Trends In Time Series Research Articles

A novel industrial process situation awareness model based on multi‐time scale dynamic feature fusion with applications to float glass manufacturing

AbstractModern industrial manufacturing processes often focus on the changes of performance indicators and the evolution of operation conditions. However, the dynamics of variables are vague within processes and even more difficult to be characterised across different processes. The time‐series dynamic characteristics of the different scales among variables, processes, and production cycles develop and transfer with changes in parameters, equipment, and processes. It is difficult to accurately show the quality index and operating condition trend at the same time. To solve these problems, a situation awareness (SA) framework integrating multi‐time scale dynamic features for manufacturing processes is proposed. First, data are condensed and reconstructed through the denoising long short term memory autoencoder to reveal the time series dynamic characteristics to get neighbourhood features which contain features among neighbourhood samples. Second, the neighbourhood features divided into window blocks are fused into the stage features by statistical analysis. Finally, a multi‐scale isometric convolution network is designed to extract the local and global features, which can effectively show the development of dynamic features on a long time scale, and profoundly describe the influence of full cycle features on variables and operating conditions. The proposed model is verified on the real data set of a float glass manufacturing process, and the SA model can well predict the future trend of long time series.

Towards efficient similarity embedded temporal Transformers via extended timeframe analysis

Price prediction remains a crucial aspect of financial market research as it forms the basis for various trading strategies and portfolio management techniques. However, traditional models such as ARIMA are not effective for multi-horizon forecasting, and current deep learning approaches do not take into account the conditional heteroscedasticity of financial market time series. In this work, we introduce the similarity embedded temporal Transformer (SeTT) algorithms, which extend the state-of-the-art temporal Transformer architecture. These algorithms utilise historical trends in financial time series, as well as statistical principles, to enhance forecasting performance. We conducted a thorough analysis of various hyperparameters including learning rate, local window size, and the choice of similarity function in this extension of the study in a bid to get optimal model performance. We also experimented over an extended timeframe, which allowed us to more accurately assess the performance of the models in different market conditions and across different lengths of time. Overall, our results show that SeTT provides improved performance for financial market prediction, as it outperforms both classical financial models and state-of-the-art deep learning methods, across volatile and non-volatile extrapolation periods, with varying effects of historical volatility on the extrapolation. Despite the availability of a substantial amount of data spanning up to 13 years, optimal results were primarily attained through a historical window of 1–3 years for the extrapolation period under examination.

КЛАСТЕРИЗАЦІЯ ЧАСОВИХ РЯДІВ НА ОСНОВІ НАЯВНОСТІ ТРЕНДУ ЗА ДОПОМОГОЮ МАШИННОГО НАВЧАННЯ

The article discusses the use of machine learning methods for clustering time series according to specified criteria, allowing to determine the presence of a trend component. Several clustering methods are used including k-means . The article explores several criteria for detecting trends in short time series, which are often noisy . Experimental results indicate that using these criteria as features demonstrates high clustering accuracy, with both quantitative and qualitative metrics. The results depend on various factors, such as the length of the time series, the type of trend, and the choice of clustering method.

The Impact of Quality Control Methods on Vegetation Monitoring Using MODIS FPAR Time Series

Monitoring vegetation dynamics (VD) is crucial for environmental protection, climate change research, and understanding carbon and water cycles. Remote sensing is an effective method for large-scale and long-term VD monitoring, but it faces challenges due to changing data uncertainties caused by various factors, including observational conditions. Previous studies have demonstrated the significance of implementing proper quality control (QC) of remote sensing data for accurate vegetation monitoring. However, the impact of different QC methods on VD results (magnitude and trend) has not been thoroughly studied. The fraction of absorbed photosynthetically active radiation (FPAR) characterizes the energy absorption capacity of the vegetation canopy and is widely used in VD monitoring. In this study, we investigated the effect of QC methods on vegetation monitoring using a 20-year MODIS FPAR time series. The results showed several important findings. Firstly, we observed that the Mixed-QC (no QC on the algorithm path) generally produced a lower average FPAR during the growing season compared to Main-QC (only using the main algorithm). Additionally, the Mixed-QC FPAR showed a very consistent interannual trend with the Main-QC FPAR over the period 2002–2021 (p < 0.05). Finally, we found that using only the main algorithm for QC generally reduced the trend magnitude (p < 0.1), particularly in forests. These results reveal differences in FPAR values between the two QC methods. However, the interannual FPAR trends demonstrate greater consistency. In conclusion, this study offers a case study on evaluating the influence of different QC methods on VD monitoring. It suggests that while different QC methods may result in different magnitudes of vegetation dynamics, their impact on the time series trends is limited.

Revised Architecture and Two New Super-Earths in the HD 134606 Planetary System

Multiplanet systems exhibit a diversity of architectures that diverge from the solar system and contribute to the topic of exoplanet demographics. Radial velocity (RV) surveys form a crucial component of exoplanet surveys, as their long observational baselines allow for searches for more distant planetary orbits. This work provides a significantly revised architecture for the multiplanet system HD 134606 using both HARPS and UCLES RVs. We confirm the presence of previously reported planets b, c, and d with periods of 12.0897−0.0018+0.0019 , 58.947−0.054+0.056 , and 958.7−5.9+6.3 days and masses of 9.14−0.63+0.65 , 11.0 ± 1, and 44.5 ± 2.9 Earth masses, respectively, with the planet d orbit significantly revised to over double that originally reported. We report two newly detected super-Earths, e and f, with periods of 4.31943−0.00068+0.00075 and 26.9−0.017+0.019 days and masses of 2.31−0.35+0.36 and 5.52−0.73+0.74 Earth masses, respectively. In addition, we identify a linear trend in the RV time series, and the cause of this acceleration is deemed to be a newly detected massive companion with a very long orbital period. HD 134606 now displays four low-mass planets in a compact region near the star, one gas giant further out in the habitable zone, an additional companion in the outer regime, and a low-mass M dwarf stellar companion at large separation, making it an intriguing target for system formation/evolution studies. The location of planet d in the habitable zone proves to be an exciting candidate for future space-based direct imaging missions, whereas continued RV observations of this system are recommended for understanding the nature of the massive, long-period companion.

Variability and time series trend analysis of streamflow in response to climatic and anthropogenic factors over Narmada River basin, India

Variability and time series trend analysis of streamflow in response to climatic and anthropogenic factors over Narmada River basin, India

Spatiotemporal distribution of carbon monoxide dynamics and probabilistic health risk assessment: A comprehensive sensitivity analysis across India's state capitals

Carbon monoxide (CO) is a harmful trace gas produced from the incomplete combustion of fossil fuels. Carbon monoxide (CO) poses a substantial threat to human health. The objectives of this study were to investigate carbon monoxide (CO) annual and seasonal variations with meteorological factors, study the marginal effects of weather factors on CO, analyze the impact of independent variables (meteorological factors) importance on CO distribution, time series trend analysis (Mann-Kendall and Sen's slope, additive decomposition model, and linear trend model), study the spatial distribution of CO in the capital cities and non-carcinogenic health risk in India from 2019 to 2022. The study showed that CO concentrations varied, notably across different years (from 2019 to 2022), months, and seasons. The maximum annual CO (0.91 ± 0.17 mg/m3) concentration was recorded in 2019, while the lowest was (0.67 ± 0.12 mg/m3) in 2022. Seasonally, the winter season had the highest load of CO (ranging from 0.17 to 2.17 mg/m3) in the ambient atmosphere compared to the monsoon (0.08 to 1.38 mg/m3) and summer (0.10 to 1.50 mg/m3) seasons. The analysis of the importance of the independent variables (meteorological factors) on CO indicated that temperature had the most substantial impact on CO variation, followed by wind direction, relative humidity, and wind speed. The trend analysis using different models showed a declining trend in CO concentrations over the study period. The non-carcinogenic health risk HQ values were less than 1, suggesting no notable risk. The Monte Carlo simulation revealed that the 5th and 95th percentile range of HQ risk was 0.02 to 0.07 and the sensitivity analysis of non-carcinogenic risk at the state capital level was conducted, and revealed that Delhi was at a higher risk than Chandigarh, Patna, Gandhinagar, Bengaluru, Mumbai, and others, respectively. The results of this study on CO emissions will be beneficial in creating focused plans and policies to reduce CO pollution in seasonal influences and its effects on ambient air quality and public health in the capital cities of Indian states.

Well googled is half done: Multimodal forecasting of new fashion product sales with image‐based google trends

AbstractNew fashion product sales forecasting is a challenging problem that involves many business dynamics and cannot be solved by classical forecasting approaches. In this paper, we investigate the effectiveness of systematically probing exogenous knowledge in the form of Google Trends time series and combining it with multi‐modal information related to a brand‐new fashion item, in order to effectively forecast its sales despite the lack of past data. In particular, we propose a neural network‐based approach, where an encoder learns a representation of the exogenous time series, while the decoder forecasts the sales based on the Google Trends encoding and the available visual and metadata information. Our model works in a non‐autoregressive manner, avoiding the compounding effect of large first‐step errors. As a second contribution, we present VISUELLE, a publicly available dataset for the task of new fashion product sales forecasting, containing multimodal information for 5,577 real, new products sold between 2016 and 2019 from Nunalie, an Italian fast‐fashion company. The dataset is equipped with images of products, metadata, related sales, and associated Google Trends. We use VISUELLE to compare our approach against state‐of‐the‐art alternatives and several baselines, showing that our neural network‐based approach is the most accurate in terms of both percentage and absolute error. It is worth noting that the addition of exogenous knowledge boosts the forecasting accuracy by 1.5% in terms of Weighted Absolute Percentage Error (WAPE), revealing the importance of exploiting informative external information. The code and dataset are both available online (at https://github.com/HumaticsLAB/GTM-Transformer).

Determining a Piecewise Linear Trend of a Nonstationary Time Series Based on Intelligent Data Analysis. II. Machine Experiments and Solution of the Practical Problem

Determining a Piecewise Linear Trend of a Nonstationary Time Series Based on Intelligent Data Analysis. II. Machine Experiments and Solution of the Practical Problem

Predicting the core thermal hydraulic parameters with a gated recurrent unit model based on the soft attention mechanism

Accurately predicting the thermal hydraulic parameters of a transient reactor core under different working conditions is the first step toward reactor safety. Mass flow rate and temperature are important parameters of core thermal hydraulics, which have often been modeled as time series prediction problems. This study aims to achieve accurate and continuous prediction of core thermal hydraulic parameters under instantaneous conditions, as well as test the feasibility of a newly constructed gated recurrent unit (GRU) model based on the soft attention mechanism for core parameter predictions. Herein, the China Experimental Fast Reactor (CEFR) is used as the research object, and CEFR 1/2 core was taken as subject to carry out continuous predictive analysis of thermal parameters under transient conditions., while the subchannel analysis code named SUBCHANFLOW is used to generate the time series of core thermal-hydraulic parameters. The GRU model is used to predict the mass flow and temperature time series of the core. The results show that compared to the adaptive radial basis function neural network, the GRU network model produces better prediction results. The average relative error for temperature is less than 0.5 % when the step size is 3, and the prediction effect is better within 15 s. The average relative error of mass flow rate is less than 5 % when the step size is 10, and the prediction effect is better in the subsequent 12 s. The GRU model not only shows a higher prediction accuracy, but also captures the trends of the dynamic time series, which is useful for maintaining reactor safety and preventing nuclear power plant accidents. Furthermore, it can provide long-term continuous predictions under transient reactor conditions, which is useful for engineering applications and improving reactor safety.

아파트 매매가격, 전세가격, 월세가격 변동성 결정요인

In this study, the financial time series trend is predicted using Hamilton's Markov regime-switching model to observe the factors of apartment sales, jeonse, and monthly rent prices that change from time to time from a Markov perspective. Inflation since the coronavirus pandemic has pushed the United States interest rate higher, and South Korea is also raising its benchmark interest rate to stabilize the economy. Due to the impact of this increase in the base rate, the mortgage interest rate has also risen, which has a chain effect on the sale price of apartments, the price of jeonse, and the price of monthly rent. Changes in the trend over the long term can be interpreted as exogenous shocks, economic measures, or structural changes. As a result of the Markov model analysis, apartment sales prices, jeonse prices, and monthly rent prices all showed a higher probability of maintaining a recession period, and among them, the recession period for apartment sales price, and jeonse prices was relatively more likely to last longer. In the case of the expected duration period, the duration of the depression phase of apartment sales prices and jeonse prices was similar, while the duration of the depression phase of apartment monthly rent prices was short and highly volatile. As a cause of volatility, loan interest rates were found to be a determinant.

ANN-LSTM-A Water Consumption Prediction Based on Attention Mechanism Enhancement

To reduce the energy consumption of domestic hot water (DHW) production, it is necessary to reasonably select a water supply plan through early predictions of DHW consumption to optimize energy consumption. However, the fluctuations and intermittence of DHW consumption bring great challenges to the prediction of water consumption. In this paper, an ANN-LSTM-A water quantity prediction model based on attention mechanism (AM) enhancement is improved. The model includes an input layer, an AM layer, a hidden layer, and an output layer. Based on the combination of artificial neural network (ANN) and long short-term memory (LSTM) models, an AM is incorporated to address the issue of the traditional ANN model having difficulty capturing the long-term dependencies, such as lags and trends in time series, to improve the accuracy of the DHW consumption prediction. Through comparative experiments, it was found that the root mean square error of the ANN-LSTM-A model was 15.4%, 13.2%, and 13.2% lower than those of the ANN, LSTM, and ANN-LSTM models, respectively. The corresponding mean absolute error was 17.9%, 11.5%, and 8% lower than those of the ANN, LSTM, and ANN-LSTM models, respectively. The results showed that the proposed ANN-LSTM-A model yielded better performances in predicting DHW consumption than the ANN, LSTM, and ANN-LSTM models. This work provides an effective reference for the reasonable selection of the water supply plan and optimization of energy consumption.

Creation and environmental applications of 15-year daily inundation and vegetation maps for Siberia by integrating satellite and meteorological datasets

As a result of climate change, the pan-Arctic region has seen greater temperature increases than other geographical regions on the Earth’s surface. This has led to substantial changes in terrestrial ecosystems and the hydrological cycle, which have affected the distribution of vegetation and the patterns of water flow and accumulation. Various remote sensing techniques, including optical and microwave satellite observations, are useful for monitoring these terrestrial water and vegetation dynamics. In the present study, satellite and reanalysis datasets were used to produce water and vegetation maps with a high temporal resolution (daily) and moderate spatial resolution (500 m) at a continental scale over Siberia in the period 2003–2017. The multiple data sources were integrated by pixel-based machine learning (random forest), which generated a normalized difference water index (NDWI), normalized difference vegetation index (NDVI), and water fraction without any gaps, even for areas where optical data were missing (e.g., cloud cover). For the convenience of users handling the data, an aggregated product is provided, formatted using a 0.1° grid in latitude/longitude projection. When validated using the original optical images, the NDWI and NDVI images showed small systematic biases, with a root mean squared error of approximately 0.1 over the study area. The product was used for both time-series trend analysis of the indices from 2003 to 2017 and phenological feature extraction based on seasonal NDVI patterns. The former analysis was used to identify areas where the NDVI is decreasing and the NDWI is increasing, and hotspots where the NDWI at lakesides and coastal regions is decreasing. The latter analysis, which employed double-sigmoid fitting to assess changes in five phenological parameters (i.e., start and end of spring and fall, and peak NDVI values) at two larch forest sites, highlighted a tendency for recent lengthening of the growing period. Further applications, including model integration and contribution to land cover mapping, will be developed in the future.

TS-Fastformer: Fast Transformer for Time-series Forecasting

Many real-world applications require precise and fast time-series forecasting. Recent trends in time-series forecasting models are shifting from LSTM-based models to Transformer-based models. However, the Transformer-based model has a limited ability to represent sequential relationships in time-series data. In addition, the transformer-based model suffers from slow training and inference speed due to the bottleneck incurred by a deep encoder and step-by-step decoder inference. To address these problems, we propose a time-series forecasting optimized Transformer model, called TS-Fastformer. TS-Fastformer introduces three new optimizations: First, we propose a Sub Window Tokenizer for compressing input in a simple manner. The Sub Window Tokenizer reduces the length of input sequences to mitigate the complexity of self-attention and enables both single and multi-sequence learning. Second, we propose Time-series Pre-trained Encoder to extract effective representations through pre-training. This optimization enables TS-Fastformer to capture both seasonal and trend representations as well as to mitigate bottlenecks of conventional transformer models. Third, we propose the Past Attention Decoder to forecast target by incorporating past long short-term dependency patterns. Furthermore, Past Attention Decoder achieves high performance improvement by removing a trend distribution that changes over a long period. We evaluate the efficiency of our model with extensive experiments using seven real-world datasets and compare our model to six representative time-series forecasting approaches. The results show that the proposed TS-Fastformer reduces MSE by 10.1% compared to state-of-the-art model and demonstrates 21.6% faster training time compared to the existing fastest transformer, respectively.

Long-Term Time Series Trend Analysis of Climatic Variable for Jaisalmer District in Thar Desert, Western Rajasthan, India, Using Statistical Approaches

Long-Term Time Series Trend Analysis of Climatic Variable for Jaisalmer District in Thar Desert, Western Rajasthan, India, Using Statistical Approaches

An Automatic Frequency Control System for Transmission Machinery Based on Back-Propagation Neural Network Algorithm in the Internet of Things Environment

Abstract In order to reduce the error of automatic frequency control of transmission machinery under different disturbances, an automatic frequency control system of transmission machinery based on back-propagation (BP) neural network algorithm is designed. The gradient descent method and Newton method are combined to optimize the BP neural network. Aiming at the periodicity and trend of the actual mechanical vibration frequency time series, the first order backward difference is processed by the difference method, and the autoregressive sequence is derived to obtain the control model. The experimental results show that the system can realize the automatic control of transmission mechanical frequency with small error under different types of interference, and the control effect is ideal.

Spatial and Temporal Variation in Primary Forest Growth in the Northern Daxing’an Mountains Based on Tree-Ring and NDVI Data

We used tree-ring width data of Larix gmelinii and Pinus sylvestris var. mongolica from the northern region of the Daxing’an Mountains, China; normalized difference vegetation index (NDVI) data; and microtopographic information (elevation, slope direction, slope gradient, and topographic location index) to assess spatiotemporal dynamics in the growth of the boreal forest and topographic patterns of forest decline under the background of climate warming. Forest growth trends were determined based on tree growth decline indicators and NDVI time series trends, and topographic patterns of forest decline were analyzed using the C5.0 decision tree model. More climatic information was present in the radial growth of the trees at higher elevations, and P. sylvestris var. mongolica was influenced strongly by climatic factors of the previous year. Since 1759, tree radial growth trends in the study area have experienced two recessions during 1878–1893 and 1935–1943, which were characterized by persistent narrow whorls of tree rings of below-average growth. Changes in NDVI and tree-ring information were similar, and they together indicate a high risk of declining forest growth in the northern Daxing’an Mountains after 2010, especially at higher elevations. The NDVI time series showed that the high temperatures in 2003 negatively affected forest growth in the study area, which was confirmed by the tree-ring data. The decision tree terrain model results had an accuracy of 0.861, and elevation was the most important terrain factor affecting forest decline. The relative importance of elevation, topographic position index, aspect, and slope was 58.41%, 17.70%, 16.81%, and 7.08%, respectively. Classification rule-based decision tree models can be used to quantify the effects of terrain factors on tree growth. This research methodology can aid the management of regional forestry resources and the conservation of forest resources under the background of climate change, which increases the risk of forest decline.

Water resource security evaluation and barrier analysis in Henan Province utilizing the DPSIR framework

Water resource health is one of the necessary conditions for society to achieve sustainable development. Due to the predominant focus of most studies on relatively short time spans, with limited attention to long time series and spatial trends, this study, using various regions of Henan Province as a case study, constructs a water resource security assessment framework based on the DPSIR model encompassing Drivers (D), Pressures (P), State (S), Impact (I), and Response (R) dimensions, with a selection of 19 evaluation indicators. Based on this evaluation index system, the CRITIC-TOPSIS evaluation method is formulated by integrating the CRITIC (Criteria Importance Through Intercriteria Correlation) and TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) models. This method is employed to assess the degree of water resource security in Henan Province from 2013 to 2022. And the Obstruction Degree Model is introduced to diagnose the water resource security levels in various regions of Henan Province. The assessment results indicate that over the past decade, the overall level of water resource security in various regions of Henan Province has shown an increasing trend. Irrigated area, per capita water resources, water consumption per unit of industrial value added, per acre water consumption for agricultural irrigation, the ratio of river length meeting water quality standards, groundwater supply proportion, and sewage treatment rate are identified as the primary obstacles influencing the water resource security levels in different regions of Henan Province. The research outcomes of this study can serve as theoretical foundations to enhance urban water resource security globally, ultimately facilitating sustainable development.

Rainfall trend analysis using the Mann-Kendall test with pyMannKendall: A case study of Jeli, Kelantan

Trend analysis was widely used as a tool to detect changes in climatic and hydrological time series data such as rainfall. Therefore, this study aims to analyse the rainfall trends in Jeli, Kelantan from 2009 to 2020 Simple linear regression was used to impute missing data in all rainfall stations. The rainfall trends in time series were tested using the Mann- Kendall test and Sens’ slope methods. This analysis was analysed using Python with pyMannKendall package. Kg. Jeli exhibited a statistically significant decreasing trend in rainfall, with a Kendall’s Tau of -0.1412 and a Sen’s slope of -0.0272 (p-value = 0.0121). However, no statistically significant rainfall trends were observed in the other three rainfall stations: Ldg. Kuala Balah, Kg. Gemang Bahru, and Kg. Ayer Lanas during the study period. Generally, this study concludes that the flood event was caused by high rainfall intensity.

Hepatitis B time series in Xinjiang, China (2006-2021): change point detection based on the Mann-Kendall-Sneyers test.

Hepatitis B is a major global challenge, but there is a lack of epidemiological research on hepatitis B incidence from a change point perspective. This study aimed to fill this gap by identifying significant change points and trends in hepatitis time series in Xinjiang, China. The datasets were obtained from the Xinjiang Information System for Disease Control and Prevention. The Mann-Kendall-Sneyers (MKS) test was used to detect change points and trend changes on the hepatitis B time series of 14 regions in Xinjiang, and the effectiveness of this method was validated by comparing it with the binary segmentation (BS) and segment regression (SR) methods. Based on the results of change point analysis, the prevention and control policies and measures of hepatitis in Xinjiang were discussed. The results showed that 8 regions (57.1%) with at least one change fell within the 95% confidence interval (CI) in all 14 regions by the MKS test, where five regions (Turpan (TP), Hami (HM), Bayingolin (BG), Kyzylsu Kirgiz (KK), Altai (AT)) were identified at one change point, two change points existed for two regions (Aksu (AK), Hotan (HT)) and three change points was detected in 1 region (Bortala (BT)). Most of the change points occurred at both ends of the sequence. More change points indicated an upward trend in the front half of the sequence, while in the latter half, many change points indicated a downward trend prominently. Finally, in comparing the results of the three change point tests, the MKS test showed a 61.5% agreement (8/13) with the BS and SR.