Short-term Model Research Articles

Research on a short-term forecast of industry load based on PSO-LSTM

Abstract Load prediction is crucial to maintaining a stable power grid and encouraging the use of innovative energy types like scenescape energy. The industrial load of a region accounts for a significant proportion of the total load, but the industrial load is difficult to predict due to its large fluctuations and instability. This study proposes a short-term power load prediction model, merging the Particle Swarm Optimization (PSO) algorithm and Long Short-Term Memory (LSTM) neural network. The PSO algorithm fine-tunes the LSTM network’s hyperparameters, which are then implemented on the LSTM layer. The trained model, using real-world datasets, is then tested for accuracy. Our model’s effectiveness is backed by practical examples.

A variational approach to quantum gated recurrent units

Quantum Recurrent Neural Networks are receiving an increased attention thanks to their enhanced generalization capabilities in time series analysis. However, their performances were bottlenecked by long training times and unscalable architectures. In this paper, we propose a novel Quantum Recurrent Neural Network model based on Quantum Gated Recurrent Units. It uses a learnable Variational Quantum Layer to process temporal data, interspersed with two classical layers to properly match the dimensionality of the input and output vectors. Such an architecture has fewer quantum parameters than existing Quantum Long Short-Term Memory models. Both the quantum networks were evaluated on periodic and real-world time series datasets, together with the classical counterparts. The quantum models exhibited superior performances compared to the classical ones in all the test cases. The Quantum Gated Recurrent Units outperformed the Quantum Long Short-Term Memory network despite having a simpler internal configuration. Moreover, the Quantum Gated Recurrent Units network demonstrated to be about 25% faster during the training and inference procedure over the Quantum Long Short-Term Memory. This improvement in speed comes with one less quantum circuit to be executed, suggesting that our model may offer a more efficient alternative for implementing Quantum Recurrent Neural Networks on both simulated and real quantum hardware.

Business Cycle and Early Warning Indicators for the Economy of Hong Kong– Challenges of Forecasting Work amid the COVID-19 Pandemic

This paper discusses the use of the OECD’s framework to identify early warning indicators for building a Composite Leading Indicator (CLI) and the use of the Vector Autoregressive Model (VAR) for constructing a short-term forecasting model of economic growth of Hong Kong. With the onset of the COVID-19 pandemic, this paper further evaluates the performance of the CLI and forecasting model which were built based on pre-COVID-19 parameters and identify further adjustments to enhance the model performance.

A hybrid approach to improvement of watershed water quality modeling by coupling process-based and deep learning models.

Watershed water quality modeling to predict changing water quality is an essential tool for devising effective management strategies within watersheds. Process-based models (PBMs) are typically used to simulate water quality modeling. In watershed modeling utilizing PBMs, it is crucial to effectively reflect the actual watershed conditions by appropriately setting the model parameters. However, parameter calibration and validation are time-consuming processes with inherent uncertainties. Addressing these challenges, this research aims to address various challenges encountered in the calibration and validation processes of PBMs. To achieve this, the development of a hybrid model, combining uncalibrated PBMs with data-driven models (DDMs) such as deep learning algorithms is proposed. This hybrid model is intended to enhance watershed modeling by integrating the strengths of both PBMs and DDMs. The hybrid model is constructed by coupling an uncalibrated Soil and Water Assessment Tool (SWAT) with a Long Short-Term Memory (LSTM). SWAT, a representative PBM, is constructed using geographical information and 5-year observed data from the Yeongsan River Watershed. The output variables of the uncalibrated SWAT, such as streamflow, suspended solids (SS), total nitrogen (TN), and total phosphorus (TP), as well as observed precipitation for the day and previous day, are used as training data for the deep learning model to predict the TP load. For the comparison, the conventional SWAT model is calibrated and validated to predict the TP load. The results revealed that TP load simulated by the hybrid model predicted the observed TP better than that predicted by the calibrated SWAT model. Also, the hybrid model reflects seasonal variations in the TP load, including peak events. Remarkably, when applied to other sub-basins without specific training, the hybrid model consistently outperformed the calibrated SWAT model. In conclusion, application of the SWAT-LSTM hybrid model could be a useful tool for decreasing uncertainties in model calibration and improving the overall predictive performance in watershed modeling. PRACTITIONER POINTS: We aimed to enhance process-based models for watershed water-quality modeling. The Soil and Water Assessment Tool-Long Short-Term Memory hybrid model's predicted and total phosphorus (TP) matched the observed TP. It exhibited superior predictive performance when applied to other sub-basins. The hybrid model will overcome the constraints of conventional modeling. It will also enable more effective and efficient modeling.

Research on macroeconomic indicators and stock market correlation analysis based on machine learning

The stock market is known as the barometer of the national economy, and macroeconomic factors have an important impact on the volatility of the stock market. Therefore, considering the impact of macroeconomic factors on the sustainability of stock market volatility will help to capture the time-varying characteristics of volatility persistence, so as to significantly improve the estimation and forecasting effect of volatility. In this work, the generalized autoregressive conditional heteroskedasticity-mixing data sampling (GARCH-MIDAS) model is adopted, which combines the advantages of the GARCH model in short-term volatility modeling and the advantages of MIDAS regression in integrating macroeconomic variables of different frequencies. The GARCH model provides an accurate depiction of intraday volatility in financial markets by capturing the dynamic nature of short-term volatility. MIDAS regression introduces long-term macroeconomic factors into volatility modeling by integrating macroeconomic data with different sampling frequencies, thus making up for the shortcomings of traditional measurement methods in frequency matching. By combining these two methods, the GARCH-MIDAS model can more comprehensively reflect the dynamic changes of stock market volatility, considering both the impact of short-term market volatility and the role of long-term macroeconomic factors, thus providing a more accurate and in-depth analytical tool for volatility prediction and risk management. The results show that the GARCH-MIDAS model can significantly improve the accuracy of volatility forecasting, and provide more reliable decision support for investors, policymakers and economists.

Self-attention mechanism to enhance the generalizability of data-driven time-series prediction: A case study of intra-hour power forecasting of urban distributed photovoltaic systems

The emergence of small-scale urban distributed solar generation (DSG) has urged the exploration of site-adaptive forecasting models designed to accurately predict future power outputs for unseen DSGs. In such scenarios, with numerous DSGs spread across utility-scale cities and a lack of historical data, it is not economically viable to use conventional approaches that develop individual models for each DSG. Therefore, this work aims to tackle this real-world challenge by adapting the state-of-the-art, attention-based temporal fusion transformer (TFT) model to 188 real-world operational DSG data, thereby validating the generalizability of self-attention mechanism for multi-step time series forecasting. When adapted to unseen DSGs without training data, the experiment results demonstrate that the proposed solar TFT (STFT) improves by 11.07%, 17.58%, and 22.76% over the persistence model at the 10-, 20-, and 30-minute forecasts, respectively. Even when compared to representative deep-learning models, such as a long short-term memory model specialized in time series forecasting, STFT has demonstrated improved forecast accuracy, achieving 3.34%, 4.18%, and 5.85% enhancements at the 10-, 20-, and 30-minute forecast horizons, respectively. However, the model architecture of STFT is more complex, and the computational cost associated with it is relatively higher compared to other deep learning models. This trade-off between accuracy and computational efficiency should be considered in practical applications. The forecast performance is analyzed in three typical weather conditions, namely, clear, partly cloudy, and overcast. STFT demonstrates advantages in high variability periods, especially during weather transition periods, where reference models experience lagged predictions yielding relatively large errors.

Pressure prediction of water supply network based on 1DCNN-GRU-multi-head attention

ABSTRACT The water supply network is a crucial component of urban construction and plays a pivotal role in the normal operation of the city. The complexity and periodicity of pressure variations in the water supply networks pose significant challenges to traditional prediction models. In this study, we introduce a novel short-term pressure prediction model, termed the 1DCNN-GRU-multi-head attention (CGMA) model, which incorporates a multi-head attention mechanism alongside an integrated network consisting of one-dimensional convolutional neural networks (1DCNNs) and gated recurrent units (GRUs). Initially, the model employs a 1DCNN network to extract features from the pressure data. Subsequently, the extracted features are input into the GRU neural network, leveraging its long-term dependency capabilities to improve prediction accuracy. Then, the attention mechanism is incorporated into the GRU component to highlight key information, enabling the model to focus on more important data features, thereby improving the prediction performance. We implemented this model in a real urban water distribution system equipped with five pressure sensors. The model achieved a mean absolute error of 0.00197 and a root mean square error of 0.00262. When compared with alternative approaches, our method demonstrated superior predictive performance for pressure data, thereby confirming its efficacy in practical applications.

Accuracy and transportability of machine learning models for adolescent suicide prediction with longitudinal clinical records

Machine Learning models trained from real-world data have demonstrated promise in predicting suicide attempts in adolescents. However, their transportability, namely the performance of a model trained on one dataset and applied to different data, is largely unknown, hindering the clinical adoption of these models. Here we developed different machine learning-based suicide prediction models based on real-world data collected in different contexts (inpatient, outpatient, and all encounters) with varying purposes (administrative claims and electronic health records), and compared their cross-data performance. The three datasets used were the All-Payer Claims Database in Connecticut, the Hospital Inpatient Discharge Database in Connecticut, and the Electronic Health Records data provided by the Kansas Health Information Network. We included 285,320 patients among whom we identified 3389 (1.2%) suicide attempters and 66% of the suicide attempters were female. Different machine learning models were evaluated on source datasets where models were trained and then applied to target datasets. More complex models, particularly deep long short-term memory neural network models, did not outperform simpler regularized logistic regression models in terms of both local and transported performance. Transported models exhibited varying performance, showing drops or even improvements compared to their source performance. While they can achieve satisfactory transported performance, they are usually upper-bounded by the best performance of locally developed models, and they can identify additional new cases in target data. Our study uncovers complex transportability patterns and could facilitate the development of suicide prediction models with better performance and generalizability.

Application of Deep Learning for Automatic Identification of Hazardous Materials and Urban Safety Supervision

The rapid process of urbanization and industrial development has raised significant concerns regarding the presence and management of hazardous substances. However, conventional methods employed for identifying hazardous substances and monitoring urban safety often suffer from low efficiency and accuracy. This paper proposes a novel approach that combines deep learning and genetic algorithms, which utilizes the Bidirectional Long Short-Term Memory model to capture temporal features in hazardous substance data and introduces the Attention Mechanism for weighted processing of crucial information, thereby improving recognition capability. Genetic Algorithms are employed to optimize the performance and generalization capacity of the deep learning model. Experimental validation demonstrates that the proposed approach achieves higher accuracy and faster processing speed, effectively enhancing urban safety monitoring. This research holds practical implications for urban safety management and accident prevention, offering an innovative solution to guarantee urban safety.

Advanced State of Charge Estimation Using Deep Neural Network, Gated Recurrent Unit, and Long Short-Term Memory Models for Lithium-Ion Batteries under Aging and Temperature Conditions

Accurate estimation of the state of charge (SoC) of lithium-ion batteries is crucial for battery management systems, particularly in electric vehicle (EV) applications where real-time monitoring ensures safe and robust operation. This study introduces three advanced algorithms to estimate the SoC: deep neural network (DNN), gated recurrent unit (GRU), and long short-term memory (LSTM). The DNN, GRU, and LSTM models are trained and validated using laboratory data from a lithium-ion 18650 battery and simulation data from Matlab/Simulink for a LiCoO2 battery cell. These models are designed to account for varying temperatures during charge/discharge cycles and the effects of battery aging due to cycling. This paper is the first to estimate the SoC by a deep neural network using a variable current profile that provides the SoC curve during both the charge and discharge phases. The DNN model is implemented in Matlab/Simulink, featuring customizable activation functions, multiple hidden layers, and a variable number of neurons per layer, thus providing flexibility and robustness in the SoC estimation. This approach uniquely integrates temperature and aging effects into the input features, setting it apart from existing methodologies that typically focus only on voltage, current, and temperature. The performance of the DNN model is benchmarked against the GRU and LSTM models, demonstrating superior accuracy with a maximum error of less than 2.5%. This study highlights the effectiveness of the DNN algorithm in providing a reliable SoC estimation under diverse operating conditions, showcasing its potential for enhancing battery management in EV applications.

Perceptions of young adult clients treated with time-limited art therapy

ABSTRACT Background This study examined the perceptions of young adult clients undergoing Time-Limited Art Therapy (TLAT) inspired by the Time-Limited Psychotherapy model (Mann, 1973). Twenty-one clients aged 23-35 dealing with typical young adulthood challenges participated in the TLAT intervention. Aims Examine clients' perceptions of the TLAT, how art helped focus on their central issue in therapy, the function of art in therapy, therapy outcomes, and changes in self-image a topic that has yet to be explored in the art therapy literature. Methods Therapy consisted of 26 sessions at a dedicated research clinic, with the termination date announced during the first session. The TLAT consisted of four intake sessions integrating structured art-based interventions, after which the central issue for therapy was formulated. This was followed by 22 sessions during which artmaking arose spontaneously. Semi structured interviews were conducted with all clients at the start and end of therapy. The data analysis adhered to Consensual Qualitative Research (CQR) principles. Results The CQR analysis revealed seven domains: clients' expectations when coming to TLAT, clients' perceptions of the time dimension in therapy, the therapeutic relationship, the significance of the central issue in therapy, the role of art in processing contents in therapy, the termination phase of therapy and the significance of therapy. Conclusions Artmaking in TLAT helps draw attention to the central issue, foster the therapeutic relationship, and facilitate self-exploration. These results indicate the value of short-term individual models in art therapy and the advantages of incorporating art into brief psychotherapy, especially in cases with time constraints.

Resilient energy management strategy in smart residential buildings considering price attack: An aggregative game perspective

Smart building techniques have drawn considerable attention for the potential to accommodate renewable energy resources and achieve local energy sustainability. However, information privacy, growing number of users and vulnerability of communication channels have led to energy management and cyber-security issues. In this paper, a cyber-attack resilient energy management system for smart residential buildings is proposed. An aggregative game is formulated to describe the non-cooperative interactions among smart buildings that are coupled through dynamic electricity pricing. The existence of the variational equilibrium is analyzed theoretically. Moreover, a semi-decentralized algorithm is employed to attain the equilibrium solution without disclosing private information of individual buildings. To detect potential price attacks, a cyber-attack detection mechanism is proposed by comparing the predicted and actual electricity costs and exchanged power based on the long short-term memory network model. Numerical simulations demonstrate the performance of the semi-decentralized algorithm in term of convergence speed and robustness against disturbances. Additionally, the effectiveness of the resilient energy management strategy is investigated.

Advancing building energy efficiency: A deep learning approach to early-stage prediction of residential electric consumption

Enhancing building energy efficiency underscores the critical need for innovative predictive models to mitigate environmental issues from high energy consumption, especially in residential areas with air-conditioning and heating ventilation systems. This study introduces the use of Long Short-Term Memory (LSTM) networks for early prediction of residential electric consumption, representing a significant innovation in the field. Unlike traditional Deep Neural Network (DNN) and Artificial Neural Network (ANN) models, Long Short-Term Memory networks efficiently process time-series data, predicting future energy usage with unmatched accuracy. The Long Short-Term Memory model exhibited superior training efficiency, requiring only 2.69 s for over 500 test cases, outperforming Deep Neural Network and Artificial Neural Network models, which took 5.26 and 3.88 s, respectively. Its predictive accuracy, evidenced by an R-squared value of 0.97, surpasses the 0.95 and 0.92 of Deep Neural Network and Artificial Neural Network models, respectively. This breakthrough enables accurate predictions of annual energy usage before construction starts and aids in identifying energy efficiency improvements early in the design process. Applying Long Short-Term Memory networks in this context marks a substantial advancement in predictive modeling for building energy consumption, equipping architects and engineers with a vital tool for designing energy-efficient buildings from the beginning. The innovation and quantitatively proven effectiveness of the Long Short-Term Memory model highlight its potential to revolutionize early-stage building design strategies, filling a crucial gap in the existing literature.

Prediction of In-Hospital Cardiac Arrest in the Intensive Care Unit: Machine Learning-Based Multimodal Approach.

Early identification of impending in-hospital cardiac arrest (IHCA) improves clinical outcomes but remains elusive for practicing clinicians. We aimed to develop a multimodal machine learning algorithm based on ensemble techniques to predict the occurrence of IHCA. Our model was developed by the Multiparameter Intelligent Monitoring of Intensive Care (MIMIC)-IV database and validated in the Electronic Intensive Care Unit Collaborative Research Database (eICU-CRD). Baseline features consisting of patient demographics, presenting illness, and comorbidities were collected to train a random forest model. Next, vital signs were extracted to train a long short-term memory model. A support vector machine algorithm then stacked the results to form the final prediction model. Of 23,909 patients in the MIMIC-IV database and 10,049 patients in the eICU-CRD database, 452 and 85 patients, respectively, had IHCA. At 13 hours in advance of an IHCA event, our algorithm had already demonstrated an area under the receiver operating characteristic curve of 0.85 (95% CI 0.815-0.885) in the MIMIC-IV database. External validation with the eICU-CRD and National Taiwan University Hospital databases also presented satisfactory results, showing area under the receiver operating characteristic curve values of 0.81 (95% CI 0.763-0.851) and 0.945 (95% CI 0.934-0.956), respectively. Using only vital signs and information available in the electronic medical record, our model demonstrates it is possible to detect a trajectory of clinical deterioration up to 13 hours in advance. This predictive tool, which has undergone external validation, could forewarn and help clinicians identify patients in need of assessment to improve their overall prognosis.

Short-term load distribution model for cascade giant hydropower stations with complex hydraulic and electrical connections

The cascade giant hydropower station system (CGHSS) has been widely applied in countries with abundant water resources. However, the dense hydropower stations and power grids bring complex hydraulic and power connections, including the backwater effect, multiple hydroplants sharing one reservoir and one-reservoir multiple dispatch power grid, which leads to modelling difficulties and inefficient model solving. To address these challenges, a short-term load distribution (SLD) model considering complicated hydraulic and electricity constraints is developed. First, the discharge of each hydroplant and the downstream forebay water level are used as inputs to construct a multivariate nonlinear function, and the average tailwater levels are obtained. Second, a head constraint considering the shared water level is added to the model to calculate the head of each hydroplant, making the output more accurate. Finally, the nonlinear constraints are linearly reconstructed using the Big-M method, triangulation technique and special ordered sets to reduce the model solution time. The SLD model is applied to the Xiluodu-Xiangjiaba cascade hydropower system in the Jinsha River. The results show that the developed model reduces water consumption by 4.02 %, 4.70 %, 5.79 % and 5.28 % compared to the constant proportional distribution model and the real operation, generating a more executable SLD scheme.

Hybrid deep learning-based identification of microseismic events in TBM tunnelling

For TBM’s safe and efficient tunnelling, the microseismic monitoring technique has been widely applied in rockburst warning. However, useful microseismic events are often mixed with noisy events, which seriously affects warning accuracy. To this end, this study proposes several hybrid deep learning-based microseismic identification approaches in TBM tunnelling, where recurrent neural networks directly treat monitored microseismic events as time series and avoid complex feature pre-extraction, and grey wolf optimization and attention mechanism are embedded to optimize model hyper-parameters and recognize value information. Additionally, the learning curve and early-stopping strategy are also integrated to reduce overfitting and underfitting risks. Results in a real TBM-excavated water conveyance tunnel indicate that the bidirectional long short-term memory network model combined with grey wolf optimization and attention mechanism achieves the greatest identification performance among discussed models and provides an effective support for intelligent identification of microseismic events under TBM tunnelling environments.

Impact of Biotic and Abiotic Factors on Listeria monocytogenes, Salmonella enterica, and Enterohemorrhagic Escherichia coli in Agricultural Soil Extracts.

Outbreaks of Enterohemorrhagic Escherichia coli (EHEC), Salmonella enterica, and Listeria monocytogenes linked to fresh produce consumption pose significant food safety concerns. These pathogens can contaminate pre-harvest produce through various routes, including contaminated water. Soil physicochemical properties and flooding can influence pathogen survival in soils. We investigated survival of EHEC, S. enterica, and L. monocytogenes in soil extracts designed to represent soils with stagnant water. We hypothesized pathogen survival would be influenced by soil extract nutrient levels and the presence of native microbes. A chemical analysis revealed higher levels of total nitrogen, phosphorus, and carbon in high-nutrient soil extracts compared to low-nutrient extracts. Pathogen survival was enhanced in high-nutrient, sterile soil extracts, while the presence of native microbes reduced pathogen numbers. A microbiome analysis showed greater diversity in low-nutrient soil extracts, with distinct microbial compositions between extract types. Our findings highlight the importance of soil nutrient composition and microbial dynamics in influencing pathogen behavior. Given key soil parameters, a long short-term memory model (LSTM) effectively predicted pathogen survival. Integrating these factors can aid in developing predictive models for pathogen persistence in agricultural systems. Overall, our study contributes to understanding the complex interplay in agricultural ecosystems, facilitating informed decision-making for crop production and food safety enhancement.

Full-field temperature prediction in tunnel fires using limited monitored ceiling flow temperature data with transformer-based deep learning models

In practical tunnel scenarios, full-field coverage of sensors is impractical and costly. During a tunnel fire, the available information is constrained and localized, making the prediction of full-field smoke temperature distribution becoming a noteworthy challenge. This study proposes a transformer-based deep learning model to predict full-field smoke temperature distributions during fire incidents in real-time using limited temporal data from the sensors installed in localized regions below the ceiling, considering heat release rate of the fire source is unknown. The results indicate that proposed approach can predict the longitudinal temperature distribution throughout the tunnel with a length of 750 m by leveraging temperature data from limited sensors within a monitoring length of 210 m. It can further predict the vertical temperature profiles, and eventually estimate the full-field temperature distribution within the tunnel. The transformer model achieved R2 of 0.95 and 0.87 for longitudinal and vertical temperature distribution predictions, respectively. Under the influence of the self-attention mechanism, the transformer model has an advantage over the long short-term memory model in capturing global information, enhancing the accuracy of longitudinal temperature distribution predictions by 18.8 %. This study significantly contributes to effective emergency response and rescue strategies during tunnel fire incidents.

Short-Term Photovoltaic System Output Power Prediction Based on Integrated Deep Learning Algorithms in the Clean Energy Sector

Photovoltaic power generation system plays an important role in renewable energy. Therefore, accurately predicting the short-term output power of photovoltaic system has become a key challenge for real-time power grid management. This study focuses on Yingli's green energy photovoltaic system, and uses the convolution neural network and long-term and short-term memory network fusion model (CNN-LSTM) to predict the short-term power. The model integrates CNN's data feature extraction and LSTM's time series prediction ability, showing high accuracy and stability. The experimental results show that CNN-LSTM model has a low mean and variance of prediction error, and the prediction is stable and reliable, and it is consistent in different scenarios. This provides theoretical support for the output power prediction of photovoltaic system based on deep learning.

Multistep-Ahead Prediction of Logging-While-Drilling Resistivity Curves Based on Seismic-Guided Seq2Seq-Long Short-Term Memory

Summary High-temperature and high-pressure reservoirs in complex geological conditions present primary targets and significant challenges in deepsea oil and gas exploration. Limited offshore drilling operations and lack of detailed geological data hinder accurate formation pressure prediction using geoguided and offset imaging, increasing drilling hazards. Logging-while-drilling (LWD) technology provides timely and accurate subsurface information. Resistivity closely correlates with formation pressure and lithology, aiding pressure prediction. Therefore, in this study, we developed an ahead prediction workflow for LWD curves using the resistivity curve (RD) as an example. A seismic-guided sequence-to-sequence framework with the long short-term memory model (Seq2Seq-LSTM) is used to predict the RD curve at a constant depth ahead of the drill bit, utilizing the RD curve of drilled sections and seismic attributes. The network structure incorporates a direct-recursive hybrid multistep prediction strategy based on update threshold control (Dir-Rec-Update), aligning with real-time LWD data acquisition for ahead curve prediction. Using real well data cross-prediction, baseline models such as multilayer perceptron (MLP) and extreme gradient boosting (XGBoost) were compared while also investigating the impact of different configurations on the proposed Seq2Seq-LSTM. The results demonstrate that the method outperforms conventional models, with an average performance across multiple wells under a 5-m update threshold: root mean square error (RMSE) of 0.15, correlation coefficient of 0.88, and coefficient of determination of 0.77. The Seq2Seq-LSTM model and Dir-Rec-Update strategy provide accurate LWD curves ahead of the drill bit, enabling advanced drilling decisions and preventing hazards. Advanced technologies such as empirical wavelet transform (EWT) and feature selection enhance the method’s potential for curve ahead-of-time prediction.