Advancement in Power Transformer Infrastructure and Digital Protection

Table of Contents

1. Frontmatter

2. Chapter 1. Transformer Infrastructure for Power Grid

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   Electricity infrastructure is the arrangement made by a human being with the use of available resources to generate power, transfer it from one place to another, conversion and dispense electrical power at various locations. The parameters and ratings of individual national, state, and regional grids are diverse, but they follow the same ideology and protocol for the functioning of important infrastructure. The amount of electricity produced by green energy sources is increasing day by day in contrast to major electricity being generated by conventional sources. Apart from the local generation and utilization, the majority of electricity generated by the centralized plant is transported using large power transformers and complex transmission infrastructure. Power transformer stands in electricity infrastructure to safely and consistently transport energy to consumers over a long distance. Investment in dedicated transformer infrastructure is essential to tackle some unforeseen situations of weather, storms, and internal failure of the system as well as to recover the flexibility of the grid. However, the conditions of existing transformers are unpleasant due to the aging effect and overloading stress. Thus, flexible power transformers are the need for a future smarter grid to deal with the mixture of conventional and bidirectional power flow from renewable energy sources. Moreover, the advanced control, monitoring, and stand-alone protection of such transformers increase the efficiency of the entire power/smart grid. This chapter deals with the role of large power transformers in electrical infrastructure with possible interconnections and operations. The advanced technology development, design aspect, testing of a transformer, voltage regulation by OLTC, fault analysis with DGA, and condition monitoring are discussed in this chapter. Moreover, a smart/flexible transformer is facilitated with the adoption of the change in voltage and variation in impedance in the event of a malfunction of the power grid. They are self-controllable with the use of smart sensors and can change the load pattern as well as monitor their operation with the use of artificial intelligence and IoT techniques.

3. Chapter 2. An Overview of the Protection of Power Transformers

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   A transformer is the heart of the entire power system and power is the heartbeat of the entire nation for the growth of the manufacturing, production, and industrial aspects. All terms are directly concerned with the Direct Foreign Investment (FDI) of the nation. Mostly, all types of businesses depend on the reliability and continuity of electricity. So, the “without power no business” slogan proves its usefulness. Power system protection is very crucial and complex due to having huge numbers of nodes. The contingency of various tactics is involved in the system and if it is done randomly then the mis-operation of the protective schemes occurs. Nowadays, the deregulation of the power network involves the malfunctioning of several system parameters. As far as a concern with the reliability of the power network, it is directly apprehension to the growth of the nation. As a key component in the power system, the protection of the power transformer has remarkable importance. Power transformer designing has many complexities like nonlinearity of its core, higher power rating, different voltage and current ratios, different phase angles in primary and secondary, and connection class. Having numerous complexity and different operating characteristics of power transformers, protection also becomes multifarious. Also, unwanted tripping of power transformer generates issues not only for consumer or industry but it gives an effect for the ecosystem of society, economy, political scenarios, and the entire nation. Different power transformer failure analysis is carried out by focusing on the failure of protective schemes (Rajurkar et al. in 16th National power systems conference, pp 180–185 [1]). Even, in the international market, different failure analyses with recent trends and involvement of further scope for the protection are analyzed in depth (Binder in Transformers-magazine, vol 1, no 1, pp 30–33 [2]).

4. Chapter 3. Introduction to Magnetic Inrush of Power Transformer

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   A trustworthy and devoted defensive system for a transformer is a prime necessity in power structure because of an extremely pricey and dependable apparatus. Nonlinearity feature of this device offers puzzling characteristics during inrush as well as CTs’ saturation states. Being an electrical-magnetic machine, it is mostly inducted by a heavy rush of current known as inrush. The behavior of the core becomes a significant case which may be an obstruction to discriminating the faults. Here, in this chapter, the impact of the inrush condition, precedent study performed to distinguish inrush from fault state, and a distinct technique for identification of it are explained. Around 50 research articles, a variety of books along with research theory are meticulously referred to evaluate the existing classifier methods. A relative investigation is performed to propose to filter out the pros and cons of different methodologies. Additionally, the average derivative angle of differential current-based resolution is presented to pick out the inrush case from faults and abnormalities. A detailed software simulation is performed to check the feasibility of the suggested scheme. The obtained outcome demonstrates the impact of inrush as well as the identification of faults. It is to be observed from the results that the scheme is capable of discriminating inrush and all kinds of faulty conditions within the considered transformer.

5. Chapter 4. Current Transformer Infrastructure and Its Application to Power System Protection

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   Progressively increasing demand for load forces the entire power system to uplift the infrastructure in terms of capacity and size. Moreover, the level of current increases multifold in EHV and UHV AC systems during peak load and severe fault conditions. Recording of this high value of current is necessary for the power system control and protection. Hence, current transformers (CTs) are inserted in every segment of the power system to scale down the actual current to a reasonable limit. However, due to the nonlinear core characteristic of the CTs, different parameters interfere in the measurement of the proper current during abnormal conditions like heavy fault, heavy burden, etc. Particularly in the unit type of protection, the selections of proper CTs are essential for developing an efficient protective system. Mismatch in parameters like CT ratio, Fault inception Angle (FIA), CT saturation, secondary burden, ratio and phasor error, higher retentivity, and presence of heavy remnant flux in the core gives major effects on measurement and analysis of the current signal in the protective scheme. All the said issues become a cause for the misoperation of the protective scheme. Various types of CT saturation effects occur on different types of relays like electromagnetic, static, or digital relays based on their protective arrangements. Sometimes, it is not possible to predict the outcome of the CT saturation effect on the relay protective algorithm. So, it is compulsory to have all basic fundamental knowledge regarding the possibilities of discrepancy and its probable solution on the protective schemes. Based on this knowledge, system engineers may be able to search for the solution to a problem in the protection field. Phase shifting and magnitude difference play an important role in measuring quantity. Different quantities of dissimilar frequencies cause the generation of distorted current signals. There is also non-uniformity generated in primary and secondary signals of the CT under core saturation, i.e., entering the core characteristics into a nonlinear region. This book chapter involves different parameters’ effects on CT saturation, its effect on software algorithms, and hardware analysis. The simulation is modeled using PSCAD^TM software for a wide range of data generation. In the last portion, how to detect CT saturation based on the saturation index in transformer protection has been elaborated pleasantly.

6. Chapter 5. Impact of Transitory Excessive Fluxing Condition on Power Transformer Protection

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   The unit protection method is universally accepted as the chief protective scheme for transformers. In conditions such as transitory over-fluxing, the universally accepted unit type of protection may mal-operate. Higher voltage level side and lower voltage level side currents become unequal in the transformer, because of core saturation. This chapter depicts the recognition of such conditions compared to the condition like interior fault, because of the fifth harmonic content of difference current. By identifying the amount of fifth harmonic w.r.t. the primary constituent of the difference current, the operational setting of the percentage bias feature is customized to evade false actions. When this type of circumstance is recognized, the presented algorithm triggers time delayed V/f relaying scheme to save the equipment. Here, the projected scheme is authorized on the PSCAD^TM platform, and post to that these collected data are exported to a MATLAB-based program, and then it is used for additional corroboration. The projected scheme is certified by testing various test circumstances like normal loading conditions, exterior fault, interior fault, as well as excessive fluxing in the core. One can say that the presented algorithm truthfully spots the interior fault condition inside of the transformer or can obstruct the triggering of relay operation under momentary over-fluxing (temporary excessive fluxing) circumstances. More on these, to evaluate the proficiency of the projected scheme, an experimental evaluation is checked on a one-phase transformer. Data received from diverse test conditions are recorded using DSO, the captured data then transfer to the computer to export it to the algorithm to check its proficiency. After examining the entire outcome, it can be clearly stated that the projected algorithm could never prevent the transformer separation during the occurrence of temporary excessive fluxing.

7. Chapter 6. Total Harmonic Distortion-Based Improved Transformer Protective Scheme

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   The nonlinear inherent nature of the transformer core escorts current and/or voltage wave deformation in instances of disturbance and/or fault. Harmonics involvement in current relies on the material of the transformer core, type of load condition, and anomalous states. Different odd and even Harmonics levels can be used for the discrimination of the normal state, inrush state and faulty state of the transformer. Through Fast Fourier Transformation (FFT) investigation, the Total Harmonic Distortion (THD) of an acquired quantity's wave pattern for a cycle is estimated. From the study, it is extracted that amount of THD remains higher in the instance of inrush because of early flux built in the core. On the contrary, for the faulty states, %THD will be noted lower for the reason of symmetric wave pattern. The amount of %THD beneath the prefixed level can be considered as a usual state and higher THD levels are treated as an inrush state. In case, THD falls inside the prefixed limits, it is pretended a faulty state. In this chapter, the legitimacy of this discriminative technique is checked by creating a variety of tests such as preliminary inrush, interior fault, toggling the dedicated transformer during faulty states, and CT saturation while persistent of the fault. A hardware experiment establishes the faithfulness of this discriminative technique for transformer fortification. This technique is trouble-free as well as proficient to resolve the difficult problem of cataloging of inrush state or faulty state of the transformer.

8. Chapter 7. Adaptive Biased Differential Protection Considering Over-Fluxing and CT Saturation Conditions

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   Modern-day unit-type transformer protection required a special performance that can adapt to any situation carefully with flexibility in all functions. It should cover all the abnormalities and discrimination ability against small internal faults with caution. Many possibilities of inter-turn fault, CT saturation under fault conditions, core saturation of transformer, magnetizing inrush, etc., are considered individually or with a different combination. This chapter exemplifies all the above-said conditions with combinations under unit-type protection. A smart relay is developed with the multifunctioning condition to operate under faulty conditions in the internal area of the device. An algorithm is validated in PSCAD™ software and a laboratory environment. Detection of inrush in transformer core to avoid malfunctioning in unit-type protection, itself a very crucial condition. The second derivative of differential current is elaborated here to evaluate the inrush current successfully. CT saturation conditions also cause severely misguided tools in protection. Here, in this chapter, adaptive algorithm is developed based on the saturation index. If saturation is detected in the CT, then percentage-biased characteristics shifted adaptively and avoid malfunctioning under such abnormality. Sometimes over-fluxing conditions are generated due to the amplification of system parameters. Times for over-fluxing are normally momentary or very small under starting of the transformer for a few cycles. Mostly, fifth and seventh harmonics are predominant under over-fluxing conditions. So, based on the fifth and seventh harmonic levels, detection techniques are also incorporated into the proposed algorithm to overcome this abnormality. So, here in this chapter inrush, CT saturation, and over-fluxing conditions are incorporated with discrimination of internal fault and other abnormal conditions to avoid the unnecessary operation of percentage-biased differential protection of the power transformer. A trip signal is generated as per the suggested algorithm when an internal fault or severe abnormality arises in a transformer. The considered power system diagram is simulated in PSCAD™ software. Voltage and current data are captured through the CT & PT of PSCAD™ and analyzed by the Modified Full Cycle DFT (MFCDFT) algorithm in MATLAB software. The hardware setup is developed in the laboratory environment considering the physical three-phase transformer with tap facilities on both windings. Various abnormal and fault conditions are generated and real-time testing has been performed on a developed algorithm. The suggested algorithm runs successfully on software and hardware with result analysis and proved its acceptability in a real field of power systems.

9. Chapter 8. Convolution Neural Network and XGBoost-Based Fault Identification in Power Transformer

   Nilesh Chothani, Maulik Raichura, Dharmesh Patel

   Abstract

   The power transformer is working in strained conditions due to the complex power system structure and increasing load demand by various sectors such as industry, commercial, and agriculture. Abnormal conditions and faults may arise during the operation of the power transformer, which may lead to insecurity and instability of the power system at large. Thus, it is mandatory to identify the types and locations of faults to minimize the interruption. Here, in this chapter, the convolution neural network (CNN) with the XGBoost technique has been proposed to accurately identify the transformer faults. To authenticate the proposed methodology, a small Indian power system network is simulated in PSCAD™ software. Voltage and current data are being recorded for the analysis and validation of the proposed method. The generated data is transferred to a one-dimensional CNN for accurate feature extraction. The extracted data from the CNN has been migrated to an algorithm prepared using XGBoost in MATLAB software. To authenticate the proposed method, a huge quantity of data is created in PSCAD™ software using a multi-run facility available in it. The training and testing of the data have been carried out in a high-performance CPU using considered AI techniques. The proposed technique accurately discriminates between the internal fault in the transformer with all external faults/abnormalities. Further, to demonstrate the right fullness of the suggested protective scheme, a hardware setup is prepared with a 50 KVA, 440/220 V transformer in the laboratory. It has been observed that the demonstrated method provides an accurate classification of different faults, and it operates in a short time.

10. Chapter 9. Sequential Component-Based Improvement in Percentage Biased Differential Protection of a Power Transformer

    Nilesh Chothani, Maulik Raichura, Dharmesh Patel

    Abstract

    Percentage-biased differential protection may falsely actuate for the cases like CT saturation and inrush generation. Sequential components of recorded quantity may also differ during abnormal events in terms of their phasor and/or magnitude. Here, in this chapter, a combination of the phasor difference of the sequential component is added as the parallel defensive technique of the percentage-biased differential protection to improve its operation under various anomalous conditions. Various irregularities such as high resistance internal fault, CT saturation, magnetizing inrush, and variation of the load level may appear and can become the reason for the mis-operation of the percentage-biased differential protection. Here, to validate the algorithm, a Full Cycle Discrete Fourier Transform (FCDFT) technique is used to investigate the current signals. Initially, the acquired phasor current signals are transformed into equivalent sequence components. Later, the phasor angle of all three sequential components (i.e., “ +Ve”, “−Ve” and “0” sequence components) are estimated on both primary and secondary side current signals. On the occurrence of the internal fault, the phasor angle difference of like sequence components is observed very low in degree, and for external fault, it may be observed up to 180°. Simultaneously, the percentage-biased differential relaying scheme work as a parallel defensive protective algorithm. All possible test conditions are replicated in PSCAD™ software and authenticated by the MATLAB coding of the FCDFT algorithm. Numerous test cases are carried out on the developed simulation and the suggested scheme is validated successfully for all the fault categories and anomalies.

11. Chapter 10. Current Direction Comparison-Based Transformer Protection Using Kalman Filtering

    Nilesh Chothani, Maulik Raichura, Dharmesh Patel

    Abstract

    The complexity of the power system network increases day by day as the power demand increases rapidly. Per capita power consumption is increased in the entire nation. To create comfort in life, it's required to attain the power system stability further to run all the electrical appliances. So, the reliability of the power is directly concerned with comfort and national growth. Power is transferred in a network by a transformer just as a pumping system. In a power system, a transformer has the highest efficiency due to its static nature. However, due to the different voltage/current and turns ratio, it is considered a very complicated device. Also, the core saturation of the transformer gives the worst effect on the unit-type protective scheme. Thus, there is a need to provide an accurate protective scheme for this device. Most of the study works have utilized Fast Fourier Transform (FFT)/Discrete Fourier Transform (DFT) algorithm to capture the required signals for protection purposes. This chapter represents three-state Kalman Filtering-based analyses involving the advantages over the FFT/DFT scheme. Different scenarios of the transformer mis-operation are involved with result analysis such as magnetizing inrush, CT saturations under internal and external faults, and high resistance internal fault. All the said test cases are authenticated on PSCAD™ software. After capturing data from PSCAD™ software, they are analyzed by MATLAB programming coded with the Kalman Filtering method. The proposed method is exploited to derive the phasor angle of the current and voltage signals acquired from both the primary and secondary winding of a transformer. Here, a comparative direction of the voltage and current-based analysis is carried out to discriminate all the above-said conditions. Said algorithm gives perfect tripping command for an internal fault and stays steady against external fault and all other abnormal conditions which is a prime requirement of a unit type of protection.