欧博百家乐Quantitative assessment and optimization stra
Quantitative assessment and optimization strategy of flexibility supply and demand based on renewable energy high-penetration power system
Download PDF
Download PDF
Quantitative assessment and optimization strategy of flexibility supply and demand based on renewable energy high-penetration power system
,
,
&
…
Show authors
Energy Informatics volume 7, Article number: 117 (2024)
Abstract
With the transformation of the global energy structure, the high penetration rate of renewable energy in power systems has become a trend. This article focuses on the quantitative evaluation and optimization strategies for the flexible supply and demand of renewable energy high-p penetration power systems. Using a combination of data-driven and model simulation methods, the flexibility requirements of the power system after integrating renewable energy are accurately quantified. The impact of uncertainty in renewable energy output on system flexibility was evaluated through system flexibility analysis and scenario construction techniques, and effective flexibility improvement strategies were proposed in combination with optimized scheduling design. The research results show that under high penetration of renewable energy, there is an imbalance between the supply and demand of flexibility in the power system. When the proportion of renewable energy installed capacity reaches 40%, the system flexibility gap reaches 10%. A comprehensive optimization strategy has been proposed to address this issue, including constructing energy storage facilities, demand side response, and virtual power plants. After implementing these measures, the flexibility gap of the system can be reduced to less than 5%, which can effectively ensure the stable operation of the power system.
Introduction
With the acceleration of the global energy transition, the penetration rate of renewable energy in the power system continues to increase. It has become an important component of the future energy structure. However, the intermittency, randomness, and uncertainty of renewable energy sources pose unprecedented challenges to the flexibility of the power system. In order to address these challenges and ensure the safety, stability, and economic operation of the power system, it is particularly important to conduct quantitative evaluation and optimization strategies for the flexibility of renewable energy high-penetration power systems [].
The motivation for this study stems from the severe reality of power system flexibility supply-demand imbalance in the context of high penetration of renewable energy []. With the rapid growth of renewable energy installed capacity, higher requirements have been put forward for the flexible regulation capability of the power system []. However, the existing power system’s flexibility resource allocation, scheduling strategies, and market mechanisms are not yet fully adapted to the high penetration demand of renewable energy, leading to increasingly prominent supply-demand contradictions in flexibility. Therefore, this study aims to quantitatively evaluate the flexibility supply and demand situation of renewable energy high penetration power systems, explore effective optimization strategies, and provide theoretical support and practical guidance for improving the flexibility of power systems.
While reviewing relevant literature, we found that scholars at home and abroad have extensively researched the flexibility of renewable energy power systems, covering multiple aspects such as flexibility definition, evaluation methods, optimization strategies, etc. However, existing research mostly focuses on exploring a single dimension or level, lacking comprehensive quantitative evaluation and comprehensive optimization strategies for the flexibility of the supply and demand relationship of the power system under the background of high penetration of renewable energy []. In addition, with the continuous advancement of renewable energy technologies and the deepening of electricity market reforms, the connotation and extension of power system flexibility are also constantly changing. New research results are urgently needed to guide practice [].
Based on the challenges and shortcomings in the literature, this study is an urgent necessity []. Firstly, by quantitatively evaluating the flexibility supply and demand situation of the renewable energy high penetration power system, the current problems and bottlenecks in the flexibility of the power system can be revealed []. Secondly, proposing effective optimization strategies based on the evaluation results can guide the rational allocation of flexible resources in various aspects, such as planning, design, and operation of the power system, and improve the flexible regulation capability of the power system. Finally, this study also contributes to the improvement and innovation of the electricity market mechanism, providing strong support for the further development of renewable energy and the sustainable operation of the power system.
In response to the identified gap in quantitatively assessing flexibility supply and demand within high-renewable energy penetration power systems, our study develops an innovative framework that integrates advanced data analytics and predictive modeling techniques. This approach not only provides a comprehensive evaluation of system flexibility but also outlines actionable strategies for optimization, ensuring reliable operation amidst growing renewable integration challenges. By employing sophisticated algorithms and real-time data analysis, we pinpoint critical flexibility nodes and quantify their impacts, enabling utilities and grid operators to make informed decisions regarding resource allocation and operational adjustments. The proposed methodology bridges the knowledge gap by offering a scalable solution adaptable to diverse power systems, thereby facilitating a seamless transition towards a more sustainable and resilient energy future.
The novelty of this study is mainly reflected in the following two aspects: firstly, adopting a comprehensive quantitative evaluation method to conduct a detailed analysis of the flexibility supply and demand situation of the renewable energy high penetration power system; Secondly, a series of practical and feasible optimization strategies have been proposed in combination with the reform of the electricity market and the development trend of renewable energy technologies. The main contribution of this study is to provide theoretical support and practical guidance for improving the flexibility of the power system, which will help promote the widespread application of renewable energy and the sustainable development of the power system.
This paper is divided into six chapters. The first chapter is the section, which mainly introduces the research background, motivation, literature review, research necessity, novelty and main contributions, and the organizational structure of the paper. Chapters two and three are the theoretical foundation and methodology section, which elaborates on the theoretical knowledge required for the research and the methods for quantitative evaluation and optimization strategies. Chapter 4 is the specific research content section, which introduces the quantitative evaluation process, optimization strategy design, empirical analysis, and other contents. Finally, the conclusion and outlook section summarizes the research results and proposes future research directions.
Renewable energy high penetration power system and flexibility balance
High penetration power system
Demand response is the mechanism in the power system that guides load adjustment of electricity consumption behavior in response to changes in grid demand through price signals or incentive measures. By receiving electricity price information or responding to instructions, the load adjusts its electricity consumption plan based on its own characteristics and economic interests, achieving a flexible response to the power grid. This helps to balance the supply and demand of the power grid and maintain stable operation.
To simplify the analysis of renewable energy high-penetration systems, we can treat wind and solar power generation as negative electricity demand and subtract these power generation from the original electricity demand [], thus obtaining the calculation formula (1) of net electricity demand:
$${P_{eq,t}}={P_{L,t}} - {P_{W,t}} - {P_{V,t}}$$
(1)
In Eq. (), PW, t is the power of wind farm at time t; PV, t is the power of photovoltaic power plant at time t; PL, t is the power of load demand at time t; and Peq, t is the power of net load demand at time t.
In the power system with a high penetration rate of renewable energy, traditional power sources such as thermal power still need to meet part of the load demand due to the volatility and uncertainty of renewable energy power generation []. This leads to insufficient flexibility of the power system, making it difficult to independently and stably supply power, and requiring the cooperation of other resources to give full play to the advantages of renewable energy and clean energy. In order to improve the efficiency and reliability of the power system, the system needs to be flexible so that it can effectively coordinate various resources and cope with fluctuations in load demand [] .
Figure shows the algorithm flow. Input as new energy generation forecast data (VTS), including renewable energy generation forecast sequences such as wind and solar power; Historical load data: a dataset that reflects the historical load changes of the system; System parameters: including technical parameters related to the power grid capacity, energy storage capacity, transmission limitations, etc. The output is the quantitative evaluation results of flexibility supply and demand, including the quantification of uncertainty in new energy generation, the quantification of system flexibility demand, and the calculation results of flexibility supply and demand gap.
Fig. 1
Algorithm Flow
Full size image
Power system flexibility balance conceptPower system flexibility focuses on how the system can cope with the uncertainty and intermittency of load and renewable energy generation through various regulation means (such as energy storage technology and conventional generating sets) []. This definition of balancing ability emphasizes whether the degree to which the flexibility resources of the system meet the flexibility needs reaches the allowable standard at any moment and direction.
There are three major differences between flexibility balance and traditional power balance: first, at the balance level, traditional power balance pursues equal supply and demand, while flexibility balance allows supply to exceed demand []; In terms of balancing resources, flexibility balance covers four dimensions: source, grid, load and storage, surpassing the power balance that only relies on thermal power and hydropower ; In terms of balance mechanism, power balance focuses on source matching load, while flexibility balance realizes two-way interaction between source and load, including demand response mechanism, so that load can be matched by source or actively matched source [].
Modeling system flexibility requirements
Modeling of wind power output characteristics
The output characteristics of wind power are uncertain due to the strong fluctuation, randomness, and unpredictability of wind speed in the field []. We use the fuzzy C-means clustering algorithm to construct a wind speed model to cope with the wind speed uncertainty. Figure shows our power calculation model, which can effectively handle wind speed uncertainty. Our study uses the smooth spline method to fit the wind power curve, ensuring the smoothness of the curve even when data is limited. The model adopts an encoding input and decoding input module, which is then passed into the correlation analysis module for feature encoding and decoding to obtain the final output.
Fig. 2
Electric power calculation model
Full size image
The wind speed model divides the interval of the actual wind speed v by nodes. This is shown in Eq. (). at represents the identification when the wind speed falls within the t-th interval, t0. tn represents time, and bt represents the upper limit value.
$${a_t}={t_0}<{t_1}< \cdots <{t_j}< \cdots <{t_n}={b_t}\;\;\;{\kern 1pt} j \in [0,n]$$
(2)
According to spline fitting, the wind power model can be obtained as shown in Eq. ():
$${P_{W,t={t_j}}}=F({v_j})={d_j}{({v_j} - {t_j})^3}+{e_j}{({v_j} - {t_j})^2}+{f_j}({v_j} - {t_j})+{\zeta _W}\;\;\;{\kern 1pt} {t_j} \leqslant {v_j} \leqslant {t_{j+1}}$$
(3)
In Eq. (), F(vj) is the fitted value of the wind power curve for the actual wind speed at vj; ζW is the prediction error term of the fitted function, ranging from ζW to N (0, εw); dj, ej and fj are the parameters of the wind power characteristic curve []. Equation () is the wind power calculation step, where vcut, j is the j-th cut-in wind speed; vrate is the rated wind speed.
$$\begin{array}{*{20}{c}} {\nu =\frac{1}{{{{\left( {{\nu _{cut,j}} - {\nu _{rate}}} \right)}^2}}}[2 - \frac{{{{\left( {{\nu _{cut,j}}+{\nu _{rate}}} \right)}^3}}}{{2\nu _{{rate}}^{3}}}]} \\ {=\frac{1}{{{{\left( {{\nu _{cut,j}} - {\nu _{rate}}} \right)}^2}}}[\frac{{{{\left( {{\nu _{cut,j}}+{\nu _{rate}}} \right)}^4}}}{{2\nu _{{rate}}^{3}}} - 3{\nu _{cut,j}} - {\nu _{rate}}]} \\ {=\frac{1}{{{{\left( {{\nu _{cut,j}} - {\nu _{rate}}} \right)}^2}}}[{\nu _{cut,j}}({\nu _{cut,j}}+{\nu _{rate}}) - \frac{{{{\left( {{\nu _{cut,j}}+{\nu _{rate}}} \right)}^4}}}{{2\nu _{{rate}}^{3}}}]} \end{array}$$
(4)
By fitting the wind power curve and analysing it, we are able to accurately calculate the total daily power generation of the wind farm []. As shown in Eq. (). Where tfar, day and tend, day are the first moment and the end moment of the effective output or equipment operation interval during the day, respectively.
$${E_{W,day}} = \int _0^{24}{P_{W,t}}dt = \int _{{t_{far,day}}}^{{t_{end,day}}}{P_{W,t}}dt$$
(5)
Modeling of photovoltaic output characteristicsPhotovoltaic output is influenced by light intensity and temperature; hence it’s calculated using formula (6).
$${P_{V,t}}=\frac{{{f_V}{I_C}}}{{{I_{STC}}}}[{P_{STC}}+{k_p}{P_{STC}}({T_C} - {T_{STC}})]$$
(6)
In Eq. (), fV is the power derating coefficient of the PV system, which is used to calculate the power loss caused by the stains on the PV panel, rain and snow cover, and the aging of the PV panel itself; IC is the actual solar insolation in kW/m2; ISTC is the solar insolation under the standard test conditions; PSTC is the maximum output power of the PV array under the standard test conditions; kp is the power temperature coefficient in %/°C; TSTC is the reference temperature of the PV array under the standard test conditions; TC is the current operating temperature of the PV array in °C, which can be estimated by the current ambient temperature. kp is the power temperature coefficient in %/°C. The expression (7) is as follows:
$${T_C}={T_{AIR}}+30 \times \frac{{{I_C}}}{{{I_{STC}}}}$$
(7)
In Eq. (), TAIR is the current ambient temperature, ℃. Since the prediction of PV output power is usually in error [], the fuzzy number equivalence model is used in this section to deal with the uncertainty of PV, and the PV output power after the numerical equivalence of the affiliation parameter is shown in Eq. ():
$$\left\{ {\begin{array}{*{20}{c}} {{{\tilde {P}}_{V,t}} \to E[{{\tilde {P}}_{V,t}}]=\frac{{1 - {\alpha _V}}}{2}({P_{v1.t}}+{P_{v2.t}})+\frac{{{\alpha _V}}}{2}({P_{v3.t}}+{P_{v4.t}})} \\ {{P_{vj.t}}={k_j}{P_{V.t}}, j=1,2,3,4} \end{array}} \right.$$
(8)
In Eq. (), αV is the confidence level of PV array, 0 ≤ αV ≤ 1; E is the expected calculation; Pv1,t, Pv2,t, Pv3,t, and Pv4,t are the affiliation parameters, which are used to determine the shape of the affiliation function, in which Pv1,t≤Pv2,t≤Pv3,t≤Pv4,t; and kj is the scaling coefficient for the affiliation parameter, 0 ≤ kj≤1. PV Model is used to calculate the total power generation of the PV system in one day, as shown in the following Eq. (), where EV, day is the total power generation in one day.
$${E_{V,day}} = \int _0^{24}{P_{V,t}}dt = \int _{{t_{far,day}}}^{{t_{and,day}}}{P_{V,t}}dt$$
(9)
Flexibility demand modelThere are two main types of load models discussed in this chapter: rigid loads and flexible loads. Rigid load refers to the load type that has no elastic buffer time and does not participate in system scheduling [], which is established by forming a net load curve in Eq. ().
$$\left\{ {\begin{array}{*{20}{c}} {{t_{fir,day}} \times [{U_{lj,\tau + 1,}}{U_{lj,\tau }}] \le \mathop \sum \limits_{t = \tau + 1}^{\tau + {t_{fir,day}}} {U_{lj,t}}}\\ {\tau \in [{t_{fir,day}} - 1,{t_{end,day}} - {t_{fir,day}}]} \end{array}} \right.$$
(10)
In Eq. (), tfir, day is the total intraday energy consumption of rigid loads; \(\:\tau\:\) is the total number of rigid load equipment; \(\:{U}_{lj,\tau\:+1}\) is the power demand of the j-th rigid load equipment, where the Boolean variable of the rigid load equipment is 1 for on and 0 for off.
Fig. 3
System flexibility demand model
Full size image
The system flexibility requirement model can be combined with modeling wind and solar power output, as shown in Fig. , which includes modules such as solar matrix acquisition module, battery module, power distribution, power supply, and data control. Through demand analysis, the power system can be modeled. Flexible load, that is, the demand response of the load, refers to the load type that can be planned and participate in system scheduling and can be used as a power auxiliary power system in specific scenarios. However, the demand response has certain uncertainties due to factors such as users’ willingness to respond. Hence, the influence of demand response needs to be considered when describing the system flexibility demand. Its mathematical model is as follows (11). Evaluate the potential for demand response by calculating the downward adjustment that can be provided by demand response during all periods. By comparing the flexibility contributions of different periods, user groups, or response types, it is possible to identify which parts of the demand response resources have the most potential, thereby guiding resource development and policy formulation. Based on the results, power system dispatchers and planners can develop more flexible and efficient scheduling strategies. Optimize the start-stop plan of generators, adjust the charging and discharging strategies of energy storage systems, or guide users to change their electricity consumption behavior based on the flexibility potential of demand response to match supply and demand better and reduce system costs.
$$\left\{ {\begin{array}{*{20}{c}} {{P_{DR,t}} = \mathop \sum \limits_{j = 1}^{{N_{DR}}} {P_{drj,t}} + {\zeta _{pare,t}}}\\ {{\zeta _{pare,t}} \sim N(0,1)} \end{array}} \right.$$
(11)
In Eq. (), NDR is the total number of demand response devices; Pdrj, t is the power of the j-th demand response device; and ζpare, t is the power error expected to be curtailed or shifted at the time t. Combining the above models, the system flexibility requirement expression can be obtained as shown in Eq. (), where Dfle is the system flexibility requirement, P represents electrical power, and dt represents differential operation.
$${D_{fle}} = \int _{{t_{\theta rdw}}}^{{t_{end,dey}}}({P_{L,t}} - {P_{W,t}} - {P_{V,t}} - {P_{DR,t}})dt$$
(12)
System flexibility assessment modelEfficient supply probability modelAt present, the flexibility resources of the power system are still centered on thermal power units, supplemented by energy storage devices as regulation systems and conventional thermal power units are the main source of flexibility. In order to assess the flexibility of the power system, the balance between supply and demand needs to be measured [, ]. This study mainly discusses the performance of thermal power units in three situations: maintenance and shutdown, technical limitations and different operating conditions. How the unit adjusts its output power, i.e., “climbing speed,” in these cases is studied. At the same time, two models are also established: one is a probability distribution model that describes the ability of the unit to adjust power upward and downward (upward and downward climbing capacity) []; The other is a probability distribution model that describes a unit’s ability to provide flexibility []. Specifically, it is shown in the probability distribution (13).
$${\varphi _{up.l.j}}(x;{\tau _G})={\varphi _{up.l.j}}({r_{up.l.j}};{\tau _G})=\left\{ {\begin{array}{*{20}{c}} {{\tau _G} \times {r_{up.l.j}}, \;\;{\kern 1pt} 1 - {\sigma _{fo.l}}} \\ {0, \;\;{\kern 1pt} {\sigma _{fo.l}}} \end{array}} \right.$$
(13)
\(\:{\phi\:}_{up,l,j}\) and \(\:{r}_{up,l,j}\) are the effective upward and downward ramping capacity probability distributions of unit l in the j-th output interval, respectively; \(\:{\tau\:}_{G}\) is the effective upward and downward ramping capacity distributions of the system in the j-th output interval after the loading of the first i thermal units, respectively. The effective climbing capacity distribution is given in Eq. () below:
$$\left\{ {\begin{array}{*{20}{c}} {\varphi _{{R,i,j}}^{+}(x;{\tau _G})={\varphi _{up,1,j}}(x;{\tau _G}) \times {\varphi _{up,2,j}}(x;{\tau _G}) \times \cdots \times {\varphi _{up,i,j}}(x;{\tau _G})} \\ {\varphi _{{R,i,j}}^{ - }(x;{\tau _G})={\varphi _{dn,1,j}}(x;{\tau _G}) \times {\varphi _{dn,2,j}}(x;{\tau _G}) \times \cdots \times {\varphi _{dn,i,j}}(x;{\tau _G})} \end{array}} \right.$$
(14)
Flexibility evaluation indexFlexibility margin refers to the part of the power system where the flexibility resources exceed the flexibility demand, which can represent whether the flexibility of the system is sufficient []. Its criterion is the probability that the flexibility supply capacity is greater than the flexibility demand, and its form is as follows (15):
$${\varphi _M}(z)={\varphi _S}(x) \odot {\varphi _D}(y)$$
(15)
In Eq. (), \(\:{\phi\:}_{M}\left(z\right)\) is the probability density function of the flexibility margin; \(\:{\phi\:}_{S}\left(x\right)\) and \(\:{\phi\:}_{D}\left(y\right)\) are the probability density functions of the supply and demand for flexibility, respectively; y and z are random variables; \(\:\odot\:\) is the roll-up operation; and ξ is the sufficiency level. The flexibility margin probability density is shown in Eq. (), where τ is the current time period.
$${\varphi _{_{M}}}(z, {P_{_{{L,t}}}})={\varphi _{_{S}}}(x, {P_{_{{L,t}}}}) \odot {\varphi _{_{D}}}(y, {P_{_{{L,t}}}})$$
(16)
If the full probability Eq. () is used, it can be derived from the above equation, where PL, max and PL, min are the maximum and minimum values of load demand in Eq. ().
$${\varphi _M}(z|\tau ) = \int _{{P_{L,\min }}}^{{P_{L,\max }}}f({P_{L,t}}){\varphi _M}(z|{P_{L,t}};\tau )d{P_{L,t}}$$
(17)
This results in a margin value that reflects the adequacy of system flexibility, as shown in Eq. (), where EM(z) is the reactant inventory margin and \(\:{\phi\:}_{M}\) is the proportion of spare capacity. The probability of insufficient system flexibility is shown in Eq. (), where Ki(z) is the flexible operating strategy diversity score and yz(i) is the operating strategy numerator.
$${E_M}\left( z \right) = \int _{ - \infty }^{ + \infty }z{\varphi _M}\left( z \right)dz$$
(18)
$${K_i}\left( z \right) = \int _{ - \infty }^{ + \infty }{y_z}\left( i \right)dy$$
(19)
E+ fle(x) and E− fle(x) are the upward and downward climb capacity shortfall expectations of the system, respectively. This metric reflects the amount and severity of the system flexibility shortfall and expression (20) is given below:
$$\left\{ {\begin{array}{*{20}{c}} {E_{fle}^ + (x) = \smallint _0^{{\rm{\Delta }}{D_{fle.t}}}x \cdot p_{fle}^ + (x)dx}\\ {E_{fle}^ - (x) = \smallint _0^{{\rm{\Delta }}{D_{fle.t}}}x \cdot p_{fle}^ - (x)dx} \end{array}} \right.$$
(20)
Experimental analysis of high proportion renewable energy power system
Numerical example analysis
This section will examine how the impact of wind and solar power forecast errors on system flexibility can be assessed during power peak shaving. Figure shows the analysis of demand changes. The uncertainty of wind and solar power generation must be considered to meet the need for system flexibility []. We employ an approach to address these forecasting errors by building an ensemble containing multiple generation scenarios that fuse historical generation data and forecasted data to analyze the forecasted horizons for wind and PV generation [].
Fig. 4
Changes in demand
Full size image
When analyzing the system flexibility demand, the renewable energy forecast error is used to quantify the uncertainty of power generation, and the required flexibility level is determined by integrating the load demand data of the system. Since changes in wind speed, sunshine and temperature in different seasons will affect the output of wind power and photovoltaic power generation, as well as changes in load, we have selected the four months of March, June, September and December throughout the year to study the flexibility requirements of each month under typical peak shaving scenarios []. The results of the correlation analysis are shown in Fig. . As can be seen from Fig. , in order to more clearly demonstrate the significant differences in system flexibility requirements in each quarter, we conducted a detailed statistical analysis for each quarter under typical peak shaving scenarios.
Fig. 5
Cost changes
Full size image
Table 1 Flexibility analysis
Full size table
According to the data in Table , the flexibility requirements of the system in each season have their characteristics: the demand power in winter is the highest, reaching 20012.17 MW, and the standard deviation is large, indicating a drastic change in demand. This is mainly due to the low temperature in winter in the northwest region, which leads to a decrease in photovoltaic power generation and supply capacity. The demand for power in summer is the lowest, only 18529.227 MW, while the demand fluctuation in spring is the most severe, with the highest standard deviation and the highest average demand power, indicating that the overall demand level in spring is relatively high. In contrast, the standard deviation in autumn is the smallest, indicating that demand fluctuations are relatively stable, although its average demand level is still relatively high. These differences reveal the characteristics of system flexibility requirements in different seasons. The system flexibility requirements are roughly normally distributed, mainly concentrated in the power range of 17.5GW to 20GW. When formulating scheduling strategies, it is necessary to ensure that the main power sources can cover the area and evaluate the flexibility supply capacity of upward and downward adjustments based on this. If thermal power units are the main power source with an installed capacity of 19GW, it is more suitable to equip about 2GW of energy storage devices; that is, the ratio of thermal power to energy storage configuration is about 10:1.
Analysis of the impact of renewable energy penetration rateThe penetration rate of renewable energy sources is a key factor when analyzing system flexibility []. In order to evaluate the impact of renewable energy grid connection on supply-side flexibility demand, as well as the supply-side shortage when facing flexibility demand at different times during the day, we conducted data analysis. From the average utilization rate of the infrastructure network in Fig. , it can be observed in the block diagram that before the renewable energy grid connection, the supply-side power supply was insufficient to cope with load peaks, resulting in a large upward flexibility shortage in the system during the two time periods of 9:00 to 12:00 am and 6:00 to 10:00 pm. Compared with the upward flexibility gap, the downward flexibility gap is less, which mainly occurs during the load trough period from 2:00 to 5:00 in the morning. Although the integration of renewable energy into the grid has improved the flexible supply capacity of upward adjustment and reduced the upward adjustment shortage, due to the high-power generation of wind power in the early morning, the downward adjustment shortage has increased [].
Fig. 6
Block diagram of average utilization of infrastructure network
Full size image
In order to more clearly demonstrate the impact of renewable energy grid integration on system flexibility, we analyzed the changes in flexibility indicators as the penetration rate of renewable energy gradually increases. From the transformed uncertainty perception in Fig. , it can be concluded that as the proportion of renewable energy in the power system increases, the risk of insufficient upward adjustment capacity (i.e., the ability to rapidly increase power generation) of the system decreases, but the risk of insufficient downward adjustment capacity (i.e., the ability to rapidly reduce power generation) increases. Especially after the penetration rate of renewable energy reaches 50%, this impact is more obvious. When the renewable energy penetration rate is 0%, the risk of system upward adjustment is greatest. When the penetration rate of renewable energy is 100%, the risk of downward adjustment of the system is the greatest.
Fig. 7
Uncertainty perception of transformation
Full size image
In addition to the impact of renewable energy penetration rate, it is also necessary to consider reducing the error of renewable energy forecast output in actual operation, which may be adjusted by wind abandonment, light abandonment or load shedding []. Under the background that the penetration rate of renewable energy is as high as 35%, in order to ensure the stability and reliability of the power system, we need to deeply explore the evolution of flexibility indicators in various scenarios. Referring to the “Energy Flexibility Analysis without Baseline” in Fig. , we find that after the implementation of a certain measure, the climbing pressure and flexibility demand caused by the inaccuracy of wind power output forecast are alleviated, causing various related indicators to show a downward trend. Specifically, by reducing the expectation of insufficient flexibility, we achieved a significant decrease of 15.5%.
However, in the field of photovoltaic power generation, despite the strategy of abandoning light, its impact on photovoltaic power generation at noon is relatively concentrated, and its effect on alleviating the upward climbing pressure of traditional units is not obvious []. Therefore, the expectation of insufficient upward climbing and the expectation of insufficient upward flexibility has only been reduced by 1.5% and 1.3%, respectively. These two sets of figures are obviously dwarfed by the improvement effects under wind curtailment conditions-4.8% and 3.4%. This shows that under different renewable energy utilization conditions, changes in flexibility indicators and their impacts on the system are different and need to be treated differently in actual operation. Overall, the effect of wind abandonment is better than that of light abandonment. Under load-limiting conditions, because the renewable energy power generation is small relative to the total load, and the load-limiting amount exceeds the wind/photovoltaic curtailment, the load-limiting measures have the best effect.
Fig. 8
Energy flexibility analysis without baseline
Full size image
Optimal scheduling design considering variable period control
Numerical example analysis
In this example, the characteristics of a provincial power grid in northwest China are as follows: the maximum adjustable output of thermal power units is 18,040 MW. Renewable energy accounted for 44.33%, including 6,900 MW of wind power and 2,652 MW of photovoltaics. There is a 1200 MW pumped storage power station. The peak daily load of the system reaches 23,540 MW. The maximum adjustable capacity for demand response is 800 MW.
Effect analysis of variable time scale considering variable time periodWhen constructing a virtual time series (VTS) based on a new energy high-penetration power system, we fully consider the uncertainty of new energy generation. We quantified the volatility of new energy generation predictions or actual outputs by calculating the standard deviation of each data point in VTS. The size of the standard deviation directly reflects the degree of dispersion of new energy generation data in VTS, that is, the range of changes and uncertainty level of predicted or actual power generation. A larger standard deviation means that new energy generation has higher volatility, which puts higher demands on the flexibility of the power system.
Through the comparison of two days’ data, it is found that the virtual time series (VTS) has a lower value in the thermal power dispatch plan tracking index than the fixed time series (FTS), regardless of whether pumped storage power stations (pumped storage) are involved, indicating that VTS has a better effect in thermal power dispatch plan tracking []. However, the model-solving time of VTS is longer than that of FTS. The reasons for this difference will be analyzed in depth next. Figure shows the planned thermal power output of wind farm groups processed at different time scales with or without the participation of pumped storage.
Fig. 9
Wind farm group with spatial resolution
Full size image
In Fig. , STRL forecasts accurately match net load demand, directly reflecting thermal power output needs without pumped storage. When the FTS method is used to formulate the day-ahead dispatch plan, there is a significant mismatch between the planned output curve of thermal power and the actual demand. Optimizing power system operation requires precise planning of power generation output. Excessively high output surpasses net load demand, resulting in unnecessary wind and solar energy waste. On the contrary, if the planned power generation output is too low, the system may not be able to meet the demand, and then load-shedding measures need to be taken, that is, power supply to certain users is reduced to balance supply and demand. Both scenarios affect the efficiency and economics of the power system and, therefore, need to be avoided through accurate forecasting and scheduling. In both cases, the greater the difference between the required output and the day-ahead scheduling plan, the higher the total system cost in actual operation.
Fig. 10
Prediction using STRL at different data resolutions
Full size image
The VTS method introduced in this chapter can flexibly adjust the precision of day-ahead dispatching time according to the net load characteristics, optimize the planned output of thermal power, and make it closer to the actual demand. This leads to the decrease of scheduling plan tracking indicators by 284.7% and 224.1%, respectively, compared with the FTS method. It is worth noting that this optimization process did not significantly increase the computational burden, and the solution time only increased by 0.51 s and 0.4 s, respectively. Therefore, the effectiveness of the proposed strategy in maintaining system operating performance (SOPs) can be concluded. This paper further discusses the effectiveness of the traditional strategy for pumped storage power stations (pumped storage), assisting thermal power units according to preset action reference values, and using fixed time series (FTS) for daily day-ahead scheduling. By analyzing the performance of data sets with different data resolutions in Fig. , the effectiveness of Strategy 5 in improving the efficiency of pumped storage power stations and reducing system operating costs can be quantitatively evaluated. At the same time, this analysis will also reveal the comparison between this strategy and other strategies in terms of SOP indicators and annual utilization hours, which will help to fully understand the advantages and disadvantages of each strategy.
Fig. 11
Performance on datasets with different data resolutions
Full size image
Compared with the traditional strategy and the latest strategy, the system performance of this research strategy is improved by 53.82% and 17.04% respectively. In terms of annual utilization hours, the increase rates were 55.93% and 19.94% respectively. We evaluate the impact of different strategies by comparing the output curve of pumped-storage power stations with the change curve of SOPs (system operating points). As shown in Fig. , the traditional strategy aims to balance the grid by adjusting net load fluctuations throughout the day, reducing the peak-to-valley difference of thermal power units. However, pumped-storage power stations concentrate on pumping water and generating electricity at noon and night, which leads to violent fluctuations in SOPs and frequent exceeding of safe operation limits, which not only threatens the security of the grid, but also minimizes the annual utilization hours of pumped-storage power stations.
Fig. 12
SOPs variation curves of pumped-storage power stations under different strategies
Full size image
Traditional strategies lack demand response support, resulting in SOPs temporarily exceeding the limit at noon and night hours, resulting in poor maintenance effects. In order to improve the utilization rate of pumped-storage power stations and maintain good SOPs status, the strategy proposed in this paper achieves this goal through demand response, which makes the annual utilization hours under this strategy rank first among the three strategies.
Conclusion
With the transformation of the global energy structure to low carbon, the power system with high penetration of renewable energy has become an important direction for future development. However, the randomness and volatility of renewable energy have brought unprecedented challenges to the power system. How to quantitatively evaluate and optimize the flexibility supply and demand of renewable energy high-penetration power system has become an urgent problem to be solved. In this study, the renewable energy high-penetration power system is deeply analyzed by establishing a mathematical model and simulation platform. The research results show that under the condition of high penetration of renewable energy, there is an imbalance between the supply and demand of power system flexibility.
In response to this problem, this study proposes a series of optimization strategies. By building a certain scale of energy storage facilities, the excess renewable energy can be stored for emergency needs. At the same time, through demand-side response technology, users can be guided to reduce electricity consumption when the power supply is tight and alleviate supply and demand pressure. The experiment found that the penetration rate of wind power has increased at an average annual rate of 10% in the past five years, and is expected to exceed 30% in the next five years. The daily volatility of wind power output increased by an average of 20%, and the volatility of power grid in high penetration areas increased by 25%. Through quantitative analysis, the demand for power system flexibility surges during peak hours, exceeding 30% of total demand. In view of the insufficient supply of flexible resources such as traditional thermal power, after the introduction of energy storage technology, the system flexibility is increased by about 20%, and the multi-energy complementary system can reduce the system failure rate by about 15%, effectively alleviating the dispatching pressure. Through intelligent charge and discharge control, the energy storage system smooths wind power fluctuations, improves system stability by 20%, reduces dispatching errors by 20%, and enhances system stability. The dynamic electricity price mechanism guides users to use electricity during high wind power periods, reducing peak load by 15%. In terms of demand-side management, the implementation of time-of-use electricity prices and demand response strategies has reduced the peak-to-valley difference of system load by about 10%, further balancing wind power output and load demand.
Overall, this research paves the way for enhanced understanding and effective management of the flexibility dynamics of large amounts of renewable energy connected to the grid. Further development of artificial intelligence-driven predictive modelling and real-time control mechanisms is expected to improve the accuracy of flexibility assessments, thereby supporting more flexible power system operations. Future exploration of the socio-economic impacts of flexibility measures and their global scalability will be crucial, which will contribute to an inclusive and sustainable energy transition on a global scale.
Data availability
The data supporting the findings of this study are available within the article.
References
Wang X, Wang Y, Liu Y (2020) Dynamic load frequency control for high-penetration wind power considering wind turbine fatigue load. Int J Electr Power Energy Syst, 117
Guo H, Zheng S, Zhang D, Gao P, Miao W, Zuo Z (2023) Effects on frequency Stability of Power System for Photovoltaic high-penetration ratio Grid-Connected Power Generation. Energies, 16(3)
Yu C, Li Y, Liu Y, Ge L, Wang H, Luo Y, Pan L (2023) Dispatch of highly renewable energy power system considering its utilization via a data-driven bayesian assisted optimization algorithm. Knowl Based Syst, 281
Yan K, Li G, Zhang R, Li F, Jiang T, Li X, Chen H (2023) Aggregated SFR model for VSC-HVDC interconnected power systems with high penetration of wind power. Electr Power Syst Res, 216
Williams DO, Megahed TF, Abdelkader SM (2022) Increasing Green Energy Penetration and efficient utilization through a Finite Control Set Model Predictive-based virtual Inductance Droop Control. Electric Power Systems Research, p 210
Google Scholar
Song W, Wang L, Zhao W, Zhang X, Wang Z (2022) Inertia Optimization Control and Transient Stability Analysis of Wind Power Grid-Connected System. Front Energy Res, 9
Zhang N, Jia H, Hou Q, Zhang Z, Xia T, Cai X, Wang J (2023) Data-Driven Security and Stability Rule in High Renewable Penetrated Power System Operation. Proceedings of the Ieee, 111(7), 788–805
Wu Q-H, Bose A, Singh C, Chow JH, Mu G, Sun Y, Liu Y (2023) Control and Stability of Large-Scale Power System with highly distributed renewable Energy Generation: viewpoints from six aspects. Csee J Power Energy Syst 9(1):8–14
Google Scholar
Rahman MJ, Tafticht T, Doumbia ML (2022) Power Stability and Frequency Control Techniques of DG for a high penetration wind-based Energy Storage System using integral-derivative Controller. Ieee Can J Electr Comput Eng 45(3):232–241
Article Google Scholar
Zhao Z, Su X, Yang J, Zhou P, Chen H, Lu G (2022) Droop-based active voltage regulation control and robust stability analysis for VSCs in the large-scale RES-integrated power system. Front Energy Res, 10
Li W, Cai D, Wu S, Zhang G, Zhang F (2023) The impact of supplementary active power control of wind turbine on power system low frequency oscillations. Electr Power Syst Res, 224
Xing C, Liu M, Peng J, Wang Y, Liu Y, Gao S, Liao J (2024) Disturbance frequency trajectory prediction in Power systems based on LightGBM Spearman. Electronics, 13(3)
Xiong X, Wu X (2022) Coordinated control of heat-power integrated energy system using zone model predictive control with variable zone width. Appl Therm Eng, 217
Yu G, Xu J, Wu G, Zhang Z, Zhao D (2023) Improved virtual synchronous generator control strategy for the flexible interconnection system in distribution transformer areas. Electric Power Systems Research, p 214
Google Scholar
Qian T, Shi F, Wang K, Yang S, Geng J, Li Y, Wu Q (2022) N-1 static security assessment method for power grids with high penetration rate of renewable energy generation. Electr Power Syst Res, 211
Liu Z, Liu Y, Xu H, Liao S, Zhu K, Jiang X (2022) Dynamic economic dispatch of power system based on DDPG algorithm. Energy Rep 8:1122–1129
Article Google Scholar
Salman UT, Shafiq S, Al-Ismail FS, Khalid M (2022) A review of improvements in Power System Flexibility: implementation, Operation and Economics. Electronics, 11(4)
Adewole AC, Rajapakse AD, Ouellette D, Forsyth P (2022) Protection of active distribution networks incorporating microgrids with multi-technology distributed energy resources. Electr Power Syst Res, 202
Khajeh H, Laaksonen H, Simoes MG (2023) A fuzzy logic control of a smart home with energy storage providing active and reactive power flexibility services. Electr Power Syst Res, 216
Enshasy H, Al-Haija QA, Al-Amri H, Al-Nashri M, Al-Muhaisen S (2019) A Schematic Design of HHO Cell as Green Energy Storage. Acta Electronica Malaysia 3(2):09–15
Article Google Scholar
Wang L, Jiang S, Shi Y, Du X, Xiao Y, Ma Y, Li M (2023) Blockchain-based dynamic energy management mode for distributed energy system with high penetration of renewable energy. Int J Electr Power Energy Syst, 148
Velasco JA, Amaris H, Alonso M (2020) Deep learning loss model for large-scale low voltage smart grids. Int J Electr Power Energy Syst, 121
Wang T, Wang S, Ma S, Guo J, Zhou X (2024) An Extended Continuation Power Flow Method for Static Voltage Stability Assessment of Renewable Power Generation-Penetrated Power Systems. IEEE Trans Circuits Syst Ii-Express Briefs 71(2):892–896
Google Scholar
Xingchen S-G, He Y, Zhang C, Jiang L, Spencer JW, Wu M (2020) Sampled-data based discrete and fast load frequency control for power systems with wind power. Appl Energy, 259
Li Y, Huang J, Liu Y, Ni Z, Shen Y, Hu W, Wu L (2022) Economic Dispatch with high penetration of wind power using Extreme Learning machine assisted Group Search optimizer with multiple producers considering Upside potential and Downside Risk. J Mod Power Syst Clean Energy 10(6):1459–1471
Article Google Scholar
Lim S, Choi D, Lee SH, Kang C, Park J-W (2022) Frequency Stability Enhancement of Low-Inertia large-Scale Power System based on Grey Wolf Optimization. IEEE Access 10:11657–11668
Article Google Scholar
Aljarrah R, Al-Omary M, Alshabi DA, Salem Q, Alnaser S, Ćetenović D, Karimi M (2023) Application of artificial neural network-based tool for short circuit currents estimation in power systems with high penetration of power electronics-based renewables. IEEE Access 11:20051–20062
Article Google Scholar
Khalaf MM (2020) Algorithms and optimal choice for power plants based on M-Polar fuzzy soft set decision making criterions. Acta Electronica Malaysia 4(1):11–23
Article Google Scholar
Luo F, Sun C, Yang X, Zhang T, Wang C, Xiao J, Qian M (2023) HPDS133-bus: a benchmark test system for distribution systems with high penetration of distributed generation. Csee J Power Energy Syst 9(3):963–977
Google Scholar
Li Z, Liu H, Zhao J, Bi T, Yang Q (2021) Fast power System Event Identification using enhanced LSTM Network with Renewable Energy Integration. IEEE Trans Power Syst 36(5):4492–4502
Article Google Scholar
Hassan A, Aly MM, Alharbi MA, Selim A, Alamri B, Aly M, Mohamed EA (2023) Optimized Multiloop Fractional-Order Controller for regulating frequency in diverse-sourced vehicle-to-Grid Power systems. Fractal Fract 7(12):864
Article Google Scholar
Download references
Acknowledgements
Not Applicable.
Funding
No funding.
Author information
Authors and Affiliations
Xuchang Vocational Technical College, Xuchang, 461000, China
Liangliang Zhang, Yimin Chu, Yanhua Xu & Wei Guo
Authors
Liangliang Zhang
View author publications
Search author on:PubMed Google Scholar
Yimin Chu
View author publications
Search author on:PubMed Google Scholar
Yanhua Xu
View author publications
Search author on:PubMed Google Scholar
Wei Guo
View author publications
Search author on:PubMed Google Scholar
ContributionsL.Z. investigation; Y.C. data curation & methodology; Y.X. methodology; W.G. visualization; writing—review and editing.
Corresponding authorCorrespondence to Wei Guo.
Ethics declarations
Consent for publication
Not applicable.
Consent to participateNot applicable.
Ethical approvalNot applicable.
Competing interestsThe authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit .
Reprints and permissions
About this article
Cite this article
Zhang, L., Chu, Y., Xu, Y. et al. Quantitative assessment and optimization strategy of flexibility supply and demand based on renewable energy high-penetration power system. Energy Inform 7, 117 (2024). https://doi.org/10.1186/s42162-024-00431-2
Download citation
Share this article
Anyone you share the following link with will be able to read this content:
Get shareable linkSorry, a shareable link is not currently available for this article.
Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative
Keywords
Download PDF
Advertisement