Short-term Accuracy Research Articles

It can be rational to change priors

Abstract According to a widely held norm of rationality, one should not change prior credences without new evidence. An important argument for this norm appeals to accuracy considerations, which say that changing priors does not maximize expected accuracy. This is because accuracy measures are strictly proper, and thus any probabilistically coherent person regards her own priors as uniquely maximizing expected accuracy compared with other priors. This paper attempts to resist the accuracy argument against changing priors. We argue that even if rational epistemic decisions maximize expected accuracy according to strictly proper accuracy measures, it can still be rational to change priors sometimes. The core idea of our argument is that changing priors can be rational if one wants to maximize not just one’s current, short-term accuracy but also future, long-term accuracy. Our argument, if successful, shows that considering long-term accuracy has significant ramifications for the accuracy-first project.

Applying neural networks for short and long-term hourly electricity consumption forecasting in universities: A simultaneous approach for energy management

Predictive models for energy consumption are increasingly relevant for buildings with medium-high demands, such as public buildings. These models enable effective energy management, especially in contexts of budget constraints, maximizing resource efficiency. Thus, university buildings are interesting cases because they meet the above-mentioned conditions, they can be managed by the public or private sector and have high societal impact. This work aims to simultaneously develop two predictive models for hourly electricity demand in a university building: one for the short-term (next hour) and one for the long-term (all hours of the year). The methodology consists in: analyzing the data using descriptive statistics to identify relevant variables; building the neural network models in Python; and evaluating them with performance indicators. The short-term model, which includes the previous hour's energy as independent variable, achieved a relative error of 16.0 %, an R2 = 91.17, RMSE = 0.512 kWh, and MAE = 0.325 kWh. The long-term model, which omits this variable, showed a relative error of 31.7 %, R2 = 74.69, RMSE = 0.864 kWh, and MAE = 0.566 kWh. The study highlights the effectiveness of neural networks in energy forecasting. It also emphasizes the importance of using the previous hour's energy as a variable to obtain short-term accuracy; but also the possibility of building models without it, allowing for long-term approaches. The novelty lies in the simultaneous approach of two predictive horizons in a university building. This approach can contribute by providing accurate information to energy control algorithms that integrate renewable energy, storage, grid power, etc., to optimize the use of resources and improving environmental sustainability.

Dynamic balancing method of power distribution and consumption tasks based on state iterative prediction and resource peaking shifting

In the context of the new power system, the widespread access to massive distributed new energy sources has led to the power distribution and consumption tasks characterized by multiple time scales, wide random distribution, and large demand differences, resulting in unpredictable resource peaks in the tasks computing resource demand curve. In view of this situation, a method of forecasting and dynamic balancing of computing resource demand for power distribution and consumption tasks based on state iteration was proposed: firstly, the tasks computing resource demand model was established under the analysis of the attributes and parameter demand of the power distribution and consumption tasks scenario. Secondly, on the basis of the short-term effectiveness prediction of the traditional Markov model, the first-order difference of the state is used for data training to track the state fluctuation, and the historical state and the predicted state are used for state iteration, so as to avoid the convergence of long-term prediction. Finally, a dynamic balancing model is established according to the time-scale characteristics of cyclical and non-cyclical tasks, and the optimal configuration of load imbalance is achieved through the identification and adjustment of historical data and burst data. The simulation results show that the improved Markov model based on first-order difference and state iteration has the short-term accuracy of the traditional model and the long-term traceability of data fluctuations. The dynamic balancing model can combine the characteristics of historical data and burst data to effectively reduce the imbalance of resource demand, and show good ability to cope with resource imbalance deviation.

Short-Term Performance Evaluation of Real-Time Kinematic Global Navigation Satellite System Receivers in Unmanned Aircraft Systems

HighlightsAn evaluation of relative GNSS receiver accuracy was performed using a UAS-based GNSS accuracy testing system.RTK receivers displayed minimal mean error and consistent standard deviations across deployment strategies and error directions.Non-RTK receivers exhibited significantly greater mean error and variability across deployment strategies, particularly in elevation error.UAS deployment strategy did not have a meaningful impact on RTK receiver positioning performance.Abstract. Global navigation satellite system (GNSS) receivers commonly integrated into small unmanned aircraft systems (UAS) generally function in standard or differential fix configurations, providing horizontal and vertical accuracies of approximately ±5 meters and ±15 meters, respectively. The accuracy of GNSS positioning is an important factor with widespread implications, affecting domains such as precision agriculture, meteorology, and photogrammetry. In the context of atmospheric observations, spatial accuracy plays a critical role, particularly in applications related to barometric pressure and precipitable water vapor. Similarly, UAS-based photogrammetry applications rely on high geospatial precision for tasks including topographic mapping and environmental monitoring. As such, this study aimed to improve our understanding of GNSS positioning accuracy in UAS-based observations. The main objectives included (1) deploying a UAS-based GNSS accuracy testing system and (2) evaluating the static and dynamic short-term accuracy of L1 and L1/L2 GNSS receivers in RTK and non-RTK fix modes. Results indicated significant differences across receiver configurations and deployment strategies. RTK receivers displayed minimal mean error and consistent standard deviations, while non-RTK receivers exhibited greater mean error and variability, especially in elevation. Though the study did not conclusively confirm a consistent reduction in accuracy due to UAS deployment, findings suggest that RTK receivers substantially enhance accuracy by reducing position measurement error by two orders of magnitude (1–6 cm for RTK; 50–315 cm for non-RTK), thereby mitigating measurement variability attributable to timing or deployment strategy. In conclusion, this research contributed insights into GNSS accuracy for UAS-based observations and underscored the importance of considering receiver configurations and deployment strategies for position measurement during atmospheric and photogrammetric observations. Keywords: Global navigation satellite systems (GNSS), Meteorology, Photogrammetry, Positioning accuracy, Precision agriculture, Real-time kinematic (RTK), Unmanned aircraft systems (UAS).

Prognostic Accuracy of Common Mortality Prognostic Scales in Very Old Patients with Intracerebral Haemorrhage

Background Spontaneous intracerebral haemorrhage (SICH) is the most severe form of all stroke types. Stratification of SICH severity is important for group comparisons and treatment decisions. The existing prognostic scores for clinical prediction in SICH have not been specifically validated in the very old (≥75 years). Therefore, we aimed to evaluate the accuracy of different SICH vital prognostic scores in the very old. Purpose To compare the short-term accuracy of three vital prognostic scores: Functional Outcome in Patients with Primary Intracerebral Haemorrhage (FUNC), Modified Emergency Department Intracerebral Haemorrhage (mEDICH) and the Intracerebral Haemorrhage Score (‘ICH score’) in patients aged 75 or older. Methods Comparison of the discriminative performance of three SICH prognostic scores in a consecutive case series of patients ≥75 years. The prognostic discrimination was assessed using the area under the receiver operating characteristic curve (AUROC). Additionally, a binary logistic regression was conducted to determine independent prognostic factors associated with mortality. Results The case-fatality was 40.6%. The AUROC and Younden index for the three scores was as it follows: ‘ICH score’ 0.882 and 0.648; mEDICH 0.867 and 0.571; FUNC 0.802 and 0.519. The main independent risk factors of death were presence of intraventricular extension (OR = 4.000,95% CI= 1.933–8.276), INR value (OR = 2.173, 95% CI = 1.146–4.117), haemorrhage volume (OR = 1.881, 95% CI = 1.029–3.440) and GCS (OR = 0.119, 95% CI = 0.060–0.236) for mEDICH. Haemorrhage volume (OR = 3.020, 95% CI = 1.806–5.050) and GCS (OR = 0.043, 95% CI = 0.013–0.151) for FUNC. Haemorrhage volume (OR = 4.950, 95% CI = 2.249–10.897) and intraventricular haemorrhage (OR = 3.811, 95% CI = 1.833–7.924) for ‘ICH score’. Conclusion The three scores (‘ICH score’, FUNC and mEDICH) showed an excellent capability of discriminating the group of elderly patients at risk of short-term death. Age per se may not be crucial for accurate discrimination of death in the group of elderly. Instead, the inclusion of available physiological markers of fragility would be more scientifically meaningful than age.

Short-term response of macroalgal communities to ocean warming in the Southern Bay of Biscay

Climate change is causing significant shifts in biological communities worldwide, including the degradation of marine communities. Previous research has predicted that southern Bay of Biscay canopy-forming subtidal macroalgal communities will shift into turf-forming Mediterranean-like communities by the end of the century. These predictions were based on a community-environment relationship model that used macroalgal abundance data and IPCC environmental projections. We have tested the short-term accuracy of that model by resampling the same communities and locations four years later and found the short-term predictions to be consistent with the observed communities. Changes in sea surface temperature were positively correlated with changes in the Community Temperature Index, suggesting that macroalgal communities had responded quickly to global warming. The changes over four years were significant, but canopy-forming macroalgae were more resilient in local sites with favourable temperature conditions. Our study demonstrated that updating predictive models with new data has the potential to yield reliable predictions and inform effective conservation strategies.

Optimization of Projected Phase Change Memory for Analog In‐Memory Computing Inference

AbstractPhase change memory (PCM) is one of the most promising candidates for non‐von Neumann based analog in‐memory computing–particularly for inference of previously‐trained deep neural networks (DNN). It is shown that PCM electrical properties can be tuned systematically using a projection liner, which is designed for resistance drift mitigation, in the manufacturable mushroom PCM. A systematic study of the electrical properties‐including resistance values, memory window, resistance drift, read noise, and their impact on the accuracy of large neural networks of various types and with tens of millions of weights is performed. It is sown that the DNN accuracy can be improved by the PCM with liner for both the short term and long term after programming, due to reduced resistance drift and read noise, respectively, despite the trade‐off of reduced memory window. The liner conductance, PCM device characteristics, and network inference accuracy with PCM memory window and reset state conductance is correlated, which allows us to identify the device optimization space to achieve better short term and long term accuracy for large neural networks.

Implicit Kalman filtering method for remaining useful life prediction of rolling bearing with adaptive detection of degradation stage transition point

Remaining useful life (RUL) prediction is a vital task in rolling bearing prognostics and health management (PHM) process. Kalman filtering (KF) is one of the hot spots in the research area of RUL prediction. However, three dispiriting shortcomings in KF methods are still unavoidable, including: (1) difficulty in tracking the unknown time-varying noise information, (2) the subjectivity for setting time to start prediction (TSP), and (3) short-term accuracy of the predicting results based on linear predictors. To improve the capability of KF methods, this work adopts the variational Bayesian technique to adaptively describe noise information and considers linear and nonlinear factors of multi-channel signals to recognize the degradation stage transition point of bearing as TSP. Moreover, this work proposes an implicit Kalman filtering method to predict the RUL. The effectiveness of the proposed method is validated on XJTU-SY and IMS-Rexnord bearing data. Results show that the proposed method can recognize the TSP and improve the long-term accuracy of the prediction result during the accelerated degradation stage.

1030 nm All-Fiber Closed-Loop Fiber Optic Gyroscope with High Sensitivity

Angle random walk (ARW) is a critical noise component of a typical gyroscope alongside with bias instability noise. ARW has dominant effects especially in short-term accuracy. The measurement uncertainty degrades with the deterioration in ARW resulting in the lowered overall gyroscope accuracy. Many inertial navigation applications such as satellite control and gyro-compassing require low ARW. For an interferometric fiber optic gyroscope, lowest detectable rotation is proportional to scale factor of sensor and inversely proportional to optical bandwidth of light source used in the gyroscope. In this study, using a novel pump laser control closed-loop method, gyroscope performance is able to significantly enhance including signal-to-noise ratio (SNR). Fiber optic gyroscope includes an Yb-doped amplified spontaneous emission (ASE) source with broad emission spectra of 15 nm bandwidth used as light source in order to improve gyroscope sensitivity. The final ARW performance is about 0.008°/√hr for a fiber coil of 150 m length.

Short-Term Prediction of Wind Power Considering the Fusion of Multiple Spatial and Temporal Correlation Features

As the wind power penetration increases, the short-term prediction accuracy of wind power is of great importance for the safe and cost-effective operation of the power grid in which the wind power is integrated. Traditional wind farm power prediction uses numerical weather prediction (NWP) information as an important input but does not consider the correlation characteristics of NWP information from different wind farms. In this study, a convolutional neural network–long short-term memory based short-term prediction model for wind farm clusters is proposed. Additionally, a feature map is established for multiposition NWP information, the spatial correlation of NWP information from different wind farms is fully explored, and the feature map is trained using the spatiotemporal model to obtain the short-term prediction results of wind farm clusters. Finally, as a case study, the operational data of a wind farm cluster in China are analyzed, and the proposed model outperforms traditional short-term prediction methods in terms of prediction accuracy.

Real-time methods for short and medium-term evapotranspiration forecasting using dynamic crop coefficient and historical threshold

Crop evapotranspiration (ETc) plays a fundamental role in agronomic and water resource management. Accurate forecasting of ETc is a major challenge for agricultural researchers and experts. Based on the measured ETc of the Eddy Covariance system and weather forecast data (1–15 d: short and medium-term) in North China, the real-time short (1–7 d) and medium (8–15 d) term ETc forecast models were developed by coupling with the dynamic crop coefficient and modifying the historical threshold. The results demonstrated that compared with the single crop coefficient model recommended by the Food and Agricultural Organization (FAO-56, M1), the M2 model (a modification of the M1 model developed using the dynamic crop coefficient) accurately forecasted the winter wheat and summer maize ETc, with an increased accuracy of 11%. Moreover, the ETc forecasting accuracy using the M2 model for short and medium-term was over 77%, of which the short-term accuracy was higher (greater than84%). The ETc forecasting accuracy increased with the decrease in the forecast period at different growth stages. Further, the short and medium-term accuracies of M3 model (a modification of the M2 model developed by incorporating the historical threshold) were over 81%, of which the accuracy of the 1 d forecast period was approximately 95%, which was 6% higher than that of the M2 model; the root mean square error and the mean absolute error were reduced by 0.1 mm d−1 and 0.11 mm d−1, respectively. Thus, these results indicated that the M3 model, which was developed by integrating the dynamic crop coefficient and the historical empirical threshold, can predict short and medium-term ETc more accurately.

A novel application of Decision Tree classifier in solar irradiance prediction

With an increase in integration of Solar energy in grid due to their environmental benefit, the need for correct prediction of solar irradiance becomes important. This paper aims towards improving short term accuracy of solar irradiation. Various visualization techniques have been used to show the relationship between various variables. The dataset is being pre-processed and fused with machine learning techniques to get the maximum accuracy of the model. We have used three models namely Logistic Regression (LR), Random Forest (RF) and Decision Tree (DT), in which Decision tree has shown maximum accuracy.

A novel adaptive Kalman filter for Euler-angle-based MEMS IMU/magnetometer attitude estimation

This paper introduces a novel Euler-angle-based adaptive attitude filter for micro-electro-mechanical system inertial measurement unit (IMU)/magnetometer fusion. An intelligent coordinate switch algorithm is proposed to overcome the singularity problem that exists in the Euler-angle-based attitude calculation approach. This method rotates the physical IMU body frame and forms a quasi-computational coordinate to perform the attitude and filter update. In addition, a novel noise estimation theorem based on redundant measurement systems is introduced, proofed and employed to adaptively tune the measurement covariance matrix R in the filter. This algorithm uses gyroscope short-term accuracy to evaluate the attitude performance derived from different sensors and optimally assigns weight in the filter update. Simulated and practical experiments are carried out to test the validation of the proposed adaptive attitude filter. The results achieved demonstrate that this approach can provide a promising orientation solution.

Tracking COVID-19 using online search

Previous research has demonstrated that various properties of infectious diseases can be inferred from online search behaviour. In this work we use time series of online search query frequencies to gain insights about the prevalence of COVID-19 in multiple countries. We first develop unsupervised modelling techniques based on associated symptom categories identified by the United Kingdom’s National Health Service and Public Health England. We then attempt to minimise an expected bias in these signals caused by public interest—as opposed to infections—using the proportion of news media coverage devoted to COVID-19 as a proxy indicator. Our analysis indicates that models based on online searches precede the reported confirmed cases and deaths by 16.7 (10.2–23.2) and 22.1 (17.4–26.9) days, respectively. We also investigate transfer learning techniques for mapping supervised models from countries where the spread of the disease has progressed extensively to countries that are in earlier phases of their respective epidemic curves. Furthermore, we compare time series of online search activity against confirmed COVID-19 cases or deaths jointly across multiple countries, uncovering interesting querying patterns, including the finding that rarer symptoms are better predictors than common ones. Finally, we show that web searches improve the short-term forecasting accuracy of autoregressive models for COVID-19 deaths. Our work provides evidence that online search data can be used to develop complementary public health surveillance methods to help inform the COVID-19 response in conjunction with more established approaches.

Toward the Optimization of the Region-based P00 Speller

Technology has tremendously contributed to improving communication and facilitating daily activities. Brain-Computer Interface (BCI) study particularly emerged from the need to serve people with disabilities such as Amyotrophic Lateral Sclerosis (ALS). However, with the advancements in cost-effective electronics and computer interface equipment, the BCI study is flourishing, and the exploration of BCI applications for people without disabilities, to enhance normal functioning, is increasing. Particularly, the P300-based spellers are among the most promising applications of the BCI technology. In this context, the region-based paradigm for P300 BCI spellers was introduced in an effort to reduce the crowding effect and adjacency problem that might affect the detection of P300 peak. This study extends this line of research by investigating the effect, in terms of accuracy and usability, of the letters’ distribution among the speller’s regions. For this purpose, a clustering algorithm is proposed, and two region-based layouts were generated by redistributing the letters based on their dissimilarity or their similarity. A pilot usability evaluation was also conducted in order to assess the usability of the different layouts in terms of effectiveness, efficiency, and satisfaction. The results indicate that the distribution of the letters has an effect on the classification accuracy as well as the user experience. Particularly, when considering short-term accuracy and cognitive effort, the original region-based layout outperforms other layouts.

Short-term Forecasting Ability of Hybrid Models for BRIC Currencies

This article proposes a new framework to improve short-term forecasting accuracy of exchange rates of BRIC nations, that is, Brazil (USD/BRL), Russia (USD/RUB), India (USD/INR) and China (USD/CNY). The study employs three methodologies for a 42-day forecast: hybrid models based on least square support vector machine, residual hybrid model and automatic hybrid model forecasting using R software. The results show that the proposed residual hybrid model framework, including autoregressive integrated moving average-artificial neural network (ARIMA–ANN)-TBATS, outperformed other models with Brazil and China return series reflecting the best accuracy in ANN model and India and Russia demonstrating the best accuracy in trigonometric seasonal, box-cox transformation, ARIMA residuals, trend and seasonality (TBATS) model. Further, the results indicate that Brazil and China return series follow a non-linear pattern, while India and Russia follow a non-linear complex seasonal pattern. The highest level of forecast accuracy has been observed in China followed by Brazil, India and Russia.

A Novel Neural Network-Based SINS/DVL Integrated Navigation Approach to Deal with DVL Malfunction for Underwater Vehicles

A navigation grade Strapdown Inertial Navigation System (SINS) combined with a Doppler Velocity Log (DVL) is widely used for autonomous navigation of underwater vehicles. Whether the DVL is able to provide continuous velocity measurements is of crucial importance to the integrated navigation precision. Considering that the DVL may fail during the missions, a novel neural network-based SINS/DVL integrated navigation approach is proposed. The nonlinear autoregressive exogenous (NARX) neural network, which is able to provide reliable predictions, is employed. While the DVL is available, the neural network is trained by the body frame velocity and its increment from the SINS and the DVL measurements. Once the DVL fails, the well trained network is able to forecast the velocity which can be used for the subsequent navigation. From the experimental results, it is clearly shown that the neural network is able to provide reliable velocity predictions for about 200 s–300 s during DVL malfunction and hence maintain the short-term accuracy of the integrated navigation.

A multiobjective single bus corridor scheduling using machine learning-based predictive models

Many real-life optimisation problems, including those in production and logistics, have uncertainties that pose considerable challenges for practitioners. In spite of considerable efforts, the current methods are still not satisfactory. This is primarily caused by a lack of effective methods to deal with various uncertainties. Existing literature comes from two isolated research communities, namely the operations research community and the machine learning community. In the operations research community, uncertainties are often modelled and solved through techniques like stochastic programming or robust optimisation, which are often criticised for their over conservativeness. In the machine learning community, the problem is formulated as a dynamic control problem and solved through techniques like supervised learning and/or reinforcement learning, which could suffer from being myopic and unstable. In this paper, we aim to fill this research gap and develop a novel framework that takes advantages of both short-term accuracy from mathematical models and high-quality future forecasts from machine learning modules. We demonstrate the practicality and feasibility of our approach for a real-life bus scheduling problem and two controlled bus scheduling instances that are generated artificially. To our knowledge, the proposed framework represents the first multi-objective bus-headway-optimisation method for non-timetabled bus schedule with major practical constraints being considered. The advantages of our proposed methods are also discussed, along with factors that need to be carefully considered for practical applications.

Influence of short-term rainfall forecast error on flood forecast operation: A risk assessment based on Bayesian theory

To study rainfall forecast error’s influence on flood operation, short-term (24 hours ahead) rainfall forecast accuracy in Qian river basin were analyzed, probabilities of event on flood operation risk were estimated based on Bayesian theory, and the dynamic control scheme was discussed. Results show that, accuracy rate and missing report rate decreased while vacancy rate increased with the increase of rainfall forecast magnitude. For flood operation based on current rainfall forecast information of different magnitudes, level I (“no rain”) error nearly has no impact on it, level II (“light rain”) error has a little impact on small flood operation, and level III (“no less than moderate rain”) error has a great impact on large flood. For the dynamic control scheme in flood season, it is not necessary to discharge flood ahead of schedule so that limited water level can be kept at the upper boundary to provide higher potential energy for power generation when forecasting rainfall shows “no rain” or “light rain”, while it should be discharged ahead of schedule so that limited water level can be kept at a lower value to ensure an adequate flood storage capacity when forecasting rainfall shows “no less than moderate rain”.

Impact Assessment of Various IMU Error Sources on the Relative Accuracy of the GNSS/INS Systems

GNSS/INS relative accuracy, which characterizes error variation or instability on different time/spatial scales, is increasingly required in some new applications, such as precise surveying and mapping, especially to represent the details of error variation over the short term for precise relative measurement. The INS, as an important GNSS/INS systems component, can provide high navigation accuracy over the short term, but INS navigation accuracy decreases with time due to IMU error sources, which might have a great influence on the GNSS/INS relative accuracy. This paper focuses on the impact assessment of IMU errors on GNSS/INS relative accuracy and proposes a hybrid simulation scheme based on signal grafting to analyze the impact of each IMU error on the relative accuracy. Allan variance, as one of the evaluation methods of GNSS/INS relative accuracy, is applied. The results show that the white noise of the gyroscope and accelerometer is the major factor affecting the GNSS/INS short-term accuracy, and the backward smoothing solution can further reduce the impact of nonwhite noise. This work can provide a reference for sensor selection and facilitate the use of low-end IMUs in applications with high demand for short-term relative accuracy.