Engineering and Scientific International Journal (ESIJ) ISSN 2394-7187(Online) Volume 10, Issue 1, January – March 2023 ISSN 2394 - 7179 (Print) 14 DOI: 10.30726/esij/v10.i1.2023.101003 Time Series Prediction Grounded on Neural Prophet- Temperature Forecasting D.B.Shanmugam 1* , P.M.Kavitha 2 , M.Pazhanivelrajan 3 , S.Prithiv Ganth 4 , Dilli Babu 5 1,2 Assistant Professor, Department of Computer Applications, SRM IST, Ramapuram Campus, Chennai 3 BCA Final Year Students, Department of Computer Application s(BCA), SRM IST, Ramapuram Campus, Chennai Corresponding Author: dbshanmugam@gmail.com Abstract—Temperature determining is a moderate and time series investigation cycle to estimate the condition of the temperature for a specific area in coming time. These days, agribusiness and assembling areas are for the most part reliant upon temperature so determining is essential to be exact in light of the fact that temperature admonitions can save life and property. In this work, the Prophet Forecasting Model is utilized for Myitkyina’s yearly temperature estimating utilizing authentic (2010 to 2017) time series information. Myitkyina is the capital city of the northernmost state (Kachin) in Myanmar, found 1480 kilometres from Yangon. Prophet is a particular relapse model for time series forecasts with high precision by utilizing basic interpretable boundaries that think about the impact of custom irregularity and occasions. In this review, the temperature estimating model is proposed by utilizing climate dataset given by an International foundation, National Oceanic and Atmospheric Administration (NOAA). This work executes the multi-step univariate time series expectation model and analyses the anticipated worth against the real information. Such discoveries check that the proposed anticipating model gives an effective and exact expectation for temperature in Myitkyina. Keywords — Solar; Photovoltaic; Thermal; Direct Normal Irradiance; Global Horizontal Irradiance. 1. Introduction A period series is an assortment of information focuses fixed on schedule. Time series examination is applied to investigate these time series information by consolidating various techniques to get significant data. The estimating of the time series information is a system which can assist the model with anticipating future qualities utilizing recently noticed notable qualities [1]. These days, there are significant time series issue to be addressed, for example, how much stock to keep up with, the number of individuals will go by a plane, how high the temperature will be in the following month, how much the cost of a tradable monetary resource will be close to tomorrow. For these issues, each datum researcher should know the methods for time series foreseeing. In this way, time based examples information is vital for any areas. For an association, Forecasting is a significant information science task expected for some exercises to be completed. For weather conditions determining, there are numerous different strategies accessible and numerous specialists are keen on this due to its effect on living things [2]. Subsequently, this paper expects to apply time series examination for temperature guaging of a city. In such examination of time series determining, two central matters have been seen in anticipating: (1) Complete robotized guaging strategies can be testing and too unbendable to even consider carrying out valuable suspicions. (2) Due to the prerequisite of huge involvement with information science ability, investigators can't in any case gauge in great. There exists very much a wide range of ways of determining future patterns, for example, ARCH, ARIMA, counterfeit neural organizations, backward models,. Among them, the Prophet guaging model is utilized in this work to anticipate the temperature of Myitkyina, Myanmar by managing the normal highlights of the climate information. For guaging time series information, a publicly released Prophet is a model delivered by Facebook on 23 February 2017 [3]. In this work, the proposed temperature forecast framework is assembled and it will give the future temperature worth to a city. In this way, it might give meteorologists in anticipating the future temperature esteem rapidly and honestly. The excess piece of this paper is coordinated in the accompanying manner. Segment 2 tends to prior temperature forecast frameworks with different learning calculations recently acted in writing. The main segment 3 blueprints the proposed work of utilizing Facebook Prophet to fabricate a compelling temperature determining strategy, clarifies the particulars of the time series based temperature anticipating method and shows the outcomes for tests by plotting the presentation of the proposed framework, then, at that point, the paper wraps up in Section 4. 2. Literature Review 2.1 Time Series Prediction Fundamentally, the prediction objective of a time series is to estimate the value at time i, yibased on its previous historic data yi-1, yi-2 , … If the interested data is x = { yi-k , yi-k+1 , … , yi-1 }, i= { k,...,n }, the goal aims at finding a function f(x) so that ̂ = () is as close to the