FAULT high cost of equipment. Fault detection methods

FAULT DETECTION AND PROTECTION FOR PHOTOVOLTAIC SYSTEMS

NARAYANASAMY. A                                          Dr.E. LATHA MERCY                       GAYATHRI.M

   PG Scholar/EEE                                                     Associate prof/EEE                           PG Scholar/EEE

GCT-CBE                                                                GCT-CBE                                        GCT-CBE

 

 

 

 

Abstract— The objective of the thesis is mainly focusing on dc side fault detection and protection for Photo Voltaic (PV) systems. The proposed algorithm has two variables namely ? and Larray, where ? is the ratio of instantaneous dc power to instantaneous irradiance and array losses Larray which is the difference between instantaneous expected power and actual power for distinguishing between normal operation and fault conditions in the simulation of PV array systems. This algorithm is proven to be successful in four possible scenarios: normal operation, partial shading conditions, line to line and line to ground faults and protection. The proposed algorithm does not require any ac quantities from the PV system. It overcomes the drawbacks of existing approaches of multi-signal decomposition technique of signal processing.

 

 

I.    INTRODUCTION

Renewable energy has recently attracted increasing attention of the researchers due to cleanliness, on-site availability, and absence of greenhouse gas emission. Solar energy is the most abundant, inexhaustible and clean of all renewable energy resources. Interest in photovoltaic power generation has increased in recent years thanks to its advantages. This wide distribution of photovoltaic panel production was not followed by monitoring, fault detection, and diagnosis functions to ensure better profitability. Numerous studies have been done on the diagnosis of photovoltaic systems but just a few have been reported for describing and localizing faults in PV systems. They are not directly applicable to conventional PV systems and they require a relatively high cost of equipment. Fault detection methods for photovoltaic systems are numerous (electrical characterization, visual inspection, ultrasonic inspection, infrared imaging, imaging). Some methods use appropriate equipment (thermal camera). This visual inspection or the thermal detection methods require visual checking with frequent visits on the PV module to monitor changes in its appearance as indicators of failures: browning, mechanical damage or occurrence of hot spots. Electrical diagnostic methods specifically use the electronic signature of faults 1. They continuously monitor the PV module performance until the appearance of a fault. The deformation of the resulting output provides information on the occurrence, location, and nature of the default. The first indication of module degradation is provided by a decrease in its output power. Resulting symptoms are presented by the I-V curves of electrical characterizations of the PV module. After detection, microscopic analysis can be performed to understand causes of the degradation. The latter are obtained from simulations, which take in constant inputs o geometrical configuration parameters of the array and variable inputs of meteorological data. A similar approach is adopted, but the measures considered here are irradiances, in the horizontal plane and in PV modules plane, the ambient temperature, as well as electrical quantities at the dc and ac sides of the PV system, which are fed into the simulation to calculate the normalized capture losses. A decision tree algorithm was used to detect and classify the PV array faults 2. Experimental data consisting of PV array current, voltage, irradiance, and operating temperature were used as attributes in the training set. The limitation of this method is that it may be difficult to obtain a training dataset that can cover all possible fault scenarios. In the measurements of the temperature of every PV module, array voltage, and current are used in a three-layered feed-forward neural network in order to detect the presence of a fault in the PV array. Limitations stated about training data holds for as well. A fault detection method has been proposed in which all PV string currents are measured and tested under outlier detection rules 3. namely array voltage, current, and irradiance are used as inputs in this method. which may be deemed constant for a given PV array under all conditions. A large amount of experimental training data and simulations of PV array are not required. The rest of this paper has the following organization. Section II presents the types of faults in PV array. Section III details  for the proposed fault detection. The result discussion  in Section IV.Section V conclusion  .    

 

 

 

 

 

 

 

 

 

Fig .1. Typical fault in PV Array                                                                      Fig.2. I-V Characteristics of PV array

                               

II.TYPICAL FAULT IN PV ARRAY                                                        

In general, faults that occur in PV systems could be classified as,

Ø  line–ground faults

Ø  line–line faults

A line–line fault is similar to a short-circuit fault in the grounded system Line–line fault scan be quantified based on the number of PV modules that have a mismatch. line–line fault with a larger number of PV modules mismatched ceases power generation from the faulted PV string if blocking diodes are used in the PV array. The above faults are permanent in nature. partial shading is a temporary fault. The PV array considered in this thesis has four strings with four series modules in each of them. As the maximum power voltage (Vmp) for one module is 36.5 V, the Vmp for the whole array is 146 V (36.5 × 4). With such high operating voltage, the forward voltage drop in blocking diode of around 0.7 V is negligible.

 

A. LINE TO LINE FAULT ANALYSIS

 

                         Fig 3 LL Fault

This type of fault will have the least losses among all faults (excluding partial shading) due to the least module mismatch fault is analyzed with help of P–V curves under normal and faulted conditions under standard test conditions (STC temperature of 25 °C and irradiance of 1000 W/m2). (Fig 3) shows that LL fault of PV Array.

 

B. LINE TO GROUND FAULT

A ground fault (Fig 4) is an accidental electrical short circuit involving ground and one or more normally designated current-carrying conductors. The magnitude of ground-fault current depends on fault location, fault impedance, and geographical factors. If a ground fault is not cleared by proper fault protection, the fault connection might begin to

Fig 4 LG Fault.

 

generate and sustain a DC arc, which may become a fire hazard 14.

 

 

 

 

 

 

 

 

 

III. FAULT DETECTION

It is necessary to distinguish between partial shading and permanent faults in PV arrays to improve the overall efficiency and reliability. Two variables, namely, gamma (?) and array losses (Larray), are introduced.

       

  Fig .6. Fault detection algorithm

 

1)       Gamma (?) is the ratio of instantaneous dc power to instantaneous irradiance, which is expressed as ? has SI units of m2.

 

                                ? =

Where,

            G is the instantaneous irradiance on unshaded portion of the PV array,

             VPV is the instantaneous PV array voltage,

             IPV is the instantaneous PV array current.

2)       Array losses (Larray) is the difference between instantaneous expected power and actual power which is expressed as,

 Array  loss(Larray) =

Where,

            Pm is the maximum power of PV array at a reference irradiance of Go at 1000 W/m2.  the instantaneous DC power is used to calculate Larray, which is readily available for measurement from the MPPT controller. The instantaneous solar irradiance (G) is measured using a reference Module

IV.RESULT DISCUSSION

(a)    Simulation for Normal Condition

For simulation of normal condition, data is taken using PV analyzer during cloudy days.  PV analyzer model (9018BT) available at Centre of Excellence in Alternative energy research (COE-AER Lab).

 

 

 

 

 

                    Fig 6 solar   PV analyzer                                                 Fig.7 Parameter data

 

 

 

 

 

 

 

(b)  SIMULATION OF LINE TO GROUND FAULT DETECTION

                                    

Fig 8 Simulink model for LG fault detection                  Fig 9 PV Voltage and Current for LG fault

Fig 8 shows that line to ground fault detection is one of the permanent faults. The fault detection based on the MPPT voltage, current, and irradiance of the different condition of STC

.

Fig 10 Voltage across the load under LG fault                                                 Fig 11 PV Output power

(c)     SIMULATION OF LINE TO GROUND FAULT PROTECTION

The Fig 9 Show that LG fault protection. In this Simulink model the trig value is connected to the circuit breaker and whenever trig value is 1 the circuit breaker will operate

 

Fig 12 Simulink model for LG fault protection                             Fig 13 Voltage across the load curve after protection

 

 

 

 

 

 

(D) SIMULATION OF LINE TO LINE FAULT

Fig 13 shows that line to line fault detection one of the permanent faults. The fault detection based on the MPPT voltage, current, and irradiance of the different condition of STC

.

Fig 13 Simulink model for LL fault                                       Fig 14 PV Voltage and current LL fault

 

(E) SIMULATION OF LINE TO LINE FAULT PROTECTION

Fig 15 shows that LL fault protection. In this model   the trig value is connected to the circuit breaker and whenever a trigger value is 1 the circuit breaker will operate.

 

   Fig 15 Simulink model for LL fault protection                       Fig 16 Voltage across the Load after protection

 

V.CONCLUSION

Thus in this thesis a new fault detection algorithm is using two variables is introduced, namely ? and Larray for distinguishing between normal operation and fault conditions in a simulation of PV array. Based on these two variables, a fault detection algorithm was proposed. This algorithm is proven to be successful in distinguishing between three possible scenarios: normal operation, partial shading, and permanent faults.  The proposed algorithm (i) does not require

any ac quantities from the PV system and (ii) training data from different conditions, it allows modified MPPT algorithms to be integrated alongside so that losses may be reduced under partial shading with P&O being used under normal operating conditions.

 

 

REFERENCES

 

1 M. K. Alam, F. Khan, J. Johnson, and J. Flicker, “A comprehensive review of catastrophic faults in PV arrays: Types, detection, and mitigation techniques,” IEEE J. Photovoltaic, vol. 5, no. 3, pp. 982–997, May 2015.

2 Y. Zhao et al., “Fault experiments in a commercial-scale PV laboratory and fault detection using local outlier factor,” in Proc. 40th IEEE Photovoltaic Spec. Conf., Denver, CO, USA, Jun. 8–13, 2014, pp. 3398–3403.

3 Y. Zhao, B. Lehman, R. Ball, J. Mosessian, and J. de Palma, “Outlier detection rules for fault detection in solar photovoltaic arrays,” in Proc. 28th Annu. IEEE Appl. Power. Electron. Conf. Expo, Long Beach, CA, USA, Mar. 17–21, 2013, pp. 2913–2920.

4 Y. Zhao et al., “Decision tree-based fault detection and classification in solar photovoltaic arrays,” in Proc. 27th Annu. IEEE Appl. Power Electron. Conf. Expo., Orlando, FL, USA, Feb. 5–9, 2012, pp. 93–99.

5 U. S. National Electrical Code, “Article 690—Solar Photovoltaic Systems,” 2011.

6 Y. Zhao, J. F. de Palma, J. Mosesian, R. Lyons and B. Lehman, “Line–line fault analysis and protection challenges in solar photovoltaic arrays,” IEEE Trans. Ind. Electron., vol. 60, no. 9, pp. 3784–3795, Sep.2013.

7 D. Collier and T. S. Key, “Electrical fault protection for a large photovoltaic power plant inverter,” in Photovoltaic Specialists Conference, 1988., Conference Record of the Twentieth IEEE, 1988, pp. 1035–1042 vol.2.

 

8 Outline of Investigation for Low-Voltage Fuses-Fuses for Photovoltaic Systems, UL Standard 2579, 2010.

9 A. Chouder and S. Silvestre, “Automatic supervision and fault detection of PV systems based on power losses analysis,” Energy Conversion and Management, vol. 51, no. 10, pp.1929-1937, 2010.

10 L. Bun, B. Raison, G. Rostaing, S. Bacha, A. Rumeau, et al, “Development of a Real-Time Photovoltaic Simulator in Normal and Abnormal Operations”, IECON 2011.

11 R. Isermann, “Fault-Diagnosis Systems- An Introduction from Fault Detection to Fault Tolerance”. Springer 2006, ISBN-10 3-540-241124.

12 Zhao, Y.; de Palma, J.; Mosesian, J.; Lyons, R.; Lehman, B., “Line- Line Fault Analysis and Protection Challenges in Solar Photovoltaic Arrays,” Industrial Electronics, IEEE Transactions on, vol.PP, no.99, pp.1,1, 0.

13 Hernandez, J.C.; Vidal, P.G., “Guidelines for Protection Against Electric Shock in PV Generators,” Energy Conversion, IEEE Transactions on, vol.24, no.1, pp.274,282, March 2009.

14 Ye Zhao, “Fault Analysis in Solar Photovoltaic Arrays”, MSc Thesis dissertation, Department of Electrical and Computer Engineering, Northeastern University, December 2010.

15 Takashima, T.; Yamaguchi, Junji; Otani, K.; Kato, K.; Ishida, M., “Experimental Studies of Failure Detection Methods in PV Module Strings,” Photovoltaic Energy Conversion, Conference Record of the 2006 IEEE 4th World Conference on, vol.2, no., pp.2227,2230, 7-12 May 2006.