LSTM Prediction Research Articles

Prediction of carbon dioxide levels in the European Alps based on machine learning algorithms

With the rapid development of world industrialization, the continuous increase of global carbon dioxide emissions has led to the gradual deterioration of the ecological environment and the obvious aggravation of the greenhouse effect. In this paper, the carbon dioxide content in the air over the European Alps is taken as the research object, and the RNN and LSTM neural network prediction models are respectively used to compare and predict it. The results show that the fitting effect of LSTM is better than that of RNN, and the fitting effect of the prediction model will also improve with the increase of iteration times and sample size. Since carbon monoxide and methane in the air will cause changes in carbon dioxide content, this paper adds the two factors into the LSTM prediction model as influencing factors, and the goodness-of-fit can reach 0.95, which is higher than the prediction results when there are only one influencing factor or no influencing factor. In order to reduce the running time, XGBoost, LightGBM and random forest algorithms are respectively used in this paper and Bayesian optimization algorithm is used to predict the carbon dioxide content. The results show that the prediction effect is slightly lower than LSTM. Therefore, this paper takes the above three algorithms as the base model. Linear regression experiments are carried out for the meta-model's Stacking fusion algorithm. The goodness-of-fit can reach 0.92, which significantly improves the prediction effect compared with the base model. Finally, the sensitivity analysis of the Stacking fusion model is carried out in this paper. The experimental results show that the model has strong stability.

Thermal error prediction and optimal design of cooling structure for oscillating head housing

The thermal error of machine tool has always been an important reason affecting the accuracy of machine tool. In modern methods to reduce thermal error, thermal error compensation method and thermal error isolation method are two methods with high operability. Thermal error compensation method emphasizes predicting deformation in advance and then compensating in advance. Thermal error isolation method emphasizes the reduction of numerical value in thermal error propagation path. In this paper, the machine tool swing head is taken as the research object, and the above two methods are innovatively combined; Build an experimental platform and a simulation model to ensure the accuracy of the experiment through mutual verification; The POA-LSTM prediction model is established to predict the expansion of the swing head shell structure caused by thermal load accumulation, and compared with the LSTM prediction model. The results show that the prediction accuracy of the POA-LSTM prediction model in Z direction and X direction is 98.89 % and 98.38 %, respectively, which is higher than that of LSTM's 89.33 % and 89.47 %. The experiment for shell structure optimization is designed to reduce the thermal expansion of the pendulum and improve the machining accuracy. As a result, the thermal expansion inhibition effect on the whole pendulum is 26.14 %, and the thermal expansion inhibition effects on the X-direction, Y-direction and Z-direction are 34.93 %, 33.35 % and 30.24 % respectively.

Short-Term Load Forecasting Method Based on Bidirectional Long Short-Term Memory Model with Stochastic Weight Averaging Algorithm

To accommodate the rapid development of the distribution network of China, it is essential to research load forecasting methods with higher accuracy and stronger generalization capabilities in order to optimize distribution system control strategies, ensure the efficient and reliable operation of the power system, and provide a stable power supply to users. In this paper, a short-term load forecasting method is proposed for low-voltage distribution substations based on the bidirectional long short-term memory (BiLSTM) model. First, principal component analysis (PCA) and the fuzzy C-means method based on a genetic algorithm (GA-FCM) are used to extract the main influencing factors and classify different types of user electricity consumption behaviors. Then, the BiLSTM forecasting model utilizing the stochastic weight averaging (SWA) algorithm to enhance generalization capability is constructed. Finally, the load data from a low-voltage distribution substation in China over recent years are selected as a case study. Compared with conventional LSTM and BiLSTM prediction models, the annual electricity load curves for various user types forecasted by the PCA-BiLSTM model are more closely aligned with actual data curves. The proposed BiLSTM forecasting model exhibits higher accuracy and can forecast user electricity consumption data that more accurately reflect real-life usage.

Recurrent neural network using LSTM for prediction of atmospheric pollutants in the State of Veracruz, Mexico

This article shows the use of a recurrent neural network LSTM whose objective was to make a 24-hour ozone forecast with data obtained from the INECC (National Institute of Ecology and Climate Change) and the validated database of the National Information System of Air Quality (SINAICA spanish acronym). The problem of air quality in the world is essential due to health problems. Currently, the techniques used to generate forecasts in monitoring stations are not accurate and do not allow generating alerts to avoid exposure to poor air quality. The state of Veracruz in Mexico has 7 air quality monitoring stations, showing that it is feasible to alternatively have a betterquality pollutant concentration forecasting system. The results obtained on the correlations of variables, although weak, positive or negative, allowed us to recognize their usefulness in the development of training in the RNN. The use of neural networks was demonstrated for the forecast of ozone concentration values in the metropolitan area of Poza Rica, using the LSTM using Python, Tensor Flow and Keras, with a good fit to the normalized pattern with an average RMSE of 0.0053 ppm of the days chosen for testing between the period from June 2019 to July 2020. Future work will consider PM10 and PM2.5.

The Marginal Effect and LSTM Prediction Model under the Chinese Solar Greenhouse Film

The solar greenhouse is a significant agricultural facility in China. It enables the cultivation of crops during periods that do not coincide with the natural growing season, thus alleviating the pressure on the supply of fruits and vegetables during the winter months. The primary rationale behind the necessity of greenhouse cultivation lies in the fact that the temperature conditions conducive to optimal crop growth can be precisely replicated within this controlled environment. However, it is important to acknowledge that a distinct low-temperature area persists under the film during the overwintering period, with the precise delineation of its boundaries and distribution patterns remaining uncertain. In order to investigate the characteristics of the temperature distribution within the marginal region under the solar greenhouse film, experimental studies, CFD simulations, and LSTM prediction models were employed. The results of these studies indicate that, during the overwintering period, a low-temperature region was observed with approximately equal temperatures near the film membrane. The maximum horizontal distance from the south-side bottom corner was 6130 mm, while the minimum height from the ground was 600 mm. The lowest temperature in the low-temperature region was 4 °C, and the maximum observed temperature difference within the same period in different months was 1 °C. Additionally, a region of elevated temperatures was observed under the film. The lowest temperature in this region was 36.7 °C, and the highest temperature point was within the optimal range for crop growth. The CFD numerical simulation results were consistent with the actual observations, and the LSTM prediction model demonstrated high reliability. The findings of this study offer a theoretical foundation for the distribution of high and low temperatures in solar greenhouses. Furthermore, the developed prediction model provides the necessary buffer time for control, thus enhancing the efficiency of greenhouse cultivation.

A Short-Term Vessel Traffic Flow Prediction Based on a DBO-LSTM Model

To facilitate the efficient prediction and intelligent analysis of ship traffic information, a short-term ship traffic flow prediction method based on the dung beetle optimizer (DBO)-optimized long short-term memory networks (LSTM) is proposed. Firstly, according to the characteristics of vessel traffic flow, speed, and density, the traffic flow parameters are extracted from the AIS data; secondly, the DBO-LSTM model is established, and the optimal hyperparameter combinations of the LSTM are found using the DBO algorithm to improve the model prediction accuracy; then, taking the AIS data of a part of the coastal port area in Xiangshan as an example, we compare and analyze the results of the recurrent neural network, temporal convolutional network, LSTM, and DBO-LSTM prediction models; finally, the results are displayed and analyzed by visualization. The experimental results show that each error is reduced in predicting the flow parameter, speed parameter, and density parameter, and the accuracy reaches 95%, 92%, and 95%, respectively. After predicting the three parameters in the next 24 h, the accuracy rate reaches 93%, 91%, and 94%, respectively, compared with the real data, which surpasses the comparison model and achieves better prediction accuracy, verifying the feasibility and reasonableness of the proposed prediction model.

Complex Event Information Mining and Processing for Massive Aerospace Big Data

This paper intends to analyze the existing problems of remote sensing data from the perspectives of space remote sensing information data capacity and data types. Then, a framework for rapidly analyzing and processing space remote sensing information is constructed. Then, LSTM is used to realize the fault diagnosis of remote sensing data continuity, discrete sample mixing and strong correlation of sample variation. LSTM conducts a multimodal analysis of remote-control commands, which is applied to modeling. The multi-stage LSTM prediction model is established and integrated efficiently to improve its adaptive ability in complex space environments. In this way, the anomaly recognition of remote sensing information is realized. Experiments show that the algorithm can improve the anomaly detection rate of remote sensing data. Experiments show that the algorithm is feasible. It can provide reliable data interpretation function for space remote sensing information control system.

Athlete Momentum Visualization Based on Logit Regression Model and LSTM Prediction Model

Athlete Momentum Visualization Based on Logit Regression Model and LSTM Prediction Model

Number of Cyber Attacks Predicted With Deep Learning Based LSTM Model

The increasing number of cyber attacks will result in various damages to the functioning of technological infrastructure. A prediction model for the number of cyber attacks based on the type of attack, handling actions and severity using time-series data has never been done. A deep learning-based LSTM prediction model is proposed to predict the number of cyberattacks in a time series on 3 evaluated data sets MSLE, MSE, MAE, RMSE, and MAPE, and displays the predicted relationships between prediction variables. Cyber attack dataset obtained from kaggle.com. The best prediction model is epoch 20, batch size 16, and neuron 32 with the lowest evaluation value on MSLE of 0.094, MSE of 9.067, MAE of 2.440, RMSE of 3.010, and MAPE of 10.507 (very good model because the value is less than 15) compared other variations. There is a negative correlation for INTRUSION-MALWARE, BLOCKED-IGNORED, IGNORED-LOGGED, and LOW-MEDIUM. The predicted results for the next 12 months will increase starting from the second month at the same time. The resulting predictions can be used as a basis for policy and strategy decisions by stakeholders in dealing with fluctuations in cyber attacks that occur.

Research on trade quantity prediction on ARIMA and Long Short-Term Memory Networks

In the face of the enormous threat that illegal wildlife trade poses to global ecology and biodiversity. In this paper, two prediction models, ARIMA and LSTM, are used for comparative experiments. By analyzing the data of global illegal wildlife trade from 1995 to 2021, two prediction models, ARIMA and LSTM, are established to obtain the prediction results of global illegal wildlife trade from 2022 to 2029 respectively. The experimental results show that both the ARIMA prediction model and the LSTM prediction model show an upward trend in the prediction of the amount of illegal wildlife trade. Especially noteworthy is that the prediction results of the LSTM model show a more obvious upward trend and larger prediction values compared with the ARIMA model. The projections tell us the grim story of the illegal wildlife trade. If we do not take urgent and effective interventions, the scale of the illegal wildlife trade will continue to show a worrying trend and could rise further. This not only poses a serious threat to global biodiversity, but also has potential negative impacts on ecological balance, economic and social stability. These projections can effectively guide efforts to reduce the amount of illegal wildlife trade.

Research and application of commodity demand forecasting algorithm based on improved ARIMA-LSTM

Through thorough investigation and research, it has been determined that commodities possess a certain degree of timeliness, with long production cycles and short storage times. Accurately predicting customer demand can effectively reduce costs. This paper proposes an improved model based on ARIMA-LSTM, incorporating data cross-correlation analysis for effective data classification. By leveraging the unique characteristics of both ARIMA and LSTM prediction time series models, the data is divided into stationary and non-stationary models for accurate forecasting of future demand for goods. After rigorous testing on the test set, our proposed model achieved impressive results with an accuracy rate of 0.855, precision rate of 0.834, and recall rate of 0.854 respectively. Therefore, the model proposed in this paper has a good performance in forecasting the sales volume of general commodities with time series.

Research on Deformation Prediction of VMD-GRU Deep Foundation Pit Based on PSO Optimization Parameters.

As a key guarantee and cornerstone of building quality, the importance of deformation prediction for deep foundation pits cannot be ignored. However, the deformation data of deep foundation pits have the characteristics of nonlinearity and instability, which will increase the difficulty of deformation prediction. In response to this characteristic and the difficulty of traditional deformation prediction methods to excavate the correlation between data of different time spans, the advantages of variational mode decomposition (VMD) in processing non-stationary series and a gated cycle unit (GRU) in processing complex time series data are considered. A predictive model combining particle swarm optimization (PSO), variational mode decomposition, and a gated cyclic unit is proposed. Firstly, the VMD optimized by the PSO algorithm was used to decompose the original data and obtain the Internet Message Format (IMF). Secondly, the GRU model optimized by PSO was used to predict each IMF. Finally, the predicted value of each component was summed with equal weight to obtain the final predicted value. The case study results show that the average absolute errors of the PSO-GRU prediction model on the original sequence, EMD decomposition, and VMD decomposition data are 0.502 mm, 0.462 mm, and 0.127 mm, respectively. Compared with the prediction mean square errors of the LSTM, GRU, and PSO-LSTM prediction models, the PSO-GRU on the PTB0 data of VMD decomposition decreased by 62.76%, 75.99%, and 53.14%, respectively. The PTB04 data decreased by 70%, 85.17%, and 69.36%, respectively. In addition, compared to the PSO-LSTM model, it decreased by 8.57% in terms of the model time. When the prediction step size increased from three stages to five stages, the mean errors of the four prediction models on the original data, EMD decomposed data, and VMD decomposed data increased by 28.17%, 3.44%, and 14.24%, respectively. The data decomposed by VMD are more conducive to model prediction and can effectively improve the accuracy of model prediction. An increase in the prediction step size will reduce the accuracy of the deformation prediction. The PSO-VMD-GRU model constructed has the advantages of reliable accuracy and a wide application range, and can effectively guide the construction of foundation pit engineering.

LSTM RNN-based excitation force prediction for the real-time control of wave energy converters

Wave energy is a type of abundant and dense renewable energy. Wave force prediction is a critical technology that influences power absorption efficiency in the real-time control of wave energy converter (WEC). Could wave elevation be used to predict wave excitation force directly by training artificial neural network? This method results in rapid and suitable prediction for real-time control. A long short-term memory recurrent neural network (LSTM RNN) algorithm is introduced to identify characteristics of wave excitation forces based on wave elevations. In this method, the wave elevations in front of the structure are measured to obtain sufficient time to actuate the control manipulation. A total of 180 regular wave and 12 irregular wave tests are conducted, and the LSTM RNN model is trained based on the experimental results. The performance of the LSTM algorithm is verified. According to the regular cases in the study, the LSTM prediction can identify high-order harmonic loads, and the anti-noise capability of the LSTM algorithm can filter random noises from the measure signals. In the irregular cases, the LSTM RNN algorithm performs effectively to predict the wave force excited on the structure using wave elevations measured by wave probes. The best combinations of the test setting parameters are determined to guide experimental tests and WEC prototypes.

Kernel multi-granularity double-quantitative rough set based on ensemble empirical mode decomposition: Application to stock price trends prediction

As financial markets grow increasingly complex and dynamic, accurately predicting stock price trends becomes crucial for investors and financial analysts. Effectively identifying and selecting the most predictive attributes has become a challenge in stock trends prediction. To address this problem, this study proposes a new attribute reduction model. A rough set theory model is built by simplifying the prediction process and combining it with the long short-term memory network (LSTM) to enhance the accuracy of stock trends prediction. Firstly, the Ensemble Empirical Mode Decomposition (EEMD) is utilized to decompose the stock price data into a multi-granularity information system. Secondly, due to the numerical characteristics of stock data, a kernel function is applied to construct binary relationships. Thirdly, recognizing the noise inherent in stock data, the double-quantitative rough set theory is utilized to improve fault tolerance during the construction of decision attributes' lower and upper approximations. Moreover, calculate the correlation between conditional and decision attributes, and retain highly correlated conditional attributes for prediction. The kernel multi-granularity double-quantitative rough set based on the EEMD (EEMD-KMGDQRS) model proposed identifies the key factors behind stock data. Finally, the efficacy of the proposed model is validated by selecting 356 stocks from diverse industries in the Shanghai and Shenzhen stock markets as experimental samples. The results show that the proposed model improves the generalization of attribute reduction results through a fault tolerance mechanism by combining kernel function with multi-granularity double-quantitative rough set, thereby enhancing the accuracy of stock trends prediction in subsequent LSTM prediction processes.

The Optimization of LSTM Model by Wavelet Transform and Simulated Annealing Algorithm

The main objective of this paper is to study the integration of data processing methods and intelligent algorithms to optimize the results of LSTM prediction models. In this paper, continuous wavelet transform is used to clean and preprocess time series data and improve data quality. Wavelet reconstruction is used to restore the results. At the same time, the simulated annealing algorithm is introduced as an intelligent algorithm to search globally for the best solution to achieve the optimal prediction result. The application of this comprehensive approach can also improve the quality and precision of data analysis in various fields, such as the parameter estimation of pulse signals in physics. The core challenge of the research is to optimize the data prediction results, and for this purpose, a multi-level method of continuous wavelet transform, deep learning model (LSTM) and simulated annealing algorithm is adopted.

Genetic Algorithms Application for Pricing Optimization in Commodity Markets

The perishable nature of vegetable commodities poses challenges for superstores, as reselling them is often unfeasible due to their short freshness period. Reliable market demand analysis is crucial for boosting revenue. This study simplifies the pricing and replenishment decision-making process by making reasonable assumptions about the selling time, wastage rate, and replenishment time for vegetable commodities. A single-objective planning model with the objective of profit maximization was constructed by fitting historical data using the nonparametric method of support vector regression (SVR). The study reveals a specific relationship between sales volume and cost-plus pricing for each category and predicts future cost changes using an LSTM model. Combining these findings, we substitute the relationship between sales volume and pricing as well as the LSTM prediction data into the model, and solve it using genetic algorithms in machine learning to derive the optimal replenishment volume and pricing strategy. Practical results show that the method can provide reasonable pricing and replenishment strategies for vegetable superstores, and after careful accounting, we arrive at an expected profit of RMB 22,703.14. The actual profit of the supermarket was RMB 19,732.89. The method, therefore, increases the profit of the vegetable superstore by 13.08%. By optimizing inventory management and pricing decisions, the superstore can better meet the challenges of vegetable commodities and achieve sustainable development.

Self‐paced learning long short‐term memory based on intelligent optimization for robust wind power prediction

AbstractGiven the unpredictable and intermittent nature of wind energy, precise forecasting of wind power is crucial for ensuring the safe and stable operation of power systems. To reduce the influence of noise data on the robustness of wind power prediction, a wind power prediction method is proposed that leverages an enhanced multi‐objective sand cat swarm algorithm (MO‐SCSO) and a self‐paced long short‐term memory network (spLSTM). First, the actual wind power data is processed into time series as input and output. Then, the progressive advantage of self‐paced learning is used to effectively solve the instability caused by noisy data during long short‐term memory network (LSTM) training. Following this, the improved MO‐SCSO is employed to iteratively optimize the hyperparameters of spLSTM. Ultimately, a combined MO‐SCSO‐spLSTM model is constructed for wind power prediction. This model is validated with the data of onshore wind farms in Austria and offshore wind farms in Denmark. The experimental results show that compared with the traditional LSTM prediction method, the proposed method has better prediction accuracy and robustness. Specifically, in the onshore and offshore wind power prediction experiments, the proposed method reduces the minimum MAE by 5.44% and 4.96%, respectively, and reduces the MAE range by 4.45% and 17.21%, respectively, which could be conducive to the safe and stable operation of power system.

Stock price prediction for Google based on LSTM model with sentiment analysis

Data analytics is increasingly widely used in economic and financial fields, with one of the more important applications being the prediction of stock price changes. However, the prediction of stock price changes is challenging because stock price changes are often uncertain and affected by multiple factors. This study is designed to use the LSTM model to predict stock price changes, and in the construction of the model to consider the psychological and emotional changes of investors, adding a sentiment analysis, combined with the sentiment index obtained from the sentiment analysis and the original stock price data as the input data for the prediction model. During the experiment, a comparison experiment was set up, i.e., only using the basic LSTM prediction model to predict stock price changes and the improved LSTM prediction model with the sentiment index obtained from the added sentiment analysis to predict stock price changes. After the comparison, the prediction results obtained by the LSTM model with the addition of sentiment analysis are more accurate, which on the one hand indicates that the change of investors' psychological sentiment will have an impact on the stock price change, and indicates that the prediction results obtained by the prediction model that considers the change of investors' sentiment are more accurate. The improved LSTM prediction model can help investors to effectively avoid possible risks when investing in stocks and thus gain more profit.

Progression of COVID-19 Cases in Telangana State by using ARIMA, MLP, ELM and LSTM Prediction Models by Retrospective Confirmation

Objective: The importance of this research article is to evaluate efficient model for diagnosing pandemic COVID-19 positive cases in Telangana State, India. Method: Neural Network models (Extreme Learning Machine and Multi-Layer Perception), Deep Learning Neural Network model (Long Short Term Memory-LSTM) and traditional Auto Regressive Integrated Moving Average (ARIMA) models were applied and the data was converted from non-linear to linear (stationarity) for forecasting Covid-19 positive cases. The study of the data covered from 1st. Dec 2020 to 30th May 2021. 80% of train data was taken to fit the models and then 20% of the test data was used to predict the values. The deviation between original test data and predicted data led to an error. Among these error values, the model which had minimum errors was considered as the best of the four models. Findings: LSTM model was proved to be the most efficient model, as a result of the least Root mean square error (RMSE = 71.12) compared to ARIMA (258.20), ELM (553.67) and MLP (641.86) values. Novelty: These forecasting methods succour to predict the Covid-19 positive cases in the forthcoming days. This has been suggested for taking the better preventive steps to control the Covid-19 positive cases. Keywords: COVID­19, ARIMA, LSTM, MLP, ELM Forecasting

Short-term Load Forecasting Based on CEEMDAN-LSTM-AdaBoost

High precision load forecasting is of great significance for effectively allocating resources in energy systems and improving energy utilization efficiency. In order to improve the accuracy of load forecasting, this paper proposes a new load forecasting model based on the adaptive noise complete ensemble empirical mode decomposition (CEEMDAN)-long short-term memory neural network (LSTM)-adaptive boosting algorithm (AdaBoost). Firstly, CEEMDAN is used to decompose the original load sequence to obtain a series of eigenmode components, which can reduce the impact caused by the non-stationary nature of data. Then, these eigenmode components are input into LSTM for prediction, and an integrated learning model by AdaBoost is introduced. A strong predictor is constructed through several weak predictors to increase the prediction accuracy. Finally, the prediction result of each component is superimposed and reconstructed to obtain the final prediction result. Compared with other models, the two evaluation indicators of the proposed prediction model have decreased by 35.52% and 76.61% respectively, indicating the good prediction accuracy and generalization performance of the proposed method.