Fuzzy Energy Management Controller for Smart Homes

  • Rabiya Khalid
  • Samia Abid
  • Ayesha Zafar
  • Anila Yasmeen
  • Zahoor Ali Khan
  • Umar Qasim
  • Nadeem JavaidEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 612)


Energy management plays a vital role in maintaining sustainability and reliability of smart grid. It also helps to prevent blackouts. Energy management at consumers side is a complex task. Utility provides incentives like: demand response, time of use and real time pricing models to encourage consumers to reduce electricity consumption in certain periods of time. However, changing energy consumption pattern according to these incentives becomes difficult for consumers. In this paper, we have proposed a fuzzy logic based energy management controller (EMC) for illumination system management. We have used fuzzy logic for reduction of monetary cost and energy consumption. This fuzzy based controller is fully automatic and alters illumination levels between comfort zone of a consumer. It alters illumination level according to price and other input parameters.

1 Introduction

Smart grid is a developed form of traditional grid. The integration of communication and information technologies results in smart grid. It uses real time data and optimization techniques to reduce power losses, maintain sustainability, and increase reliability [1]. Devices of several types are linked together to achieve reliability and sustainability of grid. Different sensor devices, software and automatic device controllers are also included in its network.

Communication is an important factor of smart grid. There exists a two way communication between consumer and utility. Data is sent and received on both sides. This data is used for energy optimization and to maintain grid sustainability. Demand side management (DSM) is also an important feature of smart grid. It optimally schedules load to minimize operating cost of appliances and makes efficient use of energy possible [2]. DSM has two aspects load management and demand response (DR). In load management the focus is on the efficient management of energy, it reduces the possible chances of distress and blackouts. It plays a vital role in reducing peak to average ratio (PAR), electricity bills, and power consumption; it also helps to maintain grids sustainability and reliability. Whereas, DR can be defined as actions performed by consumer in the response of dynamic price rates of electricity by utility. In smart grid renewable energy sources also included which make energy optimization a complex task.

Inefficient usage of energy in residential area is increasing as the number of automatic electric appliances is increasing day by day. In residential area, around 20% to 40% energy is consumed by illumination system. The illumination system can comprise of electric bulbs or tube lights. Optimization of illumination system can result in reduce energy consumption and monetary cost. It can also play a significant role in reduction of PAR. In this paper, we have proposed a fuzzy illumination controller. This fuzzy controller takes values of input parameters, processes them according to fuzzy rules and produces output values of output parameters accordingly. This system operates fully automatically and acts according to the values of input parameters. Rest of the paper is organized as follows: Sect. 2 contains related work. In Sect. 3, problem is described and Sect. 4 contains system model. Simulations and discussion is given in Sect. 5 and paper is concluded in Sect. 6.

2 Literature Review

Scheduling of distributed energy resources (DER) is a difficult task in smart grid. Integration of these distributed resources increases the complexity of energy optimization methods. In [3], an enhanced version of particle swarm optimization (PSO) is proposed which is named as signaled particle swarm optimization (SIPSO). This algorithm performs short term scheduling of DER because accurate forecasting of natural resources for longer time periods is not possible. SIPSO takes more time for execution as compared to its PSO and NPSO but takes less time than rest of the algorithms.

Day-ahead load forecasting (DLF) plays a significant role in efficient management of conventional and renewable energy resources (RES). In [4], a DLF model has been proposed. To forecast the demand load of next day, this model considers energy consumption patterns of previous days along with current day and their average is used as a base to forecast the load consumption patter. The proposed model has improved execution time up to 38% and accuracy of confining non-linearity in load demand curve of previous days is 97%. In spite of this there still exist errors in forecasting DLF which author tends to overcome in future by using autoregressive training model. In [5], a bio-inspired optimization algorithm is proposed to solve energy optimization problem on consumer side. The results show that cost minimization is up to 4% and 7% for first consumer and other two consumers respectively. The limitation of this work is that its comparison with already existing genetic algorithm (GA) is not provided.

Cost minimization method in smart grid using GA has been proposed in [1]. In this paper, RESs and storage devices e.g. batteries are also considered. Real time pricing (RTP) model is utilized for load optimization. When the price of energy is low batteries are changed by RES. When electricity price becomes high, energy from RES is used and batteries are also discharged to meet the energy demand. In case demand is still higher, then additional energy is purchased from grid. In this way cost of energy consumption is minimized and energy is utilized efficiently. It is also concluded that higher battery capacity results in more cost minimization.

In this paper [6], an energy management controller (EMC) is presented for cost minimization in residential area. In this model user comfort is also considered as an important entity. Author specifies maximization of user comfort is also aim of this work but user behaviors during load shifting is not considered and will be considered in future. Security and privacy issues during two way communication between utility and user are also neglected. In [7], a multi objective evolutionary algorithm is implemented. The aim of this work is to use maximum energy efficiently by paying less cost. A function is used to compute the average delay of all appliances. The aim of this work is to minimize average waiting time of appliances and minimum cost of energy consumption.

3 Problem Description

In literature, optimized energy consumption, PAR, user comfort and monetary cost are the main issues of smart grid. Different EMCs and optimization techniques have been proposed in literature to address these issues. In [10], reduction in energy consumption, monetary cost and PAR is achieved. These objectives are achieved by switching off the appliances with low priority. Priorities are assigned to appliances on the basis of user preferences. Authors in [11], reduce energy consumption and PAR using mixed integer linear programming (MILP). Set points of HVAC systems are altered according to outdoor temperature and constraints defined in system. In [12, 13], energy consumption and monetary cost is reduced by using fuzzy logic. Authors in [12], change the set points of HVAC system. These set points are adaptive and can be modified according to the preference of end user. In [8], lightning system is controlled using fuzzy logic. Indoor light intensity is controlled according to the outdoor light effect and user’s lightning comfort. Similarly, [13] reduces energy consumption and cost by controlling the illumination level of LED bulbs. In [14], fuzzy logic is implemented to control the set points and energy consumption of HVAC system. Cost is also reduced without compromising user comfort. However in [10, 11], user comfort is sacrificed. In [10], low priority appliances are switched off which affect the comfort of end user. Similarly, autonomous set points of [11] cannot be override by user which affects comfort of end user. Whereas, [12, 13, 14] consider only throttleable appliances to reduce energy consumption and monetary cost. A significant reduction in cost and PAR can be achieved by considering shiftable appliances along with throttleable appliances. The fuzzy illumination controller proposed in this paper, is designed to control and manage the illumination system. The basic aim of this paper is to reduce energy consumption, monetary cost and PAR while keeping user comfort high. Power consumption and operational cost is reduced by controlling the power consumption of electric lights.

4 System Model

The purpose of our work is to do energy management efficiently. The basic components of our proposed DSM system are: sensors, fuzzy illumination controller, smart meter and user interface. Sensors are used to observe environment and give input to the fuzzy illumination controller for further action. The input by sensors includes data related to light status and movement of user. Smart meter interacts with utility and fuzzy illumination controller and exchanges information related to price and demand of electricity.
Fig. 1.

Fuzzy illumination controller

4.1 Fuzzy Logic

Fuzzy logic is decision making model, which deals with approximate values rather than exact values. In this paper fuzzy logic is applied to adjust set points of HVAC and illumination systems. These set points directly affect the user comfort. Fuzzy logic is easy to implement and does not require large amount of information. It can be used to acquire reliable results using small amount of information. Fuzzy variables are used to generate these results. In fuzzy logic there are some input variables, set of rules and output variables. The input variables are fuzzified into membership functions, these membership functions are basically fuzzy sets and the degree of membership of a value is defined as, how much a value is closed to that set. An input value can at the same time belong to more than one sets. There are different types of membership functions, in our system we have used triangular function because of its simplicity and good performance. In fuzzy logic input and output parameters are defined in simple human understandable language. Fuzzy rules are defined to specify the relationship of input parameters and their effect on the values of output variables. Hence, fuzzy logic follows 3 basic steps: fuzzification of input values via fuzzifier, implementation of rules and defuzzification using defuzzyfier, Fig. 1 shows these steps graphically.

4.2 Fuzzy Controller

These appliances are fully controllable by EMC. In this paper, we are considering lightning bulbs as interruptible appliances. Approximately 20 energy is consumed by electric lights [8]. Therefore, optimized use of energy for lightning plays a vital role to reduce energy consumption, monetary cost and PAR. A fuzzy illumination controller is proposed to control illumination level in a certain range, while keeping user comfort as high as possible. The comfortable lightning range for a home is between 150 (lux) to 250 (lux) [9].

Inputs for Fuzzy Controller

The input parameters used for fuzzy illumination controller are: electricity price, user occupancy, outdoor light and indoor light. The output value of fuzzy controller is the adjusted set point for illumination. The graphical representation of proposed fuzzy illumination controller is given in Fig. 1. The input values are first converted into membership functions. Then rules are used to evaluate and define their relationship and to generate output from these membership functions. The process of mapping values to membership function is called fuzzification. Wheres, mapping back membership function into real value is called defuzzification. In fuzzy logic the processes of fuzzification and defuzzification are carried out by fuzzifier and defuzzifier respectively.
Fig. 2.

Power consumption pattern of illumination system

Fig. 3.

Energy consumption in one month

5 Simulation Results and Discussion

In this Section, we are going to discuss fuzzy illumination controller. Lightening system is operated in two modes: manual mode and autonomous mode. In manual mode, set points of illumination are set manually by user. Whereas, fuzzy illumination controller is used to generate set points of illumination in autonomous mode. The input parameters of fuzzy illumination controller are: Outdoor light, user occupancy, indoor lightning and electricity price. We have also studied the effect of these parameters on power consumption. Figure 2 shows the energy consumption pattern of illumination system in 24 h. It can be seen the energy consumption without supervised fuzzy logic learning (SFLL) is more than the energy consumption with SFLL. Similarly we have compared the energy consumption of SFLL with and without price. Results shows that without price energy consumption is more than SFLL with price. Energy consumption of illumination system for all three case is same till 7am. After 7am electricity price as well as demand of electricity changes and user occupancy also changes, so the pattern of energy consumption varies from 8am to 10pm. We have run this simulation for one month and results are shown in Fig. 3. The energy consumption of illumination system with SFLL enabled is less than the energy consumption of illumination system in manual mode. Moreover, illumination system consumes more energy if we neglect price in SFLL. It shows the importance of price as an input parameter. When price is included as input for decision making, illumination system consumes 140 (W) of less energy than SFLL with price disabled mode. If we compare autonomous mode with manual mode then without SFLL energy consumption is 1096 (W) more than SFLL enabled mode. As it is already mentioned in case study 1 price enabled energy management system play a very important role. Reduction of energy consumption in on-peak hours also decreases cost and that lowering the illumination set points does not effect user comfort, because SFLL system varies set points between user comfort zone.

6 Conclusion

In this paper, an autonomous fuzzy illumination system is proposed. It is an efficient system for energy optimization, it responds to the altering situations e.g. outdoor illumination level, user occupancy and electricity price. We have used time of use pricing scheme. Results show that it has reduced energy consumption up to a significant level. Reduction in energy consumption ultimately results in reduction of monetary cost. Energy consumption has been reduced but user comfort is not compromised in any manner. The fuzzy rules are designed in such a way that illumination set points vary between user comfort range. PAR has also been reduced, as system reduces the set point values of illumination system during on-peak hours.


  1. 1.
    Mary, G.A., Rajarajeswari, R.: Smart grid cost optimization using genetic algorithm. Int. J. Res. Eng. Technol. 3(07), 282–287 (2014)Google Scholar
  2. 2.
    Vardakas, J.S., Zorba, N., Verikoukis, C.V.: A survey on demand response programs in smart grids: pricing methods and optimization algorithms. IEEE Commun. Surv. Tutorials 17(1), 152–178 (2015)CrossRefGoogle Scholar
  3. 3.
    Logenthiran, T., Srinivasan, D., Shun, T.Z.: Demand side management in smart grid using heuristic optimization. IEEE Trans. Smart Grid 3(3), 1244–1252 (2012)CrossRefGoogle Scholar
  4. 4.
    Ahmad, A., Javaid, N., Alrajeh, N., Khan, Z.A., Qasim, U., Khan, A.: A modified feature selection and artificial neural network-based day-ahead load forecasting model for a smart grid. Appl. Sci. 5(4), 1756–1772 (2015)CrossRefGoogle Scholar
  5. 5.
    Mehrshad, M., Tafti, A.D., Effatnejad, R.: Demand-side management in the smart grid based on energy consumption scheduling by NSGA-II. Int. J. Eng. Pract. Res. (2013)Google Scholar
  6. 6.
    Rahim, S., Javaid, N., Ahmad, A., Khan, S.A., Khan, Z.A., Alrajeh, N., Qasim, U.: Exploiting heuristic algorithms to efficiently utilize energy management controllers with renewable energy sources. Energy Build. 129, 452–470 (2016)CrossRefGoogle Scholar
  7. 7.
    Muralitharan, K., Sakthivel, R., Shi, Y.: Multiobjective optimization technique for demand side management with load balancing approach in smart grid. Neurocomputing 177, 110–119 (2016)CrossRefGoogle Scholar
  8. 8.
    Liu, J., Zhang, W., Chu, X., Liu, Y.: Fuzzy logic controller for energy savings in a smart LED lighting system considering lighting comfort and daylight. Energy Build. 127, 95–104 (2016)CrossRefGoogle Scholar
  9. 9.
  10. 10.
    Rastegar, M., Fotuhi-Firuzabad, M., Zareipour, H.: Home energy management incorporating operational priority of appliances. Int. J. Electr. Power Energy Syst. 74, 286–292 (2016)CrossRefGoogle Scholar
  11. 11.
    Perez, K.X., Baldea, M., Edgar, T.F.: Integrated HVAC management and optimal scheduling of smart appliances for community peak load reduction. Energy Build. 123, 34–40 (2016)CrossRefGoogle Scholar
  12. 12.
    Keshtkar, A., Arzanpour, S.: An adaptive fuzzy logic system for residential energy management in smart grid environments. Appl. Energy 186, 68–81 (2017)CrossRefGoogle Scholar
  13. 13.
    Chew, I., Kalavally, V., Oo, N.W., Parkkinen, J.: Design of an energysaving controller for an intelligent LED lighting system. Energy Build. 120, 1–9 (2016)CrossRefGoogle Scholar
  14. 14.
    Keshtkar, A., Arzanpour, S., Keshtkar, F., Ahmadi, P.: Smart residential load reduction via fuzzy logic, wireless sensors, and smart grid incentives. Energy Build. 104, 165–180 (2015)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Rabiya Khalid
    • 1
  • Samia Abid
    • 1
  • Ayesha Zafar
    • 1
  • Anila Yasmeen
    • 1
  • Zahoor Ali Khan
    • 2
  • Umar Qasim
    • 3
  • Nadeem Javaid
    • 1
    Email author
  1. 1.COMSATS Institute of Information TechnologyIslamabadPakistan
  2. 2.Computer Information ScienceHigher Colleges of TechnologyFujairahUnited Arab Emirates
  3. 3.Cameron Library, University of AlbertaEdmontonCanada

Personalised recommendations