Hybrid System Energy Management and Supervision based on Fuzzy Logic Approach for Electricity Production in Remote Areas Sofiane Berrazouane, Kamal Mohammedi Laboratoire Energétique Mécanique et Ingénieries (LEMI), M Bougara University Boumerdès ,Algeria, E-mails: [email protected], [email protected]ABSRACT: The uncertainty of the renewable resources is a major impediment for successful implementation them for generation an electrical power to supply an isolated area. The combination of several renewable sources (wind and photovoltaic) with a diesel generator is suited to electricity supply of remote areas. The purpose of a hybrid power system is to produce energy at all times requested by consumers and if possible to produce it from renewable sources, the stochastic nature of renewable energy makes fluctuation of real power on the network therefore on the frequency. This problem can be solved or avoided by satisfying the real power supply–demand balance constraint in the hybrid energy system. Using battery bank system can increase the penetration of renewable energy and absorb the rapid fluctuation due to the stochastic nature of renewable energy, that, it increases the difficulty of managing all these systems in real time, ensuring the quality of energy supplied. The objective of this work is to propose a methodology to design a fuzzy logic controller coupling with stateflow to improve its performances. Keywords : renewable energy, energy management, fuzzy logic, stateflow. 1. INTRODUCTION For reducing green house gas emissions and fossil oil consumption, the development of hybrid power system such as diesel Generator, photovoltaic, wind-power for power generations in the remote areas which are not connected to the utility grid is considered economically reliably to a small isolated power system or a micro-grid [1].However, the renewable energy is very intermittent in nature. The necessary to storage excess energy at the periods of production of electricity from renewable sources is superior to the consumption and provide the load at the periods of production of electricity from renewable sources not sufficiently; therefore it is increase the penetration of renewable energy. The power generations from wind and PV may cause a serious problem of power fluctuations in a micro-grid. The diesel generator, wind turbine are characterized by slow transient response, we can use the battery unit because it is able to supply/absorb active power rapidly because it has a high transient response [2, 3].we can observe the difficulty of managing all these systems in real time, ensuring the quality of energy supplied 2. Problem formulation While some emerging control technologies are useful, the traditional power system provides important insights. Key power system concepts can be applied equally well to smart grid system. For example the frequency droop and voltage control used on large utility generators can also provide the same robustness of the traditional power system [4]. In our application, we only interested in controlling the real power through the frequency droop control [4, 5]. The purpose of a hybrid power system is to produce energy at all times requested by consumer and if possible to produce it from renewable sources, the stochastic nature of renewable energy makes fluctuation of real power on the network therefore on the frequency. This problem can be solved or avoided by satisfying the real power supply-demand balance constraint in the Micro-grid power system. Therefore, an objective function for frequency control in the micro-grid system is formulized as follows: ܨ߂ൌ ߂ .ܭ՜0 (1) Where: ߂ൌ ௦௨௬ െ ௗௗ (2) ௦௨௬ : is a power supply generated by different sources and ௗௗ is a power demand by load, reverse osmosis and battery bank when it is charging ൌ ோா ௧_ௗ (3) ൌ ௗ ௧_ (4)
Transcript
Hybrid System Energy Management and Supervision based on Fuzzy Logic Approach
for Electricity Production in Remote Areas
Sofiane Berrazouane, Kamal Mohammedi Laboratoire Energétique Mécanique et Ingénieries (LEMI), M Bougara University Boumerdès ,Algeria,
ABSRACT: The uncertainty of the renewable resources is a major impediment for successful implementation them for generation an electrical power to supply an isolated area. The combination of several renewable sources (wind and photovoltaic) with a diesel generator is suited to electricity supply of remote areas. The purpose of a hybrid power system is to produce energy at all times requested by consumers and if possible to produce it from renewable sources, the stochastic nature of renewable energy makes fluctuation of real power on the network therefore on the frequency. This problem can be solved or avoided by satisfying the real power supply–demand balance constraint in the hybrid energy system. Using battery bank system can increase the penetration of renewable energy and absorb the rapid fluctuation due to the stochastic nature of renewable energy, that, it increases the difficulty of managing all these systems in real time, ensuring the quality of energy supplied. The objective of this work is to propose a methodology to design a fuzzy logic controller coupling with stateflow to improve its performances. Keywords: renewable energy, energy management, fuzzy logic, stateflow.
1. INTRODUCTION For reducing green house gas emissions and fossil oil consumption, the development of hybrid power system such as diesel Generator, photovoltaic, wind-power for power generations in the remote areas which are not connected to the utility grid is considered economically reliably to a small isolated power system or a micro-grid [1].However, the renewable energy is very intermittent in nature. The necessary to storage excess energy at the periods of production of electricity from renewable sources is superior to the consumption and provide the load at the periods of production of electricity from renewable sources not sufficiently; therefore it is increase the penetration of renewable energy. The power generations from wind and PV may cause a serious problem of power fluctuations in a micro-grid. The diesel generator, wind turbine are characterized by slow transient response, we can use the battery unit because it is able to supply/absorb active power rapidly because it has a high transient response [2, 3].we can observe the difficulty of managing all these systems in real time, ensuring the quality of energy supplied 2. Problem formulation While some emerging control technologies are useful, the traditional power system provides important insights. Key power system concepts can be applied equally well to smart grid system. For example the frequency droop and voltage control used on large utility generators can also provide the same robustness of the traditional power system [4]. In our application, we only interested in controlling
the real power through the frequency droop control [4, 5]. The purpose of a hybrid power system is to produce energy at all times requested by consumer and if possible to produce it from renewable sources, the stochastic nature of renewable energy makes fluctuation of real power on the network therefore on the frequency. This problem can be solved or avoided by satisfying the real power supply-demand balance constraint in the Micro-grid power system. Therefore, an objective function for frequency control in the micro-grid system is formulized as follows: . 0 (1) Where:
(2)
: is a power supply generated by different sources and is a power demand by load, reverse osmosis and battery bank when it is charging _ (3) _ (4)
Fig1: Hybrid Energy System Along with the set of input information, the HES will need a strategy for making decisions. For a given set of information that an EM has access to, there are many control strategies that could be employed. Each must be judged by the HES performance it achieves and the costs associated with its implementation. Three possible control strategies are real-time optimization, expert system control, and decentralized control [6]. 3. Principle of command As depicted in Fig1. the hybrid energy system consists of the Diesel Genset DG, Wind Turbine WT, photovoltaic PV, battery and load. On the system, sum action allow to correct the power produced and power consumed to achieve a balance between production and consumption and thus keep the frequency close to its reference value. These actions will act on the production of renewable energy, diesel generator and also of battery The DG is used to supply power to the HES system when the other systems cannot sufficiently provide. Along with the set of input information, the HES will need a strategy for making decisions. There are many control strategies that could be employed. Each must be judged by the HES performance it achieves and the costs associated with its implementation. Three possible control strategies are real-time optimization, expert system control, and decentralized control [6]. For this problem, the expert system control (fuzzy logic) based supervision is chosen as it is well adapted to deal with [7,8]: • The complexity of the system, which has to be controlled and the difficulty to obtain or to use accurate models, and • The difficulty to predict the behavior of the renewable energy and the variation of the network frequency with load variations.
For known the real state of the system to increase its reliability and its dynamics, we have three inputs in the Fuzzy Logic Supervisor FLS: state of charge of bettery , frequency error between the network normal frequency ( ) and the measured frequency ( ), the difference between the power generated by renewable energy and the total power generated by hybrid generator when:
(5) The greatness δ to maximize used of renewable energy. We was added a gains on the inputs and outputs of FLS to normalize them: on the input : _ / _ (6) on the output : _ _ _ _ (7) _ _ _ _ (8) The reference power from renewable sources is the multiplication between the power controller output and the maximum power available at just local controller. _ _ _ _ (9) Therefore, the power control varies depending the power available, for example if the maximum power 16 , if the controller gives a medium power control, the power is _8 but 10 medium power control then _ 5 .That, to maximize the use of renewable energy. The block diagram of FLS is shown in Fig this program was developed on MATLAB/Simulink .
Fig2: structure of the supervisor
4. Internal structure of supervisor The supervisor of the control algorithm with fuzzy Logic make possible to improve in real time the desired performance. A fuzzy logic controller relates the controller output to the inputs using a list of if–then statements called rules; some of these rules, they evaluate each state of the system to obtain the controller output. However, if we have multi outputs and after the inference of rules, sometimes they generate contradictory actions for the reason that the activation of two or more rules at the same time when they have opposite actions. In order to solve this problem, we have introduced in this work another kind of control based on Stateflow which runs in parallel with fuzzy logic. Stateflow belongs discrete controllers. It provides the ability to model hierarchical parallel Statecharts and integrates three basic components:
• hierarchical and parallel state machines borrowed from Statecharts
• control flow diagrams allowing to design complex transitions between Stateflow states
• truth tables allowing to design complex actions
4.1. Internal structure of Fuzzy logic controller The computation of fuzzy logic is composed in three stage: fuzzification,the Inference engine and Defuzzification. In this paper the membership function for input variables and output variable of the fuzzy controller are considered as Table 1. The triangular and Trapezoidal membership functions are used as membership functions for the input and output variables. The Defuzzification method followed in this study is the “Center of Area Method”
Tab 1: The membership functions of fuzzification and Defuzzification Fuzzification Type of
function Defuzzification Type de
function Frequency N : negative [-∞ ,0.1, 0] Z : Zero [-0.1,0, 0.1] P : Positive [0,0.1, +∞]
Trapezoidal Triangular Trapezoidal
power reference of Battery NB : Negative big [-1,-1,-0.8, -0.6] NS : Negative small [-0.8,-0.6, -0.05,0] Z : zero [-0.05, 0, 0.05] PS : Positive small [0, 0.05,0.6, 0.8]
Difference of power ΔP NG : Negative Big [-∞,-0.4, -0.2] NM :Negative medium [-0.4,-0.2,0] Z : zero [-0.2, 0,0.4] PM : Positive medium [0,0.4, 0.8] PG : positive grate [0.4, 0.8, + ∞]
power reference of diesel generator Z : Zero [-1,5, -0.15] M : medium [0, 0.15, 0.8,0.95]
B :Big [0.4 ,0.7, 1.5, 1.5]
Triangular Trapezoidal Trapezoid
State of charge of Battery L : low [0 ,0 ,0.2, 0.3] M : middle [0.2, 0.3, 0.8, 0.9] H : high [0.8 ,0.9, 1,1]
Trapezoidal Trapezoidal Trapezoidal
power reference of renewable energy Z : Zero [0,0, 0.4] M: medium [0, 0.15, 0.8,0.95]
B : Big [0.6 ,0.8, 1.5, 1.5]
Triangular Trapezoidal Trapezoidal
Inference engine We must meet the following conditions:
- maintaining the state of charge of battery around its average value to protect against overload or deep discharge - The battery will be charging from renewable sources in first time or by the diesel generator for allow the last works a nominal power
- The diesel generator starts only if the power of other sources is not enough
- reduce the number of start / stop of diesel generator Therefore the fuzzy rules are: 1–if
2–if 3–if 4–if 5–if 6–if
7–if 8–if 9–if 10–if 11–if 12–if 13–if 14–if 15–if 16–if 17–if 18–if 19–if 4.2. Internal structure of stateflow controller Stateflow allows representing the system by through of state-transitions diagrams such as in this work, the state is reserved for DG (output) to manage its sequence of on/off and the transition is reserved for three inputs: state of charge of battery
, reference power of DG comes from fuzzy logic controller _ and difference power . State1: the generator diesel turns off _ 0 Condition 2: the state of charge of battery 0.5 or Condition 3: difference power 0 and it is maintained at minimum for 30s due to renewable energy State2: the generator diesel turns on (nominal power) such as its power _ 0.7 Condition 1: reference power of DG _0.3
Fig5: Stateflow controller 5. Modeling of each power sources Within the framework of this work, the power output characteristics of battery bank, diesel generator and renewable energy systems are linearized by the following first-order transfer functions based on different time constants as shown in [10, 11, and 12]
(10) 5.1. Modeling diesel generator The power of diesel generator is written as follows: _ (11) In our application, the minimum power of diesel generator equal 30% of maximum power 5.2. Modeling Renewable energy system The production system receives instructions via the fuzzy control system and the local control at the renewable energy generation. The latter is designed to maximize production (maximum power point tracer ‘MPPT‘).
. , . 12 5.3. Modeling of battery energy storage system The storage system is characterized by an efficiency charging and discharge . _ . (13) . _ . é (14) The storage level is written as follows:
(15) Table 2 shows the simulation parameters such as the penetration of renewable energy can reach 100% of the overall power of our station
6. Simulation results and discussions The initial state of charge of battery in the HES is set to 60% at the simulation starting point and the feasibility of controller is examined over the course of 24 hour. the power supplied by the generator diesel a unless than 30% of its rated power; this condition is assured by STC, this type of control protect the DG and battery bank against depth of discharge Fig3 shows the exchanges of real power between the different components constitute the system such as the power of battery is positive when charging and negative when discharging. The fluctuations caused by renewable energy is absorbed by it The battery provide the load by energy necessary (00h at 1h40) when the less than 30% the generator diesel turn on in first time by fuzzy logic controller after that, the stateflow controller maintain it a nominal power, the excess energy powered by DG it is absorbed by the battery until its state of charge attain the 50% where the DG turn off always by SFC, the operating time of generator is varied according to the power that it must be compensated beyond energy produced by renewable energy, in the first. The battery absorbs the excess energy powered by renewable energy but if its state of charge reaches 90%, our controller degrades the power of renewable sources until the load power is satisfied by them (14:20at 16:30 h),it uses the battery’s reserve when the renewable sources don't provide all power demand. When the power demand is greater than 70% of power DG commanded by SFC, the FLC imposes the power equal only the power load (equation 10) we show that in at 21h8. Fig 4 shows the power per unite where we observe that is maximized (equation 12) Fig5 and fig 6 show the power blatancy profile and state of charge of battery bank respectively.
Fig3: powers exchange with SFC
Fig 4: powers exchange per unit
Fig 5: Sate of Charge of battery
Fig 6:power balancing 8. Conclusion In this work, a dynamic control method in hybrid energy system is proposed, the fuzzy logic controller is introduced with stateflow to management HES. The proposed control helps to solve power quality issue resulting from real power imbalance fluctuations due to renewable energy, and it can protect the diesel generator against series on/off in the short time, otherwise it is protected the battery bank against the surcharge and deep discharge REFERENCES [1] M.S. Mohsen and J.O. Jaber, A photovoltaic-powered system for water desalination, Desalination, 138 (2001) 129–136. [2] D. Weiner, D. Fisher, E.J. Moses, B. Katz and G.Meron, Operation experience of a solar- and wind-powered desalination demonstration plant, Desali-nation, 137 (2001) 7–13. [3] Mohammedi K and al, Simulation and Energy Analysis of a Small Scale Seawater Desalination/ Electricity Production Prototype Powered with Renewable Energy, Int. J. of Thermal & Environmental Engineering Volume 2, No. 2 (2011) 107-112
[4]Robert H. Lasseter, Fellow, IEEE. Micro-grids and Distributed Generation .Journal of Energy Engineering, American Society of Civil Engineers, Sept. 2007 [5]Xiangjun Li a, Yu-Jin Songb, Soo-Bin Hanb . Frequency control in micro-grid power system combined with electrolyzer system and fuzzy PI controller. Journal of Power Sources 180 (2008) 468–475 [6]R., and C. Marnay. Energy Manager Design for Micro-grids. Firestone, LBNL-54447. January 2005 [7]Ghata boukettaia, and al. Fuzzy logic supervisor for power control of on islanded Hybrid Energy production units. In: international journal of electrical and engineering 279-285, 2007 [8]Vincent Courtecuisse a,b, Jonathan Sprooten a,b,Benoît Robyns a,b. A methodology to design a fuzzy logic based supervision of Hybrid Renewable Energy Systems. In: Mathematics and Computers in Simulation 81 (2010) 208–224 [10]S. Obara. Load response characteristics of a fuel cell micro-grid with control of number of units. International Journal of Hydrogen Energy 31 (2006) 1819 – 1830 [11] Shin’ya Obara. Analysis of a fuel cell micro-grid with a small-scale wind turbine generator. in International Journal of Hydrogen Energy 32 (2007) 323 – 336 [12]M. Mohamed Thameem Ansari*, S Velusami. DMLHFLC (Dual mode linguistic hedge fuzzy logic controller) for an isolated wind/diesel hybrid power system with BES (battery energy storage) unit in Energy 35 (2010) 3827e3837 [13] www.open-gain.org [14] Valenciaga F, Puleston PF. Supervisor control for a stand-alone hybrid generation system using wind and photovoltaic energy. IEEE Transactions on Energy Conversion 2005;20(2):398–440. [15] Souhir Sallem, Maher Chaabene , M.B.A. Kamoun. Energy management algorithm for an optimum control of a photovoltaic, Applied Energy 86 (2009) 2671–2680
Appendix The nominal parameters and nominal operating condition of the system are listed in Table 2 Parameters Diesel Generator _ 20 _ 20 0.3 1.7
