A power flow scheduling method based on energy storage converter control

Abstract

The present application relates to the technical field of power flow scheduling, in particular to a power flow scheduling method based on energy storage converter control, comprising: S1, using an improved forward-backward substitution method to perform power flow calculation on a distribution network to obtain power flow information of each node of the distribution network; S2, constructing a scheduling model, and scheduling and solving the power flow information based on a comprehensive objective function of the scheduling model to obtain a scheduling scheme; the comprehensive objective function comprises a power company side penalty term and a power user side penalty term, the power company side penalty term represents the calling degree of controllable resources, and the power user side penalty term represents the comfort degree of user load; S3, obtaining device-level instructions based on the scheduling scheme, and using a hybrid energy storage converter composed of a front-stage DC / DC converter and a rear-stage converter to cooperatively perform power flow scheduling on the distribution network based on the device-level instructions, so as to realize the collaborative consideration of reasonable calling of controllable resources of the power grid and user power consumption comfort.

Description

Technical Field

[0001] This invention relates to the field of power flow scheduling technology, specifically a power flow scheduling method based on energy storage converter control. Background Technology

[0002] Against the backdrop of global energy transition and carbon neutrality goals, the development and utilization of renewable energy has become a core direction for energy sector development. Solar energy, as one of the most promising renewable energy sources, is significantly affected by environmental factors such as sunlight and weather, naturally exhibiting randomness and volatility. With a high proportion of distributed power sources connected to the grid, the dynamic changes in power flow in the distribution network become significant, gradually increasing the pressure on system operation and regulation.

[0003] To buffer fluctuations in renewable energy output and enhance the flexible adjustment capabilities of the distribution network, energy storage systems are used in conjunction with photovoltaic units. Leveraging the charging and discharging regulation characteristics of energy storage, they help maintain stable regional power supply. Energy storage converters, as key devices for source-load interaction and power conversion, can achieve AC / DC power conversion and flexible power regulation, and have become indispensable core control devices in new distribution network architectures. In the development of new power systems, conducting rational scheduling based on power flow calculations is a crucial link in ensuring the continuous and stable operation of the distribution network.

[0004] Currently, most existing power dispatch schemes for distribution networks are designed from the perspective of power grid operation and management only. The overall dispatch logic is one-sided, generally ignoring the actual electricity consumption experience of users and load regulation constraints. The singular control approach is difficult to adapt to the operation scenario of distribution networks with multiple sources and loads interacting. This easily leads to problems such as poor implementation of dispatch strategies and mismatch between supply and demand. It is also impossible to take into account both the needs of grid operation and management and the electricity consumption experience of users in an environment with high penetration of new energy. Summary of the Invention

[0005] The purpose of this invention is to overcome the technical problem that existing scheduling strategies are limited to single-sided grid optimization, and to provide a power flow scheduling method based on energy storage converter control. This method combines precise power flow calculation to construct a two-dimensional comprehensive optimization objective, and relies on hybrid energy storage converters to complete command-coordinated control, thereby achieving a balance between the rational use of controllable grid resources and user power comfort, and meeting the actual needs of stable and reasonable operation of distribution networks under high-proportion photovoltaic access.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: a power flow scheduling method based on energy storage converter control, comprising: S1, performing power flow calculation on the distribution network using an improved forward-backward substitution method to obtain power flow information of each node in the distribution network; S2, constructing a scheduling model and solving the power flow information based on the comprehensive objective function of the scheduling model to obtain a scheduling scheme; the comprehensive objective function includes a power company-side penalty term and a power user-side penalty term, wherein the power company-side penalty term characterizes the degree of controllable resource utilization and the power user-side penalty term characterizes the user load comfort level; S3, obtaining equipment-level instructions based on the scheduling scheme, and using a hybrid energy storage converter composed of a front-end DC / DC converter and a back-end converter to collaboratively perform power flow scheduling on the distribution network based on the equipment-level instructions.

[0007] Optionally, the comprehensive objective function of the scheduling model is set in the following manner, including: S21, setting a power company penalty coefficient with the equipment regulation potential of the distribution network as the optimization objective; setting a power user penalty coefficient with the user comfort of the distribution network as the optimization objective; S22, obtaining a power company-side penalty term based on the power company penalty coefficient and the control cost unit price of controllable resources; obtaining a power user-side penalty term based on the power user penalty coefficient and the comfort penalty cost unit price; S23, obtaining the comprehensive objective function based on the sum of the power company-side penalty term and the power user-side penalty term.

[0008] Optionally, the step of setting the power company penalty coefficient based on the equipment regulation potential of the distribution network as the optimization target includes: S21-1, setting the power company penalty coefficient based on the regulation potential of controllable resources and the degree to which the target regulation power deviates from its regulation upper and lower limits; the step of setting the power user penalty coefficient based on the user comfort of the distribution network as the optimization target includes: S21-2, setting the power user penalty coefficient based on the regulation range of controllable resources and the degree to which the target regulation power deviates from its comfort range.

[0009] Optionally, the step of using the improved forward-backward substitution method to perform power flow calculation on the distribution network includes: S11, selecting branch current as the iteration variable of the improved forward-backward substitution method, and correcting the node number and node type to obtain the corrected node; S12, using the improved forward-backward substitution method to perform power flow calculation on the corrected node to obtain the power flow information of each node in the distribution network, wherein the power flow information includes voltage amplitude, voltage phase and power distribution.

[0010] Optionally, the node numbering can be corrected in the following ways: S11-1, the branches of the distribution network are numbered sequentially to obtain the branch number; S11-2, the branch numbering is performed sequentially from the beginning to the end of each branch according to the branch number to obtain the corrected node number.

[0011] Optionally, the node types include PQ nodes, PV nodes, and PI nodes; the node types are corrected in the following ways: S11-3, in the iteration, the reactive power of the PV node is corrected by the voltage amplitude change to obtain the reactive power correction value of the PV node; S11-4, in the iteration, the reactive power correction value of the PI node is calculated by the branch current and branch voltage.

[0012] Optionally, the step of using the improved forward-backward substitution method to perform power flow calculation on the corrected nodes includes: S12-1, performing back-substitution calculation on the inflow current of each corrected node from the load end node to the root node to obtain the branch current; S12-2, performing forward calculation from the root node to the load end node based on the branch current to obtain the node voltage; S12-3, after each iteration calculation, calculating the power difference of all nodes to determine the convergence degree of the iteration result. If the real part or imaginary part of the iteration result does not meet the convergence condition, then returning to re-execute step S12-1 until the real part and imaginary part simultaneously meet the convergence condition to obtain the power flow information of each node in the distribution network.

[0013] Optionally, the step of obtaining equipment-level instructions based on the scheduling scheme includes: S31, using a distribution network power flow tracing method based on proportional sharing to determine the electrical association between distributed power sources and loads in the distribution network; S32, allocating power adjustment amounts based on the electrical association to obtain load nodes and corresponding power allocation factors, and then obtaining equipment-level instructions for power flow scheduling.

[0014] Optionally, the downstream converter is a voltage source inverter; power flow scheduling of the distribution network is performed in the following ways: S33, when the distribution network is connected to the grid, the voltage source inverter responds to the equipment-level command to perform power flow scheduling; S34, when the distribution network is disconnected from the grid, the hybrid energy storage converter adopts virtual synchronous machine control and responds to the equipment-level command to perform power flow scheduling.

[0015] Optionally, the front-end DC / DC converter is electrically connected to the photovoltaic array and the energy storage unit respectively to perform power regulation and charge / discharge control on the photovoltaic array and the energy storage unit; the back-end converter is electrically connected to the distribution network to convert DC power into AC power that is compatible with the frequency of the distribution network.

[0016] The beneficial effects of this invention are as follows: 1. An improved forward-backward substitution method is used to calculate the power flow of the distribution network, accurately obtaining the power flow information of each node in the distribution network, providing accurate and reliable data support for subsequent power flow scheduling. 2. By constructing a comprehensive objective function that includes penalties on the power company side and penalties on the power user side, where the power company side penalty represents the degree of controllable resource utilization and the power user side penalty represents the user's load comfort level, the scheduling model can take into account both the power company's control needs and the user's electricity experience, overcoming the problem that existing scheduling methods only focus on the interests of a single entity, making it difficult to implement scheduling schemes. 3. A hybrid energy storage converter consisting of a front-end DC / DC converter and a back-end converter is used to coordinate power flow scheduling based on equipment-level instructions obtained from the scheduling scheme, achieving precise adjustment and stable control of power flow scheduling.

[0017] 2. This application employs an improved forward-backward substitution method, using branch current as the iterative variable to correct node numbers and node types, accurately calculating power flow information for each node in the distribution network. This overcomes the problem that traditional power flow calculations cannot uniformly handle multiple node types such as PQ, PV, and PI, providing accurate data support for power flow scheduling. By constructing a comprehensive objective function that includes penalties from both the power company and the power user side, and combining this with reasonably set penalty coefficients, the application balances the degree of access to controllable resources by the power company with the load comfort level of power users. This overcomes the problem that existing scheduling methods only focus on the interests of a single entity, achieving a balance between safe grid operation and user experience. This system utilizes a hybrid energy storage converter comprised of a front-end DC / DC converter and a back-end voltage source inverter to collaboratively execute device-level commands. The front-end DC / DC converter regulates the power and controls the charging and discharging of the photovoltaic array and energy storage units, while the back-end inverter converts DC power to AC power. It supports both grid-connected and off-grid operation modes. In off-grid operation, a virtual synchronous machine control ensures system stability, while in grid-connected operation, it achieves peak shaving and valley filling, guaranteeing stable grid operation under various conditions. Combined with a proportional sharing-based distribution network power flow tracing method, it clarifies the electrical relationship between distributed power sources and loads and rationally allocates power regulation, effectively mitigating power fluctuations caused by high-proportion photovoltaic (PV) integration, improving PV absorption rates, reducing power company control costs, and ensuring the safe and stable operation of the distribution network in scenarios with high-penetration renewable energy integration. Attached Figure Description

[0018] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. The drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0019] Figure 1This is a flowchart of a power flow scheduling method based on energy storage converter control in this invention; Figure 2 This is a schematic diagram of the main circuit topology of a hybrid energy storage converter in this invention; Figure 3 This is a schematic diagram illustrating the control of cost changes when a power company reduces controllable resources in this invention. Figure 4 This is a schematic diagram illustrating the control of cost changes when a power company increases controllable resources in this invention. Figure 5 This is a schematic diagram illustrating the change in control costs when a power user reduces controllable resources in this invention. Figure 6 This is a schematic diagram illustrating the change in control costs when a power user increases controllable resources in this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only one preferred embodiment of this invention and are only used to explain this invention. They do not limit the scope of protection of this invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] As one implementation method, such as Figure 1 As shown, a power flow scheduling method based on energy storage converter control includes: S1. An improved forward-backward substitution method is used to perform power flow calculations on the distribution network to obtain power flow information of each node in the distribution network.

[0022] S2. Construct a scheduling model, and solve the power flow information for scheduling based on the comprehensive objective function of the scheduling model to obtain a scheduling scheme; the comprehensive objective function includes a power company-side penalty term and a power user-side penalty term, the power company-side penalty term characterizes the degree of controllable resource utilization, and the power user-side penalty term characterizes the user load comfort level.

[0023] S3. Based on the scheduling scheme, obtain equipment-level instructions, and use a hybrid energy storage converter composed of a front-end DC / DC converter and a back-end converter to coordinate power flow scheduling of the distribution network based on the equipment-level instructions.

[0024] It should be noted that controllable resources refer to adjustable resources such as energy storage units and adjustable output of new energy sources within the distribution network. User load comfort refers to the electricity experience index corresponding to fluctuations in user electricity load and power regulation constraints. Equipment-level commands are low-level execution commands generated based on the dispatch scheme and directly adapted to the operation control of hybrid energy storage converters. Hybrid energy storage converters refer to two-stage photovoltaic-energy storage co-converter devices composed of a front-end DC / DC converter and a back-end converter. Co-conversion refers to the mutual cooperation between the front-end DC-side power regulation and the back-end grid-connected transformer regulation to jointly complete the operation of power flow optimization and regulation.

[0025] As one implementation method, step S1 includes: S11, selecting branch current as the iterative variable of the improved forward-backward substitution method, and correcting the node number and node type to obtain the corrected node. The node number is corrected in the following way: S11-1. The branches of the distribution network are numbered sequentially to obtain the branch numbers.

[0026] S11-2. Based on the branch number, sequentially number the branch from the beginning to the end of each branch to obtain the corrected node number.

[0027] The node types include PQ nodes, PV nodes, and PI nodes. The node types are corrected in the following ways: S11-3, during the iteration, the reactive power of the PV node is corrected by the change in voltage amplitude to obtain the corrected reactive power value of the PV node.

[0028] S11-4. In the iteration, the reactive power correction value of the PI node is calculated by using the branch current and branch voltage.

[0029] S12. An improved forward-backward substitution method is used to perform power flow calculations on the corrected nodes to obtain power flow information for each node in the distribution network. The power flow information includes voltage amplitude, voltage phase, and power distribution. Specifically, power flow calculations are performed on the corrected nodes in the following manner: S12-1. The inflow current of each corrected node is calculated backward from the load end node towards the root node to obtain the branch current.

[0030] S12-2. Based on the branch current, perform forward calculations from the root node towards the load end node to obtain the node voltage.

[0031] S12-3. After each iteration calculation, calculate the power difference of all nodes to determine the convergence degree of the iteration result. If the real part or imaginary part of the iteration result does not meet the convergence condition, return to step S12-1 and repeat until the real part and imaginary part simultaneously meet the convergence condition to obtain the power flow information of each node of the distribution network.

[0032] Each node is numbered, and the different branches of the low-voltage distribution network are numbered using a depth-first traversal. Corresponding branch numbers are generated for each branch, and corresponding head and tail bus numbers are generated for each branch. Then, a program is written to distinguish whether each node is a terminal node. For both terminal and non-terminal nodes, branch characteristic matrices are established, consisting of the branch resistance, branch reactance, and active and reactive power at the head and tail of each branch. This depth-first traversal statistical method is used to number each node throughout the processing of different node types, achieving uniform processing for all nodes and laying the foundation for the iterative loop of the main program.

[0033] Power flow calculation and analysis are core components of the planning and operation of high-penetration low-voltage distribution networks. By accurately solving for the voltage amplitude, phase, and power distribution at each node in the network, problems such as power flow reversal and voltage limit violations caused by distributed generation can be quantified, providing fundamental data for subsequent voltage and frequency control. Accurate power flow analysis can reveal the power flow patterns of the distribution network under different operating conditions, identify weak nodes and overload risks, and is a prerequisite for formulating power flow optimization strategies. Simultaneously, its results can support the design of voltage regulation schemes (such as reactive power compensation and energy storage charging and discharging) and frequency stability control (such as virtual synchronous machine strategies), laying a theoretical foundation for the safe and economical operation of distribution networks with high-penetration renewable energy integration.

[0034] The power flow calculation approach for a radial low-voltage distribution network is as follows: First, the transmission power of each line is calculated backwards, starting from the load-end nodes of the distribution network and working towards the power source side (root node), calculating the power distribution on each line, including active and reactive power. Then, starting from the power source side (root node) and working towards the load end, the voltage of each node is calculated. Using the current and impedance voltage drop on the line, the voltage of the downstream node is calculated from the voltage of the upstream node, and the voltage is decomposed into a real part e and an imaginary part f to facilitate iterative calculation. After each round of calculation, the change in node voltage is observed. When the maximum change in voltage of all nodes is less than a set threshold, such as 0.001, the power flow calculation is considered to have converged, and the iteration stops.

[0035] Power flow calculation in microgrids is a core analytical method for determining the voltage of each node, branch power, and losses within the system, used to understand the actual operating state of the power grid. The forward-backward substitution method is a classic power flow algorithm applicable to radial microgrids. Its process involves: first, calculating branch currents backward from the load end to the root node; then, calculating node voltages forward from the root node to the load end, iterating repeatedly until voltage convergence. Traditional forward-backward substitution methods can only calculate PQ nodes with ordinary loads. This application adopts an improved forward-backward substitution method, no longer using branch power as the iteration variable, but instead using branch current. Through node numbering and type correction, it achieves unified solutions for multiple node types (PQ, PV, PI), improving the accuracy and applicability of power flow calculations under high-proportion photovoltaic (PV) integration.

[0036] The basic process and steps of the forward-backward substitution method in microgrid power flow calculation are as follows: Calculate the inflow current to each node. Based on the calculation results, perform a backward substitution, calculating the branch current from the load end towards the root node, starting from the last branch and stopping the calculation when reaching the root node. Based on the calculation results, perform a forward calculation to obtain the node voltages, following the reverse calculation order from the root node towards the last branch. The convergence of the calculation iteration results is assessed on the [number]th iteration. After each iteration, the convergence of the iteration result is determined by calculating the power difference of all nodes. If the real or imaginary part of the result does not meet the convergence condition, the forward and backward iteration steps described above are repeated until both the real and imaginary parts meet the convergence condition. At this point, the forward and backward iteration calculation is stopped, as shown in the following formula: In the formula: For the first The nodes obtained in the next iteration The inflow current, Indicates the first The nodes obtained in the next iteration voltage value, Represents a node The inflow power, Represents a node Parallel admittance; For the first Branches during the next iteration The current flowing in from above, For the first Node at the next iteration The current flowing into it, For nodes The sum of the currents on each subsequent branch, M is the current at the node. The set of all connected branches; Indicates the first The nodes obtained from the next iteration voltage, Indicates the first The nodes obtained from the next iteration voltage, Indicates a branch The impedance value; Indicates the first Next iteration node The power difference Represents a node voltage value, To indicate complex conjugate, Represents a node The original injection power.

[0037] In each forward-backward iteration, the reactive power of the PQ node is dynamically adjusted according to the node voltage; in each forward-backward iteration, the reactive power of the PV node can be corrected by changes in voltage amplitude; in each forward-backward iteration, the reactive power of the PI node is obtained by solving for branch current and active power, thus realizing unified power flow calculation for the three types of distributed units. The reactive power correction equations for the iterative processes of PQ, PV, and PI nodes are as follows: In the formula, This indicates that the PQ node is at the 1st rank. The reactive power of the next forward iteration. This represents the node voltage of the PQ node; Indicates the PV node at the 1st The reactive power of the next forward iteration. Indicates the first The reactive power of the next forward iteration. Indicates the first The voltage amplitude change at the corresponding node during the next forward and backward iterations; Indicates the PI node at the 1st epoch. The reactive power of the next forward iteration. Indicates the first The corresponding branch current after the next forward pushback substitution. This represents the active power of the corresponding PI node. Indicates the first The voltage amplitude of the corresponding PI node after the next iteration.

[0038] As one implementation method, step S2 further includes: setting the comprehensive objective function of the scheduling model in the following ways: S21, setting a power company penalty coefficient with the equipment regulation potential of the distribution network as the optimization objective; setting a power user penalty coefficient with the user comfort of the distribution network as the optimization objective. Further, setting a power company penalty coefficient based on the regulation potential of controllable resources and the degree to which the target regulation power deviates from its regulation upper and lower limits. Setting a power user penalty coefficient based on the regulation range of controllable resources and the degree to which the target regulation power deviates from its comfort range.

[0039] S22. Obtain the power company-side penalty item based on the power company's penalty coefficient and the unit price of controllable resource control costs; obtain the power user-side penalty item based on the power user's penalty coefficient and the unit price of comfort penalty costs.

[0040] S23. The comprehensive objective function is obtained based on the sum of the penalty terms on the power company side and the penalty terms on the power user side.

[0041] The optimization functions include the power company's optimization function and the power user's optimization function. Assume that... At any given time, in response to load control targets, controllable resources with greater downward adjustment potential are required to undertake more load reduction tasks, while those with greater upward adjustment potential are required to undertake more load increase tasks. The power company aims to maximize the use of controllable resources and, during the target allocation process, requires users to assume their maximum adjustable capacity. (Refer to...) Figure 3 , 4 This reflects the change in control cost as the relative position of the response target from the upper and lower boundaries changes. It can be seen that the control cost exhibits a linear trend with the change of the response target. Target (obj): Minimizes the sum of deviations between the total controllable power of the entire network and the target value. The smaller the deviation, the more accurately the dispatch instructions are tracked, and the lower the cost. The power company's optimization function is as follows. In the formula, express Time needs to be allocated to the first The target adjustable power of the controllable resource is the independent variable in the optimization function. express Time of the first Historical output power of controllable resources Indicates the first on the power grid side The unit price of controllable resources. These represent the zero-prevention bias constants in the upward and downward directions, respectively, and are pre-specified small fixed values ​​to ensure... It is always a positive value. They represent Time of the first The upper and lower limits of power for controllable resources are used to constrain load increases or decreases. express Time of the first The penalty coefficient for controllable resources, as a factor in the formula, determines the priority of that resource in target allocation. The denominator is the total adjustable potential of the equipment, a fixed normalization factor. The numerator is the distance between the target value and the boundary threshold. The farther the target value is from the boundary threshold, the smaller the numerator and the smaller the penalty coefficient. Resources closer to their upper limit have larger penalty coefficients, higher costs, and are more likely to be allocated less frequently, with priority given to other relatively cheaper / more adjustable resources. Conversely, smaller penalty coefficients have lower costs, are more likely to be allocated more frequently, and are more suitable for handling more response targets. By setting the penalty coefficient, a scheduling strategy that avoids extremes and prioritizes moderate allocation is achieved.

[0042] Reference Figure 5 , 6 This reflects the change in control cost as the relative position of the response target from the upper and lower boundaries changes. It can be seen that the control cost exhibits a non-linear trend with the change of the response target. Target (obj): Minimizes the sum of deviations between the total controllable power of the entire network and the target value. The smaller the deviation, the more accurately the dispatch instructions are tracked, and the lower the cost. The power user optimization function is as follows: In the formula, Indicates the user side The unit price of comfort penalty for controllable resources. express Time of the first The penalty coefficient for power users of controllable resources, as a penalty coefficient in the formula, determines the priority of that controllable resource in target allocation. The numerator is the total adjustable potential of the equipment, a fixed normalization factor. The denominator is the distance between the calculated target value and the boundary threshold. The farther away from the boundary threshold, the larger the denominator, the smaller the penalty coefficient, the lower the cost, and the more willing users are to cooperate with this adjustment. The closer a resource is to its upper limit, the smaller the denominator, the larger the penalty coefficient, the higher the cost, and the less willing users are to cooperate with this adjustment. By setting the penalty coefficient, a scheduling strategy that avoids limits and prioritizes moderate allocation is achieved.

[0043] The comprehensive optimization function can be: The optimization functions include the power company's optimization function and the power user's optimization function. The constraints of the optimization functions are as follows: power balance constraint (st), the total power allocated to each resource must be equal to the total target demand on the user side, and the resource adjustment amount cannot exceed the physical limits (upper and lower limits), as follows: In the formula, express The overall goal of participating in goal allocation at all times. They represent Time of the first The upper and lower limits of power for controllable resources are used to constrain load increases or decreases.

[0044] The power company's perspective coefficient, focusing on equipment regulation potential, prioritizes resources with ample regulation space to avoid operating equipment under extreme conditions and ensure grid safety and stability. The user's perspective coefficient, focusing on user comfort, prioritizes regulation amounts with minimal impact on users, avoiding adjusting equipment to unacceptable extreme conditions. In the optimization process, both the power company's penalty coefficient and the power user's penalty coefficient are incorporated as penalty weights into the objective function. By finding the scheduling scheme with the minimum overall cost, an automatic balance is achieved between grid-side economy, security, and user-side comfort.

[0045] As one implementation method, step S3 further includes: S31, using a distribution network power flow tracing method based on proportional sharing to determine the electrical association between distributed power sources and loads in the distribution network.

[0046] S32. Based on the electrical association power adjustment amount, the load nodes and corresponding power allocation factors are obtained, and then the equipment-level instructions are issued to each node for power flow scheduling.

[0047] S33. When the distribution network is connected to the grid, the voltage source inverter responds to the equipment-level command to perform power flow scheduling.

[0048] S34. When the distribution network is off-grid, the hybrid energy storage converter uses virtual synchronous machine control to perform power flow scheduling in response to the equipment-level commands. Furthermore, the front-end DC / DC converter is electrically connected to both the photovoltaic array and the energy storage unit to perform power regulation and charge / discharge control on both. The rear-end converter is electrically connected to the distribution network to convert DC power into AC power compatible with the distribution network's frequency.

[0049] Specifically, refer to Figure 2This invention utilizes a photovoltaic hybrid energy storage system for power flow scheduling. The photovoltaic hybrid energy storage system mainly consists of a photovoltaic array, energy storage units, a hybrid energy storage converter, and a distribution network. The energy storage units include battery banks and supercapacitors, providing energy storage for the system. The hybrid energy storage converter adopts a two-stage topology structure: a front-end bidirectional DC / DC converter and a back-end voltage source inverter (converter). It is the core control and conversion unit of the entire system. On one hand, the front-end DC / DC converter enables power regulation and charging / discharging control of the photovoltaic array, battery banks, and supercapacitors. On the other hand, the back-end inverter converts DC power into AC power compatible with the grid frequency. The voltage source inverter, acting as an uninterruptible power supply, effectively supports stable system operation when connected to the power system. When the energy storage system is off-grid, it uses virtual synchronous machine control, equivalent to a voltage source independently supporting stable system operation, thus achieving grid-connected / off-grid operation control.

[0050] Figure 2 middle, and These are the three-phase voltages and three-phase currents on the AC side of the converter. This is the filtered current. This is the grid voltage. For bus voltage and bus current, For filter inductors and filter capacitors, The equivalent resistance of the line. These are the six switching transistors in the upper and lower arms of the subsequent converter. For DC-side output current and converter input current, It is a voltage regulator capacitor. and These represent the output voltage and output current across the battery, photovoltaic array, and capacitor, respectively. and These are two pairs of complementary conducting switching transistors. For the switching transistors in the front-end Boost converter of the photovoltaic module, Let be the inductance of the battery, photovoltaic array, and capacitor. Assuming the grid voltage waveform is symmetrical and undistorted, according to Kirchhoff's laws, a mathematical model of the hybrid energy storage converter in the three-phase stationary coordinate system of phases a, b, and c can be obtained. In the formula, This refers to the voltage across capacitor C in the AC side filter circuit. This is the control signal for the three-phase bridge arm switching transistors in the subsequent converter (subsequent converter).

[0051] By performing the Park transformation to convert the three-phase stationary coordinate system into a two-phase rotating coordinate system, and converting the three-phase AC quantities into DC quantities in the synchronous rotating coordinate system, the mathematical model of the hybrid energy storage converter in the dq rotating coordinate system can be obtained: In the formula: The angular frequency of the grid voltage. and These represent the grid voltage, grid current, capacitor voltage, and switch control signal for the dq axes, respectively.

[0052] The state equation of the front-end DC / DC converter is: In the formula: The inductors are for the front-end DC converter side of the battery, photovoltaic array, and capacitor. It is a voltage regulator capacitor. They are switching transistors On duty cycle, switching transistor The off duty cycle. and These represent the output voltage and output current across the battery, photovoltaic array, and capacitor, respectively. This is the bus voltage.

[0053] For renewable energy utilization, a distribution network power flow tracing method based on proportional sharing is adopted to determine the electrical relationship between distributed generation (photovoltaics) and loads in the distribution network, and to dynamically allocate power regulation. This method can effectively mitigate node voltage deviations and power fluctuations caused by distributed photovoltaics, maximize photovoltaic absorption, and improve the utilization rate of new energy. The solution of the Newton-Raphson three-phase power flow in a distribution network containing renewable energy is ultimately transformed into solving the following equations. In the formula: This represents the three phases a, b, and c. For Jacobian matrices, Let be the column vector of three-phase active and reactive power mismatch at the node. Let be the corrected column vector of the three-phase voltage at each node. Use this corrected column vector to correct the voltage values ​​at each node, and proceed to the next iteration until the convergence criterion is met. Based on the above power flow solution method, relevant power flow information of the distribution network can be obtained and applied to the distribution network power flow tracing algorithm. Define the distribution network node label set, ... Power output aggregation of distributed photovoltaic systems at all times Time Node The set of power allocation factors generated by distributed photovoltaic power generation and the set of controllable loads on the demand side of distribution network access are as follows. In the formula, Represents the set of distribution network node labels; This represents the combined output of a distributed photovoltaic system. express Power output aggregation of distributed photovoltaic systems at all times express Time-based load nodes Distributed photovoltaic power output, This indicates the number of nodes connected to the distributed photovoltaic system. express Time Node The set of power allocation factors generated by distributed photovoltaic power. express Time Node Distributed photovoltaic nodes Demand-side controllable load power allocation factor. This represents the set of controllable loads on the demand side of the distribution network. Indicates at photovoltaic nodes The demand-side controllable load of the access network This indicates the number of nodes connected to controllable loads in the distribution network. This indicates the total number of nodes in the distribution network.

[0054] This application employs a distribution network power flow tracing method based on the proportional sharing principle to determine the electrical association between power sources and loads. Forward distribution network power flow tracing is performed from the power source node to obtain the associated load nodes and their corresponding power allocation factors. Based on existing photovoltaic data, this method is applied to the aforementioned distributed photovoltaic systems. Select nodes Taking the research object as the subject, calculate Time Node Changes in photovoltaic power output The power distribution system dispatch center can obtain corresponding adjustable upper and lower limits by collecting the status of various controllable loads; combined with... Time Node The magnitude of the change in photovoltaic output ultimately determines the total regulation required by the controllable load on the demand side. In the formula, express The difference between the change in photovoltaic power output and the load regulation capability at any given time. Represents a node Adjustable amount of controllable load, This indicates controllable load on the demand side; This indicates the number of controllable load types on the demand side. Represents a node No. The maximum and minimum adjustable amounts of demand-side controllable load.

[0055] When the difference between the change in photovoltaic output and the load regulation capacity is greater than 0, the load regulation required to smooth out the fluctuations in photovoltaic output is greater than the upper limit of the sum of the load regulation capacity. At this time, the regulation capacity undertaken by the load is: .

[0056] When the difference between the change in photovoltaic output and the load regulation capacity is equal to 0, that is, when the required load regulation is within the sum of the adjustable amounts of controllable load on the demand side, the load regulation amount is: .

[0057] In the formula, express Time-based load nodes Distributed photovoltaic power output, express Time Node Distributed photovoltaic nodes The demand-side controllable load power allocation factor. When the difference between the photovoltaic output change and the load regulation capacity is less than 0, that is, when the required load regulation is less than the lower limit of the sum of the controllable loads on the demand side, the regulation undertaken by the load is: .

[0058] A single node typically contains various types of responsive power electronic components within its controlled area, such as energy storage units, energy storage converters, and photovoltaic inverters. Optimized scheduling is required to assign control signals to each of these controllable components. The goal of optimized scheduling is to minimize the total control cost while ensuring that the assigned response targets meet the grid-side objectives and the upper and lower boundary constraints of the response load.

[0059] This application utilizes coordinated control between the front-end DC / DC converter and the back-end converter to effectively address the randomness and volatility issues arising from high-proportion renewable energy integration. It achieves peak and valley power regulation during grid connection and provides voltage support during off-grid operation, ensuring stable grid operation. In power flow calculation, an improved forward-backward substitution method is employed, selecting branch current as the iterative variable. After node numbering and type correction, it can uniformly solve power flows for various node types, accurately obtaining the voltage amplitude, phase, and power distribution of each node in the distribution network. This provides reliable basic data for voltage and frequency control, improving the accuracy of planning and operation. For renewable energy utilization, the distribution network power flow tracing algorithm determines the electrical association between photovoltaic (PV) and loads, dynamically allocating power regulation. This effectively mitigates node voltage offsets and power fluctuations caused by distributed PV, maximizing PV absorption and improving renewable energy utilization. An optimization function is constructed from both the power company and user perspectives. While ensuring grid stability and user comfort, it minimizes the total control cost, balancing economic and social benefits, providing technical support for the safe and economical operation of distribution networks with high renewable energy penetration.

[0060] Compared with the prior art, the present invention has the following beneficial effects based on the above embodiments: 1. An improved forward-backward substitution method is employed to perform power flow calculations on the distribution network, accurately acquiring power flow information at each node and providing precise and reliable data support for subsequent power flow scheduling. By constructing a comprehensive objective function that includes penalties on both the power company's and user's sides—where the power company's penalty represents the degree of controllable resource utilization and the user's load comfort—the scheduling model can balance the power company's control needs with the user's electricity experience, overcoming the problem of existing scheduling methods focusing only on the interests of a single entity, leading to difficulties in implementing scheduling schemes. A hybrid energy storage converter, consisting of a front-end DC / DC converter and a back-end converter, is used to collaboratively perform power flow scheduling based on equipment-level commands obtained from the scheduling scheme, achieving precise adjustment and stable control of power flow.

[0061] 2. This application employs an improved forward-backward substitution method, using branch current as the iterative variable to correct node numbers and node types, accurately calculating power flow information for each node in the distribution network. This overcomes the problem that traditional power flow calculations cannot uniformly handle multiple node types such as PQ, PV, and PI, providing accurate data support for power flow scheduling. By constructing a comprehensive objective function that includes penalties from both the power company and the power user side, and combining this with reasonably set penalty coefficients, the application balances the degree of access to controllable resources by the power company with the load comfort level of power users. This overcomes the problem that existing scheduling methods only focus on the interests of a single entity, achieving a balance between safe grid operation and user experience. This system utilizes a hybrid energy storage converter comprised of a front-end DC / DC converter and a back-end voltage source inverter to collaboratively execute device-level commands. The front-end DC / DC converter regulates the power and controls the charging and discharging of the photovoltaic array and energy storage units, while the back-end inverter converts DC power to AC power. It supports both grid-connected and off-grid operation modes. In off-grid operation, a virtual synchronous machine control ensures system stability, while in grid-connected operation, it achieves peak shaving and valley filling, guaranteeing stable grid operation under various conditions. Combined with a proportional sharing-based distribution network power flow tracing method, it clarifies the electrical relationship between distributed power sources and loads and rationally allocates power regulation, effectively mitigating power fluctuations caused by high-proportion photovoltaic (PV) integration, improving PV absorption rates, reducing power company control costs, and ensuring the safe and stable operation of the distribution network in scenarios with high-penetration renewable energy integration.

[0062] The specific embodiments described above are preferred embodiments of a power flow scheduling method based on energy storage converter control according to this application, and are not intended to limit the specific implementation scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. A power flow scheduling method based on energy storage converter control, characterized in that, include: S1. An improved forward-backward substitution method is used to perform power flow calculations on the distribution network to obtain power flow information of each node in the distribution network. S2. Construct a scheduling model, and solve the power flow information based on the comprehensive objective function of the scheduling model to obtain a scheduling scheme; the comprehensive objective function includes a power company-side penalty term and a power user-side penalty term, the power company-side penalty term characterizes the degree of controllable resource utilization, and the power user-side penalty term characterizes the user load comfort level; S3. Based on the scheduling scheme, obtain equipment-level instructions, and use a hybrid energy storage converter composed of a front-end DC / DC converter and a back-end converter to coordinate power flow scheduling of the distribution network based on the equipment-level instructions.

2. The power flow scheduling method based on energy storage converter control according to claim 1, characterized in that, The comprehensive objective function of the scheduling model is set in the following ways: S21. Set the power company penalty coefficient based on the equipment regulation potential of the distribution network as the optimization objective; set the power user penalty coefficient based on the user comfort of the distribution network as the optimization objective. S22. Obtain the power company-side penalty item based on the power company's penalty coefficient and the unit price of controllable resource control costs; obtain the power user-side penalty item based on the power user's penalty coefficient and the unit price of comfort penalty costs; S23. The comprehensive objective function is obtained based on the sum of the penalty terms on the power company side and the penalty terms on the power user side.

3. The power flow scheduling method based on energy storage converter control according to claim 2, characterized in that, The step of setting the power company penalty coefficient based on the equipment regulation potential of the distribution network as the optimization target includes: S21-1, setting the power company penalty coefficient based on the regulation potential of controllable resources and the degree to which the target regulation power deviates from its regulation upper and lower limits; The step of setting a power user penalty coefficient with the user comfort of the distribution network as the optimization target includes: S21-2, setting a power user penalty coefficient based on the adjustment range of controllable resources and the degree to which the target adjustment power deviates from its comfort range.

4. The power flow scheduling method based on energy storage converter control according to claim 1, characterized in that, The steps for performing power flow calculations on the distribution network using the improved forward-backward substitution method include: S11. Select the branch current as the iterative variable of the improved forward-backward substitution method, and modify the node number and node type to obtain the modified node. S12. The improved forward-backward substitution method is used to perform power flow calculation on the corrected nodes to obtain the power flow information of each node in the distribution network. The power flow information includes voltage amplitude, voltage phase and power distribution.

5. A power flow scheduling method based on energy storage converter control according to claim 4, characterized in that, The node numbers are corrected in the following ways: S11-1. Number the branches of the distribution network sequentially to obtain the branch numbers; S11-2. Based on the branch number, sequentially number the branch from the beginning to the end of each branch to obtain the corrected node number.

6. The power flow scheduling method based on energy storage converter control according to claim 4, characterized in that, The node types include PQ nodes, PV nodes, and PI nodes; The node type is corrected in the following ways, including: S11-3. In the iteration, the reactive power of the PV node is corrected by the change in voltage amplitude to obtain the corrected reactive power value of the PV node. S11-4. In the iteration, the reactive power correction value of the PI node is calculated by using the branch current and branch voltage.

7. A power flow scheduling method based on energy storage converter control according to claim 4, characterized in that, The step of performing power flow calculations on the corrected nodes using the improved forward-backward substitution method includes: S12-1. For the inflow current of each node after the correction process, perform back-substitution calculation from the load end node to the root node to obtain the branch current. S12-2. Based on the branch current, perform forward calculations from the root node to the load end node to obtain the node voltage; S12-3. After each iteration calculation, calculate the power difference of all nodes to determine the convergence degree of the iteration result. If the real part or imaginary part of the iteration result does not meet the convergence condition, return to step S12-1 and repeat until the real part and imaginary part simultaneously meet the convergence condition to obtain the power flow information of each node of the distribution network.

8. The power flow scheduling method based on energy storage converter control according to claim 1, characterized in that, The step of obtaining device-level instructions based on the scheduling scheme includes: S31. The distribution network power flow tracing method based on proportional sharing is used to determine the electrical relationship between distributed power sources and loads in the distribution network. S32. Based on the electrical association power adjustment amount, the load node and the corresponding power allocation factor are obtained, and then the equipment-level instructions are obtained to perform power flow scheduling.

9. A power flow scheduling method based on energy storage converter control according to claim 1, characterized in that, The downstream converter is a voltage source inverter; power flow dispatching of the distribution network is performed in the following ways: S33. When the distribution network is connected to the grid, the voltage source inverter responds to the equipment-level command to perform power flow scheduling; S34. When the distribution network is off-grid, the hybrid energy storage converter adopts virtual synchronous machine control and performs power flow scheduling in response to the equipment-level instructions.

10. A power flow scheduling method based on energy storage converter control according to claim 1, characterized in that, The front-end DC / DC converter is electrically connected to the photovoltaic array and the energy storage unit respectively to perform power regulation and charge / discharge control on the photovoltaic array and the energy storage unit; the back-end converter is electrically connected to the distribution network to convert DC power into AC power that is compatible with the frequency of the distribution network.

Images