New energy carrying capacity evaluation method based on two-stage scenario reduction and multi-time scale time series simulation

Abstract

The application provides a new energy carrying capacity evaluation method based on a two-stage scene reduction multi-time scale timing simulation, and belongs to the technical field of new energy carrying capacity evaluation; solves the problem of calculation complexity caused by multi-time scale coupling of new energy carrying capacity evaluation; the method comprises the following steps: initialization; taking the minimum unit operation cost as an objective function, and punishing light abandonment, wind abandonment, water abandonment and load shedding; considering the operation constraints of aggregated thermal power units, the operation constraints of pumped storage units, the positive and negative spinning reserve constraints, the flexibility resource constraints, the load constraints and the power balance constraints; constructing a two-stage scene reduction framework, combining absolute probability distance, relative probability distance and marginal relative probability distance indexes to determine the number and specific scenes of the reduced scenes; establishing a comprehensive new energy carrying capacity evaluation index system to evaluate the power grid; and the application is applied to a power system containing new energy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy carrying capacity evaluation, in particular to a new energy carrying capacity evaluation method based on two-stage scenario reduction and multi-time scale time series simulation. BACKGROUND

[0002] Under the trend of global energy transformation, new energy, with its clean and sustainable characteristics, continues to rise in the proportion of power systems. According to the data of the International Energy Agency (IEA), the installed capacity of new energy power generation has achieved a leap in the past decade, and this trend is particularly pronounced in China. However, the large-scale access of new energy to the power system has brought many challenges.

[0003] The large-scale access of new energy to the power system has brought many challenges to power planning and operation. The randomness, volatility and uncertainty of new energy output have seriously disturbed the power balance. Traditional new energy carrying capacity evaluation methods cannot meet the development needs of new energy characteristics and power systems. The evaluation index system of traditional new energy carrying capacity evaluation method is single, and focuses on new energy consumption capacity, ignoring key factors such as grid operation flexibility, source-load interaction characteristics and power supply adequacy. Some methods only measure the consumption capacity by the proportion of new energy generation, without considering the impact of new energy output fluctuation on grid stability, resulting in a gap between the evaluation results and the actual operation.

[0004] At the same time, the operation characteristics of the power system differ significantly at daily, weekly, monthly and seasonal time scales. Within a day, the load demand shows obvious peak-valley characteristics, and the new energy output is closely related to the daily change of light and wind speed; on a seasonal scale, the large fluctuations of cooling load in summer and heating load in winter, as well as the differences in new energy resources in different seasons, increase the complexity of power system planning and operation. However, existing evaluation methods mostly fail to fully consider the dynamic characteristics of power systems at multiple time scales, making it difficult to meet the requirements of fine planning and scheduling of power grids. The coupling of multiple time scales further exacerbates the complexity of calculation, making it difficult for traditional evaluation methods to balance between calculation efficiency and evaluation accuracy.

[0005] Under this background, it is necessary to develop a method that can effectively respond to the characteristics of new energy output, take into account multiple time scales, and accurately and efficiently evaluate the carrying capacity of new energy, which is the key to promoting the safe, stable and efficient operation of power systems, and promoting the scientific planning and rational layout of new energy. SUMMARY

[0006] In order to solve the problem of calculation complexity caused by the coupling of multiple time scales in new energy carrying capacity evaluation, the present application proposes a new energy carrying capacity evaluation method based on two-stage scenario reduction and multi-time scale time series simulation.

[0007] The technical scheme adopted by the application is a new energy carrying capacity evaluation method based on a two-stage scene reduction multi-time scale time sequence simulation, comprising the following steps:

[0008] Step 1: initialization, input power system installed capacity, load curve, new energy output data;

[0009] Step 2: build a power system time sequence production simulation model: take the minimum unit operation cost as the objective function, and punish light, wind, water and load shedding; and consider the operation constraints of aggregated thermal power units, the operation constraints of pumped storage units, the positive and negative spinning reserve constraints, the flexibility resource constraints, the load constraints and the power balance constraints;

[0010] Step 3: build a two-stage scene reduction framework, determine the number and specific scenes of the reduced scenes by combining absolute probability distance, relative probability distance and marginal relative probability distance index;

[0011] Step 4: establish a comprehensive new energy carrying capacity evaluation index system to evaluate the influence of high proportion of new energy access on power supply capacity, reliability and operation safety of the power grid.

[0012] Further, the objective function is to minimize the unit operation cost while punishing light, wind, water and load shedding, and the expression of the objective function is as follows:

[0013] ;

[0014] In the formula, is the generation cost of the thermal power unit; is the total cost of starting and stopping the thermal power unit; is the total cost of climbing of the thermal power unit, is the pumped storage power generation cost; is the load shedding penalty, flexible resource penalty item such as wind / light / water, is the reward item about the spinning reserve coefficient ; , and are the quadratic term, linear term and constant term coefficients of the fuel cost-power function of the thermal power unit g respectively. and are the starting cost and stopping cost of the thermal power unit at each time; represents the total output of the aggregated gas turbine or coal-fired unit at the t'th period; represents the unit generation cost of the pumped storage unit, represents the total output of the pumped storage unit at the t'th period; and These represent the penalty costs for load shedding and curtailment of flexible resources, respectively. and These represent the load shedding power and the power curtailed from flexible resources at time t, respectively.

[0015] Furthermore, the operational constraints of aggregated thermal power units are established by classifying thermal power units into coal-fired units, peak-shaving gas-fired units, and combined heat and power gas-fired units, and by aggregating units of the same type. The operational constraints of aggregated thermal power units include:

[0016] Output upper and lower limits constraints for each coal-fired unit and peak-shaving gas-fired unit, output upper and lower limits constraints and heating load constraints for cogeneration gas-fired units, ramp-up constraints for polymerization units, start-up and shutdown constraints for polymerization units, start-up duration constraints for coal-fired units, and start-up and shutdown cost constraints for polymerization units.

[0017] Furthermore, the operational constraints of pumped storage units include: upper and lower limits of pumped storage unit output and reservoir capacity constraints.

[0018] Furthermore, a two-stage scene reduction framework was constructed based on the GAMS-SCENRED tool.

[0019] Furthermore, the first stage of the two-stage scene reduction framework is a scene reduction process performed on a single random variable, while the second stage is a scene reduction of the combined scene set after the first stage of scene reduction, and the combined scene reduction is evaluated using the average first-order moment difference and the average second-order moment difference indices.

[0020] Furthermore, initial scenario sets for new energy sources and load power are obtained based on historical data, and are denoted as follows: and And let the initial scene set be For scenarios that include random variables such as new energy sources and load power, a reduced scenario set is set. For an empty set, the first stage of scene reduction involves the following steps:

[0021] Select the first scene from the initial scene set. Select the scene that is most representative of the overall scene set and add it to the reduced scene set. ;

[0022] Iteratively select subsequent scenes, in each iteration from the remaining set of initial scenes. - Select one scene to add to the reduced scene set ;

[0023] Termination condition determination: After each new scene is added, check whether the termination condition is met.

[0024] Scenario probability reassignment, after determining the final reduced scenario set After that, the probability of each scenario in the reduced scenario set needs to be reassigned.

[0025] Further, the first scenario is selected according to the absolute probability distance between each scenario and all other scenarios, and the scenario that makes the absolute probability distance minimum is selected to join the reduced scenario set ;

[0026] The subsequent iteration selects a scenario according to the fact that after the scenario is added, the relative probability distance between the reduced scenario set and the original scenario set decreases most, that is, the scenario that makes the marginal relative probability distance maximum is selected;

[0027] The termination condition is that if the marginal relative probability distance is lower than a set target value and the cardinality of the reduced scenario set does not exceed the maximum limit, the next scenario is selected by iteration; if the marginal relative probability distance is lower than the target value and the cardinality of the reduced scenario set reaches the maximum limit, or the marginal relative probability distance is not lower than the target value but no scenario can be found to make the relative probability distance significantly decrease, the algorithm stops.

[0028] Further, it is characterized in that:

[0029] The absolute probability distance is calculated according to the following formula:

[0030] ;

[0031] In the formula: represents a random variable; represents the probability of the original scenario s; is the scenario distance normalized between the original scenario s and the scenario retained after scenario reduction in unit time, which is calculated by L1-norm;

[0032] The relative probability distance between the original scenario set and the reduced scenario set is the absolute probability distance divided by the probability distance when the reduced scenario set has only one scenario, and the relative probability distance is calculated according to the following formula:

[0033] ;

[0034] In the formula: is the probability distance when the reduced scenario set has only one scenario, which is calculated according to the following formula:

[0035] The marginal relative probability distance is calculated according to the following formula:

[0036] .

[0037] Further, the comprehensive new energy carrying capacity evaluation index system includes a power grid source-load characteristic index, an electricity adequacy index, a new energy consumption index and a power grid flexibility index.

[0038] The application has the following beneficial effects relative to the prior art:

[0039] In terms of evaluation accuracy: the constructed power system time series production simulation model takes the minimum unit operation cost as the target, comprehensively considers various costs such as the start-stop cost and the power generation operation cost of thermal power units, and incorporates various operation constraints of aggregated thermal power units and pumped storage units, while penalizing light and wind curtailment, which can better fit the actual power system operation; the two-stage scenario reduction framework combines various indexes such as absolute probability distance and relative probability distance, the first stage is aimed at reducing a single random variable, and the second stage reduces the combined scenario and adds the average first moment difference and second moment difference indexes, which can effectively retain the core characteristics of the original scenario and guarantee the scenario information accuracy.

[0040] In terms of calculation efficiency: the two-stage scenario reduction framework based on the GAMS-SCENRED tool can balance the scenario information accuracy while greatly reducing the number of scenarios, solve the calculation complexity problem caused by multi-time scale coupling, and has shorter calculation time compared with the method based on all time series and the method of selecting typical days, which improves the evaluation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0041] The application will be further described below in conjunction with the accompanying drawings:

[0042] Figure 1 The method flowchart provided for the embodiments of the application;

[0043] Figure 2 The load mean curve of a certain province provided for the embodiments of the application. DETAILED DESCRIPTION

[0044] As Figure 1 and 2As shown, the application provides a new energy carrying capacity evaluation method based on two-stage scenario reduction of multi-time scale timing simulation, to accurately and efficiently evaluate the new energy carrying capacity. First, a power system timing production simulation model is constructed, with the minimum unit operation cost as the objective, to punish light, wind, water and load shedding, considering the operation constraints of aggregated thermal power units and pumped storage units. Then, a two-stage scenario reduction framework based on GAMS-SCENRED tool is proposed, combined with absolute probability distance, relative probability distance, marginal relative probability distance and other indicators to determine the number and specific scenarios of reduced scenarios, balancing the accuracy and computational efficiency of scenario information. Finally, a comprehensive new energy carrying capacity evaluation index system is established to evaluate the impact of high proportion of new energy access on power supply capacity, reliability and operation safety of the power grid.

[0045] Specifically includes the following steps:

[0046] Step 1: initialization, input power system installed capacity, load curve, new energy output data.

[0047] Step 2: build a power system timing production simulation model: take the minimum unit operation cost as the objective function, punish light, wind, water and load shedding. And consider the operation constraints of aggregated thermal power units and pumped storage units.

[0048] The application considers the economic efficiency of thermal power units and pumped storage units and the orientation of preferential clean energy consumption, sets the objective function as the minimum unit operation cost while punishing light, wind, water and load shedding, i.e. introduces four penalty factors for abandoned electricity, the unit operation cost includes thermal power unit start-stop cost, power generation operation cost, pumped storage unit start-up cost and abandoned electricity penalty cost, the expression of the objective function is as follows:

[0049] (1);

[0050] In the formula, is the power generation cost of thermal power units; is the total cost of thermal power unit start-stop; is the total cost of thermal power unit climbing, is the pumped storage power generation cost; is the load shedding penalty, flexible resource penalty item such as abandoned wind / light / water, is the reward item about rotating reserve coefficient ; , and are the quadratic term, linear term and constant term coefficients of the fuel cost-power function of thermal power units respectively. and respectively represent the start-up cost and the shut-down cost of the thermal power unit at each time point; represents the total output of the aggregated thermal power unit at the t'th time period; represents the unit generation cost of the pumped storage unit, represents the total output of the pumped storage unit at the t'th time period; and respectively represent the penalty cost of load shedding and the penalty cost of electricity abandonment of the flexible resource, and respectively represent the load shedding power at the t'th time point and the electricity abandonment power of the flexible resource.

[0051] In the present application, the thermal power units are classified into coal-fired units, peak-shaving gas-fired units and heat and power gas-fired units according to the type to establish the operation condition constraints of the aggregated thermal power units.

[0052] For the coal-fired units and the peak-shaving gas-fired units, the upper and lower output constraints are as follows:

[0053] (2);

[0054] In the formula, t' represents the time period; g is the type sequence number of the aggregated unit, which can represent the coal-fired unit, the peak-shaving gas-fired unit and the heat and power gas-fired unit. represents the total output of the aggregated thermal power unit at the t'th time period; represents the total number of start-ups of the aggregated thermal power unit at the t'th time period; and are the maximum and minimum generation powers of the aggregated thermal power unit at the corresponding time period.

[0055] For the heat and power gas-fired unit, in addition to satisfying the upper and lower output constraints, it should also satisfy the basic requirement curve of the heating load.

[0056] (3);

[0057] In the formula, is the heating load value at each time point t.

[0058] For each type of unit, the ramping constraint should be satisfied, that is, the power change cannot exceed the ramping rate in the normal operation state but can break through the ramping rate limit when starting and stopping.

[0059] (4);

[0060] In the formula, and respectively represent the maximum ramping power and the maximum sliding power of the aggregated thermal power unit at adjacent time periods, and The number of units of the thermal power generating units at the two adjacent time points.

[0061] The start-stop constraint of the aggregated units:

[0062] (5);

[0063] In the formula: and respectively represent the minimum and maximum number of units of the aggregated thermal power generating units at time t.

[0064] The start-stop duration constraint of the aggregated units:

[0065] For coal-fired units, the coal-fired thermal power generating units are not allowed to be started and stopped again after being started at the first time point in a day.

[0066] (6).

[0067] For peak-shaving gas-fired units and cogeneration gas-fired units, no constraint is imposed on the start duration due to the convenience of start and stop.

[0068] The start-stop cost constraint of the thermal power generating units includes:

[0069] The start cost constraint:

[0070] (7);

[0071] The stop cost constraint:

[0072] (8);

[0073] In the formula: and respectively represent the cost of starting and stopping the aggregated thermal power generating units in the time period t', and are the unit start-stop cost of the thermal power generating units (unit: yuan / MW).

[0074] The power constraint of the thermal power generating units:

[0075] (9);

[0076] In the formula: j represents the type of the aggregated thermal power generating units, N is the number of coal / gas-fired units, is the output power of the aggregated thermal power generating units in the time period t', is the installed capacity corresponding to the type of the thermal power generating units.

[0077] The constraints related to pumped storage units include:

[0078] The upper and lower limit constraints of the output of the pumped storage units:

[0079] (10);

[0080] wherein: is the upper limit of the equivalent pumped storage unit output, whose value is equal to the installed capacity of pumped storage.

[0081] Pumped storage unit reservoir capacity constraint:

[0082] (11);

[0083] wherein: represents the equivalent generation / pumping capacity of pumped storage units in a day. The coefficient is actually a limit on the total amount of pumped storage units that can continuously pump / generate.

[0084] Other operation constraints include:

[0085] Positive and negative spinning reserve constraints:

[0086] (12);

[0087] wherein: represents the upper limit of the output of each type of unit, represents the lower limit of the output of each type of unit, is the spinning reserve coefficient, is the load value at time t, 0≤ ≤0.05.

[0088] Flexibility resource constraints and load constraints:

[0089] (13);

[0090] wherein: i takes wind power, photovoltaic, small hydropower, or load, represents the curtailed power of a certain type of power source or the cut-off power of load at time t, is the output or load value of the corresponding power source or load at time t.

[0091] Power balance constraint:

[0092] (14);

[0093] wherein: is the unbalanced power; , , are the actual outputs of coal, natural gas, and heat and cold at time t, respectively; is the actual output of pumped storage units at time t; represents the actual output of wind power, photovoltaic, and small hydropower at time t; For the load shedding power at the moment.

[0094] Third step: Put forward the two-stage scenario reduction framework based on GAMS-SCENRED tool, combined with absolute probability distance, relative probability distance, marginal relative probability distance and other indicators to determine the number and specific scenarios of reduced scenarios, balance the accuracy and computational efficiency of scenario information.

[0095] According to the historical data, the initial scenario set of new energy and load power is obtained, denoted as and , respectively. The scenario data is preprocessed, and the day-ahead new energy and load power scenario values are normalized by minimum-maximum.

[0096] 1. First-stage scenario reduction

[0097] Scenario reduction is performed for a single random variable, and scenario reduction is performed on and , combined with absolute probability distance, relative probability distance, marginal relative probability distance and other indicators, and the fast forward selection algorithm of GAMS-SCENRED is used for scenario reduction. At this time, the initial scenario set is , which contains scenarios of random variables such as new energy and load power, and the reduced scenario set is set to empty set. Set the termination condition of the algorithm, set the target value of marginal relative probability distance and the maximum cardinality of the reduced scenario set.

[0098] Among them, the absolute probability distance is calculated as follows:

[0099] (15);

[0100] In the formula: represents the random variable (new energy and load power); represents the probability of the original scenario s; is the normalized scenario distance of random variable p between scenarios s and in unit time, calculated by L1-norm, and the calculation formula is as follows:

[0101] (16);

[0102] In the formula: , are the normalized data of random variable p at time t under original scenario s and , respectively, where represents the scenario retained after scenario reduction.

[0103] The relative probability distance between the original scenario set and the reduced scenario set is the absolute probability distance divided by a constant (i.e. ). is the probability distance of a single scenario in the reduced scenario set.

[0104] (17);

[0105] where: is the probability distance of the reduced scenario set with only one scenario, which is calculated as follows:

[0106] (18);

[0107] where: is the normalized scenario distance between the original scenario s and in unit time, is the only scenario left after scenario reduction.

[0108] The marginal relative probability distance is calculated as follows:

[0109] (19);

[0110] The above formula reflects the marginal change in the relative probability distance when one scenario is added to the reduced scenario set each time.

[0111] The first scenario is selected from the initial scenario set The scenario that best represents the overall scenario set is selected from the initial scenario set and added to the reduced scenario set . In the selection, the absolute probability distance, the relative probability distance, and the marginal relative probability distance are used to evaluate the representation of each scenario. The absolute probability distance between each scenario and all other scenarios is calculated, and the scenario with the smallest absolute probability distance is selected to be added to because this scenario is the least different from the other scenarios and can best represent the characteristics of the original scenario set.

[0112] The subsequent scenarios are selected iteratively, and in each iteration, a scenario is selected from the remaining initial scenario set to be added to the reduced scenario set . The criterion for selection is that the scenario added should result in the greatest decrease in the relative probability distance between the reduced scenario set and the original scenario set , i.e., the scenario with the largest marginal relative probability distance is selected. The marginal relative probability distance after each remaining scenario is added to is calculated, and the scenario with the largest marginal relative probability distance is selected. The addition of this scenario can best improve the approximation of the reduced scenario set to the original scenario set.

[0113] Termination condition judgment, after adding a new scenario each time, check whether to meet the termination condition. If the marginal relative probability distance is lower than the set target value, and the cardinality of the reduced scenario set does not exceed the maximum limit, continue to iterate to select the next scenario; if the marginal relative probability distance is lower than the target value and the cardinality of the reduced scenario set reaches the maximum limit, or the marginal relative probability distance has not decreased significantly although the relative probability distance has decreased significantly, the algorithm stops.

[0114] Scenario probability redistribution, after determining the final reduced scenario set , it is necessary to redistribute the probability of scenarios in the reduced scenario set. This is because during the reduction process, some scenarios are removed. In order to ensure that the reduced scenario set can accurately reflect the probability distribution characteristics of the original scenario set, according to the probability distribution of the original scenario set and the relationship between the reduced scenario and the original scenario, the probability of each scenario in the reduced scenario set is determined again, so that the reduced scenario set is closer to the original scenario set in the probability level.

[0115] 2、Second stage scenario reduction

[0116] Combine the new energy and load power scenario set after the first step of reduction to obtain the combined scenario set. Again, use the fast forward selection algorithm of GAMS-SCENRED to reduce the combined scenario set, and add the average first moment difference and the average second moment difference index to evaluate the combined scenario reduction based on the first stage. The average first moment difference calculation formula is as follows:

[0117] (20);

[0118] In the formula: is the total number of time periods within the time span; is the first moment of the scenario set of the random variable p at time t; , the calculation formula is:

[0119] (21);

[0120] In the formula: is the normalized data of the random variable p at time t in the combined scenario S.

[0121] The average second moment difference calculation formula is as follows:

[0122] (22);

[0123] In the formula: indicates the second moment of the random variable p at time t in the scenario set The general formula of the second moment of the upper data is that the probability is , and the formula is:

[0124] (23).

[0125] The absolute probability distance, the relative probability distance, the marginal relative probability distance, the average first moment difference, and the average second moment difference are continuously calculated. When the marginal relative probability distance value is lower than the target value, and the average first moment difference and the average second moment difference indicators are significantly reduced in comprehensive analysis, it is determined that the combination scene set radix is the scene number of the final reduced scene set S, and the scene reduction process is completed.

[0126] The fourth step is to establish a comprehensive new energy carrying capacity evaluation index system to evaluate the influence of high proportion of new energy access on the power supply capacity, reliability and operation safety of the power grid.

[0127] The new energy carrying capacity of the power grid can be evaluated from four aspects, namely, power grid source-load characteristic index, power adequacy index, new energy consumption index, and power grid flexibility index. Among them, the power grid source-load characteristic index includes: new energy-load time correlation, net load peak-valley difference and change rate, net load positive / negative ramp rate and change rate. The power adequacy index includes: power deficiency probability, time expectation, power outage power expectation value, power supply shortage caused by power shortage expectation value, power deficiency frequency. The new energy consumption index includes: new energy consumption power, power consumption rate and power abandonment rate, wind power / photovoltaic power abandonment probability and power abandonment power expectation value. The power grid flexibility index includes: flexible regulation power supply proportion, flexibility deficiency probability, regulation capacity deficiency frequency.

[0128] The power system time sequence production simulation model constructed in the second step of the present application is the core simulation basis, which is based on the reduced scene determined in the third step two-stage scene reduction framework to simulate the operating state, and the output data such as unit output and power abandonment provide calculation basis for the fourth step comprehensive new energy carrying capacity evaluation index system; The representative scene is selected by the scene reduction of the third step to provide efficient input for the second step model under the premise of ensuring the core characteristics of the original scene, its accuracy affects the reliability of the model output data, and then affects the effectiveness of the fourth step index; The three form a closed loop of "input-simulation-evaluation", which jointly serves the goal of accurately and efficiently evaluating the new energy carrying capacity.

[0129] The method of the present application will be further described below according to specific embodiments.

[0130] As Figure 1 shown, taking the actual power data of a province as an example, the new energy carrying capacity evaluation method based on two-stage scene reduction of the present application includes the following steps:

[0131] Step 1: initialization, input power system installed capacity of power supply, load curve, new energy output data. The installed capacity of conventional power sources in a certain province is shown in Table 1; the installed capacity of new energy is shown in Table 2; the load average curve of a certain province is shown in Figure 2 .

[0132] Table 1 Installed capacity of conventional power sources in a certain province

[0133]

[0134] Table 2 Installed capacity of new energy

[0135]

[0136] Step 2: build a time series production simulation model of the power system: build a target function with the minimum unit operation cost as the objective function, the expression of which is shown in formula (1), consider the operation constraints of aggregated thermal power units and pumped storage units, such as formula (2)-(14).

[0137] Step 3: propose a two-stage scenario reduction framework based on GAMS-SCENRED tool, combined with absolute probability distance, relative probability distance, marginal relative probability distance and other indicators, determine the number and specific scenarios of reduced scenarios, balance the accuracy and computational efficiency of scenario information.

[0138] Step 4: establish a comprehensive new energy carrying capacity evaluation index system to evaluate the impact of high proportion of new energy access on power supply capacity, reliability and operation safety of the power grid.

[0139] To verify the effectiveness of the method of the application, the method of the application is compared with the existing method. Method 1: based on all time series, the method simulates the use of all as a scenario without screening. Method 2: select 18 typical days in the order of natural days. Table 3 summarizes the average absolute normalized error, root mean square normalized error and calculation time of multi-time series time series simulation using the three methods. Compared with the baseline method, the calculation time of the method of the application and method 2 is smaller, and the calculation time of the method of the application is smaller than that of method 2. In addition, the two indexes using the method of the application are better than the approximate performance of method 2.

[0140] Table 3 Comparison table

[0141] .

[0142] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction, characterized in that: Includes the following steps: Step 1: Initialization, input the installed power capacity, load curve, and renewable energy output data of the power system; Step 2: Construct a time-series production simulation model for the power system: with the objective function of minimizing unit operating costs, penalties are imposed for curtailment of solar power, wind power, hydropower, and load shedding; and the operating constraints of integrated thermal power units, pumped storage units, positive and negative spinning reserve constraints, flexibility resource constraints, load constraints, and power balance constraints are also considered. The objective function is to minimize the unit's operating cost while penalizing curtailment of solar, wind, and hydropower, as well as load shedding. The expression for the objective function is as follows: ； In the formula, The power generation cost of thermal power units; To represent the total cost of starting and stopping a thermal power unit; The total cost of ramping uphill for thermal power units. Cost of pumped-storage hydroelectric power generation; This includes penalties for load shedding and penalties for curtailing wind / solar / hydro resources. It concerns the spinning reserve factor. Rewards; , and These are the coefficients of the quadratic, linear, and constant terms of the fuel cost-power generation function for thermal power units, respectively. and These are the start-up and shutdown costs of thermal power units at various times; This represents the total operating output of the combined gas-fired or coal-fired power units during time period t'. This indicates the unit power generation cost of a pumped storage hydroelectric power unit. This represents the total operating output of the pumped storage unit during time period t'. and These represent the penalty costs for load shedding and curtailment of flexible resources, respectively. and These represent the load shedding power and the power curtailed from flexible resources at time t, respectively. Step 3: Construct a two-stage scenario reduction framework, combining absolute probability distance, relative probability distance, and marginal relative probability distance indicators to determine the number and specific scenarios to be reduced; Step 4: Establish a comprehensive new energy carrying capacity assessment index system to evaluate the impact of the power grid undertaking a high proportion of new energy access on the power grid's power supply capacity, reliability, and operational safety.

2. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 1, characterized in that: The operational constraints of aggregated thermal power units are established by classifying thermal power units into coal-fired units, peak-shaving gas-fired units, and combined heat and power gas-fired units, and then aggregating the constraints of units of the same type. The operational constraints of aggregated thermal power units include: Output upper and lower limits constraints for each coal-fired unit and peak-shaving gas-fired unit, output upper and lower limits constraints and heating load constraints for cogeneration gas-fired units, ramp-up constraints for polymerization units, start-up and shutdown constraints for polymerization units, start-up duration constraints for coal-fired units, and start-up and shutdown cost constraints for polymerization units.

3. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 1, characterized in that: The operational constraints of pumped storage units include: upper and lower limits of pumped storage unit output and reservoir capacity constraints.

4. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 1, characterized in that: A two-stage scene reduction framework was built based on the GAMS-SCENRED tool.

5. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 4, characterized in that: The first stage of the two-stage scene reduction framework is a scene reduction process for a single random variable. The second stage is a scene reduction of the combined scene set after the first stage of scene reduction. The combined scene reduction is evaluated using the average first-order moment difference and the average second-order moment difference indices.

6. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 5, characterized in that: The initial scenario sets for new energy sources and load power are obtained based on historical data, and are denoted as follows: and And let the initial scene set be For scenarios that include random variables of new energy sources and load power, a reduced scenario set is set. For an empty set, the first stage of scene reduction involves the following steps: Select the first scene from the initial scene set. Select the scene that is most representative of the overall scene set and add it to the reduced scene set. ; Iteratively select subsequent scenes, in each iteration from the remaining set of initial scenes. - Select one scene to add to the reduced scene set ; Termination condition determination: After each new scene is added, check whether the termination condition is met. Scene probabilities are redistributed to determine the final reduced scene set. Afterwards, the probability of reducing the number of scenes needs to be redistributed.

7. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 6, characterized in that: The selection of the first scenario is based on calculating the absolute probability distance of each scenario to all other scenarios, and choosing the scenario that minimizes the absolute probability distance. ; The selection of scenes in subsequent iterations is based on whether adding a scene reduces the scene set. With the original scene set The scenario that maximizes the marginal relative probability distance is the one in which the relative probability distance decreases the most. The termination condition is as follows: if the marginal relative probability distance is lower than the set target value and the cardinality of the reduced scene set does not exceed the maximum limit, then continue iterating to select the next scene; if the marginal relative probability distance is lower than the target value and the cardinality of the reduced scene set reaches the maximum limit, or if the marginal relative probability distance is not lower than the target value but no scene can be found that can significantly reduce the relative probability distance, then the algorithm stops.

8. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 7, characterized in that: Absolute probability distance The calculation formula is as follows: ； In the formula: Represents a random variable; Represents the probability of the original scene s; Let p be the random variable within a unit of time, representing the original scene s and the scene retained after scene reduction. The scene distance between them is normalized and calculated using the L1 norm; The relative probability distance between the original scene set and the reduced scene set is the absolute probability distance divided by the probability distance when the reduced scene set contains only one scene. The calculation formula is as follows: ； In the formula: This is the probabilistic distance when the scene set contains only one scene, calculated using the following formula: Marginal relative probability distance The calculation formula is as follows: 。 9. The method for assessing the carrying capacity of new energy sources based on multi-timescale time-series simulation with two-stage scenario reduction as described in claim 1, characterized in that: The comprehensive evaluation index system for renewable energy carrying capacity includes grid source-load characteristics index, power sufficiency index, renewable energy absorption index, and grid flexibility index.

Images