A Hidden Markov Model and Fuzzy Logic Forecasting Approach for Solar Geyser Water Heating

Time-based smart home controllers govern their environment with a predefined routine, without knowing if this is the most efficient way. Finding suitable model to predict energy consumption could prove be an optimal method manage electricity usage. The work presented in paper outlines development of prediction that controls home, adapting external environmental conditions and occupation. A backup geyser element solar solution identified as metric for more control than time-based controller. system able record multiple remote sensor readings from Internet Things devices, built based on ESP8266 microcontroller, central SQL database includes hot water usage heating patterns. Official weather predictions replace physical sensors, provide data conditions. Fuzzification categorises warm recordings into four linguistic terms (None, Low, Medium High). Partitioning clustering determines relationship patterns between efficiency. Next, hidden Markov predicts efficiency, Viterbi algorithm calculating predictions, Baum–Welch training system. Warm efficiency are used calculate time periods heat through electrical energy. Simulations historical evaluation validation approach, by comparing against heating. In simulation, intelligent controller 19.9% controller, higher temperatures during day. Furthermore, it demonstrated knowledge conditions, can switched 728 times less