Explore the vast range of titles available.

The refined control of heating substations is of great significance for on-demand heating provision and for the efficient operation of district heating systems (DHSs). This paper proposes an integrated control strategy for substations based on the prediction of the water-supply temperature and indoor temperature. Firstly, online sequential extreme learning machine (OS-ELM) is used to predict the water-supply temperature. Then, a linear prediction model is established to predict the indoor temperature. Finally, the integrated regulation strategy is established with the goal of minimizing operational costs, aiming at ensuring heating quality and meeting the limits of the flow rate and of the supply- and return-water temperatures. The heat-saving rate, power-saving rate and indoor-temperature satisfactory rate are introduced to evaluate the regulation effect of the proposed method. The field study results show that the performance index of operation executed with the regulation strategy proposed in this paper is 9.31%, 16.33% and 20.87% higher than that without our energy-saving regulation strategy respectively. The fluctuations in the water-supply pressure and differential pressure of the secondary network are significantly reduced, and the energy-saving effect is obvious.

This paper presents a model for predicting the water temperature of the reservoir incorporating with solar radiation to analyze and evaluate the water temperature of large high-altitude reservoirs in western China. Through mutual information inspection, the model shows that the dependent variable has a good correlation with water temperature, and it is added to the sample feature training model. Then, the measured water temperature data in the reservoir for many years are used to establish the support vector regression (SVR) model, and genetic algorithm (GA) is introduced to optimize the parameters, so as to construct an improved support vector machine (M-GASVR). At the same time, root-mean-square error, mean absolute error, mean absolute percentage error, and Nash–Sutcliffe efficiency coefficient are used as the criteria for evaluating the performance of SVR model, ANN model, GA-SVR model, and M-GASVR model. In addition, the M-GASVR model is used to simulate the water temperature of the reservoir under different working conditions. The results show that ANN model is the worst among the four models, while GA-SVR model is better than SVR model in terms of metric, and M-GASVR model is the best. For non-stationary sequences, the prediction model M-GASVR can well predict the vertical water temperature and water temperature structure in the reservoir area. This study provides useful insights into the prediction of vertical water temperature at different depths of reservoirs.

Sea surface temperature (SST) significantly influences the dynamics of the global climate system, impacting climate change, marine ecosystems, and marine engineering. Traditional SST prediction methods, such as time series and machine learning models, often focus solely on temporal features and neglect spatial distribution patterns. In contrast, current deep learning techniques typically limit predictions to short-term periods. This paper introduces a novel SST prediction model that integrates both temporal and spatial dimensions, employing parallel prediction and a spatio-temporal attention mechanism to enhance accuracy. The model achieves long-term SST forecasting, significantly reduces the parameter count and computational effort, and maintains high prediction precision. Experiments in the El Niño 3.4 region and the East China Sea show that this method outperforms existing deep learning approaches, accurately predicting SST over periods ranging from 7 to 60 days with superior efficiency and accuracy. Overall, this work presents an effective new approach for SST prediction with crucial implications for climate change research, marine ecosystems, and marine engineering.

Sea surface temperature is an important physical parameter in marine research. Accurate prediction of sea surface temperature is important for coping with climate change, marine ecological protection, and marine economic development. In this study, the SST prediction performance of ConvLSTM and ST-ConvLSTM with different input lengths, prediction lengths, and hidden sizes is investigated. The experimental results show that: (1) The input length has an impact on the prediction results of SST, but it does not mean that the longer the input length, the better the prediction performance. ConvLSTM and ST-ConvLSTM have the best prediction performance when the input length is set to 1, and the prediction performance gradually decreases as the input length increases. (2) Prediction length affects SST prediction. As the prediction length increases, the prediction performance gradually decreases. When other parameters are kept constant and only the prediction length is changed, the ConvLSTM gets the best result when the prediction length is set to 2, and the ST-ConvLSTM gets the best result when the prediction length is set to 1. (3) The setting of the hidden size has a great influence on the prediction ability of the sea surface temperature, but the hidden size cannot be set blindly. For ST-ConvLSTM, although the prediction performance of SST is better when the hidden size is set to 128 than when it is set to 64, the consequent computational cost increases by about 50%, and the performance only improves by about 10%.

During machining processes, a large amount of heat is generated due to plastic deformation, in a very small area of the cutting tool. This high temperature strongly influences chip formation mechanisms, tool wear, tool life, and workpiece surface integrity and quality. In this sense, knowing the temperature at various points of tool, chip, and workpiece during machining processes is of utmost importance to effectively optimize cutting parameters, improve machinability and product quality, reduce machining costs, and increase tool life and productivity. This paper presents a review of the various methods for temperature measurement and prediction in machining processes, being the different methods discussed and evaluated regarding its merits and demerits. The most suitable method for a given application depends on several aspects, such as cost, size, shape, accuracy, response time, and temperature range. Lastly, some future perspectives for real-time cutting temperature monitoring in the scope of Industry 4.0 and 5.0 are outlined, as well as being presented a new field of tools capable of measuring and controlling cutting temperature, called smart cutting tools.

The prediction accuracy of hourly air temperature is generally poor because of random changes, long time series, and the nonlinear relationship between temperature and other meteorological elements, such as air pressure, dew point, and wind speed. In this study, two deep-learning methods-a convolutional neural network (CNN) and long short-term memory (LSTM)-are integrated into a network model (CNN-LSTM) for hourly temperature prediction. The CNN reduces the dimensionality of the time-series data, while LSTM captures the long-term memory of the massive temperature time-series data. Training and validation sets are constructed using 60,133 hourly meteorological data (air temperature, dew point, air pressure, wind direction, wind speed, and cloud amount) obtained from January 2000 to October 2020 at the Yinchuan meteorological station in China. Mean absolute error (MAE), mean absolute percentage error (MAPE), and goodness of fit are used to compare the performances of the CNN, LSTM, and CNN-LSTM models. The results show that MAE, MAPE, RMSE, and PBIAS from the CNN-LSTM model for hourly temperature prediction are 0.82, 0.63, 2.05, and 2.18 in the training stage and 1.02, 0.8, 1.97, and −0.08 in the testing stage. Average goodness of fit from the CNN-LSTM model is 0.7258, higher than the CNN (0.5291), and LSTM (0.5949) models. The hourly temperatures predicted by the CNN-LSTM model are highly consistent with the measured values, especially for long time series of hourly temperature data.

PurposeThe transportation of fresh food requires cold chain logistics to maintain a low-temperature environment, which can reduce food waste and ensure product safety. Therefore, temperature control is a major challenge that cold chain logistics face.Design/methodology/approachThis research proposes a prediction model of refrigerated truck temperature and air conditioner status (air speed and air temperature) based on hybrid mayfly algorithm (MA) and extreme learning machine (ELM). To prove the effectiveness of the proposed method, the mayfly algorithm–extreme learning machine (MA-ELM) is compared with the traditional ELM and the ELM optimized by classical biological-inspired algorithms, including the genetic algorithm (GA) and particle swarm optimization (PSO). The assessment is conducted through two experiments, including temperature prediction and air conditioner status prediction, based on a case study.FindingsThe prediction method is evaluated by five evaluation indicators, including the mean relative error (MRE), mean absolute error (MAE), mean squared error (MSE), root mean square error (RMSE) and coefficient of determination (R2). It can be concluded that the biological algorithm, especially the MA, can improve the prediction accuracy. This result clearly proves the effectiveness of the proposed hybrid prediction model in revealing the nonlinear patterns of the cold chain logistics temperature.Research limitations/implicationsThe case study illustrates the effectiveness of the proposed temperature prediction method, which helps to keep the product fresh. Even though the performance of MA is better than GA and PSO, the MA has the disadvantage of premature convergence. In the future, the modified MA can be designed to improve the performance of MA-ELM.Originality/valueIn prior studies, many scholars have conducted related research on the subject of temperature monitoring. However, this monitoring method can only identify temperature deviations that have occurred that harmed fresh food. As a countermeasure, research on the temperature prediction of cold chain logistics that can actively identify temperature changes has become the focus. Once a temperature deviation is predicted, temperature control measures can be taken in time to resolve the risk.

In order to accurately and easily predict the variation of the average temperature of the concrete rectangular hollow pier. Firstly, the temperature distribution of the concrete rectangular hollow pier of Changjiahe Special Bridge was observed for 212 days. Then, based on the observation data, the functional relationship between the average concrete hollow pier temperature and the outside air temperature, and the air temperature inside the hollow pier was studied. Finally, based on this functional relationship, a prediction model for the range of variation of the mean temperature of the hollow pier was given and verified. The results of the study show that: the change of external shade temperature can be regarded as the superposition of different cyclic changes and random changes; the change rule of the average temperature of the concrete hollow pier is the same as that of the air temperature, both presenting day-by-day cyclic and step changes; the linear correlation coefficients between the daily maximum and daily minimum average hollow pier temperature and the daily average air temperature are R = 0.980 and R = 0.973, respectively; the daily average air temperature, the daily average air temperature inside the pier and daily average hollow pier temperature are R = 0.980, R = 0.998; the daily variation of the average hollow pier temperature and the daily variation of the air temperature are approximately linear, with a correlation coefficient of R = 0.899; assuming that the average temperature of the concrete hollow pier is a folding change, based on the above relationship, a method of predicting the time-by-time average temperature of the concrete hollow piers is proposed, as well as two methods of predicting the average temperature of the test piers. Average temperature change range method. Comparing the predicted values with the measured values, it is found that the predicted values are in good agreement with the measured values.

This paper proposes an Informer-based temperature prediction model to leverage data from an automatic weather station (AWS) and a local data assimilation and prediction system (LDAPS), where the Informer as a variant of a Transformer was developed to better deal with time series data. Recently, deep-learning-based temperature prediction models have been proposed, demonstrating successful performances, such as conventional neural network (CNN)-based models, bi-directional long short-term memory (BLSTM)-based models, and a combination of both neural networks, CNN–BLSTM. However, these models have encountered issues due to the lack of time data integration during the training phase, which also lead to the persistence of a long-term dependency problem in the LSTM models. These limitations have culminated in a performance deterioration when the prediction time length was extended. To overcome these issues, the proposed model first incorporates time-periodic information into the learning process by generating time-periodic information and inputting it into the model. Second, the proposed model replaces the LSTM with an Informer as an alternative to mitigating the long-term dependency problem. Third, a series of fusion operations between AWS and LDAPS data are executed to examine the effect of each dataset on the temperature prediction performance. The performance of the proposed temperature prediction model is evaluated via objective measures, including the root-mean-square error (RMSE) and mean absolute error (MAE) over different timeframes, ranging from 6 to 336 h. The experiments showed that the proposed model relatively reduced the average RMSE and MAE by 0.25 °C and 0.203 °C, respectively, compared with the results of the CNN–BLSTM-based model.

This research conducts a comparative analysis of nine Machine Learning (ML) models for temperature and humidity prediction in Photovoltaic (PV) environments. Using a dataset of 5,000 samples (80% for training, 20% for testing), the models—Support Vector Regression (SVR), Lasso Regression, Ridge Regression (RR), Linear Regression (LR), AdaBoost, Gradient Boosting (GB), Decision Tree (DT), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost)—were evaluated based on Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the Coefficient of Determination (R²). For temperature prediction, XGBoost demonstrated the best performance, achieving the lowest MAE of 1.544, the lowest RMSE of 1.242, and the highest R² of 0.947, indicating strong predictive accuracy. Conversely, SVR had the weakest performance with an MAE of 4.558 and an R² of 0.674. Similarly, for humidity prediction, XGBoost outperformed other models, achieving an MAE of 3.550, RMSE of 1.884, and R² of 0.744, while SVR exhibited the lowest predictive power with an R² of 0.253. This comprehensive study serves as a benchmark for the application of ML models to environmental prediction in PV systems, a research area that is relatively important. Notably, the results underscore the performance advantage of ensemble-based approaches, especially for XGBoost and RF compared to simpler, linear-based methods such as LR and SVR, when it comes to well-dispersed environmental interactions. The proposed machine-learning based power generation analysis approach shows significant improvements in predictive analytics capabilities for renewable energy systems, as well as a means for real-time monitoring and maintenance practices to improve PV performance and reliability.