The total generation in the 2022 LSP case is 117 GW; with approximately 13.9% instantaneous penetration of wind and 1.1% instantaneous penetration of solar PV. To increase PV penetration the existing synchronous generators are replaced by solar PV systems in each area, so no transmission upgrades are required. The generators are replaced in the sequence of coal power plants, gas generators, hydro generators, and then nuclear generators. As synchronous generators gradually get replaced by PV systems, the net system inertia decreases almost linearly, as shown in Figure 1.15 [2]. The system equivalent inertia level for the base case is 3.4 seconds. The inertia levels for 20, 40, 60, and 80% renewable penetration levels are noted to be 3.17, 2.31, 1.68, and 0.92 seconds, respectively.
The system frequency response is examined in terms of three metrics: (i) ROCOF which reflects the inertia response; (ii) frequency nadir which is determined by both inertia and PFR; and (iii) settling frequency which represents the PFR of the system.
Three severe contingencies are considered as follows:
1 Largest N‐2 contingency involving loss of two largest generators in the Palo Verde nuclear plant, equivalent to a loss of 2525 MW capacity.
2 Typical N‐2 contingency involving loss of two largest generators in the Colstrip coal power plant, representing a loss of 1514 MW capacity.
3 Typical N‐1 contingency involving loss of one large generator in the Comanche power plant, indicative of a loss of 904 MW capacity.
To demonstrate the severity of frequency decline as a result of different renewable penetration levels, models of all the protection systems including UFLS are disabled in this simulation. The frequency response for the largest N‐2 and typical N‐2 contingencies are portrayed in Figure 1.16 [2]. For both contingencies, increasing PV penetration worsens the overall frequency response. It increases ROCOF, lowers frequency nadir, and decreases settling frequency. The 2.6 GW generation loss with 80% PV penetration, causes the frequency to drop below 59.4 Hz triggering the first stage of UFLS. This implies that the Western Interconnection is not capable of hosting 80% solar PV due to its low inertia and diminished frequency response.
Figure 1.15 System equivalent inertia at different PV penetration levels.
Source: Modified from Tan et al. [2].
Figure 1.16 WECC frequency response under high PV penetration scenarios: (a) 2625 MW generation loss, (b) 1514 MW generation loss.
Source: Tan et al. [2].
Frequency response obligation (FRO) for the Western Interconnection is described as the power that the system must provide during the first couple of seconds after an event to prevent the decline of frequency and avoid activation of first stage of UFLS. The FRO is set at 906 MW/0.1 Hz which implies that the power from all the generators should increase by 906 MW for a frequency decline of 0.1 Hz. This FRO is updated by NERC every year, based on system measurements.
Studies on the Western Interconnection reveal that the prescribed FRO is met with 20, 40, and 60% renewables; however, it is not satisfied in case of 80% renewable penetration. The FRO is noted to be 599, 692, and 724 for the largest N‐2 contingency, typical N‐2 contingency, and typical N‐1 contingency, respectively. These values are much lower than the NERC specified FRO of 906 MW/0.1 Hz. This deficit of FRO will endanger the reliability and safety of the interconnected system.
1.2.18.2 Over Frequency Response
An overfrequency event is caused by loss of a large load or an exporting tie line (e.g. HVDC line). Displacement of conventional fossil‐fuel‐based conventional generators by IBRs such solar PV systems decreases system inertia. This exacerbates over frequency response of power systems. The typical behavior of power systems with different levels of stored kinetic energy during an overfrequency event is depicted in Figure 1.17 [11].
Systems with low inertia (e.g. 100 GWs) experience a higher “zenith” (peak frequency) subsequent to loss of a large load, than systems with higher inertia. The ROCOF is positive for all systems but higher for low inertia systems.
Overfrequency events have the following features [11]:
1 Lesser risk of system collapse as compared to underfrequency events.
2 Synchronous generators may provide an unexpected response due to sudden rise in frequency and may disconnect in some cases. Still, there is a low risk of cascaded loss of generation.
3 Less likelihood of disconnection of domestic loads.
Since there are no substantial adverse impacts on the system by overfrequency events, system operators are generally less concerned about these events. Overfrequency events are typically alleviated by reduction in power output from synchronous generators or IBRs.
1.2.19 Angular Stability Issues due to Reduced Inertia
Low‐frequency electromechanical power oscillations (typically 0.1–2 Hz) are recognized as one of the major limiting factors in power transfer over long transmission lines [85]. Conventionally, these oscillations are damped by Power System Stabilizers (PSSs) integrated with synchronous generators. Displacement of fossil fuel‐based steam turbine generators by inertialess IBRs such as solar PV systems results in overall reduction in damping of electromechanical and inter‐area oscillations, as conventional generators [86–89]. This especially becomes a concern during system transients such as faults or large equipment/line outages.
Figure 1.17 Typical behavior of power systems with different levels of stored kinetic energy during an overfrequency event.
Source: Reprinted with permission from EPRI [11].
An example study of reduced damping due to high penetration of solar PV systems is presented in [87]. Eigenvalue analysis and transient stability studies are performed on a test system representing the entire Western Electricity Coordinating Council (WECC) network ranging from 34.5 to 500 kV. The synchronous generators are modeled with excitation systems, PSS, and governors. Solar PV systems comprising both rooftop systems and utility‐scale plants are added in a region with a high potential of their growth. The utility‐scale PV systems are fixed at 600 MW while the amount of rooftop PV systems are varied to achieve different PV penetration scenarios.
The percent PV penetration is considered to be the ratio of total PV generation to total system generation. The solar PV penetration is increased by displacing conventional generators while still keeping critical generators providing reactive power support in service. To maintain the generation–load balance, the outputs of the critical generators are reduced to accommodate the increased penetration of solar PV systems. The rooftop PV systems are modeled with unity power factor while the utility‐scale solar PV systems are represented by full converter model and having reactive power based voltage