A distribution network power planning method and system considering safety domain and expected loss

Abstract

This disclosure belongs to the field of distribution network planning technology and provides a distribution network power planning method and system that takes into account the safety domain and expected losses. The method includes the following steps: acquiring basic data of the distribution network; constructing a distribution network power planning model that takes into account the safety domain and expected losses based on the acquired data, with the minimum annual comprehensive cost as the objective function; linearizing the nonlinear terms in the objective function and its constraints, taking into account investment and network losses, and transforming the constructed distribution network power planning model into a linear programming model; and iteratively solving the linear programming model to obtain a distribution network power planning scheme.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of distribution network planning, and in particular relates to a distribution network power planning method and system taking into account safety domain and expected loss. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] Due to the shortage of energy in the distribution network, distributed generation (DG) has been built in large quantities in recent years. Distribution network power planning is an important issue in distribution network planning. Distribution network power planning that takes into account both effectiveness and economy can ensure the safe and economic operation of the power system and meet the development needs of the power market. The definition of distribution network power planning is to determine when and where to build what type of power supply to meet the requirements of future annual load growth and power grid development. The results of distribution network power planning directly affect the investment, income, and economy, reliability and stability of future distribution network operation. DG access in the distribution network can improve the efficiency and economy of the power system. However, a high proportion of DG in the distribution network will bring a series of problems to the distribution network, such as bidirectional flow of power flow, increased short-circuit current, three-phase imbalance, harmonic pollution, etc., resulting in a decrease in the system reliability level. Therefore, the power planning problem of the distribution network has become an important part of distribution network planning.

[0004] The safety domain was first proposed in the real-time dispatching problem of power grids. It is defined as the maximum disturbance range that the system can accept, which is usually the disturbance of wind power and load. However, with the increasing number of distributed power sources in the distribution network, the problem of uncertain random disturbances in the distribution network cannot be ignored. Therefore, it is of great significance to consider the safety domain of the system when planning the power supply of the distribution network. However, in the power supply planning problem of the distribution network, the optimal planning cost and operating cost are generally used as the objective function. If the full acceptance of all node disturbances is considered, although the safety is met, the economy will be significantly reduced. Therefore, the random statistical law of node disturbances can be considered, and the disturbances that exceed the acceptance capacity can be punished by calculating the expected loss, so as to compromise the safety and economy in the planning process. Although this makes the model more reasonable, the accurate expected loss cost needs to be calculated in the form of integration. The traditional linear expected loss method will introduce a large number of 0-1 variables, which will affect the calculation efficiency of the planning model. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a distribution network power planning method and system taking into account the safety domain and expected loss, adopts the affine adjustable robust optimization method (AARO) model for processing, and further considers the random characteristics of wind power disturbances, introduces the wind power expected loss index, and adopts a variety of linearization methods to simplify the expression of expected losses. The distribution network power planning model considering the safety domain and expected loss calculation is converted into a linear programming (LP) model, which greatly improves the calculation efficiency of the planning model and realizes the efficient and accurate calculation of the expected loss cost under the premise of considering the safety domain.

[0006] According to some embodiments, a first solution of the present disclosure provides a power distribution network planning method taking into account the safety domain and expected loss, using the following technical solutions:

[0007] A power distribution network planning method taking into account safety domain and expected loss comprises the following steps:

[0008] Obtain basic data of distribution network;

[0009] Based on the acquired data, a distribution network power planning model taking into account the safety domain and expected losses is constructed with the minimum annual comprehensive cost as the objective function;

[0010] Taking investment and network loss into consideration, the nonlinear terms in the objective function and the constraints of the objective function are linearized, and the distribution network power planning model that takes into account the safety domain and expected loss is transformed into a linear planning model.

[0011] The linear programming model is solved by alternate iteration to obtain a planning scheme for the power supply of the distribution network.

[0012] As a further technical limitation, the basic data obtained include line capacity, line resistance and reactance in the distribution network topology, active and reactive loads of each node, and predicted output of renewable energy in the system.

[0013] As a further technical definition, the annual comprehensive cost includes investment cost, operating cost, reserve adjustment cost and expected loss cost.

[0014] Furthermore, piecewise linear programming is used to linearize the expected loss cost.

[0015] As a further technical limitation, the objective function constraints include constraints based on net load prediction values ​​and constraints considering node net load disturbances; the constraints based on net load prediction values ​​include total power balance constraints, line flow constraints, line capacity constraints, node voltage constraints, controllable unit output constraints and standby adjustment constraints; the constraints considering node net load disturbances include effective safety domain range constraints, random parameter fluctuation range constraints, disturbance power balance constraints, line flow constraints, line capacity constraints, node voltage constraints and controllable unit output constraints.

[0016] As a further technical limitation, the random parameters in the constraints are processed by an affine adjustable robust optimization method to obtain a solvable optimization model, that is, a linear programming model.

[0017] Furthermore, the specific process of model transformation is as follows: assuming that there is an affine relationship between the net load disturbance and the unit output adjustment; introducing auxiliary variables to transform the random parameter fluctuation range constraints to obtain the corresponding relationship between the load disturbance and the safety domain; supplementing the reserve response disturbance capacity constraints, when the net load is disturbed within the effective safety domain, it can fully cope with the load disturbance demand; according to the affine adjustable robust optimization method, the inequality constraints with random parameters are processed to obtain a linear programming model.

[0018] As a further technical limitation, the specific process of the alternating iterative solution is: fixing the unit allocation coefficient to solve the planning model, thereby optimizing the unit's predicted output and the net load power disturbance range boundary that each node can accept as a first-level linear programming problem; fixing the unit's predicted output and the acceptable disturbance range boundary, optimizing the unit allocation coefficient, as a second-level linear programming problem, and obtaining a new unit allocation coefficient value; performing alternating iterative solutions to the two-level linear optimization problem, and when the difference in allocation coefficients obtained by the two-level optimization problem is less than 0.001, exiting the loop, ending the iteration and obtaining the optimal solution, thereby achieving joint optimization of the unit allocation coefficient, predicted output and safety domain, and obtaining the final cost of each part and planning scheme.

[0019] According to some embodiments, a second solution of the present disclosure provides a distribution network power planning system taking into account the safety domain and expected loss, which adopts the following technical solutions:

[0020] A power distribution network planning system taking into account safety domain and expected loss, comprising:

[0021] An acquisition module, configured to acquire basic data of the distribution network;

[0022] A construction module is configured to construct a distribution network power planning model taking into account the safety domain and expected loss based on the acquired data and taking the minimization of the annual comprehensive cost as the objective function;

[0023] A conversion module is configured to comprehensively consider investment and network loss, linearize nonlinear terms in the objective function and the constraints of the objective function, and convert the constructed distribution network power planning model taking into account the safety domain and expected loss into a linear planning model;

[0024] The planning module is configured to perform alternate iterative solutions to the linear planning model to obtain a planning scheme for the power supply of the distribution network.

[0025] According to some embodiments, a third solution of the present disclosure provides a computer-readable storage medium, which adopts the following technical solution:

[0026] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps in the distribution network power planning method taking into account the safety domain and expected loss as described in the first aspect of the present disclosure.

[0027] According to some embodiments, a fourth solution of the present disclosure provides an electronic device, which adopts the following technical solution:

[0028] An electronic device comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for power distribution network planning taking into account safety domains and expected losses as described in the first aspect of the present disclosure are implemented.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The present invention constructs a distribution network power planning model that takes into account the safety domain and a simplified calculation method for expected power loss. The expected power loss is usually calculated in the form of an integral, which brings difficulties to the solution of the model. The piecewise linear method is used to linearize the expected power loss, and the problem of introducing 0-1 variables to reduce calculation efficiency in traditional linear methods is solved, making the model establishment more comprehensive, the planning results more comprehensive and reliable, and the solution speed faster, thereby improving the engineering application value of distribution network planning.

[0031] (2) The present invention adopts the AC method to analyze the line flow. Compared with the DC method, it can simultaneously analyze the influence of node voltage and phase angle on the line flow, and simplifies the AC flow to make it linear, which can improve the reliability and calculation efficiency of the planning results.

[0032] (3) The present invention transforms the bilinear optimization problem into the iterative solution of two linear optimization problems, and then adopts a linear solver to solve it, thereby solving the problem of unstable solution of the nonlinear solver and greatly improving the computational efficiency.

[0033] (4) With the continuous access of renewable energy, the uncertainty of disturbances in the distribution network is increasing. Therefore, it is particularly important to consider the system's ability to cope with disturbances in distribution network planning. This paper introduces the concept of safety domain into the distribution network planning problem for the first time, reasonably considers the safety indicators in the planning process, and adopts a compromise optimization plan between safety and economy, which is conducive to the safe and stable operation of the system and takes into account the economy of the planning. It has important practical significance.

[0034] (5) The present invention fully considers the random statistical laws of wind power and load disturbances, and organically integrates the random optimization method with the robust optimization method. While maintaining the statistical superiority of the random optimization results, it retains the advantage of robust optimization in ensuring the absolute satisfaction of the operating constraints within the disturbance set, further improving the effectiveness of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

[0036] Figure 1 is a flow chart of a power distribution network planning method taking into account security domain and expected loss in the first embodiment of the present disclosure;

[0037] Figure 2 is a schematic diagram of the linearization process of the probability density function in the first embodiment of the present disclosure;

[0038] Figure 3 is a schematic diagram of the expected loss cost at the segmentation point in the first embodiment of the present disclosure;

[0039] Figure 4 is a graph of the expected loss cost function after approximate linearization in the first embodiment of the present disclosure;

[0040] Figure 5 is a schematic diagram of an alternating iterative algorithm in Embodiment 1 of the present disclosure;

[0041] Figure 6 It is a structural block diagram of a distribution network power planning system taking into account safety domain and expected loss in the second embodiment of the present disclosure. DETAILED DESCRIPTION

[0042] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0043] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0044] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0045] In the absence of conflict, the embodiments in the present disclosure and the features in the embodiments may be combined with each other.

[0046] Embodiment 1

[0047] Embodiment 1 of the present disclosure introduces a power distribution network planning method taking into account safety domain and expected loss.

[0048] like Figure 1 A power distribution network planning method taking into account safety domain and expected loss is shown, comprising the following steps:

[0049] Construct a distribution network power planning model that takes into account the safety domain and the simplified calculation method of expected power loss, taking the minimum annual comprehensive cost as the optimization goal, and comprehensively considering line investment, unit investment, unit operation, power purchase costs of the upper power grid and network loss costs;

[0050] Analyze and process nonlinear terms in objective functions and constraints, such as expected loss cost, power flow expressions, and branch capacity constraints, and linearize them;

[0051] Through the AARO theory, the random parameters in the constraints are processed and converted into a solvable optimization model;

[0052] Through the alternating iterative algorithm, the LP solver is used to perform alternating iterative solutions to achieve joint optimization of the allocation coefficient and the safety domain, and obtain the planning cost and planning scheme.

[0053] As one or more implementation methods, distribution network power planning is an important issue in the power system, which can be defined as follows: in order to meet the requirements of future annual load growth and power grid development, determine when, where and what type of distributed power generation equipment to build. The results of distribution network planning directly affect the investment, income and future safety, economy and stability of the distribution network operation. Sufficient controllable power sources in the distribution network can improve the efficiency and economy of the power system, and because renewable distributed power sources DG have the characteristics of low operating cost, clean and pollution-free, they are widely used in distribution networks. However, a high proportion of renewable energy will bring a series of disturbance problems to the distribution network, such as bidirectional power flow, increased short-circuit current, three-phase imbalance, harmonic pollution, etc., resulting in a decrease in the system reliability level. Therefore, distribution network power planning that considers the disturbance acceptance capacity of the distribution network is of great significance.

[0054] In the power planning of distribution networks, the power backup cost for dealing with power disturbances is usually ignored. However, the planning model that ignores the backup cost is not sophisticated enough and may underestimate the total cost of power planning, making the planning results too optimistic, thus affecting the optimal solution of the planning problem. Therefore, the concept of safety domain is introduced to consider the system's ability to cope with disturbances in the planning process, taking into account the backup cost and the expected power loss cost of the cost-effectiveness compromise. However, the precise expected power loss cost is usually calculated in the form of integration, and the traditional linearization method introduces a large number of 0-1 variables, which has low calculation efficiency and brings difficulties to the model solution.

[0055] Usually in the expected power loss calculation, the expected power loss in the objective function is the expected loss cost beyond the acceptable disturbance range of the system, which includes the expected load shedding cost and the expected wind abandonment cost. The expected power loss expression is:

[0056]

[0057] Where 365 represents the number of days in a typical year; N T and N I is the set of optimized time period t and node i; c up With c low are the expected load shedding cost coefficient and the expected wind curtailment cost coefficient, in $ / MWh; X maxi,t With X mini,t are the upper and lower limits of the net load disturbance at node i at time period t, respectively, and are given parameters; X upi,t With X lowi,t are the upper and lower limits of the disturbance that the system can tolerate at node i in time period t, and are the variables to be optimized; pr i,t (x i,t) is the probability density function of the net load disturbance corresponding to node i in time period t. In this paper, it is assumed that all node disturbances obey a normal distribution with the net load prediction value as the center value. The expected power loss calculation form is an integral form, which makes the model difficult to solve. Therefore, it is very important to simplify the calculation formula appropriately in the calculation of the model.

[0058] The objective function of the distribution network power planning model considering the safety domain and the simplified calculation method of expected power loss is:

[0059] min(C inv +C ope +C adj +C risk ) (2)

[0060] In the distribution network planning problem, the objective function is usually to minimize the total cost, including the investment cost C inv , operating cost C ope , reserve adjustment cost C adj and the expected loss cost C risk These costs can be explicitly expressed as follows:

[0061]

[0062]

[0063]

[0064]

[0065]

[0066] In the formula, is the construction cost coefficient of the unit capacity unit; is the residual value of the equipment, taking the initial construction cost 5%; is the equipment operation and maintenance cost, taking the initial construction cost 3% of cap g is the optimized construction capacity of unit g; R g Indicates the coefficient of converting the present value of power investment into equivalent annual value; d and y g are the discount rate and economic useful life respectively; N G and They respectively represent the collection of existing generating sets and generating sets to be expanded; They represent the active operation cost coefficient, reactive operation cost coefficient, upward reserve cost coefficient and downward reserve cost coefficient of generator g respectively; are respectively the predicted active output value, reactive output value, upward reserve capacity and downward reserve capacity of generator g in time period t; is the net load forecast value of node i in time period t; and are the maximum upward and downward disturbances of the net load of node i in time period t, respectively; and are the maximum upward and downward disturbances of the net load that node i can accept in time period t. From formula (2) and formula (7), we can know that:

[0067]

[0068] The constraints of the objective function of the distribution network power planning model considering the safety domain and the simplified calculation method of expected power loss include constraints based on the net load forecast value and constraints considering the node net load disturbance.

[0069] Constraints based on net load forecasts include:

[0070] (1) Total power balance constraint

[0071]

[0072] (2) Line flow constraints

[0073]

[0074] Where ref represents the assumed reference node; and It is the active and reactive power flow of the line based on the predicted net load value; and are the corresponding active and reactive power generation load transfer factors in the distribution network; G k,ref With B k,ref Represents the block diagonal matrix of the conductance and susceptance matrices of the system.

[0075] (3) Line capacity constraints

[0076]

[0077] In the formula, is the upper limit of the apparent power of the line. Formula (10) limits the power flow in line ij to not exceed the upper limit of the line capacity.

[0078] (4) Node voltage constraints

[0079]

[0080] In the formula, Represents the node voltage value based on the net load prediction value; V baseis the voltage value at the reference node; V i min With V i max are the upper and lower limits of the node voltage.

[0081] (5) Controllable unit output constraints

[0082]

[0083]

[0084] Among them, formula (12) is the active power output constraint of the unit, is the maximum installed capacity of unit g, and the installed capacity of the unit must be less than the maximum installed capacity of the unit; in formula (13), S g is the apparent power upper limit of unit g, S g =cap g / γ g , γ g is the power factor coefficient. Formula (13) limits the power of unit g to not exceed its apparent power upper limit.

[0085] (6) Reserve adjustment constraints

[0086]

[0087]

[0088] Among them, formula (14) gives the adjustment amount constraints of the upward reserve and the downward reserve, that is, the upward reserve and the downward reserve must satisfy both the capacity constraint and the ramp rate constraint, where g = 1 represents the upper power grid, and ρ is the maximum proportion coefficient of the reserve that can be provided in the upper power grid to the predicted output.

[0089] The constraints considering node net load disturbances include:

[0090] (1) Constraints on the scope of effective security domain

[0091]

[0092] Among them, formula (16) indicates that the security domain must be within the node fluctuation range (parameter) to constitute a valid security domain.

[0093] (2) Random parameter fluctuation range constraints

[0094]

[0095] In the formula, and are the random disturbances of node i at time t, and their disturbance ranges must be within the given upper and lower bounds;i,t is the reactive disturbance coefficient, which indicates the proportional relationship between reactive disturbance and active disturbance at node i in time period t.

[0096] (3) Disturbance power balance constraint

[0097]

[0098] Among them, formula (18) indicates that the total power disturbance of the system is equal to the power adjustment of the unit responding to the disturbance, that is, the disturbance power is balanced.

[0099] (4) Line flow constraints

[0100]

[0101] Among them, formula (19) is the line power flow calculation formula after considering the disturbance, and It is the active and reactive power flow of the line based on the disturbance power.

[0102] (5) Line capacity constraints

[0103]

[0104] in, and are the active and reactive power flows of the line under the disturbance power; formula (20) limits the power flow in the line ij after the disturbance to not exceed the upper limit of the line capacity.

[0105] (6) Node voltage constraints

[0106]

[0107] In the formula, Represents the node voltage value based on the net load prediction value; V base is the voltage value at the reference node; V i min With V i max are the upper and lower limits of the node voltage.

[0108] (7) Controllable unit output constraints

[0109]

[0110]

[0111] Among them, formula (2) is the physical constraint of the unit output after the power disturbance, and formula (23) limits the unit g output after the disturbance to not exceed its apparent power upper limit.

[0112] The expected loss cost before linearization is as follows:

[0113]

[0114] This embodiment provides a new expected loss cost linearization method, and the specific process is as follows:

[0115] Step 1: Assume that the random disturbance of power at the node obeys the normal distribution function: The disturbance domain is divided into 10 segments (segment number s = 10), and the upper disturbance domain segmentation point Lower disturbance domain segmentation point in At the point and The normal distribution function is truncated at , and the probability density value at the segmentation point k is obtained according to the probability density function of the normal distribution. Then, the slope and intercept of each segment of the approximate linearization function are obtained according to the coordinates at the segmentation point, and the probability density approximate piecewise linear function is further obtained: like Figure 2 As shown in the figure, the above perturbation domain is taken as an example to illustrate the above linearization process. The black curve is Pr(x i,t ), the blue curve is the linearized Pr * (x i,t ), the vertical coordinate of the black dot is the probability density function value at the segmentation point.

[0116] Step 2: Use the linearized piecewise function Pr * (x i,t ) approximates the original probability density function Pr(x i,t ), since the system can accept domain is the variable to be optimized, and its range is It is not easy to solve in the integral formula, so we can temporarily set the admissible region at each segment point, that is, to find out the expected load shedding loss of the system when the coordinates of the admissible region coincide with the coordinates of the segment point. The expression is as follows:

[0117]

[0118] In formula (24), the expression is simply The function of the integral upper and lower limits and It only represents the coordinate value at the segmentation point, which is a given parameter. According to formula (24), the expected load shedding loss cost at each segmentation point can be obtained, and its value is shown in the attached figure. Figure 3 As shown in the figure, the horizontal coordinate value of the red point is the segment point value, and the vertical coordinate value is the function value obtained in formula (24).

[0119] Step 3: The expected loss cost value when the system's acceptable region falls on the segmentation point has been obtained. The same method as in step 1 is used according to Figure 3 The coordinates of the segmentation points, and the slope of each segment of the approximate linearization function With intercept Further, we can find the approximate piecewise linear function curve of the expected load shedding loss cost, as shown in the attached figure. Figure 4 The expected load shedding loss cost can be expressed as formula (25):

[0120]

[0121] Similarly, the expected wind curtailment cost can be expressed as formula (26):

[0122]

[0123] in, and To identify the 0-1 variable whether the actual admissible disturbance domain is located on the red line segment s, the supplementary constraint is shown in formula (27):

[0124]

[0125] Among them, formula (27) indicates that the actual acceptable domain can only be taken to as well as One of the passages in the introduction and The positioning function of the actual acceptable domain is realized.

[0126] Step 4: After the first three steps, the expected power loss cost has been converted into a linear expression, but formula (25) and formula (26) contain a large number of 0-1 variables ( and ), when the number of segments increases, the calculation efficiency will be significantly reduced, and the more segments there are, the more accurate the objective function will be. Taking the upward disturbance of net load as an example, if the slope in the cost function formula (25) As the number of segments s increases, the cost function curve for uniform segmentation (such as Figure 3 ), when the admissible domain is in [P 0i,t ,k up10,i,t ] changes from small to large, the expected load shedding loss cost decreases more and more slowly. Since the objective function is to minimize the expected load shedding loss cost, the value of the admissible domain in the optimization process must first satisfy the smaller segment s, so as to achieve the purpose of minimizing the objective function. The same is true when the net load is disturbed downward.

[0127] Based on the above analysis, the concept of segmented security domain can be defined, and the sum of the segmented security domains can be used to replace the system's admissible domain. The 0-1 variable in step 3 is removed, and the additional constraints are shown in formula (28):

[0128]

[0129] Wherein, formula (28) expresses the actual admissible domain as the equation of the sum of the net load and the segmented safety domain, where and Defined as a continuous segmented security domain variable, and The meaning is the same. By adding the above constraints, the cost function formula (25) and formula (26) can be transformed into formula (29) and formula (30):

[0130]

[0131]

[0132] in, and Where is the expected load shedding loss cost and the expected wind curtailment loss cost when the acceptable domain is 0.

[0133] The slope of the following complementary cost function is Proof that it increases monotonically with the increase of the number of segments s:

[0134] Introduce the expected load shedding function:

[0135] Will They are listed below:

[0136]

[0137] From this, the difference in expected load shedding power between each segment point can be calculated:

[0138]

[0139] Since the segments are evenly spaced, The difference in expected load shedding power between each segment point can be calculated as shown in formula (34):

[0140]

[0141] Therefore, it can be deduced The expression is shown in formula (35):

[0142]

[0143] Among them, the probability density function It is always positive in the domain of definition, so the integral part in formula (35) is always positive, and we get for is always negative, just prove It can be monotonically increased with the increase of s. The proof process is further given below.

[0144] Defining functions prove Monotonically increasing is equivalent to proving that the function g up (s) is always greater than 0. From formula (35), it is easy to know that g up (s) is expressed as follows:

[0145]

[0146] G up (s) is split into two functions and

[0147]

[0148]

[0149] Among them, formula (38) is always greater than 0. It is sufficient to prove that formula (37) is always greater than 0. Integrate formula (37) to simplify it:

[0150]

[0151] because In x i,t >0 is monotonically decreasing, so therefore Always established, and Always holds true, so g up (s)>0 is always true, so the conclusion is proved.

[0152] Since there are square terms in formula (10), formula (13), formula (20) and formula (23), the power circle linearization method is used to effectively solve them. The above expressions can be converted into formulas (40)-(43) respectively:

[0153]

[0154]

[0155]

[0156]

[0157] in, and is the coefficient of the linearized power circle, and c is the number of segments of the polygon inscribed in the power circle.

[0158] Model constraints contain random parameters and This embodiment constructs an affine relationship between the unit adjustment amount and the net load disturbance amount, and further adopts the AARO method to transform the constraints. The specific steps are as follows:

[0159] In the first step, it is assumed that there is an affine relationship between the net load disturbance and the unit output adjustment, that is:

[0160]

[0161] In the formula and are the distribution coefficients of the unit to cope with active and reactive load disturbances respectively.

[0162] The second step is to introduce the auxiliary variable z i,t , transform equation (17) into equation (45), thereby constructing the corresponding relationship between load disturbance and safety domain:

[0163]

[0164] Among them, formula (45) indicates that the net load disturbance is within the safety domain and constitutes an effective safety domain.

[0165] The third step is to add the reserve disturbance response capability constraints:

[0166]

[0167] Among them, formula (45) indicates that when the net load is disturbed within the effective safety domain, the system reserve can fully cope with the load disturbance demand.

[0168] The fourth step is to process the inequality constraints with random parameters according to the AARO theory. Taking formula (43) as an example, according to the mapping relationship and auxiliary variables introduced in steps 1 and 2, formula (43) can be transformed into formula (47):

[0169]

[0170] Through the strong duality theory, the optimization problem formula (47) is transformed into formula (48):

[0171]

[0172] Where η i,g,t,cFor the introduced dual auxiliary variables, formula (21) and formula (42) can be transformed by the same method, which will not be described here for simplicity.

[0173] The constraint still contains bilinear terms (the product of the distribution coefficient and the safety domain), and it is still a nonlinear model. Therefore, this embodiment adopts the alternating iteration method to solve, and transforms the original bilinear problem into the iterative solution of two linear optimization problems, thereby realizing the joint optimization of the adjustment coefficient and the safety domain, and can obtain better optimization effect and faster convergence speed.

[0174] like Figure 5 As shown in the figure, the specific process of solving the problem by alternating iteration method is as follows:

[0175] (1) The nonlinear expected loss power in integral form is linearized and the 0-1 variables are removed to obtain the linear expression of the expected loss cost;

[0176] (2) Simplify the AC power flow expression to obtain the linear expression of the active and reactive power flows of the distribution network lines;

[0177] (3) The power circle linearization method is used to linearize the unit output constraint and the line capacity constraint to obtain a linear expression of the branch capacity constraint;

[0178] (4) The AARO method is used to process the random perturbation parameters in the constraints and transform them into an optimization model with a bilinear term (distribution coefficient multiplied by the safety region);

[0179] (5) For the bilinear terms in the model, an alternating iterative heuristic algorithm is adopted to transform the original bilinear problem into two linear optimization problems, thereby transforming the optimization model into a mixed integer linear programming model and solving it using the LP=CPLEX solver.

[0180] This embodiment adopts the affine adjustable robust optimization method (AARO) model for processing, and further considers the random characteristics of wind power disturbances, introduces the wind power expected loss index, and uses a variety of linearization methods to simplify the expression of the expected loss. The distribution network power planning model that considers the safety domain and the expected loss calculation is converted into a linear programming (LP) model, which greatly improves the calculation efficiency of the planning model and realizes the efficient and accurate calculation of the expected loss cost under the premise of considering the safety domain.

[0181] Embodiment 2

[0182] Embodiment 2 of the present disclosure introduces a power distribution network planning system that takes into account safety domains and expected losses.

[0183] like Figure 6 A power distribution network planning system considering security domain and expected loss is shown, comprising:

[0184] An acquisition module, configured to acquire basic data of the distribution network;

[0185] A construction module is configured to construct a distribution network power planning model taking into account the safety domain and expected loss based on the acquired data and taking the minimization of the annual comprehensive cost as the objective function;

[0186] A conversion module is configured to comprehensively consider investment and network loss, linearize nonlinear terms in the objective function and the constraints of the objective function, and convert the constructed distribution network power planning model taking into account the safety domain and expected loss into a linear planning model;

[0187] The planning module is configured to perform alternate iterative solutions to the linear planning model to obtain a planning scheme for the power supply of the distribution network.

[0188] The detailed steps are the same as the distribution network power planning method taking into account the safety domain and expected loss provided in Example 1, and will not be repeated here.

[0189] Embodiment 3

[0190] A third embodiment of the present disclosure provides a computer-readable storage medium.

[0191] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps in the distribution network power planning method taking into account the safety domain and expected loss as described in the first embodiment of the present disclosure.

[0192] The detailed steps are the same as the distribution network power planning method taking into account the safety domain and expected loss provided in Example 1, and will not be repeated here.

[0193] Embodiment 4

[0194] A fourth embodiment of the present disclosure provides an electronic device.

[0195] An electronic device comprises a memory, a processor and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for power distribution network planning taking into account safety domains and expected losses as described in the first embodiment of the present disclosure are implemented.

[0196] The detailed steps are the same as the distribution network power planning method taking into account the safety domain and expected loss provided in Example 1, and will not be repeated here.

[0197] Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Technical personnel in the relevant field should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims

1. A power distribution network planning method taking into account safety domain and expected loss, characterized in that: The following steps are involved: Obtain basic data of distribution network; Based on the acquired data, a distribution network power planning model taking into account the safety domain and expected losses is constructed with the minimum annual comprehensive cost as the objective function; Taking investment and network loss into comprehensive consideration, the nonlinear terms in the objective function and the objective function constraints are linearized, and the constructed distribution network power planning model taking into account the safety domain and expected loss is converted into a linear programming model; wherein the objective function constraints include constraints based on the net load forecast value and constraints considering the node net load disturbance; the constraints based on the net load forecast value include total power balance constraints, line flow constraints, line capacity constraints, node voltage constraints, controllable unit output constraints and standby adjustment constraints; the constraints considering the node net load disturbance include effective safety domain range constraints, random parameter fluctuation range constraints, disturbance power balance constraints, line flow constraints, line capacity constraints, node voltage constraints and controllable unit output constraints; wherein the effective safety domain range constraints refer to that the safety domain must be within the node fluctuation parameter to constitute an effective safety domain; the random parameter fluctuation range constraints refer to that the disturbance range must be within the given upper and lower limits; the disturbance power balance constraint refers to that the total power disturbance of the system is equal to the power adjustment amount of the unit responding to the disturbance; The process of converting into a linear programming model is: processing the random parameters in the constraints by an affine adjustable robust optimization method to obtain a solvable optimization model, that is, a linear programming model, specifically: assuming that there is an affine relationship between the net load disturbance and the unit output adjustment; introducing auxiliary variables to convert the random parameter fluctuation range constraints to obtain the corresponding relationship between the load disturbance and the safety domain; supplementing the reserve response disturbance capacity constraint, when the net load is disturbed within the effective safety domain, it can fully cope with the load disturbance demand; processing the inequality constraints with random parameters according to the affine adjustable robust optimization method to obtain a linear programming model; The linear programming model is solved by alternate iteration to obtain a planning scheme for the power supply of the distribution network.

2. A power distribution network power planning method taking into account safety domain and expected loss as described in claim 1, characterized in that: The annual comprehensive cost includes investment cost, operating cost, reserve adjustment cost and expected loss cost.

3. A method for power distribution network planning taking into account safety domain and expected loss as claimed in claim 2, characterized in that: The expected loss cost is linearized using piecewise linear programming.

4. A power distribution network power planning method taking into account safety domain and expected loss as claimed in claim 1, characterized in that: The specific process of the alternating iterative solution is as follows: the unit allocation coefficient is fixed to solve the planning model, thereby optimizing the predicted output of the unit and the net load power disturbance range boundary that each node can accept as the first-level linear programming problem; the unit predicted output and the acceptable disturbance range boundary are fixed, the unit allocation coefficient is optimized, and a new unit allocation coefficient value is obtained as the second-level linear programming problem; The two-layer linear optimization problem is solved alternately and iteratively. When the difference in the allocation coefficients of the two-layer optimization problem is less than 0.001, the loop is exited and the optimal solution is obtained at the end of the iteration. The joint optimization of the unit allocation coefficient, predicted output and safety domain is achieved to obtain the final cost and planning scheme for each part.

5. A distribution network power planning system taking into account safety domain and expected loss, characterized in that: include: An acquisition module, configured to acquire basic data of the distribution network; A construction module is configured to construct a distribution network power planning model taking into account the safety domain and expected loss based on the acquired data and taking the minimization of the annual comprehensive cost as the objective function; The conversion module is configured to comprehensively consider investment and network loss, linearize the nonlinear terms in the objective function and the objective function constraints, and convert the constructed distribution network power planning model taking into account the safety domain and expected loss into a linear programming model; wherein the objective function constraints include constraints based on the net load forecast value and constraints considering the node net load disturbance; the constraints based on the net load forecast value include total power balance constraints, line flow constraints, line capacity constraints, node voltage constraints, controllable unit output constraints and standby adjustment constraints; the constraints considering the node net load disturbance include effective safety domain range constraints, random parameter fluctuation range constraints, disturbance power balance constraints, line flow constraints, line capacity constraints, node voltage constraints and controllable unit output constraints; wherein the effective safety domain range constraints refer to that the safety domain must be within the node fluctuation parameter to constitute an effective safety domain; the random parameter fluctuation range constraints refer to that the disturbance range must be within the given upper and lower limits; the disturbance power balance constraint refers to that the total power disturbance of the system is equal to the power adjustment amount of the unit responding to the disturbance; The process of converting into a linear programming model is: processing the random parameters in the constraints by an affine adjustable robust optimization method to obtain a solvable optimization model, that is, a linear programming model, specifically: assuming that there is an affine relationship between the net load disturbance and the unit output adjustment; introducing auxiliary variables to convert the random parameter fluctuation range constraints to obtain the corresponding relationship between the load disturbance and the safety domain; supplementing the reserve response disturbance capacity constraint, when the net load is disturbed within the effective safety domain, it can fully cope with the load disturbance demand; processing the inequality constraints with random parameters according to the affine adjustable robust optimization method to obtain a linear programming model; The planning module is configured to perform alternate iterative solutions to the linear planning model to obtain a planning scheme for the power supply of the distribution network.

6. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps of the power distribution network power planning method taking into account the safety domain and expected loss as described in any one of claims 1 to 4 are implemented.

7. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for power distribution network planning taking into account the safety domain and expected loss are implemented as described in any one of claims 1 to 4.

Images