Jinxin Xu

Climaate prediction plays a vital role in various sectors, including agriculture, disaster management, and urban planning. Traditional methods for climate forecasting often rely on complex physical models, which require substantial computational resources and may not accurately capture local weather patterns. This study explores the potential of Long Short-Term Memory (LSTM) networks, a type of recurrent neural network, for predicting daily climate variables such as temperature, precipitation, and humidity. Utilizing historical climate data from the city of Delhi, we developed an LSTM model to forecast short-term climate trends. The model consists of two LSTM layers followed by three Dense layers and is compiled with the Adam optimizer, mean squared error loss, and mean absolute error as a metric. Our results demonstrate the model's capability to capture temporal dependencies in climate data, achieving a satisfactory level of accuracy in temperature forecasting. This research underscores the potential of machine learning techniques, particularly LSTM networks, in enhancing climate prediction and contributing to more informed decision-making in weather-sensitive sectors.

This research explores the application of Long Short-Term Memory (LSTM) networks for stock market analysis and prediction, focusing on four major technology stocks: Apple Inc. (AAPL), Google LLC (GOOG), Microsoft Corporation (MSFT), and Amazon.com Inc. (AMZN). Historical stock price data from Yahoo Finance spanning from January 1, 2012, to the present is utilized. The study aims to develop and evaluate an LSTM-based prediction model for forecasting future stock prices. The LSTM model consists of two LSTM layers with 128 and 64 units, respectively, followed by two dense layers. The model is trained using the Adam optimizer and mean squared error (MSE) loss function. Evaluation of the model is done using the root mean squared error (RMSE) metric. The results demonstrate the potential of LSTM models in capturing complex patterns in stock price movements and making reasonably accurate predictions.

The National Institute of Metrology (NIM, China) proposed a joule balance method to measure the Planck constant in 2006, and built the first prototype NIM-1 to verify its principle with a relative uncertainty of 8.9 × 10–6 by 2013. Since 2013, a new joule balance NIM-2 has been designed, with a series of improvements to reduce the measurement uncertainty. By April 2017, NIM-2 has been constructed and can be employed to measure the Planck constant in vacuum. A first measurement on NIM-2 yields a determination of the Planck constant is 6.626 069 2(16) × 10–34 Js with a relative uncertainty of 2.4 × 10–7. The determination differs in relative terms by −1.27 × 10–7 from the CODATA 2014 value. Further improvement of NIM-2 is still in progress towards 10–8 level uncertainty in the future.

The joule balance experiment has been carried out at the National Institute of Metrology, China (NIM) since 2007. By the end of 2013 the first generation of the joule balance (NIM-1) achieved a measurement uncertainty of 7.2 × 10−6 (k = 1). To reduce the measurement uncertainty further, the next generation of the joule balance apparatus (NIM-2) system is under construction. A new coil system using ferromagnetic material is being adopted in NIM-2 to reduce self-heating in the coils. However, the effects on the measurement of the mutual inductance from the nonlinearity and hysteresis of the ferromagnetic material will bring a considerable measurement uncertainty. Inspired by the watt balance, the measurement of the mutual inductance is replaced by an equivalent measurement of the magnetic flux linkage difference. The nonlinearity and hysteresis will not be a problem in the measurement of the magnetic flux linkage difference. This technique comes from the watt balance method. It is called the generalized joule balance method, which is actually a modification of the watt balance method. However, it still represents a valid change that can reduce the difficulty of dynamic measurement experienced using the watt balance. Permanent magnets can also be adopted in the generalized joule balance. To check the feasibility of the generalized joule balance method, some preliminary experiments have been performed on NIM-1. A yokeless permanent magnet system has been designed and used to replace the exciting coils in NIM-1. In this paper, the structure of the yokeless permanent magnet system is introduced. Furthermore, a determination of the Planck constant with the permanent magnet system is presented. The value of the Planck constant h we obtained is 6.626 069(17) × 10−34 J s with a relative standard uncertainty of 2.6 × 10−6.