A method for calculating the new energy consumption capacity based on the output characteristics of new energy

Abstract

The present invention relates to the technical field of power system dispatching, and particularly to a method for calculating the new energy consumption capacity. Calculating the new energy utilization rate under different boundary conditions can best represent the new energy consumption capacity. Then, based on the new energy output within a certain specified time period, combined with the magnitude of the grid load and external power transmission, and the regulation capacity of conventional power sources, the theoretical power and actual power of the new energy are determined, and further the new energy utilization rate is determined. The present invention realizes the calculation of indicators such as new energy power generation, power generation ratio, and utilization rate. Compared with the traditional time series method, on the basis of taking into account accuracy and rapidity, it realizes the determination of three coupling relationships: installed capacity time series, peak shaving capacity, and new energy utilization rate.

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching, and particularly to a method for calculating the new energy consumption capacity based on the output characteristics of new energy. Background Technique

[0002] The output of new energy has strong volatility. Conventional power sources not only need to follow the changes in load but also balance the fluctuations in the output of wind and light. To ensure power consumption during low output of new energy, it is necessary to increase the startup of conventional power sources, which will squeeze the consumption space during high output of new energy on the same day. On the contrary, if the startup of conventional power sources is reduced to ensure the consumption during high output of new energy, it will inevitably lead to insufficient power supply capacity during low output of new energy. Therefore, in a power system with a high proportion of new energy, there will be a situation where electricity is wasted during peak photovoltaic output at noon and a serious shortage of power supply during the evening peak, posing a great challenge to the safe operation and stable supply of the power system. To achieve the dual-carbon goal, it is inevitable to increase the power generation proportion of new energy, which will further increase the peak-valley difference of the power grid and put forward higher requirements for the peak regulation ability of the power grid. Therefore, accurately evaluating the new energy consumption capacity in long cycles such as months and years is of great significance for guiding the grid connection layout and timing of new energy, overall arranging the national power and electricity plan, planning the new energy transmission grid, establishing a high-proportion new energy power market, and formulating new energy consumption measures.

[0003] There are mainly three existing methods for calculating the new energy consumption capacity and power and electricity balance of the power system: the traditional typical day analysis method, the time series production simulation method commonly used in recent years, and the newly emerged stochastic production simulation method. The typical day analysis method is beneficial for analyzing the consumption capacity of the power grid as a whole, but it does not consider the randomness and volatility of new energy. The time series production simulation method calculates the new energy consumption capacity per hour by optimizing the unit startup in multiple weeks. The calculation method is relatively mature and is commonly used in power production planning. However, it requires a large amount of data, complex calculations, and is extremely unfavorable for sensitivity analysis after changes in wind and light resources, installed capacity, and electricity consumption. At the same time, it cannot reverse the installed capacity based on targets such as the new energy power generation proportion and new energy utilization rate. The method based on stochastic production simulation greatly improves the calculation speed of the new energy consumption capacity through operations between probability distributions, but it also brings the problem of a slight decrease in accuracy, and it does not solve the problem of reversing the installed capacity based on the target.

[0004] The biggest difficulty and the most time-consuming part of the chronological production simulation method lie in how to arrange the output of conventional units through optimal planning. The conventional chronological simulation method needs to model each unit to form constraint conditions, plan the startup mode for each time period, and ensure the maximum consumption of new energy under the premise of meeting the maximum load demand. Whether it is the modeling of units or the optimization of startups, it is easy to cause deviations and an increase in the amount of calculation. When analyzing the new energy consumption capacity by the chronological production simulation method, the chronological output curve of the theoretical power of new energy should be adopted, but how to form this curve is already a major difficulty.

[0005] To sum up, there is a contradiction between the accuracy rate and the calculation speed in the current new energy measurement methods, and none of them can reverse the installation time sequence and layout through the power generation ratio and utilization rate targets. And this is exactly the function urgently needed in the process of vigorously developing new energy in the future, which is to not only increase the proportion of new energy but also ensure the safe operation and reliable power supply of the power grid. This method will establish a model for forming the chronological output curve of the theoretical power of new energy, adopt the aggregation method to reduce the amount of calculation for optimizing the startup of thermal and hydro power units, and propose calculation methods for the power generation ratio and peak shaving capacity demand of new energy, which are used for the planning of the installation scale, time sequence, and layout of new energy. Summary of the Invention

[0006] The purpose of the present invention is to propose a double-layer stochastic optimization method for local consumption of new energy in multiple scenarios to solve the problems of low accuracy and economy in the source-grid-load-storage planning in the prior art.

[0007] To achieve the above purpose, the specific steps of the technical solution adopted by the present invention are as follows:

[0008] A method for measuring the new energy consumption capacity based on the output characteristics of new energy calculates the most representative new energy consumption capacity through the new energy utilization rate under different boundary conditions, and then, based on the new energy output within a certain specified time period, combines the magnitude of the grid load and power transmission and the regulation capacity of conventional power sources to determine the theoretical electricity and actual electricity of new energy, and further determine the utilization rate of new energy.

[0009] A method for measuring the new energy consumption capacity based on the output characteristics of new energy includes the following steps:

[0010] S1: Analyze the output characteristics of new energy based on historical data, determine the normalized 8760-point sequence of new energy output, and then form the theoretical output sequence of new energy according to the installed capacity;

[0011] S2: Form the new energy consumption space through section constraints, peak shaving constraints, conventional unit output constraints, and load characteristics;

[0012] S3: Based on the new energy theoretical output sequence obtained in step S1 and combined with the new energy accommodation space obtained in step S2, establish a model for measuring new energy accommodation.

[0013] S4: Find the optimal solution of the model by adjusting various parameters including the startup conditions of conventional units, then adjust the scale, layout, and timing of the newly added new energy installed capacity to measure the final new energy electricity quantity and utilization rate. Finally, through multiple adjustments, form the coupling relationship between the grid installed capacity, layout, and timing and the final new energy electricity quantity and utilization rate.

[0014] Further, the utilization rate of the new energy is:

[0015]

[0016] In the formula, Q fact represents the actual power generation of new energy after being restricted by the grid accommodation capacity; Q theo represents the theoretical power generation of new energy, that is, the power that new energy can generate when the grid accommodation capacity is large enough.

[0017] Further, in step S1, the theoretical output sequence of the new energy adopts the normalized time series of historical new energy theory power, and is matched hour by hour in combination with the change of installed capacity to form the theoretical output of new energy in the measurement period.

[0018] P NE =[P NE,1 , P NE,2 , P NE,3, ......, P NE,8760 T (1)

[0019]

[0020] In the formula, define the matrix P NE as the hourly output sequence of new energy, is the hourly resource normalization sequence of wind and light, S wind , S thotov are the hourly installed capacity sequences of wind and light, and the matrices are all in the structure of 1×8760. S wind,j represents the wind power installed capacity corresponding to the jth hour; S thotov,j represents the photovoltaic installed capacity corresponding to the jth hour; respectively represent the output and installed capacity of wind power in the jth hour of the previous year; respectively represent the output and installed capacity of photovoltaic power in the jth hour of the previous year.

[0021] Further, the sum of each item of the theoretical output sequence of the new energy exactly corresponds to the annual resource hours, that is ​is the annual wind resource hours is the annual light resource utilization hours; generally, the change in the wind and light resource hours in a region is not significant. If there is a trend of getting stronger or weaker in climate prediction, the resource hours can be adjusted by multiplying each item in the sequence by a coefficient.

[0022] Furthermore, the new energy consumption space is the maximum power value of new energy that the power grid can consume at a certain moment. By comparing the new energy output with the power grid consumption space hour by hour through the time series production simulation method, when the new energy output is less than the power grid consumption space, the new energy is fully consumed; when the new energy output is greater than the power grid consumption space, the new energy electricity consumed is equal to the integral of the new energy consumption space size over time, and the calculation period is divided into N time periods with a duration of T for optimization one by one. The expression of the new energy electricity consumed is:

[0023]

[0024] P ne (t) = P wind (t) + P thotov (6)

[0025]

[0026] P ne (t) represents the new energy output at time t, P wind (t) represents the wind power output at time t, P space (t) represents the consumption space at time t, μ t is the calculation factor.

[0027] Furthermore, the new energy consumption space is formed by adding the in-network electricity consumption and the electricity transmitted out, and subtracting the minimum output of conventional energy. Considering the safety constraints, the starting capacity of conventional power sources should be able to ensure that their maximum output can meet the maximum electricity demand and the safe and stable operation of the power grid during a certain period. Combining the operating characteristics and start-stop characteristics of thermal power units, the calculation period T = 3 - 7 days, with one week being the most typical, taking T = 7, then

[0028]

[0029] P g,i,min = β i P g,i,max (10)

[0030] ∑P g,i,max >max{P con (t) + P tra (t)}(0 < t ≤ T)(11)

[0031] Among them, P con(t) represents the electricity consumption within the grid; P tra (t) represents the power transmitted out through the tie line; P res Indicates standby startup. In provinces where it is difficult to accommodate new energy, to ensure the maximum accommodation of new energy, it can take a negative value when there is sufficient standby in the regional power grid; P g,i,max Represents the maximum output of conventional units; P g,i,min Represents the minimum output of conventional units; β i Represents the minimum technical output coefficient of the i-th unit, λ i Represents the startup status of the i-th unit (i.e., startup is 1, shutdown is 0).

[0032] Furthermore, the section constraint conditions described in step S2 are:

[0033] P l (t) ≤ P l，max (t) (12)

[0034]

[0035] In the formula: P l (t), P l，max Respectively represent the actual transmission power and the maximum transmission power of the section; Represents the lower limit of the maximum transmission power of the transmission section (i.e., how much it can be reduced under the most unfavorable operating conditions), Represents the upper limit of the maximum transmission power of the transmission section (i.e., how much it can be increased under the most favorable operating conditions)

[0036] Furthermore, the peak shaving constraint conditions described in step S3 are:

[0037] ∑P wind (t) + ∑P thotov (t) + ∑P m (t) + ∑P h (t) = P(t) (14)

[0038] ∑P wind (t) represents the total wind power output, ∑P thotov (t) represents the total photovoltaic power output, ∑P m (t) represents the total thermal power output, ∑P h (t) represents the total hydropower output, P(t) represents the total power generation output.

[0039] After the grid peak shaving capacity increases, the new energy accommodation capacity will also increase. In grid planning and production operation, it is often necessary to determine the quantitative relationship between the newly increased peak shaving capacity and the newly accommodated new energy electricity or the new energy utilization rate. In the method described in the present invention, the calculation can be conveniently completed. After the grid peak shaving capacity increases by ΔP, the new grid accommodation space is:

[0040] p′ space p(t) = p space (t) + Δp (15)

[0041] Then substitute P′ space (t) into Formulas 1 and 5, and the latest new - energy consumption electricity quantity and utilization rate can be obtained. Through multiple calculations of different peak - shaving capabilities, the coupling relationship between the peak - shaving ability and the utilization rate can be calculated, and then the marginal additional new - energy electricity quantity brought by the increase in the peak - shaving ability can be determined, which helps to formulate a reasonable peak - shaving ability improvement plan.

[0042] Furthermore, the start - up capacity of the conventional units in step S2 directly determines whether the power supply capacity is sufficient when the new energy generates little electricity and the size of the new - energy consumption space when the new energy generates a large amount of electricity. And the constraints of the conventional units mainly consider the regulation capabilities and ramp - up / down constraints of thermal power units and hydropower units. Their constraint conditions:

[0043]

[0044] In the formula, respectively represent the minimum technical output and the maximum technical output of the thermal power unit, respectively represent the minimum technical output and the maximum technical output of the hydropower unit, S f represents whether the thermal power unit has undergone flexibility transformation. "1" means it has been transformed, and "0" means it has not been transformed; β represents the output reduction ratio that can be further reduced due to the flexibility transformation; represents the down - ramp rate and the up - ramp rate of the thermal power unit; represents the down - ramp rate and the up - ramp rate of the thermal power unit. To improve the operation speed, the conventional unit model is optimized, and all thermal power units are aggregated into one aggregated unit; the output change of the aggregated unit is:

[0045]

[0046] In the formula is the output of the aggregated unit, ρ i represents the start - up state of the i - th unit, where 1 means it is started up and 0 means it is shut down; after sorting the units in ascending order of capacity, a thermal - power - unit start - up matrix is formed: [1, 1, 1, 1, 0, 0, 0]. If the start - up state of the i - th unit is 1, then the start - up states of the previous i - 1 units are all 1; an output matrix of the aggregated thermal - power unit is formed in ascending order of the capacity and deep - peak - shaving ability of each unit; in the subsequent optimization algorithm, the regulation range of the aggregated unit is directly read from the output matrix:

[0047]

[0048] Aggregate the hydropower units into one hydropower unit. The adjustment range is:

[0049]

[0050] In the formula represents the output of the runoff hydropower unit, represents the minimum and maximum technical outputs of the hydropower unit with reservoir capacity, represents the output of the aggregated hydropower unit;

[0051] Considering that the measurement of new energy consumption capacity is mainly used for power grid planning and design and the annual power balance plan, and the unit ramp-up capacity is mainly used for day-ahead and intra-day plan arrangements in actual production, the unit ramp-up constraint is no longer considered in the aggregated units.

[0052] Furthermore, the optimal solution of the model in step S4 is: on the basis of satisfying the constraint conditions in steps S1 - S3, the new energy utilization rate is the highest, that is, the optimization objective is:

[0053] min(η)(25).

[0054] And in summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are:

[0055] The model established by the present invention is a mixed integer linear programming model. The constraint and optimization programs are written through GAMS software, and the CPLE solver is used for solving. Considering that the electricity load and wind-solar resources have obvious quarterly and monthly characteristics, the installed capacity in the same period of the previous year is used as the benchmark value for monthly optimization. Compared with the traditional time series simulation method that optimizes each unit one by one in the weekly optimization model, the calculation amount is much smaller.

[0056] In power grid planning, it is often necessary to conduct sensitivity analysis on various conditions. In the model of the present invention, the normalized time series can be matched with the measured boundary conditions to form the latest time series, and the boundary change adjustment or sensitivity analysis can be carried out by adjusting the change of the coefficient. Different scales, layouts and time sequences of newly added new energy installations will all affect the final new energy electricity quantity and utilization rate. Through adjustment, the above-mentioned boundary changes can be adjusted. Through multiple adjustments, the coupling relationship between the power grid installed capacity, layout and time sequence and the final new energy electricity quantity and utilization rate can be formed.

[0057] It can be seen that based on the time-series characteristics of new energy output, the present invention designs a time-series simulation method based on scenario reproduction. Starting from the overall situation, it avoids complex modeling and variable relevant factors, conducts power and electricity balance analysis from the overall perspective, realizes the calculation of indicators such as new energy power generation, power generation proportion, and utilization rate. Compared with the traditional time-series method, on the basis of considering both accuracy and rapidity, it determines the three coupling relationships of installation time series, peak shaving capacity, and new energy utilization rate, and proposes a simple and effective method for determining reasonable utilization rate. Description of the Drawings

[0058] Figure 1 Flow chart for calculating new energy consumption capacity.

[0059] Figure 2 Schematic diagram of new energy output and power curtailment on a typical day.

[0060] Figure 3 Schematic diagram of suppressing wind power volatility with multi-year average value.

[0061] Figure 4 Monthly theoretical hours of wind power from 2018 to 2020.

[0062] Figure 5 Monthly theoretical hours of photovoltaic power from 2018 to 2020.

[0063] Figure 6 Distribution map of new energy output and electricity load in a typical year. Detailed Implementation Modes

[0064] In order to make the purpose, technical solutions and advantages of the invention clearer, the following further details the invention in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention and are not used to limit the invention.

[0065] A method for measuring new energy consumption capacity based on new energy output characteristics calculates the new energy consumption capacity that can best represent new energy through the calculation of new energy utilization rate under different boundary conditions, and then determines the theoretical power and actual power of new energy based on the new energy output within a certain specified time period, combined with the size of the grid load and power transmission, and the regulation ability of conventional power sources, and further determines the utilization rate of new energy.

[0066] 1.1 Overall Idea

[0067] The first step: Based on historical data analysis of new energy output characteristics, determine the normalized 8760-point sequence of new energy output, and then form the theoretical output sequence of new energy according to the installed capacity;

[0068] The second step: Form the new energy consumption space through section constraints, peak shaving constraints, conventional unit output constraints, and load characteristics;

[0069] Step 3: Based on the new energy theoretical output sequence and combined with the new energy accommodation space, establish a model for calculating new energy accommodation.

[0070] Step 4: Find the optimal solution of the model by adjusting various parameters including the startup conditions of conventional units. Calculate the final new energy electricity quantity and utilization rate by adjusting the scale, layout, and time sequence of newly added new energy installations. Through multiple adjustments, form the coupling relationship between the grid installation scale, layout, and time sequence and the final new energy electricity quantity and utilization rate.

[0071] The new energy accommodation capacity of the grid is related to both the load characteristics and regulation ability of the grid itself, and the installation situation and output characteristics of new energy. Calculating the new energy utilization rate under different boundary conditions can best characterize the new energy accommodation capacity. The essence of utilization rate calculation is the power and electricity balance problem, that is, by combining the new energy output, the grid load, the size of power transmission, and the regulation ability of conventional power sources within a specified time period, determine the theoretical electricity quantity and actual electricity quantity of new energy, and then determine the utilization rate of new energy. That is, the utilization rate of new energy is:

[0072]

[0073] In the formula, Q fact represents the actual power generation of new energy restricted by the grid accommodation capacity; Q theo represents the theoretical power generation of new energy, that is, the electricity quantity that new energy can generate when the grid accommodation capacity is large enough [9].

[0074] Calculating Q fact and Q theo is the key to measuring the utilization rate. To accurately calculate Q fact , it is necessary to determine the new energy accommodation capacity of the grid. When the grid accommodation capacity is large enough, Q fact = Q theo , that is, as much new energy as there is can be accommodated. When new energy power generation exceeds the grid accommodation capacity, the excess electricity can only be "abandoned". Define the new energy accommodation space as the maximum power value of new energy that the grid can accommodate at a certain moment. The time-sequential production simulation method compares the new energy output and the grid accommodation space for each time period. When the new energy output is less than the grid accommodation space, new energy is fully accommodated; when the new energy output is greater than the grid accommodation space, the accommodated new energy electricity quantity is equal to the integral of the new energy accommodation space size over time. If the calculation period is divided into N time periods with a duration of T for optimization one by one, the calculation method is shown in Equation 2-4.

[0075]

[0076] P ne (t) = P wind (t) + Pthotov (t) (4)

[0077]

[0078] Wherein, P ne (t) represents the new energy output at time t, P wind (t) represents the wind power output at time t, P space (t) represents the accommodation space at time t, μ t is a calculation factor.

[0079] 2.2 Calculation of accommodation space

[0080] The new energy accommodation space is formed by adding the in-network electricity consumption and the power transmitted out through tie lines to form an electricity consumption space, and subtracting the minimum output of conventional energy. Considering safety constraints, the starting capacity of conventional power sources should be able to ensure that their maximum output can meet the maximum electricity demand and the safe and stable operation of the power grid during a certain period. Combining the operating characteristics and start-stop characteristics of thermal power units, the calculation period T = 3 - 7 days, with one week being the most typical, that is, T = 7 is taken.

[0081]

[0082] P g,i,min = β i P g,i,max (7)

[0083] ∑P g,i,max >max{P con (t) + P tra (t)} (0 < t ≤ T) (8)

[0084] Wherein P con (t) represents the in-network electricity consumption; P tra (t) represents the power transmitted out through tie lines; P res represents the standby start-up. In provinces where new energy accommodation is difficult, in order to ensure the maximum accommodation of new energy, it can take a negative value when the regional power grid standby is sufficient; P g,i,max represents the maximum output of conventional units; P g,i,min represents the minimum output of conventional units; β i represents the minimum technical output coefficient of the i-th unit, λ i represents the start-up state of the i-th unit (i.e., 1 for start-up and 0 for shutdown). Figure 1 For a typical day with high wind power generation, when new energy generation is high, the output of thermal power is maintained at the minimum technical output, and the part of the new energy output that exceeds the accommodation space forms curtailment.

[0085] 2 Calculation of new energy accommodation capacity

[0086] The factors involved in the consumption capacity of new energy are numerous, and some factors are difficult to analyze through modeling. For example, when a certain province has difficulties in consuming new energy, it can transmit it to adjacent provinces for consumption. However, it is also possible that the adjacent provinces are unable to receive it at this time. To conduct a detailed analysis, it is necessary to simulate each adjacent province hour by hour, which is equivalent to analyzing the new energy consumption capacity of each adjacent province one by one, and the computational workload will increase exponentially. Another example is that after the improvement of fan manufacturing technology, newly put into production fans can generate electricity at lower wind speeds (for example, the cut-in wind speed of previous fans was generally 5 m / s, and the cut-in wind speed of the latest fans can be reduced to 2.5 m / s)

[18] . The windward area of the blades is larger, and these factors will lead to an increase in the theoretical power of new energy

[19] . However, it is also very difficult and complex to model and quantitatively analyze this impact on a network-wide scale.

[0087] Analyzing the above various factors by traditional time-series production simulation methods is difficult and involves a large amount of computation. Through research, it can be found that, first, although some factors (such as new fans) are not easy to count, their influence effects are stable and there is no randomness. Second, some factors (such as cross-provincial support capabilities) have strong randomness with respect to time, but as the time scale is extended, the total amounts on a monthly and annual basis are basically constant. Third, the action results of all factors will be reflected in the final output of new energy.

[0088] Based on the above considerations, starting from the overall situation and the impact results, we adopt the method of scenario reproduction, with the year as the calculation period. The historical data of the previous year is formed into a normalized time series to reflect the probability distribution law of all factors. Subsequently, combined with the maintenance plan and the installation time sequence, the normalized sequence is restored to the new energy output sequence. The reason for using the previous year is that the earlier years have a lower degree of completeness in reflecting the latest factors, and the multi-year average data cannot reflect the volatility of wind and light. As Figure 1 can be seen, the peak-valley difference of the multi-year average data will be significantly lower than that of any single year.

[20] . On the premise of ensuring a certain ratio of each value within the sequence, the multi-year average utilization hours on a monthly and annual basis are reflected by overall proportional amplification or reduction, as follows.

[0089] 2.1 New energy output characteristics

[0090] Analyzing the new energy output of a certain northern province for five consecutive years, it is found that as the time scale is extended, the volatility of new energy is decreasing. Although the new energy fluctuates violently and without rules within a day and within several days, when the time scale is extended to a month or a year, as Figure 2 shown, it will be found that some months (such as April and May) have the largest wind power output in each year, while some other months (such as January and December) always have the smallest wind power output. As shown in Table 1, the difference between the total theoretical hours of each year and the multi-year average value is within 5%.

[0091] Table 1 Monthly theoretical hours of wind power from 2018 to 2020

[0092]

[0093]

[0094] The seasonal characteristics of photovoltaic power and the stability of the annual theoretical hours are more obvious than those of wind power. As Figure 3 shown, the output of photovoltaic power is low from November to February of the following year, and there is a significant increase from March to April. As shown in Table 1, the difference between the annual theoretical hours and the multi-year average value is basically within 3%.

[0095] Table 2 Monthly theoretical hours of photovoltaic power from 2018 to 2020

[0096]

[0097]

[0098] The overall slight upward trend in theoretical hours is due to the progress of wind turbine manufacturing technology, which enables newly commissioned wind turbines to capture lower wind speeds for power generation. The model based on the present invention well embodies these factors in the calculation process, solving the deficiencies of the traditional time series production simulation method.

[0099] Considering the above factors, the present invention adopts the normalized time series of historical new energy theoretical power and combines the changes in installed capacity to form the new energy theoretical output of the measurement period by matching hour by hour. Define the matrix P NE as the hourly output sequence of new energy, as the hourly resource normalization sequence of wind and light, S wind 、S thotov as the hourly installed capacity sequences of wind and light. The above proofs are all in the structure of 1×8760. Taking P NE as an example, it is constituted as shown in Equation 9:

[0100] P NE =[P NE,1 ,P NE,2 ,P NE,3, ......,P NE,8760 T (9)

[0101] The operation relationships between the matrices are as follows:

[0102]

[0103] In the formula, define the matrix P NE as the hourly output sequence of new energy, as the hourly resource normalization sequence of wind and light, S​wind , S thotov is the hourly installed capacity sequence of wind and light, and the matrices are all in the structure of 1×8760. S wind,j represents the wind power installed capacity corresponding to the j-th hour; S thotov,j represents the photovoltaic installed capacity corresponding to the j-th hour; respectively represent the output and installed capacity of wind power in the j-th hour of the previous year; respectively represent the output and installed capacity of photovoltaic power in the j-th hour of the previous year.

[0104] The sum of each item in the theoretical output sequence of the above new energy exactly corresponds to the annual resource hours, that is, is the annual wind resource hours, is the annual light resource utilization hours. Generally, the change of wind and light resource hours in a region is not large. If there is a trend of getting stronger or weaker in climate prediction, the resource hours can be adjusted by multiplying the coefficients of each item in the sequence.

[0105] 2.2 Section Constraint

[0106] The transmission section between the new energy power plant and the main grid is restricted by the transmission capacity, which is expressed in the traditional time series simulation method as:

[0107] P l (t) ≤ P l，max (13)

[0108] In the formula: P l (t), P l，max respectively represent the actual transmission power and the maximum transmission power of the section. Such section constraint conditions seem simple, but they are not reasonable because the quota of the section is also a variable and changes at any time with different operating states. That is, the constraint condition is:

[0109] P l (t) ≤ P l，max (t) (14)

[0110]

[0111] In addition to the change range of the maximum transmission quota being determinable, the specific change law is restricted by the change of the operating state and is difficult to model and describe. In the time series production simulation method of the present invention, the change of the section quota is based on the distribution sequence of the previous year, and at the same time, combined with the quota increase caused by grid transformation and the coupling relationship between the section quota and the grid operating state, the latest section quota sequence is formed, thus simulating the real operating situation to the greatest extent.

[0112] 2.3 Peak Regulation Constraint

[0113] In a power grid with a high proportion of new energy, on the one hand, conventional power sources need to meet the load changes, and on the other hand, they need to balance the fluctuations in the output of new energy, and need to meet the constraint conditions:

[0114] ∑P wind (t)+∑P thotov (t)+∑P m (t)+∑P h (t)=P(t) (16)

[0115] After the power grid peaking capacity increases, the new energy consumption capacity will also increase. In power grid planning and production operation, it is often necessary to determine the quantitative relationship between the newly added peaking capacity and the newly added new energy electricity consumption or the new energy utilization rate. In the method described in the present invention, the calculation can be conveniently completed. After the power grid peaking capacity increases by ΔP, the new power grid consumption space is:

[0116] p′ space (t)=p space (t)+Δp (17)

[0117] Then substitute P′ space (t) into Formulas 1 and 6 to obtain the latest new energy consumption electricity and utilization rate. Through multiple measurements of different peaking capacities, the coupling relationship between the peaking capacity and the utilization rate can be measured, and then the marginal additional new energy electricity generated by the increase in the peaking capacity can be determined, which helps to formulate a reasonable peaking capacity improvement plan.

[0118] 2.4 Conventional unit output constraint

[0119] The starting capacity of conventional units directly determines whether the power supply capacity is sufficient when new energy generates little power and the size of the new energy consumption space when new energy generates a large amount of power. The constraints on conventional units mainly consider the regulation capabilities and ramp constraints of thermal power units and hydropower units:

[0120]

[0121] In the formula, respectively represent the minimum technical output and the maximum technical output of thermal power units, respectively represent the minimum technical output and the maximum technical output of hydropower units, S f represents whether the thermal power unit has undergone flexibility transformation, "1" indicates that it has been transformed, and "0" indicates that it has not been transformed. β represents the output reduction ratio that can be further reduced by the flexibility transformation. represents the down-ramp rate and the up-ramp rate of the thermal power unit. represents the down-ramp rate and the up-ramp rate of the thermal power unit.

[0122] The model established by the present invention is based on reasonable adjustment of the time series of the previous year. To improve the operation speed, we optimize the conventional unit model, aggregate all thermal power units into one aggregated unit, and the output change of the aggregated unit is as follows:

[0123]

[0124] In the formula is the output of the aggregated unit, and ρ i represents the starting state of the i-th unit, where 1 means starting and 0 means stopping. After sorting the units in ascending order of capacity, a thermal power unit starting matrix is formed: [1, 1, 1, 1, 0, 0, 0]. If the starting state of the i-th unit is 1, then the starting states of the previous i - 1 units are all 1. An output matrix of the aggregated thermal power unit is formed in ascending order of the capacity and deep peak shaving capacity of each unit. In the subsequent optimization algorithm, the adjustment range of the aggregated unit is directly read from the output matrix:

[0125]

[0126] Similarly, the hydroelectric units are also aggregated into one hydroelectric unit, and the adjustment range is shown in formula (26).

[0127]

[0128] In the formula represents the output of the runoff hydroelectric unit, represents the minimum technical output and the maximum technical output of the hydroelectric unit with reservoir capacity, represents the output of the aggregated hydroelectric unit.

[0129] Considering that the measurement of new energy consumption capacity is mainly used for power grid planning and design and annual power and electricity balance plan, and the unit ramp-up capacity is mainly used for day-ahead and intra-day plan arrangements in actual production, the unit ramp-up constraint is no longer considered in the aggregated unit.

[0130] Through the above processing, the amount of calculation is reduced, the operation efficiency is improved, and conditions are created for flexibly carrying out various sensitivity analyses.

[0131] 2.5 Optimization Objectives and Methods

[0132] The overall optimization objective is to maximize the new energy utilization rate on the basis of meeting the above constraints, that is, the optimization objective is:

[0133] min(η)(27)

[0134] The model established in the present invention is a mixed-integer linear programming model. The constraints and optimization programs are written through GAMS software, and the CPLE solver is used for solving. Considering that the electricity load and wind-solar resources have obvious quarterly and monthly characteristics, the installed capacity in the same period of the previous year is used as the benchmark value for monthly optimization. Compared with the traditional time-series simulation method that optimizes each unit one by one in the weekly optimization model, the computational amount is much smaller.

[0135] In power grid planning, it is often necessary to conduct sensitivity analysis on various conditions. In the model of the present invention, the normalized time series can be matched with the measured boundary conditions to form the latest time series, and the boundary changes can be adjusted or sensitivity analysis can be carried out by changing the adjustment coefficient. Different scales, layouts and time sequences of newly added new energy installations will all affect the final new energy electricity quantity and utilization rate. Through adjustment, the above-mentioned boundary changes can be adjusted. Through multiple adjustments, the coupling relationship between the installed capacity scale, layout and time sequence of the power grid and the final new energy electricity quantity and utilization rate can be formed.

[0136] 3 Case Study

[0137] 3.1 Case Introduction

[0138] Taking the actual power grid of a major new energy province as an example, the time-series production simulation method based on scenario reproduction is used to calculate the new energy consumption situation in 2021. The calculation uses the actual data in 2020 to form the normalized sequences of new energy resources, electricity load, the start-up time sequence of conventional units, and the section limit sequence. Table 3 lists the calculation boundary conditions. Since the number of items in each time series is too large (8760 items), they are not specifically listed.

[0139] Table 3 Boundary Conditions (Unit: 100 million kWh, 10,000 kW)

[0140] Table 3 boundary conditions (Unit: 108kWh, 104kW)

[0141]

[0142] 3.2 Analysis of Case Results

[0143] At the beginning of the year, it was predicted that the cumulative power generation in 2021 would be 42.7 billion kWh, and the actual cumulative power generation was finally 43.6 billion kWh, with a prediction error of 2.06%; the predicted annual utilization rate was 95.11%, and the actual utilization rate was 96.83%, with a deviation of 1.72 percentage points. Since the installed capacity was concentrated in the fourth quarter of the year, its impact was mainly in the second year. Calculating the data in 2022, the theoretical new energy electricity quantity was 55.8 billion, the new energy power generation was 50.6 billion, and the utilization rate was only 90.72%.

[0144] Table 4 Calculation Results (Unit: 100 million kWh, 10,000 kW)

[0145]

[0146]

[0147] According to the existing load, power transmission, and the minimum output level of conventional power sources, the annual consumption space is 73.5 billion, which is greater than the theoretical power of new energy of 57.9 billion. Analyzing the consumption space minus the available power of new energy for 8,760 hours, the consumption space is greater than the available output of new energy for 6,331 hours (accounting for 72%), as Figure 5 shown. By counting the section-limited and peak-regulation-limited power for the whole year on an hourly basis, the contradiction of abandoned power is most prominent during the lunchtime period (11:00 - 17:00), and the abandoned power accounts for 69.94% of the whole year. Especially at 15:00, it is the most serious, with the abandoned power exceeding 520,000 kW (corresponding to 0.125 billion kWh for the whole day), and the abandoned power accounts for 12% of the whole year.

[0148] Analyzing the consumption space minus the available power of new energy month by month, the space is relatively small in October and February. Analyzing the consumption space minus the available power of new energy hour by hour, the consumption space is negative from 11 to 15 o'clock.

[0149] Analyzing the increase in peak-regulation capacity from 500,000 to 2.5 million, after increasing to 1.5 million peak-regulation capacity, the utilization rate can be close to 95%, but the marginal increased power generation brought by the unit peak-regulation capacity is decreasing, as shown in Table 5.

[0150] Table 5. Sensitivity Analysis Table of Peak-Regulation Capacity

[0151]

[0152] The increase in other peak-regulation capacities can only be achieved through energy storage. However, energy storage is restricted by both power and capacity. Therefore, the analysis of energy storage configuration is carried out, and the results are shown in Table 6.

[0153] Table 6. Sensitivity Analysis Table of Increasing Energy Storage

[0154]

[0155]

[0156] Taking 2-hour energy storage as an example, to achieve a utilization rate of 95%, 6 million energy storage needs to be configured, but at this time, the utilization efficiency of energy storage is very low. The utilization efficiency is relatively high when configuring 1 million - 2 million energy storage. Therefore, it is recommended to configure 1.5 million * 2-hour energy storage, and take 92.5% as the reasonable utilization rate.

[0157] Taking 3-hour energy storage as an example, to achieve a utilization rate of 95%, 4 million energy storage needs to be configured. However, at this time, the utilization efficiency of the energy storage is very low. The utilization efficiency is relatively high when 0.5 million - 1.5 million energy storage is configured. Therefore, it is recommended to configure 1 million * 3-hour energy storage and take 92.5% as the reasonable utilization rate.

[0158] The above are the preferred embodiments of the invention, which are not intended to limit the invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the invention shall be included within the protection scope of the invention.

Claims

1. A method for calculating the new energy consumption capacity based on the output characteristics of new energy, characterized in that: Calculating the utilization rate of new energy under different boundary conditions can best characterize the absorption capacity of new energy. Then, based on the output of new energy within a specified time period, combined with the grid load, the magnitude of power transmission, and the regulation capacity of conventional power sources, the theoretical and actual electricity quantities of new energy are determined, and further the utilization rate of new energy is determined; It includes the following steps: S1: Analyze the output characteristics of new energy based on historical data, determine the normalized 8760-point sequence of new energy output, and then form the theoretical output sequence of new energy according to the installed capacity; S2: Form the new energy absorption space through section constraints, peak shaving constraints, output constraints of conventional units, and load characteristics; S3: According to the theoretical output sequence of new energy obtained in step S1, combined with the new energy absorption space obtained in step S2, establish a model for measuring new energy absorption; S4: Find the optimal solution of the model by adjusting various parameters including the startup conditions of conventional units, then adjust the scale, layout, and time sequence of newly added new energy installations to measure the final new energy electricity quantity and utilization rate. Finally, through multiple adjustments, establish the coupling relationship between the grid installed capacity, layout, and time sequence and the final new energy electricity quantity and utilization rate; The theoretical output sequence of new energy described in step S1 adopts the normalized time sequence of historical new energy theoretical power, and is matched hour by hour in combination with the change of installed capacity to form the theoretical output of new energy in the measurement period; P NE = [P NE,1 , P NE,2 , P NE,3, ......, P NE,8760 T (1)​ Wherein, matrix P is defined NE is the hourly output sequence of new energy, is the hourly resource normalization sequence of wind and light, S wind , S thotov is the hourly installed capacity sequence of wind and light, and the matrices are all in the structure of 1×8760. S wind,j represents the installed wind power capacity corresponding to the j-th hour; S thotov,j represents the installed photovoltaic capacity corresponding to the j-th hour; respectively represent the output and installed capacity of wind power at the j-th hour of the previous year; respectively represent the output and installed capacity of photovoltaic power at the j-th hour of the previous year.

2. A method for measuring the new energy consumption capacity based on the output characteristics of new energy according to claim 1, characterized in that: The utilization rate of the new energy is: Where Q fact represents the actual power generation of new energy after being restricted by the grid's accommodation capacity; Q theo represents the theoretical power generation of new energy, that is, the power that new energy can generate when the grid's accommodation capacity is large enough.

3. A method for calculating the new energy consumption capacity based on the output characteristics of new energy according to claim 1, characterized in that: The sum of each item in the theoretical output sequence of the new energy is exactly equal to the annual resource hours, that is is the annual wind resource hours, is the annual light resource utilization hours; generally, the change of the wind and light resource hours in a region is not significant. If the climate prediction shows a trend of getting stronger or weaker, the resource hours can be adjusted by multiplying each item in the sequence by a coefficient.

4. A method for measuring the new energy consumption capacity based on the output characteristics of new energy according to claim 1, characterized in that: The new energy absorption space is the maximum power value of new energy that the grid can absorb at a certain moment. By comparing the new energy output hour by hour with the grid absorption space through the time series production simulation method, when the new energy output is less than the grid absorption space, the new energy is fully absorbed; when the new energy output is greater than the grid absorption space, the absorbed new energy electricity quantity is equal to the integral of the new energy absorption space size over time, and the calculation period is divided into N time periods with a duration of T for optimization one by one. The expression of the absorbed new energy electricity quantity is: P ne (t) = P wind (t) + P thotov (7) P ne (t) represents the new energy output at time t, P wind (t) represents the wind power output at time t, P space (t) represents the consumption space at time t, μ t is a calculation factor.

5. A method for measuring the new energy consumption capacity based on the output characteristics of new energy according to claim 4, characterized in that: The new energy absorption space is formed by adding the in-network electricity consumption and the power transmission electricity to form the electricity consumption space, and subtracting the minimum output of conventional energy. Considering safety constraints, the startup capacity of conventional power sources should be able to ensure that their maximum output can meet the maximum electricity demand and the safe and stable operation of the grid within a certain period. Combining the operating characteristics and startup and shutdown characteristics of thermal power units, take the calculation period T = 3 - 7 days, with one week being the most typical, take T = 7, then P g,i,min = β i P g,i,max (11) ∑P g,i,max > max{P con (t) + P tra (t)} (0 < t ≤ T) (12) Among them, P con (t) represents the electricity consumption within the network; P tra (t) represents the electricity transmitted through the tie line; P res represents the reserve start-up. In provinces with difficulties in absorbing new energy, in order to ensure the maximum absorption of new energy, it can take a negative value when the regional power grid has sufficient reserve; P g,i,max represents the maximum output of the conventional unit; P g,i,min represents the minimum output of the conventional unit; β i represents the minimum technical output coefficient of the i-th unit, λ i represents the start-up status of the i-th unit.

6. The method for measuring the new energy consumption capacity based on the output characteristics of new energy according to claim 5, wherein: The section constraint conditions described in step S2 are: P l ψ(t) ≤ P l，max ψ(t) (13) Where: P l (t) and P l，max respectively represent the actual transmission power and the maximum transmission power of the section; represents the lower limit of the maximum transmission power of the transmission section, represents the upper limit of the maximum transmission power of the transmission section; The peak shaving constraint conditions described in step S2 are: ∑P wind (t) + ∑P thotov (t) + ∑P m (t) + ∑P h (t) = P(t) (15) Where, ∑P wind (t) represents the total wind power output, ∑P thotov (t) represents the total photovoltaic power output, ∑P m (t) represents the total thermal power output, ∑P h (t) represents the total hydropower output, and P(t) represents the total power generation output; After the grid peak shaving capacity increases, the absorption capacity of new energy will also increase. In grid planning and production operation, it is often necessary to determine the quantitative relationship between the newly added peak shaving capacity and the newly added absorbed new energy electricity quantity or the utilization rate of new energy; after the grid peak shaving capacity increases by ΔP, the new grid absorption space is: P′ space (t) = P space (t) + ΔP (16) Then substitute P′ space (t) into Formulas 5 and 6 to obtain the latest new energy consumption electricity and utilization rate. Through multiple calculations of different peak shaving capabilities, the coupling relationship between the peak shaving ability and the utilization rate can be measured, and then the marginal additional new energy electricity brought by the increase in the peak shaving ability can be determined, which helps to formulate a reasonable peak shaving ability improvement plan; The startup capacity of conventional units in step S2 directly determines whether the power supply capacity is sufficient when new energy has low output and the size of the new energy absorption space when new energy has high output. And the constraints of conventional units mainly consider the regulation capacity and ramp constraints of thermal power units and hydropower units. The constraint conditions are: In the formula, respectively represent the minimum technical output and the maximum technical output of the thermal power unit, respectively represent the minimum technical output and the maximum technical output of the hydropower unit, S f indicates whether the thermal power unit has undergone flexibility transformation. "1" indicates that it has been transformed, and "0" indicates that it has not been transformed; β represents the proportion of output that can continue to be reduced due to flexibility retrofit; represent the down-ramp rate and up-ramp rate of thermal power units; represent the down-ramp rate and up-ramp rate of thermal power units. To improve the calculation speed, the conventional unit model is optimized, and all thermal power units are aggregated into one aggregated unit; the output change of the aggregated unit is: where is the output of the aggregation unit, and ρ i represents the starting state of the i-th unit, where 1 means starting and 0 means stopping; after sorting the units in ascending order of capacity, a thermal power unit starting matrix is formed: [1, 1, 1, 1, 0, 0, 0]. If the starting state of the i-th unit is 1, then the starting states of the previous i - 1 units are all 1; an output matrix of the aggregated thermal power units is formed in ascending order of the capacity and deep peak shaving capacity of each unit; in the subsequent optimization algorithm, the adjustment range of the aggregation unit is directly read from the output matrix: Aggregate the hydropower units into one hydropower unit, adjustment range: In the formula represents the output of a radial flow hydroelectric generating unit represents the minimum and maximum technical outputs of a hydroelectric generating unit with reservoir capacity represents the output of the aggregated hydroelectric generating units Considering that the measurement of new energy consumption capacity is mainly used for power grid planning and design and the annual power and energy balance plan, and the unit ramp-up capacity is mainly used for day-ahead and intra-day planning in actual production, the unit ramp-up constraint is no longer considered in the aggregated units.

7. A method for measuring the new energy consumption capacity based on the output characteristics of new energy according to claim 1, characterized in that: The optimal solution of the model in step S4 is: on the basis of satisfying the constraint conditions in steps S1 - S3, the new energy utilization rate is the highest, that is, the optimization objective is: min(η)(26).

Images