Compensation/consumption hierarchical control strategy based on wind-solar-hydrogen coupling grid connection

LÜ Xin， BAI Zhifeng， CHE Jiangxuan， REN Bing， WANG Juan， RUAN Xiaoguang

(1. Shaanxi Key Laboratory of Nanomaterials and Nanotechnology，Xi’an University of Architecture and Technology， Xi’an 710055， China；2. Xi’an Key Laboratory of Clean Energy， Xi’an University of Architecture and Technology， Xi’an 710055；3. School of Mechanical and Electrical Engineering， Xi’an University of Architecture and Technology， Xi’an 710055, China；4. Shaanxi Applied Physics and Chemistry Research Institute， Xi’an 710061, China)

Abstract： In the process of grid-connected wind and solar power generation, there are problems of high rate of abandoning wind and light and insufficient energy. In order to solve these problems, we construct a grid-connected wind-solar hydrogen storage (alkaline electrolyzer(AE)-hydrogen storage tank-battery-proton exchange membrane fuel cell(PEMFC)) coupled system architecture. A grid-connected compensation/consumption hierarchical control strategy based on wind-solar hydrogen coupling is proposed. During the grid-connected process of wind and solar power generation, the upper-level control allocates power reasonably to the hydrogen energy storage system by dispatching the power of wind and solar power generation. At the same time, the control strategy ensures that the pressure of the hydrogen storage tank is within the safety range limit, and the lower control completes the control of the duty cycle of the converter in the system. Due to the randomness of wind and light, the hydrogen energy storage system is divided into three working conditions, namely compensation, balance and consumption, and five working modes. The simulation results show that the hydrogen energy storage system compensates for 40% of the power shortage, and consumes 27.5% of the abandoned wind and solar energy, which improves the utilization rate of clean energy.

Key words： compensation； consumption； hierarchical control strategy； wind-solar-hydrogen coupling； power control

0 Introduction

With the intensification of the energy crisis and the worsening of environmental pollution, the use of efficient and clean energy has become an objective requirement for current social development[1-3]. Compared with traditional fossil energy, new energy sources such as wind and solar have developed rapidly because of their characteristics of pollution-free, clean, and high-efficiency[4]. However, the high proportion of new energy sources connected to the power system increases the burden of system regulation. The conventional power supply needs to follow the load change, also needs to balance the output fluctuation of the new energy source[5]. When the output of new energy exceeds the adjustment range of the system, the output must be controlled to ensure the dynamic balance of the system, which will result in the abandonment of wind and light[6]. To make full use of wind and solar resources, a hydrogen energy storage system is added to the existing wind and solar farms. Hydrogen energy storage has the characteristics of large storage capacity, high efficiency, and high energy density. Therefore, the combination of the two can largely solve the problems of peak fluctuating difficulties, abandoning wind, and abandoning light[7-8].

Scholars have achieved certain results in the study of wind-solar storage systems. For hybrid renewable energy off-grid systems, Benlahbib et al.[9]proposed an energy management strategy for island-type wind power photovoltaic, proton exchange fuel cell, battery state of charge, and conducts experimental verification. Yang et al.[10]proposed a multi-mode power coordination strategy for AC-DC hybrid microgrid while considering the changes of sub-network operating state and energy storage state of charge (SoC). The purpose is to improve the reliability and utilization efficiency of the AC-DC hybrid microgrid. Kamel et al.[11]combined photovoltaics, proton exchange membrane fuel cells, supercapacitors and batteries to design a new hybrid energy management system. The hydrogen consumption of proton exchange membrane fuel cell is reduced by 19.6%, and the battery SoC is increased by 5.4%. Kafetzis et al.[12]proposed an island energy management strategy based on renewable energy to solve the problem of islands and remote areas relying on fossil fuels. Yun et al.[13]took the new energy power generation system composed of solar thermal power generation, photovoltaic power generation and energy storage system as the research object, and proposed a multi-objective optimization strategy of the photovoltaic storage power generation system to solve the problems of new energy generation and consumption.

In a hybrid renewable energy grid-connected system. Kong et al.[14]presented a strategy for hybrid power generation system based on hydrogen storage and supercapacitors, which realizes the friendly grid connection of hybrid power generation system. Cai et al.[15]put forward a control strategy for hydrogen storage-based photovoltaic power generation system with the goal of grid integration. To improve the penetration rate of wind energy, stabilize the DC bus voltage, and smooth the grid power, Abadlia et al.[16]proposed an adaptive fuzzy control grid-connected solar-hydrogen hybrid power generation system based on genetic algorithm optimization. Gu et al.[17]proposed the wind-solar complementary coupling evaluation index, and established an output scenario model based on the demand for peak shaving. This model optimizes the peaking operation when a high percentage of new energy is connected to the grid.

Based on the above analysis, the current research mainly focuses on the wind-hydrogen coupled off/on-grid system, or the light-hydrogen coupled off/on-grid system. The research on wind-solar hydrogen coupling grid-connected control is mainly to solve the problems of abandoning wind and light, voltage fluctuation, and other problems in the process of wind-solar grid connection. However, it ignores the problem of power shortage when the wind and light resources are insufficient. In this work, a hydrogen storage system is coupled based on the existing wind-solar electric field. A grid-connected wind-solar hydrogen storage (alkaline electrolyzer(AE)-hydrogen storage tank-battery-proton exchange membrane fuel cell(PEMFC) coupled system architecture is constructed, and a grid-connected compensation/consumption hierarchical control strategy based on wind-solar hydrogen coupling is proposed. When the power of the wind and solar is greater than the dispatching power of the grid, the electrolyzer and the battery can consume the power difference. When the power of the wind and solar is less than the grid dispatch power, PEMFC and the battery compensate for the power difference. This control strategy realizes the compensation/consumption of grid-connected power in a large-scale wind-solar-hydrogen coupling system. Finally, the proposed control strategy is simulated and verified by Matlab/Simulink.

1 System specification

When wind power and photovoltaics are connected to the grid, due to the randomness and intermittency of wind and solar resources, large-scale abandonment of wind and solar energy or energy shortage will occur. The wind-solar-hydrogen coupling compensation/consumption layered control strategy proposed in this work can effectively solve the problems of abandoning wind and solar energy and power shortage. The control strategy can effectively improve the utilization rate of clean energy.

2 Control model

The mathematical models of wind turbines, photovoltaic arrays, and hydrogen energy storage systems are detailed in Refs.[3,18-19]. This article will not go into detail. Fig.1 shows the topology of wind-solar hydrogen including wind turbine unit, photovoltaic unit, and hydrogen energy storage unit. The wind turbine unit rectifies the alternating current into direct current through the AC/DC converter and collects it on the DC bus. The DC/DC converter at the photovoltaic unit side collects the DC power to the DC bus through the Boost booster circuit. The battery is connected to the DC bus through the Buck-Boost circuit. The DC bus is connected to the electrolyzer through the Buck circuit. The hydrogen produced by the electrolyzer is stored in a hydrogen storage tank. When the PEMFC is required to start, the hydrogen in the high-pressure hydrogen storage tank is delivered to the PEMFC. The electricity of the PEMFC is collected in the DC bus through the DC/DC conversion of the Boost booster circuit, and the electricity collected in the DC bus is sent to the power grid through the DC/AC inverter.

Fig.1 Structure diagram of wind-solar hydrogen storage coupling

2.1 Control strategies for lower layer units

The lower layer control units are controlled by the upper layer control units based on the system requirements for the purpose of stable and smooth Internet access. The control strategy block diagram of the lower unit of wind-solar hydrogen storage is shown in Fig.2.

2.1.1 Control strategy for wind turbine unit

For the wind power control unit, as shown in Fig.2, it is mainly controlled and adjusted by the machine-side inverter, and thed-axis adopts single-loop control. The current reference valueiw,d,refis 0. The error betweeniw,d,refandiw,dgenerates the voltage control signal through the PI controller. The voltage control variable interacts with theq-axis voltage coupling itemωLw,qiw,qto generate thed-axis pulse signalmw,d. Theq-axis adopts a double-loop control method, that is, power outer loop control and current inner loop control. We can find out the power reference value corresponding to the rotor angular velocityωby the look-up table methodPw,q,ref. The error value betweenPw,q,refand the actual powerPwgenerates the current reference valueiw,q,refof theq-axis through the PI controller. The difference betweeniw,q,refandq-axis actual currentiw,qproduces voltage control through the PI controller. The voltage control variable and the no-load electromotive forceωφof the permanent magnet synchronous motor and the voltageωLw,diw,dof thed-axis generate the pulse signalmw,qof theq-axis under the joint action.

Under thedqcoordinates, the control equation of the wind power control unit is expressed as

(1)

(2)

whereφis permanent magnet flux linkage;Udcis DC bus voltage;Lw,dis the filter inductance of thed-axis machine side;Lw,qis the filter inductance of theq-axis machine side;kp1andki1are expressed as the proportional coefficient and integral coefficient of theq-axis current inner loop PI controller, respectively;kp2andki2are expressed as the proportional coefficient and integral coefficient of theq-axis voltage outer loop PI controller, respectively;kp3andki3are expressed as the proportional coefficient and integral coefficient of thed-axis PI controller, respectively.

2.1.2 Control strategy for PV array

The photovoltaic (PV) array adopts the variable step perturbation observation method for the maximum point power tracking (MPPT). Through voltage disturbance, the PV output can reach its corresponding maximum output power under any light intensity. According to the characteristics of voltage-current, voltage-power, and MPPT, the control equation is expressed as

(3)

wherekppvandkipvare the proportional and integral gains of the PI controller of MPPT, respectively;Umppis the voltage of maximum power point;Upvis the terminal voltage of PV array; andmpvis the duty cycle of the trigger pulse of the photovoltaic array unit converter.

2.1.3 Control strategies for battery, PEMFC, AE

The reference powersPb,ref,Pf,refandPe,ref(Pb,ref,f,ref,e,ref) of the battery, PEMFC, and AE are generated by the upper power control.

The reference power is compared with the actual powersPb,PfandPeto generate the reference currentsib,ref,if,refandie,ref(ib,ref,f,ref,e,ref). The error between the reference current and the actual currentsib,ifandiegenerates control signalsDb,DfandDethrough the PI controller. As shown in Fig.2, the control equation is expressed as

(4)

wherekpb1,pf1,pe1andkib1,if1,ie1are expressed as (battery, PEMFC, electrolyzer) proportional coefficient and integral coefficient of current inner loop PI controller, respectively;kpb2,pf2,pe2andkib2,if2,ie2are expressed as (battery, PEMFC, electrolyzer) proportional coefficient and integral coefficient of power outer loop PI controller, respectively.

2.1.4 Control strategy for grid-connected unit

For the grid-connected control unit, the control block diagram is shown in Fig.2. Theq-axis uses the current single-loop control mode. Theq-axis current reference valueig,q,refis the ratio of the reactive powerQg,refto the actual grid-side voltageUdc. The error value between the actualq-axis currentig,qandig,q,refgenerates the voltage control quantity via the PI controller. Theq-axis pulse signalmg,qis generated under the combined action of the voltage control variable, thed-axis coupling termωgLgig,q, and the DC bus voltageUdc. Thed-axis uses the dual-loop control mode, that is, the current inner loop voltage outer loop. The error value between the reference voltageUdc,refandUdcof the DC bus through the PI controller to generate thed-axis current reference valueig,d,ref. Grid-side currentsig,dandig,d,refgenerate voltage control quantities via the PI controller. Thed-axis pulse signalmg,dis generated under the combined action of the voltage control quantity and the grid-side voltageUg,d, and theq-axis coupling item -ωgLgig,d. The control equations are expressed as

(5)

(6)

whereωgis the grid synchronization angular frequency;Lgis the filter inductor on the grid side;kpr1andkir1are expressed as the proportional coefficient and integral coefficient of the grid-sideq-axis PI controller, respectively;kpr2andkir2are expressed as the proportional coefficient and integral coefficient of the grid-sided-axis voltage outer loop PI controller, respectively;kpr3andkir3are expressed as the proportional coefficient and integral coefficient of the grid-sided-axis current inner loop PI controller, respectively.

2.2 Upper-level control strategies

The upper-level power control working mode switch of the wind-solar hydrogen multi-energy system is shown in Fig.3.

Fig.3 Upper-level power control energy management strategy

According to the characteristics of wind and solar power generation and the power dispatching requirements of the power grid, the wind and solar hydrogen storage system can be summarized into three working stages to discuss: hydrogen energy storage compensation stage, hydrogen energy storage consumption stage, and power balance stage.

According to power balance equation

Pr(t)=Pw(t)+PP(t)-Pl(t),

(7)

wherePr(t),Pw(t),PP(t),Pl(t) are expressed as the remaining power at timet, wind turbine power, photovoltaic power, and grid power demand, respectively, the power abandonment rate and loss of power supply possibility of the wind-solar hydrogen storage system in a certain period can be expressed as[20]

(8)

Whenη>0, it means that there is too much wind and solar power generation, which needs to be consumed by the hydrogen energy storage system. Whenη<0, it means that the wind and solar power generation are insufficient, and the hydrogen energy storage system needs to compensate for it. Whenη=0, the wind and solar power generation meets the grid demand.At the consumption stage of hydrogen energy storage system, PEMFC is on standby, that is,

Pf=0,

(9)

wherePfdenotes the PEMFC power generation.

a) When hydrogen storage tank pressure meets

Pqc>Pmax,

(10)

wherePqcis the pressure of hydrogen tank,Pmaxis the maximum pressure limit of hydrogen tank; and the power of electrolyzer meets

PrPemax,

(11)

the operating status of electrolyzer is

Pe=0,

(12)

wherePeis the working power of electrolyzer,Peminis the minimum working power limit of electrolyzer, andPemaxis the maximum working power limit of electrolyzer; and the operating status of the battery is

(13)

wherePbcis the operating power of the battery;Uris the voltage corresponding toPr,Umaxis the maximum operating voltage of the battery, andPbceis the rated charging power of the battery.

b) When the hydrogen tank pressure meets

Pqc

(14)

the operating status of theelectrolyzer is

(15)

wherePeeis the rated power of electrolyzer.

At the compensation stage of hydrogen energy storage system, the electrolyzer is on standby, that is,

Pe=0.

(16)

a) When hydrogen storage tank pressure meets

Pmin>Pqc,

(17)

wherePminis the minimum pressure of the hydrogen tank, and the power meets

Pf min>|Pr|,

(18)

wherePf minis the minimum power limit for PEMFC work, the operating status of the battery is

(19)

wherePbdis the discharge power of the battery, andPbdeis the rated discharge power of the battery.

b) When the hydrogen tank pressure meets

Pmin

(20)

and PEMFC power meets

Pfmin<|Pr|,

(21)

the operating status of the PEMFC is

(22)

wherePfis the rated working power of PEMFC.

At the balance stage of hydrogen energy storage system, wind and PV power generation and grid electricity demand are equal, namely

Pr=Pw(t)+PP(t)-Pl(t)=0,

(23)

at this moment, we have

Pe=Pb=Pf=0,

(24)

(25)

The consumption and compensation range of the hydrogen energy storage system depends on the wind and solar power abandonment rate of the wind and solar farm and the capacity of the hydrogen energy storage configured. According to this control strategy, the consumption and compensation range can be defined as

(26)

wherePxandPbbare the consumption power and the compensation power, respectively.

According to the characteristics of randomness and intermittence of wind and light, combined with the upper-level control strategy in Fig.3, the wind-solar hydrogen storage system is classified into three working conditions: hydrogen energy storage and consumption, hydrogen energy storage compensation, and balance, and five working modes, as shown in Table 1.

Table 1 Operation mode of wind-solar hydrogen storage system

3 Simulation

The parameters are set as follows: PV rated power, 40 kW; wind power rated power, 100 kW; electrolyzer rated power, 30 kW (Pemin=20 kW,Pemax=50 kW); PEMFC power rated power, 50 kW (Pfmin=20 kW,Pfmax=70 kW); battery rated power, 15 kW (Pbmin=1 kW,Pbmax=20 kW); the hydrogen storage tank capacity, 100 N·m3; the power grid demand, 100 kW; the DC side voltage, 800 V; the grid side voltage, 380 V; and the frequency, 50 Hz.

The change curve of wind speed and solar irradiation within 15 s is shown in Fig.4. The wind speed changes within the range of 7 m/s-12 m/s, and the solar irradiation changes within the range of 700 W/m2-1 300 W/m2. The range of change is large and random.

Fig.4 Change curves of wind speed and solar irradiation

The output power curve corresponding to the wind power PV power output following the change of wind speed and light intensity is shown in Fig.5.

Fig.5 Wind power PV grid electricity demand

The output power of wind power isPw. Following the wind speed, its output power varies in the range of 40 kW-100 kW. The photoelectric output power isPp, and the output range varies within 15 kW-40 kW. The total wind and solar power generationPallvaries within 60 kW-140 kW, and the grid electricity demandP1is 100 kW. During the first 0.15 s of the start-up stage of the wind and solar system, the overall system fluctuates to a certain extent. In 0.15 s-3 s, the wind, and solar power generation are insufficient, the system is short of 8.8 kW. In 3 s-6 s, the wind, and solar power generation are seriously insufficient, and the system is short of 40 kW. In 6 s-9 s, the wind and solar power generation, and grid electricity demand balance. In 9 s-12 s, the wind and solar power generation are greater than grid demand, and hydrogen energy storage consumption is 10.6 kW. In 12 s-15 s, the wind and solar power generation is greater than grid demand, hydrogen energy storage consumption is 38.5 kW. According to Eq.(8) and Table 1, the working modes of hydrogen energy storage and the valuePrare shown in Table 2.

Table 2 System operating conditions

The wind and solar power generation is in the state of shortage in the first 6 s, and the state of abandoning the wind and abandoning the light in the last 6 s. The hydrogen energy storage system correspondingly solves the problems of power shortage and consumption in the wind and solar system. As shown in Fig.6, the system is short of -8.8 kW in 0-3 s.

Fig.6 Change curve of wind-solar hydrogen storage power

The battery starting conditions are met (Pbmin=1 kW,Pbmax=20 kW). The battery starts in 0.2 to compensate for the system shortage. In 3 s-6 s, the PEMFC start-up conditions are met (Pfmin=20 kW,Pfmax=70 kW). In the hydrogen energy storage system, the PEMFC starts to supply power to the system. In 5.9 s-6.1 s, wind power generation and grid electricity demand load tend to balance, andPfmeets the battery start-up condition. In 6.1 s-9 s, the wind and solar power generation are balanced with the power demand of the grid, and the hydrogen energy storage system is in a standby state. In 9 s-15 s, the total wind and solar power generation are greater than the grid demand, and the energy storage system is required to consume the excess electricity. The state of charge of the battery is satisfied in 9 s-12 s. In 12 s-15 s, the conditions of electrolysis of water for hydrogen production are met (Pemin=20 kW,Pemax=50 kW). The electrolyzer electrolyzes water to produce hydrogen, and stores the produced hydrogen in a hydrogen storage tank.

The battery is divided into two stages: charging and discharging, as shown in Fig.7.

Fig.7 Battery operation curve

Before 0.15 s, the battery does not work, the SoC is set to an initial value of 45%, and the current, voltage, and power are all 0. In 0.15 s-3 s, the wind and solar power generation is insufficient, and the shortfall power is 7.8 kW, which meets the battery discharge conditions. In the first 0.15 s, the wind power PV system starts, the battery reference power value suddenly changes to 20 kW, and the current increases accordingly. In the following 0.25 s-3 s, the battery operating powerPbfollows the reference powerPbreffor normal operation, the SoC drops from 45% to 44.998 5%, and the voltage stabilizes within 241 V-242 V, the current is stable at 33 A. In 3 s-9 s, the battery working conditions are not met, so the SoC remains at a level. In 5.9 s-6.1 s, when the system shortage is greatly reduced, the battery start-up conditions are met, so the battery fluctuates within 0.2 s. In 9 s-12 s, the battery charging conditions are met, and the SoC curve protrudes outward during the increase process. This is because the battery starts to work in 9.1 s-9.3 s, and the battery power increases rapidly during the startup process. In 9.3 s-12 s, the battery SoC rises smoothly, and the current and voltage are stable. In 12 s-15 s, battery does not work.

The operating curve corresponding to the PEMFC is shown in Fig.8. The PEMFC reference powerPf,refand the PEMFC actual power generation powerPftrack well after 3.15 s. In 3 s-6 s, the PEMFC starts up, and the corresponding current, voltage, and hydrogen flow rate change within 3 s-6 s, respectively. The current fluctuates up and down in the range of 80 A, the voltage fluctuates in the range of 540 V, and the hydrogen consumption flow rate during this time period is 413 slpm.

Fig.8 Operation curve of PEMFC

In 12 s-15 s, due to the excessive wind power photovoltaic power generation, which is greater than the maximum charging power of the battery, the excess electricity is consumed by the electrolyzer, as shown in Fig.9. The reference powerPerefis 38.5 kW, the electrolyzer responds quickly to start within 0.1 s, and the electrolysis powerPeof the electrolyzer reaches the reference power value in 12.3 s. Then water is electrolyzed to produce hydrogen according to the reference powerPe,ref. The corresponding current and voltage are 139 V and 277 A, respectively, and the hydrogen production rate of electrolysis of water is 91 slpm.

Fig.9 Operation curve of electrolyzer

The hydrogen storage tank provides fuel for the PEMFC, and the hydrogen produced by electrolysis of water is stored in the hydrogen storage tank. The operating curve of the hydrogen storage tank is shown in Fig.10. When the hydrogen storage tank has 50 N·m3of hydrogen, and the initial pressure is 831.5 Pa. In 0 s-3 s, hydrogen storage tank does not work; In 3 s-6 s, the PEMFC is started, and the total amount of hydrogen in the hydrogen storage tank decreases from 50 N·m3to 49.982 N·m3with time. The pressure drops to 831.1 Pa. In 12 s-15 s, the electrolyzer electrolyzes water to produce hydrogen, the pressure in the hydrogen storage tank rises to 831.19 Pa, and the total amount of hydrogen rises to 49.987 N·m3.

Fig.10 Operation curve of hydrogen tank

In the 15 s simulation, the hydrogen energy storage system is switched from the compensation state to the hydrogen energy storage consumption state. The working mode of hydrogen energy storage has successively experienced five working modes: battery discharge, PEMFC discharge, hydrogen energy storage standby, battery charging, and electrolyzer consumption. The simulation results show that the hydrogen energy storage system compensates for 40% of the power shortage and consumes 27.5% of the abandoned wind and solar energy.

4 Conclusions

In this work, we construct a grid-connected wind-solar hydrogen storage (AE-hydrogen storage tank-battery-PEMFC) coupled system architecture. A grid-connected compensation/consumption hierarchical control strategy based on wind-solar hydrogen coupling is proposed. Under any environmental conditions, the hydrogen energy storage system can be divided into three working conditions: compensation, balance, and consumption, and five working modes. From Matlab/Simulink simulation, we can draw the conclusions as follows:

1) In the wind-solar hydrogen coupling system, when the power variation rangePris within ±20 kW, the battery responds and tracksPrwithin 0.2 s, which effectively conpensates the change of compensation/consumption power in a small range.

2) WhenPris less than -20 kW and exceeds the storage range of the battery, the electrolyzer responds quickly and consumes 38.5 kW of electricity.

3) WhenPris greater than 20 kW and exceeds the compensation range of the battery, the PEMFC responds and starts within 0.15 s. Under the condition that the pressure of the hydrogen storage tank is satisfied, the PEMFC compensates for 40 kW of electricity.

Compared with the independent grid connection of wind and solar, this study reuses the amount of abandoned wind and abandoned light, which effectively improves the utilization rate of clean energy and further reduces the cost of wind and solar power generation.