Battery Energy Storage Management to reduce electricity cost in USB Building
Buildings microgrid is arising as an important solution for electrical energy distribution networks, which provides the ability to integrate renewable energy generation, storage and consumption devices. Among these microgrid functionalities, the combination of energy storage batteries and energy management method allows the system to achieve different kind of benefits such as increase network robustness, reduce energy cost and reduce energy consumption costs.
The objective of this literature review is to give an overview of the main concepts regarding battery storage management in microgrid. The first section gives an introduction about smart grids including background and functionalities. The second section explains about the battery storage definition and functionalities. The third section, gives an approach of battery modelling process considering two different methods. The fourth section presents two battery storage management strategies that controls the battery charge and discharge process aiming to decrease electricity cost and load demand peaks. The fifth section describes the project context and the proposed work. Finally, conclusions are presented in section 6.
Definition: A microgrid is defined as a small scale combined energy systems to localise the electricity production by the means of distributed generators, renewable resources and loads .This system can be defined as a part of a smart grid which can be interconnected one another to form a big smart system.
Figure 1 Microgrid basic configuration
On the one hand, this intelligent system has several functionalities that distinguishes it from a distribution network:
- Voltage and frequency stability: the system includes methods to control voltage and frequency in the microgrid based on active and reactive power control. The battery storage system is incorporated to the system in order to contribute to limit or control the power transmission through the line .
- Saving costs and energy management functionalities: the system can handle energy to obtain optimal, reliable and efficient operation between all stakeholders (For example: distribution companies, customers, government). The battery storage application reduces the peak demand and consequently the billing and the investment for peak power handling .
- Managing fluctuations in renewable resources: With the using of battery energy storage systems, the microgrid system can solve issues such as uncertainty and intermittency or renewable resources. The battery storage systems improves the response against wide load fluctuations frequency and voltage instabilities by using optimal dispatch strategies.
On the other hand, there is a big economic potential for decentralized distributed generation because it allows to companies to have control even though it covers big areas.
The constant need for efficient energy storage system has brought new technologies that have improved the robustness and price of the system. An advanced battery storage can be defined as an electrochemical solution that accumulates energy and supplies when it is needed. This system includes several components such as computers or controller, battery stacks, electrolyte tanks, pumps and forecasting processor; however, the components may vary due to its material type and field application . Regarding emerging market needs, Energy Storage Systems (ESS) is one of the most important solutions to deal with energy fluctuation, service continuity and frequency issues when renewable sources are connected to grids .For example, the expectative in the United States for system is around 152GW by 2050, the main users will be the individual and small consumers.
The combination of renewable energy resources and unpredictable environment are increasing the complexity and uncertainty in distribution systems. As a result of this, the application of BESS has significantly grown in power systems area in the recent years. Furthermore, in many countries, BESS has been an alternative to ensure stability against power variations and control frequency along and with other applications.Besides, it is playing a vital role due to its advantage of fast dispatching of energy when there is a high demand either in high and low power applications such as homes and buildings. Maximizing economic benefit is the main challenge; therefore, multiple revenue streams which means the using of several BESS functionalities in the same system in order to co optimize and increase the revenue against the investment.
Significant development is going on the battery storage technology. In the table 1 and the table 2 the Solid and Flow batteries characteristics are explained respectively:
Table 1 Solid batteries characteristics
adapted from *[6, 9]
|Lead Acid batteries||Service life: 6 to 15 years
Cycle life:1500 cycles
Self discharging rate: from 4 to 5% month
|Usable capacity decrease when high power is discharged.
Lower density and has hazardous materials.
|Lithium Ion batteries (LI-ION)
|Adaptability in terms of voltage.
Efficiency: 95% and 98%.
Number of cycles: 5000.
Self discharging rate: from 2% to 10% month
|Safety: Toxic at high temperatures; however, there are complementary components that minimize this risk.|
|Nickel cadmium/Nickel metal hydride batteries (Ni-CD/Ni-MH)||Batteries that are very successful for low temperatures between -20 and -40 degrees.||Toxicity of cadmium
Robust and safer than Li-ION
Self discharging rate: 30% month
|Sodium sulphur batteries (NAS):||Efficiency:75%
Cycle life: 4500
Power quality and time shift
|Maintain temperature requires extra energy.|
|Sodium nickel chloride battery (NaNiCl)||Limited overcharge and discharge.
More robust when there is a serial connection.
Table 2 Flow batteries characteristics
*adapted from 
|Redox flow battery||Fast charging and discharging||Low energy density|
|Hybrid flow battery (HFB)||Fast charging and discharging||Low energy density|
In the recent years, many efforts has been dedicated to develop battery mathematical models that can give insights about its operation. To ensure safe and efficient operation, a proper model is required to predict battery behaviour in different operations cases such as charging and discharging. According several articles, there are two main methods to model this system: electric circuit equivalent and state of charge mathematical approach.
In  ,the author states that a common alternative to create a model for battery is employing a electrical circuit which enables characterization with the using of resistors and capacitors, as shown in Fig.2. The strategy is model each cell and progressively extend it to get pack configurations. Depending on the number of connections and the architecture, the analysis can increase its complexity regarding data processing requirement and system parameters identification.
Figure 2 Electrical circuit model for battery cell 
In , a standard battery model is presented based on state of charge. This model is based on a simple series configuration that simplifies algebraic loop problems and it is used for four types of battery including lithium-ion. This model can provide the information about the current voltage (As shown in Fig.2) of the battery which is calculated considering its SOC (State of charge). It is important to indicate that the modelling is considering ideal conditions of temperature and negligible wiring effects. This particular battery model provide non-linear curves for charging and discharging cases what makes it accurate comparing it with real behaviour.
Figure 3 Current discharge curve using the presented model
Battery Energy management one of the most important technologies that plays a central role to improve efficiency in the grid, reduce costs and increase safety. This management system controls how the storage system will be used offering a robust operation . Moreover, this control scheme is based on algorithms that aims different objectives such as reducing load demand peaks, reducing electricity billing, increasing network robustness, and etcetera. In this literature review, two BEMS (Battery Energy Management System) are presented: Load peak shaving and Arbitrage methods.
In , the author presents an Energy Management strategy whose objective is to reduce the load peaks and reduce the billing price. This proposed method defines the battery charging and discharging points over the 24 hours approximated curve. This shape is calculated with a mathematical equation and the load demand historical data, the curve results in the blue bold curve shown in Fig.6.
Using the approximated curve is possible to find the extreme power values which are essential to determine the charging (
t0,te)and discharging points (
tb,tf).Moreover, these charging points must consider the energy capacity of the battery because it defines how much stored energy can be dispatched to the network. In this sense, the optimization strategy is to set the charging/discharging periods the closest to extreme demand values (
Pminas presented in Fig.4). The main advantage of this strategy is the using of the found mathematical equation of the load, which makes the calculation of charging and discharging points more efficient and faster.
Figure 4 Approximated load curve and charging/discharging points 
: Minimum instant power
: Maximum instant power
: Discharging efficiency
: Energy charged to battery
: Energy delivered from battery to distribution grid
In , the article presents a solution to minimize the electricity bill in the shown building microgrid in Fig.5 by reducing the grid energy supply during high price periods. This solution is based in a dynamic model which considers three cases:
- Battery storage is providing the energy; therefore, the energy flow goes from BESS to the load and the grid.
- Photovoltaic generation is enough to give energy to the grid and the batteries, the energy flow goes from PV to BESS and Grid.
- Battery storage is providing the energy; hence, the energy flow goes from BESS to Grid and Load.
The solution algorithm considers three cases because the grid efficiency and electricity price values changes depending on the energy source. Hence, it is possible to calculate the energy costs throughout the operation and evaluate the effects of control schemes in the network system in terms of economic benefits. In this article, the control algorithm must define an optimum charging schedule to reduce electricity costs.
In order determine the charging and discharging periods (schedule), a PSO scheme (particle swarm optimization) is implemented to process the energy flow information. This scheme is applied to forecast the load demand requirements and PV generation energy generation and make the decision making process more accurate. Having this demand information, it is possible to define an optimal operation to minimize the grid energy consumption during high price periods by charging/discharging the battery storage with the algorithm control.
Figure 5 Case study considering PV, BESS, Grid and Load 
The University of Newcastle has invested in a new building called “Urban Sciences building” that has multidisciplinary laboratories for engineering and science studies. During the last 3 months of the year 2017, a demand curve has been measured using the building energy meters. As shown in Fig.8, the curves of maximum, minimum and average demand have their biggest values between 9:00 am and 7:00 pm. However, if we compare Fig.8 with the Fig.7, the demand load peak and high price periods there is a period of time in which both trends indicates a high consumption and high cost (between 4:30pm and 7:00 pm). Due to this fact, the electricity bill has been increasing with the continuous growing demand of the building.
In order to minimize this effect, the proposed solution is to install a battery storage system of 200kW (As presented in Fig.6). According , this system has 8 battery cabinets of 624 VDC put in series, each cabinet has 13 modules of 48VDC; hence, the one of the main challenges of the control is to find a model that can describe the modules group behaviour when there are different inputs. In order to decrease the electricity bill cost, the battery bank will dispatch its energy to diminish the demand loads during high market price periods; however, the battery system does not have any device or scheme to execute this control strategy.
Figure 6 Battery Bank configuration
Figure 7 Generic energy costs according distribution operators 
Figure 8 Measured demand profile from building between January and March 2018
The main aim of this research is to define a control strategy for the battery storage of the Urban Sciences Building (USB) and evaluate the implementation techno-economic feasibility. This system has to enable participation in demand response reducing demand charges using a controller that will allow the charging and discharging of the building battery bank. The system will work with the extracted data from USB Building energy meter and the UK national market prices. The controller has to be implemented with a control algorithm will process the building data and define the optimum times for battery charging and discharging in such a way that the energy load during high cost periods is reduced.
Next, in order to evaluate the control scheme performance, all the data will be processed in a simulation software using the battery storage charging and discharging schedule using different scenarios. As shown in Fig.9, the simulation will be carried out in MATLAB- Simulink considering all the elements in the internal distribution grid such as the solar panel, battery bank, controller, demand load and the grid supply.Finally, the simulation output trends and measurements will be merged with the market data to elaborate a techno-economical evaluation of the system.
Figure 9 Project testing model
There is a big gap between microgrid and common distribution systems; one of the main differences is the ability to include a management component which determines the optimal operation for different kind of applications. Among these energy management applications, battery storage energy management is arising as an important solution to reduce load demand peaks, reduce electricity bill and other benefits. In the project approach, the management strategy and battery storage system will be implemented to reduce the building electricity bill for the internal demand.
In order to reduce the billing prices, the control system must reduce the energy demand during high price periods by compensating it with the battery-stored energy. Furthermore, the battery must charge when the market price is the lowest to avoid negative effects on the economic balance. The decision making process on the management controller must include an algorithm that will calculate optimum operation times for the battery by using market prices (UK National Grid) and load demand profile.
The Battery Storage in USB Building project comprises three main stages: data processing and battery modelling, algorithm selection and simulation stage. First, in the data processing stage, the building meters information is extracted (Demand and PV power) and the electricity tariff is obtained from the distribution supplier. After, the demand and PV power are averaged and classified by their time periods in order analyse it with the market prices. It will be possible to define what kind of algorithm is more suitable prior to the analysis of energy demand trends and market prices.
Next, since the importance to investigate the battery effects in the network, the battery bank has to be accurately modelled considering its electrical characteristics. There are two main approach for battery modelling: electrical equivalent and SOC (State of Charge) model. On one hand, the electrical equivalent assumes an equivalent RC circuit and extends it depending on the number of cells in the battery bank. This first method makes the mathematical modelling simpler; however, there are strong simplification assumptions that decreases its reliability. On the other hand, the SOC approach provides information about the battery bank as one unit using mathematical approximations; nonetheless, the SOC fitting process can be complex depending on the battery characteristics. Therefore, the battery model selection has to be defined using the processed data and the battery bank datasheets.
Finally, the last step consists in building a Simulink system model that will process the management algorithm, the battery model, the distribution network model (Load and source) and the historical inputs such as market prices and demand profile. The Simulink model will allow stressing out the algorithms to find an optimal charging and discharging schedule for the battery bank. Furthermore, this simulation model allows carrying out the analysis of the power output merged with the market energy prices and evaluating the techno-economic feasibility of the project implementation in the USB building.
 R.H.Lasseter;P.Piagi, “Microgrid: A Conceptual Solution,” presented at the 35th Annual EEEE Power electronics Specialist Conference, Germany, 2004.
 R. Garmabdari, M. Moghimi, F. Yang, E. Gray, and J. Lu, “Battery energy storage capacity optimisation for grid-connected microgrids with distributed generators,” in 2017 Australasian Universities Power Engineering Conference (AUPEC), 2017, pp. 1-6.
 M. Kermani, “Transient voltage and frequency stability of an isolated microgrid based on energy storage systems,” in 2016 IEEE 16th International Conference on Environment and Electrical Engineering (EEEIC), 2016, pp. 1-5.
 T. Morstyn, B. Hredzak, and V. G. Agelidis, “Control Strategies for Microgrids with Distributed Energy Storage Systems: An Overview,” IEEE Transactions on Smart Grid, vol. PP, no. 99, pp. 1-1, 2016.
 D. Rosewater, S. Ferreira, D. Schoenwald, J. Hawkins, and S. Santoso, “Battery Energy Storage State-of-Charge Forecasting: Models, Optimization, and Accuracy,” IEEE Transactions on Smart Grid, vol. PP, no. 99, pp. 1-1, 2018.
 IEC. International Electrotechnical Commission.
 Y. Shi, B. Xu, D. Wang, and B. Zhang, “Using Battery Storage for Peak Shaving and Frequency Regulation: Joint Optimization for Superlinear Gains,” IEEE Transactions on Power Systems, vol. PP, no. 99, pp. 1-1, 2017.
 H. Setiadi, N. Mithulananthan, and M. J. Hossain, “Impact of battery energy storage systems on electromechanical oscillations in power systems,” in 2017 IEEE Power & Energy Society General Meeting, 2017, pp. 1-5.
 A. Joseph and M. Shahidehpour, “Battery storage systems in electric power systems,” in 2006 IEEE Power Engineering Society General Meeting, 2006, p. 8 pp.
 L. Guo, M. Ramaswamy, M. Pyne, B. J. Yurkovich, and S. Yurkovich, “Modeling and experimental validation of series connected battery pack modules,” in 2017 IEEE Conference on Control Technology and Applications (CCTA), 2017, pp. 859-865.
 M. T. Lawder et al., “Battery Energy Storage System (BESS) and Battery Management System (BMS) for Grid-Scale Applications,” Proceedings of the IEEE, vol. 102, no. 6, pp. 1014-1030, 2014.
 N. H. Samrat, N. B. Ahmad, I. A. Choudhury, and Z. B. Taha, “Modeling, Control, and Simulation of Battery Storage Photovoltaic-Wave Energy Hybrid Renewable Power Generation Systems for Island Electrification in Malaysia,” The Scientific World Journal, vol. 2014, p. 21, 2014, Art. no. 436376.
 T. Zhang, “The Economic Benefits of Battery Energy Storage System in Electric Distribution System,” Master of Science, Electrical and Computer Engineering, Worcester Polytechnic Insititute, 2013.
 R. K. Bonthu, H. Pham, R. P. Aguilera, and Q. P. Ha, “Minimization of building energy cost by optimally managing PV and battery energy storage systems,” in 2017 20th International Conference on Electrical Machines and Systems (ICEMS), 2017, pp. 1-6.
 “User’s Manual Lithium-Ion Battery System 8*(624V / 39Ah),” SAFT, Ed., ed: INDUSTRIAL BATTERY GROUP, 2008.
 CIBSE, “Capacity for change,” CIBSE, vol. 1, p. 36, 2017.