Power grid operation state evaluation method and system based on combination of data driving and time sequence characteristics

Abstract

The invention discloses a power grid operation state evaluation method and system based on combination of data driving and time sequence characteristics, and the method constructs a power grid operation state multi-dimensional index evaluation system, and can carry out the multi-dimensional and comprehensive evaluation of the power grid operation state, thereby guaranteeing the stability and safety of power supply. And effective utilization of renewable energy sources can be promoted. According to the method, primary assignment based on data driving is carried out through a confidence interval where primary index data is located, and secondary correction based on time sequence characteristics is carried out on a primary assignment result according to duration of a severe scene in a primary index, so that historical information and time sequence characteristics of power grid operation are mined; accurate evaluation of the power grid operation state is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid operation status assessment, and in particular relates to a power grid operation status assessment method and system based on a combination of data drive and time series characteristics. Background Art

[0002] In the context of new power systems and new energy systems, building a power grid evaluation index system is an important means to cope with the transformation of the power market and achieve safety and economic sustainable development. Through a scientific and reasonable evaluation system, not only can the stability and safety of power supply be guaranteed, but also the effective use of renewable energy can be promoted, and low-carbon goals can be further achieved. Therefore, it is urgent to build and improve the power grid evaluation system. Power grid operation involves many factors, and there is a lack of multi-dimensional and comprehensive power grid planning evaluation indicators and standard systems. In addition, it is difficult to effectively grasp the actual operation level of the power grid simply by a large amount of data statistics. Therefore, it is necessary to propose a new evaluation system and evaluation method, which fully considers the multi-dimensional operation status of the power grid, fully explores the historical information and time series characteristics of the power grid operation, realizes the accurate evaluation of the power grid operation status, and helps to improve the power grid's power supply capacity and the level of new energy consumption. Summary of the invention

[0003] In order to solve the problems existing in the prior art, the present invention proposes a method and system for evaluating the operation status of a power grid based on a combination of data drive and time series characteristics.

[0004] The technical solution of the present invention is as follows:

[0005] A method for evaluating power grid operation status based on a combination of data-driven and time series features, comprising:

[0006] Obtaining grid operation parameter information;

[0007] Constructing a multi-dimensional indicator evaluation system for the operation status of a power grid, wherein the multi-dimensional indicator evaluation system for the operation status of a power grid has a number of first-level indicators, and each first-level indicator has a number of second-level indicators;

[0008] Based on the power grid operation parameter information, various secondary indicators are quantified to obtain various secondary indicators of the power grid hour by hour throughout the year;

[0009] Based on the various secondary indicators of the power grid every hour throughout the year, the entropy weight method is used to obtain the weights of various secondary indicators, and the various primary indicators of the power grid every hour throughout the year are obtained;

[0010] Based on the hourly primary indicators of the power grid throughout the year, the daily probability distribution of each primary indicator is statistically analyzed, and a Gaussian mixture model is used for fitting to obtain the confidence intervals of the daily probability distribution of each primary indicator at different confidence levels; for each primary indicator data of the day to be evaluated, a data-driven initial scoring is performed according to the confidence interval in which it is located;

[0011] Based on the hourly primary indicators of the power grid throughout the year, the duration of the severe scenarios in each primary indicator is counted, and the initial scoring results are corrected secondary based on the time series characteristics according to the time.

[0012] Furthermore, the grid operation parameter information includes network topology, node admittance matrix, node voltage, branch flow, load data information, and installed capacity and output information of generator sets, new energy units, and energy storage equipment.

[0013] Furthermore, the first-level indicators of the multi-dimensional indicator evaluation system for the power grid operation status include safety indicators, economic indicators, flexibility indicators and low-carbon indicators.

[0014] Furthermore, the secondary indicators of the safety index include power supply margin of the power grid, frequency safety margin, line overload rate, voltage deviation rate and main transformer load rate;

[0015] The expression of the power supply margin of the power grid is:

[0016]

[0017] Where, ψ represents the power supply margin of the power grid; P sup Indicates the power supply capacity of the partition; Indicates the maximum load of the partition;

[0018] The expression of the frequency safety margin is:

[0019]

[0020] in,

[0021]

[0022] Where Δf max represents the maximum frequency disturbance; ΔP represents the power disturbance; D represents the load damping constant; R represents the speed regulator adjustment constant; T R represents the time constant; H represents the system inertia constant; F H represents turbine parameters; ω n , α, ω r , t nadir are the first, second, third, fourth, and fifth intermediate variables, respectively;

[0023] The expression of the line overload rate is:

[0024] δ=n br / N br (A-9)

[0025] In the formula, δ represents the line overload rate, n br Indicates the number of lines with transmission power exceeding 80% of capacity, N br Indicates the total number of lines;

[0026] The expression of the voltage deviation rate is:

[0027] γ=n node / N node (A-10)

[0028] In the formula, γ represents the voltage deviation rate, n node Indicates the number of nodes whose voltage fluctuation exceeds 0.05V, N node Indicates the total number of nodes;

[0029] The expression of the main transformer load rate is:

[0030] β=P i sub / P i max (A-11)

[0031] In the formula, β represents the load factor of the main transformer, P i sub Represents the power transmitted through the main transformer i, P i max represents the transmission capacity of main transformer i;

[0032] The secondary indicators of the economic indicators include operating cost, capacity factor, line loss rate, and main transformer load imbalance; wherein the operating cost is expressed as:

[0033]

[0034] In the formula, C OPE represents the total operating cost, c fuel Represents the unit power generation cost of the generator set, P i g represents the active output of conventional unit i, and They represent the unit load shedding cost and the unit wind and solar curtailment cost respectively. and They represent the load shedding and wind and solar power abandonment at node i respectively;

[0035] The expression of the capacity factor is:

[0036]

[0037] In the formula, Φ LF represents the capacity factor, P i WT and P i PV They represent the active output of wind turbine and photovoltaic unit at node i respectively, and They represent the installed capacity of wind turbines and photovoltaic units at node i respectively;

[0038] The expression of the line loss rate is:

[0039]

[0040] Where N p Indicates line loss rate, ΔP bz and ΔP l Represent the transmission loss of transformer and line respectively, P r represents the overall transmission power;

[0041] The expression of the main transformer load imbalance is:

[0042]

[0043] In the formula, Φ t,σ Indicates the unbalance degree of the main transformer at time t, P i,t represents the transmission power of main transformer i at time t, P i,max represents the transmission capacity of the main transformer, and n represents the total number of main transformers;

[0044] The secondary indicators of the flexibility index include unit reserve, energy storage capacity for regulation and line transmission margin;

[0045] The expression of the unit standby is:

[0046]

[0047] In the formula, and They represent the upper and lower reserve capacities provided by the conventional unit v at time t, Indicates the maximum climbing power of conventional unit v, p v,t represents the active output of the conventional unit v at time t, and They represent the lower and upper limits of the output of the conventional unit v, respectively. It represents the total reserve capacity of the power grid at time t;

[0048] The expression of the energy storage that can participate in regulating capacity is:

[0049]

[0050] In the formula, and They represent the upward and downward regulation power provided by the energy storage at time t, and They represent the upper limits of energy storage charging and discharging power, and They respectively represent the charging and discharging states of the energy storage at time t, and Respectively represent the lower and upper limits of energy storage capacity, η CH and η DIS Respectively represent the charging efficiency and discharging efficiency of energy storage, P t ESS Represents the overall adjustable capacity provided by energy storage;

[0051] The expression of the line transmission margin is:

[0052]

[0053] In the formula, p v,t and p k,t They represent the active power output of conventional unit v and new energy unit k at time t, d m,t represents the load active power demand of node m at time t, P r Indicates the line transmission margin, and Represent the network topology parameters of conventional units, new energy units and loads respectively;

[0054] The secondary indicators of the low-carbon index include carbon dioxide emissions, new energy penetration rate and wind and solar power abandonment rate;

[0055] The expression of carbon dioxide emission is:

[0056]

[0057] In the formula, represents carbon dioxide emissions, E f Indicates the unit carbon emission of conventional units;

[0058] The expression of the new energy penetration rate is:

[0059]

[0060] Where η res represents the penetration rate of new energy units, represents the load demand of node i.

[0061] The expression of wind and solar power abandonment rate is:

[0062]

[0063] In the formula, and They respectively represent the actual consumption and power generation of the new energy unit at node i.

[0064] Furthermore, based on the various secondary indicators of the power grid hourly throughout the year, the entropy weight method is used to obtain the weights of various secondary indicators, and the specific steps of obtaining various primary indicators of the power grid hourly throughout the year include:

[0065] 1) Forward normalization of indicators. First, the indicators are normalized, that is, the cost indicators are converted into benefit indicators according to the indicator type, and then the indicators are dimensionless. The specific method is as follows:

[0066] For an indicator attribute value matrix F with m moments and n indicators m×n

[0067]

[0068] In the formula, f ij is the initial value of the jth indicator corresponding to the i-th moment, i∈[1,m], j∈[1,n]; that is, the following formula is used:

[0069] If the jth indicator is a profit indicator:

[0070]

[0071] Where b ij is the normalized value of the jth indicator corresponding to the i-th moment;

[0072] If the jth indicator is a cost indicator:

[0073]

[0074] 2) Calculate the index entropy value

[0075] After the above indicators are normalized, the matrix B is formed m×n

[0076]

[0077] For the jth indicator, its information entropy is E j :

[0078]

[0079] In the formula, p ijis the weight of the jth indicator corresponding to the i-th moment:

[0080]

[0081] 3) Calculate weight

[0082]

[0083] In the formula, is the weight of the jth indicator;

[0084] 4) Primary indicator assignment

[0085]

[0086] in, Indicates the assignment result of the first-level index i, Represents the indicator data of the secondary indicator j under the primary indicator i.

[0087] Furthermore, based on the various first-level indicators of the power grid hourly throughout the year, the daily probability distribution of each first-level indicator is statistically analyzed, and a Gaussian mixture model is used for fitting to obtain the confidence intervals of the daily probability distribution of each first-level indicator at different confidence levels; for each first-level indicator data of the day to be evaluated, the specific steps of performing data-driven initial scoring according to the confidence interval in which it is located include:

[0088] 1) Data Normalization

[0089]

[0090] In the formula, x represents the specific value of each primary indicator, x min and x max They represent the minimum and maximum values ​​of the corresponding indicators respectively, and y represents the normalized indicator;

[0091] 2) The daily probability distribution of each primary indicator is statistically analyzed and fitted using a Gaussian mixture model. The expression of the Gaussian mixture model is:

[0092]

[0093] In the formula, F(y) is the Gaussian mixture model, k represents a single Gaussian component, K represents the number of Gaussian components, and α k , μ k and Respectively represent the expected and variance of the weight coefficient of Gaussian component k, represents the probability density function of Gaussian component k;

[0094] 3) Correspond to the fitted daily probability distribution curves of each primary indicator, specify the confidence level and provide the corresponding score table;

[0095] 4) Obtain the confidence interval according to the specified confidence level based on the fitted daily probability distribution curves of each primary indicator;

[0096]

[0097] In the formula, ε represents the confidence level, δ up and δ dn represent the upper and lower bounds of the confidence interval, respectively.

[0098] 5) According to the confidence interval of each primary indicator data on the day to be evaluated, the value is assigned according to the interval, and the amplitude expression is:

[0099]

[0100] In the formula, S (1) Indicates the initial scoring result of the first-level indicator. Represents an assignment function.

[0101] Furthermore, based on the various first-level indicators of the power grid hourly throughout the year, the duration of the severe scenarios in each first-level indicator is counted, and the specific steps of performing a secondary correction on the initial scoring result based on the time series characteristics according to the time include:

[0102] The first-level indicators of the power grid are divided into natural days throughout the year, and the number of hours when the first-level indicators are less than or equal to the first set threshold of the corresponding first-level indicators is obtained. If the number of hours is greater than the second set threshold, the day is regarded as a severe scenario day, and the time of the severe scenario day throughout the year is recorded as the duration of the severe scenario t bad , based on the duration t of the severe scenario bad Perform secondary correction based on time series characteristics. The secondary correction formula is:

[0103] ΔS=-f(X bad ,t bad ) (A-43)

[0104] S (2) =S (1) +ΔS (A-44)

[0105] In the formula, ΔS represents the secondary correction based on the time series characteristics, X bad represents a bad scenario, f represents a correction function, S (2) Indicates the scoring result after secondary correction.

[0106] A power grid operation status evaluation system based on a combination of data drive and time series characteristics, comprising a data acquisition module, a system construction module, a secondary indicator calculation module, a primary indicator calculation module, a primary scoring module and a secondary scoring module;

[0107] The data acquisition module is used to acquire power grid operation parameter information;

[0108] The system construction module is used to construct a multi-dimensional indicator evaluation system for the operation status of a power grid. The multi-dimensional indicator evaluation system for the operation status of a power grid has several first-level indicators, and each first-level indicator has multiple second-level indicators.

[0109] The secondary index calculation module is used to quantify various secondary indicators based on the power grid operation parameter information to obtain various secondary indicators of the power grid hour by hour throughout the year;

[0110] The primary index calculation module is used to obtain the weights of various secondary indicators based on the secondary indicators of the power grid hourly throughout the year by using the entropy weight method to obtain various primary indicators of the power grid hourly throughout the year;

[0111] The initial scoring module is used to calculate the daily probability distribution of each primary indicator based on the hourly primary indicators of the power grid throughout the year, and use a Gaussian mixture model for fitting to obtain the confidence intervals of the daily probability distribution of each primary indicator at different confidence levels; for each primary indicator data of the day to be evaluated, perform data-driven initial scoring according to the confidence interval in which it is located;

[0112] The secondary scoring module is used to count the duration of the severe scenarios in each primary indicator based on the primary indicators of the power grid hour by hour throughout the year, and perform secondary corrections to the initial scoring results based on the time series characteristics according to the time.

[0113] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute any of the methods described above.

[0114] A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0115] Compared with the prior art, the present invention has the following beneficial effects:

[0116] The present invention proposes a method and system for evaluating the operation status of a power grid based on a combination of data-driven and time series features. The method constructs a multi-dimensional indicator evaluation system for the operation status of a power grid, which can perform a multi-dimensional and comprehensive evaluation of the operation status of the power grid, thereby ensuring the stability and safety of the power supply and promoting the effective use of renewable energy.

[0117] The method of the present invention performs a data-driven initial scoring through the confidence interval of the primary indicator data, and then performs a secondary correction on the initial scoring result based on the time series characteristics according to the duration of the severe scenario in the primary indicator, thereby mining the historical information and time series characteristics of the power grid operation and realizing accurate evaluation of the power grid operation status. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] Figure 1 It is a flow chart of a method for evaluating power grid operation status based on a combination of data-driven and time series features in an embodiment;

[0119] Figure 2 It is a principle block diagram of a power grid operation status assessment method based on a combination of data drive and time series features in an embodiment;

[0120] Figure 3 The frequency distribution histogram and fitting curve diagram of each indicator in the application embodiment;

[0121] Figure 4 It is a diagram of the evaluation system in three typical scenarios in the application examples. DETAILED DESCRIPTION

[0122] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent modifications of the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0123] Embodiment 1:

[0124] The present invention provides a method for evaluating the operation status of a power grid based on the combination of data drive and time series characteristics, such as Figure 1 As shown, the method includes:

[0125] S1. Obtaining grid operation parameter information;

[0126] S2. Construct a multi-dimensional indicator evaluation system for the operation status of the power grid. The multi-dimensional indicator evaluation system for the operation status of the power grid has several first-level indicators, and each first-level indicator has multiple second-level indicators;

[0127] S3. Based on the grid operation parameter information, various secondary indicators are quantified to obtain various secondary indicators of the grid hourly throughout the year;

[0128] S4. Based on the various secondary indicators of the power grid every hour throughout the year, the entropy weight method is used to obtain the weights of various secondary indicators, and the various primary indicators of the power grid every hour throughout the year are obtained;

[0129] S5. Based on the hourly primary indicators of the power grid throughout the year, the daily probability distribution of each primary indicator is statistically analyzed, and the Gaussian mixture model is used for fitting to obtain the confidence intervals of the daily probability distribution of each primary indicator at different confidence levels; for the primary indicator data of the day to be evaluated, the data-driven initial scoring is performed according to the confidence interval in which it is located;

[0130] S6. Based on the first-level indicators of the power grid every hour throughout the year, the duration of the severe scenarios in each first-level indicator is counted, and the initial scoring results are corrected secondary based on the time series characteristics.

[0131] Embodiment 2:

[0132] This embodiment is further designed on the basis of the first embodiment in that the grid operation parameter information in this example includes network topology, node admittance matrix, node voltage, branch flow, load data information, and installed capacity and output information of generator sets, new energy units, and energy storage equipment.

[0133] Embodiment three:

[0134] This embodiment is further designed on the basis of the second embodiment in that the primary indicators of the multi-dimensional indicator evaluation system of the power grid operation status in this embodiment include safety indicators, economic indicators, flexibility indicators and low-carbon indicators.

[0135] Embodiment 4:

[0136] This embodiment is further designed on the basis of the third embodiment in that the secondary indicators of the safety index in this embodiment include power supply margin of the power grid, frequency safety margin, line overload rate, voltage deviation rate and main transformer load rate;

[0137] The expression of power supply margin of power grid is:

[0138]

[0139] Where, ψ represents the power supply margin of the power grid; P sup Indicates the power supply capacity of the partition; Indicates the maximum load of the partition;

[0140] The expression of frequency safety margin is:

[0141] Given the lowest frequency point, the maximum power disturbance at this time can be obtained, and this power is defined as the frequency safety margin:

[0142]

[0143] Among them, the general frequency expression based on the inertia constant of the power system, the turbine parameters of the generator and the speed governor regulation constant is adopted:

[0144]

[0145] Where Δf max represents the maximum frequency disturbance; ΔP represents the power disturbance; D represents the load damping constant; R represents the speed regulator adjustment constant; T R represents the time constant; H represents the system inertia constant; F H represents turbine parameters; ω n , α, ω r , t nadir They are the first, second, third, fourth and fifth intermediate variables, respectively, and have no clear physical meaning;

[0146] The line overload rate describes the line's carrying capacity for transmission power. The expression of the line overload rate is:

[0147] δ=n br / N br (A-9)

[0148] In the formula, δ represents the line overload rate, n br Indicates the number of lines with transmission power exceeding 80% of capacity, N br Indicates the total number of lines;

[0149] Voltage deviation describes the power quality of the power grid during operation. The expression of voltage deviation rate is:

[0150] γ=n node / N node (A-10)

[0151] In the formula, γ represents the voltage deviation rate, n node Indicates the number of nodes whose voltage fluctuation exceeds 0.05V, N node Indicates the total number of nodes;

[0152] The expression of main transformer load factor is:

[0153] β=P i sub / P i max (A-11)

[0154] In the formula, β represents the load factor of the main transformer, P i sub Represents the power transmitted through the main transformer i, P i max represents the transmission capacity of main transformer i;

[0155] The secondary indicators of economic indicators include operating cost, capacity factor, line loss rate, and main transformer load imbalance; among them, the operating cost takes into account the power generation cost, load shedding cost, and wind and solar power abandonment cost. Among them, the power generation cost mainly comes from the fuel cost, and renewable energy power generation does not require fuel consumption, so only the fuel cost of conventional thermal power units is considered. The expression of operating cost is:

[0156]

[0157] In the formula, C OPE represents the total operating cost, c fuel Represents the unit power generation cost of the generator set, P i g represents the active output of conventional unit i, and They represent the unit load shedding cost and the unit wind and solar curtailment cost respectively. and They represent the load shedding and wind and solar power abandonment at node i respectively;

[0158] The capacity factor indicates the utilization degree of the actual power generation capacity of the generator set within a certain period of time relative to its rated capacity. The expression of the capacity factor is:

[0159]

[0160] In the formula, Φ LF represents the capacity factor, P i WT and P i PV They represent the active output of wind turbine and photovoltaic unit at node i respectively, and They represent the installed capacity of wind turbines and photovoltaic units at node i respectively;

[0161] The line loss rate describes the active power loss caused by resistance during the transmission of electric energy. Considering the transformer and line losses, the expression of the line loss rate is:

[0162]

[0163] Where N p Indicates line loss rate, ΔP bz and ΔP l Represent the transmission loss of transformer and line respectively, P r represents the overall transmission power;

[0164] The load imbalance index of the main transformer can well describe the voltage boost / voltage reduction balance of the main transformer in the sub-grid under typical conditions. The imbalance between the main transformers is quantitatively described by the mean square error. The expression of the load imbalance of the main transformer is:

[0165]

[0166] In the formula, Φ t,σ Indicates the unbalance degree of the main transformer at time t, P i,t represents the transmission power of main transformer i at time t, P i,max represents the transmission capacity of the main transformer, and n represents the total number of main transformers;

[0167] The secondary indicators of flexibility indicators include unit reserve, energy storage capacity and line transmission margin;

[0168] The expression of unit reserve is:

[0169]

[0170] In the formula, and They represent the upper and lower reserve capacities provided by the conventional unit v at time t, Indicates the maximum climbing power of conventional unit v, p v,t represents the active output of the conventional unit v at time t, and They represent the lower and upper limits of the output of the conventional unit v, respectively. It represents the total reserve capacity of the power grid at time t;

[0171] Energy storage technology provides important support for real-time balancing of power supply and demand. It stores electricity when there is excess power and releases it when there is a power shortage, thus providing real-time flexibility and effectively smoothing load fluctuations and output fluctuations of renewable energy. The expression of energy storage that can participate in regulating capacity is:

[0172]

[0173]

[0174] In the formula, and They represent the upward and downward regulation power provided by the energy storage at time t, and They represent the upper limits of energy storage charging and discharging power, and They respectively represent the charging and discharging states of the energy storage at time t, and Respectively represent the lower and upper limits of energy storage capacity, η CHand η DIS Respectively represent the charging efficiency and discharging efficiency of energy storage, P t ESS Represents the overall adjustable capacity provided by energy storage;

[0175] Whether the flexible resources at the node can fully respond to power imbalance also depends on the power transmission of the line, that is, network flexibility. The expression of line transmission margin is:

[0176]

[0177] In the formula, p v,t and p k,t They represent the active power output of conventional unit v and new energy unit k at time t, d m,t represents the load active power demand of node m at time t, P r Indicates the line transmission margin, and Represent the network topology parameters of conventional units, new energy units and loads respectively;

[0178] The secondary indicators of low-carbon index include carbon dioxide emissions, new energy penetration rate and wind and solar power abandonment rate;

[0179] Coal-fired power generation is one of the main sources of global carbon emissions. Considering the carbon emissions from conventional power generation, the expression of carbon dioxide emissions is:

[0180]

[0181] In the formula, represents carbon dioxide emissions, E f Indicates the unit carbon emission of conventional units;

[0182] The new energy penetration rate refers to the percentage of the actual power generation of clean energy to the regional load. This indicator quantifies the actual power supply capacity and power supply proportion of different new energy units from the dimension of lack of power generation, thereby evaluating the low-carbon level of the power grid. The expression of the new energy penetration rate is:

[0183]

[0184] Where η res represents the penetration rate of new energy units, represents the load demand of node i.

[0185] The wind and solar power abandonment rate is an important indicator to measure the utilization of clean energy in the power grid. The lower the wind and solar power abandonment rate, the more significant progress the power grid has made in absorbing clean energy and reducing carbon emissions. The expression of the wind and solar power abandonment rate is:

[0186]

[0187] In the formula, and They respectively represent the actual consumption and power generation of the new energy unit at node i.

[0188] Embodiment five:

[0189] This embodiment is further designed on the basis of the first embodiment in that, in this embodiment, based on the various secondary indicators of the power grid hourly throughout the year, the entropy weight method is used to obtain the weights of various secondary indicators. The specific steps of obtaining various primary indicators of the power grid hourly throughout the year include:

[0190] 1) Forward normalization of indicators. First, the indicators are normalized, that is, the cost indicators are converted into benefit indicators according to the indicator type, and then the indicators are dimensionless. The specific method is as follows:

[0191] For an indicator attribute value matrix F with m moments and n indicators m×n

[0192]

[0193] In the formula, f ij is the initial value of the jth indicator corresponding to the i-th moment, i∈[1,m], j∈[1,n];

[0194] That is, the following formula is used:

[0195] If the jth indicator is a profit indicator:

[0196]

[0197] Where b ij is the normalized value of the jth indicator corresponding to the i-th moment;

[0198] If the jth indicator is a cost indicator:

[0199]

[0200] To avoid ij The presence of zero elements in the result may cause problems in subsequent calculations. The lowest normalized interval can start from 0.002.

[0201] 2) Calculate the index entropy value

[0202] After the above indicators are normalized, the matrix B is formed m×n

[0203]

[0204] For the jth indicator, its information entropy is E j :

[0205]

[0206] In the formula, p ij is the weight of the jth indicator at the i-th moment,

[0207]

[0208] 3) Calculate weight

[0209]

[0210] In the formula, is the weight of the jth indicator;

[0211] 4) Primary indicator assignment

[0212]

[0213] in, Indicates the assignment result of the first-level index i, Represents the indicator data of the secondary indicator j under the primary indicator i.

[0214] Embodiment six:

[0215] This embodiment is further designed on the basis of the first embodiment in that, in this example, based on the various first-level indicators of the power grid hourly throughout the year, the daily probability distribution of each first-level indicator is statistically calculated, and a Gaussian mixture model is used for fitting to obtain the confidence intervals of the daily probability distribution of each first-level indicator at different confidence levels; for each first-level indicator data of the day to be evaluated, the specific steps of performing data-driven initial scoring according to the confidence interval in which it is located include:

[0216] 1) Data Normalization

[0217]

[0218] In the formula, x represents the specific value of each primary indicator, x min and x max They represent the minimum and maximum values ​​of the corresponding indicators respectively, and y represents the normalized indicator;

[0219] 2) The daily probability distribution of each primary indicator is statistically analyzed and fitted using Gaussian mixture models. The expression of the Gaussian mixture model is:

[0220]

[0221] In the formula, F(y) is the Gaussian mixture model, k represents a single Gaussian component, K represents the number of Gaussian components, and α k , μ k and Respectively represent the expected and variance of the weight coefficient of Gaussian component k, represents the probability density function of Gaussian component k;

[0222] 3) Corresponding to the fitted daily probability distribution curve of each primary indicator, specify the confidence level (confidence degree) and provide the corresponding score table;

[0223] 4) Obtain the confidence interval according to the specified confidence level based on the fitted daily probability distribution curves of each primary indicator;

[0224]

[0225] In the formula, ε represents the confidence level, δ up and δ dn represent the upper and lower bounds of the confidence interval, respectively.

[0226] 5) According to the confidence interval of each primary indicator data on the day to be evaluated, the value is assigned according to the interval, and the amplitude expression is:

[0227]

[0228] In the formula, S (1) Indicates the initial scoring result of the first-level indicator. Represents an assignment function.

[0229] Embodiment seven:

[0230] This embodiment is further designed on the basis of the first embodiment. In this embodiment, based on the first-level indicators of the power grid hourly throughout the year, the duration of the severe scenarios in each first-level indicator is counted, and the specific steps of performing a secondary correction based on the time series characteristics on the initial scoring result according to the time include:

[0231] The first-level indicators of the power grid are divided into natural days throughout the year, and the number of hours when the first-level indicators are less than or equal to the first set threshold of the corresponding first-level indicators is obtained. If the number of hours is greater than the second set threshold, the day is regarded as a severe scenario day, and the time of the severe scenario day throughout the year is recorded as the duration of the severe scenario t bad , based on the duration t of the severe scenario bad Perform secondary correction based on time series characteristics. The secondary correction formula is:

[0232] ΔS=-f(X bad ,t bad ) (A-43)

[0233] S (2) =S (1) +ΔS (A-44)

[0234] In the formula, ΔS represents the secondary correction based on the time series characteristics, X bad represents a bad scenario, f represents a correction function, S (2) Indicates the scoring result after secondary correction.

[0235] (In this example, an example is used to illustrate the above secondary correction process. The example corrects the safety index, and the other primary indicators are also processed separately using this method;

[0236] First, the hourly safety index of the power grid throughout the year is divided into natural days to form 365 hourly safety indexes corresponding to the 365 days of the year. The number of hours in each day when the safety index is less than or equal to the first set value of the safety index is counted. The following table shows the hourly safety index of August 8.

[0237] time Safety indicators time Safety indicators time Safety indicators 1 0.82 9 0.18 17 0.61 2 0.88 10 0.18 18 0.16 3 0.83 11 0.28 19 0.14 4 0.83 12 0.44 20 0.11 5 0.91 13 0.48 21 0.46 6 0.75 14 0.80 22 0.67 7 0.56 15 0.65 23 0.66 8 0.20 16 0.57 24 0.77

[0238] The first setting value of the safety index is 0.2, and the number of hours that the safety index is less than or equal to the first setting value of the safety index is 6. The second setting value is 5, and the day is a severe scenario day.

[0239] According to the above method, the duration of each level 1 indicator severe scenario is counted. bad , then based on the duration of the severe scenario t bad Each primary indicator is corrected twice. In this example, the following formula is used for correction.

[0240] S (2) =S (1) +ΔS

[0241]

[0242] Where, T total For the whole year; S (1) is the initial scoring result; S (2) This is the scoring result after the second correction;)

[0243] Embodiment eight:

[0244] The present invention provides a power grid operation status evaluation system based on the combination of data drive and time series characteristics, comprising a data acquisition module, a system construction module, a secondary index calculation module, a primary index calculation module, a primary scoring module and a secondary scoring module;

[0245] A data acquisition module is used to obtain power grid operation parameter information;

[0246] The system construction module is used to construct a multi-dimensional indicator evaluation system for the operation status of the power grid. The multi-dimensional indicator evaluation system for the operation status of the power grid has several first-level indicators, and each first-level indicator has multiple second-level indicators.

[0247] The secondary index calculation module is used to quantify various secondary indicators based on the grid operation parameter information and obtain various secondary indicators of the grid hour by hour throughout the year;

[0248] The primary index calculation module is used to obtain the weights of various secondary indexes based on the hourly secondary indexes of the power grid throughout the year by using the entropy weight method to obtain various primary indexes of the power grid hourly throughout the year;

[0249] The initial scoring module is used to calculate the daily probability distribution of each first-level indicator based on the hourly first-level indicators of the power grid throughout the year, and use the Gaussian mixture model for fitting to obtain the confidence intervals of the daily probability distribution of each first-level indicator at different confidence levels; for the first-level indicator data of the day to be evaluated, the initial scoring based on data-driven is performed according to the confidence interval in which it is located;

[0250] The secondary scoring module is used to calculate the duration of the severe scenarios in each primary indicator based on the hourly primary indicators of the power grid throughout the year, and make secondary corrections to the initial scoring results based on the time series characteristics.

[0251] Embodiment nine:

[0252] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute any of the methods described above.

[0253] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0254] Application examples:

[0255] This example evaluates the operating status of a power grid in a certain area of ​​Jiangsu Province. The power grid in this area has 64 nodes, 2 photovoltaic power stations, 5 wind farms, 1 thermal power plant, and 85 lines. The total capacity of the photovoltaic power stations is 1450MW, the total capacity of the wind farms is 3203MW, and the capacity of the thermal power plants is 1320MW.

[0256] Figure 2 A principle block diagram of the evaluation method of the present invention is given, wherein the indicator data are all standardized and positively processed.

[0257] Figure 3 The frequency distribution histogram and fitting curve of each indicator are given. Figure 3It can be seen that the Gaussian mixture model can better fit the probability distribution of multimodality and skewness of each indicator by superimposing multiple Gaussian distributions.

[0258] Figure 4 The evaluation system is shown in three typical scenarios. As can be seen from the figure, economy and safety show a "seesaw" effect: high safety leads to poor economy (scenario 2), and high economy leads to poor safety (scenario 3). The low carbon nature of the power grid is positively correlated with economy: increasing the output of new energy can reduce both the cost of power generation and carbon emissions.

[0259] The method of the present invention can effectively mine historical data information and provide an objective reference for the interval assignment of indicators. In addition, it further considers the timing operation characteristics of the power grid and makes score corrections based on the duration of severe scenarios, thus facilitating power grid planning decisions and risk warnings.

[0260] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for evaluating power grid operation status based on the combination of data drive and time series characteristics, characterized in that: include: Obtaining grid operation parameter information; Constructing a multi-dimensional indicator evaluation system for the operation status of a power grid, wherein the multi-dimensional indicator evaluation system for the operation status of a power grid has a number of first-level indicators, and each first-level indicator has a number of second-level indicators; Based on the power grid operation parameter information, various secondary indicators are quantified to obtain various secondary indicators of the power grid hour by hour throughout the year; Based on the various secondary indicators of the power grid every hour throughout the year, the entropy weight method is used to obtain the weights of various secondary indicators, and the various primary indicators of the power grid every hour throughout the year are obtained; Based on the hourly primary indicators of the power grid throughout the year, the daily probability distribution of each primary indicator is statistically analyzed, and a Gaussian mixture model is used for fitting to obtain the confidence intervals of the daily probability distribution of each primary indicator at different confidence levels; for each primary indicator data of the day to be evaluated, a data-driven initial scoring is performed according to the confidence interval in which it is located; Based on the hourly primary indicators of the power grid throughout the year, the duration of the severe scenarios in each primary indicator is counted, and the initial scoring results are corrected secondary based on the time series characteristics according to the time.

2. The power grid operation status assessment method based on the combination of data drive and time series characteristics according to claim 1 is characterized in that: The grid operation parameter information includes network topology, node admittance matrix, node voltage, branch flow, load data information, and installed capacity and output information of generator sets, new energy units, and energy storage equipment.

3. The power grid operation status assessment method based on the combination of data drive and time series characteristics according to claim 2 is characterized in that: The primary indicators of the multi-dimensional indicator evaluation system for power grid operation status include safety indicators, economic indicators, flexibility indicators and low-carbon indicators.

4. The method for evaluating power grid operation status based on the combination of data drive and time series characteristics according to claim 3 is characterized in that: The secondary indicators of the safety index include power supply margin of the power grid, frequency safety margin, line overload rate, voltage deviation rate and main transformer load rate; The expression of the power supply margin of the power grid is: Where, ψ represents the power supply margin of the power grid; P sup Indicates the power supply capacity of the partition; Indicates the maximum load of the partition; The expression of the frequency safety margin is: in, Where Δf max represents the maximum frequency disturbance; ΔP represents the power disturbance; D represents the load damping constant; R represents the speed regulator adjustment constant; T R represents the time constant; H represents the system inertia constant; F H represents turbine parameters; ω n , α, ω r , t nadir are the first, second, third, fourth, and fifth intermediate variables, respectively; The expression of the line overload rate is: δ=n br / N br (A-9) In the formula, δ represents the line overload rate, n br Indicates the number of lines with transmission power exceeding 80% of capacity, N br Indicates the total number of lines; The expression of the voltage deviation rate is: γ=n node / N node (A-10) In the formula, γ represents the voltage deviation rate, n node Indicates the number of nodes whose voltage fluctuation exceeds 0.05V, N node Indicates the total number of nodes; The expression of the main transformer load rate is: In the formula, β represents the load factor of the main transformer, represents the power transmitted through the main transformer i, represents the transmission capacity of main transformer i; The secondary indicators of the economic indicators include operating cost, capacity factor, line loss rate, and main transformer load imbalance; wherein the operating cost is expressed as: In the formula, C OPE represents the total operating cost, c fuel represents the unit power generation cost of the generator set, represents the active output of conventional unit i, and They represent the unit load shedding cost and the unit wind and solar curtailment cost respectively. and They represent the load shedding and wind and solar power abandonment at node i respectively; The expression of the capacity factor is: In the formula, Φ LF represents the capacity factor, and They represent the active output of wind turbine and photovoltaic unit at node i respectively, and They represent the installed capacity of wind turbines and photovoltaic units at node i respectively; The expression of the line loss rate is: Where N p Indicates line loss rate, ΔP bz and ΔP l Represent the transmission loss of transformer and line respectively, P r represents the overall transmission power; The expression of the main transformer load imbalance is: In the formula, Φ t,σ Indicates the unbalance degree of the main transformer at time t, P i,t represents the transmission power of main transformer i at time t, P i,max represents the transmission capacity of the main transformer, and n represents the total number of main transformers; The secondary indicators of the flexibility index include unit reserve, energy storage capacity for regulation and line transmission margin; The expression of the unit standby is: In the formula, and They represent the upper and lower reserve capacities provided by the conventional unit v at time t, Indicates the maximum climbing power of conventional unit v, p v,t represents the active output of the conventional unit v at time t, and They represent the lower and upper limits of the output of the conventional unit v, respectively. It represents the total reserve capacity of the power grid at time t; The expression of the energy storage that can participate in regulating capacity is: In the formula, and They represent the upward and downward regulation power provided by the energy storage at time t, and They represent the upper limits of energy storage charging and discharging power, respectively. and They respectively represent the charging and discharging states of the energy storage at time t, and Respectively represent the lower and upper limits of energy storage capacity, η CH and η DIS They represent the charging efficiency and discharging efficiency of energy storage, Represents the overall adjustable capacity provided by energy storage; The expression of the line transmission margin is: In the formula, p v,t and p k,t They represent the active output of conventional unit v and new energy unit k at time t, respectively, and d m,t represents the load active power demand of node m at time t, P r Indicates the line transmission margin, and Represent the network topology parameters of conventional units, new energy units and loads respectively; The secondary indicators of the low-carbon index include carbon dioxide emissions, new energy penetration rate and wind and solar power abandonment rate; The expression of carbon dioxide emission is: In the formula, represents carbon dioxide emissions, E f Indicates the unit carbon emission of conventional units; The expression of the new energy penetration rate is: Where η res represents the penetration rate of new energy units, represents the load demand of node i; The expression of wind and solar power abandonment rate is: In the formula, and They respectively represent the actual consumption and power generation of the new energy unit at node i.

5. The method for evaluating power grid operation status based on the combination of data drive and time series characteristics according to claim 1, characterized in that: The specific steps of obtaining the weights of the secondary indicators based on the hourly secondary indicators of the power grid throughout the year by using the entropy weight method to obtain the hourly primary indicators of the power grid throughout the year include: 1) Forward normalization of indicators. First, the indicators are normalized, that is, the cost indicators are converted into benefit indicators according to the indicator type, and then the indicators are dimensionless. The specific method is as follows: For an indicator attribute value matrix F with m moments and n indicators m×n In the formula, f ij is the initial value of the jth indicator corresponding to the i-th moment, i∈[1,m], j∈[1,n]; That is, the following formula is used: If the jth indicator is a profit indicator: Where b ij is the normalized value of the jth indicator corresponding to the i-th moment; If the jth indicator is a cost indicator: 2) Calculate the index entropy value After the above indicators are normalized, the matrix B is formed m×n For the jth indicator, its information entropy is E j : In the formula, p ij is the weight of the jth indicator at the i-th moment, 3) Calculate weight In the formula, is the weight of the jth indicator; 4) Primary indicator assignment in, Indicates the assignment result of the first-level index i, Represents the indicator data of the secondary indicator j under the primary indicator i.

6. The method for evaluating power grid operation status based on the combination of data drive and time series characteristics according to claim 1, characterized in that: The daily probability distribution of each first-level indicator is statistically calculated based on the first-level indicators of the power grid hourly throughout the year, and a Gaussian mixture model is used for fitting to obtain the confidence interval of the daily probability distribution of each first-level indicator at different confidence levels; for each first-level indicator data of the day to be evaluated, the specific steps of performing data-driven initial scoring according to the confidence interval in which it is located include: 1) Data normalization In the formula, x represents the specific value of each primary indicator, x min and x max They represent the minimum and maximum values ​​of the corresponding indicators respectively, and y represents the normalized indicator; 2) The daily probability distribution of each primary indicator is statistically analyzed and fitted using a Gaussian mixture model. The expression of the Gaussian mixture model is: In the formula, F(y) is the Gaussian mixture model, k represents a single Gaussian component, K represents the number of Gaussian components, and α k , μ k and Respectively represent the expected and variance of the weight coefficient of Gaussian component k, represents the probability density function of Gaussian component k; 3) Correspond to the fitted daily probability distribution curves of each primary indicator, specify the confidence level and provide the corresponding score table; 4) Obtain the confidence interval according to the specified confidence level based on the fitted daily probability distribution curves of each primary indicator; In the formula, ε represents the confidence level, δ up and δ dn Represent the upper and lower bounds of the confidence interval respectively; 5) According to the confidence interval of each primary indicator data on the day to be evaluated, the value is assigned according to the interval, and the amplitude expression is: In the formula, S (1) Indicates the initial scoring result of the first-level indicator. Represents an assignment function.

7. The method for evaluating power grid operation status based on the combination of data drive and time series characteristics according to claim 1, characterized in that: The specific steps of counting the duration of the severe scenarios in each first-level indicator based on each hour of the power grid throughout the year and performing a secondary correction on the initial scoring result based on the time series characteristics according to the time include: The first-level indicators of the power grid are divided into natural days throughout the year, and the number of hours when the first-level indicators are less than or equal to the first set threshold of the corresponding first-level indicators is obtained. If the number of hours is greater than the second set threshold, the day is regarded as a severe scenario day, and the time of the severe scenario day throughout the year is recorded as the duration of the severe scenario t bad , based on the duration t of the severe scenario bad Perform secondary correction based on time series characteristics. The secondary correction formula is: ΔS=-f(X bad ,t bad ) (A-43) S (2) =S (1) +ΔS (A-44) In the formula, ΔS represents the secondary correction based on the time series characteristics, X bad represents a bad scenario, f represents a correction function, S (2) Indicates the scoring result after secondary correction.

8. A power grid operation status assessment system based on the combination of data drive and time series characteristics, characterized in that: It includes data acquisition module, system construction module, secondary indicator calculation module, primary indicator calculation module, primary scoring module and secondary scoring module; The data acquisition module is used to acquire power grid operation parameter information; The system construction module is used to construct a multi-dimensional indicator evaluation system for the operation status of a power grid. The multi-dimensional indicator evaluation system for the operation status of a power grid has several first-level indicators, and each first-level indicator has multiple second-level indicators. The secondary index calculation module is used to quantify various secondary indicators based on the power grid operation parameter information to obtain various secondary indicators of the power grid hour by hour throughout the year; The primary index calculation module is used to obtain the weights of various secondary indicators based on the secondary indicators of the power grid hourly throughout the year by using the entropy weight method to obtain various primary indicators of the power grid hourly throughout the year; The initial scoring module is used to calculate the daily probability distribution of each primary indicator based on the hourly primary indicators of the power grid throughout the year, and use a Gaussian mixture model for fitting to obtain the confidence intervals of the daily probability distribution of each primary indicator at different confidence levels; for each primary indicator data of the day to be evaluated, perform data-driven initial scoring according to the confidence interval in which it is located; The secondary scoring module is used to count the duration of the severe scenarios in each primary indicator based on the primary indicators of the power grid hour by hour throughout the year, and perform secondary corrections to the initial scoring results based on the time series characteristics according to the time.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.

Images