Time sequence-probability electric power and electric quantity balance quantitative evaluation method for electric power system

Abstract

The invention provides a time sequence-probability electric power and electric quantity balance quantitative evaluation method for an electric power system. The method comprises the following steps: step 1, establishing quantitative evaluation indexes of time sequence-probability electric power and electric quantity balance; the quantitative evaluation indexes comprise a balance probability, an unbalance amount and an unbalance risk; 2, constructing a robust operation simulation model; with the purpose of minimizing the total operation cost of the worst scene, production simulation calculation of the worst scene in an adjustment demand joint confidence interval under a given confidence level is realized; 3, calculating quantitative evaluation indexes; the joint confidence interval as large as possible is found by adjusting the joint confidence interval so as to approximate a balance domain, and accurate calculation of each quantitative evaluation index is realized. By applying the technical scheme, the defects in the prior art can be overcome, and the traditional probability reliability index and the deterministic load loss / abandoned power index are expanded.

Description

Technical Field

[0001] The invention relates to the technical field of power systems, and in particular to a method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system. Background Art

[0002] With the large-scale grid connection of new energy and the continuous improvement of the power market, the operation and management of the power system are facing huge challenges. As the foundation and core of the power system, the power and electricity balance calculation and analysis technology is crucial to ensure the stable, economic and safe operation of the power system. Through the research on the power and electricity balance calculation and analysis technology of the new power system, the operation efficiency and management level of the power system can be improved, the grid loss and load pressure can be reduced, and the efficient utilization of new energy and the healthy development of the power market can be promoted. Determining appropriate quantitative evaluation indicators is the basis for the quantitative analysis of time-series-probabilistic power and electricity balance. In general, the existing power and electricity balance evaluation indicators can be divided into probability and time-series categories. However, there is currently a lack of a unified indicator system that can simultaneously reflect the time-series and probability characteristics of power and electricity balance. How to propose a new indicator system that is compatible with the time-series-probabilistic power and electricity balance theory has become a technical problem that needs to be solved urgently.

[0003] At the same time, designing an accurate and effective calculation method for evaluation indicators is the key to the quantitative analysis of time series-probabilistic power and electricity balance. The existing calculation methods for power and electricity balance evaluation indicators mainly include analytical solution method, Monte Carlo simulation method, typical scenario method and time-division probabilistic balance method. However, these methods are insufficient in the application of time series-probabilistic power and electricity balance analysis. The theory of time series-probabilistic power and electricity balance involves the key concept of balance domain, but the effect of time series coupling makes the form of balance domain very complex when the number of time sections is high, and it is difficult to characterize analytically. The underlying logic of the Monte Carlo simulation method is to approximate the probability results with the statistical frequency results. The problem is that in order to achieve algorithm convergence, as the number of time sections increases, the number of scenarios required to be simulated will explode, and the calculation time will be unbearable. Although the typical scenario method avoids the high computational complexity of the Monte Carlo simulation method, its evaluation results depend on the quality of typical scenarios. However, typical scenarios are often defined based on the typicality of time series characteristics, which is not equivalent to the typicality of power and electricity balance, which may lead to large deviations in the evaluation results. Both the Monte Carlo simulation method and the typical scenario method are implemented based on the time series unit combination. The time period probability balance law is based on the traditional probability balance method, and the balance result of the previous period is used as the initial condition for the probability balance of the next period. However, this approach has insufficient forward-looking prediction capabilities and incomplete modeling of time series coupling characteristics. Therefore, how to design an accurate and effective calculation method for evaluation indicators is also an urgent problem to be solved. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method for quantitatively evaluating the timing-probabilistic power and electricity balance of a power system to address the deficiencies in the prior art and to expand traditional probabilistic reliability indicators and deterministic load loss / abandonment indicators.

[0005] To achieve the above object, the present invention adopts the following technical solution: a method for quantitatively evaluating the time series-probabilistic power and quantity balance of a power system, comprising the following steps:

[0006] Step 1: Establishment of quantitative evaluation indicators for time series-probabilistic power and quantity balance; the quantitative evaluation indicators include balance probability, imbalance amount and imbalance risk;

[0007] Step 2: Construction of a robust operation simulation model; with the goal of minimizing the total operating cost of the worst scenario, while considering the operating constraints of various adjustable resources such as direct control and aggregate response, active power balance constraints, uncertain set constraints of system regulation requirements, and variable logic constraints; using the KKT conditions of the inner layer problem of the constructed model, the double-layer optimization model is transformed into a single-layer minimization model for solution, and the production simulation calculation of the worst scenario within the joint confidence interval of the regulation demand under a given confidence level is realized;

[0008] Step 3: Calculation of quantitative evaluation indicators; by adjusting the joint confidence interval to find the largest possible joint confidence interval to approximate the balance domain, accurate calculation of each quantitative evaluation indicator is achieved.

[0009] In a preferred embodiment, in step 1, the balance probability BP is defined as the probability that the system regulation demand is covered by the balance domain, that is:

[0010]

[0011] Unbalanced quantity: a point ξ∈Ω in the unbalanced domain U The corresponding unbalance vector Δp ​​is defined as the distance from this point to the closest point p on the equilibrium domain. * The absolute value of the difference, that is:

[0012] Δp:=|ξ-p * |

[0013]

[0014] According to ξ-p * The positive or negative value of each element can determine whether the imbalance is caused by insufficient upward adjustment or insufficient downward adjustment; the upward imbalance vector Δp + and reduce the unbalance vector Δp - The calculation formula is as follows:

[0015] Δp + :=(ξ-p* ) +

[0016] Δp - :=-(ξ-p * ) -

[0017] Among them, (a) - =min{a,0}, (a) + =max{a,0};

[0018] The imbalance risk is defined as the expectation of the imbalance amount of all scenarios that cannot balance the time series, that is:

[0019]

[0020] Among them, bold 1 represents a column vector of all 1s;

[0021] Based on the upward adjustment of the unbalance value Δp + and reduce the unbalance Δp - We can define upward imbalance risk and downward imbalance risk, denoted as UR + and UR - , the calculation expression is:

[0022]

[0023]

[0024] From the perspective of spatial analysis, imbalance risk describes the expected imbalance amount at all points in the imbalance domain, while from the perspective of temporal analysis, imbalance risk reflects the expected load loss or power abandonment in all imbalance regulation demand scenarios; the imbalance risk index is an extension of the traditional single-section expectation reliability index.

[0025] In a preferred embodiment, the objective function in step 2 specifically includes: the goal of the robust operation simulation model is to minimize the total operating cost of the worst scenario; the operating cost includes the thermal power generation cost and the penalty cost of load loss and power abandonment, that is:

[0026]

[0027] in, c loss 、c a are the cost coefficient of unit power generation, unit load loss and unit power abandonment of thermal power cluster i respectively; is the power generation vector of thermal power cluster i; is the load loss power vector; is the power vector of abandoned power; D represents the total number of time sections; nG represents the number of thermal power clusters; is the system adjustment demand variable vector, which is the decision variable of the inner optimization problem;

[0028] p sys =[p G ;p H ;p S ;p H2 ;p AGG ] is the active power variable vector of various adjustable resources, which is the decision variable of the outer optimization problem. The superscripts H, S, H2 and AGG represent hydropower, energy storage, large-scale hydrogen production adjustable loads and distributed adjustable resource clusters with aggregated access, respectively.

[0029] In a preferred embodiment, the constraints of the robust operation simulation model in step 2 include operation constraints of various adjustable resources such as direct control type and aggregate response type, active power balance constraints, uncertain set constraints of system regulation requirements, and variable logic constraints, which are specifically expressed as follows;

[0030] (A) Operational constraints of directly controllable resources

[0031] Thermal power operation constraints:

[0032]

[0033] in, and are the polyhedron feasible domain coefficient matrix and parameter vector of thermal power cluster i respectively;

[0034] Energy storage operation constraints:

[0035]

[0036] in, is the net discharge power vector of energy storage i, and are respectively the coefficient matrix and parameter vector of the polyhedron feasible domain of energy storage i, n S Indicates the amount of directly regulated energy storage;

[0037] Large-scale hydrogen production with adjustable load operation constraints:

[0038]

[0039] in, is the power consumption vector of the adjustable load i for hydrogen production by electricity, and are respectively the polyhedron feasible domain coefficient matrix and parameter vector of the adjustable load i for hydrogen production by electricity, n H2 Indicates the quantity of electric hydrogen production load;

[0040] (B) Operational constraints of aggregated responsive adjustable resources

[0041] Aggregate power constraints for each aggregation access point:

[0042]

[0043] in, is the power consumption vector of the aggregated access point i, and are the feasible domain coefficient matrix and parameter vector of the polyhedron of the aggregated access point i respectively;

[0044] (C) System active power balance constraints

[0045]

[0046] (D) Variation range of system regulation requirements

[0047] cl≤ξ sys ≤cu

[0048] Among them, cl and cu correspond to the joint confidence lower bound and upper bound of the joint confidence interval under the 1-α confidence level;

[0049] (E) Variable logical constraints

[0050] p loss ≥0

[0051] p a ≥0.

[0052] In a preferred embodiment, the original model in step 2 is a minimum-maximum robust optimization problem, which can be converted into a single-layer minimization model for solution by using the KKT condition of the inner layer problem;

[0053] First, the original model is expressed as a generalized abstract model:

[0054]

[0055]

[0056] The abstract model can be decomposed into an outer model and an inner model, namely:

[0057]

[0058] sA out p sys ≤ b out

[0059]

[0060]

[0061] Among them, the KKT condition of the inner optimization problem is:

[0062]

[0063] Thus, by combining the KKT conditions with the outer model constraints, the final model after transformation can be obtained, as shown in the following formula. The transformed optimization model can be solved by a commercial solver;

[0064]

[0065] sA out p sys ≤ b out

[0066]

[0067] In a preferred embodiment, the calculation of the approximate balance domain in step 3 is specifically as follows: when the imbalance amount of the worst scenario is found to be non-zero, the confidence level requirement is lowered, and then the calculation is recalculated; the robust evaluation solution only takes the same confidence level for each time section during the initial solution and obtains a joint confidence interval, and then only reduces the confidence interval corresponding to the time section where the imbalance amount of the worst scenario is non-zero to form a new joint confidence interval;

[0068] Assume that the confidence interval of the t-th time section before adjustment is [cl t ,cu t ]; The adjustment strategies of confidence intervals are different when the imbalance amount is adjusted upward or downward;

[0069] Assume that the upward imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the upper bound of the confidence interval: Recorded as strategy 1;

[0070] Assume that the downward adjustment imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the lower bound of the confidence interval: This is called strategy 2.

[0071] In a preferred embodiment, the calculation of each quantitative index in step 3 is specifically as follows:

[0072] Combined with the probability density distribution function of the regulation demand, it is easy to obtain the integral result and obtain the evaluation indicators of the time-series-probabilistic power balance;

[0073] From the simplified integral domain, the equilibrium probability calculation formula can be calculated by the following formula:

[0074]

[0075] The calculation formulas for upward adjustment of unbalance, downward adjustment of unbalance and unbalance can be simplified as follows:

[0076] Δp + :=(ξ-cu) +

[0077] Δp - :=-(ξ-cl) -

[0078] Δp=Δp + +Δp -

[0079] Based on the calculation results of the imbalance amount, the calculation formulas for upward imbalance risk, downward imbalance risk and imbalance risk index can be calculated by the following formulas respectively:

[0080]

[0081]

[0082]

[0083] in, Represents the inverse function of the marginal distribution of regulation demand at the tth time section.

[0084] Compared with the prior art, the present invention has the following beneficial effects: a unified indicator system capable of simultaneously reflecting the timing and probability characteristics of power and electricity balance, and realizing mutual adaptation between the indicator system and the timing-probabilistic power and electricity balance theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 A schematic flow chart of a robust evaluation method according to a preferred embodiment of the present invention;

[0086] Figure 2 A schematic diagram of the unbalanced amount in the quantitative evaluation index of the time series-probabilistic power and quantity balance analysis in a preferred embodiment of the present invention;

[0087] Figure 3 The figure is a flowchart of a robust evaluation solution algorithm according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0088] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0089] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0090] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0091] A quantitative evaluation method for the timing-probabilistic power and electricity balance of a power system, referring to Figure 1-3 , including: Step 1: Establishment of quantitative evaluation indicators for time series-probabilistic power and electricity balance analysis, which is used to simultaneously reflect the time series and probabilistic characteristics of power and electricity balance, and realize the expansion of traditional probabilistic reliability indicators and deterministic load loss / abandonment indicators; the quantitative evaluation indicators of time series-probabilistic power and electricity balance analysis specifically include: balance probability, imbalance and imbalance risk, etc. Arrangement 2: Robust operation simulation modeling, with the robust operation simulation model as the core, seek the maximum joint confidence interval that can be balanced in the worst regulation demand scenario, and use it as the approximate result of the balance domain; the robust operation simulation model specifically includes: constructing the total operation cost function of the worst scenario as the objective function of the robust operation simulation model; constructing the operation constraints of various adjustable resources such as direct control type and aggregate response type, active balance constraints, uncertain set constraints of system regulation demand and variable logic constraints as constraints of robust operation simulation. Step 3: Quantitative evaluation indicator calculation method, used to find the largest possible joint confidence interval to approximate the balance domain, and realize the accurate calculation of each quantitative evaluation indicator. The steps of the quantitative evaluation index calculation method include: adjusting the distribution function of demand, the feasible domain and confidence level of adjustable resources; solving the joint confidence upper and lower bounds under a given confidence level, and then solving the robust operation simulation model; adjusting the joint confidence interval according to the production simulation results of the worst adjustment demand scenario; and calculating the quantitative evaluation indicators of the time series-probabilistic power and electricity balance analysis according to the joint confidence interval results: balance probability, imbalance amount and imbalance risk.

[0092] The main steps are as follows:

[0093] Step 1: Establishment of quantitative evaluation indicators for time-series-probabilistic power and electricity balance.

[0094] Based on the concepts of balance domain and unbalance domain in the time series-probabilistic power and electricity balance theory, quantitative evaluation indicators for time series-probabilistic power and electricity balance analysis are proposed, including balance probability, unbalance amount and unbalance risk. Among them, balance probability (BP) is defined as the probability that the system regulation demand is covered by the balance domain, that is:

[0095]

[0096] Unbalanced quantity: a point ξ∈Ω in the unbalanced domain U The corresponding unbalance vector Δp ​​is defined as the distance from this point to the closest point p on the equilibrium domain. * The absolute value of the difference, that is:

[0097] Δp:=|ξ-p * |

[0098]

[0099] According to ξ-p * The positive or negative value of each element can determine whether the imbalance is caused by insufficient upward adjustment or insufficient downward adjustment. + and reduce the unbalance vector Δp - The calculation formula is as follows:

[0100] Δp + :=(ξ-p * ) +

[0101] Δp - :=-(ξ-p * ) -

[0102] Among them, (a) - =min{a,0}, (a) + =max{a,0}.

[0103] The schematic diagram of the spatial representation of the unbalanced quantity is as follows Figure 2 The red area in the figure represents the balanced domain, and the blue area represents the unbalanced domain. Take a point ξ in the unbalanced domain, which is closest to the point p in the balanced domain. * The absolute value of the difference constitutes the unbalance vector: Δp = |ξ-p * |=[a,b] T According to the calculation formula of upward and downward unbalance, Δp can be obtained. + =(ξ-p * ) + =[0,0] T , and Δp -=-(ξ-p * ) - =[a,b] T , indicating that the unbalanced consequences of the example are caused by insufficient downward adjustment capacity in the two periods.

[0104] The imbalance risk is defined as the expectation of the imbalance amount of all scenarios that cannot balance the time series, that is:

[0105]

[0106] Among them, the bold 1 represents a column vector of all 1s.

[0107] Furthermore, based on the increase in the unbalance value Δp + and reduce the unbalance Δp - We can define upward imbalance risk and downward imbalance risk, denoted as UR + and UR - , the calculation expression is:

[0108]

[0109]

[0110] From the perspective of spatial analysis, the imbalance risk describes the expected imbalance amount of all points in the imbalance domain, while from the perspective of temporal analysis, the imbalance risk reflects the expected load loss or power abandonment in all imbalance regulation demand scenarios. The imbalance risk index is an extension of the traditional single-section expectation reliability index.

[0111] Step 2: Construction of robust operation simulation model.

[0112] For the operation simulation link, a robust operation simulation model is constructed to realize the production simulation calculation of the worst scenario within the joint confidence interval of the adjustment demand under a given confidence level.

[0113] (1) Objective function

[0114] The goal of the robust operation simulation model is to minimize the total operating cost of the worst scenario. The operating cost includes the cost of thermal power generation and the penalty cost of load loss and power abandonment, namely:

[0115]

[0116] in, c loss 、c a are the cost coefficient of unit power generation, unit load loss and unit power abandonment of thermal power cluster i respectively; is the power generation vector of thermal power cluster i; is the load loss power vector; is the power vector of abandoned power; D represents the total number of time sections; n G represents the number of thermal power clusters; is the system adjustment demand variable vector, which is the decision variable of the inner optimization problem; sys =[p G ;p H ;p S ;p H2 ;p AGG ] is the active power variable vector of various adjustable resources, which is the decision variable of the outer optimization problem. The superscripts H, S, H2 and AGG represent hydropower, energy storage, large-scale hydrogen production adjustable loads and distributed adjustable resource clusters with aggregated access, respectively.

[0117] (2) Constraints

[0118] The constraints of the robust operation simulation include the operation constraints of various adjustable resources such as direct control type and aggregate response type, active power balance constraints, uncertain set constraints of system regulation requirements, and variable logic constraints, which are specifically expressed as follows.

[0119] (A) Operational constraints of directly controllable resources

[0120] Thermal power operation constraints:

[0121]

[0122] in, and are the polyhedron feasible domain coefficient matrix and parameter vector of thermal power cluster i respectively.

[0123] Energy storage operation constraints:

[0124]

[0125] in, is the net discharge power vector of energy storage i, and are respectively the coefficient matrix and parameter vector of the polyhedron feasible domain of energy storage i, n S Indicates the amount of directly regulated energy storage.

[0126] Large-scale hydrogen production with adjustable load operation constraints:

[0127]

[0128] in, is the power consumption vector of the adjustable load i for hydrogen production by electricity, and are respectively the polyhedron feasible domain coefficient matrix and parameter vector of the adjustable load i for hydrogen production by electricity, n H2Indicates the quantity of electric hydrogen production load.

[0129] (B) Operational constraints of aggregated responsive adjustable resources

[0130] Aggregate power constraints for each aggregation access point:

[0131]

[0132] in, is the power consumption vector of the aggregated access point i, and are the feasible domain coefficient matrix and parameter vector of the polyhedron of aggregated access point i respectively.

[0133] (C) System active power balance constraints

[0134]

[0135] (D) Variation range of system regulation requirements

[0136] cl≤ξ sys ≤cu

[0137] Among them, cl and cu correspond to the joint confidence lower bound and upper bound of the joint confidence interval under the 1-α confidence level.

[0138] (E) Variable logical constraints

[0139] p loss ≥0

[0140] p a ≥0

[0141] (3) Model transformation

[0142] The aforementioned original model is a minimum-maximum robust optimization problem. By using the KKT conditions of the inner layer problem, it can be converted into a single-layer minimization model for solution.

[0143] First, the original model is expressed as a generalized abstract model:

[0144]

[0145]

[0146] The abstract model can be decomposed into an outer model and an inner model, namely:

[0147]

[0148] sA out p sys ≤ b out

[0149]

[0150]

[0151] Among them, the KKT condition of the inner optimization problem is:

[0152]

[0153] Therefore, by combining the KKT conditions with the outer model constraints, the final model after transformation can be obtained, as shown in the following formula. The transformed optimization model can be solved by a commercial solver.

[0154]

[0155] sA out p sys ≤ b out

[0156]

[0157] Step 3: Calculation of quantitative evaluation indicators.

[0158] (1) Calculation of approximate equilibrium domain

[0159] The accurate calculation of each quantitative evaluation index requires finding the largest possible joint confidence interval to approximate the equilibrium domain. The solution of the robust operation simulation model requires a given confidence level as a boundary condition, but the initial given confidence level may not be appropriate. Therefore, this paper considers lowering the confidence level requirement and then recalculating when the imbalance amount of the worst scenario is found to be non-zero.

[0160] The specific solution algorithm process is as follows Figure 3 shown.

[0161] Considering that imbalances often occur at specific time sections in practice, if the confidence intervals of each section are reduced in the same proportion, the joint confidence interval will be too small and the evaluation result will be too conservative. In this regard, the robust evaluation solution in this paper only takes the same confidence level for each time section in the initial solution and obtains the joint confidence interval, and then only reduces the confidence interval corresponding to the time section with non-zero imbalance in the worst scenario to form a new joint confidence interval.

[0162] Assume that the confidence interval of the t-th time section before adjustment is [cl t ,cu t ]. The adjustment strategies of the confidence interval are different for different situations of upward and downward adjustment of the imbalance.

[0163] Assume that the upward imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the upper bound of the confidence interval: This is called strategy 1.

[0164] Assume that the downward adjustment imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the lower bound of the confidence interval: This is called strategy 2.

[0165] (2) Calculation methods of various quantitative indicators

[0166] Since the balance domain is approximated by the result of the joint confidence interval, and the joint confidence interval is a cuboid in space, the integral domain form can be greatly simplified. Combined with the probability density distribution function of the regulation demand, it is easy to obtain the integral result and obtain the evaluation indicators of the time-series-probabilistic power and electricity balance.

[0167] From the simplified integral domain, the equilibrium probability calculation formula can be calculated by the following formula:

[0168]

[0169] The calculation formulas for upward adjustment of unbalance, downward adjustment of unbalance and unbalance can be simplified as follows:

[0170] Δp + :=(ξ-cu) +

[0171] Δp - :=-(ξ-cl) -

[0172] Δp=Δp + +Δp -

[0173] Based on the calculation results of the imbalance amount, the calculation formulas for upward imbalance risk, downward imbalance risk and imbalance risk index can be calculated by the following formulas respectively:

[0174]

[0175]

[0176]

[0177] in, Represents the inverse function of the marginal distribution of regulation demand at the tth time section.

Claims

1. A method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system, characterized in that: The following steps are involved: Step 1: Establishment of quantitative evaluation indicators for time series-probabilistic power and quantity balance; the quantitative evaluation indicators include balance probability, imbalance amount and imbalance risk; Step 2: Construction of a robust operation simulation model; with the goal of minimizing the total operating cost of the worst scenario, while considering the operating constraints of various adjustable resources such as direct control and aggregate response, active power balance constraints, uncertain set constraints of system regulation requirements, and variable logic constraints; using the KKT conditions of the inner layer problem of the constructed model, the double-layer optimization model is transformed into a single-layer minimization model for solution, and the production simulation calculation of the worst scenario within the joint confidence interval of the regulation demand under a given confidence level is realized; Step 3: Calculation of quantitative evaluation indicators; by adjusting the joint confidence interval to find the largest possible joint confidence interval to approximate the balance domain, accurate calculation of each quantitative evaluation indicator is achieved.

2. A method for quantitatively evaluating the time-series-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: In step 1, the balance probability BP is defined as the probability that the system regulation demand is covered by the balance domain, that is: Unbalanced quantity: a point ξ∈Ω in the unbalanced domain U The corresponding unbalance vector Δp ​​is defined as the distance from this point to the closest point p on the equilibrium domain. * The absolute value of the difference, that is: Δp:=|ξ-p * | According to ξ-p * The positive or negative value of each element can determine whether the imbalance is caused by insufficient upward adjustment or insufficient downward adjustment; the upward imbalance vector Δp + and reduce the unbalance vector Δp - The calculation formula is as follows: Δp + :=(ξ-p * ) + Δp - :=-(ξ-p * ) - Among them, (a) - = min{a,0}, (a) + = max{a,0}; The imbalance risk is defined as the expectation of the imbalance amount of all scenarios that cannot balance the time series, that is: Among them, bold 1 represents a column vector of all 1s; Based on the upward adjustment of the unbalance value Δp + and reduce the unbalance Δp - We can define upward imbalance risk and downward imbalance risk, denoted as UR + and UR - , the calculation expression is: From the perspective of spatial analysis, imbalance risk describes the expected imbalance amount at all points in the imbalance domain, while from the perspective of temporal analysis, imbalance risk reflects the expected load loss or power abandonment in all imbalance regulation demand scenarios; the imbalance risk index is an extension of the traditional single-section expectation reliability index.

3. A method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: The objective function in step 2 specifically includes: the goal of the robust operation simulation model is to minimize the total operating cost of the worst scenario; the operating cost includes the thermal power generation cost and the penalty cost of load loss and power abandonment, that is: in, c loss 、c a are the cost coefficient of unit power generation, unit load loss and unit power abandonment of thermal power cluster i respectively; is the power generation vector of thermal power cluster i; is the load loss power vector; is the power vector of abandoned power; D represents the total number of time sections; n G Indicates the number of thermal power clusters; is the system adjustment demand variable vector, which is the decision variable of the inner optimization problem; p sys =[p G ;p H ;p S ;p H2 ;p AGG ] is the active power variable vector of various adjustable resources, which is the decision variable of the outer optimization problem. The superscripts H, S, H2 and AGG represent hydropower, energy storage, large-scale hydrogen production adjustable loads and distributed adjustable resource clusters with aggregated access, respectively.

4. A method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: The constraints of the robust operation simulation model in step 2 include operation constraints of various adjustable resources such as direct control type and aggregate response type, active power balance constraints, uncertain set constraints of system regulation requirements, and variable logic constraints, which are specifically expressed as follows; (A) Operational constraints of directly controllable resources Thermal power operation constraints: in, and are the polyhedron feasible domain coefficient matrix and parameter vector of thermal power cluster i respectively; Energy storage operation constraints: in, is the net discharge power vector of energy storage i, and are respectively the coefficient matrix and parameter vector of the polyhedron feasible domain of energy storage i, n S Indicates the amount of directly regulated energy storage; Large-scale hydrogen production with adjustable load operation constraints: in, is the power consumption vector of the adjustable load i for hydrogen production by electricity, and are respectively the polyhedron feasible domain coefficient matrix and parameter vector of the adjustable load i for hydrogen production by electricity, n H2 Indicates the quantity of electric hydrogen production load; (B) Operational constraints of aggregated responsive adjustable resources Aggregate power constraints for each aggregation access point: in, is the power consumption vector of the aggregated access point i, and are the feasible domain coefficient matrix and parameter vector of the polyhedron of the aggregated access point i respectively; (C) System active power balance constraints (D) Variation range of system regulation requirements cl≤ξ sys ≤cu Among them, cl and cu correspond to the joint confidence lower bound and upper bound of the joint confidence interval under the 1-α confidence level; (E) Variable logical constraints p loss ≥0 p a ≥0。 5. A method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: The original model in step 2 is a minimum-maximum robust optimization problem, which can be converted into a single-layer minimization model for solution using the KKT condition of the inner layer problem; First, the original model is expressed as a generalized abstract model: The abstract model can be decomposed into an outer model and an inner model, namely: s.t.A out p sys ≤b out Among them, the KKT condition of the inner optimization problem is: Thus, by combining the KKT conditions with the outer model constraints, the final model after transformation can be obtained, as shown in the following formula. The transformed optimization model can be solved by a commercial solver; s.t.A out p sys ≤b out 6. A method for quantitatively evaluating the timing-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: The calculation of the approximate balance domain in step 3 is specifically as follows: when the imbalance amount of the worst scenario is found to be non-zero, the confidence level requirement is lowered, and then the calculation is recalculated; the robust evaluation solution only takes the same confidence level for each time section during the initial solution and obtains a joint confidence interval, and then only reduces the confidence interval corresponding to the time section where the imbalance amount of the worst scenario is non-zero, to form a new joint confidence interval; Assume that the confidence interval of the t-th time section before adjustment is [cl t ,cu t ]; The adjustment strategies of confidence intervals are different when the imbalance amount is adjusted upward or downward; Assume that the upward imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the upper bound of the confidence interval: Recorded as strategy 1; Assume that the downward adjustment imbalance of the tth time section in the worst adjustment demand timing scenario is Then adjust the lower bound of the confidence interval: This is called strategy 2.

7. A method for quantitatively evaluating the time-series-probabilistic power and quantity balance of a power system according to claim 1, characterized in that: The calculation of each quantitative index in step 3 is specifically as follows: Combined with the probability density distribution function of the regulation demand, it is easy to obtain the integral result and obtain the evaluation indicators of the time-series-probabilistic power balance; From the simplified integral domain, the equilibrium probability calculation formula can be calculated by the following formula: The calculation formulas for upward adjustment of unbalance, downward adjustment of unbalance and unbalance can be simplified as follows: Based on the calculation results of the imbalance amount, the calculation formulas for upward imbalance risk, downward imbalance risk and imbalance risk index can be calculated by the following formulas respectively: in, (F t sys ) -1 Represents the inverse function of the marginal distribution of regulation demand at the tth time section.

Images