Filename
stringlengths 22
64
| Paragraph
stringlengths 8
5.57k
|
|---|---|
Processed_The_link_between_unemployment_and_real_economic_gr.txt | Figure 3. The dGDP curve corrected to the CPI as a benchmark after and before 1979. After 2010, the curves start to diverge again. |
Processed_The_link_between_unemployment_and_real_economic_gr.txt | Figure 4. A new coefficient of 0.8 is used to fit the CPI and dGDP after 2010. The fit is good and the total factor for this period is 0.8*1.26=1.00. It seems the original GDP definition is used after 2010. |
Processed_The_link_between_unemployment_and_real_economic_gr.txt | It is important to stress that the most recent time segment has a negative constant a=-0.25. This means that the dup in (5) is negative and the rate of unemployment decerases with time even in the absence of real economic growth. The fall in the unemployment rate from 9.63% in 2010 to 3.67% in 2019 was not accompanied by a stellar economic growth. The real GDP per capita was growing by 1.6% per year on average. |
Processed_The_link_between_unemployment_and_real_economic_gr.txt | Figure 5. When the 1979-to-2010 set of coefficients are applied to the data after 2010 the predicted curve does not match the measured one. |
Processed_The_link_between_unemployment_and_real_economic_gr.txt | Figure 6 illustrates the predictive power of model (5): the measured rate of unemployment USA between 1951 and 2019 is very close to the predicted rate, with the real GDP per capita published by the BEA (2020). The rate of unemployment is borrowed from the BLS (2020). Figure 7 presents the model residual errors as a function of time. with the standard deviation of 0.49% and the mean uneployment rate of 5.8% for the entire period. The avearge annual absolute change in the unemployment rate was 0.8%. Figure 8 depicts the linear regression of the measured and predcited time series with R2=0.89. Hence, the new set of coeffcients also provides an excellent match between the measured and predicted values, i.e. the model is validated by the data between 2010 and 2019. |
Processed_Fermions_on_the_Light-Front.txt | Issues that are specific for formulating fermions in light-cone quan- tization are discussed. Special emphasis is put on the use of parity invariance in the non-perturbative renormalization of light-cone Hamil- tonians. |
Processed_Fermions_on_the_Light-Front.txt | The first of the above issues has been addressed very often in the past and we will restrict ourselves here to a brief summary.b In the LF framework, non- trivial vacuum structure can reside only in zero-modes (k+ = 0 modes). Since these are high-energy modes (actually infinite energy in the continuum) one often does not include them as explicit degrees of freedom but assumes they have been integrated out, leaving behind an effective LF-Hamiltonian. The important points here are the following. If the zero-mode sector involves spon- taneous symmetry breaking, this manifests itself as explicit symmetry breaking for the effective Hamiltonian. In general, these effective LF Hamiltonians thus have a much richer operator structure than the canonical Hamiltonian. There- fore, compared to a conventional Hamiltonian framework, the question of the vacuum has been shifted from the states to the operators and it should thus be clear that the issues of renormalization and the vacuum are deeply entangled in the LF framework. |
Processed_Fermions_on_the_Light-Front.txt | aBecause of lack of space, this important issue could not be discussed in these notes. bSee for example Ref. 1 for a more extensive discussion on this question. |
Processed_Fermions_on_the_Light-Front.txt | and thus j− contains quartic interactions making it at least as difficult to renormalize as the Hamiltonian. |
Processed_Fermions_on_the_Light-Front.txt | We know already from renormalizing P − that the canonical relation be- tween ψ(−) and ψ(+) is in general not preserved in composite operators (such as ¯ψψ). It is therefore clear that a canonical definition for j− will in general violate current conservation since it does not take into account integrating out zero-modes and other high-energy degrees of freedom. |
Processed_Fermions_on_the_Light-Front.txt | Most importantly, VCC is no problem in LF quantization and is manifest at the operator level, provided j− is defined using Eq. (3). |
Processed_Fermions_on_the_Light-Front.txt | Since VCC can easily be made manifest (by using the above construc- tion!), there is no point in testing its validity and it cannot be used as a non-perturbative renormalization condition either. |
Processed_Fermions_on_the_Light-Front.txt | For a non-interacting theory, Eq. (3) reduces to the canonical definition of j−, but in general this is not the case when P − contains interactions or even non-canonical terms. |
Processed_Fermions_on_the_Light-Front.txt | Note that (as so often on the LF) q+ appears in the denominator of Eq. (3). Therefore, as usual, one should be very careful while taking the q+ 0 limit and while drawing any conclusions about this limit. An example of this kind are the pair creation terms in j−. Naively they do not contribute for q+ 0, since the q ¯q pair emanating from j− necessarily carries positive q+. However, since j− often has very singular matrix elements for q+, such seemingly vanishing terms nevertheless survive the q+ 0 limit, which often leads to confusion. For an early example of this kind see Ref. 4. A more recent discussion can be found in Ref. 5. |
Processed_Fermions_on_the_Light-Front.txt | For most applications of LF quantization, the lack of manifest parity actu- ally does not constitute a problem — in fact, one can view it as an opportunity rather than a problem. The important point here is the following: due to the lack of manifest covariance in a Hamiltonian formulation, LF Hamiltonians in general contain more parameters than the corresponding Lagrangian. Parity invariance may be very sensitive to some of these parameters. |
Processed_Fermions_on_the_Light-Front.txt | This model actually has a lot in common with the kind of interactions that appear when one formulates QCD (with fermions) on a transverse lattice 7. |
Processed_Fermions_on_the_Light-Front.txt | The rest of the quantization procedure very much resembles the procedure for self-interacting scalar fields. |
Processed_Fermions_on_the_Light-Front.txt | Thus the bad new is that the number of parameters in the LF Hamilto- nian has increased by one (compared to the Lagrangian). The good news is that a wrong combination of ci will in general give rise to a parity violating theory: formally this can be seen in the weak coupling limit, where the correct relation (c2 3 = c2c4) follows from a covariant Lagrangian. Any deviation from this relation can be described on the level of the Lagrangian (for free massive fields, equivalence between LF and covariant formulation is not an issue) by i∂+ ψ, which is obviously parity violating, addition of a term of the form δ since parity transformations result in A± P A∓ for Lorentz vectors Aµ; i.e. → δ . This also affects physical observables, as can be seen by L considering boson fermion scattering in the weak coupling limit of the Yukawa model. At the tree level, there is an instantaneous contact interaction, which is proportional to 1 q+ . The (unphysical) singularity at q+ = 0 is canceled by a term with fermion intermediate states, which contributes (near the pole) with 1 q+ . Obviously, the singularity cancels iff mV = mkin. an amplitude Since the singularity involves the LF component q+, this singular piece obvi- ously changes under parity. This result is consistent with the fact that there is no zero-mode induced renormalization of mkin and thus mV = mkin at the tree level. This simple example clearly demonstrates that a ‘false’ combina- tion of mV and mkin leads to violations of parity for a physical observable, which is why imposing parity invariance as a renormalization condition may help reduce the dimensionality of coupling constant space. |
Processed_Fermions_on_the_Light-Front.txt | Of course there are an infinite number of parity sensitive observables, but not all of them are easy to evaluate non-perturbatively in the Hamiltonian LF framework. Furthermore, we will see below that some relations among observ- ables, which seem to be sensitive to parity violations, are actually ‘protected’ by manifest symmetries, such as charge conjugation, or by VCC. |
Processed_Fermions_on_the_Light-Front.txt | would be Compton scattering cross sections between quarks and gluons (or the corresponding fermions and bosons in other field theories) because there one could tune the external momenta such that the potential singularity enters with maximum strength. However, this is not a very good choice: first of all non-perturbative scattering amplitudes are somewhat complicated to construct on the LF. Secondly, quarks and gluons are confined particles, which makes qg Compton scattering an unphysical process. |
Processed_Fermions_on_the_Light-Front.txt | A much better choice are any matrix elements between bound states. Bound states are non-perturbative and all possible momentum transfers oc- cur in their time evolution. Therefore, any parity violating sub-amplitude would contribute at some point and would therefore affect physical observ- ables. Secondly, since one of the primary goal of LFQCD is to explore the non-perturbative spectra and structure of hadrons, matrix elements in bound states are the kind of observables for which the whole framework has been tailored. |
Processed_Fermions_on_the_Light-Front.txt | One may also consider non-conserved currents, such as the axial vector current. However, there one would have to face the issue of defining the ‘minus’ component before one can test any parity relations. Parity constraints are probably very helpful in this case when constructing the minus components, but then those relations can no longer be used to help constrain the coefficients in the LF-Hamiltonian. |
Processed_Fermions_on_the_Light-Front.txt | can be nonzero at the same time, since a state cannot be both scalar and pseudoscalar. What restricts the usefulness of this criterion is the fact that the same ‘selection rule’ also follows from charge conjugation invariance ( ¯ψ and ¯ψγ5ψ have opposite charge parity!) which is a manifest symmetry on the LF. Therefore, only for theories with two or more flavors, where one can consider operators such as ¯us and ¯uγ5s, does one obtain parity constraints that are not protected by charge conjugation invariance. But even then one may have to face the issue of how to define these operators. Parity may of course be used in this process, but then it has again been ‘used up’ and one can no longer use these selection rules to test the Hamiltonian. |
Processed_Fermions_on_the_Light-Front.txt | In summary, many parity relations are probably very useful to determine the LF representation of the operators involved, but may not be very useful to determine the Hamiltonian. |
Processed_Fermions_on_the_Light-Front.txt | c4/2π. values In this example, only for m2 5 one obtains a form factor that is a unique function of (cid:1) (cid:0) (cid:1) V ≈ Q2, i.e. only for m2 5, the result is consistent with Eq. (14). Therefore, only for this V ≈ particular value of the vertex mass, is the matrix element of the current operator consistent with both parity and current conservation. |
Processed_Fermions_on_the_Light-Front.txt | This quadratic equation has in general, for a given value of q2, two solutions for x which physically correspond to hitting the meson from the left and right respectively. The important point is that it is not manifestly true that these two values of x in Eq. (14) will give the same value for the form factor, which makes this an excellent parity test. |
Processed_Fermions_on_the_Light-Front.txt | In Ref. 6, the coupling as well as the physical masses of both the fermion and the lightest boson where kept fixed, while the “vertex mass” was tuned (note that this required re-adjusting the bare kinetic masses). Figure 1 shows a typical example, where the calculation of the form factor was repeated for three values of the DLCQ parameter K (24, 32 and 40) in order to make sure that numerical approximations did not introduce parity violating artifacts. |
Processed_Fermions_on_the_Light-Front.txt | For a given physical mass and boson-fermion coupling, there exists a “magic value” of the vertex mass and only for this value one finds that the parity condition (14) is satisfied over the whole range of q2 considered. This provides a strong self-consistency check, since there is only one free parameter, but the parity condition is not just one condition but a condition for every sin- gle value of q2 (i.e. an infinite number of conditions). In other words, keeping the vertex mass independent from the kinetic mass is not only necessary, but also seems sufficient in order to properly renormalize Yukawa1+1. |
Processed_Fermions_on_the_Light-Front.txt | In the above calculation, it was sufficient to work with a vertex mass that was just a constant. However, depending on the interactions and the cutoffs employed, it ma be necessary to introduce counter-term functions8. But even in cases where counter-term functions need to be introduced, parity constraints may be very helpful in determining the coefficient-functions for those more complex effective LF Hamiltonians non-perturbatively. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Abstract—Modern power systems with high share of renewable generation are at the risk of rapid changes in frequency resulting from contingencies. The importance of an accurate assessment of system inertia and load relief, as well as frequency changes in the critical timescales is more pronounced in these power systems. This knowledge serves as an insight which will guide the solutions required for enhanced security and resilience. A novel procedure for simultaneously assessing system inertia and load relief using a system frequency response (SFR) model to fit measured disturbance data is presented. Results of applying the proposed approach on generator contingencies in the South West Interconnected System (SWIS) in Western Australia demonstrate the validity of the method and indicate that it can overcome some inertia estimation of the limitations observed in conventional methods based on the linearised swing equation, such as the sliding window and polynomial fit methods. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Many power systems across the world are experiencing a gradual decline in synchronous inertia levels as synchronous generators are increasingly being displaced by generation interfaced to the grid via power-electronic converters [1]. As inertia levels decrease, the average frequency in the power system is subject to more severe and rapid fluctuations in response to active power disturbances, e.g. from generation or load contingencies. It is therefore becoming more crucial that estimates of system inertia are reliable and accurate. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | However, the RoCoF is arguably the most sensitive param- eter in (1). Ideally, the RoCoF would be calculated based on the sample-by-sample time derivative of a high-resolution frequency trace at the onset of the disturbance. However, high- resolution frequency measurements from Phasor Measurement Units (PMUs) or fault recorders contain white noise introduced by transducer, quantisation and signal processing errors [11]. Moreover, as depicted in Fig. 1, the frequency trace can con- tain transient [5] and/or oscillatory components [2]. Therefore, calculating the time derivative using consecutive frequency samples would likely lead to significant errors [3]. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | where KE is the system inertia estimate (MW.s), fn is the nominal frequency (e.g. 50 Hz), Pcont is the disturbance size (cid:12) (MW) and d∆f (cid:12)t=0 is an estimate of the rate of change of dt frequency (RoCoF) evaluated at the onset of the disturbance (Hz/s). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | To mitigate these issues, moving average windows or low- pass smoothing filters have been proposed in the literature, as well as in practical applications. For example, a sliding window has been adopted as a standard in Europe (e.g. [3] and [5]), where the RoCoF estimate is calculated by applying a 500-ms sliding window over the frequency trace and then finding the maximum value of the sample-by-sample time derivative (since the maximum RoCoF theoretically occurs at the onset of the contingency). Alternative techniques include fitting a polynomial function to the raw frequency measure- ments [2] or simply applying a 0.5 Hz low-pass filter to the frequency trace [4]. The common thread is that the raw frequency measurements are smoothed out in some way to prevent spurious RoCoF measurements. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | It was observed in [12] that system inertia estimates based on this approach are only accurate for events with relatively high RoCoFs, in which the assumption that the initial fre- quency decline is retarded by system inertia alone remains valid. For smaller disturbances or high inertia conditions where the RoCoF is low, the influence of primary frequency response becomes more prominent and may invalidate this assumption. Furthermore, the choice of sliding window length (or cut-off frequency for the smoothing filter) will also affect the quality of the estimate, with higher levels of smoothing leading to larger estimation errors. It was also shown that inertia can be materially overestimated when wind farms provide virtual inertia emulation or fast frequency response [13]. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Another factor that may confound system inertia estimates is the presence of load relief (or damping), which is the natural sensitivity of loads in a network to changes in system fre- quency. As frequency declines after a generation contingency, so does power consumption from frequency-sensitive loads such as induction motors [14]. Load relief acts to correct the active power imbalance after a contingency and thus helps to arrest the decline in frequency. As a result, system inertia estimates based on the linearised swing equation after a disturbance may mistake the effects of load relief as additional inertia. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | In the literature, the system-level load relief factor (LRF), which is also referred to as the load damping coefficient or load-frequency characteristic, is typically estimated separately from system inertia. Earlier studies attempted to directly measure load relief by conducting live tests such as the intentional tripping of generators and transmission lines, e.g. in Great Britain [15], Norway [16] and Ireland [17]. A less invasive approach is to estimate the system-level LRF after a contingency event using measurements of total system load from the Energy Management System (EMS). However, EMS sampling resolution may not be high enough to accurately as- sess system load changes after a contingency [18]. To mitigate EMS resolution issues, high-speed fault recorder data can be used to estimate the system load change after a contingency [12]. Finally, estimating LRF has also been investigated using a bottom-up approach, e.g. using high-speed fault recorder data at substation level [14] or at the level of individual consumer devices [19]. However, it is remains unclear how to aggregate the individual LRF estimates into a meaningful system-level LRF. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | This paper takes a different approach to system inertia and load relief estimation. Rather than using the linearised swing equation, the system inertia and load relief factor are treated as unknown parameters in a low-order system frequency response (SFR) model and the model is then fitted to post-disturbance frequency and generator output measurements. This approach exploits the SFR model’s ability to accurately predict the system frequency trajectory after a disturbance, and thus use more measurement data as inputs to the estimation model. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | both system inertia and system-level load relief. The proposed method is shown to be credible when compared with other inertia estimation methods, while requiring less discretion in the choice of parameters, e.g. sliding window length or polynomial order. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The remainder of this paper is organised as follows: In Section II, the system inertia estimation method based on that model-fitting is proposed, with the added benefit the method concurrently estimates the system-level load relief factor. This is followed in Section III by several computer simulation case studies comparing the performance of the proposed approach with other methods. In Section IV, case studies are performed using data from a real power system to illustrate some of the advantages of the proposed method over other approaches. A discussion of the results in this paper is then presented in Section V, and finally concluding remarks and avenues for future work are described in Section VI. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Rather than using the linearised swing equation in (1) to estimate system inertia, we propose to jointly estimate system inertia and load relief / damping by fitting a low-order SFR model to measured frequency data. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Active power measurements are high resolution and time- synchronised, e.g. from fault recorders or PMUs with time synchronisation via GPS or IEEE Std 1588 Precision Time Protocol [20]. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | There are measurements at each major generating facility and interruptible load in the system. Where this is not practical, there should be at least measurements at each of the main PFR providers. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Note that with this approach, the influence of any PFR not explicitly included in the measured system PFR trace will be considered as load relief. This could include the influence of behind-the-meter embedded generation or small distribution- connected generation that may not be visible by fault recorders or PMUs. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Similarly, the contingency trace is formed by extracting the active power measurement trace of the contingency (e.g. power flow from a generator or radial network element). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | where ∆f (t) is the frequency deviation at time t (Hz), fn is the nominal frequency (e.g. 50 Hz), KE is the post-contingent system inertia (in MW.s), Pcont(t) is the contingency at time t (MW), p(t) is the aggregate system PFR at time t (MW), D is the load relief factor (% MW/Hz) and Pload is the system load at the onset of the contingency (MW). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | where f is the measured frequency, ˆf is the frequency cal- culated from the SFR model and N is the total number of samples. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The time-domain RMS simulations were conducted in DIgSILENT PowerFactory v2020 software and in each case, loads were modelled to be frequency-dependent with a load relief factor of D = 4%. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | In Case 1, the IEEE 9-bus system depicts a small islanded system with three generators and 315 MW of load. The default IEEE 9-bus system was modified to include a standard turbine- governor model (TGOV) for each generator. To simulate a frequency disturbance, generator G3 was tripped from 85 MW. A summary of the best and worst inertia and load relief estimates for each of the different methods is shown in Table I. It should be noted that only a single estimate is shown for the model fitting approach as there are no adjustable parameters (such as window length or polynomial order). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | System inertia estimates using the sliding window and Inoue polynomial fit methods with varying window lengths and polynomial order are shown in Fig. 2a and Fig. 2b respectively. The figures indicate that the accuracy of the inertia estimate is sensitive to the choice of window length or polynomial order. Fig. 2c shows the simulated (measured) frequency trace vs the fitted frequency trace after applying the model fitting method. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Relative to Case 1, the New England IEEE 39-bus system depicts a larger interconnected power system comprising nine generators and 6,097 MW of load. To simulate a frequency disturbance, generator G10 was tripped from 200 MW. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Like in Case 1, a summary of the best and worst inertia and load relief estimates from each of the different methods is shown in Table II. The inertia estimates with varying window length and polynomial order are shown in Fig. 3a and Fig. 3b respectively, and the measured vs fitted frequency trace after applying the model fitting method is shown in Fig. 3c. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | In Case 3, the IEEE 9-bus system in Case 1 is augmented with a 30 MW BESS providing FFR. This case is included to examine the effects of FFR on inertia estimates and validate the findings in [13], i.e. material errors in inertia estimates can arise when the sliding window method is applied to systems with FFR. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | As in the previous cases, a summary of the best and worst inertia and load relief estimates from each of the different methods is shown in Table III. The inertia estimates with varying window length and polynomial order are shown in Fig. 4a and Fig. 4b respectively, and the measured vs fitted frequency trace after applying the model fitting method is shown in Fig. 4c. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The results indicate a deterioration in estimation accuracy from the sliding window method from Case 1, particularly with increasing window length. The Inoue polynomial fit method performs better, with a narrower range of estimates, but the accuracy is sensitive to the choice of polynomial order and it may be difficult to ascertain a priori whether the order selected is the appropriate one. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The following case studies are based on actual genera- tor contingency events from the South West Interconnected System (SWIS) in Australia. The SWIS is a medium-sized islanded power system serving approximately 1.2 million customers in the southwest region of Western Australia. The SWIS covers a vast geographic area of 261,000 square kilo- metres (larger than the land area of the United Kingdom), but with a historical record peak demand of just 4.3 GW. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | On 17 March 2020 at 18:06pm, an open-cycle gas turbine generator tripped from 161 MW causing frequency to drop to 49.55 Hz (at the nadir). At the onset of the contingency, the system load was 2,441 MW and the total post-contingent generator inertia was 15,009 MW.s (calculated by summing the known generator inertia values for all online machines). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The inertia estimates with varying window length and polynomial order are shown in Fig. 5a and Fig. 5b respectively, and the measured vs fitted frequency trace after applying the model fitting method is shown in Fig. 5c. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Unlike in the computer simulations, the actual system inertia is not known. The post-contingent generator inertia is known, but this does not include load inertia so is more representative of the floor for the system inertia estimate. The summary of the results in Table V show the range (minimum and maximum) of estimates from the sliding window and Inoue methods, as well as the single point estimate from the model fitting method. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The sliding window method in Fig. 5a exhibits a similar shape to that in Cases 1 and 3, where beyond a threshold window length (in this case 170-ms), the inertia estimate continues to rise steadily without settling. In Cases 1 and 3, the actual system inertia was close to the estimate at the threshold window length, which in this case would be 20,735 MW.s. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | At high polynomial orders, the Inoue method in Fig. 5b appears to settle at around 26,000 MW.s. However, such an estimate would suggest a load inertia of roughly 42% of system inertia, which is inconsistent with previous inertia estimates in the SWIS, as well as estimates reported in the literature (such as [25] and [26]), where load inertia is typically estimated to be 10 - 25% of system inertia. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The model fitting results in Fig. 5c show good align- ment between the simulated and measured frequency traces (RM SE = 0.0095) and the inertia estimate of 19,610 MW.s is on the lower end of the sliding window and and Inoue estimate range. This suggests that the actual system inertia is likely in the 19,000 - 21,000 MW.s range. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | of estimates from the sliding window and Inoue methods, as well as the single point estimate from the model fitting method. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | performs with the oscillatory frequency trace observed in Case 2. In that previous case, the sliding window method is more accurate beyond a ”knee” point window length, where the effects of the oscillations are filtered out. In this case, the knee point occurs at a window length of 210-ms and a system inertia estimate of 14,000 MW.s. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Similarly, the Inoue method will tend to over-fit the fre- quency trace (and particularly the transient component) at high polynomial order. This can be seen in Fig. 6b, where the system inertia estimate is below the known generator inertia at polynomial orders >15. The Inoue method does not credibly provide any useful information regarding the correct range for system inertia. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The model fitting results in Fig. 6c show reasonably good alignment between the simulated and measured frequency traces (RM SE = 0.0146) and yields an inertia estimate of 15,534 MW.s. This estimate is broadly consistent with the sliding window method using a window length near the knee point. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The results from the case studies indicate that system inertia estimation using the proposed model-fitting approach is a credible alternative to conventional approaches based on the linearised swing equation. Good alignment between the measured frequency and the simulated frequency implied by the fitted SFR model suggests that the parameters in the SFR model are able to endogenously explain the frequency trajec- tory. The proposed approach also has the practical advantage of providing a single point estimate without needing to select parameters such as the sliding window length or polynomial order. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | Interestingly, the case studies highlight that system inertia estimates using the sliding window and Inoue polynomial fit methods are very sensitive to the choice of these parameters. The computer simulations show that depending on the scenario and choice of window length or polynomial order, system iner- tia estimates can be either very accurate or have large errors, e.g. ±20% to 50%. While there are broad patterns that are observed in the case studies, for example, systematic under- estimates when using the Inoue method with high polynomial orders in frequency traces containing transient or oscillatory components, there are no specific rules or heuristics to guide parameter selection. As the case studies in Section IV show, it is difficult to select the most appropriate parameter in practice, particularly as measurement noise present in frequency traces can also affect the inertia estimate. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The proposed model-fitting approach does not suffer from the method these issues and in the computer simulations, yields system inertia estimates within 3% to 9% of the true value. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | contingency / disturbance data. Case studies demonstrated the applicability and validity of applying the new approach on computer simulations with the IEEE 9-bus and 39-bus systems, and on real events in the SWIS in Western Australia. Furthermore, the proposed method is not sensitive to the selection of algorithm parameters such as the sliding window length or polynomial order. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | An avenue for future work is to investigate heuristics for the appropriate selection of sliding window length and polynomial order, potentially using the proposed model-fitting approach to calibrate the range of estimates. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | generators. Since inertia is endogenous to models based on the swing equation, the system PFR trace can only be used if the inertial components are removed first (otherwise the effects of inertia are double counted). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | A simple washout filter is proposed to isolate inertial components from the PFR using the measured frequency as an input and the known inertia constants of the generators contributing to the system PFR as a parameter (see Fig. 7). The inertial components are then removed from the measured system PFR trace. |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | The selection of the washout time constant Tw is a trade-off between accuracy and noise. Ideally, a very small time constant is applied so that the washout filter more closely approximates a true differentiator. However, small time constants are more sensitive to noise in the frequency measurement (refer to Tw = 25-ms in Fig. 8). On the other hand, a large time constant will reduce the accuracy of the estimate despite appearing less noisy (refer to Tw = 200-ms in Fig. 8). It was empirically found that a washout time constant of 60 ms to 80 ms yielded an acceptable balance between accuracy and noise (refer to Tw = 60-ms in Fig. 8). |
Processed_Joint_Estimation_of_System_Inertia_and_Load_Relief.txt | https://docstore.entsoe.eu/documents/soc%20documents/regional groups continental europe/2018/tf freq meas v7.pdf, European Network of Transmission System Operators for Electricity (ENTSO-E), Tech. Rep., 2018. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | In the large Nc limit, the variables required to analyze the low energy struc- ture of QCD in the framework of an effective field theory necessarily include the degrees of freedom of the η′. We evaluate the decay constants of the pseu- doscalar nonet to one loop within this extended framework and show that, as a consequence of the anomalous dimension of the singlet axial current, some of the effective coupling constants depend on the running scale of QCD. The calculation relies on a simultaneous expansion in powers of momenta, quark masses and 1/Nc. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | The low energy properties of QCD are governed by an approximate, spontaneously broken symmetry, which originates in the fact that three of the quarks happen to be light. If mu, md, ms are turned off, the symmetry becomes exact. The spectrum of the theory then contains eight strictly massless pseudoscalar mesons, the Goldstone bosons connected with the spontaneous symmetry breakdown. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | If the number of colours is taken large, the quark loop graph which gives rise to the U(1)-anomaly is suppressed [1]. This implies that, in the limit Nc , → ∞ the singlet axial current is also conserved: The theory in effect acquires a higher degree of symmetry. Since the operator qq fails to be invariant under the extra = 0, implies that U(1)-symmetry, the formation of a quark condensate, this symmetry is also spontaneously broken [2]. The spectrum of QCD, therefore, contains a ninth state, the η′, which becomes massless if not only mu, md, ms are turned off, but if in addition the number of colours is sent to infinity [3]. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | that the properties of this particle are subject to analogous constraints, which may again be worked out by means of a suitable effective Lagrangian. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | In the large Nc limit, τ0 becomes independent of Nc while Fπ grows in proportion to √Nc, so that Mη′ tends to zero [3]. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | more coherent picture than the canonical treatment with ϑ8 = ϑ0. The sign of the prediction is confirmed, but the numerical value obtained for the difference between the two angles is somewhat smaller than what is required by (5). The purpose of the present paper is to discuss the corrections which this low energy theorem receives from higher order effects. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | of the singlet field ψ and the vacuum angle θ is invariant under chiral transfor- mations. Chiral symmetry does therefore not constrain the dependence of the La- grangian on this variable: At this stage, the coefficients are arbitrary functions thereof, Vn = Vn( ˜ψ). They may be viewed as potentials that control the dynamics of the singlet field ψ(x). |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | Moreover, we may discard V5, because this term does not contribute to the meson masses, decay constants or photonic transition rates. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | We add a remark concerning the comparison between the extended effective theory based on U (x) SU(3). As discussed in detail in ref. [5], the Ward identities obeyed by the Green functions of the vector and axial currents imply that the SU(3) effective Lagrangian is invariant under the full group U(3)R× U(3)L of chiral gauge transformations – up to the Wess- Zumino-Witten term, which accounts for the anomalies occurring in these identities. The transformation law U ′ = VRU V † L , however, violates the constraint det U = 1. Instead of modifying the transformation law, it is more convenient to perform a i 3 θ), i.e. to modify the constraint, which then change of variables, U takes the form det U = e−iθ. The condition amounts to the gauge invariant relation ψ(x) = θ(x), which states that the standard framework arises from the extended one if the extra variable ψ(x) is fixed at the minimum of the potential V0 = V0(ψ+θ). |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | eff are deter- mined by g, µ and by the masses of the heavy quarks c, b, t. We now discuss the manner in which the effective coupling constants depend on the running scale of QCD. In particular, we wish to work out the consequences of the fact that the ma- trix elements of the singlet axial current depend on the renormalization, because this operator carries anomalous dimension. To our knowledge, this issue is not discussed in the literature, because the matrix elements of the singlet axial current are usually considered only at leading order of the 1/Nc–expansion 5. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | The various Green functions are Lorentz invariant expressions formed with the external momenta. There are two categories: those that may be expressed in terms of the metric gµν alone (natural parity) and those for which the expression is linear in the tensor ǫµνρσ (unnatural parity). Accordingly, the effective action may be decomposed into two parts that are of natural and unnatural parity, respectively. The anomalies only show up in the latter. As we are restricting ourselves to an analysis of the decay constants, we are concerned with the natural parity part, which is invariant under the transformation (9). |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | While the scale dependence of the coupling constant and of the quark masses shows up already at leading order in the 1/Nc–expansion, the triangle graph responsible for the anomalous dimension of the singlet axial current is suppressed by one power of 1/Nc, so that ZA = 1 + O(1/Nc). |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | We now translate these properties into corresponding scaling laws for the coupling constants of the effective Lagrangian. This may be done by working out the first few terms in the chiral expansion of suitable observables. The leading term, F , is evidently scale independent: It represents the pion decay constant in the chiral limit. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | Conversely, this property ensures that the effective action generated by the effective theory is scale independent. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | differs from the corresponding SU(3) formula: µη is replaced by the term (cos2ϑµη + sin2ϑµη′). Numerically, the difference is not significant, however: The two terms differ by less than 0.02. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | which is independent of the running scale µ of QCD as well as of the scale µχ used in the renormalization of the effective theory (2 Γ5 + 3 Γ18 = 0). In view of Γ5 = 0, this also holds for ¯F0. In the ratio ¯F0/Fπ, the coupling constant 3 Γ4 − 4 drops out. Hence ¯F0 may be expressed in terms of the scale invariant quantities Lr Fπ, FK and LA. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | between the two angles, but phenomenological input is required to arrive at a reliable result. In this context, the relations (49) and (52) are very useful, because, even if the coupling constant LA is treated as a free parameter, they correlate the difference between the two angles with the magnitude of ¯F0. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | Part of this work was done while one of us (H. L.) stayed at the CSSM. He wishes to thank the organizers of the workshop for hospitality and support. Moreover, we thank Hans Bijnens, Thorsten Feldmann and Bachir Moussallam for informative discussions and comments. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | G. Veneziano, Nucl. Phys. B117 (1974) 519, B159 (1979) 213; R. J. Crewther, Phys. Lett. 70B (1977), 349 and in Field Theoretical Methods in Particle Physics, ed. W. R¨uhl (Plenum, New York, 1980); E. Witten, Nucl.Phys. B156 (1979) 213, B117 (1980) 57; P. Di Vecchia, Phys. Lett. 85B (1979) 357; P. Nath and R. Arnowitt, Phys. Rev. D23 (1981) 473. |
Processed_Pseudoscalar_decay_constants_at_large_N_c.txt | C. Rosenzweig, J. Schechter and T. Trahern, Phys. Rev. D21 (1980) 3388; E. Witten, Ann. Phys. (N.Y.) 128 (1980) 363. |
Processed_Gravitationally_Lensed_QSOs:_Optical_Monitoring_wi.txt | The aim of this contribution is to present the two first phases of the optical monitoring programme of the Gravitational Lenses group at the Universidad de Cantabria (GLUC, http://grupos.unican.es/glendama/). In an initial stage (2003 March-June), the Estaci´on de Observaci´on de Calar Alto (EOCA) was used to obtain V R frames of SBS 0909+532 and QSO 0957+561. These obser- vations in 2003 led to accurate fluxes of the two components of both double QSOs, which are being compared and complemented with data from other 1-1.5 m telescopes located in the North Hemisphere: Fred Lawrence Whipple Observatory (USA), Maidanak Observatory (Uzbekistan) and Wise Obser- vatory (Israel). On the other hand, the GLUC started the second phase of its monitoring programme in 2005 January. In this second phase, they are using the 2 m fully robotic Liverpool Telescope (LT). The key idea is the two- band photometric follow-up of four lensed QSOs with different main lensing galaxies: SBS 0909+532 (elliptical), QSO 0957+561 (giant cD), B1600+434 (edge-on spiral) and QSO 2237+0305 (face-on spiral). Thus, the light rays associated with the components of the four gravitational mirages cross differ- ent galaxy environments, and the corresponding light curves could unveil the content of these environments. While SBS 0909+532 and QSO 0957+561 are the targets for the two first years with the LT (2005-2006), the rest of targets (B1600+434 and QSO 2237+0305) will be monitored starting from 2007. |
Processed_Gravitationally_Lensed_QSOs:_Optical_Monitoring_wi.txt | by the sources), time delay and flux ratio estimates, quantitative study of the structure and chromaticity of the intrinsic signals, determination of chromatic (between two optical filters) delays, and detection of extrinsic fluctuations (due to intervening objects in the involved galaxies, e.g., microlenses or dusty clouds). These measurements are used to tackle three hot astrophysical sub- jects: current expansion rate of the Universe, structure of the main lensing galaxies and nature of the sources (QSOs). The discovery of variable field stars or supernova explosions might be a bonus of the programme. |
Processed_Gravitationally_Lensed_QSOs:_Optical_Monitoring_wi.txt | SBS 0909+532 consists of two components (double QSO) separated by about 1 arcsec [1, 2, 3, 4]. The lensing elliptical galaxy has a large effective radius with a correspondingly low surface brightness [3]. In Fig. 1 (top panel) we show a Maidanak subframe in the R band, where there are two close quasar components, but the very faint galaxy is not apparent. Although the presence of a faint galaxy has a positive aspect: simple photometric model with only two close point-like sources (we can avoid additional complications arising from the use of a galaxy profile), the faint and extended lenses are not a good business for cosmological studies. The relative astrometry of the quasar components (with respect to the centre of the lens galaxy) is not accurate, and this fact plays a crucial role in the estimation of the Hubble constant or the surface density of the lens (see below). |
Processed_Gravitationally_Lensed_QSOs:_Optical_Monitoring_wi.txt | During the Phase I monitoring (EOCA/2003), the set of optical frames cover the period between 2003 March 4 and June 2. Exposures in the V and R Johnson-Cousins filters were taken every night when clear. We have checked (via Wise data) the reliability of the EOCA differential photometry between widely separate and neighbouring stars (see the six field stars in the bottom panel of Fig. 1), and we have concluded that only the relative fluxes for neighbouring objects seem to be reliable. Therefore, as the photometric ruler (empirical PSF to compare with) is the normalised 2D profile of the ”a” star and there is clear evidence of variability of the ”c” star, measurements of the brightness of both quasar components (A and B) are always made with respect to the brightness of the nearby non-variable star ”b”. The R-band EOCA light curves are complemented with R-band fluxes from Maidanak frames, so the global monitoring period and the mean sampling rate (in the R band) are 4 months (∼ 120 days) and one point each 6 days, respectively. These are the first resolved brightness records of SBS 0909+532 [5]. |
Processed_Gravitationally_Lensed_QSOs:_Optical_Monitoring_wi.txt | Fig. 1. Top panel: Maidanak image of SBS 0909+532 (about 8” on a side). We see the two quasar components A (the brightest object) and B (the faintest object), i.e., the GL in the bottom panel. Bottom panel: EOCA image of the SBS 0909+532 field (∼ 7’ × 7’). This field contains the gravitationally lensed quasar (GL) and six bright and non-saturated stars (a-c, s1-s2 and x). |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.