Power grid real-time optimization scheduling system and method based on digital twinning

Abstract

The invention discloses a power grid real-time optimization scheduling system and method based on digital twinning, and relates to the technical field of power grid scheduling. A sensor is deployed to collect environmental parameters of key nodes in real time, a dynamic environmental condition coefficient is constructed, a power grid state is analyzed in combination with frequency stability and a relative strength index, and a multi-model fusion prediction mechanism is established, so that space-time two-dimensional accurate prediction of load and power generation is realized. A multi-objective optimization model is adopted to take'maximization of new energy consumption + minimization of scheduling cost 'as a core objective, a genetic algorithm is introduced to solve an optimal scheduling scheme, and a feasible solution is screened in combination with forward simulation of a digital twin model. Through abnormal early warning triggering, environment correlation analysis and model iterative optimization, a scheduling strategy is dynamically adjusted, and the power supply efficiency and the emergency response capability in an extreme scene are improved. According to the method, multi-source heterogeneous data are effectively fused, and real-time sensing of a power grid operation state, collaborative optimization of multiple energy resources and adaptive iteration of a scheduling model are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid dispatching, and specifically relates to a real-time power grid optimization dispatching system and method based on digital twins. Background Art

[0002] With the large-scale integration of renewable energy sources (such as photovoltaics and wind power) into the power grid, traditional grid dispatching faces challenges such as a high proportion of fluctuating power sources, increased uncertainty in load demand, and a complex equipment operating environment. Existing dispatching methods often rely on static models or single prediction algorithms, making it difficult to perceive the dynamic impact of environmental parameters (such as temperature and solar radiation) on grid load and generation in real time. This leads to problems such as insufficient frequency stability, limited renewable energy absorption, and delayed response of backup resources.

[0003] While digital twin technology has seen initial applications in the power grid (e.g., equipment status monitoring and fault prediction), existing solutions often focus on modeling historical or current data, lacking the ability to predict disturbances in future time intervals, and fail to deeply integrate correlation analysis of multi-source heterogeneous data (e.g., environmental parameters, renewable energy output, and load fluctuations). Furthermore, traditional optimization algorithms (e.g., linear programming and particle swarm optimization) are prone to falling into local optimality when addressing multi-objective coordinated scheduling (maximizing renewable energy consumption and minimizing costs), making it difficult to meet the comprehensive real-time, cost-effective, and reliable requirements of the new power system.

[0004] Therefore, there is an urgent need for a real-time grid optimization and dispatch method based on digital twins. By dynamically integrating environmental parameters, multi-model prediction (in both time and space), intelligent optimization algorithms (genetic algorithms combined with digital twin simulation), and adaptive model iteration mechanisms, this method can achieve early warning of grid operation anomalies, coordinated multi-energy scheduling, and dynamic strategy optimization, thereby improving renewable energy absorption capacity, frequency stability, and power supply efficiency. This paper proposes a real-time grid optimization and dispatch system and method based on digital twins. Summary of the Invention

[0005] In order to overcome the shortcomings and deficiencies of the above-mentioned prior art, the first purpose of the present invention is to provide a real-time optimization and dispatching system for a power grid based on digital twins; the second purpose of the present invention is to provide a real-time optimization and dispatching method for a power grid based on digital twins.

[0006] The first object of the present invention adopts the following technical solution: The real-time optimization and dispatching system for power grids based on digital twins includes: Digital twin module: used to build a high-fidelity virtual image of the power grid and support full-process simulation and deduction; Environmental perception and abnormal warning module: used for quantitative assessment and early warning of environmental risks based on multi-parameter fusion; Grid status analysis module: used for multi-dimensional quantitative assessment of grid stability; Intelligent prediction module: for precise load and power generation prediction based on time and space integration; Multi-energy collaborative scheduling module: used for real-time scheduling decision generation for global optimization; Closed-loop optimization module: Continuous improvement of scheduling strategies based on feedback.

[0007] The second purpose of the present invention adopts the following technical solution: The real-time optimization and dispatching method of the power grid based on digital twin is used for the real-time optimization and dispatching system of the power grid based on digital twin. The method flow is as follows: S1: Grid operation abnormality warning: Collect environmental parameters of key grid nodes and calculate environmental condition coefficients. Based on the abnormal threshold and duration of the environmental condition coefficients, trigger graded warnings and call the digital twin model to simulate risks. S2: Grid status analysis: Calculate the grid frequency stability and relative strength index (RSI) to determine frequency stability anomalies. When the RSI deviates from the threshold, calculate the abnormal correlation between frequency stability and environmental condition coefficients, triggering a pre-adjustment of the dispatch strategy. S3: Power generation and load forecasting: Using a multi-model fusion approach, we generate load forecasts in both time and space. Based on the deviation between the forecast and actual power generation capacity, we trigger the pre-startup of standby units or energy storage emergency preparations. S4: Multi-energy coordinated dispatch: Construct an objective function that maximizes new energy consumption and minimizes dispatch costs, combining power balance and equipment constraints to solve the problem. Genetic algorithms are used to generate dispatch plans, and feasible solutions are screened through digital twin forward simulation. S5: Scheduling model optimization: Define the power supply efficiency coefficient to evaluate the scheduling effect; implement model iterative optimization based on knowledge graph, transfer learning and genetic algorithm weight adjustment.

[0008] Preferably, the grid environmental condition coefficient is calculated as follows: ; in, is the temperature reference standard value; and are the maximum and minimum temperature values ​​during the monitoring period respectively; is the reference standard value of solar radiation; and are the maximum and minimum values ​​of solar radiation during the monitoring period, respectively; is the temperature weight coefficient, emphasizing the impact of load side temperature on the power grid; is the solar radiation weight coefficient, emphasizing the impact of fluctuations in renewable energy output on the power grid; , improve in load-sensitive scenarios , improve in new energy scenarios .

[0009] Preferably, the abnormal warning trigger logic is: Preset abnormal threshold of environmental condition coefficient (For example, the high temperature and high load scenario is set to 0.8), when three consecutive monitoring cycles meet hour: Trigger a red alert and send an equipment overload risk warning to the dispatch center; Automatically call up the digital twin model of the warning area to simulate the impact of continued load growth on the grid topology, power flow distribution, and voltage stability.

[0010] Preferably, the calculation method of the grid frequency stability and the relative strength index RSI is: Set sampling points at each voltage level busbar to calculate frequency stability in real time ;in, is the real-time frequency; is the rated frequency; Count the number of frequency increase cycles in the last 24 monitoring cycles and the number of down cycles ,calculate: when or When the frequency stability is abnormal, and is the preset threshold, and .

[0011] Preferably, the abnormal correlation calculation and scheduling strategy pre-adjustment method is: When RSI deviates from the normal threshold range When the frequency stability is calculated and environmental condition coefficient Abnormal correlation : ; in, is the frequency stability of the i-th cycle; is the environmental condition coefficient of the i-th cycle; is the mean value of the environmental condition coefficient under the baseline environmental pressure level; is the mean value of frequency stability; The number of cycles used to calculate the anomaly correlation; like , the pre-adjustment mechanism of the triggering dispatch strategy includes allocating spare capacity in advance and pre-optimizing the unit output curve; among them, represents the strong correlation threshold.

[0012] Preferably, a multi-model fusion method is used to generate a time-space dual-dimensional load forecast value as follows: Digital twin basic prediction generates the first prediction value : Based on the digital twin model of the power grid, real-time operating data including renewable energy output, energy storage SOC, and load distribution are input, and the load forecast value of each node is generated through the graph neural network; ARIMA modified forecast generates second forecast value : Perform differential processing on historical load data, build ARIMA (p, d, q) model, output short-term load correction value, and corrected forecast value ;in, is the ARIMA model correction; is the weight coefficient, Dynamically adjust the prediction accuracy of the two models to balance the digital twin and time trend prediction ARIMA; When it is detected When , it indicates that the load forecast value is lower than the actual power generation capacity, and there is a risk of load shortage. It is necessary to verify whether the load is underestimated or the power generation is overestimated: Start the preheating program of the standby conventional unit; send emergency discharge preparation instructions to the energy storage system, including: The current available power generation capacity of the power grid includes renewable energy output, traditional unit standby, and energy storage available discharge; when When the actual load is greater than the predicted load, power generation is insufficient, and the standby unit preheating and energy storage discharge preparation are triggered to quickly make up for the power shortage; is the currently available power generation capacity.

[0013] Preferably, multi-energy coordinated scheduling includes: Objective function: ;in, It is the fuel cost item of traditional units, which is used to minimize the economic cost of traditional energy power generation; The charging and discharging cost item of the energy storage system is used to balance the peak-shaving and valley-filling function of energy storage with the operating cost; is the penalty term for the new energy consumption rate, which is used to maximize the new energy consumption. For new energy power, is the total power generation; Constraints: Power balance: They are new energy power, energy storage power, and traditional unit power; Device constraints: ;in, For new energy power; is the power of traditional unit; The power range of traditional units is as follows: the lower limit is the minimum technical output and the upper limit is the rated power; The energy storage power range, the negative sign indicates charging, the positive sign indicates discharging, and the upper limit is the maximum charging and discharging power; Scheduling plan generation: A genetic algorithm is used to solve the objective function, with the chromosome encoding the output ratio of each power source. Combined with the forward simulation of the digital twin model, feasible solutions with no line overload and qualified voltage are screened, and a three-dimensional scheduling plan is output, including the new energy scheduling amount, energy storage charging and discharging plan, and the start and stop strategy of traditional units.

[0014] Preferably, the scheduling model optimization includes: Evaluating the power supply efficiency coefficient ;in, is the actual power supply in area i; Forecast electricity demand for region i; The total number of regions participating in the power supply efficiency coefficient assessment; when , extract similar scenario optimization solutions from the power grid dispatch knowledge graph, where, is the actual power supply in area i; Forecast electricity demand for region i: Perform transfer learning on the load forecasting module of the digital twin model to enhance the recognition of extreme scenario features; Adjust the fitness function weights of the genetic algorithm.

[0015] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention quantifies load / generation risks (such as air conditioning surges and photovoltaic fluctuations caused by high temperatures) through an environmental perception module. Combined with digital twin simulation and deduction, it achieves minute-level warnings of abnormalities (such as a red warning triggered by a coefficient exceeding the threshold for three consecutive cycles) and provides the dispatch center with a visual risk evolution path (such as simulating the overload impact of continuous load growth on equipment), significantly shortening the time for manual analysis and judgment.

[0016] 2. This invention uses spatiotemporal fusion forecasting (GNN spatial modeling + ARIMA temporal correction) to reduce load forecasting errors. It also dynamically adjusts the renewable energy consumption rate and energy storage charging and discharging strategies based on the supply-demand balance model. This reduces curtailment during peak photovoltaic power generation periods, decreases the number of starts and stops of traditional units, and lowers annualized dispatch costs (through the combined optimization of fuel costs and energy storage losses).

[0017] 3. By integrating dispatch effectiveness evaluation with knowledge graphs, this system automatically extracts optimization strategies for historical extreme scenarios (such as high-load, low-renewable energy conditions). Transfer learning strengthens the digital twin model's ability to identify extreme features. In scenarios involving sudden load surges, model iteration accelerates dispatch plan generation and improves voltage compliance (through dynamic adjustment of fitness weights using a genetic algorithm). BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work. Figure 1 The module diagram of the real-time optimization and dispatching system of the power grid based on digital twin of the present invention is shown; Figure 2 The flowchart of the real-time optimization and dispatching method of the power grid based on digital twin of the present invention is shown; Figure 3 The flowchart of the power grid status analysis of the present invention is shown. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] In addition, the described features, structures or characteristics can be combined in any suitable manner in one or more example embodiments. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure can be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. can be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0021] Example 1

[0022] See Figure 1 As shown, the real-time optimization and dispatching system of the power grid based on digital twin of this embodiment includes Digital twin module: used to build a high-fidelity virtual image of the power grid and support full-process simulation and deduction.

[0023] Environmental perception and abnormal warning module: used for quantitative assessment and early warning of environmental risks based on multi-parameter fusion.

[0024] Grid status analysis module: used for multi-dimensional quantitative assessment of grid stability.

[0025] Intelligent prediction module: used for accurate load and power generation prediction based on time and space fusion.

[0026] Multi-energy collaborative scheduling module: Real-time scheduling decision generation for global optimization.

[0027] Closed-loop optimization module: Continuous improvement of scheduling strategies based on feedback.

[0028] The beneficial effects of this embodiment include: the system realizes high-precision simulation and real-time deduction of the power grid through digital twins, and improves risk prevention and control capabilities by combining environmental perception and abnormal warning; multi-dimensional state analysis ensures power grid stability; intelligent prediction optimizes resource allocation, coordinated scheduling promotes the consumption of new energy, and closed-loop optimization continuously improves strategies, comprehensively improving the safety, economy and environmental protection of power grid operation.

[0029] Example 2

[0030] See Figure 2 As shown, the real-time optimization and dispatching method of the power grid based on digital twins in this embodiment has the following process: Step 1: Warning of abnormal power grid operation.

[0031] S11. Grid environmental condition detection and coefficient generation.

[0032] Deploy sensors at key grid nodes (substations / power plants / load centers) to collect real-time environmental parameters that affect grid operation: Load-related parameters: temperature in industrial / residential areas (affecting air conditioning load) and humidity (affecting equipment insulation performance).

[0033] Power generation related parameters: solar radiation intensity in photovoltaic areas, wind speed / direction in wind farms (affecting new energy output).

[0034] Calculation of the power grid environmental coefficient: linearly normalize the temperature T and solar radiation intensity S (map them to the [0,1] interval) to construct the environmental condition coefficient: ;in, is the temperature reference standard value; and are the maximum and minimum temperature values ​​during the monitoring period respectively; is the reference standard value of solar radiation; and are the maximum and minimum values ​​of solar radiation during the monitoring period, respectively; is the temperature weight coefficient, emphasizing the impact of load-side temperature on the power grid (such as the risk of equipment overload caused by a surge in air-conditioning load); is the solar radiation weight coefficient, which emphasizes the impact of fluctuations in renewable energy output on the power grid (such as frequency / voltage fluctuations caused by sudden changes in photovoltaic power). (Load-sensitive scenarios improve , New energy scene improvement ).

[0035] S12. Set abnormal warning trigger logic.

[0036] Abnormal warning trigger logic: preset environmental coefficient Abnormal threshold (For example, the high temperature and high load scenario is set to 0.8), when there are three consecutive monitoring cycles > hour: Trigger a red alert and send an equipment overload risk warning to the dispatch center; Automatically call up the digital twin model of the area to simulate the impact of continued load growth on the power grid.

[0037] Step 2: Grid status analysis.

[0038] See Figure 3 As shown in Figure 2, the process of power grid status analysis is as follows: S21. Calculation of grid frequency stability and relative strength index.

[0039] Set sampling points at each voltage level busbar to calculate frequency stability in real time : ; in, is the real-time frequency; is the rated frequency.

[0040] Relative Strength Index RSI calculation: Count the number of frequency increase cycles in the last 24 monitoring cycles and the number of down cycles ,calculate: ; when or When , it is determined that the frequency stability is abnormal.

[0041] S22. Calculate abnormal correlation.

[0042] When RSI exceeds the normal range (30-70), calculate the frequency stability and environmental condition coefficient Abnormal correlation : ; in, is the frequency stability of the i-th cycle (time series data, reflecting the frequency fluctuation trajectory); is the mean value of frequency stability; The number of cycles used to calculate the anomaly correlation; is the environmental condition coefficient of the i-th cycle (from step 1, reflecting the environmental pressure on the load / generation side, such as high temperature and new energy fluctuations); is the mean value of the environmental condition coefficient (baseline environmental pressure level). Indicates a strong correlation threshold (frequency anomalies are directly related to environmental pressures, such as high temperature load causing a frequency drop, or photovoltaic power generation causing a frequency increase), triggering the scheduling strategy pre-adjustment mechanism.

[0043] Step 3: Power generation and load forecasting.

[0044] S31. Multi-model fusion prediction.

[0045] Digital twin basic prediction (first prediction value ): Based on the digital twin model of the power grid, real-time operating data (new energy output, energy storage SOC, load distribution) is input, and the load forecast value of each node is generated through the graph neural network (GNN).

[0046] ARIMA revised forecast (second forecast value ): Perform differential processing on historical load data (time series), build ARIMA (p, d, q) model, output short-term load correction value, and corrected forecast value : ;in, is the ARIMA model correction; The weight coefficient (dynamically adjusted based on the prediction accuracy of the two models) balances spatial dynamic prediction (digital twin) and temporal trend prediction (ARIMA) to achieve dual-dimensional fusion of time and space; When it is detected , indicating that the load forecast value is lower than the actual power generation capacity, and there is a risk of load shortage (it is necessary to verify whether the load is underestimated or the power generation is overestimated): Start the preheating program of the standby conventional unit; send emergency discharge preparation instructions to the energy storage system. It is the current available power generation capacity of the power grid (new energy output + traditional unit standby + available discharge of energy storage).

[0047] when (Actual load > forecast, insufficient power generation), triggering the standby unit preheating + energy storage discharge preparation to quickly make up for the power shortage.

[0048] Step 4: Coordinated scheduling of multiple energy sources.

[0049] S41. The supply and demand balance objective function is based on the goal of "maximizing new energy consumption and minimizing dispatch costs" to build an optimization model: ;in, The fuel cost item for traditional units (such as coal-fired and gas-fired units) is used to minimize the economic cost of traditional energy power generation and encourage low fuel consumption scheduling strategies (such as giving priority to the use of high-efficiency units); The charging and discharging cost item (including equipment loss and life-span degradation costs) of energy storage systems (such as lithium batteries and pumped hydropower storage) is used to balance the "peak shaving and valley filling" function of energy storage with operating costs, avoiding economic degradation caused by excessive charging and discharging;

[0050] is the penalty item for the new energy consumption rate ( For new energy power, is the total power generation capacity), which is used to maximize the consumption of new energy.

[0051] The constraints are as follows: Power Balance: ; 、 、 They are new energy power, energy storage power, and traditional unit power respectively.

[0052] Device constraints: ; ;in, For new energy power; is the power of traditional unit; The power range of traditional units (the lower limit is the minimum technical output, such as the coal-fired unit needs to maintain the minimum combustion load; the upper limit is the rated power); is the energy storage power range (the negative sign indicates charging, the positive sign indicates discharging, and the upper limit is the maximum charge and discharge power, such as the charge and discharge rate limit of a lithium battery).

[0053] S42: Generate an optimal scheduling plan.

[0054] Genetic algorithm is used to solve the objective function, and the chromosome encoding is the output ratio of each power source; Combined with the digital twin model of the power grid, forward simulation is carried out to screen feasible solutions with no line overload and qualified voltage; The output includes a three-dimensional scheduling plan including new energy scheduling capacity, energy storage charging and discharging plan, and traditional unit start-up and shutdown strategies.

[0055] After solving the objective function using the genetic algorithm, the final solution is screened through the following steps: Feasibility verification: Call the digital twin module to simulate candidate solutions and eliminate solutions that cause line overload or voltage limit violations; Multi-objective evaluation: Calculate the comprehensive score for the remaining feasible solutions: ;in, is the maximum frequency deviation; 、 、 is the dynamic weight; Optimal solution selection: output the solution with the highest comprehensive score.

[0056] Step 5: Scheduling model optimization.

[0057] Scheduling effect evaluation indicators: Water supply coefficient mapping → power supply efficiency coefficient : ; ;in, is the actual power supply in area i; Forecast electricity demand for region i); is the total number of regions participating in the power supply efficiency coefficient assessment.

[0058] Model iterative optimization mechanism: when (Power supply efficiency threshold): Extract optimization solutions for similar scenarios from the grid dispatch knowledge graph (e.g., "high load, low renewable energy scenario → energy storage priority discharge" strategy); Perform transfer learning on the load forecasting module of the digital twin model to enhance the recognition of extreme scenario features; Adjust the fitness function weight of the genetic algorithm to increase the priority of the response speed of traditional units.

[0059] The beneficial effects of this embodiment are as follows: the method realizes real-time perception and simulation of the grid operation status through digital twin technology, combines multi-model fusion prediction and optimization scheduling algorithm, improves the grid frequency stability and supply and demand balance capability, enhances the absorption of new energy, reduces scheduling costs, improves power supply efficiency and emergency response capabilities, and ensures the safe, economical and efficient operation of the grid.

[0060] The weighting coefficients of the present invention are used to measure the degree of influence of different factors or variables on a particular outcome or decision. A weighting coefficient is defined as a numerical value assigned to each factor when comparing and evaluating multiple factors to reflect its importance or priority. These weighting coefficients can be determined based on specific circumstances and needs, and are typically developed and confirmed by professionals or relevant stakeholders. By properly setting weighting coefficients, programs or systems can be helped to make more accurate decisions or predictions.

[0061] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

[0062] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The real-time optimization and dispatching system of power grid based on digital twin is characterized by: The system comprises: Digital twin module: used to build a high-fidelity virtual image of the power grid and support full-process simulation and deduction; Environmental perception and abnormal warning module: used for quantitative assessment and early warning of environmental risks based on multi-parameter fusion; Grid status analysis module: used for multi-dimensional quantitative assessment of grid stability; Intelligent prediction module: for accurate load and power generation prediction based on time and space integration; Multi-energy collaborative scheduling module: used for real-time scheduling decision generation for global optimization; Closed-loop optimization module: Continuous improvement of scheduling strategies based on feedback.

2. A real-time optimization and dispatching method for a power grid based on digital twins, used in the real-time optimization and dispatching system for a power grid based on digital twins as claimed in claim 1, characterized in that: The method flow is as follows: S1: Grid operation abnormality warning: Collect environmental parameters of key grid nodes and calculate environmental condition coefficients. Based on the abnormal threshold and duration of the environmental condition coefficients, trigger graded warnings and call the digital twin model to simulate risks. S2: Grid status analysis: Calculate grid frequency stability and relative strength index (RSI) to determine frequency stability anomalies; When the RSI deviates from the threshold, the abnormal correlation between the frequency stability and the environmental condition coefficient is calculated, triggering the pre-adjustment of the scheduling strategy; S3: Power generation and load forecasting: Using a multi-model fusion approach, we generate load forecasts in both time and space. Based on the deviation between the forecast and actual power generation capacity, we trigger the pre-startup of standby units or energy storage emergency preparations. S4: Multi-energy coordinated dispatch: Construct an objective function that maximizes new energy consumption and minimizes dispatch costs, combining power balance and equipment constraints to solve the problem. Genetic algorithms are used to generate dispatch plans, and feasible solutions are screened through digital twin forward simulation. S5: Scheduling model optimization: Define the power supply efficiency coefficient to evaluate the scheduling effect; implement model iterative optimization based on knowledge graph, transfer learning and genetic algorithm weight adjustment.

3. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The calculation method of the power grid environmental condition coefficient is: Taking the temperature reference standard value as a benchmark, the temperature sensitivity index is constructed through the actual temperature fluctuation range during the monitoring period. Taking the solar radiation reference standard value as a benchmark, the power generation fluctuation index is constructed through the actual radiation intensity fluctuation range during the monitoring period. The temperature influence component and the solar radiation influence component are weighted and summed to obtain the comprehensive environmental condition coefficient.

4. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The abnormal warning trigger logic is: Preset abnormal threshold of environmental condition coefficient , when three consecutive monitoring cycles meet When: a red alert is triggered and an equipment overload risk warning is sent to the dispatch center; Automatically call up the digital twin model of the warning area to simulate the impact of continued load growth on the grid topology, power flow distribution, and voltage stability.

5. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The calculation method of the grid frequency stability and the relative strength index RSI is: Set sampling points at each voltage level busbar to calculate frequency stability in real time ; Count the number of frequency increase cycles in the last 24 monitoring cycles and the number of down cycles ,calculate ;when or When the frequency stability is abnormal, and is the preset threshold, and .

6. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The method for calculating the abnormal correlation and pre-adjusting the scheduling strategy is as follows: When RSI deviates from the normal threshold range When the frequency stability is calculated and environmental condition coefficient Abnormal correlation ; like The pre-adjustment mechanism for triggering the dispatching strategy includes allocating spare capacity in advance and pre-optimizing the unit output curve; represents the strong correlation threshold.

7. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The multi-model fusion method is used to generate the time-space dual-dimensional load forecast value as follows: Digital twin basic prediction generates the first prediction value : Based on the digital twin model of the power grid, real-time operating data including renewable energy output, energy storage SOC, and load distribution are input, and the load forecast value of each node is generated through the graph neural network; ARIMA modified forecast generates second forecast value :Perform differential processing on historical load data, build ARIMA (p, d, q) model, output short-term load correction value, and the corrected forecast value is ; When it is detected When , it indicates that the load forecast value is lower than the actual power generation capacity, and there is a risk of load shortage. It is necessary to verify whether the load is underestimated or the power generation is overestimated: Start the preheating program of the standby conventional unit; send emergency discharge preparation instructions to the energy storage system, including: The current available power generation capacity of the power grid includes renewable energy output, traditional unit standby, and energy storage available discharge; when When the actual load is greater than the predicted load, power generation is insufficient, and the standby unit preheating and energy storage discharge preparation are triggered to quickly make up for the power shortage; is the currently available power generation capacity.

8. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The multi-energy coordinated scheduling includes: Objective function: ;in, It is the fuel cost item of traditional units, which is used to minimize the economic cost of traditional energy power generation; The charging and discharging cost item of the energy storage system is used to balance the peak-shaving and valley-filling function of energy storage with the operating cost; is the penalty term for the new energy consumption rate, which is used to maximize the new energy consumption. For new energy power, is the total power generation; Constraints: Power Balance: They are new energy power, energy storage power, and traditional unit power; Device constraints: ; in, For new energy power; is the power of traditional unit; The power range of traditional units is as follows: the lower limit is the minimum technical output and the upper limit is the rated power; The energy storage power range, the negative sign indicates charging, the positive sign indicates discharging, and the upper limit is the maximum charging and discharging power; Dispatch plan generation: A genetic algorithm is used to solve the objective function, with chromosomes encoding the output ratios of each power source. Combined with forward simulation using the digital twin model, feasible solutions with no line overload and acceptable voltage are screened, and a three-dimensional dispatch plan is output, including the amount of new energy dispatch, energy storage charging and discharging plans, and traditional unit start and stop strategies. After solving the objective function using the genetic algorithm, the final solution is screened through the following steps: Feasibility verification: Call the digital twin module to simulate candidate solutions and eliminate solutions that cause line overload or voltage limit violations; Multi-objective evaluation: Calculate a comprehensive score for the remaining feasible solutions ; Optimal solution selection: output the solution with the highest comprehensive score.

9. The real-time optimization and dispatching method for power grid based on digital twin according to claim 2 is characterized in that: The scheduling model optimization includes: Evaluating the power supply efficiency coefficient ; in, is the actual power supply in area i; Forecast electricity demand for region i; is the total number of regions participating in the power supply efficiency coefficient assessment; When , similar scenario optimization solutions are extracted from the power grid dispatch knowledge graph, where is the actual power supply in area i; Forecast electricity demand for region i: Perform transfer learning on the load forecasting module of the digital twin model to enhance the recognition of extreme scenario features; Adjust the fitness function weights of the genetic algorithm.

Images