Optimization scheduling of power system under load winter and summer double-peak scenario

In recent years, due to people’s need for a better life and the impact of extreme weather factors, the annual load curve shows a winter-summer double-peak pattern in summer and winter. In addition, due to the large-scale grid integration of renewable energy sources such as wind power, the uncertainty of its output has further increased the net load during the double-peak scenario, putting the power system under even greater pressure to supply electricity. In this paper, we study the day-ahead hierarchical optimal scheduling model of the power system under the double-peak scenario, considering the pumped-storage flexibility space and the extreme boundary flexibility demand. We simulate and analyze the impacts of different operation modes of the pumped-storage power plant on the scheduling results. The simulation shows that the proposed method can give full play to the effect of pumped storage peak shaving and valley filling and guarantee the reliability of the power supply under the double-peak scenario.


Introduction
With China's economic development and the improvement of people's living standards, the electricity consumption of residents and enterprises is growing rapidly.The power system is facing the situation of expanding load peak and valley differences.Especially in winter and summer load peaks, due to the impact of extreme weather factors, temperature-controlled loads grow significantly so that the peakvalley difference in the power grid is further increased; this further increases the difficulty of regulating the power grid [1] .On the other hand, wind and other clean, renewable energy sources are gradually known and widely developed, becoming the core of the energy revolution in the new era [2][3] .Moreover, China has abundant reserves of renewable energy, such as wind power, which has great potential for development.However, the large-scale grid integration of wind power and other renewable energy sources and the uncertainty of its output have put forward higher requirements for the operation and scheduling of the power system.And when there is extreme heat and cold without wind, it will make the net load during the winter and summer double-peak periods further increase.
Therefore, energy storage devices will play an important role in the optimal scheduling of the power system.In the power system, the main functions of pumped storage power plants are peak shaving and valley filling, voltage and frequency regulation, accidental backup, and black start [4] .With the large-scale grid connection of renewable new energy sources, the role of pumped storage power stations in voltage and frequency regulation and accidental backup has become increasingly prominent in order to smooth out the impact of new energy uncertainty [5] .T E represent the lower energy limit and upper energy limit of the pumped flexible space.
(2) Pumped storage units constraints The pumped storage units constraints mainly include upper reservoir capacity constraints, upper and lower upper reservoir storage capacity constraints, upper reservoir storage capacity daily balance constraints, upper and lower pumping and generating power constraints, and unit status constraints. where

Electrochemical energy storage-related constraints.
Electrochemical energy storage-related constraints mainly include electrochemical energy storage capacity constraints, electrochemical energy storage capacity upper and lower limit constraints, electrochemical energy storage daily equilibrium constraints, electrochemical energy storage active output constraints, and electrochemical energy storage state constraints.12 , , , where , where 1 , C is the startup cost of the nth thermal unit at time t.
n n t n t n t n t up Wind turbine 300 A net load dynamic scenario is selected for a given day and its extreme boundaries are determined, as shown in Figure 1.The maximum net load of 2080.56WM occurs at 12:00 in scenario 5.The probabilities for scenarios 1 through 5 are 0.056, 0.162, 0.088, 0.234, and 0.460, respectively.

Analysis of results
Mode 1: The pumped storage power plant adopts the conventional optimal scheduling of power system day-ahead stratified optimal scheduling under the bimodal scenario, with the pumping and storage units pumping and generating at regular intervals.The pumping time is from 0 to 6 o'clock, and the generating time is from 9 to 12 o'clock and 17 to 20 o'clock.Mode 2: The pumped storage power plant adopts day-ahead hierarchical optimal dispatching of the power system under the bimodal scenario with full dispatching.The pumped storage units flexibly determine the pumping time.Mode 3: The pumped storage power plant adopts day-ahead stratified optimal scheduling of the power system under the bimodal scenario with full dispatch.The pumped storage units determine the pumping time flexibly and consider the pumping and storage flexibility space.As can be seen from Table 5, after upper-level dispatch, the total expectation of equivalent net load variance of Mode 2 is reduced by 36.31% compared with Mode 1.The total expectation of equivalent net load variance of Mode 3 is further reduced by 8.64% compared with Mode 2. This indicates that from the perspective of smoothing the fluctuation of peaks and valleys of the net loads, the use of full dispatch for a pumped storage plant is obviously superior to the conventional optimal dispatch and can have a better effect after considering the flexibility of pumping and storage space.The startup cost of thermal power units in the three modes is the same.This indicates that the number of thermal power units started in the three modes is the same as the number of times.The coal consumption cost of the three modes decreases in turn, of which the smallest in Mode 3 is partly due to the fact that the pumped storage sends out part of the power allowed by the flexible space to reduce the thermal power units' output.All three modes of load shedding penalties are 0, and there is no involuntary load cutting.In Figure 2, it can be found that after the participation of the pumped storage power plant and electrochemical storage power plant, the fluctuation of the equivalent net load in all three modes becomes smoother compared with the desired netload.The results show that the participation of a pumped storage plant and an electrochemical storage plant can effectively realize the net load shaving under the bimodal scenario.
In Figure 3, the statistics of the electricity consumed by pumping and the electricity issued by releasing water in the pumped storage power plant on the same day are both improved by 84.65% in Mode 2 compared to Mode 1.They are reduced by 8.61% and improved by 1.55% in Mode 3 compared to Mode 2, respectively.When the pumped storage power station adopts the full dispatch operation mode, the utilization efficiency of the pumped storage power station can be effectively improved.After considering the pumped storage flexible space, the pressure of the energy supply under the bimodal scenario is also reduced to a certain extent.In Figures 4 and 5, all three approaches have more ample flexibility margins.It shows that all three approaches can ensure that the power system still has enough flexibility to regulate when extreme scenarios occur.In Figure 6, there is a small difference in the operating time of each thermal unit in the three modes, but the number of starts of each thermal unit is the same, and the thermal unit output fluctuates more in Mode 1.

Conclusion
In order to cope with the pressure of power system supply preservation under the double-peak scenario, a hierarchical optimal scheduling model under the double-peak scenario is proposed.The impacts of conventional optimal scheduling and full scheduling as well as pumped storage power plant considering pumped storage flexible space on the day-ahead scheduling results are investigated in the simulation under this model, and the following conclusions are drawn.
(1) When a pumped storage power plant adopts the full dispatch mode, it can fully utilize the effect of pumped storage to cut peaks and fill valleys.The effect can be further improved after considering the storage flexible space and, to some extent, alleviating the pressure on the power supply in a double-peak scenario.(2) With the flexibility constraints after considering the extreme boundaries, the power system can have enough flexibility to regulate, guaranteeing the reliability of the power supply in the bimodal scenario.

tE
is the energy storage in the upper reservoir at time t.psf k  and psc k  are the generation efficiency and pumping efficiency of the kth pumped storage unit, respectively., psc kt P is the generating power of the kth pumped storage machine at time t, which is non-negative., psf kt P is the pumping power of the kth pumping unit at time t, which is non-positive.max E and min E represent the maximum and minimum values of the energy storage in the upper reservoir, respectively.0 E and T E are the energy storage at the beginning and end of the schedule, respectively.

Figure 1 .
Figure 1.Net load dynamic scenarios and extreme boundary maps.

Figure 2 .
Figure 2. Comparison of equivalent net loads.Figure 3. Comparison of upper reservoir storage volumes.

Figure 3 .
Figure 2. Comparison of equivalent net loads.Figure 3. Comparison of upper reservoir storage volumes.

Figure 6 .
Figure 6.Expected output of thermal units for the three modes.

A day-ahead optimal scheduling model for power system under a bimodal scenario
represent the load power and wind power at time t of the sth scenario before the scenario cut, respectively.max, P B represent the upper and lower extreme boundaries of the load at time t.max, B represent the upper and lower extreme boundaries of wind power at time t.3. , mtis the amount of electricity stored in the mth energy storage plant at moment t. , Lower-level objective function.The operating cost of a conventional thermal power unit is divided into start-up cost and coal consumption cost.
is the startup state of the nth thermal power unit at time t.represents the generation power of the nth thermal power unit in the sth scenario at time t.
I P

Table 1 .
− represent the upward flexibility and downward flexibility at time t in the sth scenario due to the start and stop of the thermal unit at the next moment in time.−represent the upward and downward flexibility of the power system at time t, respectively.System structure and netload dynamic scenarios.A power system model is constructed, containing thermal power units, pumped storage units, electrochemical energy storage plants, and wind farms, with the specific installed capacity of each unit as shown in Table1.Installed capacity of units.
t F + and t F

Table 2 .
Parameters related to thermal units.

Table 3 .
Parameters related to electrochemical energy storage plants

Table 4 .
Parameters related to pumped storage power plants.

Table 5 .
Objective function results.