International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-1, October 2019 461 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: A9555109119/2019©BEIESP DOI: 10.35940/ijeat.A9555.109119 Real Time Prediction of Temperature using ANFIS- SUGENO Model Rashmi Bhardwaj, Varsha Duhoon Abstract: Temperature patterns are continuously change over time and but dependences on the temperature of most of the industries has not yet changed hence making it important for the scientists to predict temperature on a regular basis as most share of the industries in GDP of the economy of country has its dependence on the weather and hence the average of the weather patterns known as climate. In relation to this, the need is to generate a system which can foretell the temperature so that it can help in the various policy making and foreseeing the upcoming catastrophic event. AdaptiveNeuroFuzzyInferenceSystem (ANFIS) and SUGENOmodel a tools & techniques under Artificial Intelligence used for analyzing the data set and foretell the behavior for upcoming reference. ANFIS-SUGENO model used to analyses weather parameters like Humidity, Maximum & Minimum temperature, speed of wind, Bright sunshine (BSS), Evaporation for Delhi daily data set from January 1, 2017 up till February 28, 2018 and further from January 1, 2017 up till November 30, 2018 is used for foretelling and it is observed that the observed and predicted values are much related. Key words: Artificial Intelligence, AdaptiveNeuroFuzzyInferenceSystem (ANFIS), SUGENO Model, Fuzzy Logic, Non-linear I. INTRODUCTION Temperature patterns are continuously change over time and but dependences on the temperature of most of the industries has not yet changed hence making it important for the scientists to predict temperature on a regular basis as most share of the industries in GDP of the economy of country has its dependence on the weather and hence the average of the weather patterns known as climate. Weather Forecasting is the most taken but difficult task at the same time and so scientists have been working is this field to attain the highest accuracy. Artificial Intelligence is the ability of man-made machines that is computers to deal in a way human mind would do. Weather forecasting is the most difficult task due to its chaotic nature as Earth‟s climate is chaotic in nature, hence the vagueness, uncertainty and intuitions are looked by the Artificial Intelligence. Fuzzy deals with vagueness in the time series which is most seen and observed in Meteorological data for forecasting. Meteorological data‟s prediction involves set of reasoning, rules, concepts for prediction. Temperature being a stochastic process makes it Revised Mainscript Received October 05, 2019 Rashmi Bhardwaj*, Professor of Mathematics, University School of Basic & Applied Sciences (USBAS), Head, Non-Linear Dynamics Research Lab, Guru Gobind Singh Indraprastha University, Delhi, India. Varsha Duhoon, Research Scholar(s), USBAS, GGS Indraprastha University, Delhi, India, a rigid rather complex parameter to be predicted as it involves analysis and involvement of other weather affecting parameters too. Fuzzy Logic works with the same nature as it includes set of rules for it to function and hence generate output based on the analysis of other parameters. Weather on Earth is considered to be Chaotic in nature and hence has large impact even if there is some minute change at any point on the surface. Scientist have worked rigorously in the field of climate change analysis and weather prediction using different mathematical models. Climate change is the most challenging problem faced in the present time. The climate has been affected adversely result of which can be seen as the monsoons have shifted and the rainfall also has decreased as compared to the past 10 years. The change in the Climate is result of increasing global warming which has affected the raise in temperature annually.Climate change has significantly affected the rate of growth of a developing country like India, where agriculture is the most taken occupation. Aim is to analyze the relationship between various climatic factors and temperature in the capital of India. Bai. et.al. (2018) applied Artificial Intelligence model, Numerical forecast Methods and Hybrid Models and predicted air pollutants using the forecasting models by comparing statistical models [1]. Chow et.al. (1996) studied application of fuzzy logic to predict load [2]. Ebert et.al. (2008) studied that the High- resolution forecasts from numerical models, further compared to traditional metrics [3]. Grauel et.al. (1999) application of fuzzy sets and system in various aspects [4]. Hossain et.al. predicted air pollution using hidden markov model [5]. Kaloop et.al. (2017) compared ANFIS and ANFIS-WNN models [6]. Kan et.al. (2002) studied a novel fuzzy KNN algorithm for weather prediction [7]. Jing et.al. (2014) applied NFIS-WPM based model to predict daily fuzzy precipitation hence proving traditional ANN had less predictive accuracy [8]. Mukhopadhyay et.al. (2018) applied fuzzy logic to forecast future load on short-term- basis further to predict the electricity load [9]. Setyaningurum et.al. (2015) applied ANFIS to forecast weather parameters ANFIS showed 100% accuracy [10]. Telesca et.al. (2017) applied stochastic weather generator & neuro-fuzzy network; efficiency was tested on basis of the RMSE and MAE [11]. Tilva et.al. (2014) studied climate data in order to help the farmers using Fuzzy-Logic structure to forecast plant disease [12]. Williams et.al. (2009) studied environmental science problems using fuzzy approach [13]. II. METHODOLOGY: