Risk operation method of electricity-gas coupled distribution network considering wind power uncertainty

Abstract

The present invention discloses a risk operation method for an electric-gas coupled distribution network considering the uncertainty of wind power. First, a system model of an electric coupled distribution network is established, including the objective functions and constraints of the electric power system and the natural gas system, to form a deterministic scheduling model of an electric coupled system containing wind power; then, the uncertainty set of wind power is established through the maximum cross-correlation entropy, and the basic model of IGDT based on the maximum cross-correlation entropy is obtained; then, a cost deviation coefficient is given to obtain a risk aversion scheduling and opportunity seeking scheduling model; the model is solved using a CPLEX solver to obtain a risk scheduling result of an electric coupled distribution network considering the uncertainty of wind power. The operation strategy proposed by the present invention can quantify the uncertainty risk of wind power, make scheduling results of the two strategies of opportunity seeking and risk aversion, and provide different scheduling strategies for decision makers with different risk preferences.

Description

Technical Field

[0001] The present invention belongs to the technical field of optimized operation of electric-gas coupled energy distribution network of large-scale distributed wind power grid connection, and specifically relates to a risk operation method of electric-gas coupled distribution network based on maximum cross-correlation entropy improved IGDT. Background Art

[0002] With the rapid development of new energy sources such as wind power, the uncertainty brought by the grid connection of a large number of wind power has a greater risk of affecting the power system, which will cause adverse effects such as wind abandonment and load shedding. The integrated energy system can increase the flexibility of energy consumption and improve the economy and safety of system operation. The electrically coupled distribution network is a specific manifestation of the integrated energy system. It is composed of the power system and the natural gas system. The two systems are deeply coupled by power-to-gas technology and gas turbines. Therefore, the adverse effects of wind power on the power system will further affect the safe and economic operation of the natural gas grid. Therefore, it is urgent to carry out risk operation of electrically coupled distribution networks that take into account the uncertainty of wind power.

[0003] Domestic and foreign scholars have successively carried out some research on the risk operation considering the uncertainty of wind power. It can be mainly divided into two methods. One is based on the conditional risk value theory (CVaR) in finance. This method describes the risk of wind abandonment and load shedding under uncertainty, and adds the risk cost (the cost of dealing with wind abandonment and load shedding) as the target value to the objective function, so that the model can minimize the risk of power outage while minimizing the operating cost. The other is the information gap decision theory (IGDT) in informatics. This method is a non-probabilistic and non-fuzzy mathematical optimization method that effectively handles Knight's uncertainty. Information refers to the uncertain information that affects the objective function, and the gap is the error between the predicted value and the actual value of the uncertainty factor. Among them, the description of the information gap mainly includes envelope constraints, ellipsoid constraints and other methods, but in fact, electrical quantities have the characteristics of non-Gaussianity. The maximum cross-correlation entropy is a theory that effectively describes non-Gaussianity. It is a good method to use it to describe the information gap to establish an uncertainty set, and then establish an electrical coupling distribution network risk operation model based on IGDT. Summary of the invention

[0004] The purpose of the present invention is to provide a risk operation method for an electric-gas coupled distribution network taking into account the uncertainty of wind power, thereby solving the problem in the prior art that wind power resources cannot be fully utilized.

[0005] The technical solution adopted by the present invention is: a distributed optimization method for an electric-gas coupled distribution network considering the uncertainty of wind power, and the specific operation steps are as follows:

[0006] Step 1: Establish a deterministic dispatch model for the electrical coupled distribution network, wherein the deterministic dispatch model includes objective functions and constraints of the power system and the natural gas system;

[0007] Step 2: Establish the uncertainty set of wind power through the maximum cross-correlation entropy, and obtain the basic model of improved IGDT based on the maximum cross-correlation entropy;

[0008] Step 3: Based on the basic model in step 2, the cost deviation coefficients are given respectively, and two risk-averse scheduling and opportunity-seeking scheduling models considering wind power uncertainty are established;

[0009] Step 4: Use the CPLEX solver to solve the risk-averse scheduling and opportunity-seeking scheduling models to obtain the risk scheduling results of the electrically coupled distribution network considering the uncertainty of wind power.

[0010] The present invention is also characterized in that:

[0011] Step 1 mainly includes the mathematical models of power-to-gas, gas turbines and gas storage tanks and related safety operation constraints. The objective functions of the distribution network and gas distribution network are:

[0012]

[0013]

[0014] In formula (1), C e,t and C g is the cost coefficient of electricity purchase and gas turbine gas purchase; T is the total dispatch period; P t b,e To purchase electric power from the main grid; is the gas purchasing power from the gas grid; N G is the total number of gas turbines; p is the total number of P2G; and C c is the daily operating cost of P2G and the wind curtailment cost coefficient; For abandoned wind power; The revenue coefficient of P2G selling gas to the gas grid; is the amount of natural gas sold to the gas grid; The penalty cost for shedding load; P t s,e is the load shedding capacity; in formula (2), C well is the cost coefficient of purchasing gas from natural gas wells; P t b,g is the gas purchase volume from natural gas wells; C s The penalty cost for cutting gas; P t s,g is the gas cutting load.

[0015] In step 2, the wind power uncertainty set is described by the maximum cross-correlation entropy as follows:

[0016]

[0017] In the formula, U represents the fluctuation range of wind power output, represents the predicted value of wind power output, α represents the deviation coefficient of the predicted value of wind power output, represents the wind power output value, σ represents the kernel width of the kernel function in the cross-correlation entropy criterion;

[0018] Furthermore, in step 2, the basic model of IGDT based on maximum cross-correlation entropy is obtained:

[0019]

[0020] In the formula, f(x,y) is the objective function, x is the decision variable, y is the input parameter, m and n are the equality and inequality constraints respectively;

[0021] Therefore, the basic model of the improved IGDT is determined, and its solution is set to be x0.

[0022] The risk-averse scheduling model and opportunity-seeking scheduling model in step 3 are as follows:

[0023] The risk-averse scheduling model can be expressed as:

[0024]

[0025] In the formula, μ r is the robust participation factor, set by the user; α r is the risk uncertainty;

[0026] The opportunity seeking scheduling model can be expressed as:

[0027]

[0028] Among them, α o is the chance uncertainty.

[0029] In step 4, the CPLEX solver is used to first solve the established deterministic scheduling model to obtain the deterministic scheduling results of the electrical coupling distribution network containing wind power; further, the deterministic scheduling results are used as parameters of the risk-averse scheduling and opportunity-seeking scheduling models to solve the risk-averse scheduling and opportunity-seeking scheduling models, thereby obtaining the risk scheduling results of the electrical coupling distribution network considering the uncertainty of wind power.

[0030] The beneficial effects of the present invention are as follows: the IGDT model improved based on the maximum cross-correlation entropy proposed in the present invention can effectively combine the IES scheduling operation cost with the deviation coefficient of the uncertain parameters quantitatively, and provide quantitative decision-making basis for decision makers. At the same time, unlike the previous uncertainty set construction, a large number of physical observation phenomena in actual electrical engineering show non-Gaussian behavior, such as wind power generation, AC and DC loads and electric vehicle charging, environmental noise, communication instability, etc. Under such behavior, it will no longer be appropriate to use the mean square error based on the Gaussian assumption as the cost function. The uncertainty set established by the present invention takes into account the characteristics of non-Gaussian data in practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of the risk operation method of the electric-gas coupled distribution network of the invention based on the maximum cross-correlation entropy improvement of IGDT.

[0032] Figure 2 The figure is a schematic diagram of a distribution network system in which the method of the present invention is applied to an example.

[0033] Figure 3 It is a curve diagram of electric load and gas load in the example system of the method of the present invention.

[0034] Figure 4 It is a graph of electricity price and gas price in the example system of the method of the present invention. DETAILED DESCRIPTION

[0035] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] The process of the distributed optimization method of the electric-gas coupled distribution network considering the uncertainty of wind power in the present invention is as follows: Figure 1 The specific steps are as follows:

[0037] Step 1: Establish a deterministic dispatch model for the electrical coupled distribution network, wherein the deterministic dispatch model includes objective functions and constraints of the power system and the natural gas system;

[0038] Step 2: Establish the uncertainty set of wind power through the maximum cross-correlation entropy, and obtain the basic model of improved IGDT based on the maximum cross-correlation entropy;

[0039] Step 3: Based on the basic model in step 2, the cost deviation coefficients are given respectively, and two risk-averse scheduling and opportunity-seeking scheduling models considering wind power uncertainty are established;

[0040] Step 4: Use the CPLEX solver to solve the risk-averse scheduling and opportunity-seeking scheduling models to obtain the risk scheduling results of the electrically coupled distribution network considering the uncertainty of wind power.

[0041] Step 1 mainly includes the mathematical models of power-to-gas, gas turbines and gas storage tanks and related safety operation constraints. The optimization objectives of the distribution network and gas distribution network are:

[0042]

[0043]

[0044] In formula (1), C e,t and C g is the cost coefficient of electricity purchase and gas turbine gas purchase; T is the total dispatch period; P t b,e is the power purchased from the main grid; is the gas purchasing power from the gas grid; N G is the total number of gas turbines; p is the total number of P2G; and C c is the daily operating cost of P2G and the wind curtailment cost coefficient; For abandoned wind power; The revenue coefficient of P2G selling gas to the gas grid; is the amount of natural gas sold to the gas grid; The penalty cost for shedding load; P t s,e is the load shedding capacity; in formula (2), C well is the cost coefficient of purchasing gas from natural gas wells; P t b,g is the gas purchase volume from natural gas wells; C s The penalty cost for cutting gas; P t s,g is the gas cutting load.

[0045] Furthermore, in step 1, the obtained distribution network, gas distribution network and other constraints include:

[0046] (1) Distribution network constraints:

[0047]

[0048] In the formula, is the active power output of fan w, is the active power output of gas turbine g; π(j) represents the set of branch head nodes with j as the terminal node in the distribution network, ω(j) represents the set of branch terminal nodes with j as the head node in the distribution network; P ij and Q ij is the active and reactive power of the ij section of the distribution network line; express and Indicates the predicted active and reactive electrical load values; represents the gas cut-off load at the end node of the branch with j as the head node; P jk，t and Q jk，t It represents the active and reactive power of the branch jk segment with j as the head node in the distribution network; and Indicates wind curtailment and gas curtailment load; U jt represents the voltage of node j, I ij，t Represents the current of the ij section of the distribution line; R ij and X ij Indicates the resistance and reactance values ​​of the ij section of the distribution line; and Indicates the upper and lower limits of the gas turbine ramp rate. and Indicates the upper and lower limits of fan climbing; and Indicates the upper and lower limits of gas turbine output, and Indicates the upper and lower limits of fan output; and Indicates the minimum start-up and shutdown time of unit g; I g (t) is a binary integer variable representing the start and stop status of unit g in time period t, with a value of 1 for start and 0 for stop; and Indicates the start / stop duration.

[0049] (2) Gas distribution network constraints:

[0050]

[0051] In the formula, C pq represents the natural gas pipeline constant; V(p) and U(p) represent the set of natural gas pipelines with node p as the first node and the last node; ω q，t represents the natural gas flow injected into node p, ω pq，t represents the pipeline flow rate of the natural gas pipeline pq, ω pr，t represents the pipeline flow of natural gas pipeline qr; p, q, r are natural gas node indexes; θ p，t represents the node pressure; θ p，min With θ p，max Represents the upper and lower limit constraints of node air pressure; ω pq，min With ω pq，max Represents the upper and lower limit constraints of pipeline flow.

[0052] (3) Other constraints:

[0053]

[0054] In the formula, and Indicates the minimum and maximum active power exchanged between the distribution network and the upper main network; and Indicates the minimum and maximum active power exchanged between the distribution grid and the natural gas supplier. Indicates the volume of natural gas converted by the power-to-gas equipment; Indicates the actual electrical power consumed by the power-to-gas equipment; η P2G It indicates the conversion efficiency of power-to-gas equipment; HHV indicates the higher heating value of natural gas; Indicates the maximum conversion power of the power-to-gas device; I P2G Indicates the working status of the power-to-gas device.

[0055] In step 2, the wind power uncertainty set is described by the maximum cross-correlation entropy as follows:

[0056]

[0057] In the formula, U represents the fluctuation range of wind power output, represents the predicted value of wind power output, α represents the deviation coefficient of the predicted value of wind power output, represents the wind power output value, and σ represents the kernel width of the kernel function in the cross-correlation entropy criterion.

[0058] Furthermore, in step 2, the basic model of IGDT based on maximum cross-correlation entropy is obtained:

[0059]

[0060] Where f(x,y) is the objective function, x is the decision variable, y is the input parameter, m and n are equality and inequality constraints, respectively.

[0061] Therefore, a deterministic model is established, and its solution is assumed to be x0.

[0062] In step 3, the cost deviation coefficient is formulated according to the decision maker's preference. The uncertainty of wind power may be harmful or beneficial. These two possibilities are quantified, and a robust (risk avoidance) model and an opportunity (opportunity profit) model are established on the basic model.

[0063] The risk-averse scheduling model can be expressed as:

[0064]

[0065] In the formula, μ r is the robust participation factor, set by the user, α r The uncertainty of risk.

[0066] The opportunity seeking scheduling model can be expressed as:

[0067]

[0068] Among them, α o is the chance uncertainty.

[0069] In step 4, the CPLEX solver is used to first solve the established deterministic scheduling model to obtain the deterministic scheduling results of the electrical coupling distribution network containing wind power; further, the deterministic scheduling results are used as parameters of the risk-averse scheduling and opportunity-seeking scheduling models to solve the risk-averse scheduling and opportunity-seeking scheduling models, thereby obtaining the risk scheduling results of the electrical coupling distribution network considering the uncertainty of wind power.

[0070] Example

[0071] The beneficial effects of the present invention are described below by means of specific embodiments.

[0072] In order to illustrate the system topology of the proposed model method, the power-gas coupled energy system structure consisting of IEEE 33-node distribution network and 24-node gas distribution network is shown in the attached figure. Figure 2 As shown in the figure, it includes two gas turbines G1 and G2, two wind turbines W1 and W2, two power-to-gas devices P2G1 and P2G2, and the P2G device is connected to the system as a supporting device for the wind turbines, as well as two gas storage tanks S1 and S2. The system electrical load and gas load are as follows Figure 2 , time-of-use electricity and gas prices Figure 3 . The maximum output of the gas source is 1600m3 / h, the P2G gas price is 0.98 times the gas source price, the maximum gas storage capacity of the gas tank is 2500m3 and 2500m3, the minimum gas storage capacity is 200m3 and 250m3, the wind turbine capacity is 2MW and 2.5MW, and the gas turbine rated power is 1MW and 1.2MW. Node 1 of the distribution network is the power purchase node, and node 1 of the natural gas system is the gas purchase node. The total scheduling cycle is 24 hours, with 1 hour as an optimization interval. This example is carried out in the MATLAB2016a software environment.

[0073] Through the model proposed in the present invention, a comparative analysis is conducted on the deterministic dispatch including wind power and the risk operation considering the uncertainty of wind power.

[0074] The deterministic scheduling results are shown in the following table:

[0075] Table 1 Deterministic scheduling results

[0076]

[0077]

Claims

1. A risk operation method for an electricity-gas coupled distribution network considering the uncertainty of wind power, characterized in that: The specific operation process includes the following steps: Step 1: Establish a deterministic dispatch model for the electrical coupled distribution network, wherein the deterministic dispatch model includes objective functions and constraints of the power system and the natural gas system; Step 2: Establish the uncertainty set of wind power through the maximum cross-correlation entropy, and obtain the basic model of improved IGDT based on the maximum cross-correlation entropy; In step 2, the wind power uncertainty set is described by the maximum cross-correlation entropy as follows: In the formula, U represents the fluctuation range of wind power output, represents the predicted value of wind power output, α represents the deviation coefficient of the predicted value of wind power output, represents the wind power output value, σ represents the kernel width of the kernel function in the cross-correlation entropy criterion; Furthermore, in step 2, the basic model of IGDT based on maximum cross-correlation entropy is obtained: In the formula, f(x,y) is the objective function, x is the decision variable, y is the input parameter, m and n are the equality and inequality constraints respectively; Therefore, the basic model of the improved IGDT is determined, and its solution is x0 Step 3: Based on the basic model in step 2, the cost deviation coefficients are given respectively, and two risk-averse scheduling and opportunity-seeking scheduling models considering wind power uncertainty are established; Step 4: Use the CPLEX solver to solve the risk-averse scheduling and opportunity-seeking scheduling models to obtain the risk scheduling results of the electrically coupled distribution network considering the uncertainty of wind power.

2. The risk operation method of the electricity-gas coupled distribution network considering the uncertainty of wind power according to claim 1 is characterized in that: Step 1 mainly includes the mathematical models of power-to-gas, gas turbines and gas storage tanks and related safety operation constraints. The objective functions of the distribution network and gas distribution network are: In formula (1), C e and C g Cost coefficients for purchasing electricity and gas for gas turbines; T is the total dispatch period; P t b,e To purchase electric power from the main grid; is the gas purchasing power from the gas grid; N G is the total number of gas turbines; p is the total number of P2G; and C c is the daily operating cost of P2G and the wind curtailment cost coefficient; For abandoned wind power; The revenue coefficient of P2G selling gas to the gas grid; is the amount of natural gas sold to the gas grid; The penalty cost for shedding load; P t s,e is the load shedding capacity; in formula (2), C well is the gas purchase cost coefficient from the natural gas well; P t b,g is the gas purchase volume from natural gas wells; C s The penalty cost for cutting gas; P t s,g The gas cutting load.

3. The risk operation method of the electric-gas coupled distribution network considering the uncertainty of wind power according to claim 2 is characterized in that: The risk-averse scheduling model and opportunity-seeking scheduling model in step 3 are as follows: The risk-averse scheduling model can be expressed as: In the formula, μ r is the robust participation factor, set by the user; α r is the risk uncertainty; The opportunity seeking scheduling model can be expressed as: Among them, α o is the chance uncertainty.

4. The risk operation method of the electricity-gas coupled distribution network considering the uncertainty of wind power according to claim 1 is characterized in that: In step 4, the established deterministic scheduling model is first solved using the CPLEX solver to obtain the deterministic scheduling results of the electrical coupling distribution network containing wind power; further, the deterministic scheduling results are used as parameters of the risk-averse scheduling and opportunity-seeking scheduling models to solve the risk-averse scheduling and opportunity-seeking scheduling models, thereby obtaining the risk scheduling results of the electrical coupling distribution network considering the uncertainty of wind power.

Images