In this paper, the hybrid photovoltaic-thermoelectric generator (PV-TEG) combined dynamic voltage restorer (DVR) system is proposed for the power quality disturbances compensation in a single-phase distribution system. The stable and precise level of input voltage is essential for the smooth and trouble-free operation of the electrically sensitive loads which are connected at the utility side to avoid system malfunctions. In this context, the hybrid PV-TEG energy module combined DVR system is proposed in this paper. With the support of the hybrid energy module, the DVR will perform the power quality disturbances compensation effectively with needed voltage and /or power. In the proposed system, the PV and TEG energy sources are connected electrically in series to produce adequate voltage for the DVR operation and the fractional factor-based variable incremental conduction (FFVINC) maximum power point tracking (MPPT) control algorithm is employed to extract the possible maximum power from the PV array. The intelligent fuzzy logic controller (FLC) is chosen for implementing the MPPT control algorithm. The half-bridge voltage source inverter (VSI) circuit and in-phase voltage compensation technique are used in the DVR for better power quality disturbances compensation. The performance and usefulness of the proposed DVR system are investigated by an extensive simulation study with four different modes of operation, the study results are confirmed that the proposed system promptly identifies the power quality disturbances for compensation. Moreover, the investigation proved that the combined PV and TEG energy module can provide better energy efficiency in converting solar irradiation into electricity.

Nowadays, sensitive electrical and electronic loads are widely used for network communication, computing, safety and monitoring in residential as well as commercial buildings. The sensitive loads which are connected in the distribution supply system are mainly affected by the power quality disturbances, which leads to various system malfunctions such as data losses, system halts, communication interference, reducing the life period of the equipment etc. To overcome such a kind of malfunctions, the voltage level of the distribution system must be maintained correctly [

The uses of renewable energy resources for electric power generation are essential to supply the required power for the increasing electricity demand and protect the environment [

The appropriate control topology is also essential for the DVR to decide the proper magnitude and phase of the injecting voltage to retain the magnitude and phase of the load voltage. Various control strategies and circuit configurations have been suggested in the literature [

In this paper, the H-bridge VSI circuit-based DVR with a hybrid PV-TEG renewable energy module is proposed for compensating the power quality disturbances of the sensitive loads connected in a single-phase distribution system. The proposed DVR system is designed that to compensate deep and long-term voltage disturbances by supplying adequate real and reactive power from the PV-TEG hybrid energy source. Also, the energy conservation mode of the DVR will conserve the customer energy consumption from the utility grid which results in the potential panel tariff for the consumers could be reduced at a reasonable level. A low power DC-DC converter is chosen to extract the maximum power generated on the PV array and a coupled inductor based high step-up DC-DC boost converter is used to amplify the DC voltage suitably for the H-bridge VSI circuit. The in-phase compensation method is employed in the DVR to mitigate the various types of voltage disturbances that occurred in the low voltage power distribution system. The simulation study of the proposed DVR system is carried out under various operating modes to demonstrate its effectiveness.

The schematic of the proposed PV and TEG combined single-phase DVR system for the power quality disturbances compensation is shown in

Parameter | Value |
---|---|

Input Capacitor | 70 μF |

Inductor | 27 mH |

Switching Frequency | 25 kHz |

Output Capacitor | 210 μF |

The PWM technique-based H-bridge VSI circuit of the DVR converts the DC-link voltage into a sinusoidal AC voltage at the required amplitude and frequency. The H-bridge VSI inverter is the most suitable inverter for the single-phase DVR system due to its simplicity. The circuit configuration of the VSI is shown in _{1} and Q_{2} are ON, the input voltage V_{dc} will be transfer to the load similarly, if the switches Q_{3} and Q_{4} are ON, –V_{dc} will be transferred to the load. The Low Pass Filter (LPF) added in this circuit will eliminate the unwanted higher-order harmonics present in the VSI output and make a pure sinusoidal waveform suitable for the load voltage compensation [

The output voltage can be expressed in the following Eqs. (2) and (3)

The controller unit for the VSI circuit is depicted in _{S}| then it is compared with the reference (V_{ref}) value to generate an error (e) signal. The proportional-integral (PI) controller produces an angle delta (δ) to reduce e to 0. The reference voltage generator produces the U_{ref} signal for the PWM generator using the angle δ, the pulse output of PWM generator is applied to VSI circuit. The detailed operation of the VSI controller can be found [

If S is turned ON, the input voltage V_{in} will charge the inductor L_{m} at the same time the coupled inductor induces a voltage on the secondary (N_{s}). The voltage (V_{L2}) of the coupled inductor allows V_{in}, V_{c1}, V_{c2} and V_{c3} to provide sufficient energy to the load in series. The load voltage V_{o} can be expressed as

where V_{c1}, V_{c2} and V_{c3} are the voltages across the capacitor C_{1}, C_{2} and C_{3}, respectively. The current flow takes in the converter circuit when the switch S is closed is illustrated in _{m} will release the stored energy through N_{s} to charge the capacitors C_{2} and C_{3}. The current flow that takes place in the converter during switch S is OFF is depicted in

The capacitors C_{2} and C_{3} voltage is given as

where D is the duty cycle, k is the coupling coefficient and n is the turns ratio of a coupled inductor. The PI controller of the control unit will produce the required gate pulses to operate the switch S. The capacitor and inductor values are chosen based on the load current and voltage drop, a detailed design procedure can be found in Hsieh et al. [

The PV cells are usually connected in parallel and series form in a PV array, the PV cell generates the electricity directly from the solar irradiation [_{pv}) can be represented using Kirchhoff’s current law,

where I_{ph} is the photocurrent in A, I_{D} is the diode current in A. The current flow through the diode can be expressed using the Shockley diode equation,

Now the

The current flow through the diode is negligible during the short-circuit condition so the current I_{pv} is approximately equal to I_{sc}. By considering the parallel and series-connected PV cells the above expression can be expressed as

where I_{sat} is the reverse saturation current in A, V_{pv} denotes voltage output in V, q denotes the electron charge in C, k represents the Boltzmann constant in JK^{–1}, T denotes the cell junction temperature in K, A denotes the ideality factor of the diode, n_{s} and n_{p} are the numbers of series and parallel-connected PV cells, respectively. The model of the PV array is developed based on the theory discussed above and the important design parameters are shown in

Specifications | Value (unit) |
---|---|

Maximum power | 150 W |

Current at maximum power | 4.35 A |

Voltage at maximum power | 34.6 V |

Open-circuit voltage | 43.6 V |

Short-circuit current | 4.85 A |

Area of the solar panel | 1480 mm × 670 mm |

Number of PV arrays | (2 × 4) 8 |

TEG converts the thermal energy directly into electricity based on the theory of the Seebeck effect. The structure of a typical TEG is shown in

where S denotes the Seebeck coefficient in V/K, T_{c} and T_{h} are the cold-end and hot-end temperatures respectively in K. The power output of the TEG module can be calculated by

where R_{int} and R_{L} are the internal resistance of the TE and load resistance in ohm, respectively.

The model of the TEG is developed based on the thermal and electrical properties of the bismuth telluride Bi_{2}Te_{3} type thermocouple and the model parameters are shown in _{teg}), load voltage (V_{T}) and output power (P_{teg}) are plotted for the temperature difference (ΔT) of 0 to 50 K as shown in

Specifications | Value |
---|---|

Length (L) | 1.6 mm |

Area (A) | 1.4 mm^{2} |

Electrical conductivity p-type (σ_{p}) |
2.18 × 105 S/m |

Electrical conductivity n-type (σ_{n}) |
0.825 × 105 S/m |

Thermal conductivity p-type (k_{p}) |
1.44 W/mK |

Thermal conductivity n-type (k_{n}) |
1.34 W/mK |

Seebeck coefficient p-type (α_{p}) |
384 µV/K |

Seebeck coefficient n-type (α_{n}) |
–634 µV/K |

Total Number of thermocouples (N_{tc}) |
110 |

The FFVINC type maximum power tracking control algorithm is proposed in this paper to overcome the drawbacks of the conventional incremental conduction (INC) MPPT control algorithm. The control strategy of the proposed FFVINC technique will use a larger step size for tracking the maximum power point (MPP) when the operating point of the PV is not closer to the optimum level in the P-V curve, the step size will be gradually reduced when the PV array operating point is approaching the peak point of maximum power level. A variable factor (β) used in this control algorithm supports the FLC to produce the variable tracking step size based on the PV operating condition. The value for the β is set between 0 to 1 using the difference in present instant voltage (V_{pv}) and previous instant voltage (V_{pv} (t–1)) of the PV array. The difference in voltages is zero then β is equal to 1 otherwise, β will be between 0 and 0.9 based on the magnitude of the voltage difference. The variable factor β will modify the error input of the FLC by which the variable tracking step size is achieved. The concept to use the variable factor for the proposed method is developed from the fractional calculus [

where γ() denotes γ function and β denotes the order of the derivative in the range of 0 < β ≤ 1. When β is 0 < β < 1, then the control scheme will be a fractional order control. Otherwise, if β = 1, the control scheme becomes a conventional integer order control. Therefore, the variable factor β can be applied in the MPP tracking scheme to achieve the variable tracking step size which leads to an enhanced MPP tracking performance in terms of speed and accuracy. Using the fractional order differentiator Eq.

where P_{pv} and P_{pv}(t–1) are the present and previous moment PV array power output in W, V_{pv} and V_{pv}(t–1) are the present and previous moment voltage output. From

Now the variation in power output to the voltage variation in unit time is represented by

Using the factor β, the

The change in power (ΔP) of the PV array in unit time is given as

The main purpose of the proposed DVR system is to maintain supply power quality for the sensitive loads to avoid malfunctions. This can be achieved by injecting a dynamically controlled voltage (V_{DVR}) generated by a forced commutated VSI, the voltage injection to the utility grid line is made using a series-connected voltage injection transformer.

The overall DVR system configuration proposed is shown in _{1}, S_{2}, and S_{3}, are illustrated in

Mode of | Grid Supply Voltage in % | Status of the switches | ||
---|---|---|---|---|

S_{1} |
S_{2} |
S_{3} |
||

Idle | 100% | ON | OFF | ON |

Compensation-voltage sag | < 100% | ON | OFF | OFF |

Compensation-voltage swell | > 100% | ON | OFF | OFF |

UPS | 0% | OFF | ON | OFF |

PV-TEG module voltage level | Status of switches | Battery Charging | |
---|---|---|---|

R_{1} |
R_{2} |
||

Normal | ON | OFF | PV Array |

Below normal | ON | ON | Grid supply and PV Array |

Zero | OFF | ON | Grid supply |

When the DVR identifies any voltage disturbances in the distribution system, the compensating voltage will be injected at the utility side using the series injection transformer with the desired phase angle, magnitude, and wave shape. In this mode, the switch S_{3} and switch S_{2} are OFF, and the DVR will start to compensate voltage sag/swell. The switch S_{1} is ON to provide the continuous supply to the load, now the voltage across the load is the sum of the voltage of the utility grid supply plus the compensation voltage supplied from the DVR system.

In ideal mode, the grid supply is at the normal level and DVR is made idle, the DC supply is blocked by turning OFF the upper and lower switches of the inverter legs. The switch S_{3} is ON to short circuit the secondary winding of the injection transformer and make V_{DVR} = 0, the switch S_{1} ON and switch S_{2} is OFF to allow the utility grid to supply the required power for the load. The PV-TEG module voltage is at normal then it will charge the battery otherwise the battery will be connected to the grid supply by closing the switch R_{2} for charging.

The proposed DVR system can also support saving power consumption from the utility grid by the consumer in the case of the hybrid PV-TEG module produces sufficient output power to meet out the load requirement. In this energy conservation mode, the utility grid supply will be disconnected, therefore the series injection transformer and the load will be in parallel such that to transfer the power from the PV-TEG module to the load via battery storage. In order to perform this operation, the switch R_{1} is ON and switch R_{2} is OFF. By activating this mode, the power utilization from the gridline can be minimized thereby the electricity tariff to the consumer will be reduced.

The proposed MPPT algorithm is developed to achieve the MPP of the PV quickly and accurately. The FLC is chosen to implement the MPPT algorithm, the control strategy is developed to use a larger tracking step size when the PV operating point is not nearer to the peak level in the P-V curve. Alternatively, the step size will be made smaller when the operating point of the PV module is closer to the peak level in the P-V curve. The dynamic variable incremental conduction is achieved by modifying the fuzzy input variable e(t) using the variable factor β. Based on error e(t) the present position of the PV operating point is identified, which is summarized as

Also, the fuzzy input signal ΔP(t) confirms the movement of the PV array operating point, the positive ΔP(t) confirms that the PV array operating point is moving towards the maximum level in the P-V curve, otherwise, the negative ΔP(t) indicates that the PV operating point is moving away from the maximum level in the P-V curve. To develop the fuzzy control rules, the range of input and output parameters are split into five fuzzy sets and each fuzzy set is represented with a specific linguistic term. The linguistic terms used for the inputs are PL: Positive Large, PS: Positive Small, ZE: Zero, NS: Negative Small, NL: Negative Large. Also, the output linguistic terms are VL: Very Large, LG: Large, ME: Medium, SM: Small, VS: Very Small. The triangular shape fuzzy sets are chosen for the fuzzy variables as shown in

u(t) | e(t) | |||||
---|---|---|---|---|---|---|

NL | NS | ZE | PS | PL | ||

ΔP(t) | NL | VS | VS | SM | ME | SM |

NS | VS | SM | SM | LG | ME | |

ZE | SM | SM | LG | ME | LG | |

PS | ME | ME | LG | VL | VL | |

PL | LG | LG | VL | VL | VL |

The performance of the VSI-based DVR system is investigated for the single-phase distribution system under different operating modes using MATLAB software. The simulation study parameters used for the present module are shown in ^{2} respectively and the temperature difference between cold and hot sides of the TEG as 20 K. Based on the DVR operational modes, the analysis and discussion are carried out with four different cases. In the initial part of the investigation, the suitability of the hybrid PV-TEG energy source for the DVR operation is verified, the power output of the hybrid energy source under changing solar irradiation is tested as shown in

Description | Parameter | Value |
---|---|---|

Utility grid supply | RMS Voltage and Frequency | 230 V, 50 Hz |

LC Filter | Inductance and Capacitance | 38 mH, 20 µF |

DC Bus | Voltage | 290 V |

RL Load | Inductance and Resistance | 0.5 mH, 120 Ω |

Storage Battery Bank | Nominal Voltage & Rating | 24 V, 300 Ah |

High Step-Up |
Inductance | Lk = 0.25 µH, Lm = 48 µH |

Switching Frequency | 25 kHz | |

Capacitance | C1 = 3.151 µF /450 V, C2 = 1.05 µF, C3 = 1.05 µF/450 V, Co = 500 µF/450 V |

Similarly, the energy conversion efficiency of the hybrid PV-TEG study module is also verified by comparing the theoretically calculated efficiency of the standalone PV array. The efficiency of the PV array is calculated based on its output power and the area of the PV panel absorbing solar irradiation [

where A_{P} denotes the area of PV panel in m^{2} and G denotes solar irradiation in W/m^{2}. Similarly, the efficiency of the hybrid PV-TEG module can be determined based on the following expression.

where P_{teg} denotes the output power of the TEG in watts. The efficiency comparison depicted in

In voltage sag compensation, the hybrid PV-TEG integrated DVR performance for the voltage drop (sag) of 20% (46 V) and 50% (115 V) at the utility grid side is studied and system response is illustrated in

The power flow during the voltage sag condition is also verified as shown in

The performance of the DVR is analyzed for the voltage swell of 15%, the voltage swell is taking place from 0.2 s to 0.3 s the actual voltage supplied from the utility grid is increased from its normal level of 230 V to 264.5 V with a phase jump of 0° as depicted in

In outage compensation, the failure in grid supply voltage will be compensated suitably by the DVR to provide the continuous supply to the sensitive load for the uninterruptable operation. The DVR will supply the full voltage of 230 V to the load with the help of the hybrid PV-TEG module power output and the energy stored in the battery. To examine the outage compensation, the grid supply voltage is made into 0 V from 0.2 s to 0.4 s as shown in

When the hybrid PV-TEG module generates sufficient or more than the load power demand, the energy conservation mode will be activated by making the utility grid supply voltage as 0 V concurrently the DVR operation is changed into the compensation mode. Therefore, the DVR will provide the total power directly to the load using the PV-TEG module output and the energy stored in the battery by which the power consumption from the utility grid is saved. The energy conservation mode operation is studied by making the grid supply voltage as 0 V, the grid supply voltage, injected voltage and load voltage of the proposed hybrid PV-TEG integrated DVR is shown in

The real and reactive power distribution during the energy conservation mode is illustrated in

A new configuration of the H-bridge VSI-based DVR system integrated with a hybrid PV-TEG energy source has been proposed for the single-phase sensitive load power quality disturbances compensation. The hybrid renewable energy source was integrated with DVR to improve the system ability for the deep and long-term voltage disturbances compensation and energy conservation for the consumer. In the simplified H-bridge VSI circuit, the in-phase voltage compensation method was employed to maintain the supply power quality of the sensitive loads to prevent malfunctions and trouble-free operation. The FFVINC MPPT control algorithm has effectively tracked the MPP of the PV array to extract the maximum power in terms of fast and accuracy, the proposed hybrid energy module by combining the TEG with the PV array gives better power conversion efficiency which is essential for the DVR during the voltage sag/swell and outage compensation. The overall performance of the developed single-phase DVR system has been tested for voltage compensation and utility grid energy conservation. The study results were proved that the PV-TEG integrated DVR configuration can be able to maintain the sensitive load supply power quality for smooth operation. The results of outage compensation mode have confirmed that the DVR system could provide uninterruptable supply for the consumer loads using the hybrid energy module suitably to prevent the system malfunctions. Moreover, the investigation results of the energy conservation mode demonstrated that the PV combined TEG hybrid module can be able to provide adequate real and reactive powers for the DVR system thereby the electricity tariff for the consumer could be reduced.

^{λ}D

^{μ}controller for a class of first-order delay-time systems