Power system and protection pdf

For other applications such as protection and as- significantly smaller number of matrix multiplications, as the sociated controls, DSE needs to operate at faster time scales and calculations of observer gain are usually done offline assuming typically uses SV measurements or dynamic phasors.

An over- certain error bounds of process and measurement models [16]. Kalman filters, however, still produce Gear Train some useful results even with a limited number of sensors. However, design- Wind PWM switched ing a good observer when the system is subjected to complex Switching Logic Converter Control Frequency nonlinearities is very challenging as it involves nonlinear opti- Supplementary Controller mization [18].

Also, the Jacobian matrix is needed for some Estimated States nonlinear observer designs. As a result, special tech- power system, they are the main points of control-actuation and niques are needed to establish system stability. DSE-based control only for small-signal dynamics [5], and are similar to the tradi- design can be centralized, decentralized, or hierarchical Fig. Dynamic state disturbance?

No in the LQR costs to find control gains. However, this is usually not an issue as transient for small stability disturbance? The estimated rotor speed of each gen- erator is used as a feedback signal in the MPC for excitation Centralized and Hierarchical Control control. A con- three-phase fault at the infinite bus in a single machine infinite trol law is then obtained using the system model and the dy- bus system.

This example uses a highly accurate Koopman namic state estimates via various control techniques, including model trained with random excitations.

Some limitations of centralized con- trol that restrict its field implementation are: communication la- tencies can impact performance, an accurate model of the whole system at the central location is required, and communication Fig. Nonlinear control methods are needed to ensure the stability A partial solution to the problems of centralized control is of the system against large disturbances, i.

These methods machine locations and the estimates are sent to a central loca- either find a partially linear transformation of system dynamics, tion or PMU measurements are sent to a central location for called feedback linearization [27]-[28], or find a positive scalar performing DSE in a decentralized manner using a federation with a negative time derivative, called a Lyapunov function of estimators [22]. Global control laws are then developed at [29]-[30].

Feedback linearization is easier to formulate and use the central location along with local control laws for decentral- than Lyapunov functions, but the latter has better asymptotic ized locations. The local and global controls form the two levels stability characteristics. In [27], detailed generator modeling of hierarchical control, as shown in Fig.

In the worst-case scenario, detailed load modeling for the same. In [29], a Lyapunov func- such control can also be used to shed run-away generators on tion is constructed for excitation and governor control, and [30] the fly to prevent instability [25]. The hierarchical control ap- proposes an optimal Lyapunov formulation for excitation con- proach also requires knowledge of a complete system model at trol and uses neural networks for computational efficiency. Although nonlinear control works well even for small dis- Decentralized Control turbances, its control costs given by the time integral of a In decentralized control, each generator is controlled inde- weighted sum of the squares of control inputs are much higher pendently, requiring only local measurements for both DSE and than the linear control costs.

Hence, it is practical to activate control. By controlling local machine dynamics, global system nonlinear control only for large disturbances. To the resulting frequency deviations in area 1 have been plotted.

The comparison of per- can use the system snapshots provided by a DSE giving access, formances of the hybrid control and PSS control system model, for instance, to the excitation system status of generators simulation conditions, and controller parameters are as those in namely: is the generator controlling its voltage or is it under [27] shows that hybrid control can ensure both small-signal field or armature current limit?

The stability and transient stability of the system Fig. Long-term voltage instability detection and control can use originally proposed ap- proaches like sensitivities in [37] and modal analysis in [39]. In addition to the dynamic states, the network state and topol- ogy are required to monitor, detect, and control voltage insta- bility. In this respect, the local DSE approaches discussed in [1] are combined with network state estimation, which can be fully based on PMUs or multi-rate measurements [38].

In principle, for long-term voltage instability, the system state can be up- dated at low rates less than 10 times per second. One of the Fig. Control performance of decentralized hybrid-control advantages of DSE is its ability to provide full state estimates, thus enabling full-state feedback MPC controllers [40]. Deviation Hz 0 0. Deviation Hz 0.

A framework for two-level long-term voltage stability VS in continuous fluctuations in system frequency and tie-line- monitoring, instability detection, and emergency control power exchanges causing them to deviate from their nominal It is a two-level scheme. The upper level is in charge of wide- values. These variations are corrected by governor-based load area monitoring, while the lower-level controllers provide re- frequency controller LFC action [32].

However, communication im- minimum transmission voltages to maintain. Those voltages pediments hinder the applicability of such LFCs. To address correspond to the point where long-term voltage instability is this issue, it is proposed that traditional LFCs should be assisted detected [41].

The phenomenon of con- are modeled as unmatched perturbations. This improves the fre- cern is faster and a DSE used for this purpose must provide the quency stability margins compared to observer-based LFCs system dynamic states at a high rate 10 to 60 times per second ; such as [36] and is illustrated in Fig. This requires even more rent transformers CTs , and MUs. Next, using the DAEs of the rigorous metering infrastructure and DSE should use a detailed line of interest, DSE estimates the dynamic states of the system, model with the full complement of state variables for the CBRs including SV voltages and currents at the remote side.

Finally, to solve voltage instability problems. Note that the remote side fre- have been proposed in the literature. In addition, estimated SV states of the dynamics and this allows for the development of a flux estima- system can also be used for other applications such as fast con- tion based control scheme [42], which achieves better fault re- trol of converters, harmonics filtering, etc.

In [43], DSE has been used to estimate the electromechanical dynamics of a V. DSE-based sliding mode control for DFIG- nents, minimize the power outage, and ensure the safety of hu- integrated power systems is developed in [44] for maximum en- man beings and the overall power system.

Protective relays are ergy extraction and power quality enhancement. The idea is fur- evolving with improved reliability. However, statistically, the ther extended to the DSE-based frequency restoration method industry in the U. DSE based control percent misoperation on average [50]. These are due to limited can also be used to damp sub-synchronous oscillations in series- measurements, mis-coordination among relays, or faults that compensated lines with wind generation [46].

In addition, most pro- B. DSE-based Control using SV Measurements tective relays are based on fundamental frequency measure- Traditional generator controls are mostly decentralized as the ments, which limit their operational speed as well as applicabil- local frequency and voltage information acts as a medium to ity to DC systems. The proliferation of CBRs has also generated bring the information of the rest of the grid to the local genera- new challenges for legacy protection systems.

Most relay set- tor. Modern power systems are evolving with a lower inertia tings are fixed; but with reduced system inertia, the settings and more complex transients with CBRs. In this case, controls need adjustment to distinguish stable and unstable swings. DSE with local frequency and voltage information may not be suffi- based protection provides a new approach to deal with these cient; remote side information such as frequency, RoCoF, and challenges. With the help of SV measurements and A.

DSE-based Protection using PMU Measurements the accurate time-domain transmission line model, DSE can ef- 1 Out-of-step Protection using Direct Stability Assessment fectively estimate the voltages and currents at the remote side Quasi dynamic state estimation has been used to monitor the of the transmission line using local information only, without stability of generators and to detect instability when it occurs. Specifi- nals of a transmission line [47]-[48].

Afterward, remote side in- cally, generator instability is a serious problem for power sys- formation such as frequency, RoCoF, and waveform distortion tems; generators must be protected against this condition; out- is extracted from the estimated voltages and currents and is uti- of-step OOS relaying is used to detect and protect the genera- lized as the input to CBR control.

When instability occurs the impedance Units PTs WTS Converters moves from the right-hand side of the impedance diagram to the Line of interest left-hand side. Because of this timetable i. DSE based converter control system using SV measurements based approach described above [51]-[52]. The DSE estimates the po- the generator and the immediate network to which the generator tential energy after the clearing of the fault and due to compu- is connected.

A simple generator rotor is at The CoO of oscillation. At this phase angle, the generator can be tripped is characterized by the fact that the frequency at that point does immediately, avoiding any additional stress on the generator. The theory states that when the total energy reaches the value of the peak potential energy stability barrier the generator becomes unstable. The dynamic state estimator provides the total energy of the gener- ator as well as the potential energy function and asserts insta- Fig.

Instability assertion time via DSE and comparison to legacy bility when the total energy reaches the stability barrier. It turns OOS Protection out that the assertion of generator instability occurs before the 2 Adaptive OOS Relay Settings Approach generator has swung away from the system and therefore can Adaptive OOS settings based on quasi dynamic model param- be tripped immediately, avoiding severe transients of an unsta- eters estimation preserves existing industrial practice and it is ble generator.

Next, we present a simplified example to illus- assisted by DSE. The OOS settings are mainly influenced by trate the performance of this method.

It consists of age, and generator inertia. As seen by the relay, these parame- three generators, three transformers, six transmission lines, five ters vary with time in the modern grid architecture with high loads, and 12 breakers. The parameters of the generators, trans- penetration of power electronic interfaced technologies. The formers, lines, etc. We consider a fault on the approach reported in [53] can be used to provide this real-time kV, a mile-long line that is successfully cleared by the pro- information for OOS relay settings recalculation.

DSE is used tection system of the line. The fault disturbs the system and os- to estimate dynamic model parameters which are used to recal- cillations are triggered. This approach has been reported in [54].

Single-line diagram of the example test system Compared to the synchrophasor measurements from PMUs, Fig. DSE can be utilized to of the generator. The trajectory is superimposed on the charac- extract the system information embedded in SV measurements, teristics of an OOS legacy protective relay to compare the DSE to provide reliable detection of fault conditions that are not re- based protection results with the legacy protection. The appar- flected by fault current levels, distortion of waveforms and ent impedance moves to a value very close to the origin upon characteristics of fault currents.

During the fault, as the generator urements provides detection of faulty conditions much faster and the system oscillate, the trajectory moves. The fault is than legacy protection functions such as overcurrent protection, cleared 0. For this system, usually require collection of enough data to compute phasors, the critical clearing time is 0. This means that the genera- resulting in fault detection delays. At widely adopted current differential protection.

By contrast, DSE based protection examines DSE combines the dynamic model and the SV measurements, whether all the physical laws of the specific protection zone are solves the dynamic states of the system, and computes the con- satisfied and an internal fault is detected with any violation of sistency between the dynamic model and the measurements via any physical law. Depending on the protection zone, physical the well-known chi-square test. Finally, the trip signal is issued laws may include KCLs, KVLs, motion laws, thermodynamic to open the circuit breaker of the transmission line if an internal laws, etc.

The primary improvements of the DSE based protec- fault is detected. Using such an approach high impedance, or tion compared to the existing protection approaches such as arcing faults are easily detected.

A similar approach can be used current differential protection include: 1 speed: DSE based to design smart auto-reclosure procedures [58]. Specifically, the exact fault location tems; 3 DSE based protection approach checks all the physical should be determined to minimize the time spent searching for laws instead of KCLs only , and therefore can detect internal the fault location by repair crews, yielding reduced power out- faults with improved sensitivity and reliability.

Note that cur- age time and operating costs. With the development of fast-trip- rent differential protection fails to detect some faults, for exam- ping techniques of protective relays, the time window of avail- ple, inter-turn faults in transformers, etc. In addition, during this lows: 1 building the dynamic model that encapsulates all the short time window, the available measurements usually experi- physical laws of the protection zone in the time domain; the ence severe transients.

This leads to compromised accuracy of model uses differential and algebraic equations DAEs that calculated phasors and therefore increases the fault location er- could include electromechanical, electromagnetic, and thermal rors for legacy phasor domain-based fault location approaches. They were inspired by the widely adopted traditional the available measurements and the dynamic model.

Low con- phasor domain fault location methods. Traditional phasor do- sistency indicates that some of the physical laws are violated main methods build the relationship between the available and therefore an internal fault is detected.

The validity of the phasor domain measurements and the location of the fault using DSE based protection comes from the following key ad- algebraic equations, which are afterward solved to identify the vantages of DSE: 1 accurately tracking the dynamics of the location of the fault. DSE based fault location methods first de- system; 2 systematically checking the consistency between the scribe the relationship between the available time-domain measurements and the dynamic model through the residuals, measurements and the location using an accurate time-domain and 3 effectively filtering out measurement errors.

DSE based dynamic model of the transmission line with fault. The time- protection schemes have been applied to transmission lines domain model can also include the model of the arc [62].

The [55]-[56], microgrids [48], transformers [57], etc. With in- dynamic model is a set of DAEs, which typically include in- creased security and dependability compared to legacy methods, stantaneous voltages and currents through the transmission line the DSE based protection can be utilized as the main protection as dynamic states of the system, and also the location of the of the component of interest protection zone.

Specifi- nals of a transmission line [47]-[48]. Afterward, remote side in- cally, generator instability is a serious problem for power sys- formation such as frequency, RoCoF, and waveform distortion tems; generators must be protected against this condition; out- is extracted from the estimated voltages and currents and is uti- of-step OOS relaying is used to detect and protect the genera- lized as the input to CBR control.

When instability occurs the impedance Units PTs WTS Converters moves from the right-hand side of the impedance diagram to the Line of interest left-hand side. Because of this timetable i. DSE based converter control system using SV measurements based approach described above [51]-[52]. The DSE estimates the po- the generator and the immediate network to which the generator tential energy after the clearing of the fault and due to compu- is connected.

A simple generator rotor is at The CoO of oscillation. At this phase angle, the generator can be tripped is characterized by the fact that the frequency at that point does immediately, avoiding any additional stress on the generator. The theory states that when the total energy reaches the value of the peak potential energy stability barrier the generator becomes unstable.

The dynamic state estimator provides the total energy of the gener- ator as well as the potential energy function and asserts insta- Fig. Instability assertion time via DSE and comparison to legacy bility when the total energy reaches the stability barrier. It turns OOS Protection out that the assertion of generator instability occurs before the 2 Adaptive OOS Relay Settings Approach generator has swung away from the system and therefore can Adaptive OOS settings based on quasi dynamic model param- be tripped immediately, avoiding severe transients of an unsta- eters estimation preserves existing industrial practice and it is ble generator.

Next, we present a simplified example to illus- assisted by DSE. The OOS settings are mainly influenced by trate the performance of this method. It consists of age, and generator inertia.

As seen by the relay, these parame- three generators, three transformers, six transmission lines, five ters vary with time in the modern grid architecture with high loads, and 12 breakers. The parameters of the generators, trans- penetration of power electronic interfaced technologies. The formers, lines, etc. We consider a fault on the approach reported in [53] can be used to provide this real-time kV, a mile-long line that is successfully cleared by the pro- information for OOS relay settings recalculation.

DSE is used tection system of the line. The fault disturbs the system and os- to estimate dynamic model parameters which are used to recal- cillations are triggered. This approach has been reported in [54]. Single-line diagram of the example test system Compared to the synchrophasor measurements from PMUs, Fig. DSE can be utilized to of the generator.

The trajectory is superimposed on the charac- extract the system information embedded in SV measurements, teristics of an OOS legacy protective relay to compare the DSE to provide reliable detection of fault conditions that are not re- based protection results with the legacy protection.

The appar- flected by fault current levels, distortion of waveforms and ent impedance moves to a value very close to the origin upon characteristics of fault currents. During the fault, as the generator urements provides detection of faulty conditions much faster and the system oscillate, the trajectory moves. The fault is than legacy protection functions such as overcurrent protection, cleared 0.

For this system, usually require collection of enough data to compute phasors, the critical clearing time is 0. This means that the genera- resulting in fault detection delays. At widely adopted current differential protection. By contrast, DSE based protection examines DSE combines the dynamic model and the SV measurements, whether all the physical laws of the specific protection zone are solves the dynamic states of the system, and computes the con- satisfied and an internal fault is detected with any violation of sistency between the dynamic model and the measurements via any physical law.

Depending on the protection zone, physical the well-known chi-square test. Finally, the trip signal is issued laws may include KCLs, KVLs, motion laws, thermodynamic to open the circuit breaker of the transmission line if an internal laws, etc. The primary improvements of the DSE based protec- fault is detected. Using such an approach high impedance, or tion compared to the existing protection approaches such as arcing faults are easily detected.

A similar approach can be used current differential protection include: 1 speed: DSE based to design smart auto-reclosure procedures [58]. Specifically, the exact fault location tems; 3 DSE based protection approach checks all the physical should be determined to minimize the time spent searching for laws instead of KCLs only , and therefore can detect internal the fault location by repair crews, yielding reduced power out- faults with improved sensitivity and reliability.

Note that cur- age time and operating costs. With the development of fast-trip- rent differential protection fails to detect some faults, for exam- ping techniques of protective relays, the time window of avail- ple, inter-turn faults in transformers, etc. In addition, during this lows: 1 building the dynamic model that encapsulates all the short time window, the available measurements usually experi- physical laws of the protection zone in the time domain; the ence severe transients.

This leads to compromised accuracy of model uses differential and algebraic equations DAEs that calculated phasors and therefore increases the fault location er- could include electromechanical, electromagnetic, and thermal rors for legacy phasor domain-based fault location approaches. They were inspired by the widely adopted traditional the available measurements and the dynamic model.

Low con- phasor domain fault location methods. Traditional phasor do- sistency indicates that some of the physical laws are violated main methods build the relationship between the available and therefore an internal fault is detected. The validity of the phasor domain measurements and the location of the fault using DSE based protection comes from the following key ad- algebraic equations, which are afterward solved to identify the vantages of DSE: 1 accurately tracking the dynamics of the location of the fault.

DSE based fault location methods first de- system; 2 systematically checking the consistency between the scribe the relationship between the available time-domain measurements and the dynamic model through the residuals, measurements and the location using an accurate time-domain and 3 effectively filtering out measurement errors.

DSE based dynamic model of the transmission line with fault. The time- protection schemes have been applied to transmission lines domain model can also include the model of the arc [62]. The [55]-[56], microgrids [48], transformers [57], etc. With in- dynamic model is a set of DAEs, which typically include in- creased security and dependability compared to legacy methods, stantaneous voltages and currents through the transmission line the DSE based protection can be utilized as the main protection as dynamic states of the system, and also the location of the of the component of interest protection zone.

Additionally, the fault as an extended state. Then, DSE is applied to systemati- DSE based protection is capable of detecting bad data through cally estimate the location of the fault. The primary advantages centralized substation protection details in section V.

Due to space limitations, the figure only cally filter out measurement errors; 2 time-domain algorithms demonstrates the relay on the left terminal of the line, with the typically possess faster convergence compared to phasor do- inter-trip signal connected to the left side breaker the relay on main algorithms; 3 time-domain algorithms are not sensitive to the right terminal is equivalent.

Instrumentation faults blown fuses, shorted CT, incor- Fig. DSE offers another advantage: upon detecting a hidden failure and bad data, The two-terminal voltage and current SV measurements are the compromised data can be replaced in real-time by estimated obtained through PTs, CTs, and the MUs. The dynamic model values to ensure the resilient and reliable operation of the pro- of the protection zone is established by describing all the phys- tection system.

Zhao, et al. Power This paper has explored the usefulness and the advantages of Syst. DSE on many control and protection applications for modern [3] V. Telukunta, J. Pradhan, A. Agrawal, M. Singh and S. Srivani, power systems. It has been shown how DSE-based solutions "Protection challenges under bulk penetration of renewable en- comprehensively respond to challenges in the control and pro- ergy resources in power systems: A review", CSEE Journal tection of modern power systems holistically.

In addition, sev- Power Energy Syst. Meliopoulos et al. Power Delivery, vol. As a result, the dy- no. This further allows [9] Z. Huang, H. Krishnaswami, G. Yuan, and R.

Venkatasubramanian, H. Schattler, and J. Sanders, J. Noworolski, X. Liu, G. The practical Trans. Khalil, Nonlinear Systems, 3rd ed. Korda and I. Nicolai, L. Lorenz-Meyer, A. Bobtsov, R. Ortega, N. Niko- control for all DERs do not cover all protection and control laev, and J.

Zhao, L. Mili, "A robust generalized-maximum likelihood unscented Kalman filter for power system dynamic state estima- age control, and protective actions.

Qi, A. Taha, J. Wang, "Comparing Kalman filters and observ- cation. DSE applications need to be supported by adequate ers for power system dynamic state estimation with model uncer- computing resources or distributed to achieve practical and tainty and malicious cyber-attacks," IEEE Access, vol.

Many DSE applications , Nugroho, A. Qi, "Robust dynamic state estimation of control and protection lack examples of their practical im- of synchronous machines with asymptotic state estimation error plementation in the field. Power Syst. How to make them compatible with the current [18] S. Stefani, H. What level of centralized or dis- versity Press, New York, tributed computing is needed?

What response times are desir- [19] A. Taha, M. Bazrafshan, S. Nugroho, N. Gatsis, J. What information should be communicated between re- bust control for renewable-integrated power networks consider- mote sites and control rooms? What training should be devel- ing input bound constraints and worst-case uncertainty measure," IEEE Trans. Network Syst. And what new stand- [20] A. Singh, R. Singh, B.

Smart Grid, vol. Ersdal, L. Imsland, K. Yingchen Zhang, Dr. It also has objective questions or MCQ's of Power systems at the end of each chapter. Also, we provide all free downloads of electrical engineering ebooks , electrical engineering lecturer notes , electrical engineering free pdf. We always try to provide quality content and downloads. K Mehta ebook " is already produced on the internet for free. We are reproducing this book for free without any copyright issues.

We always respect the copyrights of this book and sharing the download link for the benefit of electrical engineering students who are searching for it on the internet. K Mehta ebook: Contents : Introduction Generating stations Variable load on power stations Economics power generation Tariff Power factor improvement Supply systems Mechanical design of overhead lines Electrical design of overhead lines Performance of transmission lines Underground cables Distribution system general D.