Power grid load dynamic prediction and optimal scheduling method, device and equipment and medium

Abstract

The application relates to the technical field of power distribution network dispatching. Through providing a power grid load dynamic prediction and optimal dispatching method, device, equipment and medium, the method comprises the following steps: performing multi-source heterogeneous fusion processing on meteorological parameters, historical load curves and new energy output data to generate a dynamic load prediction atlas; a dynamic network model is constructed, and based on the dynamic network model, a power flow distribution simulation process is performed to obtain stable margin calculation results and a preset safety threshold boundary; a global dispatching target is subjected to stage decomposition processing to generate a gradual dispatching stage sequence; the response characteristics of power generation equipment are subjected to matching processing to generate an adaptive gradual dispatching instruction sequence; the adaptive gradual dispatching instruction sequence is executed, and the real-time state of the power grid is subjected to feedback processing to generate a dynamic adjustment instruction, so that multi-dimensional data correlation modeling, stable margin quantitative analysis and dynamic instruction optimization are realized, thereby improving the load prediction accuracy, reducing fault diffusion and transient oscillation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network dispatching, in particular to a power grid load dynamic prediction and optimal dispatching method, device, equipment and medium. BACKGROUND

[0002] With the accelerated construction of new power systems, the high proportion of new energy access and the rapid growth of diversified loads have put higher requirements on power grid dispatching technology.

[0003] However, the related power grid dispatching technology has the following problems in multi-source heterogeneous data dynamic coupling modeling: the dynamic load prediction accuracy is limited by the lack of multi-dimensional correlation modeling of weather, new energy and historical load; the dispatching model lacks quantitative mapping relationship of new energy volatility and power grid dynamic stability margin, which easily induces cascading failure risk; the dispatching instruction generation mechanism and device dynamic response characteristics are mismatched, and the closed-loop feedback mechanism is lacking, which leads to weakening of transient oscillation suppression ability and dynamic deterioration of trajectory deviation. SUMMARY

[0004] Therefore, it is necessary to provide a power grid load dynamic prediction and optimal dispatching method, device, equipment and medium to realize multi-dimensional data correlation modeling, stability margin quantitative analysis and dynamic instruction optimization, so as to improve the load prediction accuracy, reduce fault propagation and transient oscillation.

[0005] In a first aspect, the present application provides a power grid load dynamic prediction and optimal dispatching method, which comprises:

[0006] Obtaining weather parameters, historical load curves and new energy output data, performing multi-source heterogeneous fusion processing on the weather parameters, historical load curves and new energy output data, and generating a dynamic load prediction atlas;

[0007] Analyzing and processing the power grid topology structure, constructing a dynamic network model, simulating the power flow distribution based on the dynamic network model, and obtaining the stability margin calculation results and preset safety threshold boundary of each node;

[0008] According to the dynamic load prediction atlas and the stability margin calculation results, the global dispatching target is processed by stage decomposition, and a gradual dispatching stage sequence is generated;

[0009] Matching the response characteristics of the power generation equipment, generating an adaptive gradual dispatching instruction sequence based on the gradual dispatching stage sequence;

[0010] Executing the adaptive gradual dispatching instruction sequence and feeding back the real-time state of the power grid, and generating a dynamic adjustment instruction.

[0011] Further, according to the dynamic load prediction map and the stability margin calculation result, the global scheduling target is processed by stage decomposition to generate a progressive scheduling stage sequence, including:

[0012] Using the following formula, according to the load gradient distribution characteristics of the dynamic load prediction map and the stability margin calculation result, the bottleneck area of the whole network transmission channel is identified and processed to generate a multi-dimensional risk heat map:

[0013]

[0014] Wherein, L g represents the load gradient index, n represents the total number of nodes, P i represents the active power of node i, V i represents the voltage amplitude of node i, ΔV i represents the voltage deviation, S m represents the stability margin index, λ max represents the maximum eigenvalue, λ min represents the minimum eigenvalue, m represents the total number of eigenvalues, λ k represents the kth eigenvalue;

[0015] Based on the spatio-temporal evolution trend of the multi-dimensional risk heat map, the load transfer priority of the global scheduling target is sorted and the time window is segmented to generate a progressive scheduling stage sequence.

[0016] Further, according to the load gradient distribution characteristics of the dynamic load prediction map and the stability margin calculation result, the bottleneck area of the whole network transmission channel is identified and processed to generate a multi-dimensional risk heat map, including:

[0017] According to the load gradient distribution characteristics of the dynamic load prediction map, the load change direction and rate of the whole network are processed by trend analysis to generate a gradient evolution path;

[0018] Based on the stability margin calculation result and the gradient evolution path, the node bearing capacity attenuation area of the transmission channel is located and processed to generate an initial bottleneck node set;

[0019] Using the following formula, according to the topological connection relationship of the initial bottleneck node set, the power coupling strength of the adjacent transmission channel is processed by correlation analysis to generate a risk diffusion path:

[0020]

[0021] Wherein, P ij represents the power coupling strength between node i and node j, α ij represents the coupling coefficient, N i represents the neighbor node set of node i, d ibdenotes the physical distance between node i and node b, σ denotes the standard deviation parameter of the Gaussian kernel function, R i denotes the risk propagation intensity of node i, β denotes the risk propagation coefficient, ω ij denotes the weight coefficient between node i and node j, t ij denotes the signal propagation time, τ denotes the time decay constant, h denotes the total number of nodes in the network;

[0022] The gradient evolution path, the initial bottleneck node set and the risk diffusion path are fused to generate a multi-dimensional risk thermal map including risk levels, propagation directions and emergency level labels.

[0023] Further, the response characteristics of the power generation equipment are matched, and an adaptive gradual scheduling instruction sequence is generated based on the gradual scheduling phase sequence, including:

[0024] The response characteristics of the power generation equipment are dynamically capability-mapped to generate a ramp rate constraint domain;

[0025] Based on the spatiotemporal scope division result of the gradual scheduling phase sequence, the capacity matching degree of the load adjustment demand of each sub-phase is calculated and processed to generate a phase instruction amplitude upper limit;

[0026] According to the ramp rate constraint domain and the phase instruction amplitude upper limit, the timing execution interval of the scheduling instruction is dynamically planned to generate an instruction step optimization parameter;

[0027] The instruction step optimization parameter and the priority weight of the gradual scheduling phase sequence are fused to generate an adaptive gradual scheduling instruction sequence.

[0028] Further, according to the ramp rate constraint domain and the phase instruction amplitude upper limit, the timing execution interval of the scheduling instruction is dynamically planned to generate an instruction step optimization parameter, including:

[0029] Based on the ramp rate constraint domain's margin decay rate and the phase instruction amplitude upper limit's load mutation threshold, the timing interval of the scheduling instruction is analyzed and processed in a reverse propagation risk to generate a safe execution time window;

[0030] According to the boundary constraint condition of the safe execution time window, the instruction superposition effect between adjacent scheduling phases is dynamically time-window-divided to generate an instruction execution interval threshold;

[0031] The safe execution time window and the instruction execution interval threshold are fused to generate an instruction step optimization parameter.

[0032] Further, the adaptive gradual scheduling instruction sequence is executed, and the real-time state of the power grid is fed back to generate a dynamic adjustment instruction, including:

[0033] According to the space-time constraint condition of the adaptive progressive scheduling instruction sequence, step-by-step instruction issuing processing is performed on the generator set and the energy storage device to generate an initial scheduling execution trajectory;

[0034] Real-time state monitoring processing is performed on the node voltage, frequency and power flow of the initial scheduling execution trajectory to generate a power grid real-time state data set;

[0035] Dynamic trajectory deviation analysis processing is performed on the power grid real-time state data set and the expected scheduling path to generate trajectory deviation parameters and safety risk level labels;

[0036] Based on the trajectory deviation parameters and the safety risk level labels, dynamic compensation processing is performed on the step size and direction of the adaptive progressive scheduling instruction sequence to generate dynamic adjustment instructions.

[0037] Further, the dynamic trajectory deviation analysis processing is performed on the power grid real-time state data set and the expected scheduling path to generate trajectory deviation parameters and safety risk level labels, including:

[0038] Temporal and spatial alignment processing is performed on the power grid real-time state data set to generate a standardized state monitoring sequence matching the expected scheduling path;

[0039] Based on the time-domain waveform characteristics of the standardized state monitoring sequence and the expected scheduling path, dynamic trajectory similarity comparison processing is performed to generate a trajectory deviation degree parameter;

[0040] Time window cumulative effect analysis processing is performed on the trajectory deviation degree parameter to generate a risk diffusion trend prediction index;

[0041] According to the risk diffusion trend prediction index and the preset safety threshold boundary, risk level classification processing is performed on the deviation region to generate trajectory deviation parameters and safety risk level labels.

[0042] In a second aspect, the present application also provides a power grid load dynamic prediction and optimized scheduling device, which comprises:

[0043] A multi-source data fusion module is configured to obtain meteorological parameters, historical load curves and new energy output data, perform multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data, and generate a dynamic load prediction map;

[0044] A network modeling and power flow analysis module is configured to perform analysis processing on the power grid topology, construct a dynamic network model, perform power flow distribution simulation processing based on the dynamic network model, and obtain stable margin calculation results and a preset safety threshold boundary of each node;

[0045] A stage decomposition optimization module is configured to perform stage decomposition processing on a global scheduling target based on the dynamic load prediction map and the stable margin calculation results, and generate a progressive scheduling stage sequence;

[0046] a response characteristic matching module, configured to perform matching processing on the response characteristic of the power generation device, generate an adaptive progressive scheduling instruction sequence based on the progressive scheduling stage sequence;

[0047] a dynamic feedback execution module, configured to execute the adaptive progressive scheduling instruction sequence and perform feedback processing on the real-time state of the power grid to generate a dynamic adjustment instruction.

[0048] In a third aspect, the present application also provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps of any method in the first aspect of the present application when executing the computer program.

[0049] In a fourth aspect, the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of any method in the first aspect of the present application.

[0050] The technical scheme provided by the present application includes the following technical effects: by providing the power grid load dynamic prediction and optimal scheduling method, device, equipment and medium, the method comprises: acquiring meteorological parameters, historical load curve and new energy output data, performing multi-source heterogeneous fusion processing on the meteorological parameters, historical load curve and new energy output data to generate a dynamic load prediction atlas; performing analysis processing on the power grid topology structure, constructing a dynamic network model, performing power flow distribution simulation processing based on the dynamic network model to obtain the stable margin calculation result of each node and the preset safety threshold boundary; based on the dynamic load prediction atlas and the stable margin calculation result, performing stage decomposition processing on the global scheduling target to generate a progressive scheduling stage sequence; performing matching processing on the response characteristic of the power generation device, generating an adaptive progressive scheduling instruction sequence based on the progressive scheduling stage sequence; executing the adaptive progressive scheduling instruction sequence and performing feedback processing on the real-time state of the power grid to generate a dynamic adjustment instruction, so as to realize multi-dimensional data correlation modeling, stable margin quantitative analysis and dynamic instruction optimization, thereby improving the load prediction accuracy, reducing fault diffusion and transient oscillation. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical scheme in the embodiments of the present application or related technologies, the following will briefly introduce the drawings needed to be used in the embodiment or related technology description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0052] Figure 1 Flowchart of the power grid load dynamic prediction and optimal scheduling method in an embodiment of the present application;

[0053] Figure 2 For the matching process of the response characteristics of the power generation equipment in an embodiment of the application, a flowchart for generating an adaptive progressive scheduling instruction sequence based on a sequence of progressive scheduling stages is provided.

[0054] Figure 3 For the structure diagram of the power grid load dynamic prediction and optimal scheduling device in an embodiment of the application. DETAILED DESCRIPTION

[0055] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation mode of the present application will be described in detail below. In the following description, many specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the scope of the application, so the present application is not limited to the specific embodiments disclosed below.

[0056] As shown in Figure 1 The present application provides a power grid load dynamic prediction and optimal scheduling method, which comprises:

[0057] S101: Obtain meteorological parameters, historical load curves and new energy output data, and perform multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data to generate a dynamic load prediction atlas.

[0058] Specifically, real-time meteorological parameters, historical load curves and new energy output data are collected, which are derived from multiple channels such as power grid operation monitoring systems, meteorological departments, power trading markets and new energy power stations, and cover key information such as temperature, humidity, wind speed, historical period electricity load data and photovoltaic, wind power and other new energy power generation curves. Then, multi-source heterogeneous fusion processing is performed, which includes data standardization conversion, unifies different sources and different formats of data to a standard format framework, ensures the comparability and fusibility between data, such as adapting the time resolution of meteorological data to the time resolution of power data, and performing feature extraction to identify the key feature dimensions in the data set that can reflect the trend of power grid load change, such as extracting temperature and humidity indicators from meteorological data that affect electricity load, extracting load peak, trough and fluctuation characteristics from historical load curves, etc. At the same time, data correlation integration is also performed to establish a correlation model between meteorological parameters, historical load and new energy output, analyze their mutual influence relationship, and then generate a dynamic load prediction atlas.

[0059] S102: Analyze the topology structure of the power grid, construct a dynamic network model, perform power flow distribution simulation based on the dynamic network model, and obtain the stable margin calculation results of each node and the preset safety threshold boundary.

[0060] Specifically, the connection information of the power grid is acquired, including the line connection mode, substation distribution, power station access point, and load center location, etc., to determine the node and branch composition of the power grid. Then, according to the electrical parameters of the power grid elements, such as line impedance, transformer ratio, generator output limit, etc., a network model capable of reflecting the dynamic characteristics of the power grid is constructed, and the steady-state operation parameters and transient response characteristics of the power grid are included in the model system. On this basis, the power flow distribution simulation is carried out on the constructed dynamic network model by using the power flow calculation algorithm such as Newton-Raphson method or PQ decomposition method, etc. The power balance, voltage level, and reactive power distribution of the power grid under different operating conditions are comprehensively considered to determine the stable margin calculation results of each node, and the stable operation range of the power grid under the current state and the predicted state is determined. At the same time, combined with the historical data and safe operation standard of the power grid operation, a preset safety threshold boundary is set for each node to monitor the operation state of the power grid in real time during the dispatching process, and to provide key basis for subsequent dispatching decision.

[0061] S103: According to the dynamic load prediction map and the stable margin calculation result, the global scheduling target is processed by stage decomposition to generate a progressive scheduling stage sequence.

[0062] Specifically, the dynamic load prediction map is analyzed in depth to obtain the load change trend, fluctuation amplitude, and time period of load peak and valley of each future period, etc. key information, and the stable margin calculation results of each node are detailed to determine the stability limit and safety boundary of the power grid under different operating conditions. Based on the comprehensive understanding of the load characteristics and stable margin, the global scheduling target is reasonably split according to the time dimension and space dimension. In the time dimension, according to the natural law and fluctuation characteristics of the load change, the scheduling stage is divided to ensure the relative consistency of the load characteristics of each stage, so as to formulate the targeted scheduling strategy; in the space dimension, the geographical distribution, load density, and power supply access point of the power grid are considered to reasonably allocate the scheduling tasks of different regions. Through the above more refined decomposition method, a sequence containing multiple progressive scheduling stages is generated, each stage has a clear load adjustment target, power balance requirement, and safety and stability constraint, to gradually realize the stable operation and optimal scheduling of the power grid.

[0063] S104: The response characteristics of the power generation equipment are matched and processed to generate an adaptive progressive scheduling instruction sequence based on the progressive scheduling stage sequence.

[0064] Specifically, the response characteristics of the power generation equipment are analyzed in depth, including key indicators such as the power regulation capability of the equipment, the climb rate limit, and the response sensitivity to frequency and voltage changes. Then, the gradual scheduling phase sequence is disassembled to clarify the specific tasks of each phase, such as load adjustment requirements, power balance targets, and node voltage and frequency stability requirements. According to the response characteristics of the power generation equipment, an appropriate combination of power generation equipment is matched for each scheduling task, and the output adjustment amplitude and timing of each device are determined, while considering the operating constraints of the equipment, such as maximum output limit, minimum stable operating power, etc.

[0065] The scheduling instructions are refined into specific operation steps, including generator set power increase and decrease instructions, start and stop operation instructions, and reactive power regulation instructions, etc., and are arranged in chronological order and logical relationship to generate a preliminary scheduling instruction sequence. The generated scheduling instruction sequence is optimized and adjusted to ensure the feasibility of the instructions and the stability of the power grid operation, and to verify whether the instruction sequence meets the safe operation standards and economic operation requirements of the power grid, thereby generating an adaptive gradual scheduling instruction sequence to guide the reasonable adjustment of power generation equipment in different stages to meet the scheduling needs of the power grid.

[0066] S105: Execute the adaptive gradual scheduling instruction sequence and perform real-time state feedback processing on the power grid to generate dynamic adjustment instructions.

[0067] Specifically, executing the adaptive gradual scheduling instruction sequence involves passing the scheduling instructions down to each power generation unit and energy storage device, etc. execution unit, adjusting its power output according to the time and amplitude specified in the instructions. At the same time, a real-time monitoring system is established to continuously track the operation state of the power grid, focusing on monitoring key indicators such as whether the node voltage is stable within the rated range, whether the frequency is maintained at the standard level, and whether the power flow is evenly distributed, etc. Real-time data is collected through sensors and monitoring equipment installed at each node of the power grid, which can clearly reflect the actual operation status of the power grid after executing the scheduling instructions. Then, the data obtained through real-time monitoring is compared and analyzed with the expected target of the scheduling instructions to obtain the deviation between the actual operation state and the expected target, such as node voltage deviation, frequency offset, and power imbalance, etc. The above deviation will be used as the basis for subsequent adjustment.

[0068] According to the results of the deviation analysis, the scheduling optimization algorithm is used to recalculate and generate dynamic adjustment instructions, which will correct and optimize the original scheduling plan to ensure that the power grid can operate safely, stably, and efficiently. The dynamic adjustment instructions will be issued to each execution unit again, forming a closed-loop scheduling control process that continuously adjusts according to the actual operation of the power grid to achieve the optimal operation state of the power grid.

[0069] An embodiment of the present application also provides a power grid load dynamic prediction and optimal scheduling method, comprising: acquiring meteorological parameters, historical load curves and new energy output data, performing multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data, and generating a dynamic load prediction atlas; performing analysis processing on the power grid topology structure, constructing a dynamic network model, performing power flow distribution simulation processing based on the dynamic network model, and obtaining stable margin calculation results and a preset safety threshold boundary of each node; performing stage decomposition processing on the global scheduling target according to the dynamic load prediction atlas and the stable margin calculation results, and generating a gradual scheduling stage sequence; performing matching processing on the response characteristics of the power generation equipment, generating an adaptive gradual scheduling instruction sequence based on the gradual scheduling stage sequence; executing the adaptive gradual scheduling instruction sequence, and performing feedback processing on the real-time state of the power grid, generating a dynamic adjustment instruction, so as to realize multi-dimensional data correlation modeling, stable margin quantitative analysis and dynamic instruction optimization, thereby improving the load prediction accuracy, reducing fault diffusion and transient oscillation.

[0070] Further, the global scheduling target is subjected to stage decomposition processing according to the dynamic load prediction atlas and the stable margin calculation results, and a gradual scheduling stage sequence is generated, comprising:

[0071] The following formula is used to identify the bottleneck area of the whole network transmission channel according to the load gradient distribution characteristics of the dynamic load prediction atlas and the stable margin calculation results, and generate a multi-dimensional risk heat map:

[0072]

[0073]

[0074] wherein, L g represents a load gradient index, n represents the total number of nodes, P i represents the active power of node i, V i represents the voltage amplitude of node i, ΔV i represents the voltage deviation, S m represents a stable margin index, λ max represents the maximum eigenvalue, λ min represents the minimum eigenvalue, m represents the total number of eigenvalues, λ k represents the kth eigenvalue;

[0075] Based on the time-space evolution trend of the multi-dimensional risk heat map, the load transfer priority of the global scheduling target is sorted and the time period window is segmented, and a gradual scheduling stage sequence is generated.

[0076] Specifically, the load gradient distribution characteristics in the dynamic load prediction atlas are analyzed to determine the change rate and direction of the load in different regions and time periods, and the stability limit and risk level of each node of the power grid are determined by combining the stability margin calculation results. Based on the above information, the transmission channels of the entire network are comprehensively evaluated using a specific algorithm to identify potential bottleneck regions, including congestion points during peak load periods and weak links with low stability margins. Then, the identified bottleneck regions and their risk levels are integrated to generate a multi-dimensional risk heat map that not only reflects the current risk distribution of the power grid but also predicts its spatiotemporal evolution trend.

[0077] According to the evolution trend of the heat map, the global dispatching target is refined: on the one hand, according to the urgency and impact range of the risk, the load transfer task is prioritized, and the load congestion problem in high-risk areas is solved first; on the other hand, the dispatching period is reasonably divided, and corresponding dispatching strategies are developed for time windows with different risk levels and load characteristics, thereby generating a phased and prioritized progressive dispatching stage sequence, providing clear guidance and planning for subsequent dispatching operations.

[0078] Further, according to the load gradient distribution characteristics of the dynamic load prediction atlas and the stability margin calculation results, the bottleneck region identification process is performed on the transmission channels of the entire network to generate a multi-dimensional risk heat map, including:

[0079] According to the load gradient distribution characteristics of the dynamic load prediction atlas, the trend analysis process is performed on the load change direction and rate of the entire network to generate a gradient evolution path.

[0080] Based on the stability margin calculation results and the gradient evolution path, the node carrying capacity decay region of the transmission channel is located to generate an initial bottleneck node set.

[0081] Using the following formula, the power coupling strength of adjacent transmission channels is analyzed based on the topological connection relationship of the initial bottleneck node set to generate a risk diffusion path:

[0082]

[0083] where Pij represents the power coupling strength between node i and node j, α represents the coupling coefficient, N(i) represents the neighbor node set of node i, di,j represents the physical distance between node i and node b, σ represents the standard deviation parameter of the Gaussian kernel function, R(i) represents the risk propagation strength of node i, β represents the risk propagation coefficient, ωij represents the weight coefficient between node i and node j, and t represents the time. ij ij i ib i ij ij ​​​​​​denotes signal propagation time, τ denotes time decay constant, and h denotes total number of nodes in the network;

[0084] The gradient evolution path, the initial bottleneck node set and the risk diffusion path are fused to generate a multi-dimensional risk heat map including risk level, propagation direction and emergency level label.

[0085] Specifically, based on the dynamic load prediction map, the direction and rate information of load change in each region are extracted to construct a gradient evolution path and determine the development trend of load in the future space-time range. Then, combined with the calculation result of stable margin, the carrying capacity of nodes in the transmission channel is analyzed to identify the carrying capacity attenuation nodes that may occur under high load and generate an initial bottleneck node set. Then, the topological connection relationship is introduced, and the coupling strength of power between adjacent transmission channels is calculated by analyzing the physical distance, connection strength and other parameters between the initial bottleneck node and its adjacent nodes, so as to determine the diffusion path of risk. The gradient evolution path, the initial bottleneck node set and the risk diffusion path are comprehensively analyzed to generate a multi-dimensional risk heat map with risk level, propagation direction and emergency level as dimensions, which provides a more comprehensive and intuitive risk assessment basis for power grid dispatching decision.

[0086] As Figure 2 described, the response characteristics of the power generation equipment are matched, and an adaptive gradual scheduling instruction sequence is generated based on a gradual scheduling stage sequence, including:

[0087] S201: dynamically mapping the response characteristics of the power generation equipment to generate a ramping rate constraint domain;

[0088] S202: based on the time and space scope division result of the gradual scheduling stage sequence, calculating the capacity matching degree of the load adjustment demand of each sub-stage to generate a stage instruction amplitude upper limit;

[0089] S203: dynamically planning the time sequence execution interval of the scheduling instruction according to the ramping rate constraint domain and the stage instruction amplitude upper limit to generate an instruction step length optimization parameter;

[0090] S204: fusing the instruction step length optimization parameter and the priority weight of the gradual scheduling stage sequence to generate an adaptive gradual scheduling instruction sequence.

[0091] Specifically, the response characteristics of the power generation equipment are comprehensively mapped to obtain the ramp rate limit of the power output, and a clear ramp rate constraint domain is generated to ensure that the equipment does not exceed its adjustment capability range during scheduling. Then, based on the spatiotemporal scope division result of the sequence of progressive scheduling stages, the capacity matching degree of the power generation equipment is calculated for the load adjustment requirement of each sub-stage, so as to determine the upper limit of the instruction amplitude of each stage, and to ensure that the load demand is met while avoiding equipment overload. By comprehensively considering the ramp rate constraint domain and the upper limit of the stage instruction amplitude, the dynamic programming technique is used to reasonably arrange the time sequence execution interval of the scheduling instruction, generate scientific instruction step optimization parameters, and ensure the orderliness and executability of the scheduling instruction. Then, the instruction step optimization parameters and the priority weight of the sequence of progressive scheduling stages are fully integrated, and the emergency and importance of each stage are comprehensively considered to generate an adaptive progressive scheduling instruction sequence, so as to realize more accurate, efficient and stable scheduling of the power grid, and ensure safe and reliable operation of the power grid in different stages.

[0092] Further, according to the ramp rate constraint domain and the upper limit of the stage instruction amplitude, the time sequence execution interval of the scheduling instruction is dynamically programmed to generate instruction step optimization parameters, including:

[0093] Based on the margin decay rate of the ramp rate constraint domain and the load mutation threshold of the upper limit of the stage instruction amplitude, the time sequence interval of the scheduling instruction is analyzed by the backpropagation risk analysis technique to generate a safe execution time window;

[0094] According to the boundary constraint condition of the safe execution time window, the instruction superposition effect between adjacent scheduling stages is dynamically time windowed to generate an instruction execution interval threshold;

[0095] The safe execution time window and the instruction execution interval threshold are integrated to generate the instruction step optimization parameters.

[0096] Specifically, the margin decay rate of the ramp rate constraint domain and the load mutation threshold of the upper limit of the stage instruction amplitude are combined to evaluate the time sequence interval of the scheduling instruction by the backpropagation risk analysis technique, identify potential risk points, and then determine the safe execution time window to ensure safe and stable operation of the power grid. According to the boundary constraint condition of the safe execution time window, the power grid load fluctuation and other problems caused by the instruction superposition between adjacent scheduling stages are analyzed, the time interval is reasonably divided by the dynamic time windowing technique, the instruction execution interval threshold is formulated, and the ordered execution of the instructions of each stage is ensured without interference. The safe execution time window and the instruction execution interval threshold are organically integrated, and the safety, stability and economy of the power grid operation and other factors are comprehensively considered, and the instruction step optimization parameters are calculated and generated by using the optimization algorithm, which provides a key basis for generating a scientific and reasonable adaptive progressive scheduling instruction sequence, and realizes more accurate and efficient scheduling control of the power grid.

[0097] Further, the adaptive progressive scheduling instruction sequence is executed, and the real-time state of the power grid is processed for feedback to generate dynamic adjustment instructions, including:

[0098] According to the space-time constraint conditions of the adaptive progressive scheduling instruction sequence, the generator set and the energy storage device are processed for step-by-step instruction issuing to generate an initial scheduling execution trajectory;

[0099] The node voltage, frequency and power flow of the initial scheduling execution trajectory are processed for real-time state monitoring to generate a power grid real-time state data set;

[0100] The power grid real-time state data set and the expected scheduling path are processed for dynamic trajectory deviation analysis to generate trajectory deviation parameters and safety risk level labels;

[0101] Based on the trajectory deviation parameters and the safety risk level labels, the step and direction of the adaptive progressive scheduling instruction sequence are processed for dynamic compensation to generate dynamic adjustment instructions.

[0102] Specifically, according to the space-time constraint conditions of the instruction sequence, the instructions are issued step by step to the generator set and the energy storage device, thereby generating an initial scheduling execution trajectory, ensuring that each device adjusts gradually according to the predetermined plan. At the same time, a real-time monitoring system is established to continuously track the operation state of the power grid, focusing on monitoring key indicators such as whether the node voltage is stable, whether the frequency is normal, and whether the power flow is balanced, and collecting and integrating the above real-time data to generate a power grid real-time state data set. Then, the real-time state data set is compared and analyzed with the expected scheduling path, through dynamic trajectory deviation analysis technology, the deviation parameters between the actual running trajectory and the expected trajectory are calculated, and according to the size and trend of the deviation, combined with the safety risk assessment model, the corresponding safety risk level label is generated. Based on the trajectory deviation parameters and the safety risk level labels, the step and direction of the adaptive progressive scheduling instruction sequence are dynamically compensated and adjusted using a scheduling optimization algorithm to generate dynamic adjustment instructions to correct the deviation and reduce the risk, ensuring the accuracy and stability of the power grid scheduling.

[0103] Further, the power grid real-time state data set and the expected scheduling path are processed for dynamic trajectory deviation analysis to generate trajectory deviation parameters and safety risk level labels, including:

[0104] The power grid real-time state data set is processed for space-time alignment to generate a standardized state monitoring sequence matching the expected scheduling path;

[0105] Based on the time-domain waveform characteristics of the standardized state monitoring sequence and the expected scheduling path, dynamic trajectory similarity comparison processing is performed to generate a trajectory deviation degree parameter;

[0106] The trajectory deviation parameter is subjected to time window cumulative effect analysis processing to generate a risk diffusion trend prediction index.

[0107] According to the risk diffusion trend prediction index and the preset safety threshold boundary, the deviation region is subjected to risk level classification processing to generate a trajectory deviation parameter and a safety risk level label.

[0108] Specifically, the real-time state data set is subjected to spatio-temporal alignment processing to ensure that the time sequence and spatial distribution of the data match the expected scheduling path, and a standardized state monitoring sequence is generated. Based on the time domain waveform features of the standardized sequence and the expected path, a dynamic trajectory similarity comparison algorithm is used to calculate the trajectory deviation parameter, which quantifies the difference between the actual running trajectory and the expected trajectory. Through time window cumulative effect analysis, the cumulative influence of the trajectory deviation parameter in the time dimension is evaluated, the risk diffusion trend is predicted, and a risk diffusion trend prediction index is generated. In combination with the risk diffusion trend prediction index and the preset safety threshold boundary, the risk level of the region with deviation is classified to determine its risk level, and a trajectory deviation parameter and a corresponding safety risk level label are generated to provide a basis for subsequent scheduling adjustment.

[0109] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0110] In one embodiment, as shown in Figure 3 The present application also provides a power grid load dynamic prediction and optimized scheduling device 300, which comprises:

[0111] A multi-source data fusion module 301 is configured to acquire meteorological parameters, historical load curves and new energy output data, and perform multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data to generate a dynamic load prediction atlas.

[0112] A network modeling and power flow analysis module 302 is configured to analyze the topology structure of the power grid, construct a dynamic network model, and perform power flow distribution simulation based on the dynamic network model to obtain stable margin calculation results of each node and a preset safety threshold boundary.

[0113] a stage decomposition optimization module 303 configured to perform stage decomposition processing on the global scheduling target according to the dynamic load prediction atlas and the stability margin calculation result, and generate a progressive scheduling stage sequence;

[0114] a response characteristic matching module 304 configured to perform matching processing on the response characteristics of the power generation equipment, and generate an adaptive progressive scheduling instruction sequence based on the progressive scheduling stage sequence;

[0115] a dynamic feedback execution module 305 configured to execute the adaptive progressive scheduling instruction sequence, and perform feedback processing on the real-time state of the power grid, and generate a dynamic adjustment instruction.

[0116] Specifically, the multi-source data fusion module 301 acquires meteorological parameters, historical load curves, and new energy output data. The above data sources are diverse and have different formats. The module fuses the multi-source heterogeneous data into a dynamic load prediction atlas of a unified format through data cleaning, standardized conversion, and feature extraction processing steps, thereby providing a basis for subsequent analysis. The network modeling and power flow analysis module 302 is responsible for analyzing the topology structure of the power grid, constructing a dynamic network model including nodes, branches, and electrical parameters, and simulating the power flow distribution of the power grid by using a power flow calculation algorithm to obtain the stability margin and safety threshold boundary of each node. The stage decomposition optimization module 303 combines the dynamic load prediction atlas and the stability margin result, and uses time sequence segmentation and priority sorting methods to decompose the global scheduling target into a plurality of progressive scheduling stage sequences, so that the scheduling task is more operable. The response characteristic matching module 304 analyzes the response characteristics of the power generation equipment, such as the power regulation capability and the climbing rate, and generates an adaptive progressive scheduling instruction sequence matched with the equipment characteristics according to the requirements of the progressive scheduling stage sequence. The dynamic feedback execution module 305 executes the scheduling instruction sequence, monitors the state of the power grid in real time, generates a dynamic adjustment instruction through a feedback mechanism, and ensures that the operation of the power grid meets the scheduling target.

[0117] The stage decomposition optimization module 303 is further configured to:

[0118] use the following formula to perform bottleneck area identification processing on the transmission channels of the entire power grid according to the load gradient distribution characteristics of the dynamic load prediction atlas and the stability margin calculation result, and generate a multi-dimensional risk heat map:

[0119]

[0120] wherein, L g represents a load gradient index, n represents the total number of nodes, P i represents the active power of node i, V i represents the voltage amplitude of node i, ΔV i represents the voltage deviation, S m represents a stability margin index, and λ maxdenotes the maximum eigenvalue, λ min denotes the minimum eigenvalue, m denotes the total number of eigenvalues, λ k denotes the kth eigenvalue;

[0121] Based on the spatio-temporal evolution trend of the multi-dimensional risk thermodynamic map, the global scheduling target is prioritized for load transfer and the time window is segmented to generate a gradual scheduling stage sequence.

[0122] The stage decomposition optimization module 303 is further configured to:

[0123] According to the load gradient distribution characteristics of the dynamic load prediction map, the trend of the global load change direction and rate is analyzed to generate a gradient evolution path;

[0124] Based on the stability margin calculation result and the gradient evolution path, the node carrying capacity attenuation area of the transmission channel is located to generate an initial bottleneck node set;

[0125] Using the following formula, the power coupling strength of adjacent transmission channels is analyzed according to the topological connection relationship of the initial bottleneck node set to generate a risk diffusion path:

[0126]

[0127] wherein, P ij denotes the power coupling strength between node i and node j, α ij denotes the coupling coefficient, N i denotes the neighbor node set of node i, d ib denotes the physical distance between node i and node b, σ denotes the standard deviation parameter of the Gaussian kernel function, R i denotes the risk propagation strength of node i, β denotes the risk propagation coefficient, ω ij denotes the weight coefficient between node i and node j, t ij denotes the signal propagation time, τ denotes the time attenuation constant, h denotes the total number of nodes in the network;

[0128] The gradient evolution path, the initial bottleneck node set and the risk diffusion path are fused to generate a multi-dimensional risk thermodynamic map including risk level, propagation direction and emergency level label.

[0129] The response characteristic matching module 304 is further configured to:

[0130] The response characteristics of the power generation equipment are dynamically mapped to generate a ramp rate constraint domain;

[0131] Based on the spatio-temporal scope division result of the gradual scheduling stage sequence, the load adjustment demand of each sub-stage is calculated to generate a stage instruction amplitude upper limit;

[0132] According to the ramp rate constraint domain and the upper limit of the phase instruction amplitude, the timing execution interval of the scheduling instruction is dynamically planned to generate an instruction step optimization parameter;

[0133] The priority weight of the instruction step optimization parameter and the gradual scheduling phase sequence is fused to generate an adaptive gradual scheduling instruction sequence.

[0134] The response characteristic matching module 304 is also used for:

[0135] Based on the margin decay rate of the ramp rate constraint domain and the load mutation threshold of the upper limit of the phase instruction amplitude, the timing interval of the scheduling instruction is analyzed by reverse propagation risk to generate a safe execution time window;

[0136] According to the boundary constraint condition of the safe execution time window, the instruction superposition effect between adjacent scheduling phases is dynamically time window divided to generate an instruction execution interval threshold;

[0137] The safe execution time window and the instruction execution interval threshold are fused to generate an instruction step optimization parameter.

[0138] The dynamic feedback execution module 305 is also used for:

[0139] According to the space-time constraint condition of the adaptive gradual scheduling instruction sequence, the generator set and the energy storage device are processed by step instruction issuing to generate an initial scheduling execution trajectory;

[0140] The node voltage, frequency and power flow of the initial scheduling execution trajectory are processed by real-time state monitoring to generate a power grid real-time state data set;

[0141] The power grid real-time state data set and the expected scheduling path are processed by dynamic trajectory deviation analysis to generate a trajectory deviation parameter and a safety risk level label;

[0142] Based on the trajectory deviation parameter and the safety risk level label, the step and direction of the adaptive gradual scheduling instruction sequence are dynamically compensated to generate a dynamic adjustment instruction.

[0143] The dynamic feedback execution module 305 is also used for:

[0144] The power grid real-time state data set is processed by space-time alignment to generate a standardized state monitoring sequence matched with the expected scheduling path;

[0145] Based on the time domain waveform characteristics of the standardized state monitoring sequence and the expected scheduling path, a dynamic trajectory similarity comparison is performed to generate a trajectory deviation degree parameter;

[0146] The trajectory deviation degree parameter is processed by time window cumulative effect analysis to generate a risk diffusion trend prediction index;

[0147] According to the risk diffusion trend prediction index and the preset safety threshold boundary, the deviation area is classified by risk level, and a trajectory deviation parameter and a safety risk level label are generated.

[0148] In one embodiment, the present application also provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0149] In one embodiment, the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps in the above method embodiments.

[0150] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts are described in the method embodiments. The device embodiments described above are only schematic, and the components described as separate components can or can not be physically separate, and the components displayed as a unit can or can not be a physical unit, i.e. they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present disclosure according to actual needs. Those skilled in the art can understand and implement it without creative labor.

[0151] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the protection scope of the present application.

Claims

1. A method for dynamic prediction and optimal dispatch of power grid load, characterized in that, The method comprises: acquiring meteorological parameters, historical load curves and new energy output data, performing multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data, and generating a dynamic load prediction atlas; performing analysis processing on the power grid topology structure, constructing a dynamic network model, performing power flow distribution simulation processing based on the dynamic network model, and obtaining stable margin calculation results and a preset safety threshold boundary of each node; based on the dynamic load prediction atlas and the stable margin calculation results, performing stage decomposition processing on the global scheduling target, and generating a gradual scheduling stage sequence; the stage decomposition processing on the global scheduling target based on the dynamic load prediction atlas and the stable margin calculation results, and the generation of the gradual scheduling stage sequence, comprise: using the following formula, based on the load gradient distribution characteristics of the dynamic load prediction atlas and the stable margin calculation results, performing bottleneck area identification processing on the whole network transmission channel, and generating a multi-dimensional risk heat map: ； ； wherein, denotes the load gradient indicator, n denotes the total number of nodes, denotes the active power of node i, denotes the voltage amplitude of node i, denotes the voltage deviation amount, denotes the stability margin indicator, denotes the maximum eigenvalue, denotes the minimum eigenvalue, m denotes the total number of eigenvalues, denotes the kth eigenvalue; based on the time and space evolution trend of the multi-dimensional risk heat map, performing load transfer priority sorting and time period window segmentation processing on the global scheduling target, and generating the gradual scheduling stage sequence; the bottleneck area identification processing on the whole network transmission channel based on the load gradient distribution characteristics of the dynamic load prediction atlas and the stable margin calculation results, and the generation of the multi-dimensional risk heat map, comprise: based on the load gradient distribution characteristics of the dynamic load prediction atlas, performing trend analysis processing on the load change direction and rate of the whole network, and generating a gradient evolution path; based on the stable margin calculation results and the gradient evolution path, performing positioning processing on the node bearing capacity attenuation area of the transmission channel, and generating an initial bottleneck node set; using the following formula, based on the topological connection relationship of the initial bottleneck node set, performing correlation analysis processing on the power coupling strength of adjacent transmission channels, and generating a risk diffusion path: ； ； wherein, denotes the power coupling strength between node i and node j, denotes the coupling coefficient, denotes the neighbor node set of node i, denotes the physical distance between node i and node b, denotes the standard deviation parameter of the Gaussian kernel function, denotes the risk propagation strength of node i, denotes the risk propagation coefficient, denotes the weight coefficient between node i and node j, denotes the signal propagation time, denotes the time decay constant, h denotes the total number of nodes in the network; fusing the gradient evolution path, the initial bottleneck node set and the risk diffusion path, and generating the multi-dimensional risk heat map including risk level, propagation direction and emergency level label; performing matching processing on the response characteristics of the power generation equipment, and generating an adaptive gradual scheduling instruction sequence based on the gradual scheduling stage sequence; executing the adaptive gradual scheduling instruction sequence and performing feedback processing on the real-time state of the power grid, and generating a dynamic adjustment instruction. 2.The power grid load dynamic prediction and optimal scheduling method according to claim 1, characterized in that, the matching processing on the response characteristics of the power generation equipment, and the generation of the adaptive gradual scheduling instruction sequence based on the gradual scheduling stage sequence, comprise: performing dynamic capability mapping processing on the response characteristics of the power generation equipment, and generating a ramping rate constraint domain; based on the time and space scope division results of the gradual scheduling stage sequence, performing capacity matching degree calculation processing on the load adjustment demand of each sub-stage, and generating a stage instruction amplitude upper limit; based on the ramping rate constraint domain and the stage instruction amplitude upper limit, performing dynamic planning processing on the time sequence execution interval of the scheduling instruction, and generating an instruction step length optimization parameter; fusing the instruction step length optimization parameter and the priority weight of the gradual scheduling stage sequence, and generating the adaptive gradual scheduling instruction sequence. 3.The power grid load dynamic prediction and optimal scheduling method of claim 2, wherein, The dynamic programming processing is performed on the time sequence interval of the scheduling instruction according to the ramp rate constraint domain and the upper limit of the stage instruction amplitude, and an instruction step length optimization parameter is generated, including: Based on the margin decay rate of the ramp rate constraint domain and the load mutation threshold of the upper limit of the stage instruction amplitude, the time sequence interval of the scheduling instruction is subjected to a backward propagation risk analysis processing, and a safe execution time window is generated; According to the boundary constraint condition of the safe execution time window, the instruction superposition effect between adjacent scheduling stages is subjected to a dynamic time window division processing, and an instruction execution interval threshold is generated; The safe execution time window and the instruction execution interval threshold are fused to generate the instruction step length optimization parameter. 4.The power grid load dynamic prediction and optimal scheduling method of claim 1, wherein, The adaptive gradual scheduling instruction sequence is executed, and a dynamic adjustment instruction is generated by performing feedback processing on the real-time state of the power grid, including: According to the space-time constraint condition of the adaptive gradual scheduling instruction sequence, a step-by-step instruction issuing processing is performed on the generator set and the energy storage device to generate an initial scheduling execution trajectory; The node voltage, frequency and power flow of the initial scheduling execution trajectory are subjected to real-time state monitoring processing to generate a power grid real-time state data set; The power grid real-time state data set and the expected scheduling path are subjected to dynamic trajectory deviation analysis processing to generate a trajectory deviation parameter and a safety risk level label; Based on the trajectory deviation parameter and the safety risk level label, a dynamic compensation processing is performed on the step length and direction of the adaptive gradual scheduling instruction sequence to generate the dynamic adjustment instruction.

5. The power grid load dynamic prediction and optimal scheduling method according to claim 4, characterized in that, The power grid real-time state data set and the expected scheduling path are subjected to dynamic trajectory deviation analysis processing to generate a trajectory deviation parameter and a safety risk level label, including: The power grid real-time state data set is subjected to space-time alignment processing to generate a standardized state monitoring sequence matched with the expected scheduling path; Based on the time domain waveform characteristics of the standardized state monitoring sequence and the expected scheduling path, a dynamic trajectory similarity comparison processing is performed to generate a trajectory deviation degree parameter; The trajectory deviation degree parameter is subjected to time window cumulative effect analysis processing to generate a risk diffusion trend prediction index; According to the risk diffusion trend prediction index and the preset safety threshold boundary, a risk level classification processing is performed on the deviation region to generate the trajectory deviation parameter and the safety risk level label.

6. A power grid load dynamic prediction and optimal dispatching device, characterized in that, The device includes: A multi-source data fusion module is configured to acquire meteorological parameters, historical load curves and new energy output data, and perform multi-source heterogeneous fusion processing on the meteorological parameters, historical load curves and new energy output data to generate a dynamic load prediction graph; A network modeling and power flow analysis module is configured to analyze the topology structure of a power grid, build a dynamic network model, and perform power flow distribution simulation based on the dynamic network model to obtain stable margin calculation results and a preset safety threshold boundary for each node; A stage decomposition optimization module is configured to perform stage decomposition processing on a global scheduling target based on the dynamic load prediction graph and the stable margin calculation results to generate a gradual scheduling stage sequence; The global scheduling target is stage-decomposed and processed according to the dynamic load prediction map and the stable margin calculation result to generate a progressive scheduling stage sequence, including: According to the load gradient distribution characteristics of the dynamic load prediction map and the stable margin calculation result, a bottleneck area of a global transmission channel is identified and processed to generate a multi-dimensional risk heat map using the following formula: ； ； wherein, denotes the load gradient indicator, n denotes the total number of nodes, denotes the active power of node i, denotes the voltage amplitude of node i, denotes the voltage deviation amount, denotes the stability margin indicator, denotes the maximum eigenvalue, denotes the minimum eigenvalue, m denotes the total number of eigenvalues, denotes the kth eigenvalue; Based on the time-space evolution trend of the multi-dimensional risk heat map, the global scheduling target is subjected to load transfer priority sorting and time period window segmentation processing to generate the progressive scheduling stage sequence; The bottleneck area of the global transmission channel is identified and processed according to the load gradient distribution characteristics of the dynamic load prediction map and the stable margin calculation result to generate a multi-dimensional risk heat map, including: According to the load gradient distribution characteristics of the dynamic load prediction map, the trend of the global load change direction and rate is analyzed and processed to generate a gradient evolution path; Based on the stable margin calculation result and the gradient evolution path, a node bearing capacity attenuation area of the transmission channel is located and processed to generate an initial bottleneck node set; According to the topological connection relationship of the initial bottleneck node set, the power coupling strength of adjacent transmission channels is associated and analyzed to generate a risk diffusion path using the following formula: ； ； wherein, represents the power coupling strength between node i and node j, represents the coupling coefficient, represents the neighbor node set of node i, represents the physical distance between node i and node b, represents the standard deviation parameter of the Gaussian kernel function, represents the risk propagation strength of node i, represents the risk propagation coefficient, represents the weight coefficient between node i and node j, represents the signal propagation time, represents the time decay constant, h represents the total number of nodes in the network; The gradient evolution path, the initial bottleneck node set and the risk diffusion path are fused to generate the multi-dimensional risk heat map including risk level, propagation direction and emergency level label; A response characteristic matching module is configured to match the response characteristics of the power generation equipment, generate an adaptive progressive scheduling instruction sequence based on the progressive scheduling stage sequence; A dynamic feedback execution module is configured to execute the adaptive progressive scheduling instruction sequence and perform real-time feedback processing on the power grid state to generate a dynamic adjustment instruction. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the power grid load dynamic prediction and optimization scheduling method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the power grid load dynamic prediction and optimization scheduling method in any one of claims 1 to 5.

Images