. A nonlinear spring-mass model is used to simulate rigid-barrier vehicle frontal impact. A Gaussian process regression (GPR) model is then applied in conjunction with a Gaussian mixture model to capture the model bias of the nonlinear spring constant under a dynamic analysis scheme. In the second scenario, we propose a time-domain method (deceleration versus time) based on temporal convolutional network (TCN) and transfer learning. An initial TCN model is first trained by fusing CAE data with physical test data of the first vehicle type based on data augmentation. This data-augmented TCN model is then fine-tuned through transfer learning using CAE and test data of the second vehicle type. It leverages the domain-invariant representations of the two types of vehicles to enhance the CAE model prediction accuracy of the second vehicle type. Case studies are used to validate the proposed approaches and to demonstrate their efficacy in improving the prediction accuracy of the CAE models.

Do Expressway Interchanges Increase Crash Injury Severities? Insights Using Temporal Instability and Unobserved Heterogeneity

Expressway crashes in interchange areas are a critical concern in China, posing significant economic and social challenges. Utilizing three years of crash data from the Beijing–Shanghai Expressway, this study investigates the transferability and heterogeneity of crash characteristics between interchange and noninterchange areas, as well as the temporal shifts in factors influencing injury severity levels. The research employs four series of random parameters logit models to estimate three potential crash injury severity outcomes of severe injury, minor injury, and no injury (based on the most severely injured individual in each crash) and to identify key determinants, encompassing driver, vehicle, roadway, environmental, temporal, traffic, and crash attributes. Likelihood ratio tests and out-of-sample predictions are utilized to assess the temporal stability and transferability of crash area characteristics. Additionally, the marginal effects of various determinants are calculated to understand their influence across different year periods and crash types. Key variables such as overspeed, single-vehicle, AADT (annual average daily traffic volume), Ls min , and other crash type indicators are identified as significant random parameters, demonstrating heterogeneity in means and variances. Notable distinctions are observed between interchange and noninterchange crashes, indicating nontransferability, with most significant indicators revealing temporal instabilities. Furthermore, factors such as multivehicle involvement, buses, and nighttime conditions are identified as risk indicators, notably increasing the likelihood of severe injuries. These insights are invaluable for expressway designers and decision-makers, aiding in understanding the contributing mechanisms of various elements. This study suggests that stricter enforcement measures are crucial to prevent random lane changes, particularly at interchange entrances and exits. Additionally, effective management strategies and enforcement countermeasures should be implemented to mitigate crash injury outcomes in both interchange and noninterchange areas.

Practical guide for retaining correlated climate variables and unthinned samples in species distribution modeling, using random forests

Species distribution models contain bias, or inaccurate predictions, due to predictors and species occurrences. Collinearity of predictor variables is a concern limited to standard error of estimates for data models, although issues remain about identification of important variables and model transferability, whereas thinning of species occurrences is not likely to improve samples without information about species characteristics. Here, I present a case study of selection of correlated climate variables and thinning of species samples for distributions of 80 tree species in North America, modeled with random forests. Robust important variable identification and temporal transferability with correlated climate variables by random forests was displayed by 1) greatest accuracy with all 14 input predictor variables, at 0.99 sensitivity and 0.97 specificity, whether from current or future climate, compared to variable selection approaches to reduce correlation, 2) greater accuracy for two variable models from foremost ranked important variables than least important variables, 3) models with more than two fixed input variables agreed on the four most important variables, 4) models with all 14 input variables from current and future climates identified the same five most important variables, and 5) mapped models of current species observations were similar with use of current or future climates. Unconstrained distributions were apparent in mapped models with only one preselected temperature and precipitation variable. Thinning samples produced no clear benefits to unthinned samples. Following standard guidelines to remove correlated variables before modeling will result in information loss and may lead to incorrect omission of relevant variables, at least with random forests. Predictive modeling is a practical approach for informing variable selection, including retention of correlated climate variables, which create a reliable basis for accurate predictions.

Neural decoding of inferior colliculus multiunit activity for sound category identification with temporal correlation and transfer learning

ABSTRACT Natural sounds are easily perceived and identified by humans and animals. Despite this, the neural transformations that enable sound perception remain largely unknown. It is thought that the temporal characteristics of sounds may be reflected in auditory assembly responses at the inferior colliculus (IC) and which may play an important role in identification of natural sounds. In our study, natural sounds will be predicted from multi-unit activity (MUA) signals collected in the IC. Data is obtained from an international platform publicly accessible. The temporal correlation values of the MUA signals are converted into images. We used two different segment sizes and with a denoising method, we generated four subsets for the classification. Using pre-trained convolutional neural networks (CNNs), features of the images were extracted and the type of heard sound was classified. For this, we applied transfer learning from Alexnet, Googlenet and Squeezenet CNNs. The classifiers support vector machines (SVM), k-nearest neighbour (KNN), Naive Bayes and Ensemble were used. The accuracy, sensitivity, specificity, precision and F1 score were measured as evaluation parameters. By using all the tests and removing the noise, the accuracy improved significantly. These results will allow neuroscientists to make interesting conclusions.

Frequency response of certain temporal and complex transfer functions for the Interfacial 1D temperature wave

Frequency response of certain temporal and complex transfer functions for the Interfacial 1D temperature wave

Temporal transferability of the housing price component of an integrated land use and transportation model

Housing price is a critical component of integrated urban models (IUMs). The housing market in Canada was affected by COVID-19 including significant price hikes. This poses the question: how transferable are our pre-pandemic travel and land use models to the pandemic period? This question is specifically relevant for IUM’s land use modeling components, such as housing market, since these models tend to make longer-range predictions. With this motivation, this study investigates temporal transferability of housing price micromodel of an IUM. Separate latent segmentation-based autoregressive (LSA) models are developed for pre-pandemic and pandemic periods using sales data from 2016 to 2021 for Central Okanagan region. The key feature of LSA models is to accommodate unobserved heterogeneity, spatial autocorrelation, and temporal correlation. The parameter estimation results of the two models suggest significant differences exist between two periods. For example, buyers are likely to spend more to reside in mixed land use areas during pre-pandemic period. During the pandemic, people are willing to spend more for areas with homogeneous land uses such as residential areas. Furthermore, the pre-pandemic model is applied to pandemic context to examine temporal transferability based on statistical and predictive measures. Results reveal significant differences between the two models. Findings provide empirical evidence on the need for re-estimation or re-calibration of the pre-pandemic model parameters prior to applying for the pandemic period. Developed model will be included in an IUM which is expected to enhance the capacity of IUM to test impacts of socioeconomic shocks like the COVID-19 pandemic on housing market.

Evaluating the Transferability of Spectral Variables and Prediction Models for Mapping Forest Aboveground Biomass Using Transfer Learning Methods

Forests, commonly viewed as the Earth’s lungs, play a crucial role in mitigating greenhouse gas emissions, regulating the globe, and maintaining ecological equilibrium. The assessment of aboveground biomass (AGB) serves as a pivotal indicator for evaluating forest quality. By integrating remote sensing images with a small number of ground-measured samples to map, forest AGBs can significantly reduce time and labor costs. Current research mainly focuses on improving the accuracy of mapping forest AGBs, such as integrating multiple-sensors remote sensing data and models. However, due to uncertainties associated with remote sensing images and complexities inherent in forest structures, the accuracy of mapping forest AGBs is constrained by both the quantity and distribution of ground samples available. The development of transfer learning methods can fully utilize ground-based measurement data and enable the application of samples across regions and time. To evaluate the potential of transfer learning methods in mapping forest AGBs, this study conducted a spatial–temporal transfer of spectral variables (SVs) and prediction models (PMs) using a direct-push transfer method, and a new evaluation metric, relative change of R-squared (RCRS), was proposed to assess the transferability of SVs and PMs. The results showed that the transferability of SVs and PMs in the spatial target domain is obviously greater than that in the temporal target domain. Compared to the temporal target domain, the RCRS for transfer SVs in the spatial target domain was lower by 20.89 (oak) and 20.88 (Chinese fir) and for transfer PMs by 24.16 (oak) and 24.79 (Chinese fir). Tree species is also one of the main factors affecting the spatial and temporal transfer of SVs, and it is challenging to transfer SVs between different tree species. The results also show that nonparametric models have better generalization performance, and their transferability is much greater than that of parametric models.

Comparative Evaluation of State-of-the-Art Semantic Segmentation Networks for Long-Term Landslide Map Production.

The production of long-term landslide maps (LAM) holds crucial importance in estimating landslide activity, vegetation disturbance, and regional stability. However, the availability of LAMs remains limited in many regions, despite the application of various machine-learning methods, deep-learning (DL) models, and ensemble strategies in landslide detection. While transfer learning is considered an effective approach to tackle this challenge, there has been limited exploration and comparison of the temporal transferability of state-of-the-art deep-learning models in the context of LAM production, leaving a significant gap in the research. In this study, an extensive series of tests was conducted to evaluate the temporal transferability of typical semantic segmentation models, specifically U-Net, U-Net 3+, and TransU-Net, using a 10-year landslide-inventory dataset located near the epicenter of the Wenchuan earthquake. The experiment results disclose the feasibility and limitations of implementing transfer-learning methods for LAM production, particularly when leveraging the power of U-Net 3+. Furthermore, following an assessment of the effects of varying data volumes, patch sizes, and time intervals, this study recommends appropriate settings for LAM production, emphasizing the balance between efficiency and production performance. The findings from this study can serve as a valuable reference for devising an efficient and reliable strategy for large-scale LAM production in landslide-prone regions.

Remaining Flying Time Prediction of Unmanned Aerial Vehicles Under Different Load Conditions

Unmanned aerial vehicles (UAVs) are forecast to be widely used in the military and civilian domains. The remaining flying time is a critical parameter to monitor during a flight to ensure the safety of electric UAVs (e-UAVs). However, accurate remaining flying time prediction under different load conditions requires a large amount of data and is computationally expensive for online applications. To address these issues, a deep learning approach based on temporal convolutional networks and transfer learning is developed for lithium-ion battery systems for e-UAVs. A temporal convolutional network is used to extract features from monitoring data and predict the remaining flying time of flights under one load condition. A layer transfer strategy is then used to transfer the knowledge learned from one load condition to another load condition. Battery health monitoring data collected from a fixed-wing e-UAV are used to demonstrate the effectiveness of the proposed method. Experimental results show that the proposed temporal convolutional network with the transfer learning strategy can predict the remaining flying time of the e-UAV under two load conditions more efficiently and accurately than a temporal convolutional network without transfer learning.

DTTM: A deep temporal transfer model for ultra-short-term online wind power forecasting

Accurate wind power forecasting (WPF) is vital for grid stability. Most existing studies rely on the combination methods, and the multi-source information (MSI) related to the wind power, but rarely consider the model performances in online prediction scenarios, where an efficient model structure and the timeliness of MSI are worth considering. In this study, a deep temporal transfer model (DTTM) and a synchronous training strategy (STS) are initially introduced. Firstly, temporal synchronous data is mapped to external temporal data (ETD), not only avoiding the asynchronous obtaining time of the MSI, but also making the past and future spatial-temporal information relevant. Secondly, to expand the forecasting model structure for adapting the ETD, a lightweight model named parallel network (PN) is developed as a forecasting cell. Based on deep parallel network (DPN), to fully train ETD for acquiring superior online prediction effect, a highway convolutional neural network (HCNN) is utilized to link the ETD to DPN, mitigating the degradation caused by insufficient information transmission. Through applying DTTM structure adopted STS, the Normalized Mean Absolute Error (NMAE) reduces by up to 74.42 % and at least 38.11 % compared with other models, achieving better online WPF performance.

Mapping integrated crop-livestock systems in Brazil with planetscope time series and deep learning

Accurate mapping of crops with high spatiotemporal resolution plays a critical role in achieving the Sustainable Development Goals (SDGs), especially in the context of integrated crop-livestock systems (ICLS). Stakeholders can make informed decisions and implement targeted strategies to achieve multiple SDGs related to agriculture, rural development, and sustainable livelihoods by understanding the spatial dynamics of these systems. Accurate information on the extent of ICLS derived from multitemporal remote sensing and emerging map techniques such as deep learning can help in the implementation of sustainable agricultural practices. However, far too little attention has been paid to ICLS map accuracy because it may not be at the forefront of research agendas compared to those of other agricultural practices. This paper aims to map ICLS using high spatiotemporal resolution imagery and deep learning neural network classifiers at two different sites located in Brazil. The pipeline involves four interpretation approaches based on the ICLS class: evaluating deep neural network classifiers with different image composition intervals, explaining commission and omission errors, evaluating the temporal transferability of the method, and evaluating the influence of variables. The study area consists of two locations in São Paulo (study site 1, SS1) and Mato Grosso state (study site 2, SS2), Brazil. We derived nine spectral variables from PlanetScope (PS) images and four metrics through object-based image analysis (OBIA) using two time intervals, 10 and 15 days, to generate the image compositions. These input variables were used in three deep neural network classifiers: convolutional neural network in one dimension (Conv1D), long short-term memory (LSTM), and LSTM with a fully convolutional network (LSTM-FCN). Our results showed that mapping dynamic land use such as ICLS is possible by using high-spatiotemporal-resolution imagery and deep neural network classifiers. The 15-day LSTM-FCN classifier returned the highest map accuracies for both sites, with the following class-level accuracies: producer accuracy (PA) = 97.0% and user accuracy (UA) = 97.0% for SS1 and PA = 82.0% and UA = 96.5% for SS2. Meanwhile, we found map uncertainties arising from the diverse crop calendars and spectro-temporal similarities between ICLS and other land use. The best approaches revealed that temporal generalization was suitable for mapping ICLS, but some classifiers could not generalize due to the inherent characteristics of the class. Most variables were considered efficient for predicting ICLS, although spectral indices revealed better functional relationships, while the PS bands had a lower influence on the predictions. The accuracies achieved with the proposed method represent promising opportunities for the sufficiently accurate mapping of ICLS and other complex crop activities.

Behavioral characterization of the cochlear amplifier lesion due to loss of function of stereocilin (STRC) in human subjects

Loss of function of stereocilin (STRC) is the second most common cause of inherited hearing loss. The loss of the stereocilin protein, encoded by the STRC gene, induces the loss of connection between outer hair cells and tectorial membrane. This only affects the outer hair cells (OHCs) function, involving deficits of active cochlear frequency selectivity and amplifier functions despite preservation of normal inner hair cells. Better understanding of cochlear features associated with mutation of STRC will improve our knowledge of normal cochlear function, the pathophysiology of hearing impairment, and potentially enhance hearing aid and cochlear implant signal processing. Nine subjects with homozygous or compound heterozygous loss of function mutations in STRC were included, age 7–24 years. Temporal and spectral modulation perception were measured, characterized by spectral and temporal modulation transfer functions. Speech-in-noise perception was studied with spondee identification in adaptive steady-state noise and AzBio sentences with 0 and -5 dB SNR multitalker babble. Results were compared with normal hearing (NH) and cochlear implant (CI) listeners to place STRC−/− listeners’ hearing capacity in context. Spectral ripple discrimination thresholds in the STRC−/− subjects were poorer than in NH listeners (p < 0.0001) but remained better than for CI listeners (p < 0.0001). Frequency resolution appeared impaired in the STRC−/− group compared to NH listeners but did not reach statistical significance (p = 0.06). Compared to NH listeners, amplitude modulation detection thresholds in the STRC−/− group did not reach significance (p= 0.06) but were better than in CI subjects (p < 0.0001). Temporal resolution in STRC−/− subjects was similar to NH (p = 0.98) but better than in CI listeners (p = 0.04). The spondee reception threshold in the STRC−/− group was worse than NH listeners (p = 0.0008) but better than CI listeners (p = 0.0001). For AzBio sentences, performance at 0 dB SNR was similar between the STRC−/− group and the NH group, 88 % and 97 % respectively. For -5 dB SNR, the STRC−/− performance was significantly poorer than NH, 40 % and 85 % respectively, yet much better than with CI who performed at 54 % at +5 dB SNR in children and 53 % at + 10 dB SNR in adults. To our knowledge, this is the first study of the psychoacoustic performance of human subjects lacking cochlear amplification but with normal inner hair cell function. Our data demonstrate preservation of temporal resolution and a trend to impaired frequency resolution in this group without reaching statistical significance. Speech-in-noise perception compared to NH listeners was impaired as well. All measures were better than those in CI listeners. It remains to be seen if hearing aid modifications, customized for the spectral deficits in STRC−/− listeners can improve speech understanding in noise. Since cochlear implants are also limited by deficient spectral selectivity, STRC−/− hearing may provide an upper bound on what could be obtained with better temporal coding in electrical stimulation.

On Frequency/Time Invariance of Certain Temporal and Complex Transfer Functions for the One-Dimensional Interfacial Monochromatic Neutron Density Wave

Abstract Optimization of dynamical multibeam neutron cancer therapy has recently been shown to be possible via employment of the beam frequencies of neutron waves as a control variable. The concepts of transfer functions (TF), addressed in this paper, can be essential ingredients of such optimization. Accordingly, the paper studies the dynamics of a one-dimensional (1D) monochromatic neutron density wave generated by time modulation of a boundary neutron current. It is demonstrated that a certain temporal transfer function (TTF) of both parabolic (diffusion) and low frequency hyperbolic (P−1 transport) interfacial neutron density wave happens to be frequency noninvariant with a vibrating boundary neutron current. It is proved that, only at high frequencies, both parabolic and hyperbolic interfacial neutron waves turn out to have a fully frequency-invariant and time-invariant temporal transfer function relative to such a vibrating neutron beam at the boundary. The frequency response of an associated complex transfer function is studied and demonstrated to change behavior, from a lag compensator to a fixed gain amplifier, with changing the frequency, neutron absorption and employed theory for neutron diffusion. A highlight of this paper is its illustration that mere continuity of these transfer functions can be a reflection of the correctness of the transport theory employed for modeling the neutron density waves.

Secondary dynamic midface reanimation with gracilis free muscle transfer after failed reconstruction attempt: A 15-year experience

The quantitative outcome of secondary reanimation after a failed primary reconstruction attempt for facial paralysis is rarely reported in the literature. This study aimed to investigate the feasibility of secondary reanimation with gracilis free muscle transfer (GFMT) and whether this outcome is influenced by the primary reconstruction. Twelve patients with previously failed static procedures (static group, n=6), temporal muscle transfer (temporal transfer group, n=2), and GFMT (GFMT group, n=4) were all secondarily reanimated with GFMT. The clinical outcome was graded with the eFACE metric. The objective oral commissure excursion was measured with Emotrics, and the artificial intelligence software FaceReader evaluated the intensity score (IS) of emotional expression. The mean follow-up was 40±27 months. The eFACE metric showed a statistically significant (p<0.05) postoperative improvement in the dynamic and smile scores across all groups. In the GFMT group, oral commissure with smile (75.75±20.43 points), oral commissure excursion while smiling with teeth showing (32.7±4.35mm), and the intensity of happiness emotion while smiling without teeth showing (IS of 0.37±0.23) were significantly lower as compared with the static group postoperatively (98.83±2.86 points, p=0.038; 41.7±4.35mm, p=0.025; IS 0.83±0.16, p=0.01). Our data suggest that secondary dynamic reconstruction with GFMT is feasible should the primary reconstruction fail. The secondary GFMT appears to improve the outcome of primary GFMT; however, the oral commissure excursion while smiling might be lower than that in patients who had static procedures as primary reconstruction.

Growth response of oaks to insect defoliation: Immediate and intermediate perspectives

In this study, we investigate the immediate and short-term consequences of defoliation by the spongy moth Lymantria dispar on secondary growth of oaks (Quercus robur L. and Quercus petraea (Matt.) Liebl.), including the interplay between defoliation and water availability during the defoliation period within a large-scale field experiment in south-eastern Germany. Furthermore, the impact of defoliation on oak secondary growth is explored retrospectively based on tree core measurements.Within the large-scale filed experiment, secondary growth of 880 oaks in 44 oak-dominated stands was monitored using permanent girth tapes over a three-year period following a spongy moth outbreak. Insecticide treatments were applied to half of the plots to obtain trees without defoliation, and canopy cover dynamics were subsequently monitored ground based and from satellites in all plots.We found that moth defoliation significantly reduced oak secondary growth by 10–60% during the outbreak year, with the impact on secondary growth being directly proportional to defoliation intensity. The negative impact of defoliation on secondary growth was aggravated by increased water availability during the outbreak year.In the post-outbreak year, secondary growth of oaks that had endured mild defoliation was no longer different from that of non-defoliated oaks. In contrast, oaks that had experienced substantial defoliation still exhibited a significant 10–30% reduction in secondary growth. Regardless of defoliation intensity, no further reduction in secondary growth was detected two years after the event compared to control trees. Our complementary retrospective analysis, utilizing core samples from oaks previously subjected to a defoliation event under distinct weather conditions, disclosed a strikingly analogous recovery of secondary growth from defoliation instigated by the spongy moth. Thus, validating that our experimental findings possess broad temporal transferability.

EEG-based auditory attention decoding with audiovisual speech for hearing-impaired listeners.

Auditory attention decoding (AAD) was used to determine the attended speaker during an auditory selective attention task. However, the auditory factors modulating AAD remained unclear for hearing-impaired (HI) listeners. In this study, scalp electroencephalogram (EEG) was recorded with an auditory selective attention paradigm, in which HI listeners were instructed to attend one of the two simultaneous speech streams with or without congruent visual input (articulation movements), and at a high or low target-to-masker ratio (TMR). Meanwhile, behavioral hearing tests (i.e. audiogram, speech reception threshold, temporal modulation transfer function) were used to assess listeners' individual auditory abilities. The results showed that both visual input and increasing TMR could significantly enhance the cortical tracking of the attended speech and AAD accuracy. Further analysis revealed that the audiovisual (AV) gain in attended speech cortical tracking was significantly correlated with listeners' auditory amplitude modulation (AM) sensitivity, and the TMR gain in attended speech cortical tracking was significantly correlated with listeners' hearing thresholds. Temporal response function analysis revealed that subjects with higher AM sensitivity demonstrated more AV gain over the right occipitotemporal and bilateral frontocentral scalp electrodes.

Philosophical Reflections on the Art of John Cage's 4 Minutes 33 Seconds

John Cage (1912-1992) was the most direct challenger of traditional music, a composer and musical inventor with considerable achievements in both destruction and reconstruction. He undoubtedly became the representative of the 20th century avant-garde music, and his works brought the people of the world a kind of shock and a kind of thinking. In opposition to traditional European musical art, 4 Minutes and 33 Seconds blurs the boundaries between sound and soundlessness, music creation and music appreciation, creating contemporary musical art that is purely centred on reflection. Because of the peak nature of the presentation of contemporary art, the philosophy of art also extends to the thinking of the entire philosophical community, so is the philosophical aspect of art at this time still called artistic thinking? In fact, it extends to people's thinking about the meaning of the existence of art and what is art. In Zhu Di's book Contemporary Western Philosophy of Art, he proposes that the concept of art is a "floating reference", which has a temporal transfer and presents different entities in different times.

Analyzing the syntax and salience of causal links embedded within semantic links in concept maps: Implications for temporal flow and learning transfer

Including causal links in concept maps enables learners to meaningfully relate concepts to a larger context or problem in terms of how and where concepts apply within the chains of causal events that lead to a given goal or outcome. Given that higher quality maps are produced when students link and sequence events to flow temporally and sequentially in a consistent direction towards a target outcome in a map, it is highly plausible that students can improve learning transfer (the ability to apply concepts to diagnose and solve problems) by including and making more salient the sequences of causal links nestled in the semantic links in concept maps. To lay the groundwork to empirically test this proposition in future research, this study: 1) analyzes 16 concept maps presented in the Proceedings of the 8th Int. Conference on Concept Mapping to codify the diverse and sometimes incongruent syntaxes used to convey causal relationships; 2) examines how the causal link syntaxes combined with semantic links work jointly (or against each other) to create temporal flow; and 3) explore how causal and semantic links can be integrated to increase the saliency and quality of the causal networks connecting concepts to outcomes. A better understanding of how causal links are expressed, integrated, and made more salient in concept maps can reveal ways to help students create concept maps that are more accurate, meaningful, and effective in improving the ability to apply concepts to solve complex problems.

An analytical method for sizing energy storage in microgrid systems to maximize renewable consumption and minimize unused storage capacity

This paper presents a novel analytical method to optimally size energy storage in microgrid systems. The method has fast calculation speeds, calculates the exact optimal, and handles non-linear models. The method first constructs a temporal storage profile of stored energy, based on how storage charges and discharges in response to renewable generation and load demand. The storage is sized according to the largest cumulative charge or discharge in the profile. In essence, the storage profile represents how storage is utilized within a given system, and the method sizes optimal storage to maximize that profile, such that storage utilization is maximized, and unutilized or wasted storage is eliminated. Maximizing storage utilization also maximizes renewable consumption and minimizes load shedding, as storage utilization is the temporal transfer of energy from renewable generation to load demand. The proposed method is extended iteratively to account for storage’s energy limits, power limits, and energy leakage. Two solar–battery case studies demonstrate the method. The first study shows that optimally sized storage does not have wasted capacity due to over-sizing, nor cause energy deficits due to under-sizing. The second case study shows increasing the storage size reduces the marginal increase in energy provided by storage, indicating diminishing returns. The diminishing return thresholds are defined by the largest daily and annual storage designs. The result shows the largest daily design only requires 3% of the annual design’s storage size, but provides 80% of the energy provided by the annual design. The proposed method can be used as a decision support tool for energy analysts, to determine required storage capacity when coupled with known renewable generation and load demand.