Institutional Repository
Technical University of Crete
EN  |  EL



My Space

Energy and security management in homes based on algorithm for optimal control, using an experimental setup with low cost Arduino/Android technologies

Kostoglou Stylianos

Full record

Year 2017
Type of Item Diploma Work
Bibliographic Citation Stylianos Kostoglou, "Energy and security management in homes based on algorithm for optimal control, using an experimental setup with low cost Arduino/Android technologies", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2017
Appears in Collections


One of the main goals of automation in a smart home is the regulation and control of the home entertainment systems and household activities, in order to achieve the atmospheric comforts (e.g. heating, ventilation, atmosphere humidity) and simultaneous keep the user’s desired electric power consumption threshold. Therefore, the aim of this thesis is to present an embedded artificial intelligence control system based on a proposed algorithm and applied to the Arduino platform, for decision support in power flow management. For this, an embedded Arduino system is developed which among other receives the environmental values (e.g. temperature), to be processed in a Fuzzy controller and return the output as a recommendation to control the electric home devices.To achieve the above goals, a real time control algorithm based on fuzzy logic theory is developed, in order to be implemented on the ARDUINO UNO microcontroller board, using sensors and Android technologies.At first, the proposed fuzzy algorithm is designed and simulated using MATLAB’s Fuzzy Toolkit. Thereafter, this fuzzy algorithm is implemented through ARDUINO UNO tool kit, using the eFLL (Embedded Fuzzy Logic Library) library which is a standard library for Embedded Systems. The eFLL is developed by Robotic Research Group (RRG) at the State University of Piauí (UESPI-Teresina).The fuzzy algorithm consists of two collaborated fuzzy controllers. The first one has as input: the internal temperature of the home and the external temperature of the environment. The output is the current power level settings of home appliances. The second fuzzy controller has input: the output of the first fuzzy controller, the user’s preferences for the daily upper limit of power consumption, the devices’ total energy consuption and the data of the internal sensor temperature. The output is the set point value of the heating boiler.The platform consists of an Arduino based node with network connectivity which is able to exchange data with the developed Android app. The app can be used to monitor sensor data and set the power level settings or deactivate appliances remotely. Sensor data can also be graphically represented on charts or specifically designed Internet of Things gauges on the Android app.Automations are also built to control energy consumption. The user is able to activate or deactivate specifically programmed automatic modes. Once such modes are activated, appliances will automatically shut down when certain conditions are met, e.g. once the reading of the light sensor reaches a certain point, all the light emitting devices will automatically turn off. The Android app is built according to the best practices and offers a wide variety of options to the end user. Many other state of the art features are incorporated in the platform, such as voice commands, ajax navigation that contribute to the best possible user experience.The platform is also open source, modular and can easily integrate any other type of sensor or actuator with minimal or no code changes. Moreover, the industrial design is sized and implemented taking care of the platform cooling.

Available Files