Adaptive optimization method for distributed photovoltaic bearing capacity of power grid based on intelligent scheduling

Abstract

The invention relates to an adaptive optimization method for distributed photovoltaic bearing capacity of a power grid based on intelligent scheduling. The adaptive optimization method comprises the following steps: S1, monitoring running state data of the power grid in real time; s2, establishing a topological model of a regional power grid, fusing distributed photovoltaic access characteristics, and generating a dynamic multi-time scale power grid operation model; s3, constructing a digital twinborn model of a power grid operation state based on the topology and operation model; s4, outputting a simulation state based on a digital twinborn model, optimizing distributed photovoltaic bearing capacity, dynamically evaluating power grid access capacity, and dynamically optimizing an area with insufficient potential; S5, determining local photovoltaic access bearing capacity through load flow calculation, voltage stability analysis and transient stability analysis, and realizing interactive evaluation with overall operation constraint of the whole network. And obtaining the upper and lower limits of the photovoltaic access capability and an optimal access scheme. According to the invention, the maximization of the photovoltaic access capability and the high-efficiency and stable operation of the power grid are realized, the photovoltaic access capability is improved, and the flexibility and reliability of the power grid are enhanced.

Description

Technical Field

[0001] The present invention relates to the field of power grid operation optimization, and in particular to an adaptive optimization method for distributed photovoltaic carrying capacity of a power grid based on intelligent scheduling. Background Art

[0002] With the continuous growth of global energy demand and the improvement of environmental awareness, more and more countries and regions have begun to promote the large-scale application of renewable energy, especially solar photovoltaic power generation. Distributed photovoltaic systems have become a key component of today's energy transformation due to their flexible installation locations and low environmental impact. However, the large-scale access of distributed photovoltaics has posed a huge challenge to traditional power systems.

[0003] Traditional power grid design is mainly based on centralized power generation and one-way transmission, while photovoltaic power generation has significant volatility and intermittency, and its output is closely related to natural conditions such as weather and season. This has brought new problems to the dispatch and management of the power grid, such as voltage over-limit, frequency fluctuation and line overload. These problems limit the maximum carrying capacity of distributed photovoltaics in the power grid and affect the stable operation of the power grid. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide an adaptive optimization method for distributed photovoltaic carrying capacity of power grid based on intelligent scheduling, which realizes the maximization of photovoltaic access capacity and the efficient and stable operation of power grid, which not only improves the photovoltaic access capacity, but also enhances the flexibility and reliability of the power grid.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] An adaptive optimization method for distributed photovoltaic carrying capacity of a power grid based on intelligent scheduling includes the following steps:

[0007] S1: Real-time monitoring of grid operation status data through smart meters, synchronized phasor measurement units and distributed data terminals;

[0008] S2: Based on the collected network operation status data, a topological model of the regional power grid is established, the access characteristics of distributed photovoltaics are integrated, and a dynamic multi-time scale power grid operation model is generated;

[0009] S3: Build a digital twin model of the grid operation status based on the topology and operation model, simulate the impact of PV access on the grid operation status, and generate dynamic feedback;

[0010] S4: Based on the simulation status output by the digital twin model, the distributed photovoltaic carrying capacity is optimized, the grid access capacity is dynamically evaluated, and the areas with insufficient potential are dynamically optimized;

[0011] S5: Based on the dynamic optimization results, in the regional power grid, the local photovoltaic access carrying capacity is determined through power flow calculation, voltage stability analysis and transient stability analysis, and interactive evaluation is achieved with the overall operation constraints of the entire network to obtain the upper and lower limits of the photovoltaic access capacity and the optimal access plan.

[0012] Further, S1 is as follows:

[0013] The grid operation status data collected by smart meters, synchronized phasor measurement units and distributed data terminals include:

[0014] Grid topology information: nodes, lines, transformers, load access points, photovoltaic access points;

[0015] Real-time operation data:

[0016] Photovoltaic output: Active power P pv , reactive power Q pv , voltage V pv ;

[0017] Grid status: node voltage V i 、Current I i , phase angle θ i , frequency f;

[0018] Environmental data: light intensity G, ambient temperature T env ;

[0019] Time series data: grid and photovoltaic operation data at different time scales, including intraday fluctuations, seasonal changes, and short-term fluctuations.

[0020] Furthermore, S2 is specifically:

[0021] The topological model of the power grid is described by the connection relationship between nodes and lines, combined with the access characteristics of distributed photovoltaics, forming the basis of the dynamic operation model;

[0022] Node definition:

[0023] Each bus or device access point in the power grid is defined as a node i; node attributes include voltage amplitude V i , phase angle θ i , Active power P i , reactive power Q i ;

[0024] Line definition:

[0025] The transmission line between nodes is defined as branch ij, and its parameters include:

[0026] Impedance Z ij =R ij +jXij ; Admittance Y ij =1 / Zij;

[0027] Definition of photovoltaic access point:

[0028] The characteristics of distributed photovoltaic access points include: photovoltaic output P pv , Q pv and voltage control characteristics;

[0029] Active power P pv The ambient light G and temperature T env Decide:

[0030] P pv =ηAG;

[0031] Where, η is the efficiency of the photovoltaic module; A is the area of ​​the photovoltaic module;

[0032] Photovoltaic reactive power Q pv Determined by the inverter control strategy; the photovoltaic inverter uses reactive power Q pv Regulation supports voltage stability, satisfying the following relationship:

[0033]

[0034] Among them, V ref is the reference voltage; V pv is the voltage at the photovoltaic access point; X inv is the reactive power regulation coefficient of the inverter;

[0035] Node admittance matrix Y bus To generate according to the topological relationship of the power grid and used for power flow calculation:

[0036]

[0037] Among them, Y shunt is the shunt admittance of the node;

[0038] Based on the topological model and photovoltaic access characteristics, a dynamic multi-time scale grid operation model is constructed.

[0039] Furthermore, according to the topological model and photovoltaic access characteristics, a dynamic multi-time scale grid operation model is constructed, including a short-term dynamic model, an intraday model and a seasonal model, as follows:

[0040] The short-term dynamic model reflects the rapid fluctuation of photovoltaic power and the dynamic response of the power grid, based on the dynamic equation of the power system:

[0041]

[0042] Among them, θ i is the node phase angle, wi is the node frequency offset, M i is the node inertia, D i is the damping coefficient, P m,i is the mechanical power, P e,i is the electrical power;

[0043] The intraday model reflects the intraday variation of load and photovoltaic output based on the power flow equation

[0044]

[0045]

[0046] Among them, G ij ,B ij are the real and imaginary parts of the line admittance;

[0047] The seasonal model reflects the long-term trend of photovoltaic output and load. Combined with historical data, a prediction model is built through LSTM to obtain the long-term output characteristics of photovoltaics.

[0048] Furthermore, the digital twin model consists of a physical layer, a data layer, and a twin modeling and simulation layer. Specifically:

[0049] Physical layer: Deployed sensors and monitoring devices, including smart meters, PMUs, and DTUs, transmit dynamic data on real-time grid operation status, including collected topology, operating parameters, and photovoltaic status;

[0050] The data layer is responsible for data integration, preprocessing and cleaning. It converts the data input from the physical layer into a data format that can be used by the twin model and stores historical data for model training and simulation driving.

[0051] Twin modeling and simulation layer: Based on the topology model, the dynamic operation model of the power grid, and the characteristics of photovoltaic access, an interactive dynamic simulation model of the power grid is constructed to support high-precision real-time simulation and dynamic feedback optimization of the power grid status.

[0052] Furthermore, S4 is specifically:

[0053] Through power flow calculation and steady-state analysis, the key constraints limiting PV access are identified. The contribution of PV access points to the grid is:

[0054] P net,i =P load,i -P pv,i ;

[0055] Q net,i =Q load,i -Q pv,i ;

[0056] Voltage limit constraint:

[0057] V min ≤V i ≤V max ;

[0058] When the node voltage exceeds the limit due to photovoltaic access, adjust the reactive power Q pv Return the voltage to a safe range:

[0059]

[0060] Line power flow S ij The amplitude should meet the rated capacity of the line limit:

[0061]

[0062] Dynamic evaluation combines light, load and environmental factors at different time scales to gradually and dynamically evaluate the photovoltaic access capacity, including intraday characteristics analysis and seasonal characteristics analysis of light and load distribution and abnormal working condition simulation analysis:

[0063] Intraday characteristic analysis shows that photovoltaic power output and grid load vary significantly over time. The dynamic load capacity is calculated using time series data:

[0064] P pv,max,i (t) = min[P overload (t),P voltage (t)];

[0065] Among them, P overload (t) is the maximum photovoltaic access power calculated according to the line load limit; P voltage (t) is the maximum photovoltaic access power calculated based on the voltage limit;

[0066] The seasonal characteristics of light and load distribution are used to build a prediction model through LSTM to obtain the long-term output characteristics of photovoltaic power generation and predict the long-term changes of photovoltaic power generation;

[0067] Abnormal working condition simulation analysis, simulating low-load and high-PV periods, and analyzing whether there is a risk of voltage collapse or reactive power shortage;

[0068] The goal of optimizing distributed photovoltaic carrying capacity is to maximize photovoltaic access capacity while ensuring safe and stable grid operation. The optimization objective function is defined as:

[0069]

[0070] Where: P pv,i is the active power of photovoltaic access to node i; C comp,i is the activation cost of reactive power compensation equipment; C ess,iis the activation cost of the energy storage system; cost is the cost weight coefficient;

[0071] Design multi-level dynamic optimization measures for areas with insufficient PV access potential;

[0072] Use mixed integer linear programming to obtain the best optimization solution:

[0073] Import the grid topology and operation data, initialize the MLP model, and set the initial values ​​of continuous and integer variables based on the constraints and objective function;

[0074] The MILP solver is used to calculate the optimal solution, including photovoltaic access power, reactive compensation, and energy storage power output; the optimization results are fed back to the digital twin model for verification.

[0075] Further, multi-level dynamic optimization measures are as follows:

[0076] Reactive power compensation optimization, configure dynamic reactive power compensation equipment at low voltage nodes:

[0077]

[0078] Among them, Q comp is the reactive power compensation power; V actual is the actual voltage of the node; X comp is the equivalent reactance of the reactive power compensation equipment;

[0079] Add energy storage system near the photovoltaic access point to achieve peak-valley load balance and frequency support. Energy storage charging and discharging power:

[0080] P ess (t) = P pv (t)-P load( t), P ess,min ≤P ess (t)≤P ess,max :

[0081] Among them, P ess (t) is the charging and discharging power of the energy storage system at time t; P pv (t) is the power generation of the photovoltaic system; P load (t) is the load power demand;

[0082] Optimize the electricity consumption side, implement demand-side response, and reduce the risk of grid overload during photovoltaic peak hours.

[0083] Furthermore, S5 is specifically:

[0084] Import dynamic optimization results of power flow calculation, reactive power compensation, and energy storage regulation; import regional photovoltaic power generation forecast data, load forecast data, and network topology;

[0085] Use power flow calculation to determine the voltage adjustment and line power flow distribution after local photovoltaic access; evaluate the stability margin of local photovoltaic access nodes through voltage stability analysis, and further confirm the impact of photovoltaic short-term fluctuations on local stability through transient stability analysis;

[0086] Combining multi-node power flows with grid-wide constraints, the global impact of PV access is comprehensively analyzed to determine the dynamic impact of local PV access on the overall grid operation. Feedback is then used for multiple rounds of optimization to obtain the upper and lower limits of PV access capacity and the optimal access plan.

[0087] The present invention has the following beneficial effects:

[0088] 1. The present invention realizes the maximization of photovoltaic access capacity and the efficient and stable operation of the power grid, which not only improves the photovoltaic access capacity, but also enhances the flexibility and reliability of the power grid;

[0089] 2. The topological model of the regional power grid constructed by the present invention combines the access characteristics of distributed photovoltaics to generate a dynamic multi-time scale operation model, which can reflect the operation status of the power grid at different time scales and the dynamic impact of photovoltaic access on the power grid. It is also expanded into a digital twin model based on the topological and dynamic operation models, which can accurately simulate the impact of photovoltaic access on the voltage fluctuation and power flow changes of the power grid;

[0090] 3. The present invention comprehensively evaluates and optimizes the access capability of distributed photovoltaic power generation through power flow calculation, voltage constraint analysis and timing dynamic simulation, combined with reactive power compensation, energy storage regulation and demand response, and finally provides a dynamically feasible photovoltaic configuration scheme through optimization algorithm, and realizes optimization verification in digital twin technology;

[0091] 4. The present invention realizes interactive optimization of local photovoltaic access capacity and the operation status of the whole network through the overall evaluation of power flow calculation, voltage stability analysis and transient stability analysis, combined with the overall operation constraints of the whole network, which not only ensures the maximization of photovoltaic carrying capacity in the region, but also ensures the operation safety and stability of the whole network. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0093] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0094] refer to Figure 1 In this embodiment, a method for adaptive optimization of distributed photovoltaic carrying capacity of a power grid based on intelligent scheduling is provided, comprising the following steps:

[0095] S1: Real-time monitoring of grid operation status data through smart meters, synchronized phasor measurement units (PMUs) and distributed data terminals (DTUs);

[0096] S2: Based on the collected network operation status data, a topological model of the regional power grid is established, the access characteristics of distributed photovoltaics are integrated, and a dynamic multi-time scale power grid operation model is generated;

[0097] S3: Build a digital twin model of the grid operation status based on the topology and operation model, simulate the impact of PV access on the grid operation status (such as voltage fluctuations and changes in power flow distribution), and generate dynamic feedback;

[0098] S4: Based on the simulation status output by the digital twin model, the distributed photovoltaic carrying capacity is optimized, the grid access capacity is dynamically evaluated, and the areas with insufficient potential are dynamically optimized:

[0099] S5: Based on the dynamic optimization results, in the regional power grid, the local photovoltaic access carrying capacity is determined through power flow calculation (PF), voltage stability analysis (VSA) and transient stability analysis, and interactive evaluation is achieved with the overall operation constraints of the entire network to obtain the upper and lower limits of the photovoltaic access capacity and the optimal access plan.

[0100] In this embodiment, S1 is as follows:

[0101] The grid operation status data collected by smart meters, synchronized phasor measurement units and distributed data terminals include:

[0102] Grid topology information: nodes (busbars), lines (impedance parameters), transformers, load access points, photovoltaic access points;

[0103] Real-time operation data:

[0104] Photovoltaic output: Active power P pv , reactive power Q pv , voltage V pv ;

[0105] Grid status: node voltage V i 、Current I i , phase angle θ i , frequency f;

[0106] Environmental data: light intensity G, ambient temperature T env ;

[0107] Time series data: grid and photovoltaic operation data at different time scales, including intraday fluctuations, seasonal changes, and short-term fluctuations.

[0108] In this embodiment, S2 is specifically:

[0109] The topological model of the power grid is described by the connection relationship between nodes and lines, combined with the access characteristics of distributed photovoltaics, forming the basis of the dynamic operation model;

[0110] Node definition:

[0111] Each bus or device access point in the power grid is defined as a node i; node attributes include voltage amplitude V i , phase angle θ i , Active power P i , reactive power Q i ;

[0112] Line definition:

[0113] The transmission line between nodes is defined as branch ij, and its parameters include:

[0114] Impedance Z ij =R ij +jX ij ; Admittance Y ij =1 / Zij;

[0115] Definition of photovoltaic access point:

[0116] The characteristics of distributed photovoltaic access points include: photovoltaic output P pv , Q pv and voltage control characteristics (such as constant power factor mode or constant voltage mode);

[0117] Active power P pv The ambient light G and temperature T env Decide:

[0118] P pv =ηAG;

[0119] Where, η is the efficiency of the photovoltaic module; A is the area of ​​the photovoltaic module;

[0120] Photovoltaic reactive power Q pv Determined by the inverter control strategy; the photovoltaic inverter uses reactive power Q pv Regulation supports voltage stability, satisfying the following relationship:

[0121]

[0122] Among them, V ref is the reference voltage; V pv is the voltage at the photovoltaic access point; X inv is the reactive power regulation coefficient of the inverter;

[0123] Node admittance matrix Y bus To generate according to the topological relationship of the power grid and used for power flow calculation:

[0124]

[0125] Among them, Y shunt is the shunt admittance of the node;

[0126] Based on the topological model and photovoltaic access characteristics, a dynamic multi-time scale grid operation model is constructed.

[0127] In this embodiment, a dynamic multi-time scale grid operation model is constructed based on the topological model and photovoltaic access characteristics, including a short-term dynamic model, an intraday model and a seasonal model, as follows:

[0128] The short-term dynamic model reflects the rapid fluctuation of photovoltaic power and the dynamic response of the power grid, based on the dynamic equation of the power system:

[0129]

[0130] Among them, θ i is the node phase angle, w i is the node frequency offset, M i is the node inertia, D i is the damping coefficient, P m,i is the mechanical power, P e,i is the electrical power;

[0131] The intraday model reflects the intraday variation of load and photovoltaic output based on the power flow equation

[0132]

[0133]

[0134] Among them, G ij ,B ij are the real and imaginary parts of the line admittance;

[0135] The seasonal model reflects the long-term trend of photovoltaic output and load. Combined with historical data, a prediction model is built through LSTM to obtain the long-term output characteristics of photovoltaics.

[0136] In this embodiment, the digital twin model consists of a physical layer, a data layer, and a twin modeling and simulation layer. Specifically:

[0137] Physical layer: Deployed sensors and monitoring devices, including smart meters, PMUs, and DTUs, transmit dynamic data on real-time grid operation status, including collected topology, operating parameters, and photovoltaic status;

[0138] The data layer is responsible for data integration, preprocessing and cleaning. It converts the data input from the physical layer into a data format that can be used by the twin model and stores historical data for model training and simulation driving.

[0139] Twin modeling and simulation layer: Based on the topology model, the dynamic operation model of the power grid, and the characteristics of photovoltaic access, an interactive dynamic simulation model of the power grid is constructed to support high-precision real-time simulation and dynamic feedback optimization of the power grid status.

[0140] In this embodiment, S4 is specifically:

[0141] Through power flow calculation and steady-state analysis, the key constraints limiting PV access are identified. The contribution of PV access points to the grid is:

[0142] P net,i =P load,i -P pv,i ;

[0143] Q net,i =Q load,i -Q pv,i ;

[0144] Voltage limit constraint:

[0145] V min ≤V i ≤V max ;

[0146] When the node voltage exceeds the limit due to photovoltaic access, adjust the reactive power Q pv Return the voltage to a safe range:

[0147]

[0148] Line power flow S ij The amplitude should meet the rated capacity of the line limit:

[0149]

[0150] Dynamic evaluation combines light, load and environmental factors at different time scales to gradually and dynamically evaluate the photovoltaic access capacity, including intraday characteristics analysis and seasonal characteristics analysis of light and load distribution and abnormal working condition simulation analysis:

[0151] Intraday characteristic analysis shows that photovoltaic power output and grid load vary significantly over time. The dynamic load capacity is calculated using time series data:

[0152] P pv,max,i (t) = min[P coverload (t),P voltage (t)];

[0153] Among them, P overload (t) is the maximum photovoltaic access power calculated according to the line load limit; P voltage(t) is the maximum photovoltaic access power calculated based on the voltage limit;

[0154] The seasonal characteristics of light and load distribution are used to build a prediction model through LSTM to obtain the long-term output characteristics of photovoltaic power generation and predict the long-term changes of photovoltaic power generation;

[0155] Abnormal working condition simulation analysis, simulating low-load and high-PV periods, and analyzing whether there is a risk of voltage collapse or reactive power shortage;

[0156] The goal of optimizing distributed photovoltaic carrying capacity is to maximize photovoltaic access capacity while ensuring safe and stable grid operation. The optimization objective function is defined as:

[0157]

[0158] Where: P pv,i is the active power of photovoltaic access to node i; C comp,i is the activation cost of reactive power compensation equipment; C ess,i is the activation cost of the energy storage system; cost is the cost weight coefficient;

[0159] Design multi-level dynamic optimization measures for areas with insufficient PV access potential;

[0160] Use mixed integer linear programming to obtain the best optimization solution:

[0161] Import the grid topology and operation data, initialize the MLP model, and set the initial values ​​of continuous and integer variables based on the constraints and objective function;

[0162] The MILP solver is used to calculate the optimal solution, including photovoltaic access power, reactive compensation, and energy storage power output; the optimization results are fed back to the digital twin model for verification.

[0163] In this embodiment, the multi-level dynamic optimization measures are as follows:

[0164] Reactive power compensation optimization, configure dynamic reactive power compensation equipment (such as SVG, STATCOM) at low voltage nodes:

[0165]

[0166] Among them, Q comp is the reactive power compensation power; V actual is the actual voltage of the node; X comp is the equivalent reactance of the reactive power compensation equipment;

[0167] Add an energy storage system (ESS) near the photovoltaic access point to achieve peak-valley load balancing and frequency support. The energy storage charging and discharging power is:

[0168] P ess (t) = P pv (t)-P load (t), P ess,min ≤P ess (t)≤P ess,max :

[0169] Among them, P ess (t) is the charging and discharging power of the energy storage system at time t; P pv (t) is the power generation of the photovoltaic system; P load (t) is the load power demand;

[0170] Optimize the electricity consumption side, implement demand-side response, and reduce the risk of grid overload during photovoltaic peak hours.

[0171] In this embodiment, S5 is specifically:

[0172] Import dynamic optimization results of power flow calculation, reactive power compensation, and energy storage regulation; import regional photovoltaic power generation forecast data, load forecast data, and network topology;

[0173] Use power flow calculation to determine the voltage adjustment and line power flow distribution after local photovoltaic access; evaluate the stability margin of local photovoltaic access nodes through voltage stability analysis, and further confirm the impact of photovoltaic short-term fluctuations on local stability through transient stability analysis;

[0174] Combining multi-node power flow with grid-wide constraints (eligible voltage range, line power limit, frequency constraint, etc.), the global impact of PV access is comprehensively analyzed to determine the dynamic impact of local PV access on the overall grid operation. Feedback is then conducted for multiple rounds of optimization to obtain the upper and lower limits of PV access capacity and the optimal access solution.

[0175] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0177] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0178] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0179] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. An adaptive optimization method for distributed photovoltaic carrying capacity of power grid based on intelligent scheduling, characterized in that: The following steps are involved: S1: Real-time monitoring of grid operation status data through smart meters, synchronized phasor measurement units and distributed data terminals; S2: Based on the collected network operation status data, a topological model of the regional power grid is established, the access characteristics of distributed photovoltaics are integrated, and a dynamic multi-time scale power grid operation model is generated; S3: Build a digital twin model of the grid operation status based on the topology and operation model, simulate the impact of PV access on the grid operation status, and generate dynamic feedback; S4: Based on the simulation status output by the digital twin model, the distributed photovoltaic carrying capacity is optimized, the grid access capacity is dynamically evaluated, and the areas with insufficient potential are dynamically optimized: S5: Based on the dynamic optimization results, in the regional power grid, the local photovoltaic access carrying capacity is determined through power flow calculation, voltage stability analysis and transient stability analysis, and interactive evaluation is achieved with the overall operation constraints of the entire network to obtain the upper and lower limits of the photovoltaic access capacity and the optimal access plan.

2. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 1 is characterized in that: The S1 is specifically as follows: The grid operation status data collected by smart meters, synchronized phasor measurement units and distributed data terminals include: Grid topology information: nodes, lines, transformers, load access points, photovoltaic access points; Real-time operation data: Photovoltaic output: Active power P pv , reactive power Q pv , voltage V pv ; Grid status: node voltage V i 、Current I i , phase angle θ i , frequency f; Environmental data: light intensity G, ambient temperature T env ; Time series data: grid and photovoltaic operation data at different time scales, including intraday fluctuations, seasonal changes, and short-term fluctuations.

3. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 1 is characterized in that: The S2 is specifically: The topological model of the power grid is described by the connection relationship between nodes and lines, combined with the access characteristics of distributed photovoltaics, forming the basis of the dynamic operation model; Node definition: Each bus or device access point in the power grid is defined as a node i; node attributes include voltage amplitude V i , phase angle θ i , Active power P i , reactive power Q i ; Line definition: The transmission line between nodes is defined as branch ij, and its parameters include: Impedance Z ij =R ij +jX ij ; Admittance Y ij =1 / Zij; Definition of photovoltaic access point: The characteristics of distributed photovoltaic access points include: photovoltaic output P pv , Q pv and voltage control characteristics; Active power P pv The ambient light G and temperature T env Decide: P pv =ηAG; Where, η is the efficiency of the photovoltaic module; A is the area of ​​the photovoltaic module; Photovoltaic reactive power Q pv Determined by the inverter control strategy; the photovoltaic inverter uses reactive power Q pv Regulation supports voltage stability, satisfying the following relationship: Among them, V ref is the reference voltage; V pv is the voltage at the photovoltaic access point; X inv is the reactive power regulation coefficient of the inverter; Node admittance matrix Y bus To generate according to the topological relationship of the power grid and used for power flow calculation: Among them, Y shunt is the shunt admittance of the node; Based on the topological model and photovoltaic access characteristics, a dynamic multi-time scale grid operation model is constructed.

4. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 3 is characterized in that: According to the topological model and photovoltaic access characteristics, a dynamic multi-time scale grid operation model is constructed, including a short-term dynamic model, an intraday model and a seasonal model, as follows: The short-term dynamic model reflects the rapid fluctuation of photovoltaic power and the dynamic response of the power grid, based on the dynamic equation of the power system: Among them, θ i is the node phase angle, w i is the node frequency offset, M i is the node inertia, D i is the damping coefficient, P m,i is the mechanical power, P e,i is the electrical power; The intraday model reflects the intraday variation of load and photovoltaic output based on the power flow equation Among them, G ij ,B ij are the real and imaginary parts of the line admittance; The seasonal model reflects the long-term trend of photovoltaic output and load. Combined with historical data, a prediction model is built through LSTM to obtain the long-term output characteristics of photovoltaics.

5. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 4 is characterized in that: The digital twin model consists of a physical layer, a data layer, and a twin modeling and simulation layer. Specifically: Physical layer: Deployed sensors and monitoring devices, including smart meters, PMUs, and DTUs, transmit dynamic data on real-time grid operation status, including collected topology, operating parameters, and photovoltaic status; The data layer is responsible for data integration, preprocessing and cleaning. It converts the data input from the physical layer into a data format that can be used by the twin model and stores historical data for model training and simulation driving. Twin modeling and simulation layer: Based on the topology model, the dynamic operation model of the power grid, and the characteristics of photovoltaic access, an interactive dynamic simulation model of the power grid is constructed to support high-precision real-time simulation and dynamic feedback optimization of the power grid status.

6. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 1 is characterized in that: The S4 is specifically: Through power flow calculation and steady-state analysis, the key constraints limiting PV access are identified. The contribution of PV access points to the grid is: P net,i =P load,i -P pv,i ; Q net,i =Q load,i -Q pv,i ; Voltage limit constraint: In min ≤V i ≤V max ; When the node voltage exceeds the limit due to photovoltaic access, adjust the reactive power Q pv Return the voltage to a safe range: Line power flow S ij The amplitude should meet the rated capacity of the line limit: Dynamic evaluation combines light, load and environmental factors at different time scales to gradually and dynamically evaluate the photovoltaic access capacity, including intraday characteristics analysis and seasonal characteristics analysis of light and load distribution and abnormal working condition simulation analysis: Intraday characteristic analysis shows that photovoltaic power output and grid load vary significantly over time. The dynamic load capacity is calculated using time series data: P pv,max,i (t)=min[P overload (t),P voltage (t)]; Among them, P overload (t) is the maximum photovoltaic access power calculated according to the line load limit; P voltage (t) is the maximum photovoltaic access power calculated based on the voltage limit; The seasonal characteristics of light and load distribution are used to build a prediction model through LSTM to obtain the long-term output characteristics of photovoltaic power generation and predict the long-term changes of photovoltaic power generation; Abnormal working condition simulation analysis, simulating low-load and high-PV periods, and analyzing whether there is a risk of voltage collapse or reactive power shortage; The goal of optimizing distributed photovoltaic carrying capacity is to maximize photovoltaic access capacity while ensuring safe and stable grid operation. The optimization objective function is defined as: Where: P pv,i is the active power of photovoltaic access to node i; C comp,i is the activation cost of reactive power compensation equipment; C ess,i is the activation cost of the energy storage system; cost is the cost weight coefficient; Design multi-level dynamic optimization measures for areas with insufficient PV access potential; Use mixed integer linear programming to obtain the best optimization solution: Import the grid topology and operation data, initialize the MLP model, and set the initial values ​​of continuous and integer variables based on the constraints and objective function; The MILP solver is used to calculate the optimal solution, including photovoltaic access power, reactive compensation, and energy storage power output; the optimization results are fed back to the digital twin model for verification.

7. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 6 is characterized in that: The multi-level dynamic optimization measures are as follows: Reactive power compensation optimization, configure dynamic reactive power compensation equipment at low voltage nodes: Among them, Q comp is the reactive power compensation power; V actual is the actual voltage of the node; X comp is the equivalent reactance of the reactive power compensation equipment; Add energy storage systems near photovoltaic access points to achieve peak-valley load balancing and frequency support. Energy storage charging and discharging power: P ess (t)=P pv (t)-P load (t),P ess,min ≤P ess (t)≤P ess,max ; Among them, P ess (t) is the charging and discharging power of the energy storage system at time t; P pv (t) is the power generation of the photovoltaic system; P load (t) is the load power demand; Optimize the electricity consumption side, implement demand-side response, and reduce the risk of grid overload during photovoltaic peak hours.

8. The adaptive optimization method of distributed photovoltaic carrying capacity of power grid based on intelligent scheduling according to claim 1 is characterized in that: The S5 is specifically: Import dynamic optimization results of power flow calculation, reactive power compensation, and energy storage regulation; import regional photovoltaic power generation forecast data, load forecast data, and network topology; Use power flow calculation to determine the voltage adjustment and line power flow distribution after local photovoltaic access; evaluate the stability margin of local photovoltaic access nodes through voltage stability analysis, and further confirm the impact of photovoltaic short-term fluctuations on local stability through transient stability analysis; Combining multi-node power flows with grid-wide constraints, the global impact of PV access is comprehensively analyzed to determine the dynamic impact of local PV access on the overall grid operation. Feedback is then used for multiple rounds of optimization to obtain the upper and lower limits of PV access capacity and the optimal access plan.

Images