To improve energy efficiency, extensive studies have focused on the cutting parameters optimization in the machining process. Actually, non-cutting activities (NCA) occur frequently during machining and this is a promising way to save energy through optimizing NCA without changing the cutting parameters. However, it is difficult for the existing methods to accurately determine and reduce the energy wastes (EW) in NCA. To fill this gap, a novel Therblig-embedded Value Stream Mapping (TVSM) method is proposed to improve the energy transparency and clearly show and reduce the EW in NCA. The Future-State-Map (FSM) of TVSM can be built by minimizing non-cutting activities and Therbligs. By implementing the FSM, time and energy efficiencies can be improved without decreasing the machining quality, which is consistent with the goal of lean energy machining. The method is validated by a machining case study, the results show that the total energy is reduced by 7.65%, and the time efficiency of the value-added activities is improved by 8.12% , and the energy efficiency of value-added activities and Therbligs are raised by 4.95% and 1.58%, respectively. This approach can be applied to reduce the EW of NCA, to support designers to design high energy efficiency machining processes during process planning.
The wide use of machining processes has imposed a large pressure on environment due to energy consumption and related carbon emissions. The total power required in machining include power consumed by the machine before it starts cutting and power consumed to remove material from workpiece. Accurate prediction of energy consumption in machining is the basis for energy reduction. This paper investigates the prediction accuracy of the material removal power in turning processes, which could vary a lot due to different methods used for prediction. Three methods, namely the specific energy based method, cutting force based method and exponential function based method are considered together with model coefficients obtained from literatures and experiments. The methods have been applied to a cylindrical turning of three types of workpiece materials (carbon steel, aluminum and ductile iron). Methods with model coefficients obtained from experiments could achieve a higher prediction accuracy than those from literatures, which can be explained by the inability of the coefficients from literatures to match the specific machining conditions. When the coefficients are obtained from literatures, the prediction accuracy is largely dependent on the sources of coefficients and there is no definitive dominance of one approach over another. With model coefficients from experiments, the cutting force based model achieves the best accuracy, followed by 2 the exponential function based method and specific energy based method. Furthermore, the power prediction methods can be used in process design stage to support energy consumption reduction of a machining process.
The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.