A method, system, device and storage medium for evaluating voltage sags in a distribution network

Abstract

The present invention relates to the technical field of voltage sag assessment, and discloses a method, system, device and storage medium for voltage sag assessment in a distribution network, including calculating an initial amplitude of the terminal voltage sag corresponding to a fault in a photovoltaic output scenario of a photovoltaic power station, and establishing an initial equivalent model of the power station; connecting the initial equivalent model of the power station to the distribution network for fault simulation to obtain a reactive power control coefficient and the number of unit operating conditions, and iteratively updating the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until a final equivalent model of the power station is obtained; establishing a set of voltage sag severity according to the amplitude of the node voltage sag corresponding to the final equivalent model of the power station, and performing weighted summation on the set of voltage sag severity according to the scenario weight to obtain a voltage sag assessment index. The present invention can accurately assess the voltage sag severity under the condition of high-proportion photovoltaic power stations connected to the distribution network, thereby improving the safety and reliability of power grid operation.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage sag assessment, and particularly to a method, a system, a device and a storage medium for assessing voltage sags in a distribution network. Background Art

[0002] The trend of high - proportion distributed new - energy power stations accessing the distribution network is becoming increasingly obvious, which promotes the optimization of the energy industrial structure and realizes low - carbon environmental protection and ecological harmonious development. However, the access of large - scale distributed new - energy power stations will affect the stable structure of the distribution network. Taking photovoltaic power sources as an example, after grid connection, they exhibit dynamic characteristics different from traditional power sources, and their low - voltage ride - through characteristics and islanding characteristics have received extensive attention for their impact on active distribution networks.

[0003] Voltage sag is currently the most frequent and serious power - quality problem, which causes great economic losses to sensitive users. With the increasing access of new energy to the distribution network, the voltage sag problem has been affected to a certain extent. Photovoltaic power stations with low - voltage ride - through capabilities will compensate reactive power during faults, alleviating voltage sags to a certain extent. However, severe voltage sags may cause continuous islanding of the internal photovoltaics in the station, further deepening the severity of voltage sags in the distribution network.

[0004] For the voltage sag assessment of high - proportion distributed new - energy power stations accessing the distribution network, most of the existing assessment methods based on fault simulation ignore the fault characteristics of the photovoltaic power generation system during grid faults and only simplify it as a traditional power source. However, the equivalent modeling of a single photovoltaic power source is difficult to depict the output characteristics of photovoltaic power stations under various uncertain factors and the mutual correlation between different photovoltaic power sources within the station. Therefore, it is difficult to accurately depict the impact of large - scale photovoltaic power stations under different operating conditions on the severity of voltage sags, resulting in inaccurate voltage sag assessment results. To achieve accurate assessment of voltage sags when large - scale new - energy power stations access the distribution network, there is an urgent need for a method for assessing voltage sags in a distribution network that considers the operating conditions of the internal units of photovoltaic power stations. Summary of the Invention

[0005] To solve the above - mentioned technical problems, the present invention provides a method, a system, a device and a storage medium for assessing voltage sags in a distribution network, and achieves the technical effect of accurately assessing the severity of voltage sags when a high - proportion photovoltaic power station accesses the distribution network by accurately depicting the interactive influence between voltage sags and the operating conditions of the internal units of the photovoltaic power station.

[0006] In the first aspect, the present invention provides a method for assessing voltage sags in a distribution network, and the method includes:

[0007] Calculate the initial transient voltage drop value of the generator terminal of the photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault under each photovoltaic output scenario of the photovoltaic power station, and establish an initial substation equivalent model based on the number of unit operating conditions according to the initial transient voltage drop value of the generator terminal;

[0008] Connect the initial substation equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and iteratively update the reactive power control coefficient and the number of unit operating conditions according to the preset algorithm until the final substation equivalent model is obtained;

[0009] Establish a voltage sag severity set according to the transient voltage drop value of the node corresponding to the final substation equivalent model, and perform weighted summation on the voltage sag severity set according to the scenario weight to obtain the voltage sag evaluation index.

[0010] Further, the steps of calculating the initial transient voltage drop value of the generator terminal of the photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault under each photovoltaic output scenario of the photovoltaic power station, and establishing an initial substation equivalent model based on the number of unit operating conditions according to the initial transient voltage drop value of the generator terminal include:

[0011] Perform disconnection analysis on the photovoltaic units of the photovoltaic power station according to the photovoltaic output scenario and the corresponding fault set to obtain the photovoltaic disconnection probability;

[0012] Adopt the Monte Carlo method to establish an initial simulation scheme based on the photovoltaic output scenario, fault information and initial reactive power control coefficient, and simulate the initial simulation scheme to obtain the initial transient voltage drop value of each node corresponding to each fault under each photovoltaic output scenario;

[0013] Taking the initial connection point in the initial transient voltage drop value as the slack node, perform power flow calculation on the photovoltaic power station according to the initial reactive power control coefficient to obtain the initial transient voltage drop value of the generator terminal of the photovoltaic unit;

[0014] Compare the initial transient voltage drop value of the generator terminal with the voltage threshold, and obtain the initial number of unit operating conditions under different operating conditions according to the comparison result and the photovoltaic disconnection probability;

[0015] Establish an initial substation equivalent model of the photovoltaic power station according to the initial reactive power control coefficient and the corresponding initial number of unit operating conditions.

[0016] Further, the steps of connecting the initial substation equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions include:

[0017] Connect the initial substation equivalent model to the distribution network for fault simulation to obtain the transient voltage drop value of the node;

[0018] Based on the first connection point voltage in the node voltage temporary drop amplitude, through a preset coefficient identification model, obtain the first reactive power control coefficient;

[0019] Based on the first connection point voltage and the first reactive power control coefficient, obtain the first generator terminal voltage temporary drop amplitude, and based on the first generator terminal voltage temporary drop amplitude, determine the first unit operating condition quantity under different operating conditions.

[0020] Further, the coefficient identification model is composed of a neural network model and a parameter calculation model. The input data of the neural network model is the connection point voltage temporary drop amplitude, the output data of the neural network model is the generator terminal voltage temporary drop amplitude, the input data of the parameter calculation model is the connection point voltage temporary drop amplitude and the generator terminal voltage temporary drop amplitude, and the output data of the parameter calculation model is the reactive power control coefficient.

[0021] Further, the step of iteratively updating the reactive power control coefficient and the unit operating condition quantity according to a preset algorithm until the final substation equivalent model is obtained includes:

[0022] Based on the first reactive power control coefficient and the first unit operating condition quantity, establish the substation equivalent model of the photovoltaic substation;

[0023] Connect the substation equivalent model to the distribution network for fault simulation to obtain the second connection point voltage, and judge whether the difference between the second connection point and the first connection point is greater than the difference threshold;

[0024] If it is greater, based on the second connection point voltage and the coefficient identification model, obtain the second reactive power control coefficient, and use the second reactive power control coefficient as the updated reactive power control coefficient; otherwise, use the first reactive power control coefficient as the updated reactive power control coefficient;

[0025] Based on the updated reactive power control coefficient, obtain the second unit operating condition quantity, and based on the updated reactive power control coefficient and the second unit operating condition quantity, update the substation equivalent model to obtain the updated substation equivalent model;

[0026] Judge whether the second unit operating condition quantity is consistent with the first unit operating condition quantity. If they are consistent, use the updated substation equivalent model as the final substation equivalent model. If they are not consistent, connect the updated substation equivalent model to the distribution network for fault simulation iteration until the final substation equivalent model is obtained.

[0027] Further, the step of establishing a voltage sag severity set according to the node voltage temporary drop amplitude corresponding to the final substation equivalent model and performing weighted summation on the voltage sag severity set according to the scenario weight to obtain the voltage sag assessment index includes:

[0028] Based on the voltage sag magnitudes of each node corresponding to each fault under each photovoltaic output scenario in the final substation equivalent model, a voltage sag severity set is established.

[0029] According to the entropy weight method and the probabilities of photovoltaic output scenarios, the scenario weights of each photovoltaic output scenario are obtained.

[0030] The voltage sag severity set is weighted and summed according to the scenario weights to obtain the voltage sag assessment index.

[0031] Furthermore, the step of taking the initial grid connection point in the initial voltage sag magnitude as the slack node, and performing power flow calculation on the photovoltaic power station according to the initial reactive power control coefficient to obtain the initial terminal voltage sag magnitude of the photovoltaic generator set includes:

[0032] According to the reactive power priority control strategy of the photovoltaic power station during grid faults and the initial reactive power control coefficient, the current reference value is obtained.

[0033] Taking the initial grid connection point in the initial voltage sag magnitude as the slack node, and performing power flow calculation on the photovoltaic power station according to the initial reactive power control coefficient and the current reference value to obtain the initial terminal voltage sag magnitude of the photovoltaic generator set.

[0034] In a second aspect, the present invention provides a distribution network voltage sag assessment system, and the system includes:

[0035] An initial model construction module, configured to calculate the initial terminal voltage sag magnitude of the photovoltaic generator set according to the initial reactive power control coefficient corresponding to each fault under each photovoltaic output scenario of the photovoltaic power station, and establish an initial substation equivalent model based on the number of unit operating conditions according to the initial terminal voltage sag magnitude.

[0036] A model iteration and update module, configured to connect the initial substation equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and perform iterative update on the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until a final substation equivalent model is obtained.

[0037] An assessment index calculation module, configured to establish a voltage sag severity set according to the voltage sag magnitudes of the nodes corresponding to the final substation equivalent model, and perform weighted summation on the voltage sag severity set according to the scenario weights to obtain the voltage sag assessment index.

[0038] In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the steps of the above method are implemented.

[0039] Fourthly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0040] The present invention provides a method, a system, a device and a storage medium for evaluating voltage sags in a distribution network. Through simulation iterative calculation, the present invention can accurately depict the interaction relationship between voltage sags and the operating conditions of the units inside a photovoltaic power station, and can accurately evaluate the severity of voltage sags when a high-proportion photovoltaic power station is connected to the distribution network, thereby providing accurate data support for improving the weak links of the grid voltage and enhancing the power supply quality of the grid, and further improving the safety and reliability of the grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic flowchart of the method for evaluating voltage sags in a distribution network according to an embodiment of the present invention;

[0042] Figure 2 is a schematic diagram of dividing the uncertain region of low-voltage ride-through of a photovoltaic unit according to an embodiment of the present invention;

[0043] Figure 3 is a schematic structural diagram of the system for evaluating voltage sags in a distribution network according to an embodiment of the present invention;

[0044] Figure 4 is an internal structure diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Please refer to Figure 1 , a method for evaluating voltage sags in a distribution network proposed in the first embodiment of the present invention, which includes steps S10 to S30:

[0047] Step S10: Calculate the initial amplitude of the terminal voltage sag of the photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault in each photovoltaic output scenario of the photovoltaic power station, and establish an initial equivalent model of the power station based on the number of unit operating conditions according to the initial amplitude of the terminal voltage sag;

[0048] Step S20: Connect the initial substation equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and iteratively update the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until the final substation equivalent model is obtained;

[0049] Step S30: Establish a voltage sag severity set based on the node voltage sag amplitude corresponding to the final substation equivalent model, and perform weighted summation on the voltage sag severity set according to the scenario weight to obtain a voltage sag assessment index.

[0050] In the present invention, for a distribution network with a high proportion of distributed new energy substations connected, the randomness of photovoltaic output makes the bus voltage sag amplitude of the distribution network also show uncertainty. Therefore, under different output conditions, the unit operating conditions in the photovoltaic substation caused by the same short - circuit fault may be different. According to the terminal voltage sag amplitude of the photovoltaic unit, the unit operating conditions can be divided into three states, namely, normal operating state, low - voltage ride - through state, and islanding state. Among them, due to the fuzziness of the low - voltage ride - through of the photovoltaic inverter and the uncertainty of the islanding of the photovoltaic unit, the equivalent model during the fault of the photovoltaic substation will change. Therefore, it is necessary to consider the uncertainties of photovoltaic output, fault, and photovoltaic islanding, and analyze the internal unit operating conditions of the photovoltaic substation to accurately characterize the output characteristics of the photovoltaic substation and thus accurately evaluate the voltage sag of the distribution network.

[0051] In a preferred embodiment, first analyze the uncertainties of grid power output, fault probability, and power islanding, and establish a multiple - uncertainty model of photovoltaic unit operating conditions. Since the reactive power control coefficients of photovoltaic inverters are different under different photovoltaic output scenarios and different distribution network faults, the reactive power control parameters are also uncertain, which further affects the power flow calculation inside the photovoltaic substation and causes changes in unit operating conditions. Therefore, in this embodiment, consider the influence of control parameter uncertainty on unit operating conditions, and based on the multiple - uncertainty model of photovoltaic unit operating conditions, establish an equivalent model of a photovoltaic substation considering multiple operating condition uncertainties. The specific steps are as follows:

[0052] Conduct islanding analysis on the photovoltaic units of the photovoltaic substation according to the photovoltaic output scenario and the corresponding fault set to obtain the photovoltaic islanding probability;

[0053] Adopt the Monte Carlo method to establish an initial simulation scheme based on the photovoltaic output scenario, fault information, and initial reactive power control coefficient, and simulate the initial simulation scheme to obtain the initial node voltage sag amplitude corresponding to each fault under each photovoltaic output scenario;

[0054] Taking the initial connection point in the initial node voltage dip amplitude as the balancing node, performing a power flow calculation on the photovoltaic power station according to the initial reactive power control coefficient, and obtaining the initial terminal voltage dip amplitude of the photovoltaic unit;

[0055] Comparing the initial terminal voltage dip amplitude with the voltage threshold, and obtaining the initial unit condition quantity under different working conditions according to the comparison result and the photovoltaic disconnection probability;

[0056] Establishing an initial equivalent model of the photovoltaic power station according to the initial reactive power control coefficient and the corresponding initial unit condition quantity.

[0057] In this embodiment, the photovoltaic output scenarios are divided into n scenarios (S1, S2, S3,..., Sn), and the probability of each scenario is calculated based on the probability density function f(P pv ) of the Beta distribution function, and the calculation formula of the probability density function is:

[0058]

[0059] In the formula, P pv is the output power of the solar cell, P pv,max is the maximum output power of the solar cell, is the Beta distribution function, and are the parameters related to the Beta distribution function.

[0060] Then, the probability of a specific photovoltaic output range is calculated according to the probability density function. Specifically, first, the photovoltaic output range [X 1 , X 2 is defined, indicating the photovoltaic power generation output power within the interval; secondly, the probability of the photovoltaic output in the range [X pv , X 1 , X 2 is calculated according to f(P pv ), and among them, integrating the probability density function gives:

[0061]

[0062] Then, the probability within the interval is calculated:

[0063]

[0064] In the formula, f(*) represents the probability density function, P represents the photovoltaic output scenario probability, and X represents the photovoltaic output power.

[0065] Using the Monte Carlo method to perform random sampling of faults for each scenario, obtaining the fault set F S1,i , F S2,i , F S3,i , …, F Sn,i, where \(i = 1, 2, 3, \ldots, m\), and \(m\) represents the number of faults in the fault set. The analysis of the disconnection of the internal units of the PV power station is carried out for the fault sets in different scenarios. Considering the uncertainty of PV low-voltage ride-through, a data-driven method is used to fit the PV disconnection probability. The specific fitting steps are as follows:

[0066] According to the influence of the characteristics of the sag event on the PV disconnection characteristics, the low-voltage ride-through uncertain region is divided into three regions, namely A, B, and C. \(t\) min and \(t\) max represent the minimum and maximum values of the voltage sag duration \(t\) respectively, and \(U\) pv,min and \(U\) pv,max represent the minimum and maximum values of the voltage sag magnitude \(U\) respectively. As Figure 2 shown, the disconnection probability of region A depends on the voltage sag magnitude \(U\) and the voltage sag duration \(t\), and is expressed as ; the disconnection probability of region B depends on the voltage sag magnitude \(U\), and is expressed as ; the disconnection probability of region C depends on the voltage sag duration \(t\), and is expressed as ; \(a\), \(b\), \(c\), and \(d\) represent the variable influence parameters, which are obtained by data-driven fitting.

[0067]

[0068] Taking the sag event set \(R\) as the training input and the corresponding PV disconnection situation set \(T\) of the sag event set as the training output, the variable influence parameter characteristics are trained to determine the parameters \(a\), \(b\), \(c\), and \(d\).

[0069] Using Monte Carlo sampling, the PV output scenarios and the corresponding fault information are determined, and the initial reactive power control coefficients for the corresponding scenarios are generated, thereby establishing an initial simulation scheme based on the PV output scenarios, fault information, and initial reactive power control coefficients. Among them, the reactive power control coefficients are represented in the form of a set:

[0070]

[0071] In the formula, \((K\) q1 , \(K\) q2 ) ni represents the reactive power control coefficient combination for the \(i\)-th fault in the \(n\)-th PV output scenario, and both \(K\) q1 and \(K\) q2 are reactive power control coefficients.

[0072] After obtaining the initial simulation scheme, fault simulation is performed on the initial simulation scheme to obtain the set of initial node voltage transient drop values under the i-th fault in the n-th photovoltaic output scenario. The nodes in this embodiment refer to power grid nodes. Then, according to the voltage transient drop values of the power grid nodes, the initial terminal voltage transient drop values of the photovoltaic units in the photovoltaic power station are calculated. The specific calculation steps are as follows:

[0073] According to the reactive power priority control strategy of the photovoltaic power station during grid faults and the initial reactive power control coefficient, the current reference value is obtained;

[0074] Taking the initial grid connection point in the initial node voltage transient drop value as the equilibrium node, according to the initial reactive power control coefficient and the current reference value, power flow calculation is performed on the photovoltaic power station to obtain the initial terminal voltage transient drop value of the photovoltaic unit.

[0075] In this embodiment, during grid faults, the photovoltaic uses a reactive power priority control strategy, and the output current of the photovoltaic quickly tracks the current reference value. The current reference value is shown in the following formula:

[0076]

[0077] In the formula, I qref and I dref are the component reference values of the q-axis and d-axis on the AC side of the inverter respectively, I dref0 is the initial value of the d-axis current reference value, I N is the rated current, I max is the maximum allowable current value of the photovoltaic inverter during faults, preferably taking 1.2 times of I N , U x is the terminal voltage transient drop value of the photovoltaic unit during faults, U th,lvrt is the voltage threshold for the unit to enter low voltage ride through, U th,off is the voltage threshold for the unit to enter the islanding state, K q1 is the reactive power control coefficient when the voltage is between [0.2, 0.9], K q2 is the reactive power compensation coefficient when the voltage is between [0, 0.2], k u represents the degree of voltage dip.

[0078] Combining the output current of the photovoltaic during faults and the output power of the photovoltaic unit to obtain the expression of the photovoltaic output power during grid faults. Taking the grid connection point of the photovoltaic power station as the equilibrium node, the power flow equation during internal faults of the photovoltaic power station is listed. Among them, the expression of the photovoltaic unit output power is:

[0079]

[0080] The output current reference value expression during the fault period and the photovoltaic unit output expression are combined to obtain the photovoltaic unit output expression during the fault period:

[0081]

[0082] Where P x , Q x is the expression of active and reactive output of photovoltaic units, K P is the inverter active output parameter, P x,fault , Q x,fault are respectively the active output and reactive output of the photovoltaic unit during the fault period, P x,ref is the active output reference value, which depends on the sunlight at the time of the fault, that is, g(U x )、h(U x ) represent the active and reactive power of the photovoltaic unit with respect to U x expression.

[0083] Based on the output expression of the photovoltaic unit during the fault period, the Jacobian matrix during the fault period is derived, and the power flow calculation is performed using the Newton-Raphson method to obtain the voltage sag amplitude at the terminal of each photovoltaic unit. The specific power flow calculation steps can refer to the conventional calculation steps and will not be repeated here.

[0084] After obtaining the terminal voltage sag amplitude of the photovoltaic unit, the unit is divided into three operating conditions according to the comparison relationship between the terminal voltage sag amplitude and the above-mentioned photovoltaic low voltage protection action threshold, among which U ni,pvx It represents the voltage drop amplitude of the generator terminal under the ith fault of the nth photovoltaic output scenario. When the photovoltaic unit is in normal operation, When the PV system is in low voltage ride-through state, When the PV system is disconnected from the grid.

[0085] Then, according to the probability of photovoltaic off-grid, the number of unit operating conditions corresponding to each operating condition is calculated:

[0086]

[0087] Where M on,ni 、M lvrt,ni and M off,ni They represent the number of unit operating conditions under normal operation, low voltage ride-through and off-grid conditions, respectively. th,lvrt is the voltage threshold for the unit to enter low voltage ride through, U th,off is the voltage threshold for the unit to enter the off-grid state, m on,ni Indicates the number of units in normal working condition, m lvrt,niThe number of units with voltage lower than the low voltage ride-through threshold, m off,ni The number of units with voltage lower than the low voltage protection threshold, P off Indicates the probability of PV disconnection from the grid.

[0088] Finally, based on the number of unit operating conditions under the three operating conditions corresponding to each reactive power control coefficient combination in the reactive power control coefficient set, a set of unit operating conditions M is established ni,pv , and an initial equivalent model Z of the PV power station is established based on the set of unit operating conditions corresponding to the reactive power control coefficient pv0 , where M ni,pv ={M on,ni ,M lvrt,ni ,M off,ni}}.

[0089] In the above steps, based on the multiple uncertainties of the PV unit operating conditions, an initial equivalent model of the PV power station is established. In the actual operation of the distribution network, the value of the reactive power control coefficient depends on the grid connection point voltage of the PV power station, and at the same time, as a variable in the internal power flow calculation of the PV power station, it affects the calculation results, further affecting the division of the unit operating conditions inside the PV power station, and further affecting the assessment results of the voltage sag in the distribution network, that is, the two affect each other. At the same time, for the disconnected units, the PV disconnection will exacerbate the severity of the voltage sag, and a more severe voltage sag may cause other PVs inside the station to disconnect in series, and the two also interact with each other. Based on this, the present invention establishes an interaction model between the PV power station and the distribution network by considering the interaction between the voltage sag and the unit operating conditions inside the PV power station on the basis of the multiple uncertainty model of the unit operating conditions, and updates the unit operating conditions inside the PV power station for different scenarios and different faults through simulation iterative calculations.

[0090] First, connect the initial equivalent model of the power station to the distribution network for fault simulation to obtain a set of node voltage sag amplitudes, select the grid connection point voltage of the PV power station from the set of node voltage sag amplitudes, and determine the reactive power control coefficient and the corresponding number of unit operating conditions according to the grid connection point voltage. The specific steps include:

[0091] Connect the initial equivalent model of the power station to the distribution network for fault simulation to obtain the node voltage sag amplitude;

[0092] According to the first grid connection point voltage in the node voltage sag amplitudes, through a preset coefficient identification model, obtain the first reactive power control coefficient;

[0093] According to the first grid connection point voltage and the first reactive power control coefficient, obtain the first terminal voltage sag amplitude, and determine the number of the first unit operating conditions under different operating conditions according to the first terminal voltage sag amplitude.

[0094] In this embodiment, the initial substation equivalent model is connected to the distribution network for fault simulation, and the voltage of the grid connection point of the photovoltaic power station is selected from the node voltage sag values of the grid nodes obtained from the simulation as the first grid connection point voltage U ni,pv1 , and the first grid connection point voltage is input into the coefficient identification model to obtain the first reactive power control coefficient (K q1 , K q2 ) 1 . Here, the coefficient identification model is a pre-established model for generating reactive power control coefficients. The coefficient identification model is constructed based on a neural network model and a parameter calculation model. Among them, the neural network model uses a fully connected neural network. Its input data is the grid connection point voltage sag value, and the output data is the terminal voltage sag value. Its training dataset is obtained by connecting the internal detailed model of the photovoltaic power station to the distribution network for fault simulation. The training dataset includes the grid connection point voltage sag value of the photovoltaic power station, the set of reactive power control coefficients, and the terminal voltage sag value of the units inside the power station. A parameter calculation model is spliced behind the neural network model. The parameter calculation model is constructed based on the reverse derivation of the calculation steps of the terminal voltage sag value in the above steps, that is, the reactive power control coefficient is obtained by reverse derivation of the node voltage sag value and the terminal voltage sag value. At this time, the input data and output data of the neural network model are jointly used as the input data of the parameter calculation model, and the predicted value of the reactive power control coefficient can be obtained. The loss function is calculated with its true value and used for parameter update of the fully connected neural network. Finally, the trained coefficient identification model is obtained. According to the coefficient identification model, only the grid connection point voltage needs to be input to obtain the accurate set of reactive power control coefficients:

[0095]

[0096] In the formula, U pv is the grid connection point voltage of the photovoltaic power station, (K q1 , K q2 ) i represents the i-th type of reactive power control coefficient combination, and U i is the voltage boundary value corresponding to each type of reactive power control coefficient combination.

[0097] After obtaining the first grid connection point voltage U ni,pv1 , it is input into the trained coefficient identification model, and the corresponding first reactive power control coefficient (K q1 , K q2 ) 1 can be output. Similar to the above steps of establishing the initial substation equivalent model, according to the first grid connection point voltage and the first reactive power control coefficient, the first terminal voltage sag value U ni,pvx1 of the photovoltaic unit is calculated. According to the first terminal voltage sag value U ni,pvx1Determine the operating conditions of the photovoltaic units to obtain the number of the first unit operating conditions of the photovoltaic power station, thereby establishing the first unit operating condition set M ni,pv1 ={M on1,ni ,M lvrt1,ni ,M off1,ni}, and then establish the substation equivalent model Z of the photovoltaic power station according to the first unit operating condition set pv1 .

[0098] In this embodiment, by performing a fault simulation on the initial substation equivalent model, a substation equivalent model is obtained, and then the substation equivalent model is connected to the distribution network for the simulation of the same fault. Through simulation iteration, the final substation equivalent model is obtained. The specific steps of iterative update include:

[0099] Establish the substation equivalent model of the photovoltaic power station according to the first reactive power control coefficient and the number of the first unit operating conditions;

[0100] Connect the substation equivalent model to the distribution network for fault simulation to obtain the second connection point voltage, and determine whether the difference between the second connection point and the first connection point is greater than the difference threshold;

[0101] If it is greater, then according to the second connection point voltage and the coefficient identification model, obtain the second reactive power control coefficient, and use the second reactive power control coefficient as the updated reactive power control coefficient; otherwise, use the first reactive power control coefficient as the updated reactive power control coefficient;

[0102] According to the updated reactive power control coefficient, obtain the number of the second unit operating conditions, and update the substation equivalent model according to the updated reactive power control coefficient and the number of the second unit operating conditions to obtain the updated substation equivalent model;

[0103] Determine whether the number of the second unit operating conditions is the same as the number of the first unit operating conditions. If they are the same, use the updated substation equivalent model as the final substation equivalent model. If they are not the same, connect the updated substation equivalent model to the distribution network for fault simulation iteration until the final substation equivalent model is obtained.

[0104] In this embodiment, connect the substation equivalent model to the distribution network for the simulation of the same fault to obtain the second connection point voltage U ni,pv2 , determine the difference threshold according to the difference between the voltage boundary values corresponding to each type of reactive power control coefficient combination, and determine whether it is necessary to update the first reactive power control coefficient according to the comparison relationship between the difference between the second connection point voltage and the first connection point voltage and the difference threshold:

[0105] If , , then update the reactive power control coefficient, that is, input the second connection point voltage into the coefficient identification model to obtain the second reactive power control coefficient (K q1 ,Kq2 ) 2 ; If , the reactive power control coefficient is not updated and the first reactive power control coefficient (K q1 , K q2 ) 1 is still used.

[0106] For unified description, the reactive power control coefficient after the coefficient update determination is called the updated reactive power control coefficient. According to the updated reactive power control coefficient, the corresponding number of operating conditions M of the second unit is calculated using the above steps ni,pv2 ={M on2,ni , M lvrt2,ni , M off2,ni}, and based on the updated reactive power control coefficient and the number of operating conditions of the second unit, the substation equivalent model Z pv1 is updated to obtain the updated substation equivalent model Z pv2 .

[0107] Then, it is judged whether the number of operating conditions of the second unit is equal to that of the first unit, that is, whether the number of units under different operating conditions is the same. If they are all the same, the updated substation equivalent model Z pv2 is used as the final substation equivalent model. If the number of any unit is inconsistent, the updated substation equivalent model Z pv2 is connected to the distribution network to continue the fault simulation, and the iterative calculation is carried out according to the above steps until the number of operating conditions of the unit obtained from the two fault simulations is exactly the same, that is, the final substation equivalent model is obtained.

[0108] Then, according to the node voltage sag amplitude corresponding to the final substation equivalent model, a voltage sag severity set is established, and the voltage sag severity set is weighted and summed according to the scenario weight to obtain the voltage sag evaluation index. The specific steps include:

[0109] According to the node voltage sag amplitude corresponding to each fault under each photovoltaic output scenario in the final substation equivalent model, a voltage sag severity set is established;

[0110] According to the entropy weight method and the photovoltaic output scenario probability, the scenario weight of each photovoltaic output scenario is obtained;

[0111] The voltage sag severity set is weighted and summed according to the scenario weight to obtain the voltage sag evaluation index.

[0112] In this embodiment, first, according to the final substation equivalent model, the node voltage sag amplitude of the corresponding grid node is obtained, and a voltage sag severity set S sag,n is established:

[0113]

[0114] where U j,n represents the expected value of voltage sag at each node under the nth photovoltaic output scenario, and U i,j represents the amplitude of the voltage sag at the jth node under the ith fault, and m represents the total number of faults.

[0115] Then, according to the probability of the photovoltaic output scenario, the scenario weight is set by the entropy weight method. Specifically, based on the photovoltaic output scenario probability P i , a weight evaluation matrix is constructed, and all probability indicators are normalized by Min-Max to obtain the C i indicator, and the evaluation matrix is obtained:

[0116]

[0117]

[0118] where P i is the probability of the ith photovoltaic output scenario, and c i is the normalized C i indicator corresponding to the nth characteristic variable.

[0119] According to the evaluation matrix C, calculate the information entropy H n of the nth scenario, and calculate the information redundancy D n , and further obtain the weight of the nth scenario:

[0120]

[0121]

[0122]

[0123] Calculate the evaluation index of the voltage sag severity considering the uncertainty of photovoltaic output according to the obtained weight:

[0124]

[0125] A method for evaluating voltage sags in a distribution network provided in this embodiment. The present invention combines the obtained voltage sag severity set with the tolerance characteristics of sensitive users, and evaluates the voltage sag risk by considering the possibility of voltage sag occurrence and the severity of user impact, realizing an accurate evaluation of the voltage sag risk in a distribution network considering the influence of the operating conditions of photovoltaic units under multiple uncertainties. By using the voltage sag evaluation index, the weak links of the grid voltage are improved, which can improve the power supply quality of the grid, reduce the social losses caused by voltage sags, and further improve the safety and reliability of the grid operation.

[0126] Please refer to Figure 3, based on the same inventive concept, a distribution network voltage sag evaluation system proposed in the second embodiment of the present invention includes:

[0127] An initial model construction module 10, configured to calculate the initial terminal voltage sag amplitude of a photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault under each photovoltaic output scenario of a photovoltaic power station, and establish an initial substation equivalent model based on the number of unit operating conditions according to the initial terminal voltage sag amplitude;

[0128] A model iterative update module 20, configured to connect the initial substation equivalent model to the distribution network for fault simulation, obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and iteratively update the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until a final substation equivalent model is obtained;

[0129] An evaluation index calculation module 30, configured to establish a voltage sag severity set according to the node voltage sag amplitude corresponding to the final substation equivalent model, and perform weighted summation on the voltage sag severity set according to the scenario weight to obtain a voltage sag evaluation index.

[0130] The technical features and technical effects of the distribution network voltage sag evaluation system proposed in the embodiment of the present invention are the same as those of the method proposed in the embodiment of the present invention, and will not be elaborated herein. Each module in the above distribution network voltage sag evaluation system can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above modules.

[0131] In addition, an embodiment of the present invention further proposes a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0132] Please refer to Figure 4, the internal structure diagram of a computer device in an embodiment. The computer device may specifically be a terminal or a server. The computer device includes a processor, a memory, a network interface, a display, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements the method for evaluating the voltage sag of a distribution network. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0133] Those of ordinary skill in the art can understand that Figure 4 the structure shown in

[0134] is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computing device may include more or fewer components than those shown in the figure, or combine some components, or have the same component arrangement.

[0135] In summary, a method, a system, a device, and a storage medium for evaluating voltage sags in a distribution network proposed by an embodiment of the present invention calculate the initial amplitude of the terminal voltage sag of a photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault in each photovoltaic output scenario of a photovoltaic power station, and establish an initial equivalent model of the power station based on the number of unit operating conditions according to the initial amplitude of the terminal voltage sag; connect the initial equivalent model of the power station to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and iteratively update the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until the final equivalent model of the power station is obtained; establish a voltage sag severity set according to the amplitude of the node voltage sag corresponding to the final equivalent model of the power station, and perform weighted summation on the voltage sag severity set according to the scenario weight to obtain a voltage sag evaluation index. The present invention comprehensively considers the uncertainty of photovoltaic output, the uncertainty of faults, and the uncertainty of photovoltaic disconnection to analyze the operating conditions of the internal units of a photovoltaic power station in a distribution network, and establishes an equivalent model of the photovoltaic power station suitable for voltage sag analysis according to different operating conditions. Through simulation iterative calculation, it accurately depicts the interactive influence between voltage sags and the operating conditions of the internal units of the photovoltaic power station, accurately evaluates the severity of voltage sags under the access of a high-proportion photovoltaic power station to the distribution network, thereby providing accurate data support for improving the weak links of the grid voltage and enhancing the power supply quality of the grid, and further improving the safety and reliability of grid operation.

[0136] Each embodiment in this specification is described in a progressive manner. For parts that are the same or similar in each embodiment, they can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not conflict, it should be considered as the scope described in this specification.

[0137] The above embodiments only represent several preferred implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the technical principle of the present invention, several improvements and substitutions can be made, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the protection scope of the claimed rights.

Claims

1. A method for evaluating voltage sag in a distribution network, characterized in that: include: According to the initial reactive power control coefficient corresponding to each fault of the photovoltaic station under each photovoltaic output scenario, the initial terminal voltage sag amplitude of the photovoltaic unit is calculated, and according to the initial terminal voltage sag amplitude, the initial station equivalent model based on the number of unit operating conditions is established; The initial station equivalent model is connected to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and the reactive power control coefficient and the number of unit operating conditions are iteratively updated according to the preset algorithm until the final station equivalent model is obtained; According to the node voltage sag amplitude corresponding to the final station equivalent model, a voltage sag severity set is established, and the voltage sag severity set is weighted and summed according to the scenario weight to obtain the voltage sag evaluation index; The step of connecting the initial station equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions includes: The initial station equivalent model is connected to the distribution network for fault simulation to obtain the node voltage sag amplitude; According to the first grid connection point voltage in the node voltage sag amplitude, a first reactive power control coefficient is obtained through a preset coefficient identification model; According to the first grid connection point voltage and the first reactive power control coefficient, a voltage sag amplitude of the first generator terminal is obtained, and according to the voltage sag amplitude of the first generator terminal, the number of first unit operating conditions under different operating conditions is determined; The step of iteratively updating the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until the final station equivalent model is obtained includes: Establishing a station equivalent model of the photovoltaic station according to the first reactive power control coefficient and the number of first unit operating conditions; Connecting the station equivalent model to the distribution network for fault simulation, obtaining the voltage of the second grid connection point, and determining whether the difference between the second grid connection point and the first grid connection point is greater than a difference threshold; If it is greater than, then the second reactive power control coefficient is obtained according to the second grid connection point voltage and the coefficient identification model, and the second reactive power control coefficient is used as the updated reactive power control coefficient; otherwise, the first reactive power control coefficient is used as the updated reactive power control coefficient; According to the updated reactive power control coefficient, the number of operating conditions of the second unit is obtained, and according to the updated reactive power control coefficient and the number of operating conditions of the second unit, the station equivalent model is updated to obtain an updated station equivalent model; Determine whether the number of operating conditions of the second unit is consistent with the number of operating conditions of the first unit. If they are consistent, the updated station equivalent model is used as the final station equivalent model. If they are inconsistent, the updated station equivalent model is connected to the distribution network for fault simulation iteration until the final station equivalent model is obtained.

2. The method for evaluating voltage sag in distribution network according to claim 1, characterized in that: The steps of calculating the initial terminal voltage sag amplitude of the photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault of the photovoltaic station in each photovoltaic output scenario, and establishing the initial station equivalent model based on the number of unit operating conditions according to the initial terminal voltage sag amplitude include: According to the photovoltaic output scenario and the corresponding fault set, the photovoltaic units of the photovoltaic station are analyzed for off-grid disconnection, and the photovoltaic off-grid probability is obtained; Using the Monte Carlo method, an initial simulation scheme based on photovoltaic output scenarios, fault information and initial reactive power control coefficients is established, and the initial simulation scheme is simulated to obtain the initial node voltage sag amplitude corresponding to each fault under each photovoltaic output scenario; The initial grid connection point in the initial node voltage sag amplitude is taken as the balancing node, and the power flow calculation of the photovoltaic station is performed according to the initial reactive power control coefficient to obtain the initial terminal voltage sag amplitude of the photovoltaic unit; The initial terminal voltage sag amplitude is compared with the voltage threshold, and the number of initial unit operating conditions under different operating conditions is obtained according to the comparison result and the photovoltaic grid-off probability; According to the initial reactive power control coefficient and the corresponding number of initial unit operating conditions, an initial station equivalent model of the photovoltaic station is established.

3. The method for evaluating voltage sag in distribution network according to claim 1, characterized in that: The coefficient identification model is composed of a neural network model and a parameter calculation model, the input data of the neural network model is the voltage sag amplitude at the grid connection point, the output data of the neural network model is the voltage sag amplitude at the machine end, the input data of the parameter calculation model are the voltage sag amplitude at the grid connection point and the voltage sag amplitude at the machine end, and the output data of the parameter calculation model is the reactive power control coefficient.

4. The method for evaluating voltage sag in distribution network according to claim 1, characterized in that: The step of establishing a voltage sag severity set according to the node voltage sag amplitude corresponding to the final station equivalent model, and performing weighted summation on the voltage sag severity set according to the scenario weight to obtain a voltage sag evaluation index comprises: According to the node voltage sag amplitude corresponding to each fault under each photovoltaic output scenario in the final station equivalent model, a voltage sag severity set is established; According to the entropy weight method and the probability of photovoltaic output scenarios, the scene weights of each photovoltaic output scenario are obtained; The voltage sag severity set is weighted and summed according to the scenario weights to obtain the voltage sag assessment index.

5. The method for evaluating voltage sag in distribution network according to claim 2, characterized in that: The step of taking the initial grid-connected point in the initial node voltage sag amplitude as the balancing node and performing power flow calculation on the photovoltaic station according to the initial reactive power control coefficient to obtain the initial terminal voltage sag amplitude of the photovoltaic unit comprises: According to the reactive power priority control strategy and the initial reactive power control coefficient of the photovoltaic station during the power grid fault, a current reference value is obtained; The initial grid-connected point in the initial node voltage sag amplitude is taken as the balancing node. According to the initial reactive power control coefficient and the current reference value, the power flow calculation is performed on the photovoltaic station to obtain the initial terminal voltage sag amplitude of the photovoltaic unit.

6. A distribution network voltage sag assessment system, characterized in that: include: The initial model building module is used to calculate the initial terminal voltage sag amplitude of the photovoltaic unit according to the initial reactive power control coefficient corresponding to each fault of the photovoltaic station under each photovoltaic output scenario, and to establish the initial station equivalent model based on the number of unit operating conditions according to the initial terminal voltage sag amplitude; The model iteration and update module is used to connect the initial station equivalent model to the distribution network for fault simulation, obtain the reactive power control coefficient and the corresponding number of unit operating conditions, and iteratively update the reactive power control coefficient and the number of unit operating conditions according to a preset algorithm until the final station equivalent model is obtained; The method of connecting the initial station equivalent model to the distribution network for fault simulation to obtain the reactive power control coefficient and the corresponding number of unit operating conditions includes: The initial station equivalent model is connected to the distribution network for fault simulation to obtain the node voltage sag amplitude; According to the first grid connection point voltage in the node voltage sag amplitude, a first reactive power control coefficient is obtained through a preset coefficient identification model; According to the first grid connection point voltage and the first reactive power control coefficient, a voltage sag amplitude of the first generator terminal is obtained, and according to the voltage sag amplitude of the first generator terminal, the number of first unit operating conditions under different operating conditions is determined; The reactive power control coefficient and the number of unit operating conditions are iteratively updated according to a preset algorithm until the final station equivalent model is obtained, including: Establishing a station equivalent model of the photovoltaic station according to the first reactive power control coefficient and the number of first unit operating conditions; Connecting the station equivalent model to the distribution network for fault simulation, obtaining the voltage of the second grid connection point, and determining whether the difference between the second grid connection point and the first grid connection point is greater than a difference threshold; If it is greater than, then the second reactive power control coefficient is obtained according to the second grid connection point voltage and the coefficient identification model, and the second reactive power control coefficient is used as the updated reactive power control coefficient; otherwise, the first reactive power control coefficient is used as the updated reactive power control coefficient; According to the updated reactive power control coefficient, the number of operating conditions of the second unit is obtained, and according to the updated reactive power control coefficient and the number of operating conditions of the second unit, the station equivalent model is updated to obtain an updated station equivalent model; Determine whether the number of operating conditions of the second unit is consistent with the number of operating conditions of the first unit. If they are consistent, the updated station equivalent model is used as the final station equivalent model. If they are inconsistent, the updated station equivalent model is connected to the distribution network for fault simulation iteration until the final station equivalent model is obtained. The evaluation index calculation module is used to establish a voltage sag severity set according to the node voltage sag amplitude corresponding to the final station equivalent model, and to perform weighted summation on the voltage sag severity set according to the scenario weight to obtain the voltage sag evaluation index.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images