Distribution network distributed photovoltaic access capacity and position optimization method based on sensitivity

Abstract

The invention relates to a distribution network distributed photovoltaic access capacity and position optimization method based on sensitivity, and belongs to the technical field of distributed photovoltaic power generation. The method comprises the following steps: determining a node model containing distributed photovoltaic access; determining characteristic parameters of a power distribution network line, and establishing a sensitivity coefficient matrix; based on the sensitivity coefficient matrix, establishing a line correction model for distributed photovoltaic capacity calculation, and correcting line parameters; and solving the distributed photovoltaic optimal access capacity and position by using an improved particle swarm algorithm. The distributed photovoltaic optimal access capacity and position can be effectively evaluated, and the method is of great significance to stable operation of a distribution network system in the later period.

Description

Technical Field

[0001] This invention belongs to the field of distributed photovoltaic power generation technology, specifically relating to a sensitivity-based method for optimizing the capacity and location of distributed photovoltaic grid access. Background Technology

[0002] Accelerating the development of renewable energy has become a consensus in society. Among these, photovoltaic (PV) power generation, with its outstanding advantages such as inexhaustible resources, unrestricted distribution, and safety and reliability, has received increasing attention and development. In the initial construction phase, determining the capacity and location of distributed PV grid connections is crucial for the stable operation of the subsequent distribution network system.

[0003] Traditional distributed photovoltaic (PV) systems are mostly located close to load points, and the random fluctuations in distributed PV power directly affect the power quality for users and the safe and stable operation of the distribution network. Therefore, for distribution networks containing large-scale distributed PV systems, there is an urgent need to establish a scientific and effective system for assessing and optimizing the maximum grid connection capacity of distributed PV systems to accurately reduce the operational risks of the distribution network and improve economic efficiency. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies and provide a sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections includes the following steps: Step (1): Based on the distribution network structure, establish a node model with distributed photovoltaic access; the node model with distributed photovoltaic access includes a weighted objective function and constraints; the constraints include power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits. Step (2): Obtain the characteristic parameters of the distribution network line; based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, calculate the sensitivity of the photovoltaic absorption ratio to the line length, wire diameter, load rate, and distribution transformer capacity, and consider the influence of location factors on the sensitivity, thereby obtaining the sensitivity coefficient matrix considering location factors. Step (3): Based on the sensitivity coefficient matrix, establish a line correction model for distributed photovoltaic capacity calculation and correct the line parameters; Step (4): The optimal access capacity and location of distributed photovoltaics are solved by improving the particle swarm optimization algorithm. When solving, the distributed photovoltaic capacity at the access point is used as the particle position information to calculate the fitness of each particle. The weighted objective function is used as the fitness. After the initial particle position is substituted into the line parameters in step (3), the initial particle position is substituted into the node model with distributed photovoltaic access in step (1). The node voltage, node active power output, node reactive power output and branch current data after the line parameters are corrected are solved by the forward push-back method to calculate the fitness of each particle. When the optimal access capacity of distributed photovoltaic (PV) at the access point is not 0, it means that distributed PV needs to be installed at that location. When the optimal access capacity of distributed PV at the access point is 0, it means that distributed PV does not need to be installed at that access point.

[0006] Furthermore, in step (1), a weighted objective function is established by using a linear weighting method to minimize the objective function of power grid loss cost and the objective function of minimizing electricity purchase cost. The specific establishment steps are as follows: Step (1.1), establish the objective function for active power loss cost: Power flow calculations for the distribution network are performed using the forward-backward substitution method at the PQ node. The formula for calculating active power loss in the distribution network is as follows: in, For active power loss in the distribution network; The total number of branches; Let be the conductance of the Kth branch; the voltage amplitudes at nodes i and j are respectively , ; Let be the voltage phase angle difference between nodes i and j; Then, the network loss cost is calculated using the active power loss and network loss of the distribution network, as shown in the following formula; in, Cost of network loss; This refers to the maximum annual load hours. For real-time electricity prices; Step (1.2), establish the objective function for electricity purchase cost: in, Total system capacity; For the total photovoltaic active power output; and These represent the active power grid losses before and after photovoltaic (PV) grid connection; For electricity purchase costs; Step (1.3) uses a linear weighting method to merge the two objective functions into a weighted objective function: in, The weighted objective function; and These are weighting coefficients, and they satisfy... ; Step (1.4) establishes the grid connection constraints for distributed photovoltaics within the region, including power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits; ① Power balance constraints: in, To summarize the points for the system; , They are nodes Injecting active power and reactive power; , These represent the real and imaginary components of the node voltage, respectively. , They are nodes and The electrical conductance and susceptance of the lines between them; ② Node voltage amplitude constraints: in, , , These are the lower limit of node voltage, the node voltage value, and the upper limit of node voltage, respectively. ③ Line maximum transmission power constraint: in, For nodes arrive The transmission power; branch road Maximum transmission power; ④ Total capacity limit for photovoltaic grid connection: in, Node i outputs active power from the distributed photovoltaic system; Total load capacity; This is the correction factor for the total load capacity.

[0007] further, The value is 1 / 4.

[0008] Furthermore, in step (2), the characteristic parameters of the distribution network line include line length l, wire diameter S, and load factor. Distribution transformer capacity S T .

[0009] Furthermore, in step (2), based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, the sensitivity of the photovoltaic absorption ratio to line length, wire diameter, load rate, and distribution transformer capacity is calculated, and the influence of location factors on the sensitivity is considered, thereby obtaining a sensitivity coefficient matrix considering location factors; the specific method is as follows: Step (2.1), Calculation of the sensitivity of photovoltaic absorption ratio to line length: Multiple sets of experiments were set up, with fixed wire diameter, load factor, and transformer capacity, and the line length increased at equal intervals from 200m to 2000m. The maximum photovoltaic grid connection capacity under voltage constraints at different line lengths was calculated using simulation software, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of the photovoltaic power absorption ratio to the line length; Step (2.2), Calculation of the sensitivity of photovoltaic absorption ratio to wire diameter: Multiple sets of experiments were set up, with fixed line lengths, constant load rates and transformer capacities, and wire diameters increasing at equal intervals from LJG-25 to LJG-115. The maximum photovoltaic grid connection capacity under voltage constraints was calculated using simulation software for different wire diameters, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of the photovoltaic power absorption ratio to the wire diameter; Step (2.3), Calculation of the sensitivity of photovoltaic grid connection ratio to load factor: Multiple sets of experiments were set up, with fixed line length, wire diameter, and transformer capacity. The load rate was increased at equal intervals from 5% to 50%. The maximum photovoltaic grid connection capacity under different load rates under voltage constraints was calculated using simulation software, and the data were fitted to obtain a form such as... The curve, This is the sensitivity coefficient of photovoltaic power absorption ratio to load factor; Step (2.4), Calculation of the sensitivity of photovoltaic absorption ratio to distribution transformer capacity: Multiple sets of experiments were set up, with fixed line length, wire diameter, and load rate. The transformer capacity was increased from equal intervals. The maximum photovoltaic access capacity under voltage constraints with different transformer capacities was calculated using simulation software, and the data were fitted to obtain a form like... The curve, This is the sensitivity coefficient of photovoltaic absorption ratio to distribution transformer capacity; Step (2.5): Calculate the sensitivity coefficient matrix considering location factors. In the formula, The sensitivity coefficient matrix considering location factors, where ; ; This refers to the location coefficient for distributed photovoltaic (PV) grid connection.

[0010] Furthermore, in steps (2.1) to (2.4), the experiments are all set to 10 groups; In step (2.5), when the access point for distributed photovoltaic power is segment 0 to 1 / 2, The value is 1.3; when the distributed photovoltaic grid is connected within the 1 / 2 to 3 / 4 section, The value is 1.1; when the distributed photovoltaic system is connected to the end of the grid (3 / 4 to the end of the grid), The value is 0.95.

[0011] Furthermore, the specific method for step (3) is as follows: The formulas for the node voltages at each node are as follows: in, P represents the initial node voltage; Q represents the active power; R represents the reactive power; and X represents the resistance. Node voltage; in, In the formula, For active power loss in the distribution network; For reactive power loss in the distribution network; This refers to the active power loss of the transformer. This refers to the reactive power loss of the transformer. Inject active power into the nodes; This indicates the reactive power injected into the node; Resistivity; It is the cross-sectional area; The vacuum permeability; For frequency; Based on the sensitivity coefficient matrix, a line correction model for distributed photovoltaic capacity calculation is established, as shown in the following equation: in, This is the corrected line length; , , , These are the sensitivity coefficients after considering location factors for line length, wire diameter, load rate, and transformer capacity, respectively; C is the correction coefficient.

[0012] Furthermore, in step (4), the position information of the particles must meet the total photovoltaic access capacity limit; the adaptive weight is used to adjust the weight, update the velocity and position of the particles, and recalculate the particle fitness; the fitness of each particle is compared with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, the particle position is updated and the global extreme value is determined. Otherwise, the optimal position of the particles and the best position passed by all particles remain unchanged; repeat the above steps until the distributed photovoltaic output capacity corresponding to the current particle position meets the overall constraint or reaches the maximum number of iterations, stop the iteration, and output the optimal distributed photovoltaic optimal access capacity.

[0013] Furthermore, in step (4), the improved particle swarm algorithm is to improve the velocity inertia weight w in the particle swarm algorithm, specifically as follows: In the formula, , These are the preset minimum and maximum inertia weights; Let be the average fitness of all particles at the d-th iteration; is the maximum fitness of all particles at the d-th iteration.

[0014] This invention also provides a sensitivity-based method and system for optimizing the capacity and location of distributed photovoltaic (PV) grid connections in distribution networks. The method, employing the aforementioned sensitivity-based approach, includes: The acquisition module obtains characteristic parameters of the distribution network lines within the region (for power flow calculation, line length (l, m), wire diameter (S, mm)). 2 ), load rate ( ,%), transformer capacity (S) T ,kW)); The calculation module, by inputting parameters obtained from the acquisition module, calculates the optimal maximum access capacity of distributed photovoltaic power generation using the method of assessing and optimizing the maximum access capacity of distributed photovoltaic power generation in the distribution network.

[0015] Based on the data of distribution network line nodes within the region, this invention establishes regional distributed photovoltaic grid connection constraints, namely power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits.

[0016] In this invention, the characteristic parameters of the distribution network line include line length (l, in meters) and wire diameter (S, in mm). 2 ), load rate ( (unit: %), transformer capacity (S) T (unit: kW).

[0017] In this invention, and It has a linear relationship with the line load; and It has a linear relationship with the transformer capacity and line impedance; In this invention, the improved particle swarm optimization algorithm (improved PSO) is as follows: Particle swarm position and velocity update formulas: in, , c1 and c2 are the velocities at times t and t+1, respectively; c1 and c2 are the individual learning factor and the social learning factor, respectively (preferably 0.5 and 0.5); w is the inertial weight of the velocity. Let t be the position of the particle at time t; and denoted as the best position traversed by particle i in the d-th iteration and the best position traversed by all particles in the d-th iteration, respectively; j is the dimension of the search space.

[0018] In existing PSO (Problem Searching) systems, the velocity inertia weight *w* is a fixed value, which easily leads to local optima. In the improved PSO, *w* is adjusted by comparing the difference between the example position and the global optimum. When the difference is large, the value of *w* needs to be increased to ensure its global search capability; conversely, the value of *w* is decreased to give it better local search capability. The specific calculation process is as follows: In the formula, , These are the preset minimum and maximum inertia weights; Let be the average fitness of all particles at the d-th iteration; is the maximum fitness of all particles at the d-th iteration.

[0019] The specific steps for improving the particle swarm optimization algorithm are as follows: 1. Initialize and improve PSO parameters, including population size 30, number of iterations 50, learning factors c1 and c2 0.5 and 0.5 respectively, and velocity -1 to 1, determine the original data of the distribution network, and adopt the IEEE 33-bus distribution system; 2. Using the distributed photovoltaic capacity at the access point as the particle position information, calculate the fitness of each particle; use a weighted objective function as the fitness; after correcting the line parameters in step (3) by substituting the initial particle position, substituting the initial particle position into the node model containing distributed photovoltaic access in step (1), and using the forward pushback method to solve the node voltage, node active power output, node reactive power output, and branch current data after the line parameters are corrected to calculate the fitness of each particle; the particle position information must meet the total photovoltaic access capacity limit; 3. Adaptive weighting is used to adjust the weights and update the particle velocity and position, avoiding getting trapped in local optima. The particle fitness is recalculated, and the iteration count is updated. 4. Calculate the fitness of each particle and compare it with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, update the particle position and determine the global extreme value; otherwise, keep the optimal position of the particle and the best position passed by all particles unchanged. 5. Repeat steps 3-4 above to update the fitness and calculate the globally optimal particle solution; 6. Determine if the distributed photovoltaic output capacity corresponding to the current particle position meets the overall constraints or reaches the maximum number of iterations, then stop the iteration and output the optimal distributed photovoltaic optimal access capacity. Simultaneously, if the optimal distributed photovoltaic access capacity at this access point is not 0, it indicates that distributed photovoltaic installation is required at this location; if the optimal distributed photovoltaic access capacity at this access point is 0, it indicates that distributed photovoltaic installation is not required at this access point.

[0020] Compared with the prior art, the beneficial effects of this invention are as follows: (1) Through sensitivity analysis, the impact of key parameters such as line length, wire diameter, load factor and transformer capacity on photovoltaic absorption capacity was quantified; (2) The introduction of the access location correction coefficient improves the accuracy of distributed photovoltaic access and effectively solves the voltage over-limit problem caused by distributed photovoltaic access; (3) This method can effectively guide the transformation of the distribution network (such as upgrading the wire diameter and increasing the transformer capacity) and improve the distributed photovoltaic acceptance capacity. It has application value for promoting the consumption of clean energy and ensuring the safe and stable operation of the distribution network. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method for assessing and optimizing the maximum access capacity of distributed photovoltaic power in distribution networks according to the present invention. Figure 2This is the distributed photovoltaic access node model of the present invention; Figure 3 The flowchart of the improved particle swarm algorithm of this invention is shown below; Figure 4 This is a flowchart of the distribution network distributed photovoltaic maximum access capacity assessment and optimization configuration system of the present invention. Detailed Implementation

[0022] The present invention will now be described in further detail with reference to the embodiments.

[0023] Those skilled in the art will understand that the following embodiments are for illustrative purposes only and should not be construed as limiting the scope of the invention. Where specific techniques or conditions are not specified in the embodiments, they are performed in accordance with the techniques or conditions described in the literature in the field or according to the product instructions. Materials or equipment whose manufacturers are not specified are all conventional products that can be obtained by purchase.

[0024] Example 1 A sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections includes the following steps: Step (1): Based on the distribution network structure, establish a node model with distributed photovoltaic access; the node model with distributed photovoltaic access includes a weighted objective function and constraints; the constraints include power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits. Step (2): Obtain the characteristic parameters of the distribution network line; based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, calculate the sensitivity of the photovoltaic absorption ratio to the line length, wire diameter, load rate, and distribution transformer capacity, and consider the influence of location factors on the sensitivity, thereby obtaining the sensitivity coefficient matrix considering location factors. Step (3): Based on the sensitivity coefficient matrix, establish a line correction model for distributed photovoltaic capacity calculation and correct the line parameters; Step (4): The optimal access capacity and location of distributed photovoltaics are solved by improving the particle swarm optimization algorithm. When solving, the distributed photovoltaic capacity at the access point is used as the particle position information to calculate the fitness of each particle. The weighted objective function is used as the fitness. After the initial particle position is substituted into the line parameters in step (3), the initial particle position is substituted into the node model with distributed photovoltaic access in step (1). The node voltage, node active power output, node reactive power output and branch current data after the line parameters are corrected are solved by the forward push-back method to calculate the fitness of each particle. When the optimal access capacity of distributed photovoltaic (PV) at the access point is not 0, it means that distributed PV needs to be installed at that location. When the optimal access capacity of distributed PV at the access point is 0, it means that distributed PV does not need to be installed at that access point.

[0025] Example 2 A sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections includes the following steps: Step (1): Based on the distribution network structure, establish a node model with distributed photovoltaic access; the node model with distributed photovoltaic access includes a weighted objective function and constraints; the constraints include power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits. Step (2): Obtain the characteristic parameters of the distribution network line; based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, calculate the sensitivity of the photovoltaic absorption ratio to the line length, wire diameter, load rate, and distribution transformer capacity, and consider the influence of location factors on the sensitivity, thereby obtaining the sensitivity coefficient matrix considering location factors. Step (3): Based on the sensitivity coefficient matrix, establish a line correction model for distributed photovoltaic capacity calculation and correct the line parameters; Step (4): The optimal access capacity and location of distributed photovoltaics are solved by improving the particle swarm optimization algorithm. When solving, the distributed photovoltaic capacity at the access point is used as the particle position information to calculate the fitness of each particle. The weighted objective function is used as the fitness. After the initial particle position is substituted into the line parameters in step (3), the initial particle position is substituted into the node model with distributed photovoltaic access in step (1). The node voltage, node active power output, node reactive power output and branch current data after the line parameters are corrected are solved by the forward push-back method to calculate the fitness of each particle. When the optimal access capacity of distributed photovoltaic (PV) at the access point is not 0, it means that distributed PV needs to be installed at that location. When the optimal access capacity of distributed PV at the access point is 0, it means that distributed PV does not need to be installed at that access point.

[0026] In step (1), a weighted objective function is established by using a linear weighting method to minimize the objective function of power grid loss cost and the objective function of minimizing electricity purchase cost. The specific establishment steps are as follows: Step (1.1), establish the objective function for active power loss cost: Power flow calculations for the distribution network are performed using the forward-backward substitution method at the PQ node. The formula for calculating active power loss in the distribution network is as follows: in, For active power loss in the distribution network; The total number of branches; Let be the conductance of the Kth branch; the voltage amplitudes at nodes i and j are respectively , ; Let be the voltage phase angle difference between nodes i and j; Then, the network loss cost is calculated using the active power loss and network loss of the distribution network, as shown in the following formula; in, Cost of network loss; This refers to the maximum annual load hours. For real-time electricity prices; Step (1.2), establish the objective function for electricity purchase cost: in, Total system capacity; For the total photovoltaic active power output; and These represent the active power grid losses before and after photovoltaic (PV) grid connection; For electricity purchase costs; Step (1.3) uses a linear weighting method to merge the two objective functions into a weighted objective function: in, The weighted objective function; and These are weighting coefficients, and they satisfy... ; Step (1.4) establishes the grid connection constraints for distributed photovoltaics within the region, including power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits; ① Power balance constraints: in, To summarize the points for the system; , They are nodes Injecting active power and reactive power; , These represent the real and imaginary components of the node voltage, respectively. , They are nodes and The electrical conductance and susceptance of the lines between them; ② Node voltage amplitude constraints: in, , , These are the lower limit of node voltage, the node voltage value, and the upper limit of node voltage, respectively. ③ Line maximum transmission power constraint: in, For nodes arrive The transmission power; branch road Maximum transmission power; ④ Total capacity limit for photovoltaic grid connection: in, Node i outputs active power from the distributed photovoltaic system; Total load capacity; This is the correction factor for the total load capacity.

[0027] The value is 1 / 4.

[0028] In step (2), the characteristic parameters of the distribution network line include line length l, wire diameter S, and load factor. Distribution transformer capacity S T .

[0029] In step (2), based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, the sensitivity of the photovoltaic absorption ratio to line length, wire diameter, load factor, and distribution transformer capacity is calculated, and the influence of location factors on the sensitivity is considered, thereby obtaining a sensitivity coefficient matrix considering location factors; the specific method is as follows: Step (2.1), Calculation of the sensitivity of photovoltaic absorption ratio to line length: Multiple sets of experiments were set up, with fixed wire diameter, load factor, and transformer capacity, and the line length increased at equal intervals from 200m to 2000m. The maximum photovoltaic grid connection capacity under voltage constraints at different line lengths was calculated using simulation software, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of the photovoltaic power absorption ratio to the line length; Step (2.2), Calculation of the sensitivity of photovoltaic absorption ratio to wire diameter: Multiple sets of experiments were set up, with fixed line lengths, constant load rates and transformer capacities, and wire diameters increasing at equal intervals from LJG-25 to LJG-115. The maximum photovoltaic grid connection capacity under voltage constraints was calculated using simulation software for different wire diameters, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of the photovoltaic power absorption ratio to the wire diameter; Step (2.3), Calculation of the sensitivity of photovoltaic grid connection ratio to load factor: Multiple sets of experiments were set up, with fixed line length, wire diameter, and transformer capacity. The load rate was increased at equal intervals from 5% to 50%. The maximum photovoltaic grid connection capacity under different load rates under voltage constraints was calculated using simulation software, and the data were fitted to obtain a form such as... The curve, This is the sensitivity coefficient of photovoltaic power absorption ratio to load factor; Step (2.4), Calculation of the sensitivity of photovoltaic absorption ratio to distribution transformer capacity: Multiple sets of experiments were set up, with fixed line length, wire diameter, and load rate. The transformer capacity was increased from equal intervals. The maximum photovoltaic access capacity under voltage constraints with different transformer capacities was calculated using simulation software, and the data were fitted to obtain a form like... The curve, This is the sensitivity coefficient of photovoltaic absorption ratio to distribution transformer capacity; Step (2.5): Calculate the sensitivity coefficient matrix considering location factors. In the formula, The sensitivity coefficient matrix considering location factors, where ; ; This refers to the location coefficient for distributed photovoltaic (PV) grid connection.

[0030] In steps (2.1) to (2.4), the experiments are all set to 10 groups; In step (2.5), when the access point for distributed photovoltaic power is segment 0 to 1 / 2, The value is 1.3; when the distributed photovoltaic grid is connected within the 1 / 2 to 3 / 4 section, The value is 1.1; when the distributed photovoltaic system is connected to the end of the grid (3 / 4 to the end of the grid), The value is 0.95.

[0031] The specific method for step (3) is as follows: The formulas for the node voltages at each node are as follows: in, P represents the initial node voltage; Q represents the active power; R represents the reactive power; and X represents the resistance. Node voltage; in, In the formula, For active power loss in the distribution network; For reactive power loss in the distribution network; This refers to the active power loss of the transformer. This refers to the reactive power loss of the transformer. Inject active power into the nodes; This indicates the reactive power injected into the node; Resistivity; It is the cross-sectional area; The vacuum permeability; For frequency; Based on the sensitivity coefficient matrix, a line correction model for distributed photovoltaic capacity calculation is established, as shown in the following equation: in, This is the corrected line length; , , , These are the sensitivity coefficients after considering location factors for line length, wire diameter, load rate, and transformer capacity, respectively; C is the correction coefficient.

[0032] In step (4), the position information of the particles must meet the total photovoltaic access capacity limit; the adaptive weight is used to adjust the weight, update the velocity and position of the particles, and recalculate the particle fitness; the fitness of each particle is compared with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, the particle position is updated and the global extreme value is determined. Otherwise, the optimal position of the particle and the best position passed by all particles remain unchanged; repeat the above steps until the distributed photovoltaic output capacity corresponding to the current particle position meets the overall constraint or reaches the maximum number of iterations, stop the iteration, and output the optimal distributed photovoltaic optimal access capacity.

[0033] In step (4), the improved particle swarm algorithm is to improve the velocity inertia weight w in the particle swarm algorithm, specifically as follows: In the formula, , These are the preset minimum and maximum inertia weights; Let be the average fitness of all particles at the d-th iteration; is the maximum fitness of all particles at the d-th iteration.

[0034] Example 3 The system of sensitivity-based distribution network distributed photovoltaic (PV) grid access capacity and location optimization method adopts the aforementioned sensitivity-based distribution network distributed PV grid access capacity and location optimization method, including: The acquisition module acquires characteristic parameters of the distribution network lines within the area. The calculation module, by inputting the feature parameters obtained by the acquisition module, uses a sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections to calculate the optimal maximum capacity for distributed PV grid connections.

[0035] Example 4 A sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections includes the following steps: Step (1): Based on the distribution network structure, establish a node model with distributed photovoltaic access; the node model with distributed photovoltaic access includes a weighted objective function and constraints; the constraints include power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits. Step (2): Obtain the characteristic parameters of the distribution network line; based on the relationship between the photovoltaic absorption ratio and the distribution network line parameters, calculate the sensitivity of the photovoltaic absorption ratio to the line length, wire diameter, load rate, and distribution transformer capacity, and consider the influence of location factors on the sensitivity, thereby obtaining the sensitivity coefficient matrix considering location factors. Step (3): Based on the sensitivity coefficient matrix, establish a line correction model for distributed photovoltaic capacity calculation and correct the line parameters; Step (4): The optimal access capacity and location of distributed photovoltaic (PV) are solved by improving the particle swarm optimization algorithm. The PV capacity at the access point is used as the particle position information to calculate the fitness of each particle. The present invention uses a weighted objective function as the fitness. After correcting the line parameters in step (3), the initial particle position is substituted into the node model with PV access in step (1). The node voltage, active power output, reactive power output, and branch current data after the line parameters are corrected are solved using the forward push-back method to calculate the fitness of each particle. The particle position information must meet the total PV access capacity limit. The weights are adjusted using adaptive weights, the particle velocity and position are updated, and the particle fitness is recalculated. The fitness of each particle is compared with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, the particle position is updated to determine the global extreme value. Otherwise, the optimal position of the particle and the best position passed by all particles remain unchanged. The above steps are repeated until the PV output capacity corresponding to the current particle position meets the overall constraint or the maximum number of iterations is reached. Then the iteration is stopped and the optimal PV access capacity is output. Meanwhile, when the optimal access capacity of distributed photovoltaic (PV) at the access point is not 0, it means that distributed PV needs to be installed at that location; when the optimal access capacity of distributed PV at the access point is 0, it means that distributed PV does not need to be installed at that access point. Therefore, this method can optimize the access capacity and location of distributed PV.

[0036] Step 1 specifically includes: 1) Characteristic parameters of distribution network lines include line length (l, m) and wire diameter (S, mm). 2 ), load rate ( ,%), transformer capacity (S) T ,kW).

[0037] 2) A weighted objective function is established by using the linear weighting method to construct the objective functions of power loss cost and electricity purchase cost. The specific steps are as follows: ① Objective function of active power loss cost Power flow calculations for the distribution network are performed using the forward-backward substitution method at the PQ node. The network loss calculation formula is as follows: in, For active power loss in the distribution network; The total number of branches; The conductance of the branch is given; the voltage amplitudes at nodes i and j are respectively... , ; This represents the voltage phase angle difference.

[0038] The formula for converting network loss into cost is as follows; in, Cost of network loss; This refers to the maximum annual load hours. This refers to the real-time electricity price.

[0039] ② Objective function of electricity purchase cost in, Total system capacity; For the total photovoltaic active power output; and These represent the active power grid losses before and after photovoltaic (PV) integration.

[0040] This invention uses a linear weighting method to merge two objective functions into one objective function: in, The weighted objective function; and These are weighting coefficients, and they satisfy... .

[0041] 3) Based on the data of the distribution network line nodes in the region, establish the grid connection constraints for distributed photovoltaic power in the region.

[0042] ① Power balance constraints: in, To summarize the points for the system; , They are nodes Injecting active power and reactive power; , These represent the real and imaginary components of the node voltage, respectively. , They are nodes and The electrical conductance and susceptance of the lines between them.

[0043] ② Node voltage amplitude constraints: in, , , These are the lower limit of node voltage, the node voltage value, and the upper limit of node voltage, respectively.

[0044] ③ Line maximum transmission power: in, For nodes arrive The transmission power; This represents the upper limit of the transmission power of the branch.

[0045] ④ Total capacity limit for photovoltaic grid connection: in, Node i outputs active power from the distributed photovoltaic system; Total load capacity; This is the total load capacity correction factor. The value is 1 / 4.

[0046] Step 2 specifically includes: 1) Calculation of the sensitivity of photovoltaic power absorption ratio to line length: Ten sets of experiments were set up, with fixed wire diameter, load factor, and transformer capacity. The line length was increased at equal intervals from 200m to 2000m. The maximum photovoltaic grid connection capacity under voltage constraints at different line lengths was calculated using simulation software, and the data were fitted to obtain a form as follows: The curve, This is the sensitivity coefficient of photovoltaic power absorption ratio to line length.

[0047] 2) Calculation of the sensitivity of photovoltaic power absorption ratio to wire diameter: Ten sets of experiments were set up, with fixed line lengths, constant load rates and transformer capacities, and wire diameters increasing at equal intervals from LJG-25 to LJG-115. The maximum photovoltaic grid connection capacity under voltage constraints was calculated using simulation software for different wire diameters, and the data were fitted to obtain a model as follows: The curve, This is the sensitivity coefficient of the photovoltaic absorption ratio to the wire diameter.

[0048] 3) Calculation of the sensitivity of photovoltaic grid connection ratio to load factor: Ten sets of experiments were set up, with fixed line lengths, wire diameters, and transformer capacities. Load rates were increased at equal intervals from 5% to 50%. Simulation software was used to calculate the maximum photovoltaic (PV) capacity at different load rates under voltage constraints, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of photovoltaic power absorption ratio to load factor.

[0049] 4) Calculation of the sensitivity of photovoltaic absorption ratio to distribution transformer capacity: Ten sets of experiments were set up, with fixed line lengths, wire diameters, and load rates. The transformer capacity was increased at equal intervals from 5% to 50%. Simulation software was used to calculate the maximum photovoltaic (PV) grid connection capacity under voltage constraints at different transformer capacities, and the data were fitted to obtain a model such as... The curve, This is the sensitivity coefficient of photovoltaic absorption ratio to distribution transformer capacity.

[0050] In the formula, For distributed photovoltaic (PV) grid connection location coefficient; The sensitivity coefficient matrix considering location factors, where ; The sensitivity coefficient matrix considering location factors, where ; The values ​​can be found in the table below.

[0051] Table 1 Access location <![CDATA[C P ]]> 0~1 / 2 segment 1.3 Within the 1 / 2~3 / 4 section 1.1 3 / 4 to the end 0.95 Step 3 specifically includes: 1) The formulas for the node voltages at each node are as follows: in, P is the initial node voltage; Q is the active power; R is the reactive power; and X is the reactance.

[0052] 2) The expressions for active and reactive power in the line are: in, and It has a linear relationship with the line load; and It has a linear relationship with the transformer capacity and line impedance.

[0053] 3) The expression for line impedance is: In the formula, Indicates the active power load of the node; Indicates the reactive load of the node; For active power loss in the distribution network; For reactive power loss in the distribution network; This refers to the active power loss of the transformer. This refers to the reactive power loss of the transformer. Inject active power into the nodes; This indicates the reactive power injected into the node; Resistivity; It is the cross-sectional area; The vacuum permeability; For frequency.

[0054] 4) Based on the sensitivity coefficient matrix, a line correction model for distributed photovoltaic capacity calculation is established, as shown in the following equation: in, This is the corrected line length; , , , These are the sensitivity coefficients after considering location factors, which are applied to the line length, wire diameter, load rate, and transformer capacity, respectively, with C being the correction coefficient.

[0055] Step 4 specifically includes: The optimal grid connection configuration for distributed photovoltaic systems is solved by improving the particle swarm optimization algorithm.

[0056] 1) Based on the optimal capacity configuration of access points of the improved particle swarm algorithm, the improved particle swarm algorithm is as follows.

[0057] Particle swarm position and velocity update formulas: in, , denoted as t and t+1, respectively; c1 and c2 are the individual learning factor and the social learning factor, respectively; w is the inertia weight of the velocity. Let t be the position of the particle at time t; and denoted as the best position traversed by particle i in the d-th iteration and the best position traversed by all particles in the d-th iteration, respectively; j is the dimension of the search space.

[0058] In the improved PSO, the inertia weight w is a fixed value, which easily leads to local optima. w is improved by comparing the difference between the example position and the global optimum. When the difference is large, the value of w needs to be increased to ensure its global search capability; conversely, the value of w is decreased to give it better local search capability. The specific calculation process is as follows: In the formula, , These are the preset minimum and maximum inertia weights; Let be the average fitness of all particles at the d-th iteration; is the maximum fitness of all particles at the d-th iteration.

[0059] Initialize and improve PSO parameters, including population size 30, number of iterations 50, learning factors c1 and c2 0.5 and 0.5 respectively, and velocity -1 to 1, determine the original data of the distribution network, and adopt the IEEE 33-bus distribution system; Using the distributed photovoltaic capacity at the access point as particle position information, the fitness of each particle is calculated; a weighted objective function is used as the fitness; after correcting the line parameters in step (3) by substituting the initial particle position, the initial particle position is substituting into the node model containing distributed photovoltaic access in step (1), and the node voltage, node active power output, node reactive power output, and branch current data after the line parameters are corrected are solved using the forward pushback method to calculate the fitness of each particle; the particle position information must meet the total photovoltaic access capacity limit; Adaptive weights are used to adjust the weights and update the particle velocity and position, avoiding getting trapped in local optima. Particle fitness is recalculated, and the iteration count is updated. Calculate the fitness of each particle and compare it with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, update the particle position and determine the global extreme value; otherwise, keep the optimal position of the particle and the best position passed by all particles unchanged. Repeat steps ③ and ④ above to update the fitness and calculate the globally optimal particle solution; If the distributed photovoltaic (PV) output capacity at the current particle position satisfies the overall constraints or reaches the maximum number of iterations, the iteration stops, and the optimal distributed PV access capacity is output. Simultaneously, if the optimal distributed PV access capacity at this access point is not 0, it indicates that distributed PV needs to be installed at that location; if the optimal distributed PV access capacity at this access point is 0, it indicates that distributed PV does not need to be installed at that location.

[0060] Application Examples To verify the evaluation and optimization configuration method for the maximum access capacity of distributed photovoltaic (PV) power grids, the access capacity of the front and back-end areas was compared. An IEEE 33-node network topology was selected, and the initial distributed PV access capacity was set. The improved particle swarm optimization (PSO) algorithm parameters were initialized: population size 30, maximum number of iterations 50, learning factors c1 and c2 0.5 and 0.5 respectively; particle velocity ranged from -1 to 1, with a weighted objective function... The fitness function is used. During the iteration process, the fitness value of each particle is calculated, the individual and global optimal solutions are updated, and the search capability is dynamically adjusted using an adaptive inertia weight strategy to avoid getting trapped in local optima. By continuously updating the particle velocity and position, the fitness is re-evaluated until the voltage constraint condition is met or the maximum number of iterations is reached. Finally, the optimal maximum grid connection capacity of distributed photovoltaic power generation that meets the requirements for safe operation is output.

[0061] Table 2 Comparison of Photovoltaic Capacity Before and After Construction Comparison indicators Before renovation After renovation Accessible capacity (kW) 7 17 <![CDATA[Wire diameter (mm 2 )]]> 25 95 Network losses and electricity purchase costs (ten thousand yuan) 485 337 As shown in Table 2, the sensitivity-based method for assessing and optimizing the maximum grid-connected capacity of distributed photovoltaic power generation in distribution networks, compared to the method before the modification, reduces the wire diameter from 25mm. 2 Change to 95mm 2 The available capacity increased by 142%, while network losses and electricity purchase costs decreased by 30.5%. Actual simulations verified that the method proposed in this invention can make the maximum available capacity and configuration of distributed photovoltaic power in distribution networks more reasonable, and can be applied to actual distribution network planning.

[0062] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A sensitivity-based distribution network distributed photovoltaic access capacity and location optimization method, characterized in that, Includes the following steps: Step (1): Based on the distribution network structure, establish a node model with distributed photovoltaic access; the node model with distributed photovoltaic access includes a weighted objective function and constraints; the constraints include power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits. Step (2): Obtain the characteristic parameters of the distribution network line; based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, calculate the sensitivity of the photovoltaic absorption ratio to the line length, wire diameter, load rate, and distribution transformer capacity, and consider the influence of location factors on the sensitivity, thereby obtaining the sensitivity coefficient matrix considering location factors. Step (3): Based on the sensitivity coefficient matrix, establish a line correction model for distributed photovoltaic capacity calculation and correct the line parameters; Step (4): The optimal access capacity and location of distributed photovoltaics are solved by improving the particle swarm optimization algorithm. When solving, the distributed photovoltaic capacity at the access point is used as the particle position information to calculate the fitness of each particle. The weighted objective function is used as the fitness. After the initial particle position is substituted into the line parameters in step (3), the initial particle position is substituted into the node model with distributed photovoltaic access in step (1). The node voltage, node active power output, node reactive power output and branch current data after the line parameters are corrected are solved by the forward push-back method to calculate the fitness of each particle. When the optimal access capacity of distributed photovoltaic (PV) at the access point is not 0, it means that distributed PV needs to be installed at that location. When the optimal access capacity of distributed PV at the access point is 0, it means that distributed PV does not need to be installed at that access point.

2. The sensitivity-based distribution network distributed photovoltaic access capacity and location optimization method of claim 1, wherein, In step (1), a weighted objective function is established by using a linear weighting method to minimize the objective function of power grid loss cost and the objective function of minimizing electricity purchase cost. The specific establishment steps are as follows: Step (1.1), establish the objective function for active power loss cost: Power flow calculations for the distribution network are performed using the forward-backward substitution method at the PQ node. The formula for calculating active power loss in the distribution network is as follows: wherein, is the active power network loss of the distribution network; is the total number of branches; is the conductance of the Kth branch; the voltage amplitudes of nodes i, j are , ; is the phase angle difference of nodes i, j; Then, the network loss cost is calculated using the active power loss and network loss of the distribution network, as shown in the following formula; wherein, is the cost of network losses; is the maximum annual load hours; is the real-time electricity price; Step (1.2), establish the objective function for electricity purchase cost: wherein, is the total capacity of the system; is the total photovoltaic active output; and are the active network losses before and after the connection of the photovoltaic, respectively; is the electricity purchase cost; Step (1.3) uses a linear weighting method to merge the two objective functions into a weighted objective function: wherein, is a weighted objective function; and is a weighting coefficient, and satisfies ; Step (1.4) establishes the grid connection constraints for distributed photovoltaics within the region, including power balance constraints, node voltage amplitude constraints, line limit transmission power constraints, and total photovoltaic access capacity limits; ① Power balance constraints: wherein, is the total number of system nodes; , are the real and imaginary parts of the voltage at node injected active and reactive power; , are the real and imaginary parts of the voltage at node , are the conductance and susceptance of the line between nodes and ; ② Node voltage amplitude constraints: wherein, , , are a node voltage lower limit, a node voltage value, a node voltage upper limit, respectively. ③ Line maximum transmission power constraint: wherein, is a node to a transmission power; is a branch a transmission power upper limit; ④ Total capacity limit for photovoltaic grid connection: wherein, Node i distributed photovoltaic output active power; Total load capacity; Total load capacity correction factor.

3. The method for optimizing the capacity and location of distributed photovoltaic power grid access based on sensitivity according to claim 2, characterized in that, The value is 1 / 4.

4. The method for optimizing the capacity and location of distributed photovoltaic power grid access based on sensitivity according to claim 1, characterized in that, In step (2), the characteristic parameters of the power distribution network line include line length l, line diameter S, load rate , and distribution transformer capacity S T .

5. The sensitivity-based distribution network distributed photovoltaic access capacity and location optimization method of claim 4, wherein, In step (2), based on the relationship between the photovoltaic absorption ratio and the characteristic parameters of the distribution network line, the sensitivity of the photovoltaic absorption ratio to line length, wire diameter, load factor, and distribution transformer capacity is calculated, and the influence of location factors on the sensitivity is considered, thereby obtaining a sensitivity coefficient matrix considering location factors; the specific method is as follows: Step (2.1), Calculation of the sensitivity of photovoltaic absorption ratio to line length: Set multiple groups of experiments, fixed wire diameter, load rate and capacity of distribution unchanged, line length from 200m-2000m equal interval increase, through the simulation software to calculate the maximum photovoltaic access capacity under voltage constraint at different line length, and the data fitting to get the curve , That is the sensitivity coefficient of photovoltaic consumption ratio to line length; Step (2.2), Calculation of the sensitivity of photovoltaic absorption ratio to wire diameter: Set multiple groups of experiments, fixed line length, load rate and capacity of distribution transformer unchanged, wire diameter from LJG-25~LJG-115 increase at equal intervals, through the simulation software to calculate the maximum photovoltaic access capacity under voltage constraints at different wire diameter, and the data fitting to get the curve , That is the sensitivity coefficient of photovoltaic accommodation ratio to wire diameter; Step (2.3), Calculation of the sensitivity of photovoltaic grid connection ratio to load factor: Set multiple groups of experiments, fixed line length, wire diameter and capacity of the transformer are unchanged, load rate from 5%-50% increase at equal intervals, through the simulation software to calculate the maximum photovoltaic access capacity under voltage constraints at different load rates, and the data is fitted to get the curve , That is the sensitivity coefficient of photovoltaic consumption ratio to load rate; Step (2.4), Calculation of the sensitivity of photovoltaic absorption ratio to distribution transformer capacity: Set multiple groups of experiments, fixed line length, wire diameter and load rate unchanged, the capacity of distribution transformer from equal interval increase, through the simulation software to calculate the maximum photovoltaic access capacity under voltage constraint at different capacity of distribution transformer, and the data fitting to get the curve , That is the sensitivity coefficient of photovoltaic accommodation ratio to the capacity of distribution transformer; Step (2.5): Calculate the sensitivity coefficient matrix considering location factors. In the formula, The sensitivity coefficient matrix considering location factors, where ; ; This refers to the location coefficient for distributed photovoltaic (PV) grid connection.

6. The method for evaluating and optimizing the maximum grid connection capacity of distributed photovoltaic power generation in a distribution network according to claim 5, characterized in that, In steps (2.1) to (2.4), the experiments are all set to 10 groups; In step (2.5), when the access point for distributed photovoltaic power is segment 0 to 1 / 2, The value is 1.3; when the distributed photovoltaic grid is connected within the 1 / 2 to 3 / 4 section, The value is 1.1; when the distributed photovoltaic system is connected to the end of the grid (3 / 4 to the end of the grid), The value is 0.

95.

7. The method for optimizing the capacity and location of distributed photovoltaic power grid access based on sensitivity according to claim 1, characterized in that, The specific method for step (3) is as follows: The formulas for the node voltages at each node are as follows: in, P represents the initial node voltage; Q represents the active power; R represents the reactive power; and X represents the resistance. Node voltage; in, In the formula, For active power loss in the distribution network; For reactive power loss in the distribution network; This refers to the active power loss of the transformer. This refers to the reactive power loss of the transformer. Inject active power into the nodes; This indicates the reactive power injected into the node; Resistivity; It is the cross-sectional area; The vacuum permeability; For frequency; Based on the sensitivity coefficient matrix, a line correction model for distributed photovoltaic capacity calculation is established, as shown in the following equation: in, This is the corrected line length; , , , These are the sensitivity coefficients after considering location factors for line length, wire diameter, load rate, and transformer capacity, respectively; C is the correction coefficient.

8. The method for optimizing the capacity and location of distributed photovoltaic power grid access based on sensitivity according to claim 1, characterized in that, In step (4), the position information of the particles must meet the total photovoltaic access capacity limit; the adaptive weight is used to adjust the weight, update the velocity and position of the particles, and recalculate the particle fitness; the fitness of each particle is compared with the fitness of the best position passed by in this iteration and the fitness of the best position passed by all particles in this iteration. If it is greater than the original value, the particle position is updated and the global extreme value is determined. Otherwise, the optimal position of the particle and the best position passed by all particles remain unchanged; repeat the above steps until the distributed photovoltaic output capacity corresponding to the current particle position meets the overall constraint or reaches the maximum number of iterations, stop the iteration, and output the optimal distributed photovoltaic optimal access capacity.

9. The method for optimizing the capacity and location of distributed photovoltaic power grid access based on sensitivity according to claim 8, characterized in that, In step (4), the improved particle swarm algorithm is to improve the velocity inertia weight w in the particle swarm algorithm, specifically as follows: In the formula, , These are the preset minimum and maximum inertia weights; Let be the average fitness of all particles at the d-th iteration; is the maximum fitness of all particles at the d-th iteration.

10. A sensitivity-based method system for optimizing the capacity and location of distributed photovoltaic (PV) grid access in a distribution network, employing the sensitivity-based method for optimizing the capacity and location of distributed PV grid access in a distribution network as described in any one of claims 1 to 9, characterized in that... include: The acquisition module acquires characteristic parameters of the distribution network lines within the area. The calculation module, by inputting the feature parameters obtained by the acquisition module, uses a sensitivity-based method for optimizing the capacity and location of distributed photovoltaic (PV) grid connections to calculate the optimal maximum capacity for distributed PV grid connections.

Images