Power distribution network photovoltaic power supply power distribution method based on online power flow feedback optimization

Abstract

The invention discloses a power distribution network photovoltaic power supply power distribution method based on online power flow feedback optimization. Firstly, system parameter initialization and data acquisition are performed to provide a basis for subsequent calculation. And constructing a power flow matrix, analyzing the result, and adjusting the output of the photovoltaic power supply in real time through online power flow feedback so as to maintain the stability of the power grid. On the basis, a power distribution model is constructed, various constraint conditions are incorporated, and an interior point method is adopted to solve and obtain an optimal or satisfactory scheme. Finally, the scheme is issued to a controller, initial power adjustment is carried out firstly, then secondary adjustment is carried out based on voltage feedback, and voltage safety is ensured. According to the method, fair power distribution is realized, and the utilization efficiency of a photovoltaic power supply is improved; various operation constraints are strictly followed, and power grid stability is guaranteed; the real-time dynamic adjustment capability is realized, and the power grid adaptability is enhanced; and moreover, continuous optimization is controlled through closed-loop feedback, so that the problem existing in power distribution of the photovoltaic power supply of the existing power distribution network is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power distribution in a distribution network, and in particular to a photovoltaic power distribution method for a distribution network based on online power flow feedback optimization. Background Art

[0002] As the global demand for clean energy grows, solar energy is increasingly used in distribution networks as a sustainable clean energy source. The number of photovoltaic power sources connected to distribution networks continues to increase, but this also brings a series of technical challenges. In the prior art, the power allocation of photovoltaic power sources in distribution networks often lacks effective optimization strategies. On the one hand, traditional power allocation methods are difficult to achieve fairness goals, which easily leads to over-generation of some photovoltaic arrays and insufficient generation of other photovoltaic arrays. This not only affects the balance of photovoltaic power utilization in the entire distribution network, reduces energy utilization efficiency, but also may cause waste of photovoltaic power investment. For example, in some complex distribution network topologies, because the actual operating conditions of each photovoltaic array and the overall needs of the power grid are not taken into account, power is simply allocated according to a fixed ratio or default settings, which makes the output power of some photovoltaic arrays under favorable light and geographical conditions too high, while some photovoltaic arrays in areas that are blocked or have large line losses cannot fully exert their power generation capacity.

[0003] On the other hand, the existing power allocation schemes are insufficient in meeting the grid operation constraints. The maximum power limit of the photovoltaic array is a key factor in ensuring its safe and stable operation. However, in actual operation, the power output often exceeds the rated maximum value due to the lack of precise control strategies, which has a negative impact on the service life and performance of the photovoltaic array and may even cause equipment failure. At the same time, the grid connection point voltage constraints and the transmission capacity limitations of the collector line have not received sufficient attention. The fluctuation of the grid connection point voltage beyond the allowable range will affect the normal operation of the grid equipment, such as causing damage to electrical equipment and deterioration of power quality. The overload of the collector line transmission capacity may cause line overheating and increased losses. In severe cases, it will cause line failures, affecting the safe and stable operation of the entire distribution network.

[0004] In addition, most existing technologies lack an effective online feedback adjustment mechanism. The operating state of the distribution network is dynamically changing and is affected by multiple factors such as changes in light intensity and load fluctuations. However, traditional methods are difficult to monitor these changes in real time and adjust the output power of photovoltaic power sources in a timely manner. They cannot adapt to various dynamic changes in the operation of the power grid, making it difficult to ensure that the power grid is always in a stable operating state, reducing the reliability of the distribution network operation. Summary of the invention

[0005] In view of the above deficiencies in the prior art, the purpose of the present invention is to propose a power allocation method for photovoltaic power sources in a distribution network based on online power flow feedback optimization, which is used to solve the problems existing in the power allocation of photovoltaic power sources in the existing distribution network, such as unfair allocation, insufficient consideration of operating constraints, and lack of online feedback adjustment.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions:

[0007] The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization includes the following steps:

[0008] S1. System parameter initialization and data collection: initializing the parameters of the photovoltaic power system of the distribution network and collecting the operation data of the distribution network in real time; system parameter initialization ensures that the configuration of the photovoltaic power system of the distribution network matches the actual power grid;

[0009] S2. Construction of power flow matrix and online feedback: Based on the topological structure of the distribution network and the real-time operation data collected in S1, the power flow matrix of the distribution network is constructed. The power flow matrix is ​​used to describe the distribution of power flow in the power grid; the power flow matrix is ​​analyzed, and the analysis results of the power flow matrix are used to adjust the output of the photovoltaic power source in real time through online power flow feedback to maintain the stable operation of the power grid;

[0010] S3. Construction of power allocation model. The power allocation model is constructed based on the online power flow feedback information obtained in S2. The goal of the power allocation model is to achieve fair power distribution while meeting the operation constraints of the power grid. The maximum power limit of the photovoltaic array, the grid connection point voltage constraint, and the transmission capacity limit of the collector line are incorporated into the model.

[0011] S4. Design and implement optimization algorithms, using interior point method to solve the power output of the power allocation model until the optimal or satisfactory power allocation solution is found;

[0012] S5. Implementation and adjustment of the power allocation plan. The calculated power allocation plan is sent to each photovoltaic array controller. The photovoltaic array controller adjusts the output power of the photovoltaic inverter according to the received power allocation instruction, and monitors the voltage of the photovoltaic array grid connection point in real time. Based on the voltage feedback, the power is adjusted secondary to ensure that the voltage is within a safe range.

[0013] In S1, the parameters of the photovoltaic power system include the access location, capacity, upper and lower limits of the grid-connected point voltage, collector line impedance parameters, etc. of the photovoltaic array; the operation data of the power grid includes the active power, reactive power, grid-connected point voltage measurement value, etc. of the photovoltaic array.

[0014] The steps of constructing the power flow matrix in S2 include:

[0015] S201. Establish a node admittance matrix, analyze the topological structure of the distribution network, determine the number of nodes, the connection relationship between nodes, and the parameters of each branch, the branch parameters include resistance and reactance;

[0016] S202. Calculate the branch admittance between nodes. For each branch that has a connection relationship, calculate the branch admittance according to its impedance. The calculation formula is:

[0017]

[0018] In the formula, y ij is the branch admittance between node i and node j, R ij is the resistance between node i and node j, X ij is the reactance between node i and node j;

[0019] S203. Calculate node self-admittance and mutual admittance:

[0020] For each node i, the self-admittance Y ii The formula is:

[0021]

[0022] where y ij is the branch admittance between node i and node j, yi0 is the ground admittance of node i;

[0023] The formula between node i and node j is:

[0024] Y ij =y ij (i≠j);

[0025] S204. Construct a node admittance matrix. According to the self-admittance and mutual admittance calculated in S23, construct an n×n node admittance matrix Y. The elements of the node admittance matrix are expressed as:

[0026] Y ij [i,j=1,2,3,…n];

[0027] Among them, when i=j, it is self-admittance, and when i≠j, it is mutual admittance.

[0028] In S2, the step of analyzing the power flow matrix includes:

[0029] S211. Calculate the node injection current vector, for access node a i (i=1,2,…,m) PV arrays, m represents the number of PV arrays connected to the distribution network, and the node injection current I ai The calculation formula is:

[0030]

[0031] Among them, P' is the collected active power; Q' i is the reactive power; is the actual voltage at the grid point V ai The complex conjugate of ai =V′ ai ×V b , V b is the reference voltage, V′ ai is the per unit value of the grid connection point voltage;

[0032] According to the calculated injection currents of all nodes, the node injection current vector I = [I1, I2, ..., I n ] T , n is the number of nodes;

[0033] S212. Solve the node voltage vector. According to the node voltage equation I=YV, solve the node voltage vector V=Y -1 I, get the node voltage vector V = [V1, V2, ..., V n ] T , Y -1 is the inverse matrix of the node admittance matrix Y.

[0034] In S2, the method of adjusting the output of the photovoltaic power source in real time by online power flow feedback using the analysis result of the power flow matrix includes:

[0035] S221. Continuously monitor the key parameters calculated by the power flow matrix, including node voltage and branch power, compare the node voltage with the preset node voltage upper and lower limits, and compare the branch power with the branch power transmission capacity limit, and judge whether the node voltage is within the allowable range and whether the branch power does not exceed its transmission capacity through comparison;

[0036] S222. Determine the adjustment strategy. When the node voltage is detected to exceed the preset node voltage upper and lower limits, or the branch power exceeds the branch power transmission capacity limit, the online power flow feedback adjustment mechanism is triggered;

[0037] When the node voltage exceeds the limit, the power amount to be adjusted is calculated according to the pre-set voltage adjustment strategy; the pre-set voltage adjustment strategy calculates the adjustment amount according to the sensitivity relationship between voltage and power, that is, the power adjustment amount ΔP x , ΔP x =k v (V v -V ref ), where k v is the voltage-power sensitivity coefficient, V ref is the reference voltage value;

[0038] S223. Implement power adjustment, and adjust the calculated power adjustment amount ΔP x The PV array controller receives the received signal and adjusts the power by changing the control signal of the PV inverter. The controller adjusts the duty cycle of the PWM signal of the inverter. If ΔP x >0, that is, the power needs to be increased, then the duty cycle is increased to increase the amplitude of the inverter output voltage or current, thereby increasing the output power; if ΔP x <0, the duty cycle is reduced and the output power is reduced;

[0039] S224. Feedback and iteration. After completing the power adjustment, immediately re-monitor the grid operation data, including node voltage and branch power, etc., feed back the new monitoring data to the power flow matrix calculation link, recalculate the power flow matrix, and analyze the grid operation status again; if after adjustment, the grid still does not meet the stable operation standard, repeat the adjustment mechanism triggering, adjustment amount determination and power adjustment steps to form a closed-loop feedback control system; through continuous iterative adjustment, the grid operation status gradually converges to a stable state, that is, all node voltages are between the pre-set node voltage upper and lower limits, and all branch powers are less than or equal to the corresponding branch power transmission capacity limit.

[0040] In S3, the power allocation model is determined to minimize the sum of the deviations between the output power of each photovoltaic array and the average power, that is, the objective function of the power allocation model is expressed as:

[0041]

[0042] Where f(P) represents the objective function of the power allocation model, m is the number of photovoltaic arrays connected to the distribution network; P i is the output power of the i-th photovoltaic array; is the average power;

[0043] The constraints of the objective function are:

[0044] The maximum power limit of the photovoltaic array. For each photovoltaic array, its output power must meet the following requirements:

[0045] 0≤P i ≤P imax ;

[0046] Where P imax is the rated maximum power of the PV array; this constraint is set based on the physical characteristics of the PV array itself and the requirements for safe operation;

[0047] The grid connection point voltage constraint requires that the grid connection point voltage of each photovoltaic array be strictly controlled within the set upper and lower limits, namely:

[0048] V min ≤V i ≤V max ;

[0049] V min is the preset node voltage lower limit, V max is the preset node voltage upper limit, V i is the grid connection point voltage of each photovoltaic array;

[0050] The transmission capacity of the collector line is limited. The transmission power of each collector line must meet the following requirements:

[0051] S k ≤S kmax ;

[0052] Where S k is the transmission power of each collector line, k is the collector line number, S kmax The maximum transmission capacity of collector line k.

[0053] In S4, the step of using the interior point method to solve the power output of the power allocation model includes:

[0054] S41. Construct a barrier function, convert the constraints in the power allocation model into barrier terms, add them to the power allocation model, and construct a complete new objective function through the constructed barrier function;

[0055] S411. For the maximum power limit of the photovoltaic array, when constructing the barrier function, it is converted into two inequality constraints g i1 (P i )=-P i ≤0 and g i2 (P i )=P i -P imax ≤0; where g i1 (P i ) and g i2 (P i ) represents the function part of the inequality constraint for the maximum power limit conversion of the photovoltaic array;

[0056] The barrier function term corresponding to the maximum power limit of the photovoltaic array is constructed accordingly:

[0057]

[0058] B i (P i ) represents the barrier function term corresponding to the maximum power limit of the i-th photovoltaic array;

[0059] Adding up these barrier function terms for all PV arrays yields the total barrier function part B associated with the maximum power limit constraint. max (P):

[0060]

[0061] Where P represents the vector containing all photovoltaic array power variables; the optimization objective function F(P) with maximum power limit is constructed:

[0062] F(P)=f(P)+μB max (P);

[0063] Where f(P) represents the objective function of the power allocation model and is defined as:

[0064]

[0065] P i is the output power of the i-th photovoltaic array; is the average power; μ is the barrier parameter. In the iterative solution process, when P i Close to 0 or P imax When , the barrier function value will increase rapidly, preventing the iteration point from violating the maximum power limit constraint;

[0066] S412. For the grid connection point voltage constraint, convert it into two inequality constraints g i3 (V i )=V i -V min ≥0 and; g i4 (V i )=V max -V i ≥0; g i3 (V i ) and g i4 (V i ) represents the function part of the inequality constraint for the grid connection point voltage constraint transformation;

[0067] Construct the barrier function term corresponding to the grid connection point voltage constraint:

[0068]

[0069] B iV (V i ) represents the barrier function term corresponding to the voltage constraint of the ith grid connection point;

[0070] The total barrier function corresponding to all grid connection point voltage constraints is:

[0071]

[0072] B V (V) represents the total barrier function corresponding to all grid connection point voltage constraints; V represents the vector containing all grid connection point voltage variables, V i is the voltage variable of the grid-connected point of the i-th photovoltaic array;

[0073] The optimization objective function F(V) with grid connection point voltage constraint is:

[0074] F(V)=f(V)+μB V (V);

[0075] f(V) is a voltage-related function determined according to the characteristics of the power grid; during the iteration process, the voltage at the grid connection point is ensured to meet the constraint requirements to prevent the voltage from exceeding the allowable range;

[0076] S413. For the transmission capacity limit of the collector line, for each collector line, its transmission power S k There is a functional relationship between S and the photovoltaic array power variable, that is, k =h k (P), h k (P) is the functional relationship between the transmission power of the collector line and the power variable of the photovoltaic array, which describes how the output power of the photovoltaic array affects the transmission power of the collector line. Its specific form is related to factors such as the topological structure of the power grid, the line connection method, and the electrical characteristics. Finally, the constraint is transformed into:

[0077] g k (p) = h k (p)-S kmax ≤0;

[0078] S kmax is the maximum transmission capacity of collector line k;

[0079] Construct the barrier function term corresponding to the transmission capacity limitation of the collector line:

[0080]

[0081] B k (P) the barrier function term corresponding to the transmission capacity limit of the kth collector line;

[0082] The total barrier function corresponding to the transmission capacity limit of all collector lines is:

[0083]

[0084] B s (P) represents the part of the total barrier function corresponding to the transmission capacity limit of all collector lines; l represents the number of collector lines;

[0085] S414. The complete new objective function F(P) finally constructed is:

[0086] F(P)=f(P)+μ(B max (P)+B V (V)+B S (P));

[0087] S42. Solve the unconstrained optimization problem, find the derivative of the new objective function with respect to each photovoltaic array power variable, set the derivative to 0, and obtain a set of nonlinear equations;

[0088] The Newton's method is used to iterate the equations. The Newton's method is iteratively formulated as follows:

[0089]

[0090] Among them, P k and P k+1 They represent the power variable vectors of the current iteration point and the next iteration point in the Newton method iteration process, J F (P k ) is the value of F(P) in P k The Jacobian matrix at , whose elements are is F(P) in P k The gradient vector at , whose elements are Through continuous iteration, we gradually approach the optimal solution;

[0091] S43. Update the barrier parameter and check convergence. After each iteration, update the barrier parameter μ so that the influence of the barrier function is gradually weakened so as to get closer to the optimal solution of the original problem.

[0092] Check the convergence conditions, which include the change of the objective function value being less than the set threshold or the change of the variable being less than the set threshold;

[0093] If the convergence condition is not met, return to the step of solving the unconstrained optimization problem and continue the iteration; if the convergence condition is met, stop the iteration, and what is obtained at this time is the optimal or satisfactory power output solution of the power allocation model, that is, the optimal power allocation value of each photovoltaic array.

[0094] In S5, the implementation and adjustment of the power allocation scheme includes:

[0095] S51. Sending power allocation instructions and initial power adjustment, sending the power allocation scheme calculated by the optimization algorithm in S4 to the corresponding photovoltaic array controller through the communication network;

[0096] After receiving the power allocation instruction, the PV array controller immediately adjusts the output power of the PV inverter according to the instruction; power regulation is achieved by adjusting the duty cycle of the pulse width modulation (PWM) signal. If the instruction requires an increase in the output power of a certain PV array, the controller will increase the duty cycle of the PWM signal, thereby increasing the amplitude of the inverter output voltage or current, thereby increasing the output power; conversely, if the power is required to be reduced, the duty cycle will be reduced;

[0097] S52. After completing the initial power adjustment, the voltage of the photovoltaic array grid connection point is monitored in real time; this monitoring process is achieved by a high-precision voltage sensor installed at the grid connection point, and the sensor feeds back the real-time voltage value to the photovoltaic array controller;

[0098] The controller compares the monitored grid connection point voltage with the preset safety voltage range [V min , V max ] is compared. If it exceeds the safety range, the controller will start the secondary power adjustment mechanism according to the voltage feedback; if V i >V max , indicating that the voltage is too high, which will cause damage to the grid equipment. At this time, the controller will calculate the amount of power that needs to be reduced according to the pre-set voltage-power adjustment strategy, and then adjust the output power of the inverter accordingly to reduce the grid connection point voltage to a safe range; if V i <V min , the opposite adjustment strategy is adopted, increasing power output to increase voltage.

[0099] In summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are:

[0100] Power allocation optimization and efficient resource utilization, by building a scientific and reasonable power allocation model, with the goal of minimizing the sum of the deviations between the output power of each photovoltaic array and the average power, achieves accurate and fair distribution of power among photovoltaic power sources. It avoids the imbalance of power generation between photovoltaic arrays, enables each photovoltaic array to output appropriate power according to its actual situation and grid demand, fully taps the potential of photovoltaic power sources, improves the efficiency of photovoltaic power utilization in the entire distribution network, avoids energy waste, and achieves optimal allocation of resources.

[0101] To ensure the stable operation of the power grid, we fully consider and strictly follow the key constraints such as the maximum power limit of the photovoltaic array, the voltage constraint of the grid connection point, and the transmission capacity limit of the collector line. In the power distribution process, we ensure that the photovoltaic array does not exceed the rated maximum power, effectively extending the service life of the photovoltaic array and reducing the equipment failure rate; stabilize the grid connection point voltage within a safe range to ensure the normal operation of various equipment in the power grid and prevent equipment damage or performance degradation caused by voltage fluctuations; control the transmission power of the collector line within its capacity range to avoid overheating, increased losses and failure risks caused by line overload, and ensure the stable operation of the distribution network from multiple key links.

[0102] Real-time dynamic adjustment and enhanced adaptability, based on the online power flow feedback mechanism, real-time monitoring of grid operation data, including key parameters such as node voltage and branch power. Once an abnormality is found, such as node voltage exceeding the set upper and lower limits or branch power exceeding the transmission capacity limit, the adjustment mechanism is immediately triggered to calculate and implement accurate power adjustment. This real-time dynamic adjustment capability enables photovoltaic power sources to quickly respond to changes in the grid's operating status. Whether it is a sudden change in light intensity or a fluctuation in load, it can adjust power output in a timely manner to ensure that the grid is always in a stable operating state, greatly enhancing the adaptability of the distribution network to various complex working conditions.

[0103] Improve the reliability and economy of the power grid. By continuously optimizing power distribution and real-time feedback adjustment, the unstable factors in the operation of the power grid are reduced, and the risk of equipment failure and power outage caused by voltage fluctuations, power overload and other problems is reduced, thereby significantly improving the reliability of the power grid. At the same time, the optimized power distribution scheme improves the utilization efficiency of photovoltaic power sources and reduces unnecessary power losses. Under the premise of ensuring the stable operation of the power grid, it reduces operating costs and improves the overall economy of the distribution network, providing strong support for the sustainable development of the distribution network.

[0104] Closed-loop feedback control and continuous optimization form a complete closed-loop feedback control system, starting from system parameter initialization and data collection, through power flow matrix construction and analysis, power distribution model construction and optimization, to the implementation and adjustment of power distribution scheme, each link is interrelated and mutually influential. The adjusted grid operation data is fed back to the system, the power flow matrix is ​​recalculated and the power distribution is optimized, and continuous iteration and optimization are carried out, so that the photovoltaic power system of the entire distribution network can continuously adapt to various internal and external changes, and always maintain the optimal or near-optimal operating state, realizing the self-adjustment and continuous optimization of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Figure 1 It is a flowchart of a photovoltaic power allocation method for distribution network based on online power flow feedback optimization;

[0106] Figure 2 A flowchart of the power flow matrix construction and online feedback method;

[0107] Figure 3 A flowchart of a method for solving the power output of a power allocation model using an interior point method;

[0108] Figure 4 A flowchart of the method for implementing and adjusting the power allocation scheme is provided. DETAILED DESCRIPTION

[0109] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in combination with the embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and are not used to limit the invention.

[0110] like Figure 1 As shown, the method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization includes the following steps:

[0111] S1. System parameter initialization and data collection: initializing the parameters of the photovoltaic power system of the distribution network and collecting the operation data of the distribution network in real time; system parameter initialization ensures that the configuration of the photovoltaic power system of the distribution network matches the actual power grid, while real-time data collection provides dynamic data support for online power flow analysis;

[0112] S2. Construction of power flow matrix and online feedback: Based on the topological structure of the distribution network and the real-time operation data collected in S1, the power flow matrix of the distribution network is constructed. The power flow matrix is ​​used to describe the distribution of power flow in the power grid; the power flow matrix is ​​analyzed, and the analysis results of the power flow matrix are used to adjust the output of the photovoltaic power source in real time through online power flow feedback to maintain the stable operation of the power grid;

[0113] S3. Construction of power allocation model. The power allocation model is constructed based on the online power flow feedback information obtained in S2. The goal of the power allocation model is to achieve fair power allocation while meeting the operation constraints of the power grid. The maximum power limit of the photovoltaic array, the grid connection point voltage constraint, and the transmission capacity limit of the collector line are incorporated into the model. These constraints ensure the feasibility and safety of the power allocation scheme.

[0114] S4. Design and implement optimization algorithms, using the interior point method to solve the power output of the power allocation model until the optimal or satisfactory power allocation solution is found; algorithm implementation is the process of converting theoretical models into practical operations;

[0115] S5. Implementation and adjustment of the power allocation plan. The calculated power allocation plan is sent to each photovoltaic array controller. The photovoltaic array controller adjusts the output power of the photovoltaic inverter according to the received power allocation instruction, and monitors the voltage of the photovoltaic array grid connection point in real time. Based on the voltage feedback, the power is adjusted secondary to ensure that the voltage is within a safe range.

[0116] In S1, the parameters of the photovoltaic power system include the access location of the photovoltaic array (obtained through field investigation and grid topology drawing annotation), capacity (rated power generation capacity of the photovoltaic array), upper and lower limits of the grid connection point voltage (the allowable voltage fluctuation range of the photovoltaic array grid connection point), collector line impedance parameters, etc.; the operation data of the power grid include the active power and reactive power of the photovoltaic array (obtained by the power sensor installed at the output end of the photovoltaic array), the grid connection point voltage measurement value (obtained by the voltage transformer deployed at the photovoltaic array grid connection point), etc.

[0117] like Figure 2 As shown, the steps of constructing the power flow matrix in S2 include:

[0118] S201. Establish a node admittance matrix, conduct a comprehensive analysis of the topological structure of the distribution network, determine the number of nodes, the connection relationship between nodes (which nodes have branches), and the parameters of each branch, the branch parameters include resistance and reactance;

[0119] S202. Calculate the branch admittance between nodes. For each branch that has a connection relationship, calculate the branch admittance according to its impedance. The calculation formula is:

[0120]

[0121] In the formula, y ij is the branch admittance between node i and node j, R ij is the resistance between node i and node j, X ij is the reactance between node i and node j;

[0122] S203. Calculate node self-admittance and mutual admittance:

[0123] For each node i, the self-admittance Y ii The formula is:

[0124]

[0125] where y ij is the branch admittance between node i and node j, y i0 is the ground admittance of node i;

[0126] The formula between node i and node j is:

[0127] Y ij =y ij (i≠j);

[0128] S204. Construct a node admittance matrix. According to the self-admittance and mutual admittance calculated in S23, construct an n×n node admittance matrix Y. The elements of the node admittance matrix are expressed as:

[0129] Y ij [i,j=1,2,3,…n];

[0130] Among them, when i=j, it is self-admittance, and when i≠j, it is mutual admittance.

[0131] In S2, the step of analyzing the power flow matrix includes:

[0132] S211. Calculate the node injection current vector, for access node a i (i=1,2,…,m) PV arrays, m represents the number of PV arrays connected to the distribution network, and the node injection current I ai The calculation formula is:

[0133]

[0134] Among them, P' is the collected active power; Q' i is the reactive power; is the actual voltage at the grid point V ai The complex conjugate of ai =V′ ai ×V b , V b is the reference voltage, V′ ai is the per unit value of the grid connection point voltage;

[0135] According to the calculated injection currents of all nodes, the node injection current vector I = [I1, I2, ..., I n ] T , n is the number of nodes;

[0136] S212. Solve the node voltage vector. According to the node voltage equation I=YV, solve the node voltage vector V=Y -1 I, get the node voltage vector V = [V1, V2, ..., V n ] T , Y -1 is the inverse matrix of the node admittance matrix Y.

[0137] In S2, the method of adjusting the output of the photovoltaic power source in real time by online power flow feedback using the analysis result of the power flow matrix includes:

[0138] S221. Continuously monitor the key parameters calculated by the power flow matrix, including node voltage and branch power, compare the node voltage with the preset node voltage upper and lower limits, and compare the branch power with the branch power transmission capacity limit, and judge whether the node voltage is within the allowable range and whether the branch power does not exceed its transmission capacity through comparison;

[0139] S222. Determine the adjustment strategy. When the node voltage is detected to exceed the preset node voltage upper and lower limits, or the branch power exceeds the branch power transmission capacity limit, the online power flow feedback adjustment mechanism is triggered;

[0140] When the node voltage exceeds the limit, the power amount to be adjusted is calculated according to the pre-set voltage adjustment strategy; the pre-set voltage adjustment strategy calculates the adjustment amount according to the sensitivity relationship between voltage and power, that is, the power adjustment amount ΔP x , ΔP x =k v (V v -V ref ), where k v is the voltage-power sensitivity coefficient, V ref is the reference voltage value;

[0141] S223. Implement power adjustment, and adjust the calculated power adjustment amount ΔP x The PV array controller receives the received signal and adjusts the power by changing the control signal of the PV inverter. Specifically, the controller adjusts the duty cycle of the PWM signal of the inverter. If ΔP x >0, that is, the power needs to be increased, then the duty cycle is increased to increase the amplitude of the inverter output voltage or current, thereby increasing the output power; if ΔP x <0, the duty cycle is reduced and the output power is reduced; this adjustment method can quickly and accurately respond to power adjustment needs and effectively change the output power of the photovoltaic power source;

[0142] S224. Feedback and iteration. After completing the power adjustment, the grid operation data, including node voltage and branch power, are immediately re-monitored, and the new monitoring data is fed back to the power flow matrix calculation link, the power flow matrix is ​​recalculated, and the grid operation status is analyzed again. If the grid still fails to reach the stable operation standard after adjustment (i.e., there are still abnormal node voltages or branch power), the above adjustment mechanism triggering, adjustment amount determination and power adjustment steps are repeated to form a closed-loop feedback control system. Through continuous iterative adjustment, the grid operation status gradually converges to a stable state, i.e., all node voltages are between the pre-set node voltage upper and lower limits, and all branch powers are less than or equal to the corresponding branch power transmission capacity limit, thereby achieving the goal of maintaining stable operation of the grid. This real-time feedback and continuous optimization mechanism can adapt to various dynamic changes in the grid operation process, ensure the coordinated operation of photovoltaic power sources and the grid, and improve the reliability and stability of the grid.

[0143] In S3, the power allocation model is determined to minimize the sum of the deviations between the output power of each photovoltaic array and the average power, that is, the objective function of the power allocation model is expressed as:

[0144]

[0145] Where f(P) represents the objective function of the power allocation model, m is the number of photovoltaic arrays connected to the distribution network; P i is the output power of the i-th photovoltaic array; is the average power; the significance of the objective function is to make the output power of each photovoltaic array as close to the average power as possible, so as to achieve fair distribution of power among various photovoltaic power sources, avoid over-generation or under-generation of some photovoltaic arrays, and improve the balance of photovoltaic power utilization in the entire distribution network;

[0146] The constraints of the objective function are:

[0147] The maximum power limit of the photovoltaic array. For each photovoltaic array, its output power must meet the following requirements:

[0148] 0≤P i ≤P imax ;

[0149] Where P imax is the rated maximum power of the PV array; this constraint is set based on the physical characteristics of the PV array itself and the requirements for safe operation;

[0150] The grid connection point voltage constraint requires that the grid connection point voltage of each photovoltaic array be strictly controlled within the set upper and lower limits, namely:

[0151] V min ≤V i ≤Vmax ;

[0152] V min is the preset node voltage lower limit, V max is the preset node voltage upper limit, V i is the grid connection point voltage of each photovoltaic array;

[0153] The transmission capacity of the collector line is limited. The transmission power of each collector line must meet the following requirements:

[0154] S k ≤S kmax ;

[0155] Where S k is the transmission power of each collector line (k is the collector line number), S kmax The maximum transmission capacity of the collector line k depends on factors such as the material, cross-sectional area, length and heat dissipation conditions of the line.

[0156] In S4, the interior point method is an effective numerical calculation method suitable for solving optimization problems with inequality constraints, and is used to solve the power output in the power distribution model; the basic principle is to introduce a barrier function inside the feasible domain to transform the constrained optimization problem into a series of unconstrained optimization problems for solution; for inequality constraints in the power distribution model, such as the maximum power limit of the photovoltaic array, the grid connection point voltage constraint and the collection line transmission capacity limit, the interior point method can cleverly handle these constraints and find the optimal or satisfactory power distribution solution under the premise of satisfying the constraints;

[0157] like Figure 3 As shown, the steps of solving the power output of the power allocation model using the interior point method include:

[0158] S41. Construct a barrier function, convert the constraints in the power allocation model into barrier terms, and add them to the power allocation model;

[0159] S411. For the maximum power limit of the photovoltaic array, when constructing the barrier function, it is converted into two inequality constraints g i1 (P i )=-P i ≤0 and g i2 (P i )=P i -P imax ≤0; where g i1 (P i ) and g i2 (P i ) represents the function part of the inequality constraint for the maximum power limit conversion of the photovoltaic array;

[0160] The barrier function term corresponding to the maximum power limit of the photovoltaic array is constructed accordingly:

[0161]

[0162] B i (P i ) represents the barrier function term corresponding to the maximum power limit of the i-th photovoltaic array;

[0163] Adding up these barrier function terms for all PV arrays yields the total barrier function part B associated with the maximum power limit constraint. max (P):

[0164]

[0165] Where P represents the vector containing all photovoltaic array power variables; the optimization objective function F(P) with maximum power limit is constructed:

[0166] F(P)=f(P)+μB max (P);

[0167] Where f(P) represents the objective function of the power allocation model and is defined as:

[0168]

[0169] P i is the output power of the i-th photovoltaic array; is the average power; μ is the barrier parameter. In the iterative solution process, when P i Close to 0 or P imax When , the barrier function value will increase rapidly, preventing the iteration point from violating the maximum power limit constraint;

[0170] S412. For the grid connection point voltage constraint, convert it into two inequality constraints g i3 (V i )=V i -V min ≥0 and; g i4 (V i )=V max -V i ≥0; g i3 (V i ) and g i4 (V i ) represents the function part of the inequality constraint for the grid connection point voltage constraint transformation;

[0171] Construct the barrier function term corresponding to the grid connection point voltage constraint:

[0172]

[0173] B iV (V i ) represents the barrier function term corresponding to the voltage constraint of the ith grid connection point;

[0174] The total barrier function corresponding to all grid connection point voltage constraints is:

[0175]

[0176] B V (V) represents the total barrier function corresponding to all grid connection point voltage constraints; V represents the vector containing all grid connection point voltage variables, V i is the voltage variable of the grid-connected point of the i-th photovoltaic array;

[0177] The optimization objective function F(V) with grid connection point voltage constraint is:

[0178] F(V)=f(V)+μB V (V);

[0179] f(V) is a voltage-related function determined according to the characteristics of the power grid; during the iteration process, the voltage at the grid connection point is ensured to meet the constraint requirements to prevent the voltage from exceeding the allowable range;

[0180] S413. For the transmission capacity limit of the collector line, for each collector line, its transmission power S k There is a functional relationship between S and the photovoltaic array power variable, that is, k =h k (P), h k (P) is the functional relationship between the transmission power of the collector line and the power variable of the photovoltaic array, which describes how the output power of the photovoltaic array affects the transmission power of the collector line. Its specific form is related to factors such as the topological structure of the power grid, the line connection method, and the electrical characteristics. Finally, the constraint is transformed into:

[0181] g k (p) = h k (p)-S kmax ≤0;

[0182] S kmax is the maximum transmission capacity of collector line k;

[0183] Construct the barrier function term corresponding to the transmission capacity limitation of the collector line:

[0184]

[0185] B k (P) the barrier function term corresponding to the transmission capacity limit of the kth collector line;

[0186] The total barrier function corresponding to the transmission capacity limit of all collector lines is:

[0187]

[0188] B s (P) represents the part of the total barrier function corresponding to the transmission capacity limit of all collector lines; l represents the number of collector lines;

[0189] S414. The complete new objective function F(P) finally constructed is:

[0190] F(P)=f(P)+μ(B max (P)+B V (V)+B S (P));

[0191] S42. Solve the unconstrained optimization problem, find the derivative of the new objective function F(P) with respect to the variable P, set the derivative to 0, and obtain a set of nonlinear equations;

[0192] The Newton's method is used to iterate the equations. The Newton's method is iteratively formulated as follows:

[0193]

[0194] Among them, P k and P k+1 They represent the power variable vectors of the current iteration point and the next iteration point in the Newton method iteration process, J F (P k ) is the value of F(P) in P k The Jacobian matrix at , whose elements are is F(P) in P k The gradient vector at , whose elements are Through continuous iteration, we gradually approach the optimal solution;

[0195] S43. Update the barrier parameter and check convergence. After each iteration, update the barrier parameter μ so that the influence of the barrier function is gradually weakened so as to get closer to the optimal solution of the original problem.

[0196] Check the convergence conditions, which include the change of the objective function value being less than the set threshold or the change of the variable being less than the set threshold;

[0197] If the convergence condition is not met, return to the step of solving the unconstrained optimization problem and continue the iteration; if the convergence condition is met, stop the iteration, and what is obtained at this time is the optimal or satisfactory power output solution of the power allocation model, that is, the optimal power allocation value of each photovoltaic array.

[0198] like Figure 4As shown, in S5, the implementation and adjustment of the power allocation scheme includes:

[0199] S51. Sending power allocation instructions and initial power adjustment, the power allocation scheme (i.e., the power setting value of each photovoltaic array) calculated by the optimization algorithm in S4 is quickly and accurately sent to the corresponding photovoltaic array controllers through a reliable communication network (such as optical fiber communication, wireless communication, etc.); these controllers are key control units for the interaction between the photovoltaic array and the power grid, and are responsible for receiving and executing power allocation instructions;

[0200] After receiving the power allocation instruction, the photovoltaic array controller immediately adjusts the output power of the photovoltaic inverter according to the instruction; the specific operation is to achieve power regulation by changing the control signal inside the inverter, such as adjusting the duty cycle of the pulse width modulation (PWM) signal; if the instruction requires an increase in the output power of a photovoltaic array, the controller will increase the duty cycle of the PWM signal, thereby increasing the amplitude of the inverter output voltage or current, thereby increasing the output power; conversely, if the power is required to be reduced, the duty cycle is reduced; this adjustment method can respond to the power allocation instruction quickly and accurately, so that the output power of the photovoltaic array is close to the set value;

[0201] S52. After completing the initial power adjustment, the voltage of the photovoltaic array grid connection point is monitored in real time; this monitoring process is achieved by a high-precision voltage sensor installed at the grid connection point, and the sensor feeds back the real-time voltage value to the photovoltaic array controller;

[0202] The controller compares the monitored grid connection point voltage with the preset safety voltage range [V min , V max ], if it exceeds the safety range, the controller will start the secondary power adjustment mechanism according to the voltage feedback; for example, if V i >V max , indicating that the voltage is too high, which will cause damage to the grid equipment. At this time, the controller will calculate the amount of power that needs to be reduced according to the pre-set voltage-power adjustment strategy, and then adjust the output power of the inverter accordingly to reduce the grid connection point voltage to a safe range; if V i <V min , then take the opposite adjustment strategy, increase power output to increase voltage;

[0203] Secondary power adjustment is a dynamic, real-time process. As long as the voltage at the grid connection point is not within the safe range, the controller will continue to adjust to ensure that the voltage is always stable within the safe range, while also ensuring the stable and coordinated operation of the photovoltaic power source and the power grid. This secondary adjustment mechanism based on voltage feedback can effectively respond to various dynamic changes in the operation of the power grid, such as load fluctuations, changes in light intensity, etc., which further improves the rationality of power distribution and the stability of power grid operation.

[0204] The above description is a preferred embodiment of the invention and is not intended to limit the invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the invention should be included in the protection scope of the invention.

Claims

1. A method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization, characterized in that: The following steps are involved: S1. System parameter initialization and data collection: initializing the parameters of the photovoltaic power system of the distribution network and collecting the operation data of the distribution network in real time; system parameter initialization ensures that the configuration of the photovoltaic power system of the distribution network matches the actual power grid; S2. Construction of power flow matrix and online feedback: Based on the topological structure of the distribution network and the real-time operation data collected in S1, the power flow matrix of the distribution network is constructed. The power flow matrix is ​​used to describe the distribution of power flow in the power grid; the power flow matrix is ​​analyzed, and the analysis results of the power flow matrix are used to adjust the output of the photovoltaic power source in real time through online power flow feedback to maintain the stable operation of the power grid; S3. Construction of power allocation model. The power allocation model is constructed based on the online power flow feedback information obtained in S2. The goal of the power allocation model is to achieve fair power distribution while meeting the operation constraints of the power grid. The maximum power limit of the photovoltaic array, the grid connection point voltage constraint, and the transmission capacity limit of the collector line are incorporated into the model. S4. Design and implement optimization algorithms, using interior point method to solve the power output of the power allocation model until the optimal or satisfactory power allocation solution is found; S5. Implementation and adjustment of the power allocation plan. The calculated power allocation plan is sent to each photovoltaic array controller. The photovoltaic array controller adjusts the output power of the photovoltaic inverter according to the received power allocation instruction, and monitors the voltage of the photovoltaic array grid connection point in real time. Based on the voltage feedback, the power is adjusted secondary to ensure that the voltage is within a safe range.

2. The method for fair distribution of photovoltaic power in a distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: In S1, the parameters of the photovoltaic power system include the access location, capacity, upper and lower limits of the grid-connected point voltage, and collector line impedance parameters of the photovoltaic array; the operation data of the power grid includes the active power, reactive power, and grid-connected point voltage measurement values ​​of the photovoltaic array.

3. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: The steps of constructing the power flow matrix in S2 include: S201. Establish a node admittance matrix, analyze the topological structure of the distribution network, determine the number of nodes, the connection relationship between nodes, and the parameters of each branch, the branch parameters include resistance and reactance; S202. Calculate the branch admittance between nodes. For each branch that has a connection relationship, calculate the branch admittance according to its impedance. The calculation formula is: In the formula, y ij is the branch admittance between node i and node j, R ij is the resistance between node i and node j, X ij is the reactance between node i and node j; S203. Calculate node self-admittance and mutual admittance: For each node i, the self-admittance Y ii The formula is: where y ij is the branch admittance between node i and node j, y i0 is the ground admittance of node i; The formula between node i and node j is: AND ij =and ij (i≠j); S204. Construct a node admittance matrix. According to the self-admittance and mutual admittance calculated in S23, construct an n×n node admittance matrix Y. The elements of the node admittance matrix are expressed as: Y ij [i,j=1,2,3,…n]; Among them, when i=j, it is self-admittance, and when i≠j, it is mutual admittance.

4. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: In S2, the step of analyzing the power flow matrix includes: S211. Calculate the node injection current vector, for access node a i (i=1,2,…,m) PV arrays, m represents the number of PV arrays connected to the distribution network, and the node injection current I ai The calculation formula is: Among them, P' is the collected active power; Q' i is the reactive power; is the actual voltage at the grid point V ai The complex conjugate of ai =V′ ai ×V b , V b is the reference voltage, V′ ai is the per unit value of the grid connection point voltage; According to the calculated injection currents of all nodes, the node injection current vector I = [I1, I2, ..., I n ] T , n is the number of nodes; S212. Solve the node voltage vector. According to the node voltage equation I=YV, solve the node voltage vector V=Y -1 I, get the node voltage vector V = [V1, V2, ..., V n ] T , Y -1 is the inverse matrix of the node admittance matrix Y.

5. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: In S2, the method of adjusting the output of the photovoltaic power source in real time by online power flow feedback using the analysis result of the power flow matrix includes: S221. Continuously monitor the key parameters calculated by the power flow matrix, including node voltage and branch power, compare the node voltage with the preset node voltage upper and lower limits, and compare the branch power with the branch power transmission capacity limit, and judge whether the node voltage is within the allowable range and whether the branch power does not exceed its transmission capacity through comparison; S222. Determine the adjustment strategy. When the node voltage is detected to exceed the preset node voltage upper and lower limits, or the branch power exceeds the branch power transmission capacity limit, the online power flow feedback adjustment mechanism is triggered; When the node voltage exceeds the limit, the power amount to be adjusted is calculated according to the pre-set voltage adjustment strategy; the pre-set voltage adjustment strategy calculates the adjustment amount according to the sensitivity relationship between voltage and power, that is, the power adjustment amount ΔP x , ΔP x =k v (V v -V ref ), where k v is the voltage-power sensitivity coefficient, V ref is the reference voltage value; S223. Implement power adjustment, and adjust the calculated power adjustment amount ΔP x The PV array controller receives the received signal and adjusts the power by changing the control signal of the PV inverter. The controller adjusts the duty cycle of the PWM signal of the inverter. If ΔP x >0, that is, the power needs to be increased, then the duty cycle is increased to increase the amplitude of the inverter output voltage or current, thereby increasing the output power; if ΔP x <0, the duty cycle is reduced and the output power is reduced; S224. Feedback and iteration. After completing the power adjustment, immediately re-monitor the grid operation data, including node voltage and branch power, and feed back the new monitoring data to the power flow matrix calculation link, recalculate the power flow matrix, and analyze the grid operation status again. If the grid still does not meet the stable operation standard after adjustment, repeat the adjustment mechanism triggering, adjustment amount determination and power adjustment steps to form a closed-loop feedback control system. Through continuous iterative adjustment, the grid operation status gradually converges to a stable state, that is, all node voltages are between the pre-set node voltage upper and lower limits, and all branch powers are less than or equal to the corresponding branch power transmission capacity limit.

6. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: In S3, the power allocation model is determined to minimize the sum of the deviations between the output power of each photovoltaic array and the average power, that is, the objective function of the power allocation model is expressed as: Where f(P) represents the objective function of the power allocation model, m is the number of photovoltaic arrays connected to the distribution network; P i is the output power of the i-th photovoltaic array; is the average power; The constraints of the objective function are: The maximum power limit of the photovoltaic array. For each photovoltaic array, its output power must meet the following requirements: 0≤P i ≤P i max ; Where P i max is the rated maximum power of the PV array; this constraint is set based on the physical characteristics of the PV array itself and the requirements for safe operation; The grid connection point voltage constraint requires that the grid connection point voltage of each photovoltaic array be strictly controlled within the set upper and lower limits, namely: In min ≤V i ≤V max ; V min is the preset node voltage lower limit, V max is the preset node voltage upper limit, V i is the grid connection point voltage of each photovoltaic array; The transmission capacity of the collector line is limited. The transmission power of each collector line must meet the following requirements: S k ≤S k max ; Where S k is the transmission power of each collector line, k is the collector line number, S k max The maximum transmission capacity of collector line k.

7. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1 is characterized in that: In S4, the step of using the interior point method to solve the power output of the power allocation model includes: S41. Construct a barrier function, convert the constraints in the power allocation model into barrier terms, add them to the power allocation model, and construct a complete new objective function through the constructed barrier function; S42. Solve the unconstrained optimization problem, find the derivative of the new objective function with respect to each photovoltaic array power variable, set the derivative to 0, and obtain a set of nonlinear equations; The Newton's method is used to iterate the equations. The Newton's method is iteratively formulated as follows: Among them, P k and P k+1 They represent the power variable vectors of the current iteration point and the next iteration point in the Newton method iteration process, J F (P k ) is the value of F(P) in P k The Jacobian matrix at , whose elements are is F(P) in P k The gradient vector at , whose elements are Through continuous iteration, we gradually approach the optimal solution; S43. Update the barrier parameter and check convergence. After each iteration, update the barrier parameter μ so that the influence of the barrier function is gradually weakened so as to get closer to the optimal solution of the original problem. Check the convergence conditions, which include that the change of the objective function value is less than the set threshold or the change of the variable is less than the set threshold; If the convergence condition is not met, return to the step of solving the unconstrained optimization problem and continue the iteration; if the convergence condition is met, stop the iteration, and what is obtained at this time is the optimal or satisfactory power output solution of the power allocation model, that is, the optimal power allocation value of each photovoltaic array.

8. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 7 is characterized in that: In S41, the constraints in the power allocation model are converted into obstacle terms, added to the power allocation model, and the steps of constructing a complete new objective function through the constructed obstacle function include: S411. For the maximum power limit of the photovoltaic array, when constructing the barrier function, it is converted into two inequality constraints g i1 (P i )=-P i ≤0 and g i2 (P i )=P i -P i max ≤0; where g i1 (P i ) and g i2 (P i ) represents the function part of the inequality constraint for the maximum power limit conversion of the photovoltaic array; The barrier function term corresponding to the maximum power limit of the photovoltaic array is constructed accordingly: B i (P i ) represents the barrier function term corresponding to the maximum power limit of the i-th photovoltaic array; Adding up these barrier function terms for all PV arrays yields the total barrier function part B associated with the maximum power limit constraint. max (P): Where P represents the vector containing all photovoltaic array power variables; the optimization objective function F(P) with maximum power limit is constructed: F(P)=f(P)+μB max (P); Where f(P) represents the objective function of the power allocation model and is defined as: Pi is the output power of the i-th photovoltaic array; is the average power; μ is the barrier parameter. In the iterative solution process, when P i Close to 0 or P i max When , the barrier function value will increase rapidly, preventing the iteration point from violating the maximum power limit constraint; S412. For the grid connection point voltage constraint, convert it into two inequality constraints g i3 (V i )=V i -V min ≥0 and; g i4 (V i )=V max -V i ≥0; g i3 (V i ) and g i4 (V i ) all represent the function part of the inequality constraint for the grid connection point voltage constraint transformation; Construct the barrier function term corresponding to the grid connection point voltage constraint: B iV (V i ) represents the barrier function term corresponding to the voltage constraint of the ith grid connection point; The total barrier function corresponding to all grid connection point voltage constraints is: B V (V) represents the total barrier function corresponding to all grid connection point voltage constraints; V represents the vector containing all grid connection point voltage variables, V i is the voltage variable of the grid-connected point of the i-th photovoltaic array; The optimization objective function F(V) with grid connection point voltage constraint is: F(V)=f(V)+μB V (V); f(V) is a voltage-related function determined according to the characteristics of the power grid; during the iteration process, the voltage at the grid connection point is ensured to meet the constraint requirements to prevent the voltage from exceeding the allowable range; S413. For the transmission capacity limit of the collector line, for each collector line, there is a functional relationship between its transmission power Sk and the photovoltaic array power variable, that is, S k =h k (P), h k (P) is the functional relationship between the transmission power of the collector line and the power variable of the photovoltaic array, which describes how the output power of the photovoltaic array affects the transmission power of the collector line. Its specific form is related to the topological structure of the power grid, the line connection method and the electrical characteristics. Finally, the constraint is transformed into: g k (p)=h k (p)-S k max ≤0; S k max is the maximum transmission capacity of collector line k; Construct the barrier function term corresponding to the transmission capacity limitation of the collector line: B k (P) the barrier function term corresponding to the transmission capacity limit of the kth collector line; The total barrier function corresponding to the transmission capacity limit of all collector lines is: B s (P) represents the part of the total barrier function corresponding to the transmission capacity limit of all collector lines; l represents the number of collector lines; S414. The complete new objective function F(P) finally constructed is: F(P)=f(P)+μ(B max (P)+B V (V)+B S (P))。 9. The method for fair distribution of photovoltaic power in distribution network based on online power flow feedback optimization according to claim 1, characterized in that: In S5, the implementation and adjustment of the power allocation scheme includes: S51. Sending power allocation instructions and initial power adjustment, sending the power allocation scheme calculated by the optimization algorithm in S4 to the corresponding photovoltaic array controller through the communication network; After receiving the power allocation instruction, the PV array controller immediately adjusts the output power of the PV inverter according to the instruction; power regulation is achieved by adjusting the duty cycle of the pulse width modulation (PWM) signal. If the instruction requires an increase in the output power of a certain PV array, the controller will increase the duty cycle of the PWM signal, thereby increasing the amplitude of the inverter output voltage or current, thereby increasing the output power; conversely, if the power is required to be reduced, the duty cycle will be reduced; S52. After completing the initial power adjustment, the voltage of the photovoltaic array grid connection point is monitored in real time; this monitoring process is achieved by a high-precision voltage sensor installed at the grid connection point, and the sensor feeds back the real-time voltage value to the photovoltaic array controller; The controller compares the monitored grid connection point voltage with the preset safety voltage range [V min , V max ] is compared. If it exceeds the safety range, the controller will start the secondary power adjustment mechanism according to the voltage feedback; if V i >V max , indicating that the voltage is too high, which will cause damage to the grid equipment. At this time, the controller will calculate the amount of power that needs to be reduced according to the pre-set voltage-power adjustment strategy, and then adjust the output power of the inverter accordingly to reduce the grid connection point voltage to a safe range; if V i <V min , the opposite adjustment strategy is adopted, increasing power output to increase voltage.

Images