Power distribution network reactive power optimization control method and system based on ISA-HELM

Abstract

The invention discloses a power distribution network reactive power optimization control method and system based on an ISA-HELM, and belongs to the field of power distribution network reactive power optimization control. Comprising the following steps: constructing a network loss minimization mathematical model which comprises a network loss minimization objective function and a network loss minimization constraint condition; constructing a voltage quality optimization model, wherein the voltage quality optimization model comprises a voltage quality optimization objective function and a voltage quality optimization constraint condition; constructing a power distribution network reactive voltage control optimization model based on the game theory, wherein the power distribution network reactive voltage control optimization model comprises a reactive voltage control optimization objective function and a reactive voltage control optimization constraint condition; performing weight coefficient calculation and weight distribution on the reactive voltage control optimization objective function and the voltage quality optimization objective function based on a mean value adaptive method; solving the reactive voltage control optimization objective function and the voltage quality optimization objective function after weight distribution based on an improved sine and cosine ISCA algorithm; and solving the power flow of the power distribution network based on an all-pure embedded power flow calculation method HELM. The power distribution network loss can be effectively minimized, and the power supply voltage quality of the power distribution network can be improved.

Description

Technical Field

[0001] The present invention relates to a reactive power optimization control method and system for a distribution network based on ISCA-HELM, belonging to the field of reactive power optimization control of a distribution network. Background Art

[0002] The power industry is one of the key industries for reducing energy consumption and improving energy utilization efficiency. The application of reactive power optimization control technology can effectively reduce the energy loss of the power network, thereby reducing carbon emissions and primary energy consumption, and contributing to the achievement of the energy conservation and carbon reduction goals.

[0003] With the development of smart grid technology, the degree of automation of the power system has been improved, and the research and application of reactive power optimization control technology for distribution networks have become increasingly widespread. Reactive power optimization control is closely related to voltage. Too low or too high voltage will affect the operation of the distribution network. If the voltage of each node in the distribution network is too low, it will lead to insufficient output of equipment, and in severe cases, production accidents, personal accidents, etc. may occur, and it may also cause the collapse of the system voltage; if the voltage in the system is too high, when the operating voltage exceeds the withstand voltage of the equipment insulation, overvoltage will cause damage to the equipment insulation. The reactive power optimization control technology for distribution networks can reduce the power loss of the distribution network, improve the operation economy of the distribution network, and ensure good power quality. Therefore, it is necessary to perform real-time control of the system voltage, and improving the voltage quality as much as possible helps to ensure the stability of the power grid system. At present, adjusting the reactive power transmission of the system is an effective way to improve the voltage level. In addition, currently, new reactive power compensation devices are widely used in the distribution network, gradually breaking the traditional reactive power voltage optimization control method. The traditional reactive power voltage control relies on the Newton-Raphson method for the power flow calculation results. However, due to the low calculation accuracy of the Newton-Raphson method, the voltage deviation of each node in the distribution network is relatively large. Excessive voltage deviation will affect the safety and stability of the distribution network, and make the calculated reactive power compensation device capacity inaccurate. When performing reactive power optimization control, this voltage deviation makes the voltage of each node unable to return to the rated value. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a reactive power optimization control method for a distribution network based on ISCA-HELM, which can provide more accurate power flow calculation results, so as to provide more reliable data support for reactive power optimization control. By improving the sine-cosine algorithm to solve the optimized objective function of reactive power voltage control and the optimized objective function of voltage quality after weight distribution, the results of the optimized objective function of reactive power voltage control and the optimized objective function of voltage quality are more accurate. In addition, the power flow calculation results based on the holomorphic embedding algorithm also significantly improve the calculation accuracy. According to these accurate power flow calculation results for reactive power optimization control, not only the network loss and voltage deviation are greatly reduced, but also the operation stability and economy of the distribution network are improved.

[0005] The technical solution adopted by the present invention is as follows: A reactive power optimization control method for a distribution network based on ISCA-HELM, comprising the following steps:

[0006] Step1. Construct a mathematical model for minimizing network loss, including an objective function for minimizing network loss and constraints for minimizing network loss;

[0007] Step2. Construct an optimization model for voltage quality, including an objective function for optimizing voltage quality and constraints for optimizing voltage quality;

[0008] Step3. Construct an optimization model for reactive power and voltage control in a distribution network based on game theory, including an objective function for optimizing reactive power and voltage control and constraints for optimizing reactive power and voltage control;

[0009] Step4. Calculate the weight coefficients for the objective function of optimizing reactive power and voltage control and the objective function of optimizing voltage quality based on the mean adaptive method, and then perform weight allocation;

[0010] Step5. Solve the objective function of optimizing reactive power and voltage control and the objective function of optimizing voltage quality after weight allocation based on the improved sine-cosine ISCA algorithm;

[0011] Step6. Solve the power flow of the distribution network based on the holomorphic embedded power flow calculation method HELM.

[0012] The specific content of Step1 includes: First, considering the power loss generated on the transmission line, construct an objective function for minimizing network loss; Second, set the decision variable reactive power compensation device; Finally, construct constraints for minimizing network loss.

[0013] The specific content of Step2 includes: First, construct a function for minimizing the sum of squares of node voltage deviations as the objective function for optimizing voltage quality without weight allocation; Second, set the decision variable, and flexibly adjust the reactive power output according to the grid demand to maintain the voltage stability of the distribution network; Finally, construct constraints for optimizing voltage quality.

[0014] In Step3:

[0015] The objective function for optimizing reactive power and voltage control is:

[0016]

[0017] In the formula, F is the network loss of the distribution network; ΔP is the active network loss in the distribution network; U imax , U imin respectively represent the upper and lower limit values of the voltage allowed for system node i; λ u is the over-limit penalty coefficient, and U ilim represents the upper and lower limits of the voltage amplitude U i of node i, and its calculation method is:

[0018]

[0019] When the logical value λ u is 0, there is no problem of node voltage over-limit. When the logical value λ u is 1, there is a problem of node voltage over-limit;

[0020] The reactive power voltage control optimization constraint conditions include: the power balance equation constraint of the distribution network, the voltage amplitude of each node in the distribution network, the reactive power compensation capacity, and the inequality constraint of the tap of the on-load tap-changing transformer.

[0021] Step4 specifically includes: obtaining the weight coefficients of the reactive power voltage control optimization objective function and the voltage quality optimization objective function through the mean adaptive method, that is, dividing each objective function by the total number of objective functions, and adding the calculated weight coefficients to the reactive power voltage control optimization objective function and the voltage quality optimization objective function.

[0022] Step5 specifically includes: inputting the distributed parameters and load prediction values of each distribution network node; processing the reactive power voltage control optimization constraint conditions: adding the penalty function and the power flow equation constraint to the reactive power voltage control optimization objective function; initializing the parameters of the improved sine-cosine algorithm to obtain the fitness of the reactive power voltage control optimization objective function and the voltage quality optimization objective function; adding the weight coefficients of the reactive power voltage control optimization objective function and the voltage quality optimization objective function; introducing the Levy flight strategy; calculating the values of the reactive power voltage control optimization objective function and the voltage quality optimization objective function as fitness, recording the optimal individual and the optimal solution, and judging the iteration termination condition; outputting the optimal reactive power control strategy.

[0023] Step6 specifically includes: setting the iterative convergence threshold ε of the holomorphic embedded power flow calculation; constructing a holomorphic embedded power flow calculation model and performing the H-th power flow calculation using the holomorphic embedded power flow calculation method; calculating the difference between the voltage phasor obtained from the H-th power flow calculation and the voltage phasor obtained from the (H + 1)-th power flow calculation; judging whether the accuracy requirement is met. If the accuracy requirement is met, output the final power flow calculation result. If the accuracy requirement is not met, re-use the holomorphic embedded power flow calculation method to perform the H-th power flow calculation and perform another iteration until the accuracy requirement is met and the calculation result is output.

[0024] A distribution network reactive power optimization control system based on ISCA-HELM includes:

[0025] The first construction module is used to construct a network loss minimization model, including a network loss minimization objective function and network loss minimization constraint conditions;

[0026] A second construction module for constructing a voltage quality optimization model, including a voltage quality optimization objective function and voltage quality optimization constraint conditions;

[0027] A third construction module for constructing a reactive power voltage control optimization model based on game theory, including a reactive power voltage control optimization objective function and reactive power voltage control optimization constraint conditions;

[0028] A solution module for calculating weight coefficients of the reactive power voltage control optimization objective function and the voltage quality optimization objective function based on the mean adaptive method, and then performing weight allocation; solving the reactive power voltage control optimization objective function and the voltage quality optimization objective function after weight allocation based on the improved sine-cosine ISCA algorithm; solving the distribution network power flow based on the holomorphic embedded power flow calculation method HELM.

[0029] A processor for running a program, wherein when the program runs, it executes the above-mentioned distribution network reactive power optimization control method based on ISCA-HELM.

[0030] A computer-readable storage medium including a stored program, wherein when the program runs, it controls the device where the computer-readable storage medium is located to execute the above-mentioned distribution network reactive power optimization control method based on ISCA-HELM.

[0031] The beneficial effects of the present invention are as follows: The reactive power voltage control optimization model and voltage quality optimization model of the distribution network based on ISCA-HELM of the present invention calculate the distribution network power flow based on the holomorphic embedding method and determine the power flow of each branch of the network, and at the same time formulate corresponding reactive power compensation measures according to the constraint conditions of each node, and are applied to the overall optimization management of the system; the reactive power voltage control optimization model and voltage quality optimization model of the distribution network based on ISCA-HELM can effectively realize the reactive power optimization control of the distribution network; finally, based on the ISCA-HELM algorithm, the complex non-linear models of reactive power voltage control optimization and voltage quality optimization are solved, and the reactive power optimization control of the distribution network can be effectively realized. Description of the Drawings

[0032] Figure 1 is a block diagram of the method of the present invention;

[0033] Figure 2 is a flow chart of the solution of the improved sine-cosine ISCA algorithm model;

[0034] Figure 3 is a flow chart of the solution of the holomorphic embedded power flow calculation method HELM for solving the distribution network power flow;

[0035] Figure 4 is a block diagram of the device of the present invention. Detailed Embodiments

[0036] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, but the content of the present invention is not limited to the described scope.

[0037] Embodiment 1: As Figures 1-4 shown, a reactive power optimization control method for a distribution network based on ISCA-HELM includes the following steps:

[0038] Step1. Construct a mathematical model for minimizing network loss, including an objective function for minimizing network loss and constraint conditions for minimizing network loss;

[0039] Step2. Construct a voltage quality optimization model, including an objective function for optimizing voltage quality and constraint conditions for optimizing voltage quality;

[0040] Step3. Construct an optimization model for reactive power and voltage control in a distribution network based on game theory, including an objective function for optimizing reactive power and voltage control and constraint conditions for optimizing reactive power and voltage control;

[0041] Step4. Calculate the weight coefficients for the objective function of optimizing reactive power and voltage control and the objective function of optimizing voltage quality based on the mean adaptive method, and then perform weight allocation;

[0042] Step5. Solve the objective function of optimizing reactive power and voltage control and the objective function of optimizing voltage quality after weight allocation based on the improved sine-cosine algorithm (ISCA);

[0043] Step6. Solve the power flow of the distribution network based on the holomorphic embedded power flow calculation method (HELM).

[0044] To make the technical means, creative features, achieved objectives and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments:

[0045] I. Construct a model for minimizing network loss.

[0046] First, considering the power loss generated on the transmission line, construct an objective function for minimizing network loss; secondly, set the decision variable reactive power compensation equipment; finally, construct constraint conditions for minimizing network loss.

[0047] By constructing a model for minimizing network loss, the economy and reliability of the power system can be effectively improved, and unnecessary energy losses can be reduced. The construction of the model needs to comprehensively consider multiple factors, including the physical characteristics of the system, operation limitations and market demands, etc. Network loss (or transmission loss) refers to the energy loss caused by resistance and other factors during the power transmission process. The goal of minimizing network loss is to optimize power transmission, reduce energy losses, improve the economy and reliability of the system. The network loss of the transmission line is defined as follows:

[0048] Definition of network loss:

[0049] P loss is the total transmission loss of the power network, R l is the resistance of the l-th transmission line, I l is the current passing through the transmission line l, and G is the number of transmission lines in the distribution network.

[0050] The mathematical model for minimizing the distribution network power loss is as follows:

[0051] Objective function for minimizing power loss:

[0052] In the formula: Z is the line power loss.

[0053] (2) Constraints for minimizing power loss: P G -P load =P loss Active power balance constraint of the distribution network, |I l | ≤ I max,l Current limit of the transmission line, V min ≤ V j ≤ V max Voltage limit of each node in the distribution network, P min,j ≤ P j ≤ P max,j Active power constraint of each node in the distribution network, Q min,j ≤ Q j ≤ Q max,j Reactive power constraint of each node in the distribution network.

[0054] In the formula: I max,l is the maximum current that the l-th transmission line can withstand; V min , V max are the lower and upper limits of the voltage at node j respectively; P min,j , P max,j are the upper and lower limits of the active power at node j respectively; Q min,j , Q max,j are the lower and upper limits of the reactive power at node j respectively; P G is the total active power of the power sources in the distribution network, P load is the active load of the distribution network, I max,l is the maximum value of the limited current of the transmission line l, V j is the voltage amplitude at node j, P j is the active power at node j.

[0055] II. Construct a voltage quality optimization model.

[0056] First, construct the function that minimizes the sum of squared node voltage deviations as the voltage quality optimization objective function without weight allocation; second, set decision variables to flexibly adjust the reactive power output according to grid demands and maintain the voltage stability of the distribution network; finally, construct the voltage quality optimization constraint conditions.

[0057] Constructing a voltage quality optimization model is an important part of power system analysis, aiming to optimize the voltage quality of the power system to ensure the reliability and stability of power supply. Voltage quality involves multiple aspects, including the amplitude, frequency, waveform distortion, etc. of the voltage.

[0058] The mathematical model for optimizing the voltage quality of the distribution network is as follows:

[0059] (1) Voltage quality optimization objective function:

[0060] U is the voltage deviation value of the distribution network, V jN is the rated voltage of node j, H k is the voltage distortion index related to harmonics (usually can be represented by the effective value of harmonic current or voltage), M is the number of nodes, and X is the number of harmonics.

[0061] (2) Voltage quality optimization constraint conditions: P G -P load =P loss Active power balance constraint of the distribution network, V min ≤V j ≤V max Node voltage limit, P min,j ≤P j ≤P max,j Node active power constraint, Q min,j ≤Q j ≤Q max,j Reactive power constraint of each node in the distribution network, H k ≤H max The voltage distortion of each harmonic cannot exceed the set maximum value, |I i |≤I max,i .

[0062] I max,i is the maximum value of the current limit of transmission line i, |I i | is the current value on transmission line i.

[0063] III. Construct an optimization model for reactive power and voltage control of the distribution network based on game theory.

[0064] In the reactive power optimization of the distribution network, the present invention considers that there is a game relationship between minimizing network losses and optimizing voltage quality. In the actual operation of the power system, in order to reduce network losses, some measures may be taken to change the reactive power distribution, but this may have a certain impact on voltage quality.

[0065] By constructing a game model, the present invention finds a balance strategy between optimizing network losses and voltage quality, so that each objective reaches a relatively optimal state under the condition of mutual competition. When constructing the voltage quality optimization model and the reactive power and voltage control optimization model of the distribution network based on game theory, the maximum economic benefit generated by power supply is taken as the objective function, and the level of system network losses is taken as an important reference index to measure the rationality of the optimization result. Therefore, when realizing the maximum economic benefit, it is required that the network losses of the distribution network system are the least. During the research process, it is also necessary to consider whether other electrical parameters in the system exceed the limits, etc. Therefore, the voltage control of each node in the system is incorporated into the objective function. To sum up, the reactive power and voltage control optimization objective function of the reactive power and voltage control optimization model of the distribution network proposed by the present invention is as follows:

[0066]

[0067] In the formula, F is the network loss of the distribution network; ΔP is the active network loss in the distribution network; U imax 、U imin respectively represent the upper limit value and the lower limit value of the voltage allowed at system node i;

[0068] On the basis of the traditional optimization control method, the present invention adds an over-limit penalty coefficient λ u of the node voltage of the distribution network. U ilim represents the upper and lower limits of the voltage amplitude U i of node i, and its calculation method is:

[0069]

[0070] λ u is a logical value, and whether there is an over-limit of the node voltage is judged through the logical value. When the logical value λ u is 0, there is no problem of over-limit of the node voltage. When the logical value λ u is 1, there is a problem of over-limit of the node voltage; the penalty function is introduced in the reactive power and voltage control optimization objective function to achieve the purpose of simplifying the calculation process. If each electrical parameter does not exceed the limit, the coefficient in the penalty function is 0, and this part is ignored during the calculation process to simplify the calculation of the objective function.

[0071] In the process of solving the optimization objective function of reactive power voltage control, the value of the penalty factor plays a decisive role in the convergence speed of the penalty function in the objective function. If the penalty factor is not selected appropriately, the convergence speed of the penalty function is largely restricted by the suitability of the penalty factor. Once the penalty factor is selected improperly, many problems will be caused, the accuracy of the search results is poor, and the convergence speed is reduced. If the selected value of the penalty factor is too small, the fitness ratio of the solution will decrease accordingly, which will have a negative impact on the accuracy of the search results; if the value of the penalty factor is too large, the penalty factor coefficient in the fitness function will increase, and the proportion of the penalty function will also increase accordingly. In this way, the component of the objective function in the entire calculation process is weakened, and ultimately the reliability of the calculation results will be interfered with.

[0072] If the penalty factor is small, the component of ΔP in the objective function will increase correspondingly. When performing reactive power optimization, the population will inevitably change in the direction of smaller ΔP, further increasing the evolution speed; as the evolution process continues, while the penalty factor gradually becomes larger, the number of out-of-limit solutions will become fewer and fewer. As the optimization process increases, there will be a solution with the minimum value of ΔP.

[0073] Compared with the traditional reactive power optimization control method, the constraint conditions considered in the present invention add the constraint conditions of harmonics, and harmonics are also an important factor affecting voltage quality.

[0074] Equality constraint conditions of the distribution network, reactive and active power balance constraints:

[0075]

[0076] In the formula: P i is the active power injected into node i, Q i is the reactive power injected into node i, U i is the voltage amplitude of node i, U j is the voltage amplitude of node j, G ij is the conductance between nodes i and j, B ij is the susceptance between nodes i and j, θ ij is the phase angle difference between nodes i and j.

[0077] Inequality Constraints of Distribution Network. In the power flow calculation of the distribution network, there are clear and crucial active power, reactive power, and voltage inequality constraints for each node and branch. For nodes, the voltage must be strictly controlled within a reasonable range to ensure the normal operation of electrical equipment and power quality. Usually, the node voltage needs to satisfy, where is the lower limit of the node voltage and is the upper limit of the node voltage. This constraint avoids problems such as equipment being unable to start normally, low operating efficiency, or even damage due to too low voltage, and equipment insulation being endangered and equipment life being shortened due to too high voltage. From the perspective of active power, considering the balance between power supply output and load demand, there are generally upper and lower limits for the active power of nodes. The active power of each branch in the distribution network needs to satisfy the inequality constraint, and both the active power and reactive power of the branch need to satisfy the inequality constraint conditions to ensure the reasonable flow of power in the branch, maintain the power balance of the distribution network, and avoid adverse consequences such as voltage fluctuations and increased network losses caused by unreasonable reactive power distribution. These inequality constraint conditions cooperate with each other to provide a solid foundation for the power flow calculation of the distribution network and ensure the stable and efficient operation of the distribution network. The inequalities in the power flow calculation process of the distribution network are as follows:

[0078]

[0079] In the formula, Q jmin 、Q j 、Q jmax are respectively the lower limit, output value, and upper limit of the reactive power output of each node in the distribution network; U jmin 、U j 、U jmax are respectively the lower limit, amplitude, and upper limit of the voltage of each node in the distribution network system; q jmin 、q j 、q jmax are respectively the lower limit, transmission value, and upper limit of the reactive power transmission of each branch in the distribution network system;

[0080]

[0081] In the formula, Q Cjmin 、Q Cj 、Q Cjmax are respectively the lower limit of compensation, actual compensation power, and upper limit of the reactive power compensation device; T jmin 、T j 、T jmax are respectively the lower limit, position, and upper limit of the tap of the on-load tap-changing transformer.

[0082] IV. Calculate the weight coefficients for the reactive power-voltage control optimization objective function and the voltage quality optimization objective function based on the mean adaptive method, and then perform weight allocation.

[0083] During the process of optimizing the control of the distribution network, multiple objectives need to be optimized. The weight coefficient can take into account multiple objective functions. In the solution of multiple objective functions, considering the weight coefficient can more accurately calculate the values of each objective function. Compared with the traditional situation of solving multiple objective functions without considering the weight coefficient, it can better balance different objective functions and find the optimal solutions of each function. If the weight coefficients assigned to each objective function in the proposed weight distribution are set unreasonably, it may lead to the scheduling scheme only focusing on optimizing a certain objective while ignoring the optimization of other objectives, and often cannot obtain ideal results. The present invention selects the weight coefficient through the mean self-adaptive method, which can obtain relatively ideal weight coefficients, can well balance different objective functions, and can well find the global optimal solution, without being unable to meet the global situation due to focusing on the optimization of a certain objective, resulting in missing the global optimal solution and being unable to obtain a suitable configuration scheme.

[0084] In the process of optimizing the control of the distribution network, it is necessary to optimize multiple objectives simultaneously. However, if the weight coefficient is set unreasonably, it may cause the scheduling scheme to deviate, only focusing on optimizing one objective and ignoring other objectives, and it is often difficult to achieve ideal optimization results.

[0085] The present invention uses the mean self-adaptive method to select the weight coefficient. This method has significant advantages and can accurately obtain relatively ideal weight coefficients to balance different objective functions. When finding the global optimal solution, it will not cause the situation that the objective function has no optimal solution and over-focuses on a certain optimization objective, and can obtain a suitable configuration scheme to ensure that the optimization control of the distribution network can successfully achieve the expected goal.

[0086] For a set of q different objective functions f1(x), f2(x), f3(x), …, f q (x), randomly taking values within its range will generate a new set of objective function values f i (x1), f i (x2), f i (x3), …, f i (x m ), where i = 1, 2, 3, …, q;

[0087]

[0088] In the formula: f k (x i ) is the kth objective function; f k (x) is the weight coefficient of the kth objective function; m is the total number of objective functions.

[0089] In the process of optimizing the control of the distribution network, to ensure that each objective function in the comprehensive objective function can account for a relatively large proportion and the numerical differences between them are small, when setting a reasonable objective function later, the weight coefficients between the objective functions must be made as close as possible. In the process of modifying and calibrating the weight coefficients, only a small parameter change can achieve significant results, and then more accurately analyze the weight coefficient ratio of multiple objectives. The weight coefficient ratio affects the objective function value. To achieve the ideal effect of optimizing the control of the distribution network, the weight coefficients of each objective function should be strictly allocated. At this time, the comprehensive objective function is as follows:

[0090]

[0091] In the formula, W i is the weight of the i-th objective function. The present invention includes two objective functions: minimizing the power loss of the distribution network and minimizing the voltage deviation of each node in the distribution network. is the weight coefficient of the i-th objective function, and f i (x) is the i-th objective function, where i = 1, 2,..., n, and n is the total number of objective functions.

[0092] V. Solve the reactive power voltage control optimization objective function and the voltage quality optimization objective function after weight allocation based on the improved sine-cosine ISCA algorithm.

[0093] The optimization control model of the present invention belongs to a complex non-linear problem with multiple objectives and multiple constraints. The improved sine-cosine (ISCA) algorithm is used for solving, and the solving process is as Figure 2 shown. The specific steps are as follows:

[0094] (1) Input the distributed parameters of each part of the distribution network, the load prediction value, and the comprehensive objective function;

[0095] (2) Add their respective weight coefficients (the respective weight coefficients are ) to the reactive power voltage control optimization objective function and the voltage quality optimization objective function;

[0096] (3) Constraint condition processing. The penalty function method adds the power flow equation constraint to the reactive power voltage control optimization objective function. The voltage constraint, the reactive power source capacity constraint, and the on-load tap changer position constraint of the transformer are processed by the penalty function method or the repair strategy. The repair strategy is to adjust an individual variable to the nearest feasible value when it exceeds the constraint range;

[0097] (4) Initialize the parameters of the improved sine-cosine algorithm. At the initial stage, P random individuals are generated, and the fitness of different individuals is obtained through the reactive power voltage control optimization objective function value and the voltage quality optimization objective function value after weight allocation;

[0098] (5) Add a weight coefficient. Add the weight coefficient λ to the update formula to improve the convergence accuracy of the reactive power voltage control optimization objective function and the voltage quality optimization objective function after weight allocation. λ = 2 - 1.5e ( / / 100-1) , the addition of λ in the early stage of the search helps with global optimization, while in the later stage, more attention is paid to local optimization, thus significantly improving the convergence speed and accuracy of the improved sine-cosine algorithm. The above formula is updated as follows:

[0099]

[0100] In the formula, represents the position of the i-th individual in the j-th dimension at the t-th iteration; represents the position of the i-th individual in the j-th dimension at the (t + 1)-th iteration; is the position of the optimal solution in the j-th dimension after t iterations; r2, r3, r4 are three random numbers subject to a uniform distribution, r2 ∈ [0, 2π], r3 ∈ [0, 2], r4 ∈ [0, 1], and λ is the weight coefficient.

[0101] (6) Introduce the Levy flight strategy. Add the Levy flight strategy to the sine-cosine (SCA) algorithm. The Levy flight strategy is a movement trajectory that combines frequent short-distance movements with occasional long-distance movements, effectively avoiding the sine-cosine algorithm from falling into local optima during the process of solving the reactive power voltage control optimization objective function and the voltage quality optimization objective function. Use the normal distribution to solve the random number method as follows:

[0102]

[0103] In the formula: I is the symbol representing the normal distribution; u, v are random numbers that satisfy the normal distribution; σ v is the standard deviation of v. Use the random Levy distribution to update the optimal solution and the sub-optimal solution in each iteration. β is the step size of the flight strategy. The final global optimization formula:

[0104] If r5 < random number, then

[0105]

[0106] If r5 ≥ random number

[0107]

[0108] In the formula, is the position of the optimal solution in the j-th dimension after t iterations,

[0109]

[0110] Among them: is the fitness value; and fi t They are the fitness value of the optimal solution after the t-th iteration and the fitness value of the i-th individual respectively; γ is a number that satisfies a random distribution in [0, 2π].

[0111]

[0112] Among them: both b1 and b2 are numbers in [0, 1]; t is the current iteration number; T is the maximum iteration number; r5 is the linear coefficient of the current iteration number and the maximum iteration number.

[0113] In the formula: and f i t They are the fitness value of the optimal solution after the t-th iteration and the fitness value of the i-th individual respectively; γ is a number that satisfies a random distribution in [0, 2π]; both b1 and b2 are numbers in [0, 1], and b1 > b2.

[0114] Position update formula:

[0115]

[0116] In the formula, represents the i-th individual in the (t + 1)-th iteration.

[0117] (7) Calculate the objective function value of the reactive power voltage control optimization and the objective function value of the voltage quality optimization after weight allocation as the fitness, record the optimal individual and the optimal solution, and judge the iteration termination condition;

[0118] (8) Output the optimal reactive power control strategy, determine the control strategy according to the optimal solution, determine the switching status, compensation capacity of the reactive power compensation device in the distribution network, and the tap position of the on-load tap-changer transformer and other control parameters, form the optimal reactive power control strategy, and be used for the reactive power optimization control of the actual distribution network;

[0119] VI. Solve the power flow of the distribution network based on the Holomorphic Embedded Power Flow Method (HELM).

[0120] For the distribution network control optimization model and voltage quality optimization model of the present invention, the power flow in the model is solved by using the Holomorphic Embedded Power Flow Method, and the solution process is as Figure 3 shown:

[0121] (1) Set the iterative convergence threshold ε of the Holomorphic Embedded Power Flow Calculation, and let H = 1. ε = 1×10 -6 ;

[0122] (2) Determine the operation mode of reactive power compensation in the distribution network, and the construction of the Holomorphic Embedded Power Flow Calculation Model, and perform the H-th power flow calculation by using the Holomorphic Embedded Power Flow Calculation method;

[0123] (3) Set \(V\) i \( = 1\angle0\ p.u.\), \(V\) j \( = 1\angle0\ p.u.\), and use the following formula to calculate the reactive power output of PQ node \(i\) and PV node \(j\) respectively:

[0124]

[0125] In the formula: \(Q\) i is the injected reactive power to be solved for node \(i\), \(V\) i is the voltage amplitude of node \(i\), p.u. is the per-unit value of the voltage of each node in the distribution network, \(I\) i is the current of node \(i\) in the distribution network.

[0126] According to the principle of holomorphic embedding, the holomorphic power flow calculation model of PQ(V) nodes can be obtained from the following formula:

[0127]

[0128] In the formula, \(s\) is the embedded complex parameter, \(Y\) i is the admittance between nodes \(i\) and \(j\), \(Y\) i sh is the shunt branch admittance of node \(i\), is the current vector value of node \(i\), is the holomorphic function of the voltage vector of node \(i\), is the holomorphic function of the voltage vector of node \(i\), \(Y\) is the number of nodes in the distribution network, \(o = 1,\ldots,Y\).

[0129] The node current phasor values of node \(i\) and node \(j\) can be calculated respectively using the following formula:

[0130]

[0131] In the formula, \(P\) i sp , \(Q\) i are the known injected active power and reactive power of load node \(i\) respectively, is the current phase

[0132] value of node \(i\), is the voltage vector of node \(i\), the above \(P\) i sp \(+IQ\) i is the complex power representation form, and \(J\) is the imaginary unit.

[0133] Solve the node current phasor values of each node in the holomorphic embedded power flow calculation model through the expression of the node current phasor, and substitute the current phasor values of each node into the holomorphic embedded power flow calculation model, so as to calculate the power flow distribution of the distribution network, and extract the node voltage values from the results of the power flow calculation. Use the method of Taylor series expansion to perform Taylor series expansion on the holomorphic embedded power flow calculation model of the distribution network, deduce the recurrence formula of the power series coefficients of each node of the distribution network according to the expansion results, solve the series coefficients of the order power of 1 through the recurrence formula of the power series coefficients, and use the Padé approximation method to solve the approximate values of each node voltage.

[0134] (4) First, use the results of the H-th power flow calculation to extract the node voltage values, substitute the node voltage values of the H-th power flow calculation into the power flow equation to update the reactive power of the PV node i and the PQ node j, and at the same time calculate the node current phasor values and improve the holomorphic embedded power flow calculation models of node i and node j. After the above calculations are completed, use the holomorphic embedded power flow calculation method again to solve the distribution of the (H + 1)-th power flow of the distribution network, and then use the Padé approximation method to calculate the approximate values of each node voltage. This approximate value is the actual voltage value of each node corresponding to the distribution network at this stage, providing key data support for the precise regulation of the subsequent distribution network.

[0135] (5) Compare the voltage phasor results of the H-th power flow calculation and the (H + 1)-th power flow calculation, as shown in the following formula.

[0136]

[0137] In the formula, Y ij is the admittance between node i and node j, Y i sh is the shunt branch admittance of node i, is the current vector value of node i, is the voltage vector of node i, Y is the number of nodes in the distribution network, o = 1,..., Y, ΔV is the difference between the voltage of the (H + 1)-th power flow calculation and the voltage of the H-th power flow calculation, V L represents the voltage phasor obtained during the H-th power flow calculation, and V H+1 corresponds to the voltage phasor obtained from the (H + 1)-th power flow calculation.

[0138] e r is the maximum relative error of the voltage amplitude of each node in the distribution network, and e i is the maximum relative error of the voltage phase angle of each node. Take the maximum value of e r and e i as the error e rr . Determine the subsequent operations based on the value of e rr : If e rr≥ ε, the calculation result does not meet the accuracy requirement, and the next power flow calculation is continued, and step (4) in solving the power flow of the distribution network based on the holomorphic embedded power flow calculation method (HELM) is repeated; otherwise, if e rr < ε, the calculation result meets the accuracy standard, and the final power flow calculation result is output, providing an accurate basis for the stable operation analysis of the distribution network. ε is the error accuracy requirement, taking 1×10 -6 .

[0139] According to the second aspect of the embodiments of the present invention, a reactive power optimization control system for a distribution network based on ISCA-HELM is provided, including: a first construction module for constructing a mathematical model for minimizing network loss, including a network loss minimization objective function and network loss minimization constraint conditions; a second construction module for constructing a voltage quality optimization model, including a voltage quality optimization objective function and voltage quality optimization constraint conditions; a third construction module for constructing a reactive power and voltage control optimization model for the distribution network based on game theory. The model construction includes a reactive power and voltage control optimization objective function and reactive power and voltage control optimization constraint conditions. The reactive power and voltage control optimization constraint conditions include: constructing an active power and reactive power balance equation constraint for the distribution network, and constructing inequality constraints such as the voltage amplitude, reactive power compensation capacity, and on-load tap-changer steps of each node in the distribution network; a solving module for calculating weight coefficients for the reactive power and voltage control optimization objective function and the voltage quality optimization objective function based on the mean adaptive method, and then performing weight allocation; solving the reactive power and voltage control optimization objective function and the voltage quality optimization objective function after weight allocation based on the improved sine-cosine ISCA algorithm; and solving the power flow of the distribution network based on the holomorphic embedded power flow calculation method HELM. For the parts not detailed for each module above, reference can be made to the relevant descriptions in the embodiments. It should be noted that the above-mentioned each module can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above-mentioned each module can be located in the same processor; and / or, the above-mentioned each module is located in different processors in any combination way.

[0140] According to the third aspect of the embodiments of the present invention, a processor is provided, and the processor is used to run a program, wherein when the program runs, it executes the above-mentioned reactive power optimization control method for a distribution network based on ISCA-HELM.

[0141] According to the fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium includes a stored program, wherein when the program runs, it controls the device where the computer-readable storage medium is located to execute the above-mentioned reactive power optimization control method for a distribution network based on ISCA-HELM.

[0142] The specific embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the above embodiments, and various changes can be made without departing from the gist of the present invention within the scope of knowledge possessed by those of ordinary skill in the art.

Claims

1. A reactive power optimization control method for a distribution network based on ISCA-HELM, characterized in that, It includes the following steps: Step 1: Construct a mathematical model for minimizing network losses, including an objective function for minimizing network losses and constraints for minimizing network losses; Step 2: Construct an optimization model for voltage quality, including an objective function for optimizing voltage quality and constraints for optimizing voltage quality; Step 3: Construct an optimization model for reactive power and voltage control in a distribution network based on game theory, including an objective function for optimizing reactive power and voltage control and constraints for optimizing reactive power and voltage control; Step 4: Calculate the weight coefficients for the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality based on the mean adaptive method, and then perform weight allocation; Step 5: Solve the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality after weight allocation based on the improved sine-cosine ISCA algorithm; Step 6: Solve the power flow of the distribution network based on the holomorphic embedded load flow calculation method HELM.

2. The reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, wherein, The specific content of Step 1 includes: First, considering the power losses generated on the transmission line, construct an objective function for minimizing network losses; Second, set the decision variable of the reactive power compensation device; Finally, construct constraints for minimizing network losses.

3. A reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, characterized in that, The specific content of Step 2 includes: First, construct a function for minimizing the sum of the squares of the node voltage deviations as the objective function for optimizing voltage quality without weight allocation; Second, set the decision variable and flexibly adjust the reactive power output according to the grid demand to maintain the voltage stability of the distribution network; Finally, construct constraints for optimizing voltage quality.

4. A reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, characterized in that, In Step 3: The objective function for optimizing reactive power and voltage control is: In the formula, F is the network loss of the distribution network; ΔP is the active power loss in the distribution network; U imax , U imin represent the upper and lower limit values of the voltage allowed for system node i respectively; λ u is the out-of-limit penalty coefficient, U ilim represents the upper and lower limits of the voltage amplitude U i of node i, and its calculation method is as follows: When the logical value λ u is 0, there is no problem of node voltage exceeding the limit. When the logical value λ u is 1, there is a problem of node voltage exceeding the limit; The constraints for optimizing reactive power and voltage control include: the power balance equation constraint of the distribution network, the voltage magnitudes of each node in the distribution network, the reactive power compensation capacity, and the inequality constraints of the on-load tap changer of the transformer.

5. The reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, wherein The specific content of Step 4 includes: Obtain the weight coefficients of the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality through the mean adaptive method, that is, divide each objective function by the total number of objective functions, and add the calculated weight coefficients to the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality.

6. The reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, characterized in that, The specific content of Step 5 includes: Input the distributed parameters and load forecast values of each distribution network; Process the constraints for optimizing reactive power and voltage control; Add the penalty function and the load flow equation constraint to the objective function of optimizing reactive power and voltage control; Initialize the parameters of the improved sine-cosine algorithm to obtain the fitness of the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality; Add the weight coefficients of the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality; Introduce the Levy flight strategy; Calculate the values of the objective functions of optimizing reactive power and voltage control and the objective function of optimizing voltage quality as fitness, record the optimal individual and the optimal solution, and judge the iteration termination condition; Output the optimal reactive power control strategy.

7. The reactive power optimization control method for a distribution network based on ISCA-HELM according to claim 1, the specific implementation of Step 6 specifically includes: Set the iteration convergence threshold ε for the holomorphic embedded load flow calculation; Construct a holomorphic embedded load flow calculation model and perform the H-th load flow calculation using the holomorphic embedded load flow calculation method; Calculate the difference between the voltage phasor obtained from the H-th power flow calculation and the voltage phasor obtained from the (H + 1)-th power flow calculation; determine whether the accuracy requirement is met. If the accuracy requirement is met, output the final power flow calculation result. If the accuracy requirement is not met, re-use the holomorphic embedded power flow calculation method to perform the H-th power flow calculation and perform another iteration until the accuracy requirement is met and the calculation result is output.

8. A reactive power optimization control system for a distribution network based on ISCA-HELM, characterized in that, Including: The first construction module is used to construct a network loss minimization model, including a network loss minimization objective function and network loss minimization constraint conditions; The second construction module is used to construct a voltage quality optimization model, including a voltage quality optimization objective function and voltage quality optimization constraint conditions; The third construction module is used to construct a reactive power voltage control optimization model based on game theory, including a reactive power voltage control optimization objective function and reactive power voltage control optimization constraint conditions; The solution module calculates the weight coefficients for the reactive power voltage control optimization objective function and the voltage quality optimization objective function based on the mean adaptive method, and then performs weight allocation; Solve the reactive power voltage control optimization objective function and the voltage quality optimization objective function after weight allocation based on the improved sine-cosine ISCA algorithm; Solve the distribution network power flow based on the holomorphic embedded power flow calculation method HELM.

9. A processor, characterized in that, The processor is used to run a program, wherein, when the program runs, it executes the ISCA-HELM-based distribution network reactive power optimization control method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program runs, it controls the device where the computer-readable storage medium is located to execute an ISCA-HELM-based distribution network reactive power optimization control method according to any one of claims 1-7.

Images