CC BY
4
UDC 620.91:621.31
doi: 10.20998/2074-272X.2021.2.06
A.A. Shavelkin, J. Gerlici, 1.О. Shvedchykova, K. Kravchenko, H.V. Kruhliak
MANAGEMENT OF POWER CONSUMPTION IN A PHOTOVOLTAIC SYSTEM WITH A STORAGE BATTERY CONNECTED TO THE NETWORK WITH MULTI-ZONE ELECTRICITY PRICING TO SUPPLY THE LOCAL FACILITY OWN NEEDS
Purpose. Improving the principles of management of photovoltaic system with storage battery and with autonomous functioning during daylight hours for a local object, connected to the grid with multi-zone payment when excluding the generation of energy into the grid. Methodology. Modeling and analysis of energy processes in the photovoltaic system was performed using the Matlab software package. The simulation model of energy processes is based on calculated expressions taking into account the characteristics of the battery. Operability of the proposed solutions are confirmed on an experimental setup based on a standard hybrid inverter. Results. It's shown, that due to the battery energy during the most loaded peak hours and part of the daytime the system operates autonomously and does not depend on possible violations of the quality of electricity in the grid. Scenarios of the recommended load schedule are proposed in accordance with the ratio of the predicted value of the daily energy generation of the photovoltaic battery to its possible maximum value. A simulation model of energy processes in the system with the correction of the recommended load value was developed. Originality. A method of the recommended load calculation with current correction for the actual generation and degree of battery charge is proposed, which allows taking into account differences the actual generation of the photovoltaic battery from its predicted value and the actual load from the recommended one. Practical value. The obtained solutions are the basis for the design of new and modernization of existing photovoltaic systems of local objects using software and hardware complexes for power consumption management. References 18, figures 4.
Key words: multi-zone electricity pricing, energy redistribution, storage battery state of charge, PWM, recommended load scenarios with current correction, simulation.
Удосконалено принципи управлтня i перерозподыу енергп, яка накопичуеться в акумуляторнш батарег, в фотоелектричнш cucmeMi локального об'екта, тдключеного до мережi з багатозонною тарифжащею при виключенн генерацп енергп в мережу. За рахунок енергп батарег в найбыьш навантажен пiковi години та частково в денний час система працюе автономно i не залежить вiд можливих спотворень якоcmi електроенергп в мeрeжi. Запропоновано сценарп рекомендованого графжа навантаження вiдповiдно до вiдношeння прогнозованого значення денног генерацп енергп фотоелектричног батарег до ii можливого максимального значення. Запропоновано методику розрахунку рекомендованого навантаження з поточним корегуванням за фактичною генеращею i ступенем заряду батарег, що дозволяе врахувати вiдхuлeння фактичног генерацп фотоелектрично'1 батарег вiд прогнозного значення i фактичного навантаження вiд рекомендованого. Розроблено жтацшну модель енергетичних процеЫв в cucmeмi з корегуванням значення рекомендованого навантаження. Працездаттсть запропонованих ршень тдтверджено моделюванням в Маtlab i на експериментальнш установц на базi стандартного гiбрuдного твертора. Отриман ршення е основою для проектування нових i модертзацп icнуючuх фотоелектричних систем локальних об'екmiв з використанням програмно-техтчних комплекЫв управлтня електроспоживанням. Бiбл. 18, рис. 4.
Ключовi слова: багатозонний тариф, перерозподш енергИ, стушнь заряду акумулятора, Ш1М, сценарй" рекомендованого навантаження з поточним корегуванням, моделювання.
Усовершенствованы принципы управления и перераспределения энергии, накапливаемой в аккумуляторной батарее, в фотоэлектрической системе локального объекта, подключенного к сети с многозонной тарификацией при исключении генерации энергии в сеть. За счет энергии батареи в наиболее нагруженные пиковые часы и частично в дневное время система работает автономно и не зависит от возможных нарушений качества электроэнергии в сети. Предложено сценарии рекомендованного графика нагрузки в соответствии с отношением прогнозируемого значения дневной генерации энергии фотоэлектрической батареи к ее возможному максимальному значению. Предложена методика расчета рекомендованной нагрузки с текущей корректировкой по фактической генерации и степени заряда батареи, что позволяет учесть отличия фактической генерации фотоэлектрической батареи от прогнозного значения и фактической нагрузки от рекомендованной. Разработана имитационная модель энергетических процессов в системе с корректировкой значения рекомендованной нагрузки. Работоспособность предложенных решений подтверждена моделированием в Маtlab и на экспериментальной установке на базе стандартного гибридного инвертора. Полученные решения являются основой для проектирования новых и модернизации существующих фотоэлектрических систем локальных объектов с использованием программно-технических комплексов управления электропотреблением. Библ. 18, рис. 4.
Ключевые слова: многозонный тариф, перераспределение энергии, степень заряда аккумулятора, ШИМ, сценарии рекомендованной нагрузки с текущей корректировкой, моделирование.
Introduction. «Green» energy and, first of all, photovoltaic systems (PVSs) are becoming more widespread in modern conditions. This also applies to «small» energy, in particular, economic entities - local facilities (LFs) for various purposes. Here, the use of combined (hybrid) PVSs with a storage battery (SB) and connection to a distribution network (DN) allows: to increase the reliability of power supply due to the
uninterruptible power supply function in conditions of voltage outages when using solar generation and SB energy; to increase the power consumption of the LF, this is especially important if there is a limit on the power from the power system, when the use of PVSs is cheaper than the construction of a new power transmission line and equipment for connecting to the DN.
© A.A. Shavelkin, J. Gerlici, 1.О. Shvedchykova, K. Kravchenko, H.V. Kruhliak
The interest in the sale of electricity at the «green» tariff, along with the provision of their own needs, is not decisive for the LF which is associated with the overestimation of the power of the PVS and with the need for registration of permits. In addition, approaches to «green» tariffs are changing, and the tariffs themselves are decreasing [1]. In this regard, the approach adopted in world practice is promising, when the consumer is a prosumer [2], and energy is consumed where it is generated. This partly alleviates the problems with the management of generation in the power system and the need to ensure the balance of energy.
The effectiveness of the introduction of PVS for business entities is determined in accordance with the increase in production and improved working conditions. At the same time, the introduction of such systems should ensure a reduction in the cost of paying for electricity consumption from the DN, especially taking into account the constant growth of tariffs. It is promising to use a PVS with a SB when connected to DN with multi-zone tariffication with the redistribution of energy between tariff zones [3-5].
The real need for such PVSs is evidenced by the fact that ready-made solutions [6-8] and the development of «hybrid» inverters are widely represented on the electrical market. These solutions represent modern software and hardware complexes: with power supply of the LF from an autonomous voltage inverter (AVI) and switching the LF load to the DN (bypass) with insufficient generation of a photovoltaic battery (PB) and SB energy; with a network inverter (NI), when the load and NI operate in parallel with the DN. These solutions are intended for use in a one-part tariff, have a developed interface with the possibility of remote control of parameters, for example, with output to a smartphone. They have the ability to reconfigure the parameters, but everything is done by the operator, and their effective use requires experience and skills. There are no functions for generating recommendations and automatic adjustment according to the weather forecast.
The cycle of operation of hybrid inverters provides for the power supply of the LF load from the PB with recharging the SB; if the PB energy is insufficient, the SB energy is used. When the state of charge of the SB drops to the threshold value, the transition to the bypass mode is carried out, and the LF load is connected to the DN with the SB charge (from the PB and the DN); if the SB energy is sufficient, the LF is powered again from the SB. The improvement of PVS in the conditions of multi-zone tariffication is associated with the binding of switching of operating modes to the hours of tariff zones, control of the energy supply from the PB and the SB charge. Here, it becomes possible to exclude energy consumption during peak loads [4, 5], the discharge of the SB charged from the night (according to the night rate) in the morning hours until the morning peak is excluded, and the SB is charged until the evening peak.
In real conditions, when using worn out and overloaded DNs, the option with AVI has an advantage, which, if energy consumption is excluded from the DN during peak loads and during light time, ensures the autonomous operation of the LF. In this case, the quality
of the LF voltage is determined by the AVI, which will contribute to the normal functioning of the LF in the event of deterioration in the quality of the DN voltage. When using a NI operating in parallel with a DN, switching to stand-alone mode can be accomplished by disconnecting from the DN.
In solutions of PVS with hybrid inverters, the use of a SB is considered, the energy consumption of which WB varies within wide limits - WB = 1-4.8 kWh for a PB with power of Pr=1 kW. Obviously, the WB value is determined by the system functioning algorithm. At the same time, the cost of the SB is significant and requires justification from the condition of sufficiency.
The use of daily meteorological forecast looks promising with the formation of recommendations on the LF load schedule PL(t) which will allow planning operation modes and ensuring a reduction in energy consumption from the DN at more expensive tariffs. It becomes possible to automatically adjust according to the season and forecast. Various approaches are used using data from meteorological sites [9]. An interesting and modern solution is forecasting using neural networks [10, 11]. At the same time, the availability of the most accurate forecast does not solve the issue of efficient use of the energy of the PB and SB without linking PL(t) to it. And here it seems appropriate to use the system with the formation of the PRt) load schedule recommended in accordance with the forecast under various operating scenarios. Since it is impossible to ensure an exact match of the PLR(t) load, it is advisable to use the current Pm(t) correction. The correction can be carried out according to the actual parameters of the PVS. This issue requires further study. In this case, the forecast accuracy of the PB generation PPV(t) at the level of the meteorological site data can be sufficient for effective correction [12].
Simulation modelling is an effective tool for studying the possibilities of controlling energy processes in PVS [13-15]. Here, the introduction of additional elements makes it possible to assess the possibilities and efficiency of correcting the PLR(t) load schedule for various deviations of the actual PPVF(t) and the predicted Ppvp(i) generation of the PB.
The goal of the work is to improve the management principles for a photovoltaic system with a storage battery and autonomous operation during the daytime for a local facility connected to a multi-zone billing network while excluding the generation of energy to the network.
It is necessary to solve the following tasks:
• to study the possibility of using a limited number of scenarios of operation with the recommended load schedule based on the forecast of the PB generation with the current correction based on the actual values of generation and the degree of SB charge;
• to develop a simulation model of the system to study energy processes in the daily cycle using the recommended load correction;
• to carry out experimental validation using a standard hybrid inverter.
The structure of the power circuits of the PVS. The structure of a PVS with a SB (using the example of a single-phase version of the PVS implementation, see
Fig. 1) is based on a standard approach and contains: an autonomous inverter (VSI) with an output LC filter, PB (PV), SB, Load, DN (G) and switching devices. The PB is connected to the AVI input through a DC/DC1 step-up converter with the Maximum Power Point Tracking (MPPT) function, the SB is connected through a DC/DC2 (charge controller) with two-way conduction. As a starting point, a variant with the use of a hybrid inverter of the Axioma Energy ISMPPT 3000 type (3 kVA) is considered. The LF load is connected to the AVI output and through the K1 contactor with the VS triac - to the DN. The AVI with PWM and an output filter is used as a source of sinusoidal voltage, and when the load is connected to the network, it is used in the SB charging mode. K1 is needed to disconnect the PVS from the DN when the voltage disappears in it (emergency mode). VS is used to exclude a pause when connecting to the DN when the voltage is restored in it (first, K1 is closed, and
K2_ I
after synchronizing the AVI voltage uC with the network voltage ug, the triac is switched on). Relay K2 is used to switching off the PB. Voltage and current sensors in Fig. 1 are not shown.
Let us consider the operation of the PVS in a daily cycle with autonomous operation when the energy of the PB and the SB is sufficient to consume the LF load and switching to bypass when the energy of the PB and SB is insufficient. A variant of the load graph PL(t) is shown in Fig. 2 (bypass zones are highlighted). The following conventional zones and relative tariffs have been adopted: daytime Td=1 (t2=11:00 - t5=20:00), nighttime Tn=0.4 (t6=22:00 - ti=7:00), peak morning Tm=1.5 (tj=7:00 -t2=11:00) and evening Te=1.5 (t5=20:00 - t6=22:00). In this case, an unfavorable option is considered, when the peak loads are shifted by the hours of minimum PB generation.
SB
Fig. 1. Structure of the power circuits of the PVS
Pl2
P23 _
P56
P35
1 P61
! !
t\ Î2 ¡4 5 ¡6 t
Fig. 2. Dependence PL(t)
In this case, it is necessary to take into account the peculiarities of the charging characteristics of the SB. The standard solution uses a charge in three stages [16, 17]. In this case, the «active» is the first stage when the state of charge (SOC) of the SB Q*=100-Q/Qr<80 % (Q and Qr are the current and nominal values, respectively), when the charge current IB for different SB types can vary from 0 to iBmax = (2-3)lBr= (0.2-0.3)Cs (Cs is the SB capacitance, IBr=0.1CB is the nominal value). With further charging, the current decreases significantly and, accordingly, the ability of the SB to receive energy decreases.
The depth of discharge (DOD) and the number of discharge cycles nrc determine the SB life. Usually the acceptable DOD does not exceed 30-50 %. Thus, DOD
must be controlled, limited (not lower than 50 %), and SOC must be maintained in the active zone, for example, 75-80 %. This is possible with the use of a relay regulator acting on relay K2 (Fig. 1) with PB disconnection. This applies to the time interval when PPV>PL and it is possible to charge the SB using the energy of the PB. In the early evening, when PPV < PL, when Q* drops to the set value Q*d, the load and the inverter are connected to the DN with the SB charge from the PB and the network.
The autonomous operation of the PVS with limited SB capacitance also imposes an additional limitation - the need to reduce the generation of the SB when its energy is excessive and there is nowhere to put it.
We proceed from the data on the generation of the PB PPV(t) in the conditions of Kyiv according to the archival data [18]. As an indicator of the efficiency of PVS, we use a simplified indicator kE = C / C2 (Q and C2 are the cost of electricity consumed by the LF and the cost of electricity consumed from the DN). Let us consider various options for LF load schedules with constant generation PPV(t) and with a change in the SB capacitance. Energy WBf, given by the SB, taking into account the efficiency nB of the SB and the efficiency nc of the converter is WBf = WB^B^C (WB = UBCB, UB is the SB voltage). We accept WBf values from 1kW h to 4.8 kWh. We are considering the option of using lead-
0
acid SBs of the OPzV12-100 type (12 V, 100 Ah), which at DOD up to 50 % provide for at least 2500 discharge cycles. With a two-hour discharge at UB=1.85 V, IB=29.5 A, the discharge power is 57.6 W, and, accordingly, nB= 0.947 [17]. In this case, the SB is used in the active charging zone up to Q* = 80 %, and only in the hours preceding the peak hours, it is possible to charge the SB from the network over 80 %.
Mode of operation with constant load from 7:00 to 20:00. In the evening, with a decrease in PB generation, connection to the DN is inevitable at time t4. We accept the control values Q*1=95 %, Q*5=85 %, Q*4=62 %, Q*min>50 %. In accordance with [17], the duration of the SB charge from Q*4=62 % to Q*5=83-85 % is approximately At = 2.5 hours, respectively, t4=17:30.
In the hours of the evening peak (t5, t6) at AQ*56 = = Q*5 - Q*mm, the value of the load power P56 can be taken from the condition of providing a two-hour discharge (t56 = 2 h)
= AQ W . (1)
56 (t6 -15)100 For the interval (t1, t4) AQ*14=Q*1-Q*4, WPV14 is the energy generated by the PB on the interval (t1, t4) in accordance with PPV(t), and the load power PL=P14=P15 AQ *14 WBf
WB14 =-
Pl4 =
(t4 - ¿0100
WB14 + WPV l4nC
¿A — t\
(2)
(3)
kE =
in mind that at WB=1828 Wh, the value of the load power P12~PAVD (PAVd is the average value of the PB generation power per day), and in the evening peak P56 is almost half as low. At WB=2742 Wh, the value of P56 is close to PAVD, and P12 exceeds PAVD, i.e. this value of the SB capacitance is preferable.
The solar activity time for the summer period is limited to t3=16:30 and the kE value can be increased (up to kE=(4.1-4.6)) with a decrease in PL in the interval (t3, t5). Here, P13 increases. However, the capacity of the SB is underused during the morning peak hours.
Let us consider the option PL(t) (Fig. 2) while ensuring the maximum value of P12 and reducing the load P35. This will reduce the bypass time to 1.5 hours (t4=18:30), which is enough to charge the SB by 10 %. We assume that Q*1=95 %, Q*2=55 %, Q*3=78 % (less than 80 %), Q*4=75 %, Q*5=85 %.
The values of WB12, WB23, WB34 are determined in accordance with (2), P12, P23, P34=P45 - in accordance with (3), P56 - in accordance with (1). Then
kE
To exclude a deep discharge of the SB (Q*2>62 %) during the morning peak hours (t1, t2), a similar calculation is performed for P12. The lower power value is taken as PL.
When charging the SB from the network, energy is consumed:
• on the interval (t6, M - Wbc = 0.01AQ*a Wb;
• on the interval (t4, t5) - WBC45 = 0.01AQ*45WBf.
Part of the SB energy is compensated by the energy W1PV45 generated by the PB at this time -W1PV45=WPV45nBnC (by t1 the SB is practically charged, and the PB energy is not used).
We accept the LF night load PLN=P14/3. Then
0.4(t6 — ti)PLN+1.5[Pl(Î2 — ti) + P56(t6 —15)]+ PL(t5 —12) 0.4(t6 — ti)PLN+0.4Wbc + 1.0WBC45 + PL(t5 — ¿4) —W1PV45 ' The values of the index kE at WPVC = const (WPVC is the total energy generated by the PB) decrease with increasing WB. For example, with WB = 1828 W h -kE=3.45, with an increase in capacitance by 1.5 times (WB=2742 Wh) - kE=3.4, with an increase in capacitance by 2 times (WB=3656 Wh) - kE=3.17. When WB = 2742 W h values Q*2=Q*4 i.e. there is a balance between the energy generated by the PB and the consumed by the load. With a smaller SB capacitance, the values Q*2<Q*4, and there is an excess of PB energy after 11:00, which leads to a SB charge Q*>80 % and the need to regulate the energy supply from the PB, i.e. the energy of the PB is underused. With a larger SB capacitance, the situation is reversed and Q*4<Q*2. With close values of kE for Wb=1828 Wh and Wb=2742 Wh, it should be borne
_ 0.4(t6 — t1)PN+1.W2 —1 +Pj(fe —t5)]+P3fe —¿2) +P34(t5 —3) . 0.4(Î6 —Î1)PLN+0.4WBc+WBct5+P45(t5 —¿4)—W1PV45
According to the calculation results, we have the same situation with respect to kE at different values of the SB capacitance: WB=1828, 2742, 3656 kWh, when kE=5.47, 5.15, 4.73. Based on the possibility of providing close to PAVD values of P23 and P56 with an increase in P12, it is preferable to use a SB with WB=2742 kWh. Also, the SB must be able to accept the energy of the PB. Typically allowable charging current is /B<(0.2^0.3)CB. The maximum value of the PB power for a clear day in June [18] is PPVmax=0.76PPV. The SB with WB=2742 Wh is capable of accepting the excess energy of the PB at Pl>0.42Ppvad and current Ib=0.2Cb, and at Ib=0.277Cb -even at PL=0. Thus, PPV: WB=1:2.74 is accepted.
Solar generation depends on the season of the year and weather conditions, and the recommended load also needs to be changed. At the first stage of development, three Plr(î) scenarios were considered, which are determined by the ratio w=WPVP/WPVmax (WPVP is the energy generated by the PB for the current day according to the forecast, WPVmax is the maximum value of the PB generation according to statistical data on a clear summer day). With PB power of 1 kW in the conditions of Kyiv [18], according to WPVmax, the average power value per 24 hours is Ppvc~250, for a day Pavd=500 W. If w>0.7, the load scenario CS1 is selected, if 0.7>w>0.4 - CS2, if 0.4>w - CS3.
The CS1 scenario is focused on the maximum use of the energy of the PB and the SB during peak hours and in the daytime (Fig. 2), the calculation of power values is discussed above.
The goal of the CS2 scenario is to increase the average load power during the day, with a certain increase in the duration of the bypass in the evening from 17:30 (t4) to 20:00. In this case, we have three stages of load change: (¿1, t3=16.30), (t3, t5), (t5, te). We take the value P35=0.7P13, Q*4=62 %, Q*1=95 %. The energy consumed in the interval (t1, t4) by the load is WL14=P13(t3--Î1)+0.7P13(Î4-Î3)=P13[(Î3-Î1)+0.7(VÎ3)]. The value
WL14=nc-WPV14+WB1A. The value of P12 is also determined from the condition AQ* 12<40 %. Accordingly, the condition P13<P12 must be satisfied. The P56 value is determined by (1). Thus, until 17:30, 1/3 of the SB energy (AQ*14=33 %) is used, which is added to the PB energy, which, at w=0.7, allows providing an average value of the load power at the level of 0.8PAVD, at w=0.4 - at the level of 0.5Pavd.
In the CS3 scenario, bypass prevails, and it is realistic to exclude energy consumption from the DN during peak hours due to the SB power. Load schedule is similar to CS2. It makes no sense to take Pm by calculation, since this value is small. Therefore, the value of P12 is taken at the level of P12=0.5PAVD with a conscious increase in the cost of paying for electricity. Also, in the daytime, a restriction of Q*>60 % is introduced, since in the event of an emergency shutdown of the DN with low PB generation, it is possible to remain without electricity at night. The P56 value is determined by (1). On a clear winter day at w~0.3 in the midday hours, PPV exceeds the value of PAVD, therefore, the PVS operates autonomously at this time.
The basis for the formation of the recommended load schedule is the meteorological forecast, which can change during the day, the predicted values of PPVP differ from the actual generation PPVF. The values that the actual LF load can take have a certain discreteness and cannot exactly correspond to the recommendations; the peculiarities of the LF functioning should also be taken into account. Thus, there is a need for current adjustment of the recommendations, which is possible based on the values of PPVP and PPVF, as well as the actual degree of charge in Q*F.
The averaged value of PPVPi is used in the 5-10 min interval preceding the time t when the correction is made. The value p,=PPVF/PPVPi (PPVPi is the value at the moment t) is determined. The value of the PB energy at the corresponding stage PLR(t), starting from t, is taken equal to WPV=p iWPVPi (WPVPi is the predicted value). The PLR(t) value is recalculated according to the expressions (1)-(3) corresponding to the current stage of the schedule. At the next correction point ti+1, the procedure is similar.
To test the effectiveness of this method, a simulation model of energy processes in a PVS was developed using a Correction Load Unit (CLU). Here, it is possible to adjust the LF load in a step-by-step mode with the use of a programmable pause and subsequent continuation of the simulation. The correction step is 0.5 h. The CLU includes a set of sampling-storage devices with a 0.5 h step for measuring the values of PPVF, PPVP, Q*F. PLR(t) is calculated using expressions corresponding to the selected scenario.
WPVP values for the time intervals corresponding to the correction points 7:00, 7:30,... 20:00 (for example, for point 7:30 in the CS1 scenario this is the interval (7:30 - 11:00)), the duration of the intervals (in this case, t2-t,= 3.5 h), as well as the control values Q*c by intervals (in this case Q*C=Q*2=55 %) are set in tabular form. Also, the dependencies PPVP(t) are set in tabular form (according to archival data [18] for Kyiv city at PPVr=1 kW),
PPVF(t) and PLR1(t) is the recommended load schedule according to the forecast at the moment ti=7:00. The PL correction is carried out directly in the PLRi(t) table. The kE value per 24 h is also calculated.
The SB model is made according to the catalog data. SB charge taking into account energy losses
Q= Qs +|I 1b ■ dt, where QS is the initial value,
I1B = IB^B - when charging and I1B = IB/^B - when discharging the SB. The IB value is formed in accordance with the SB charging characteristics [17] in the form of IB(Q*). In the discharge mode, a limitation of the permissible value of IBrcmax was introduced for 30 min. This is implemented using an adjustable limit, when the upper limit is set as IB(Q*), and the lower limit is IBrcmax. The SB voltage is also set as the UB(Q*) dependence. IB value in the stand-alone mode
Ib =
K2PPV 'VC - PL UB
where K2 is the relay stay (K2=1, when the relay is on, and K2=0, when the relay is off).
Since Q*<80 % is maintained in stand-alone mode, there is no current IB limitation.
When connected to the network (bypass)
PPV -VC + Pga -VC
Ib =-
Ub
where Pga is the power consumed from the network when PPV is not enough to provide SB charging at the specified current value, subject to the limitation specified as IB(Q*).
The K2 value is set by the relay element, which switches to the state: K2=0 at PPVnc>PL and condition Q*>80 %; K2=1 at PPVnc<PL and condition Q*<75 %. The load is connected to the network provided that 20:00>t>12:00, Ppv^c<Pl, Q*<Q*d, and also regardless of other factors at Q*<50 %.
Simulation results in Matlab. The use of correction on the interval (t1, t3) is considered. Under ideal conditions for a clear day in June (PPVF(t)=PPVP(t)) and load Pwi(t), the value kE=5.63. With a slight discrepancy between the values of PPVF(t) and PPVP(t) at separate time intervals (Fig. 3) and PLRi(t), the value kE=4.91. In the case of load power correction (Fig. 3) kE=5.302.
At the value PpVi(t)=0.9PpVp(t) and PLR1(t) kE=4.01. When using correction (Fig. 4) kE=5.098. In the case when the actual generation is higher PPVF(t)>PPVP(t), there is no special need for correction, since kE increases, with PPVF(t)=1.15PPVP(t) the value kE=5.735. When using correction kE=6.206.
At present, an experimental setup has been manufactured on the basis of the Axioma Energy ISMPPT 3000 hybrid inverter (with a built-in MPPT controller for connecting a PB and a SB charge controller) with a developed software module for power consumption control. The first tests with PB in CS3 conditions (according to weather conditions of early December) were carried out, which are planned to be continued (without waiting for summer) on an experimental setup using an electronic PB emulator.
Conclusions. The key to ensuring the efficiency of PVS is the use of software and hardware complexes that provide power consumption control, as well as recommendation functions with their correction. In this case, the subject-consumer is provided with information that allows organizing energy consumption for the day ahead while reducing the cost of paying for electricity. In the absence of a response to the recommendations, the system remains operational, eliminating the SB discharge below the set level. The approaches discussed in the article can be also used with a different schedule of tariff zones during peak hours.
The proposed solutions can be used in the development of new and modernization of existing PVS with hybrid inverters when connected to a network with multi-zone tariffication. The considered solution does not imply parallel operation of the inverter with the DN. This limits the possibilities of its application, since the load power depends on solar generation. Therefore, the further development of the work is to improve the principles of PVS implementation using a network inverter, when the load is supplied from the AVI and DN.
Conflict of interest. The authors declare that they have no conflicts of interest.
REFERENCES
1. Pro vnesennia zmin do deiakykh zakoniv Ukrainy shchodo udoskonalennia umov pidtrymky vyrobnytstva elektrychnoi
enerhii z alternatyvnykh dzherel enerhii: Zakon Ukrainy vid 21 lypnia 2020 r. № 810-IX [On amendments to some laws of Ukraine regarding the improvement of conditions for supporting the production of electrical energy from alternative energy sources: Law of Ukraine of July 21, 2020 No. 810-IX]. Available at: https://zakon.rada.gov.ua/laws/show/810-20#Text (Accessed 15.08.2020). (Ukr).
2. Rao B.H., Selvan M.P. Prosumer participation in a transactive energy marketplace: a game-theoretic approach. 2020 IEEE International Power and Renewable Energy Conference, Karunagappally, India, 2020, pp. 1-6. doi: https://doi.org/10.1109/iprecon49514.2020.9315274.
3. Moira L. Nicolson, Michael J. Fell, Gesche M. Huebner. Consumer demand for time of use electricity tariffs: A systematized review of the empirical evidence. Renewable and Sustainable Energy Reviews, December 2018, vol. 97, pp. 276289. doi: https://doi.org/10.1016/i.rser.2018.08.040.
4. Shavelkin A., Shvedchykova I. Management of generation and redistribution electric power in grid-tied photovoltaic system of local object. Technical electrodynamics, 2020, no. 4, pp. 55-59. doi: https://doi.org/10.15407/techned2020.04.055.
5. El-Hendawi M., Gabbar HA., El-Saady G., Ibrahim E.-N. A. Optimal operation and battery management in a grid-connected microgrid. Journal of International Council on Electrical Engineering, 2018, vol. 8, no. 1, 195-206. doi: https://doi.org/10.1080/22348972.2018.1528662.
6. Hybrid Grid Inverter Growatt Hybrid 10000 HY. Available at: https://alfa.solar/ru/gibridnyj-setevoj-invertor-growatt-hybrid-10000-hy-id494.html (Accessed 15.08.2020). (Rus).
7. Conext SW. Hybrid Inverter. Available at: https://www.se.com/ww/en/product-range-presentation/61645-conext-sw/ (Accessed 15.08.2020).
8. ABB solar inverters. Product manual REACT-3.6/4.6-TL (from 3.6 to 4.6 kW). Available at: https://www.x-wm. it/wp-content/uploads/2017/03/REACT-3.6 4.6-TL-Product-manual-EN-RevBM0000025BG.pdf (Accessed 15.08.2020).
9. Shvedchykova I.O., Kravchenko O.P., Romanchenko J.A., Kozakov E.V. Development of a database for predicting the solar generation in the software and technical complex for the management of electrical supply of the local object. Scientific papers of Donetsk National Technical University. Series: Electrical and Power Engineering, 2020, no. 1 (22), pp. 55-61. (Ukr). doi: https://doi.org/10.31474/2074-2630-2020-1-55-61.
10. Kaplun V., Shtepa V., Makarevych S. Neuro-network model for providing electricity generation by renewable sources in energy management system of local object. Power Engineering: Economics, Technique, Ecology, 2019, no. 2, pp. 27-39. (Ukr). doi: https://doi.org/10.20535/1813-5420.2.2019.190002.
11. Naderipour A., Abdul-Malek Z., Zahedi Vahid M., Mirzaei Seifabad Z., Hajivand M., Arabi-Nowdeh S. Optimal, Reliable and Cost-Effective Framework of Photovoltaic-Wind-Battery Energy System Design Considering Outage Concept Using Grey Wolf Optimizer Algorithm - Case Study for Iran. IEEE Access, 2019, vol. 7, pp. 182611-182623. doi: https://doi.org/10.1109/access.2019.2958964.
12. Forecast. Solar. Available at: https://forecast.solar/ (Accessed 15.08.2020).
13. Traore A., Taylor A., Zohdy M., Peng F. Modeling and Simulation of a Hybrid Energy Storage System for Residential Grid-Tied Solar Microgrid Systems. Journal of Power and Energy Engineering, 2017, vol. 5, no. 5, pp. 28-39. doi: https://doi.org/10.4236/jpee.2017.55003.
14. Barelli L., Bidini G., Bonucci F., Castellini L., Castellini S., Ottaviano A., Pelosi D., Zuccari A. Dynamic Analysis of a Hybrid Energy Storage System (H-ESS) Coupled to a Photovoltaic (PV) Plant. Energies, Feb. 2018, vol. 11, no. 2, p. 396. doi: https://doi.org/10.3390/en11020396.
15. Shavolkin O., Shvedchykova I., Demishonkova S. Simulation model of the photovoltaic system with a storage battery for a local object connected to a grid with multi-zone tariffication. 2020 IEEE 7th International Conference on Energy Smart Systems (ESS), Kyiv, Ukraine, 2020, pp. 368-372. doi: https://doi.org/10.1109/ess50319.2020.9160112.
16. Lawder M.T., Suthar B., Northrop P.W.C., De S., Hoff C.M., Leitermann O., Crow M.L., Santhanagopalan S., Subramanian V.R. Battery Energy Storage System (BESS) and Battery Management System (BMS) for Grid-Scale Applications. Proceedings of the IEEE, Jun. 2014, vol. 102, no. 6, pp. 10141030. doi: https://doi.org/10.1109/jproc.2014.2317451.
17. DG12-100 (12V100Ah). Available at: https://www.ritarpower.com/products/176.html (Accessed 15.08.2020).
18. Photovoltaic geographical information system. Available at: https://re.jrc.ec.europa.eu/pvg tools/en/tools.html#SA (Accessed 15.08.2020).
Received 08.11.2020 Accepted 14.02.2021 Published 05.04.2021
A.A. Shavelkin1, Doctor of Technical Science, Professor, J. Gerlici 2, Professor, Dr. Ing.,
I.O. Shvedchykova1, Doctor of Technical Science, Professor, K. Kravchenko1, PhD, H.V. Kruhliak1,
1 Kyiv National University of Technologies and Design, 2, Nemirovich-Danchenko Str., Kyiv, 01011, Ukraine, e-mail: ishved89@gmail.com
2 University of Zilina,
Univerzitna 8215/1, SK 01026 Zilina, Slovak Republic, e-mail: juraj.gerlici@fstroj.uniza.sk kateryna.kravchenko@fstroj.uniza.sk
How to cite this article:
Shavelkin A.A., Gerlici J., Shvedchykova I.O., Kravchenko K., Kruhliak H.V. Management of power consumption in a photovoltaic system with a storage battery connected to the network with multi-zone electricity pricing to supply the local facility own needs. Electrical Engineering & Electromechanics, 2021, no. 2, pp. 36-42. doi: 10.20998/2074-272X.2021.2.06.
