System node risk assessment method considering multiple uncertainties under extreme cold weather

Abstract

The present invention discloses a method for risk assessment of system nodes under extreme cold weather conditions taking into account multiple uncertainties. A system standard probability model of the power system is established, the load value of the power node, the status of the transmission line, the output of the wind turbine and the fuel cost parameter of the natural gas unit are input, and the power cost parameter of the power node is output; based on the power cost parameter of each power node of the power system, the spatiotemporal risk index of each power node of the power system is obtained, and the risk assessment of the power nodes under extreme cold weather is performed. The method of the present invention comprehensively considers the fusion of multiple uncertainties under extreme weather conditions and the spatiotemporal distribution characteristics of power cost parameters, and can effectively identify the risk distribution of power cost parameters under extreme weather conditions. This method can not only help the power system establish a risk prevention and control strategy under extreme weather conditions, but also provide effective guidance for establishing a highly resilient power grid planning under extreme weather conditions.

Description

Technical Field

[0001] The present invention relates to a system node risk assessment method, and in particular to a system node risk assessment method considering multiple uncertainties in extreme cold weather. Background Art

[0002] In recent years, with the frequent occurrence of extreme cold weather, the power system has been impacted by prolonged spikes in power cost parameters, placing a significant burden on electricity users. Typically, extreme cold weather is accompanied by sub-zero temperatures and periods of freezing temperatures. Consequently, the impacts of extreme cold weather on the power system include generator output losses due to icing and overhead line failures. Furthermore, extreme cold temperatures can lead to a sharp increase in heating demand, and consequently, electricity demand. These impacts can severely disrupt the supply-demand balance in the power system, leading to a sharp increase in power cost parameter risks.

[0003] Considering the impact of extreme cold weather on power systems, a comprehensive assessment of node risks in power systems is necessary. On the one hand, during extreme cold weather, power cost parameters are simultaneously affected by various uncertainties within the power system, from the generation side to the load side, including surges in power load and damage to power components. On the other hand, due to the time-varying nature of weather and the uneven distribution of power system resources, power cost parameters exhibit temporal and spatial patterns, necessitating the identification of critical periods and key power nodes with the highest risk. However, current research is often limited to the impact of a single uncertain factor and a single temporal and spatial characteristic, making it difficult to comprehensively assess node risks under extreme weather conditions. Summary of the Invention

[0004] To address the problems in the background technology, the present invention provides a system node risk assessment method for extreme cold weather, taking into account multiple uncertainties. This method can effectively identify the risk distribution under extreme weather conditions and provide effective guidance for the power system to establish risk prevention and control strategies under extreme weather conditions.

[0005] The technical solution adopted in the present invention is:

[0006] The system node risk assessment method of the present invention comprises the following steps:

[0007] 1) Establish a power system, which includes several power nodes, transmission lines, wind turbines, and natural gas units. Each wind turbine and natural gas unit is located at its own power node. Establish a system standard probability model of the power system. Input the load value of each power node of the power generation system, the status of each transmission line, the output of each wind turbine, and the fuel cost parameter of each natural gas unit into the system standard probability model. The system standard probability model outputs the power cost parameter of each power node of the power system.

[0008] 2) Based on the power cost parameters of each power node in the power system, a spatiotemporal risk index is obtained for each power node. This risk index is then used to assess each power node's risk during extreme cold weather. This allows for the identification of nodes with high loads, prompting early warnings and the implementation of appropriate measures. The cost is specifically a quantity related to power generation.

[0009] In step 1), the system standard probability model of the power system established is specifically as follows:

[0010]

[0011] Among them, Ω φMPC represents the output factor of the power system linear optimization model considering system constraints, z represents the UGF parameter based on the universal generating function method UGF; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., ie, ..., nth ED Standard probability model of electric load of power nodes, n ED Indicates the total number of power nodes; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., io, ..., nth order of the power system in period t, respectively. OL Standard probability model of fault state of overhead transmission lines, n OL Indicates the total number of overhead transmission lines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., iw, ..., nth ED The output standard probability model of wind turbines, n ED Indicates the total number of wind turbines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., in, ..., nth order of the power system in period t, respectively. NGP Standard probability model of fuel cost parameters for natural gas units, n NGP Indicates the total number of natural gas units; K t represents the total number of states of the power system, ρ ie,h (t) represents the power cost parameter of the jth power node in the power system at time period t under the hth state of the power system, p h (t) represents the probability of the power system being in the hth state during time period t. The transmission lines in the power system also include non-overhead transmission lines, that is, underground transmission lines. Underground transmission lines are not considered because they are not covered with ice.

[0012] The power system linear optimization model considering system constraints is as follows:

[0013]

[0014]

[0015] Among them, f h represents the sum of the power generation cost and load reduction cost of the power system in the hth state; the power system also includes coal-fired units, coal-fired units, wind turbines and natural gas units are all generators, n g Indicates the number of generator sets in the power system; GC r,h () represents the power generation cost of the ath generator unit in the power system under the hth state; CC ie,h () represents the load reduction cost of the ieth power node in the power system under the hth state; represents the active power output of the ath generator unit in the power system in the hth state and time period t; L represents the load reduction of the ieth power node in the power system in the hth state and time period t; s represents the Lagrangian function operator; It represents the load of the ieth power node in the power system in the hth state and time period t.

[0016] The power generation cost of natural gas units is proportional to the fuel cost parameters, as follows:

[0017]

[0018] Among them, GC in,h () represents the power generation cost of the inth natural gas unit in the power system under the hth state in extreme cold weather; GC NR,in () represents the power generation cost of the in-th natural gas unit in the power system during normal winter days; represents the active power output of the inth natural gas unit in the power system in the hth state and time period t; k represents the conversion coefficient between the natural gas unit cost and fuel cost parameters.

[0019] The power constraints are as follows:

[0020] a) Power balance constraints:

[0021]

[0022] Among them, B h (t) represents the node admittance matrix of the power system in the hth state of the power system and the time period t; θ h(t) represents the node voltage phase angle vector of the power system in the hth state of the power system and the time period t; represents the set of power generation of the power system in time period t under the h-th state of the power system; It represents the load set of the power system in the hth state of the power system and the time period t.

[0023] b) Load reduction constraints:

[0024]

[0025] in, It represents the load Gaussian distribution curve of the ieth power node in the power system in the leth state and time period t.

[0026] c) Wind power output constraints:

[0027]

[0028] in, represents the active power output of the iw-th wind turbine in the power system in the h-th state and time period t; It represents the load Gaussian distribution curve of the iw-th wind turbine in the power system in the lw-th state and time period t.

[0029] d) Natural gas unit output constraints:

[0030]

[0031] in, and They represent the active power output and maximum active power output of the in-th natural gas unit in the power system in the h-th state and time period t respectively.

[0032] e) Output constraints of coal-fired units:

[0033]

[0034] in, and They represent the active power output and maximum active power output of the icth coal-fired unit in the power system in the hth state and time period t respectively.

[0035] f) Line flow constraints:

[0036]

[0037] Among them, x iek,hrepresents the impedance of the transmission line between the ieth power node and the kth power node in the power system in the hth state, θ ie,h (t) and θ k,h (t) denotes the admittance of the ieth power node and the kth power node in the power system in the hth state and the time period t, respectively; represents the maximum flow of the transmission line between the ieth power node and the kth power node in the power system; 1·() represents a true-false function, which outputs 1 when the true-false function 1×() is true and outputs 0 when it is false; It represents the state quantity of the overhead transmission line between the ieth power node and the kth power node in the power system in the hth state and the time period t.

[0038] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the electric load of the ieth power node in the power system during period t The details are as follows:

[0039]

[0040]

[0041] ln(ED ie (t))=ln(ED NR (t))-β ED (τ(t)-τ NR (t))

[0042] Among them, K ED Indicates the total state number of the electric load of the ieth power node in the power system; ED ie (t) and ED NR (t) represents the temperature τ(t) in extreme cold weather and the preset reference temperature τ in normal winter days, respectively. NR (t) the load of the ieth power node in the power system in the next period t; η ED represents the load forecast error coefficient, N() represents Gaussian distribution; represents the load of the ieth power node in the power system based on Gaussian distribution during period t; and They represent the state load and state probability of the ieth power node in the power system at the leth state of the ieth power node in the power system during period t; β ED Represents the load linear regression coefficient.

[0043] The load ED of the ieth power node in the power system during period t under the temperature τ(t) of extreme cold weather is ie(t) After Gaussian distribution processing, the load Gaussian distribution curve is obtained. The load Gaussian distribution curve is normalized and discretized and divided into seven load Gaussian distribution areas. The median load value in each load Gaussian distribution area is used as the load of the ieth power node in the power system based on Gaussian distribution, that is, the load of the ieth power node in the power system based on Gaussian distribution in period t In each state of the ieth power node in the power system, the ieth power node selects the median load value in one of the load Gaussian distribution regions as its state load. The state probability that the ieth power node in the power system selects the median load value in one of the load Gaussian distribution regions as its state load is equal to the area of ​​the selected load Gaussian distribution region. Standard probability models are generated based on the universal generating function method (UGF).

[0044] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fault state of the ioth transmission line in the power system during period t The details are as follows:

[0045]

[0046]

[0047]

[0048]

[0049] in, and Represent time period t, time period t-1 and time period tT respectively R -1The probability that the ioth overhead transmission line in the power system is in a fault state, T R Indicates the preset fault line repair time; and They represent the loth state of the ioth overhead transmission line in the power system, the state of the ioth overhead transmission line in the power system in period t, and the state probability. When the state of the ioth overhead transmission line in the power system is in operation state, When the ioth overhead transmission line in the power system is in fault state, K OL Indicates the state number of the overhead transmission line in the power system; represents the failure rate of the ioth overhead transmission line in the power system; x OL (t), x OL (tT R ), xOL (t0), x OL (t q+1 ) and x OL (t q ) represent time period t and time period tT respectively R , period t0, period t q+1 and period t q The ice thickness of each overhead transmission line in the power system; d represents the preset ice-resistant diameter of each overhead transmission line in the power system; ρ0 represents the density of ice, α1, α2, and a3(t) represent the collision efficiency, viscosity coefficient, and freezing coefficient at time period t, respectively; T p (t) and v(t) represent the precipitation intensity and ambient wind speed in time period t, respectively.

[0050] The UGF parameter z based on the universal generating function method UGF, the output standard probability model of the iw-th wind turbine in the power system during period t The details are as follows:

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] Among them, K WP WP represents the total output state number of the iw-th wind turbine in the power system; iw (t) and WP NR () represents the output of the iw-th wind turbine in the power system during period t under extreme cold weather and normal winter weather, respectively; η WP represents the output prediction error coefficient, N() represents Gaussian distribution; represents the output of the iw-th wind turbine in the power system during time period t based on Gaussian distribution; and They represent the state output and state probability of the iw-th wind turbine in the power system in the lw-th state of the iw-th wind turbine in the power system during period t; R WP () represents the processing loss of each wind turbine in the power system; x WP (t), x WP (t q+1 ) and x WP (t q ) represent time periods t and t q+1and period t q The ice thickness at the blade tip of each wind turbine in the power system; α WP () and β WP () represent the first output linear regression coefficient and the second output linear regression coefficient respectively; x wse Indicates the ice thickness threshold for wind turbine damage; v ref (t) represents the relative wind speed of the ambient wind speed relative to the rotating blades of each wind turbine in the power system during period t.

[0057] The output WP of the iw-th wind turbine in the power system during period t under extreme cold weather iw (t) After Gaussian distribution processing, the output Gaussian distribution curve of itself is obtained. The output Gaussian distribution curve is normalized and discretized and divided into seven output Gaussian distribution areas. The output median in each output Gaussian distribution area is used as the output of the iw-th wind turbine in the power system based on Gaussian distribution, that is, the output of the iw-th wind turbine in the power system based on Gaussian distribution in period t In each state of the iw-th wind turbine in the power system, the iw-th wind turbine in the power system selects the median output value in one of the wind power Gaussian distribution areas as its own state output, and the iw-th wind turbine in the power system selects the median output value in one of the load Gaussian distribution areas as its own state output. The state probability is equal to the area of ​​one of the selected output Gaussian distribution areas.

[0058] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fuel cost parameter of the in-th natural gas unit in the power system during period t The details are as follows:

[0059]

[0060]

[0061] ln(NGP in (t))=ln(NGP NR )+β NGP (HDD(t)-HDD NR )

[0062] HDD(t)=max(0,τ HR -τ ave (t))

[0063] Among them, K NGP Indicates the total number of states of the fuel cost parameters of the in-th natural gas unit in the power system; NGP in (t) and NGP NRrepresents the fuel cost parameter of the in-th natural gas unit in the power system during period t under extreme cold weather and normal winter weather, respectively; η NGP represents the fuel cost parameter prediction error coefficient, N() represents Gaussian distribution; represents the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution during period t; and They represent the fuel cost parameter and state probability of the in-th natural gas unit in the power system at the ln-th state of the in-th natural gas unit in the time period t; HDD(t) and HDD NR represents the heating degree day index of extreme cold weather and normal winter days respectively; β NGP Represents the linear regression coefficient of the cost parameter; τ ave (t) and τ HR They represent the daily average temperature in extreme cold weather and the preset heating critical reference temperature respectively.

[0064] The fuel cost parameter NGP of the in-th natural gas unit in the power system during period t under extreme cold weather in (t) After Gaussian distribution processing, the fuel cost parameter Gaussian distribution curve is obtained. The fuel cost parameter Gaussian distribution curve is normalized and discretized and divided into seven fuel cost parameter Gaussian distribution areas. The median of the fuel cost parameter in each fuel cost parameter Gaussian distribution area is used as the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution, that is, the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution in period t In each state of the in-th natural gas unit in the power system, the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the natural gas Gaussian distribution areas as its own state fuel cost parameter. The state probability that the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the load Gaussian distribution areas as its own state fuel cost parameter is equal to the area of ​​one of the selected fuel cost parameter Gaussian distribution areas.

[0065] In each state of the power system, power nodes, transmission lines, wind turbines and natural gas units are in one of their respective states.

[0066] In the step 2), the spatiotemporal risk index of each power node in the power system is obtained according to the power cost parameter of each power node in the power system. The spatiotemporal risk index includes a time dimension cost parameter risk index, a space dimension cost parameter risk index, a time dimension uncertainty risk index, and a space dimension uncertainty risk index, which are specifically as follows:

[0067] a) Time dimension cost parameter risk indicator:

[0068]

[0069]

[0070]

[0071]

[0072] in, represents the expected system power cost parameter of the power system in period t; p h (t) represents the probability that the power system is in the hth state in time period t; σ(t) represents the standard deviation of the system power cost parameter of the power system in time period t; ζ represents the probability of the peak power cost parameter of the power system; ρ a,h (t) represents the load-side weighted average node power cost parameter of the power system in period t, which is used to describe the power cost parameter level of the entire system in different time periods when it is in state h; Ψ represents the total state of the time dimension cost parameter risk indicator, that is, the product of the total state of the power system in different time periods; T represents the total number of time periods; ρ Cap represents the peak power cost parameter threshold; m Represents the state quantity of the time dimension cost parameter risk indicator of time period t.

[0073] The time dimension cost parameter risk index of the ieth power node in the power system includes the expected system cost parameter of the power system in period t The standard deviation of the system cost parameter σ(t) of the power system in period t and the probability of the peak cost parameter ζ of the power system, the expected system cost parameter of the power system in period t The higher the standard deviation σ(t) of the system cost parameter of the power system in period t and the probability ζ of the peak cost parameter of the power system, the more risky the ieth power node in the power system is.

[0074] b) Spatial dimension cost parameter risk indicator:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080] in, The expected average power cost parameter of the power nodes in the power system, represents the expected average power cost parameter of the power nodes in the power system during period t; σ ie The standard deviation of the expected average power cost parameter of the power nodes in the power system, σ ie (t) represents the standard deviation of the expected average power cost parameter of the power nodes in the power system during period t; ie represents the average peak power cost parameter probability of the power nodes of the power system;

[0081] The spatial dimension cost parameter risk index of the ieth power node in the power system includes the expected average power cost parameter of the power node in the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie , the expected average power cost parameter of the power node of the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie The higher it is, the more risky the ieth power node in the power system is.

[0082] c) Time dimension uncertainty risk indicators:

[0083]

[0084]

[0085] Where S represents a set of multiple uncertain factors, including power load, overhead transmission line failure rate, wind turbine output loss, and natural gas fuel cost parameters, and s represents an uncertain factor in the set of multiple uncertain factors S; represents the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s; It represents the total state of the power system without considering a certain uncertainty factor s; represents the power cost parameter of the jth power node in the power system in the hth state and time period t without considering a certain uncertainty factor s. represents the probability of the power system being in the hth state in time period t without considering a certain uncertainty factor s; Δρ s (t) represents the expected system power cost parameter change of the power system in time period t considering a certain uncertainty factor s, that is, relative to the original scenario.

[0086] The time dimension uncertainty risk index obtains the influence of uncertain factors on the expected system power cost parameters of the power system from the time dimension based on sensitivity analysis.

[0087] The time dimension uncertainty risk index of the ieth power node in the power system includes the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s (t), the expected system power cost parameter of the power system in period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s The higher (t) is, the more risky the ieth power node in the power system is.

[0088] d) Spatial dimension uncertainty risk indicators:

[0089]

[0090]

[0091] in, represents the expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s, It represents the change in the average power cost parameter of the power nodes of the expected system in the normalized time period t of the power system considering a certain uncertainty factor s, that is, relative to the original scenario.

[0092] The spatial dimension uncertainty risk index is used to obtain the influence of uncertain factors on the expected average power cost parameters of the power nodes of the power system based on sensitivity analysis in the spatial dimension.

[0093] The spatial dimension uncertainty risk index of the ieth power node in the power system is the expected average power cost parameter of the power node in the power system without considering a certain uncertainty factor s. The expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s The higher it is, the more risky the ieth power node in the power system is.

[0094] The beneficial effects of the present invention are:

[0095] The proposed method comprehensively considers the fusion of multiple uncertainties under extreme weather conditions and the spatiotemporal distribution characteristics of power cost parameters, effectively identifying the risk distribution of power cost parameters under extreme weather conditions. This method not only helps power systems establish risk prevention and control strategies for extreme weather conditions, but also provides effective guidance for establishing highly resilient power grid planning under extreme weather conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 It is the modified IEEE-39 node power system topology diagram;

[0097] Figure 2 Parameter diagram of expected average node power cost under different extreme low temperatures;

[0098] Figure 3 Parameter diagram of expected system power cost under different extreme low temperatures;

[0099] Figure 4 (a) is a comparison diagram of the influence of system power cost parameters;

[0100] Figure 4 (b) is a comparison diagram of the influence of average node power cost parameters;

[0101] Figure 5 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0102] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0103] The system node risk assessment method of the present invention comprises the following steps:

[0104] 1) Establish a power system, which includes several power nodes, transmission lines, wind turbines, and natural gas units. Each wind turbine and natural gas unit is located at its own power node. Establish a system standard probability model of the power system. Input the load value of each power node of the power generation system, the status of each transmission line, the output of each wind turbine, and the fuel cost parameter of each natural gas unit into the system standard probability model. The system standard probability model outputs the power cost parameter of each power node of the power system.

[0105] In step 1), the system standard probability model of the power system established is as follows:

[0106]

[0107] Among them, Ω fMPC represents the output factor of the power system linear optimization model considering system constraints, z represents the UGF parameter based on the universal generating function method UGF; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., ie, ..., nth ED Standard probability model of electric load of power nodes, n ED Indicates the total number of power nodes; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., io, ..., nth order of the power system in period t, respectively. OL Standard probability model of fault state of overhead transmission lines, n OL Indicates the total number of overhead transmission lines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., iw, ..., nth ED The output standard probability model of wind turbines, n ED Indicates the total number of wind turbines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., in, ..., nth order of the power system in period t, respectively. NGP Standard probability model of fuel cost parameters for natural gas units, n NGP Indicates the total number of natural gas units; K t represents the total number of states of the power system, ρ ie,h (t) represents the power cost parameter of the jth power node in the power system at time period t under the hth state of the power system, p h (t) represents the probability of the power system being in the hth state during time period t. The transmission lines in the power system also include non-overhead transmission lines, that is, underground transmission lines. Underground transmission lines are not considered because they are not covered with ice.

[0108] The linear optimization model of the power system considering system constraints is as follows:

[0109]

[0110]

[0111] Among them, f h represents the sum of the power generation cost and load reduction cost of the power system in the hth state; the power system also includes coal-fired units, coal-fired units, wind turbines and natural gas units are all generators, n g Indicates the number of generator sets in the power system; GC r,h () represents the power generation cost of the ath generator unit in the power system under the hth state; CC ie,h () represents the load reduction cost of the ieth power node in the power system under the hth state; represents the active power output of the ath generator unit in the power system in the hth state and time period t; L represents the load reduction of the ieth power node in the power system in the hth state and time period t; s represents the Lagrangian function operator; It represents the load of the ieth power node in the power system in the hth state and time period t.

[0112] The power generation cost of natural gas units is proportional to the fuel cost parameters, as follows:

[0113]

[0114] Among them, GC in,h () represents the power generation cost of the inth natural gas unit in the power system under the hth state in extreme cold weather; GC NR,in () represents the power generation cost of the in-th natural gas unit in the power system during normal winter days; represents the active power output of the inth natural gas unit in the power system in the hth state and time period t; k represents the conversion coefficient between the natural gas unit cost and fuel cost parameters.

[0115] The power constraints are as follows:

[0116] a) Power balance constraints:

[0117]

[0118] Among them, B h (t) represents the node admittance matrix of the power system in the hth state of the power system and the time period t; θ h (t) represents the node voltage phase angle vector of the power system in the hth state of the power system and the time period t; represents the set of power generation of the power system in time period t under the h-th state of the power system; It represents the load set of the power system in the hth state of the power system and the time period t.

[0119] b) Load reduction constraints:

[0120]

[0121] in, It represents the load Gaussian distribution curve of the ieth power node in the power system in the leth state and time period t.

[0122] c) Wind power output constraints:

[0123]

[0124] in, represents the active power output of the iw-th wind turbine in the power system in the h-th state and time period t; It represents the load Gaussian distribution curve of the iw-th wind turbine in the power system in the lw-th state and time period t.

[0125] d) Natural gas unit output constraints:

[0126]

[0127] in, and They represent the active power output and maximum active power output of the in-th natural gas unit in the power system in the h-th state and time period t respectively.

[0128] e) Output constraints of coal-fired units:

[0129]

[0130] in, and They represent the active power output and maximum active power output of the icth coal-fired unit in the power system in the hth state and time period t respectively.

[0131] f) Line flow constraints:

[0132]

[0133] Among them, x iek,h represents the impedance of the transmission line between the ieth power node and the kth power node in the power system in the hth state, θ ie,h (t) and θ k,h (t) denotes the admittance of the ieth power node and the kth power node in the power system in the hth state and the time period t, respectively; represents the maximum flow of the transmission line between the ieth power node and the kth power node in the power system; 1·() represents a true-false function, which outputs 1 when the true-false function 1·() is true and outputs 0 when it is false; It represents the state quantity of the overhead transmission line between the ieth power node and the kth power node in the power system in the hth state and the time period t.

[0134] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the electric load of the ieth power node in the power system during period t The details are as follows:

[0135]

[0136]

[0137] ln(ED ie (t))=ln(ED NR (t))-β ED (τ(t)-τ NR (t))

[0138] Among them, K ED Indicates the total state number of the electric load of the ieth power node in the power system; ED ie (t) and ED NR (t) represents the temperature τ(t) in extreme cold weather and the preset reference temperature τ in normal winter days, respectively. NR (t) the load of the ieth power node in the power system in the next period t; η ED represents the load forecast error coefficient, N() represents Gaussian distribution; represents the load of the ieth power node in the power system based on Gaussian distribution during period t; and They represent the state load and state probability of the ieth power node in the power system at the leth state of the ieth power node in the power system during period t; β ED Represents the load linear regression coefficient.

[0139] The load ED of the ieth power node in the power system during period t under the temperature τ(t) of extreme cold weather is ie (t) After Gaussian distribution processing, the load Gaussian distribution curve is obtained. The load Gaussian distribution curve is normalized and discretized and divided into seven load Gaussian distribution areas. The median load value in each load Gaussian distribution area is used as the load of the ieth power node in the power system based on Gaussian distribution, that is, the load of the ieth power node in the power system based on Gaussian distribution in period t In each state of the ieth power node in the power system, the ieth power node in the power system selects the median load value in one of the load Gaussian distribution areas as its own state load. The state probability that the ieth power node in the power system selects the median load value in one of the load Gaussian distribution areas as its own state load is equal to the area of ​​one of the selected load Gaussian distribution areas.

[0140] Standard probability models are all generated based on the universal generating function method UGF.

[0141] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fault state of the ioth transmission line in the power system during period t The details are as follows:

[0142]

[0143]

[0144]

[0145]

[0146] in, and Represent time period t, time period t-1 and time period tT respectively R -1The probability that the ioth overhead transmission line in the power system is in a fault state, T R Indicates the preset fault line repair time; and They represent the loth state of the ioth overhead transmission line in the power system, the state of the ioth overhead transmission line in the power system in period t, and the state probability. When the state of the ioth overhead transmission line in the power system is in operation state, When the ioth overhead transmission line in the power system is in fault state, K OL Indicates the state number of the overhead transmission line in the power system; represents the failure rate of the ioth overhead transmission line in the power system; x OL (t), x OL (tT R ), x OL (t0), x OL (t q+1 ) and x OL (t q ) represent time period t and time period tT respectively R , period t0, period t q+1 and period t q The ice thickness of each overhead transmission line in the power system; d represents the preset ice-resistant diameter of each overhead transmission line in the power system; ρ0 represents the density of ice, α1, α2, and α3(t) represent the collision efficiency, viscosity coefficient, and freezing coefficient at time period t, respectively; T p (t) and v(t) represent the precipitation intensity and ambient wind speed in time period t, respectively.

[0147] UGF parameter z based on the universal generating function method UGF, output standard probability model of the iw-th wind turbine in the power system during period t The details are as follows:

[0148]

[0149]

[0150] WP iw (t) = WP NR (v(t))·R WP (x WP (t))

[0151]

[0152]

[0153] Among them, K WP WP represents the total output state number of the iw-th wind turbine in the power system; iw (t) and WP NR () represents the output of the iw-th wind turbine in the power system during period t under extreme cold weather and normal winter weather, respectively; η WP represents the output prediction error coefficient, N() represents Gaussian distribution; represents the output of the iw-th wind turbine in the power system during time period t based on Gaussian distribution; and They represent the state output and state probability of the iw-th wind turbine in the power system in the lw-th state of the iw-th wind turbine in the power system during period t; R WP () represents the processing loss of each wind turbine in the power system; x WP (t), x WP (t q+1 ) and x WP (t q ) represent time periods t and t q+1 and period t q The ice thickness at the blade tip of each wind turbine in the power system; α WP () and β WP () represent the first output linear regression coefficient and the second output linear regression coefficient respectively; x wse Indicates the ice thickness threshold for wind turbine damage; v ref (t) represents the relative wind speed of the ambient wind speed relative to the rotating blades of each wind turbine in the power system during period t.

[0154] The output WP of the iw-th wind turbine in the power system during period t under extreme cold weatheriw (t) After Gaussian distribution processing, the output Gaussian distribution curve of itself is obtained. The output Gaussian distribution curve is normalized and discretized and divided into seven output Gaussian distribution areas. The output median in each output Gaussian distribution area is used as the output of the iw-th wind turbine in the power system based on Gaussian distribution, that is, the output of the iw-th wind turbine in the power system based on Gaussian distribution in period t In each state of the iw-th wind turbine in the power system, the iw-th wind turbine in the power system selects the median output value in one of the wind power Gaussian distribution areas as its own state output, and the iw-th wind turbine in the power system selects the median output value in one of the load Gaussian distribution areas as its own state output. The state probability is equal to the area of ​​one of the selected output Gaussian distribution areas.

[0155] The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fuel cost parameter of the in-th natural gas unit in the power system during period t The details are as follows:

[0156]

[0157]

[0158] ln(NGP in (t))=ln(NGP NR )+β NGP (HDD(t)-HDD NR )

[0159] HDD(t)=max(0,τ HR -τ ave (t))

[0160] Among them, K NGP Indicates the total number of states of the fuel cost parameters of the in-th natural gas unit in the power system; NGP in (t) and NGP NR represents the fuel cost parameter of the in-th natural gas unit in the power system during period t under extreme cold weather and normal winter weather, respectively; η NGP represents the fuel cost parameter prediction error coefficient, N() represents Gaussian distribution; represents the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution during period t; and They represent the fuel cost parameter and state probability of the in-th natural gas unit in the power system at the ln-th state of the in-th natural gas unit in the time period t; HDD(t) and HDD NRrepresents the heating degree day index of extreme cold weather and normal winter days respectively; β NGP Represents the linear regression coefficient of the cost parameter; τ ave (t) and τ HR They represent the daily average temperature in extreme cold weather and the preset heating critical reference temperature respectively.

[0161] The fuel cost parameter NGP of the in-th natural gas unit in the power system during period t under extreme cold weather in (t) After Gaussian distribution processing, the fuel cost parameter Gaussian distribution curve is obtained. The fuel cost parameter Gaussian distribution curve is normalized and discretized and divided into seven fuel cost parameter Gaussian distribution areas. The median of the fuel cost parameter in each fuel cost parameter Gaussian distribution area is used as the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution, that is, the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution in period t In each state of the in-th natural gas unit in the power system, the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the natural gas Gaussian distribution areas as its own state fuel cost parameter. The state probability that the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the load Gaussian distribution areas as its own state fuel cost parameter is equal to the area of ​​one of the selected fuel cost parameter Gaussian distribution areas.

[0162] In each state of the power system, power nodes, transmission lines, wind turbines and natural gas units are in one of their respective states.

[0163] 2) Based on the power cost parameters of each power node in the power system, a spatiotemporal risk index is obtained for each power node. This risk index is then used to assess each power node's risk during extreme cold weather. This allows for the identification of nodes with high loads, prompting early warnings and the implementation of appropriate measures. The cost is specifically a quantity related to power generation.

[0164] In step 2), the spatiotemporal risk index of each power node in the power system is obtained according to the power cost parameter of each power node in the power system. The spatiotemporal risk index includes the time dimension cost parameter risk index, the space dimension cost parameter risk index, the time dimension uncertainty risk index, and the space dimension uncertainty risk index, which are as follows:

[0165] a) Time dimension cost parameter risk indicator:

[0166]

[0167]

[0168]

[0169]

[0170] in, represents the expected system power cost parameter of the power system in period t; p h (t) represents the probability that the power system is in the hth state in time period t; σ(t) represents the standard deviation of the system power cost parameter of the power system in time period t; ζ represents the probability of the peak power cost parameter of the power system; ρ a,h (t) represents the load-side weighted average node power cost parameter of the power system in period t, which is used to describe the power cost parameter level of the entire system in different time periods when it is in state h; Ψ represents the total state of the time dimension cost parameter risk indicator, that is, the product of the total state of the power system in different time periods; T represents the total number of time periods; ρ Cap represents the peak power cost parameter threshold; m Represents the state quantity of the time dimension cost parameter risk indicator of time period t.

[0171] The time dimension cost parameter risk index of the ieth power node in the power system includes the expected system cost parameter of the power system in period t The standard deviation of the system cost parameter σ(t) of the power system in period t and the probability of the peak cost parameter ζ of the power system, the expected system cost parameter of the power system in period t The higher the standard deviation σ(t) of the system cost parameter of the power system in period t and the probability ζ of the peak cost parameter of the power system, the more risky the ieth power node in the power system is.

[0172] b) Spatial dimension cost parameter risk indicator:

[0173]

[0174]

[0175]

[0176]

[0177]

[0178] in, The expected average power cost parameter of the power nodes in the power system, represents the expected average power cost parameter of the power nodes in the power system during period t; σ ie The standard deviation of the expected average power cost parameter of the power nodes in the power system, σie (t) represents the standard deviation of the expected average power cost parameter of the power nodes in the power system during period t; ie Represents the average peak power cost parameter probability of the power nodes in the power system.

[0179] The spatial dimension cost parameter risk index of the ieth power node in the power system includes the expected average power cost parameter of the power node in the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie , the expected average power cost parameter of the power node of the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie The higher it is, the more risky the ieth power node in the power system is.

[0180] c) Time dimension uncertainty risk indicators:

[0181]

[0182]

[0183] Where S represents a set of multiple uncertain factors, including power load, overhead transmission line failure rate, wind turbine output loss, and natural gas fuel cost parameters, and s represents an uncertain factor in the set of multiple uncertain factors S; represents the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s; It represents the total state of the power system without considering a certain uncertainty factor s; represents the power cost parameter of the jth power node in the power system in the hth state and time period t without considering a certain uncertainty factor s. represents the probability of the power system being in the hth state in time period t without considering a certain uncertainty factor s; Δρ s (t) represents the expected system power cost parameter change of the power system in time period t considering a certain uncertainty factor s, that is, relative to the original scenario.

[0184] The time dimension uncertainty risk index obtains the influence of uncertain factors on the expected system power cost parameters of the power system from the time dimension based on sensitivity analysis.

[0185] The time dimension uncertainty risk index of the ieth power node in the power system includes the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s (t), the expected system power cost parameter of the power system in period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s The higher (t) is, the more risky the ieth power node in the power system is.

[0186] d) Spatial dimension uncertainty risk indicators:

[0187]

[0188]

[0189] in, represents the expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s, It represents the change in the average power cost parameter of the power nodes of the expected system in the normalized time period t of the power system considering a certain uncertainty factor s, that is, relative to the original scenario.

[0190] The spatial dimension uncertainty risk index is used to obtain the influence of uncertain factors on the expected average power cost parameters of the power nodes of the power system based on sensitivity analysis in the spatial dimension.

[0191] The spatial dimension uncertainty risk index of the ieth power node in the power system is the expected average power cost parameter of the power node in the power system without considering a certain uncertainty factor s. The expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s The higher it is, the more risky the ieth power node in the power system is.

[0192] The embodiments of the present invention are as follows:

[0193] 1. A modified IEEE-39 node system is used for verification analysis. Natural gas units and coal-fired units are distributed at nodes 30-38 respectively, and large wind farms are distributed at nodes 2, 6, 23 and 28. The power system topology is as follows: Figure 1Assume that the average temperature in the region during normal winter days is 5°C, which is the preset temperature parameter. When an extreme cold wave arrives, low temperatures of -4°C, -6°C, and -10°C are selected for study. Assume a wind speed of 13 m / s and a precipitation intensity of 5 mm / h. Based on these meteorological parameters, the UGF method is used to establish a standard probabilistic model of the power system.

[0194] 2. Based on the clearing power cost parameter operator, the standard probability model of the power system is obtained, and the expected average node power cost parameter risk index in the spatial dimension is calculated, that is, the power cost parameter risk index in the spatial dimension is as follows: Figure 2 As shown in the figure, the expected system power cost parameter risk index in the time dimension, that is, the power cost parameter risk index in the time dimension is as follows: Figure 3 As shown in Table 1. In terms of spatial dimension, Figure 2 The results show that as the temperature decreases, the expected average node power cost parameter values ​​and the peak average node power cost parameter values ​​increase significantly. Nodes 3, 4, 15, and 18 are the nodes with the highest power cost parameter risk. These nodes carry high power loads and are located far from key generation resources. From a temporal perspective, as the temperature decreases, the expected system power cost parameter values ​​and the probability of system peak power cost parameters also increase. Time periods with the highest power cost parameter risk often experience high power loads and severe icing. This indicates that critical nodes and time periods with high risk in extreme cold weather are effectively identified, demonstrating the effectiveness of this method.

[0195] 3. Based on the sensitivity analysis method, the extreme low temperature of -10℃ is selected as an example to compare and identify the influence of the internal factors of the power system on the power cost parameters. Figure 4 As shown in the figure, it can be seen that the surge in power load during extreme cold weather is the factor that has the greatest impact on the power cost parameter. Secondly, the overhead line failure rate has a significant impact on the load node. The impact of the natural gas cost parameter surge and wind power output loss on the power cost parameter is relatively small in this example.

Claims

1. A system node risk assessment method considering multiple uncertainties in extreme cold weather, characterized by: The method comprises the following steps: 1) Establishing a power system, which includes several power nodes, transmission lines, wind turbines, and natural gas generators. Each wind turbine and natural gas generator is located at its own power node. Establishing a standard probability model of the power system. Inputting the load value of each power node of the power generation system, the status of each transmission line, the output of each wind turbine, and the fuel cost parameter of each natural gas generator into the standard probability model, the standard probability model outputs the power cost parameter of each power node of the power system. 2) Obtaining a spatiotemporal risk index for each power node in the power system based on the power cost parameter of each power node in the power system, and conducting a risk assessment of each power node in extreme cold weather based on the spatiotemporal risk index of each power node; In step 1), the system standard probability model of the power system established is specifically as follows: Among them, Ω φMPC represents the output factor of the power system linear optimization model considering system constraints, z represents the UGF parameter based on the universal generating function method UGF; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., ie, ..., nth ED Standard probability model of electric load of power nodes, n ED Indicates the total number of power nodes; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., io, ..., nth order of the power system in period t, respectively. OL Standard probability model of fault state of overhead transmission lines, n OL Indicates the total number of overhead transmission lines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., iw, ..., nth ED The output standard probability model of wind turbines, n ED Indicates the total number of wind turbines; They represent the UGF parameter z based on the universal generating function method UGF, the 1st, 2nd, ..., in, ..., nth order of the power system in period t, respectively. NGP Standard probability model of fuel cost parameters for natural gas units, n NGP Indicates the total number of natural gas units; K t represents the total number of states of the power system, ρ ie,h (t) represents the power cost parameter of the jth power node in the power system at time period t under the hth state of the power system, p h (t) represents the probability of the power system being in the hth state during time period t; The power system linear optimization model considering system constraints is as follows: Among them, f h represents the sum of the power generation cost and load reduction cost of the power system in the hth state; the power system also includes coal-fired units, coal-fired units, wind turbines and natural gas units are all generators, n g Indicates the number of generator sets in the power system; GC r,h ( ) represents the power generation cost of the ath generator unit in the power system under the hth state; CC ie,h ( ) represents the load reduction cost of the ieth power node in the power system under the hth state; represents the active power output of the ath generator unit in the power system in the hth state and time period t; L represents the load reduction of the ieth power node in the power system in the hth state and time period t; s represents the Lagrangian function operator; It represents the load of the ieth power node in the power system in the hth state and time period t.

2. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: The system constraints are as follows: a) Power balance constraints: Among them, B h (t) represents the node admittance matrix of the power system in the hth state of the power system and the time period t; θ h (t) represents the node voltage phase angle vector of the power system in the hth state of the power system and the time period t; represents the set of power generation of the power system in time period t under the h-th state of the power system; represents the load set of the power system in the hth state of the power system and the time period t; b) Load reduction constraints: in, represents the load Gaussian distribution curve of the ieth power node in the power system in the leth state and time period t; c) Wind power output constraints: in, represents the active power output of the iw-th wind turbine in the power system in the h-th state and time period t; represents the load Gaussian distribution curve of the iw-th wind turbine in the power system in the lw-th state and time period t; d) Natural gas unit output constraints: in, and They represent the active power output and maximum active power output of the in-th natural gas unit in the power system in the h-th state and time period t respectively; e) Output constraints of coal-fired units: in, and They represent the active power output and maximum active power output of the icth coal-fired unit in the power system in the hth state and time period t respectively; f) Line flow constraints: Among them, x iek,h represents the impedance of the transmission line between the ieth power node and the kth power node in the power system in the hth state, θ ie,h (t) and θ k,h (t) denotes the admittance of the ieth power node and the kth power node in the power system in the hth state and the time period t, respectively; represents the maximum flow of the transmission line between the ie-th power node and the k-th power node in the power system; 1·( ) represents a true-false function, which outputs 1 when the true-false function 1·( ) is true and outputs 0 when it is false; It represents the state quantity of the overhead transmission line between the ieth power node and the kth power node in the power system in the hth state and the time period t.

3. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: The UGF parameter z based on the universal generating function method UGF, the standard probability model of the electric load of the ieth power node in the power system during period t The details are as follows: ln(ED ie (t))=ln(ED NR (t))-β ED (τ(t)-τ NR (t)) Among them, K ED Indicates the total state number of the electric load of the ieth power node in the power system; ED ie (t) and ED NR (t) represent the temperature τ(t) in extreme cold weather and the preset reference temperature τ in winter days, respectively. NR (t) the load of the ieth power node in the power system in the next period t; η ED represents the load forecast error coefficient, N() represents Gaussian distribution; represents the load of the ieth power node in the power system based on Gaussian distribution during period t; and They represent the state load and state probability of the ieth power node in the power system at the leth state of the ieth power node in the power system during period t; β ED represents the load linear regression coefficient; The load ED of the ieth power node in the power system during period t under the temperature τ(t) of extreme cold weather is ie (t) After Gaussian distribution processing, the load Gaussian distribution curve is obtained. The load Gaussian distribution curve is normalized and discretized and divided into seven load Gaussian distribution areas. The median load value in each load Gaussian distribution area is used as the load of the ieth power node in the power system based on Gaussian distribution, that is, the load of the ieth power node in the power system based on Gaussian distribution in period t In each state of the ieth power node in the power system, the ieth power node in the power system selects the median load value in one of the load Gaussian distribution areas as its own state load. The state probability that the ieth power node in the power system selects the median load value in one of the load Gaussian distribution areas as its own state load is equal to the area of ​​one of the selected load Gaussian distribution areas.

4. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fault state of the ioth transmission line in the power system during period t The details are as follows: in, and Represent time period t, time period t-1 and time period tT respectively R -1The probability that the ioth overhead transmission line in the power system is in a fault state, T R Indicates the preset fault line repair time; and They represent the loth state of the ioth overhead transmission line in the power system, the state of the ioth overhead transmission line in the power system in period t, and the state probability. When the state of the ioth overhead transmission line in the power system is in operation state, When the ioth overhead transmission line in the power system is in fault state, K OL Indicates the state number of the overhead transmission line in the power system; represents the failure rate of the ioth overhead transmission line in the power system; x OL (t), x OL (tT R ), x OL (t0), x OL (t q+1 ) and x OL (t q ) represent time period t and time period tT respectively R , period t0, period t q+1 and period t q The ice thickness of each overhead transmission line in the power system; d represents the preset ice-resistant diameter of each overhead transmission line in the power system; ρ0 represents the density of ice, α1, α2, and α3(t) represent the collision efficiency, viscosity coefficient, and freezing coefficient at time period t, respectively; T p (t) and v(t) represent the precipitation intensity and ambient wind speed in time period t, respectively.

5. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: The UGF parameter z based on the universal generating function method UGF, the output standard probability model of the iw-th wind turbine in the power system during period t The details are as follows: WP iw (t)=WP NR (v(t))·R WP (x WP (t)) Among them, K WP WP represents the total output state number of the iw-th wind turbine in the power system; iw (t) and WP NR ( ) represents the output of the iw-th wind turbine in the power system during period t under extreme cold weather and seasonal weather, respectively; η WP represents the output prediction error coefficient, N( ) represents Gaussian distribution; represents the output of the iw-th wind turbine in the power system during time period t based on Gaussian distribution; and They represent the state output and state probability of the iw-th wind turbine in the power system in the lw-th state of the iw-th wind turbine in the power system during period t; R WP ( ) represents the processing loss of each wind turbine in the power system; x WP (t), x WP (t q+1 ) and x WP (t q ) represent time periods t and t q+1 and period t q The ice thickness at the blade tip of each wind turbine in the power system; α WP ( ) and β WP ( ) represent the first output linear regression coefficient and the second output linear regression coefficient respectively; x wse Indicates the ice thickness threshold for wind turbine damage; v ref (t) represents the relative wind speed of the ambient wind speed relative to the rotating blades of each wind turbine in the power system during period t; The output WP of the iw-th wind turbine in the power system during period t under extreme cold weather iw (t) After Gaussian distribution processing, the output Gaussian distribution curve of itself is obtained. The output Gaussian distribution curve is normalized and discretized and divided into seven output Gaussian distribution areas. The output median in each output Gaussian distribution area is used as the output of the iw-th wind turbine in the power system based on Gaussian distribution, that is, the output of the iw-th wind turbine in the power system based on Gaussian distribution in period t In each state of the iw-th wind turbine in the power system, the iw-th wind turbine in the power system selects the median output value in one of the wind power Gaussian distribution areas as its own state output, and the iw-th wind turbine in the power system selects the median output value in one of the load Gaussian distribution areas as its own state output. The state probability is equal to the area of ​​one of the selected output Gaussian distribution areas.

6. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: The UGF parameter z based on the universal generating function method UGF, the standard probability model of the fuel cost parameter of the in-th natural gas unit in the power system during period t The details are as follows: ln(NGP in (t))=ln(NGP NR )+β NGP (HDD(t)-HDD NR ) HDD(t)=max(0,τ HR -τ ave (t)) Among them, K NGP Indicates the total number of states of the fuel cost parameters of the in-th natural gas unit in the power system; NGP in (t) and NGP NR represents the fuel cost parameter of the in-th natural gas unit in the power system during period t under extreme cold weather and winter weather respectively; η NGP represents the fuel cost parameter prediction error coefficient, N( ) represents Gaussian distribution; represents the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution during period t; and They represent the fuel cost parameter and state probability of the in-th natural gas unit in the power system at the ln-th state of the in-th natural gas unit in the time period t; HDD(t) and HDD NR Represents the heating degree day index of extreme cold weather and winter days respectively; β NGP Represents the linear regression coefficient of the cost parameter; τ ave (t) and τ HR They represent the daily average temperature in extreme cold weather and the preset critical reference temperature for heating; The fuel cost parameter NGP of the in-th natural gas unit in the power system during period t under extreme cold weather in (t) After Gaussian distribution processing, the fuel cost parameter Gaussian distribution curve is obtained. The fuel cost parameter Gaussian distribution curve is normalized and discretized and divided into seven fuel cost parameter Gaussian distribution areas. The median of the fuel cost parameter in each fuel cost parameter Gaussian distribution area is used as the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution, that is, the fuel cost parameter of the in-th natural gas unit in the power system based on Gaussian distribution in period t In each state of the in-th natural gas unit in the power system, the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the natural gas Gaussian distribution areas as its own state fuel cost parameter. The state probability that the in-th natural gas unit in the power system selects the median of the fuel cost parameters in one of the load Gaussian distribution areas as its own state fuel cost parameter is equal to the area of ​​one of the selected fuel cost parameter Gaussian distribution areas.

7. The system node risk assessment method considering multiple uncertainties in extreme cold weather according to claim 1, characterized in that: In the step 2), the spatiotemporal risk index of each power node in the power system is obtained according to the power cost parameter of each power node in the power system. The spatiotemporal risk index includes a time dimension cost parameter risk index, a space dimension cost parameter risk index, a time dimension uncertainty risk index, and a space dimension uncertainty risk index, which are specifically as follows: a) Time dimension cost parameter risk indicator: in, represents the expected system power cost parameter of the power system in period t; p h (t) represents the probability that the power system is in the hth state in time period t; σ(t) represents the standard deviation of the system power cost parameter of the power system in time period t; ζ represents the probability of the peak power cost parameter of the power system; ρ a,h (t) represents the load-side weighted average node power cost parameter of the power system in period t; Ψ represents the total state of the time dimension cost parameter risk index, that is, the product of the total state of the power system in different periods; T represents the total number of periods; ρ Cap represents the peak power cost parameter threshold; m Represents the state quantity of the time dimension cost parameter risk indicator for period t; The time dimension cost parameter risk index of the ieth power node in the power system includes the expected system cost parameter of the power system in period t The standard deviation of the system cost parameter σ(t) of the power system in period t and the probability of the peak cost parameter ζ of the power system, the expected system cost parameter of the power system in period t The higher the standard deviation σ(t) of the system cost parameter of the power system in period t and the probability ζ of the peak cost parameter of the power system, the more risky the ieth power node in the power system is; b) Spatial dimension cost parameter risk indicator: in, The expected average power cost parameter of the power nodes in the power system, represents the expected average power cost parameter of the power nodes in the power system during period t; σ ie The standard deviation of the expected average power cost parameter of the power nodes in the power system, σ ie (t) represents the standard deviation of the expected average power cost parameter of the power nodes in the power system during period t; ie represents the average peak power cost parameter probability of the power nodes of the power system; The spatial dimension cost parameter risk index of the ieth power node in the power system includes the expected average power cost parameter of the power node in the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie , the expected average power cost parameter of the power node of the power system The standard deviation of the expected average power cost parameter σ of the power nodes in the power system ie and the average peak power cost parameter probability ζ of the power nodes of the power system ie The higher it is, the more risky the ieth power node in the power system is; c) Time dimension uncertainty risk indicators: Where S represents a set of multiple uncertain factors, and s represents an uncertain factor in the set S of multiple uncertain factors; K represents the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s; s It represents the total state of the power system without considering a certain uncertainty factor s; represents the power cost parameter of the jth power node in the power system in the hth state and time period t without considering a certain uncertainty factor s. represents the probability of the power system being in the hth state in time period t without considering a certain uncertainty factor s; Δρ s (t) represents the expected system power cost parameter change of the power system in time period t considering a certain uncertainty factor s; The time dimension uncertainty risk index of the ieth power node in the power system includes the expected system power cost parameter of the power system in time period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s (t), the expected system power cost parameter of the power system in period t without considering a certain uncertainty factor s and the expected system power cost parameter change Δρ of the power system in time period t considering a certain uncertainty factor s s The higher (t) is, the more risky the ieth power node in the power system is; d) Spatial dimension uncertainty risk indicators: in, represents the expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s, represents the average power cost parameter change of the power nodes of the normalized power system in time period t considering a certain uncertainty factor s; The spatial dimension uncertainty risk index of the ieth power node in the power system is the expected average power cost parameter of the power node in the power system without considering a certain uncertainty factor s. The expected average power cost parameter of the power node of the power system without considering a certain uncertainty factor s The higher it is, the more risky the ieth power node in the power system is.

Images