Generalized unscented transformation-based comprehensive energy system global sensitivity calculation method, device and medium

Abstract

The application discloses a kind of generalized unscented transformation-based comprehensive energy system global sensitivity calculation method, equipment and medium, establishes the thermal power probability multi-energy flow model considering coupling device and coupling mode;Generalized unscented transformation is applied to extract key samples to reduce sampling times;Efficiently calculate the global sensitivity of uncertain source in comprehensive energy system.The generalized unscented transformation-based comprehensive energy system global sensitivity calculation method, equipment and medium provided by the application greatly reduce the sample size while considering the correlation of heat network load, effectively solve the problem of excessive calculation amount of traditional Monte Carlo sampling algorithm, and can quickly calculate the influence degree of uncertain source in comprehensive energy system on power grid operation.

Description

Technical Field

[0001] This invention relates to a method, device, and medium for calculating the global sensitivity of an integrated energy system based on generalized unscented transformation, belonging to the field of uncertainty analysis technology for integrated energy systems. Background Technology

[0002] Energy is the driving force and foundation of social and economic development. my country's ever-increasing energy consumption and pressing environmental problems urgently require us to develop clean energy industries and establish a low-carbon, sustainable energy system. Electrothermal coupling systems, as a major form of integrated energy systems, help improve the complementarity of electrical and thermal energy utilization, increase overall energy efficiency, and raise the proportion of renewable energy consumption. Related technological research has become a current research hotspot.

[0003] However, integrated energy systems involve the coordinated operation of electric and thermal networks. The volatility and strong correlation of loads in the heating network significantly increase the uncertainty of system operation, which may affect the operating status of the power system and the original power distribution within the grid, thereby further impacting the overall power quality of the grid. To ensure the safe operation of the system, it is necessary to quantitatively assess the impact of various uncertainties on the operation of the electric-thermal coupling system. Global sensitivity analysis is an effective tool for quantitatively assessing the impact of uncertainties. The Sobol method is a common global sensitivity analysis method. This method quantifies the impact of each input variable on the system's output variable based on analysis of variance. Sensitivity analysis assesses the degree of influence of uncertainties by quantifying the relationship between output and input, helping operators identify key factors affecting system operation.

[0004] Global sensitivity analysis requires a large number of random samples. In existing techniques, Monte Carlo simulation (MCS) is a common sampling method; however, MCS requires a large number of samples to achieve acceptable convergence, making it inefficient for analyzing global sensitivity. Therefore, more efficient sampling methods are needed for global sensitivity analysis. Summary of the Invention

[0005] Objective: To overcome the shortcomings of existing technologies, this invention provides a method, device, and medium for calculating the global sensitivity of integrated energy systems based on generalized unscented transformation. This method is applied to thermoelectric coupled integrated energy systems. By extracting a small number of key samples, it reduces the computational burden while ensuring the accuracy of global sensitivity, which is of great significance for improving the efficiency of uncertainty analysis of large systems.

[0006] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0007] Firstly, a method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform includes the following steps:

[0008] Obtain a probabilistic multi-energy flow model for a comprehensive energy system that considers multiple types of coupled devices.

[0009] The generalized unscented transform is applied to extract key samples from the integrated energy system, and statistical information of the output variables of the probabilistic multi-energy flow model is obtained based on the key samples.

[0010] Calculate the first-order sensitivity of the variables in the integrated energy system based on the statistical information of the output variables.

[0011] Optionally, obtaining the probabilistic multi-energy flow model of the integrated energy system considering multiple types of coupled devices specifically includes:

[0012] Obtain the steady-state power flow model of the heating network system.

[0013] Obtain the steady-state power flow model of the power grid system.

[0014] Obtain models of different types of electrothermal coupling devices.

[0015] Based on the steady-state power flow model of the heating network system, the steady-state power flow model of the power grid system, and the models of different types of electrothermal coupling equipment, a probabilistic multi-energy flow model of the integrated energy system is constructed.

[0016] The calculation formula for the probabilistic multi-energy flow model of the integrated energy system is as follows:

[0017] g = f(x)

[0018] Where, x={φ,P G ,P D Q G Q D} represents the uncertainty in the system, serving as the input to the integrated energy system. f represents the steady-state power flow model of the power grid and heating network system, as well as the models of different types of electrothermal coupling equipment. This represents the state variable to be determined in the system, which serves as the output of the integrated energy system.

[0019] Wherein, φ includes the following variables: φ i φ bp φ eb or φ hp P G Includes the following variable: P Gk P D Includes the following variable: P Dk Q G Includes the following variable: Q Gk Q D Includes the following variable: Q Dk φi This represents the heat power output by the heat source at node i or the heat power transferred in pipe i. φ bp φ eb φ hp These represent the heat energy generated by the back-pressure cogeneration unit, the electric furnace, and the heat pump, respectively. Gk Q Gk P represents the active power and reactive power injected at node k, respectively; Dk Q Dk These represent the active power and reactive power consumed at node k, respectively.

[0020] in, T represents the mass flow rate of water in pipe i. i s T i r V represents the temperature of the water in the inlet and outlet pipes of pipe i. V includes the following variables: V k V i δ includes the following variables: δ ki Among them, V k V i δ represents the voltage magnitude at node k and node i, respectively; ki This represents the phase angle difference between nodes k and i.

[0021] Optionally, the steady-state power flow model calculation formula for the heating network system is as follows:

[0022]

[0023]

[0024]

[0025]

[0026] Among them, T i out T i in T represents the temperature of water flowing out of and into pipe i. a Indicates ambient temperature; This represents the mass flow rate of water in pipe i; This represents the mass flow rate of water flowing from pipe i into node d; T represents the temperature of water flowing from pipe i into node d; d φ represents the temperature of the water after mixing at node d; i T represents the output heat power of the heat source at node i or the heat power transferred in pipe i; i s T ir λ represents the temperature of the water in the inlet and return pipes of pipe i; c represents the specific heat capacity of the water; λ represents the heat transfer coefficient per unit length of the pipe; L represents the length of the heating network pipes; K represents the resistance coefficient matrix of the pipes in the network; B represents the loop matrix. Represents the set of all pipes; Represents the set of all thermal nodes; This represents the set of thermal nodes connected to heating network pipe i; This represents the set of pipes through which water flows into node d.

[0027] The calculation formula for the steady-state power flow model of the power grid system is as follows:

[0028]

[0029]

[0030] Where m represents the number of power nodes; G ki B ki V represents the conductance and susceptance of the line between nodes k and i, respectively; k V i δ represents the voltage magnitude at node k and node i, respectively; ki P represents the phase angle difference between nodes k and i; Gk Q Gk P represents the active power and reactive power injected at node k, respectively; Dk Q Dk These represent the active power and reactive power consumed at node k, respectively.

[0031] The calculation formulas for the different types of electrothermal coupling devices are as follows:

[0032] φ bp =P bp η bp

[0033] φ eb =P eb η eb

[0034] φ hp =P hp COP hp

[0035] Where, φ bp φ eb φ hp These represent the heat energy generated by the back-pressure cogeneration unit, the electric furnace, and the heat pump, respectively. bp P represents the electrical energy generated by a back-pressure combined heat and power unit; eb P hpThese represent the electrical energy consumed by the electric furnace and the heat pump, respectively; η bp Indicates the heat and power ratio of a combined heat and power (CHP) unit with a CHP boiler; η eb COP hp These represent the heat production efficiency of the electric furnace and the heat pump, respectively.

[0036] Optionally, the application of generalized unscented transform to extract key samples from the integrated energy system, and the acquisition of statistical information on the output variables of the probabilistic multi-energy flow model based on the key samples, specifically includes:

[0037] S21. Calculate the mean vector of each input random variable x in the integrated energy system. Covariance matrix P, skewness vector and kurtosis vector

[0038] S22, according to P uses a symmetric sampling strategy to calculate the position of the Sigma point sampled by the generalized unscented transform method. The calculation formula is as follows:

[0039]

[0040]

[0041]

[0042] Where i∈{1,2,…,n}; χ 0 χ represents the position coordinates of the center Sigma point; i χ represents the position coordinates of the i-th Sigma point; i+n This represents the position coordinates of the (i+n)th Sigma point; a i This indicates the degree to which the i-th Sigma point deviates from the mean of the input variable; b i This indicates the degree to which the (i+n)th Sigma point deviates from the mean of the input variable. It is the matrix obtained by Cholesky decomposition of the covariance matrix P of the random variable x; yes The i-th column of the matrix.

[0043] S23, according to and Calculate the values ​​of a and b:

[0044]

[0045]

[0046] Where ⊙ represents the Hadamard product; a = {a1, a2, ..., a...} n} represents the deviation of the first to nth Sigma points from the mean, where b = {b1, b2, ..., bn} n} represents the degree of deviation of the (n+1)th to 2nth Sigma points from the mean.

[0047] S24. Calculate the weights corresponding to the Sigma point based on a and b:

[0048]

[0049]

[0050]

[0051] Where i∈{1,2,...,n}; w i Represents vector w j The i-th element in; w i+n Represents vector w j The (i+n)th element in the array. 0 Represents vector w j The 0th element in the array, where j∈{0,1,2,...,2n}.

[0052] S25, Calculate each Sigma point χ i Substituting these values ​​into the probabilistic multi-energy flow model, a corresponding output sample point is obtained.

[0053]

[0054] S26. Output all sample points Weighted calculations are performed to obtain statistical information about the output variables.

[0055]

[0056]

[0057] Where f0 represents the mean of the output variable estimated by the generalized unscented transformation, and D represents the variance of the output variable estimated by the generalized unscented transformation.

[0058] Optionally, the step of calculating the first-order sensitivity of the variables of the integrated energy system based on the statistical information of the output variables specifically includes:

[0059] S32. Based on the generalized unscented transformation, the i-th variable X in the target variable x of the sensitivity analysis... i The corresponding normal distribution α is sampled as the outer sample in the two-level sampling algorithm:

[0060]

[0061]

[0062]

[0063] Where, μ α σ represents the expectation of α; i X represents the i-th variable X in the target variable x. i Standard deviation; They represent X respectively i Three Sigma sample points after unscented transformation sampling;

[0064] S33. Based on the three extracted Sigma sample points, calculate the conditional expectation vector and conditional covariance matrix of β respectively:

[0065]

[0066]

[0067]

[0068]

[0069] Where β represents the division by X i The normal distribution of the remaining input variables, μ β Σ represents the expectation of β; β Let Σ represent the covariance of β. α Σ represents the covariance of α; βα This represents the covariance between α and β.

[0070] S34. Based on the conditional expectation vector and conditional covariance matrix of β, apply the generalized unscented transformation to the normal distribution. Perform sampling as the inner sampling in the two-level sampling algorithm:

[0071]

[0072]

[0073]

[0074] in, Indicates the relationship with the outer sample point The corresponding Sigma point in the inner sample group, Indicates the relationship with the outer sample point The corresponding Sigma point in the inner sample group. Indicates the relationship with the outer sample point The Sigma point of the (i+2)th (n-1)th element in the corresponding inner sample group; express The corresponding weights; express The corresponding weights; express The corresponding weights; express The i-th column of the matrix, a i This indicates the degree to which the i-th Sigma point deviates from the mean of the input variable; b i This indicates the degree to which the (i+2)th (n-1)th Sigma point deviates from the mean of the input variable. Let represent the conditional distribution variable of β with respect to α. Indicates and Random variables that follow the same probability distribution.

[0075] S35. Combining the Sigma points of the inner and outer layers and substituting them into the probabilistic multi-energy flow model of the integrated energy system, we obtain...

[0076]

[0077]

[0078] In the formula, Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0079] Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0080] Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0081] in, express The first (n-1) dimensions, express The nth to 2(n-1)th dimensions, where j∈{0,1,2,...,2n}, and k takes the values ​​0, 1, and 2.

[0082] S36. Based on the output samples obtained in step S35, calculate the first-order sensitivity S of the i-th variable. y :

[0083]

[0084] in, They represent and The weights corresponding to the sample points. They represent and The weights corresponding to the sample points.

[0085] In a second aspect, a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transformation as described in any of the first aspects.

[0086] Thirdly, a computer device comprising:

[0087] Memory is used to store instructions.

[0088] A processor is configured to execute the instructions, causing the computer device to perform operations as described in any of the first aspects of a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transform.

[0089] Beneficial effects: The present invention provides a method, device and medium for calculating the global sensitivity of integrated energy systems based on generalized unscented transformation, establishes a thermoelectric probabilistic multi-energy flow model considering coupling devices and coupling methods, applies generalized unscented transformation to extract key samples to reduce the number of samplings, and efficiently calculates the global sensitivity of uncertain sources in integrated energy systems.

[0090] This invention significantly reduces the sample size while considering the correlation of heating network load, effectively solving the problem of excessive computation in traditional Monte Carlo sampling algorithms, and can quickly calculate the impact of uncertain sources in integrated energy systems on grid operation. Attached Figure Description

[0091] Figure 1 This is a flowchart of a method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transformation, according to the present invention.

[0092] Figure 2 This is a system structure diagram of the regional heating network in Embodiment 1 of the present invention.

[0093] Figure 3 This is a structural diagram of the electrothermal coupling integrated energy system in Embodiment 1 of the present invention.

[0094] Figure 4 This is the sensitivity index calculation result of Embodiment 1 of the present invention. Detailed Implementation

[0095] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0096] The present invention will be further described below with reference to specific embodiments.

[0097] Firstly, such as Figure 1 As shown, a method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform includes the following steps:

[0098] S1. Establish a probabilistic multi-energy flow model for a comprehensive energy system that considers multiple types of coupled devices.

[0099] S11. Establish a steady-state power flow model for the heating network system:

[0100]

[0101]

[0102]

[0103]

[0104] Among them, T i out T i in T represents the temperature of water flowing out of and into pipe i. a Indicates ambient temperature; This represents the mass flow rate of water in pipe i; This represents the mass flow rate of water flowing from pipe i into node d; T represents the temperature of water flowing from pipe i into node d; d φ represents the temperature of the water after mixing at node d; i T represents the output heat power of the heat source at node i or the heat power transferred in pipe i; i s T i r λ represents the temperature of the water in the inlet and return pipes of pipe i; c represents the specific heat capacity of the water; λ represents the heat transfer coefficient per unit length of the pipe; L represents the length of the heating network pipes; K represents the resistance coefficient matrix of the pipes in the network; B represents the loop matrix. Represents the set of all pipes; Represents the set of all thermal nodes; This represents the set of thermal nodes connected to heating network pipe i; This represents the set of pipes through which water flows into node d.

[0105] S12. Establish a steady-state power flow model for the power grid system:

[0106]

[0107]

[0108] Where m represents the number of power nodes; G ki B ki V represents the conductance and susceptance of the line between nodes k and i, respectively; k V i δ represents the voltage magnitude at node k and node i, respectively; ki P represents the phase angle difference between nodes k and i; Gk Q Gk P represents the active power and reactive power injected at node k, respectively; Dk Q Dk These represent the active power and reactive power consumed at node k, respectively.

[0109] S13. Construct models of different types of electrothermal coupling devices:

[0110] φ bp =P bp η bp

[0111] φ eb =P eb η eb

[0112] φ hp =P hp COP hp

[0113] Where, φ bp φ eb φ hp These represent the heat energy generated by the back-pressure cogeneration unit, the electric furnace, and the heat pump, respectively.

[0114] P bp P represents the electrical energy generated by a back-pressure combined heat and power unit; eb P hp These represent the electrical energy consumed by the electric furnace and the heat pump, respectively; η bp Indicates the heat and power ratio of a combined heat and power (CHP) unit with a CHP boiler; η eb COP hp These represent the heat production efficiency of the electric furnace and the heat pump, respectively.

[0115] S14. Construct a probabilistic multi-energy flow model of a thermoelectric coupled integrated energy system based on S11-S13.

[0116]

[0117] This model can be simplified as follows:

[0118] g = f(x)

[0119] Where, x={φ,P G ,P D Q G Q D} represents the uncertainty in the system, serving as the input to the integrated energy system. f represents the steady-state power flow model of the power grid and heating network system, as well as the models of different types of electrothermal coupling equipment. This represents the state variable to be determined in the system, which serves as the output of the integrated energy system.

[0120] Wherein, φ includes the following variables: φ i φ bp φ eb or φ hp P G Includes the following variable: P Gk P D Includes the following variable: P Dk Q G Includes the following variable: Q Gk Q D Includes the following variable: Q Dk φ i This represents the heat power output by the heat source at node i or the heat power transferred in pipe i. φ bp φ eb φ hp These represent the heat energy generated by the back-pressure cogeneration unit, the electric furnace, and the heat pump, respectively. Gk Q Gk P represents the active power and reactive power injected at node k, respectively; Dk Q Dk These represent the active power and reactive power consumed at node k, respectively.

[0121] in, T represents the mass flow rate of water in pipe i. i s T i r V represents the temperature of the water in the inlet and outlet pipes of pipe i. V includes the following variables: V k V i δ includes the following variables: δ ki Among them, V k V iδ represents the voltage magnitude at node k and node i, respectively; ki This represents the phase angle difference between nodes k and i.

[0122] S2. Use generalized unscented transformation to extract key samples to reduce the number of samplings.

[0123] S21. Calculate the mean vector of each input random variable x in step S14. Covariance matrix P, skewness vector and kurtosis vector

[0124] S22, Calculated based on step S21 P, and Then, the position of the Sigma point sampled by the generalized unscented transform method is calculated using a symmetric sampling strategy:

[0125]

[0126]

[0127]

[0128] Where i∈{1,2,…,n}; χ 0 χ represents the position coordinates of the center Sigma point; i χ represents the position coordinates of the i-th Sigma point; i+n This represents the position coordinates of the (i+n)th Sigma point; a i This indicates the degree to which the i-th Sigma point deviates from the mean of the input variable; b i This indicates the degree to which the (i+n)th Sigma point deviates from the mean of the input variable. It is the matrix obtained by Cholesky decomposition of the covariance matrix P of the random variable x; yes The i-th column of the matrix.

[0129] S23, based on the result obtained in step S21 and Calculate the values ​​of a and b:

[0130]

[0131]

[0132] Where ⊙ represents the Hadamard product; a = {a1, a2, ..., a...} n} represents the degree of deviation of the first to nth Sigma points from the mean, b = {b n+1 ,b n+2,...,b 2n} represents the degree of deviation of the (n+1)th to 2nth Sigma points from the mean.

[0133] S24. Calculate the weights corresponding to the Sigma points based on a and b obtained in step S23:

[0134]

[0135]

[0136]

[0137] Where i∈{1,2,...,n}; w i Represents vector w j The i-th element in; w i+n Represents vector w j The (i+n)th element in the array. 0 Represents vector w j The 0th element in the array, where j∈{0,1,2,...,2n}.

[0138] S25. Based on the sampling nodes and weights obtained in steps S21 to S24, assign each Sigma point χ... i Substituting the values ​​into the probabilistic multi-energy flow model obtained in step S14, a corresponding output sample is obtained.

[0139]

[0140] S26. All output sample points obtained in step S25 Weighted calculations are performed to obtain statistical information about the output variables.

[0141]

[0142]

[0143] Where f0 represents the mean of the output variable estimated by the generalized unscented transformation, and D represents the variance of the output variable estimated by the generalized unscented transformation.

[0144] S3. Efficiently calculate the global sensitivity of uncertain sources in an integrated energy system.

[0145] S31. Calculate the first-order sensitivity S of the i-th variable considering the correlation between variables using the following formula. y Since the following formula is not easy to solve, it needs to be further optimized:

[0146]

[0147] It can be derived from normal variables Obtained through inverse Gaussian Copula transform:

[0148]

[0149]

[0150] Where y represents the i-th variable X i z represents the difference between X and X. i Other input variables besides; This represents the conditional distribution of variable z with respect to variable y. Indicates and Random variables that follow the same probability distribution; α, β, Representing the relationship with y, z, respectively The corresponding normal random variable; Φ(α) represents the probability density function of the normal variable α; Let represent the conditional probability density function of β with respect to α; Represents the i-th variable X i The inverse probability distribution function; This indicates that, except for the i-th variable X i The inverse joint probability distribution function of the remaining variables; f0 represents the mean of the output variable; D represents the variance of the output variable; f0 and D can be obtained by using the generalized unscented transformation in step S26.

[0151] S32, The following steps are for further optimization of S y Solve the formula by applying the generalized unscented transformation in step S22 to the i-th variable X in the target variable x of the sensitivity analysis. i The corresponding normal distribution α is sampled as the outer sample in the two-level sampling algorithm:

[0152]

[0153]

[0154]

[0155] Where, μ α σ represents the expectation of α; i X represents the i-th variable X in the target variable x. i Standard deviation; They represent X respectively i Three Sigma sample points after unscented transformation sampling.

[0156] S33. Based on the Sigma points extracted in step S32, calculate the conditional expectation and conditional covariance matrix of β respectively:

[0157]

[0158]

[0159]

[0160]

[0161] Where β represents the division by X i The normal distribution of the remaining input variables, μ β Σ represents the expectation of β; β Let Σ represent the covariance of β. α Σ represents the covariance of α; βα This represents the covariance between α and β.

[0162] S34. Based on the conditional expectation vector and conditional covariance matrix of β, apply the generalized unscented transformation to the normal distribution. Perform sampling as the inner sampling in the two-level sampling algorithm:

[0163]

[0164]

[0165]

[0166] in, Indicates the relationship with the outer sample point The corresponding Sigma point in the inner sample group, Indicates the relationship with the outer sample point The corresponding Sigma point in the inner sample group. Indicates the relationship with the outer sample point The Sigma point of the (i+2)th (n-1)th element in the corresponding inner sample group; express The i-th column of the matrix, a i This indicates the degree to which the i-th Sigma point deviates from the mean of the input variable; b i This indicates the degree to which the (i+2)th (n-1)th Sigma point deviates from the mean of the input variable. Let represent the conditional distribution variable of β with respect to α. Indicates and Random variables that follow the same probability distribution.

[0167] S35. Combine the Sigma points of the inner and outer layers and then substitute them into the probabilistic multi-energy flow model of the integrated energy system:

[0168]

[0169]

[0170] In the formula, Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0171] Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0172] Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform.

[0173] in, express The first (n-1) dimensions, express The nth to the nth 2(n-1) There are 3 dimensions, where j∈{0,1,2,...,4(n-1)} and k takes the values ​​0, 1, and 2.

[0174] S36. Based on the output samples obtained in step S35, calculate the first-order sensitivity S of the i-th variable. y :

[0175]

[0176] in, They represent and The weights corresponding to the sample points. They represent and The weights corresponding to the sample points.

[0177] In a second aspect, a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transformation as described in any of the first aspects.

[0178] Thirdly, a computer device comprising:

[0179] Memory is used to store instructions.

[0180] A processor is configured to execute the instructions, causing the computer device to perform operations as described in any of the first aspects of a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transform.

[0181] Example 1:

[0182] The integrated energy system structure of this embodiment is as follows: Figure 3 As shown. The system includes a 30-node power grid and three 35-node heating networks. The heating network structure is as follows. Figure 2 As shown. The power system consists of conventional generator units, two back-pressure cogeneration units, and one electric boiler. The heat-to-power ratio of the cogeneration is set at 1.3, and the efficiency of the electric boiler is set at 0.8. All heat loads in the same regional heating network are considered as a source of uncertainty. The mean of the total heat load of each regional heating subsystem is 10.8 MW. Furthermore, it is assumed that all heat loads follow a normal distribution with a standard deviation equal to 20% of the mean.

[0183] Regional heating network HN1 is connected to node 7 of the power grid via an electric furnace; regional heating network HN2 is connected to node 13 of the power grid via a combined heat and power (CHP) unit; and regional heating network HN3 is connected to node 22 of the power grid via a CHP unit. By varying the correlation coefficient of the loads between heating networks HN1 and HN2, the sensitivity of active power input to lines 2-6 of the power grid under different load correlation conditions is analyzed.

[0184] According to the steps of this invention, a global sensitivity analysis based on the generalized unscented transform is performed, and the calculation results are as follows: Figure 4 As shown.

[0185] Table 1 compares the computation time required using generalized unscented transform sampling and 5000 Monte Carlo sampling. As can be seen after running the steps according to the invention, the computation time using generalized unscented transform sampling is only 0.61 seconds. Compared to Monte Carlo sampling, the algorithm efficiency is improved by hundreds of times.

[0186] Table 1. Calculation time for sensitivity of different sampling methods

[0187] Sampling methods Calculation time (seconds) Generalized unscented transformation 0.61 Monte Carlo 287.6

[0188] Therefore, this method, by extracting key samples, rapidly solves the sensitivity of electrothermal coupled integrated energy systems, which is of great significance for improving the efficiency of uncertainty analysis of large-scale systems. It also provides important basis for the safe, efficient, and economical operation of multi-energy flow systems.

[0189] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0190] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

[0191] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0192] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0193] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0194] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0195] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform, characterized in that: Includes the following steps: Obtain a probabilistic multi-energy flow model for a comprehensive energy system that considers multiple types of coupled devices; The generalized unscented transform is applied to extract key samples of the integrated energy system, and statistical information of the output variables of the probabilistic multi-energy flow model is obtained based on the key samples. Calculate the first-order sensitivity of the variables in the integrated energy system based on the statistical information of the output variables; The application of generalized unscented transform to extract key samples from the integrated energy system, and the acquisition of statistical information on the output variables of the probabilistic multi-energy flow model based on these key samples, specifically includes: S21. Calculate each input random variable of the integrated energy system. mean vector covariance matrix skewness vector and kurtosis vector ; S22, according to , The location of the Sigma point sampled by the generalized unscented transform method is calculated using a symmetric sampling strategy, and the calculation formula is as follows: ； in, ; Indicates the position coordinates of the center Sigma point; Indicates the first The position coordinates of the Sigma points; Indicates the first The position coordinates of the Sigma points; Indicates the first The degree to which each Sigma point deviates from the mean of the input variable; Indicates the first The degree to which each Sigma point deviates from the mean of the input variable; It is a random variable covariance matrix The matrix obtained after Cholesky decomposition; yes The first of the matrix List; S23, according to and calculate , The possible values ​​of: ； ； in, It represents the Hadamardi (or Hadama) stack; Indicates the first to the second The degree of deviation of each Sigma point from the mean Indicates the first To the The degree of deviation of each Sigma point from the mean; S24, according to , Calculate the weights corresponding to the Sigma points: ； in, ; Representing vectors The first in One element; Representing vectors The first in One element; Representing vectors The 0th element in, where, ; S25, Set each Sigma point Substituting these values ​​into the probabilistic multi-energy flow model, a corresponding output sample point is obtained. : ； S26. Output all sample points Weighted calculations are performed to obtain statistical information about the output variables; ； ； in, This represents the mean of the output variable estimated by the generalized unscented transform. This represents the variance of the output variable estimated by the generalized unscented transformation; The calculation of the first-order sensitivity of the variables of the integrated energy system based on the statistical information of the output variables specifically includes: S32. Target variable for sensitivity analysis based on generalized unscented transformation. The Middle Variables Corresponding normal distribution Perform sampling as the outer sampling in the two-level sampling algorithm: ； in, express Expectations; Represent the target variable The Middle Variables Standard deviation; , , They represent respectively to Three Sigma sample points after unscented transformation sampling; S33. Calculate based on the three extracted Sigma sample points respectively. The conditional expectation vector and conditional covariance matrix: ； ； in, Indicates except The other input variables correspond to a normal distribution. express Expectations; express The covariance matrix, express The covariance matrix; express and The covariance matrix between them; S34, according to Given the conditional expectation vector and conditional covariance matrix, apply the generalized unscented transformation to the normal distribution. Perform sampling as the inner sampling in the two-level sampling algorithm: ； in, Indicates the relationship with the outer sample point The corresponding Sigma point in the inner sample group, Indicates the relationship with the outer sample point The corresponding inner sample group Sigma points, Indicates the relationship with the outer sample point The corresponding inner sample group Sigma points; express The corresponding weights; express The corresponding weights; express The corresponding weights; express The first of the matrix List, Indicates the first The degree to which each Sigma point deviates from the mean of the input variable; Indicates the first The degree to which each Sigma point deviates from the mean of the input variable; express about Conditional distribution variables; Indicates and Random variables that follow the same probability distribution ; S35. Combining the Sigma points of the inner and outer layers and substituting them into the probabilistic multi-energy flow model of the integrated energy system, we obtain... , : ； In the formula, Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform; Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform; Represents the Sigma sample points in normal space Obtained through inverse Gaussian Copula transform; in, express The former One dimension, express The To the There are several dimensions, among which , Choose 0, 1, or 2; S36. Based on the output sample obtained in step S35, calculate the... First-order sensitivity of each variable : ； in, They represent , and The weights corresponding to the Sigma sample points; They represent , and The weights corresponding to the Sigma sample points.

2. The method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform according to claim 1, characterized in that: The acquisition of the probabilistic multi-energy flow model of the integrated energy system considering multiple types of coupled devices specifically includes: Obtain the steady-state power flow model of the heating network system; Obtain the steady-state power flow model of the power grid system; Obtain models of different types of electrothermal coupling devices; Based on the steady-state power flow model of the heating network system, the steady-state power flow model of the power grid system, and the models of different types of electrothermal coupling equipment, a probabilistic multi-energy flow model of the integrated energy system is constructed. The calculation formula for the probabilistic multi-energy flow model of the integrated energy system is as follows: ； in, It represents the uncertainty in the system and serves as the input to the integrated energy system; Representing steady-state power flow models of power grids and heating networks, and models of different types of electrothermal coupling devices; , representing the state variable to be determined in the system, which serves as the output of the integrated energy system; in, Includes the following variables: , , or , Includes the following variables: , Includes the following variables: , Includes the following variables: , Includes the following variables: ; This indicates that the heat source is at the node. The output heat power or pipe The heat power transferred in the middle; , , These represent the heat energy generated by a back-pressure cogeneration unit, an electric furnace, and a heat pump, respectively. , Representing nodes respectively The active and reactive power injected at the point; , Representing nodes respectively The active and reactive power consumed at the point; in, Indicates pipeline Mass flow rate of reclaimed water , Indicates pipeline The temperature of the water in the inlet and outlet pipes, Includes the following variables: , , Includes the following variables: ;in, , Representing nodes respectively ,node Voltage amplitude at the location; Represents a node and The phase angle difference between them.

3. The method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform according to claim 2, characterized in that: The steady-state power flow model calculation formula for the heating network system is as follows: ； ； ； ； in, , Indicates water flowing out and into the pipe The temperature at that time; Indicates ambient temperature; Indicates pipeline Mass flow rate of reclaimed water; Indicates from the pipe Inflow node The mass flow rate of the water; Indicates water from the pipe Inflow node The temperature at that time; Indicates water at the node Temperature after mixing; This indicates that the heat source is at the node. The output heat power or pipe The heat power transferred in the middle; , Indicates pipeline The temperature of the water in the inlet and outlet pipes; This indicates the specific heat capacity of water; This represents the heat transfer coefficient per unit length of the pipe; Indicates the length of the heating network pipes; This represents the resistance coefficient matrix of the pipes in the network; Represents a loop matrix; Represents the set of all pipes; Represents the set of all thermal nodes; Indicates connection with heating network pipelines A set of connected thermal nodes; Indicates the flow of water to the node A collection of pipes.

4. The method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform according to claim 2, characterized in that: The calculation formula for the steady-state power flow model of the power grid system is as follows: ； in, Indicates the number of power nodes; , Representing nodes respectively and The conductance and susceptance of the lines between them; , Representing nodes respectively ,node Voltage amplitude at the location; Represents a node and The phase angle difference between them; , Representing nodes respectively The active and reactive power injected at the point; , Representing nodes respectively The active and reactive power consumed at the point.

5. The method for calculating the global sensitivity of a comprehensive energy system based on generalized unscented transform according to claim 2, characterized in that: The calculation formulas for the different types of electrothermal coupling devices are as follows: ； in, , , These represent the heat energy generated by a back-pressure cogeneration unit, an electric furnace, and a heat pump, respectively. This refers to the electrical energy generated by a back-pressure combined heat and power unit. , These represent the electrical energy consumed by the electric furnace and the heat pump, respectively. This indicates the heat and power ratio of a combined heat and power (CHP) unit in a CHP boiler. , These represent the heat production efficiency of the electric furnace and the heat pump, respectively.

6. A computer-readable storage medium, characterized in that: It stores a computer program, which, when executed by a processor, implements a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transformation as described in any one of claims 1-5.

7. A computer device, characterized in that: include: Memory, used to store instructions; A processor is configured to execute the instructions, causing the computer device to perform the operation of a method for calculating the global sensitivity of an integrated energy system based on generalized unscented transform as described in any one of claims 1-5.

Images