Wind and light storage and charging micro-grid distribution optimization method and system

Abstract

The invention discloses a wind and light storage and charging micro-grid distribution optimization method and system. The method comprises the following steps: acquiring parameter data of each node in a micro-grid in real time; constructing a load demand prediction model, and automatically adjusting the scheduling weight coefficient of the wind and light storage and charging equipment according to the weather type; based on a model output result, establishing a matching relationship between the line impedance and the load emergency degree, generating a power supply path set, and creating a power path selection mechanism based on a dynamic planning algorithm; calculating real-time charge-discharge efficiency based on the equivalent resistance parameter of the power supply path, and dynamically adjusting a charge-discharge rate threshold according to a prediction error; acquiring a real-time monitoring load curve, calculating the similarity between the load curve and historical data, and matching and identifying a potential sudden change risk; and according to a distributed negotiation protocol, each node achieves a scheduling consensus through a priority broadcasting mechanism. The method has the advantages that through load prediction and dynamic scheduling, optimal scheduling, power fluctuation stabilization and sudden change risk identification of the wind and light storage and charging equipment are realized.

Description

Technical Field

[0001] The present invention relates to power grid optimization technology, and particularly to a method and system for optimizing the distribution of a microgrid with wind, light, storage, and charging. Background Art

[0002] With the rapid development of renewable energy, the traditional power system is facing increasing pressure. As a new type of power system, the wind-light-storage-charging microgrid can effectively address the challenges of the volatility and intermittency of renewable energy. The method for optimizing the distribution of the wind-light-storage-charging microgrid is based on the comprehensive application of wind energy, solar energy, energy storage systems, and charging facilities. Through reasonable layout and optimized scheduling, the energy utilization efficiency and power supply reliability of the microgrid are improved.

[0003] The current methods for optimizing the distribution of the wind-light-storage-charging microgrid on the market fail to fully utilize the advantages of real-time data acquisition and intelligent prediction models, resulting in low response speed and scheduling efficiency of the system when facing complex weather changes and sudden load demands. Traditional methods usually rely on fixed scheduling strategies and lack the ability of dynamic adjustment, leading to overload or energy waste in the microgrid when the wind and light power generation fluctuates greatly or the load demand changes violently. In addition, some existing systems have weak capabilities in dealing with power fluctuations and grid stability and cannot achieve refined power regulation through virtual impedance and other means, thus affecting the overall stability and reliability of the system. In terms of risk identification, most methods also only rely on historical data for analysis and lack the dynamic matching of the similarity between the real-time monitored load curve and historical data, so they cannot accurately identify mutation risks and give early warnings. Summary of the Invention

[0004] In order to improve the existing methods and systems for optimizing the distribution of the wind-light-storage-charging microgrid, a method and system for optimizing the distribution of the wind-light-storage-charging microgrid are provided. Through real-time data acquisition, load prediction, and dynamic scheduling, the method realizes the optimized scheduling of wind-light-storage-charging devices, the suppression of power fluctuations, and the identification of mutation risks, effectively improving the operation efficiency, stability, and safety of the microgrid.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for optimizing the distribution of a wind-light-storage-charging microgrid includes:

[0007] Real-time collecting the wind and light power generation data, energy storage state data, charging load demand data, and grid interaction constraint parameters of each node in the microgrid;

[0008] Based on the collected data, decomposing the load demand into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension, constructing a load demand prediction model, and automatically adjusting the scheduling weight coefficients of the wind-light-storage-charging devices according to the weather type;

[0009] Based on the output results of the load demand prediction model, establish the matching relationship between line impedance and load urgency, generate a set of power supply paths, and create a power path selection mechanism based on the dynamic programming algorithm;

[0010] Calculate the real-time charge and discharge efficiency based on the equivalent resistance parameters of the power supply path, dynamically adjust the charge and discharge rate threshold according to the prediction error, and suppress power fluctuations by adjusting the virtual impedance;

[0011] Based on the real-time monitored power grid load data, obtain the real-time monitored load curve, calculate the similarity between the load curve and historical data, and match and identify potential mutation risks;

[0012] Based on the obtained power supply paths and load demands, according to the distributed negotiation protocol, each node reaches a scheduling consensus through the priority broadcast mechanism.

[0013] Preferably, the method of decomposing the load demand into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension based on the collected data, constructing a load demand prediction model, and automatically adjusting the scheduling weight coefficient of the wind-solar-storage-charging equipment according to the weather type specifically includes:

[0014] Obtain the type characteristics of the charging load, divide the priority according to the tolerance time of load power supply interruption, and obtain the priority sequence in the time dimension;

[0015] Based on the microgrid node topology map, use the Dijkstra algorithm to calculate the shortest electrical distance from each load point to the distributed power source, start energy storage compensation for the edge load, and obtain the power supply radius in the space dimension;

[0016] Based on the real-time obtained meteorological monitoring data, divide the weather type, preset the basic weight coefficient according to the equipment type, and dynamically adjust based on the weather type;

[0017] Prevent the sudden change of the weight through a first-order inertial link, trigger the emergency mode under extreme weather, and cancel the smooth transition and directly switch to the preset safety weight;

[0018] Based on the collected data, two-dimensional load demand data and dynamic weight coefficients, train through the LSTM neural network model to obtain the trained load demand prediction model.

[0019] Preferably, the method of establishing the matching relationship between line impedance and load urgency based on the output results of the load demand prediction model, generating a set of power supply paths, and creating a power path selection mechanism based on the dynamic programming algorithm specifically includes:

[0020] Based on the output results of the load demand prediction model, collect the parameters of each branch of the microgrid, construct a line impedance matrix, and calculate the equivalent impedance covering all possible power supply paths;

[0021] Quantify the load urgency index through comprehensive scoring of urgency, and perform urgency normalization processing;

[0022] Abstract the microgrid as a weighted directed graph, with grid nodes as the vertices of the directed graph and electrical connection lines as the edges of the directed graph. Generate an initial path through the Prim algorithm and output a set of candidate power supply paths;

[0023] Define state variables based on the current node position and the real-time state of charge of the energy storage, and design a power supply recurrence equation;

[0024] By initializing the boundary conditions, perform backward iterative solution to obtain the optimal path and generate the globally optimal power supply path.

[0025] Preferably, calculating the real-time charge and discharge efficiency based on the equivalent resistance parameter of the power supply path, dynamically adjusting the charge and discharge rate threshold according to the prediction error, and suppressing power fluctuations by adjusting the virtual impedance specifically includes:

[0026] Based on the obtained globally optimal power supply path, extract the resistance of the line, the transformer loss resistance and the contact resistance, and calculate the total equivalent resistance;

[0027] Calculate the real-time charge and discharge efficiency by defining the comprehensive efficiency of the energy storage system, and update the efficiency value through synchronized phasor measurement;

[0028] Based on the prediction error value between the real-time load demand and the predicted load demand, set an error adaptive adjustment rule, control the change rate of the charge and discharge rate through a ramp function, and constrain the upper limit of the charge rate and the lower limit of the discharge rate;

[0029] Select the values of virtual resistance, inductance and capacitance based on the equivalent resistance of the power supply path, and adjust the virtual impedance through the feedback state data to respond to power fluctuations.

[0030] Preferably, based on the real-time monitored grid load data, obtain the real-time monitored load curve, calculate the similarity between the load curve and the historical data, and match and identify potential mutation risks specifically includes:

[0031] Based on the real-time monitored grid load data, extract key time-frequency domain features and construct a feature vector set;

[0032] Divide the mutation load patterns in the historical data and add risk labels;

[0033] Calculate the similarity between the real-time grid load data and the historical data through dynamic time warping similarity to obtain the mutation probability distribution;

[0034] Perform hierarchical early warning based on the type of mutation.

[0035] Preferably, based on the obtained power supply path and load demand, and according to the distributed negotiation protocol, each node reaches a scheduling consensus through the priority broadcast mechanism, which specifically includes:

[0036] Based on the obtained power supply path and load demand, generate scheduling data for each node;

[0037] Set dynamic weights based on energy abundance and historical packet loss rate, and obtain node priorities;

[0038] Based on the obtained node priorities, eliminate conflicts through an iterative negotiation process, and broadcast in descending order.

[0039] Furthermore, a distributed optimization system for a wind-solar-storage-charging microgrid is proposed, including:

[0040] Data acquisition module: The data acquisition module is mainly used to collect the wind and solar power generation power, energy storage status, charging load demand data and grid interaction constraint parameters of each node in the microgrid in real time, providing basic data support;

[0041] Load demand prediction module: The load demand prediction module is mainly used to decompose the load demand based on the collected data, construct a prediction model in the time and space dimensions, and dynamically adjust the device scheduling weight according to the weather type;

[0042] Path selection module: The path selection module is mainly used to calculate the equivalent resistance parameters based on the power supply path, adjust the charge and discharge efficiency in real time, and suppress power fluctuations through virtual impedance regulation;

[0043] Risk identification module: By real-time monitoring the similarity between the load curve and historical data, match and identify potential mutation risks, and conduct hierarchical early warnings of mutation types;

[0044] Scheduling consensus module: The scheduling consensus module is mainly used to reach a scheduling consensus through the distributed negotiation protocol, and each node reaches a scheduling consensus through the priority broadcast mechanism to ensure the coordination and efficiency of microgrid scheduling;

[0045] Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

[0046] Compared with the prior art, the advantages of the present invention are:

[0047] By collecting data such as wind and solar power generation, energy storage status, and load demand in real time, and combining dynamic prediction models and adjustments based on weather types, accurate load demand prediction and equipment scheduling can be achieved, ensuring the optimal utilization of wind-solar-storage-charging equipment under different weather conditions. At the same time, based on the dynamic programming algorithm and path selection mechanism, the optimal power supply path can be calculated, and flexible charge and discharge management can be carried out according to the equivalent resistance of the power grid and the load urgency, effectively avoiding power fluctuations. In addition, by using the similarity analysis of real-time load curves and historical data, potential mutation risks can be accurately identified, timely warnings can be given, and the safety and stability of the system can be improved. Finally, based on the distributed negotiation protocol, each node reaches a scheduling consensus through the priority broadcast mechanism, ensuring the coordinated operation of each node in the microgrid and greatly improving the scheduling efficiency and fault tolerance of the system. Brief Description of the Drawings

[0048] Figure 1 Schematic diagram of the method proposed by the present invention;

[0049] Figure 2 Schematic diagram of the construction of the load demand prediction model proposed by the present invention;

[0050] Figure 3 Schematic diagram of path selection proposed by the present invention;

[0051] Figure 4 Schematic diagram of the adjustment of charge and discharge rates proposed by the present invention;

[0052] Figure 5 Schematic diagram of the identification of potential mutation risks proposed by the present invention;

[0053] Figure 6 Schematic diagram of the scheduling consensus proposed by the present invention;

[0054] Figure 7 Architecture diagram of the electronic device in this solution;

[0055] Figure 8 Schematic diagram of the structure of the computer-readable storage medium in this solution. Detailed Embodiment

[0056] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0057] A distributed optimization system for a wind-solar-storage-charging microgrid includes:

[0058] Data acquisition module: The data acquisition module is mainly used to collect in real time the wind and solar power generation power, energy storage status, charging load demand data and grid interaction constraint parameters of each node in the microgrid, and provide basic data support;

[0059] Load demand forecasting module: The load demand forecasting module is mainly used to collect data, decompose the load demand, construct a forecasting model in the time and space dimensions, and dynamically adjust the device scheduling weight according to the weather type;

[0060] Path selection module: The path selection module is mainly used to calculate the equivalent resistance parameters based on the power supply path, adjust the charge and discharge efficiency in real time, and suppress the power fluctuation through virtual impedance regulation;

[0061] Risk identification module: By monitoring the similarity between the real-time load curve and historical data in real time, match and identify potential mutation risks, and conduct hierarchical early warnings of mutation types;

[0062] Scheduling consensus module: The scheduling consensus module is mainly used to achieve scheduling consensus through a distributed negotiation protocol, and each node reaches a scheduling consensus through a priority broadcast mechanism to ensure the coordination and efficiency of microgrid scheduling;

[0063] Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

[0064] Refer to Figure 1 As shown in

[0065] Step 1: Collect the wind and solar power generation power data, energy storage state data, charging load demand data and grid interaction constraint parameters of each node in the microgrid in real time;

[0066] Step 2: Based on the collected data, decompose the load demand into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension, construct a load demand forecasting model, and automatically adjust the scheduling weight coefficient of the wind-solar-storage-charging equipment according to the weather type;

[0067] Step 3: Based on the output results of the load demand forecasting model, establish a matching relationship between the line impedance and the load urgency, generate a set of power supply paths, and create a power path selection mechanism based on the dynamic programming algorithm;

[0068] Step 4: Calculate the real-time charge and discharge efficiency based on the equivalent resistance parameters of the power supply path, dynamically adjust the charge and discharge rate threshold according to the prediction error, and suppress the power fluctuation by adjusting the virtual impedance;

[0069] Step 5: Based on the real-time monitored grid load data, obtain the real-time monitored load curve, calculate the similarity between the load curve and historical data, and match and identify potential mutation risks;

[0070] Step 6: Based on the obtained power supply path and load demand, each node reaches a scheduling consensus through a priority broadcast mechanism according to the distributed negotiation protocol.

[0071] Refer to Figure 2 As shown, based on the collected data, the load demand is decomposed into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension, a load demand prediction model is constructed, and the scheduling weight coefficients of the wind-solar-storage-charging equipment are automatically adjusted according to the weather type. Specifically, it includes:

[0072] Obtain the type characteristics of the charging load, divide the priority according to the tolerance time of the load power supply interruption, and obtain the priority sequence in the time dimension;

[0073] Based on the microgrid node topology map, use the Dijkstra algorithm to calculate the shortest electrical distance from each load point to the distributed power source, start energy storage compensation for the edge load, and obtain the power supply radius in the space dimension;

[0074] Based on the real-time obtained meteorological monitoring data, divide the weather type, preset the basic weight coefficient according to the equipment type, and dynamically adjust based on the weather type;

[0075] Prevent weight mutation through a first-order inertia link, trigger the emergency mode under extreme weather, cancel the smooth transition and directly switch to the preset safety weight;

[0076] Based on the collected data, two-dimensional load demand data and dynamic weight coefficients, train through the LSTM neural network model to obtain the trained load demand prediction model.

[0077] Specifically, according to the tolerance time of the load power supply interruption, it is divided into three levels:

[0078] Emergency type: The tolerance time is less than 15 minutes, including the power supply of first-aid facilities and the instantaneous charging demand of electric vehicles;

[0079] Regular type: The tolerance time is between 15 minutes and two hours, covering the basic electricity consumption of residents and the continuous production load of industrial and commercial enterprises;

[0080] Interruptible type: The tolerance time is greater than two hours, such as the charging of energy storage systems and adjustable air-conditioning loads;

[0081] The topology map of the microgrid can be represented in a graph, where each node represents a load point or a power source point, the edge represents an electrical connection, calculate the shortest path from the load point to the power source point through the Dijkstra algorithm, and calculate the electrical distance of each load point according to the shortest path. If the distance of the load point is far, the energy storage system can be enabled for compensation;

[0082] Divide the weather type according to the meteorological monitoring data. Common weather types include sunny, cloudy, rainy, and stormy. Preset the basic weight coefficient according to the equipment type. For example, fast charging stations and ordinary charging stations may have different sensitivities to weather conditions. Adjust the weight coefficient based on the real-time weather data;

[0083] To avoid drastic changes in weights during sudden weather changes, a first-order inertial link is used to smooth the weight changes, and the formula is:

[0084]

[0085] where ω i (t) is the current value of the i-th weight at time t, and ω i (t + 1) is the updated weight value at time t + 1. α is the smoothing coefficient (0 ≤ α < 1), which controls the retention ratio of historical weights, is the weight adjustment amount calculated according to new data;

[0086] When the weather conditions become extreme, the emergency mode is automatically triggered, the smooth transition is cancelled, and the weight is directly set to the preset safety value;

[0087] The LSTM (Long Short-Term Memory Network) is used for load demand prediction to accurately predict future loads and optimize power dispatching.

[0088] Refer to Figure 3 As shown, based on the output results of the load demand prediction model, a matching relationship between line impedance and load urgency is established, a set of power supply paths is generated, and the creation of a power path selection mechanism based on the dynamic programming algorithm specifically includes:

[0089] Based on the output results of the load demand prediction model, the parameters of each branch of the microgrid are collected, a line impedance matrix is constructed, and the equivalent impedance covering all possible power supply paths is calculated;

[0090] The load urgency index is quantified through a comprehensive urgency score, and urgency normalization is performed;

[0091] The microgrid is abstracted as a weighted directed graph, the grid nodes are used as the vertices of the directed graph, and the electrical connection lines are used as the edges of the directed graph. The initial path is generated through the Prim algorithm, and a set of candidate power supply paths is output;

[0092] Based on the current node position and the real-time state of charge of the energy storage, state variables are defined, and a power supply recurrence equation is designed;

[0093] By initializing the boundary conditions, reverse iteration is performed to solve the optimal path, and the global optimal power supply path is obtained.

[0094] Specifically, the electrical parameters of each branch of the microgrid are obtained, especially the impedance of the line. Let the impedance of the i-th line be Z i , then the elements of the line impedance matrix Z are the impedances between each pair of adjacent nodes;

[0095] The path impedance is calculated through series and parallel rules;

[0096] Calculate the urgency of each load point based on the load type, tolerance time, and other relevant factors, and normalize the urgency values of all load points so that their values are between [0, 1];

[0097] Abstract the microgrid topology into a graph, where nodes represent each load point and power source point in the power grid, edges represent electrical connection lines, and the weights of the edges are determined by the line impedance. Apply the Prim algorithm, starting from the high-urgency load nodes, preferentially select the edge with the smallest weighted distance, and iteratively expand the spanning tree. The constraint conditions include the node voltage deviation ΔV ≤ ±10%, and the branch load rate ≤ 85%. Output the candidate power supply path set and its characteristic parameters;

[0098] Solve the optimal power supply path through the reverse iteration method to ensure the maximization of the power supply efficiency of the microgrid system. Set the initial boundary conditions, that is, the energy storage state and load demand at the initial moment. Starting from the target node, calculate the optimal power supply path by reverse iteration. Based on the state equation and dynamic programming method, update the optimal path and power supply strategy of each node through iteration. The final optimal path solution formula is:

[0099]

[0100] Among them, P opt (t) is the optimal power supply value at time t, N is the total number of nodes in the power supply network, min p is to take the minimum value of the power supply path p, Z i is the weight coefficient of node i, P i is the power supply value passing through node i.

[0101] Refer to Figure 4 As shown, calculate the real-time charge and discharge efficiency based on the equivalent resistance parameters of the power supply path, dynamically adjust the charge and discharge rate threshold according to the prediction error, and suppress the power fluctuation by adjusting the virtual impedance. Specifically include:

[0102] Based on the obtained global optimal power supply path, extract the resistance of the line, the transformer loss resistance, and the contact resistance, and calculate the total equivalent resistance;

[0103] Calculate the real-time charge and discharge efficiency by defining the comprehensive efficiency of the energy storage system, and update the efficiency value through synchronous phasor measurement;

[0104] Based on the prediction error value between the real-time load demand and the predicted load demand, set the error adaptive adjustment rule, control the change rate of the charge and discharge rate through the ramp function, and constrain the upper limit of the charge rate and the lower limit of the discharge rate;

[0105] Select virtual resistance, inductance, and capacitance values based on the equivalent resistance of the power supply path, and adjust the virtual impedance through the feedback state data to respond to power fluctuations.

[0106] Specifically, the line resistance is the main impedance during power transmission, the transformer loss resistance represents the losses generated by the transformer during power transmission, and the contact resistance is the resistance at the connection between the power equipment and the power line components. It is usually small, but it also needs to be considered for high-precision calculations. Assuming that there are multiple resistance values to be considered in the microgrid, the total equivalent resistance can be calculated by series, parallel, etc. According to the actual grid topology and physical model, select the appropriate calculation method;

[0107] The efficiency of the energy storage system affects the overall operation performance of the microgrid. By measuring the charge and discharge power of the energy storage system in real time and combining with the synchronized phasor measurement technology, the energy storage efficiency can be updated in real time;

[0108] The synchronized phasor measurement technology is used to measure the current and voltage phasors in the power grid in real time. Through the synchronized phasors, the real-time charge and discharge state and electrical parameters of the energy storage system can be obtained. The charge and discharge power of the energy storage can be calculated by the following formula:

[0109] P(t) = V(t)·I(t)·cos(φ)

[0110] where P(t) is the charge and discharge power of the energy storage system at time t, V(t) is the effective voltage value at time t, I(t) is the effective current value at time t, cos(φ) is the power factor, and it is the cosine value of the phase difference φ between the voltage and the current;

[0111] The energy storage efficiency can be adjusted according to real-time data. For example, if there are fluctuations in the efficiency during the charging and discharging processes, it can be corrected based on historical data;

[0112] The load prediction error will affect the charge and discharge rate of the energy storage system. Set an adaptive adjustment rule, and the change rate of the charge and discharge rate is adjusted through a ramp function. The formula is:

[0113]

[0114] where ΔP storage (t) is the change rate of the charge and discharge rate, k is the control factor of the ramp function to adjust the change of the charge and discharge rate, and ΔP(t) is the load prediction error;

[0115] To ensure the stability of the microgrid, upper and lower limits need to be set for the charging rate and the discharging rate, and these rate limits can be dynamically adjusted by real-time regulating the load demand error;

[0116] Based on real-time grid status data (such as voltage, frequency, etc.), the virtual impedance is adjusted through feedback control. The adjustment of the virtual impedance helps to smooth power fluctuations. By dynamically adjusting these parameters, the response ability of the microgrid can be optimized, thereby reducing the power supply instability caused by power fluctuations.

[0117] See Figure 5 As shown, based on the real-time monitored grid load data, obtain the real-time monitored load curve, calculate the similarity between the load curve and the historical data, and the specific steps for matching and identifying potential mutation risks include:

[0118] Based on the real-time monitored grid load data, extract key features in the time-frequency domain and construct a feature vector set;

[0119] Divide the mutation load patterns in the historical data and add risk labels;

[0120] Calculate the similarity between the real-time grid load data and the historical data through dynamic time warping similarity to obtain the mutation probability distribution;

[0121] Carry out hierarchical early warning based on the type of mutation.

[0122] Specifically, analyze the grid load data through short-time Fourier transform, divide the data signal into multiple short time periods, and calculate the Fourier transform of each short time period;

[0123] Extract key information from the results of time-frequency analysis and construct a feature vector set. The following features can be extracted:

[0124] Spectrum features: such as power spectral density, center frequency, bandwidth;

[0125] Time-domain features: such as the mean, variance, kurtosis, and skewness of the load;

[0126] Time-frequency joint features: combinations of frequency-domain and time-domain features, for example, the instantaneous change of frequency-domain features;

[0127] According to the historical load data, identify the mutation patterns in it, cluster the load data through a clustering algorithm, identify different load patterns, and assign risk labels to each pattern according to the characteristics of the mutation load pattern, including:

[0128] Minor risk: When the load mutation amplitude is small and the duration is short, it is defined as a minor risk;

[0129] Medium risk: When the load mutation amplitude is large, it is defined as a medium risk;

[0130] High risk: When the load mutation amplitude is very large, it is defined as a high risk;

[0131] Calculate the similarity between real-time network load data and historical data using the dynamic time warping algorithm, so as to estimate the probability distribution of mutations. By calculating the DTW distance between real-time load data and historical data, a similarity metric is obtained. The higher the similarity, the lower the probability of mutation occurrence;

[0132] Implement hierarchical early warning according to the type and probability of mutation, including:

[0133] Low early warning: When the mutation probability is lower than a certain threshold (for example, the mutation probability is less than 0.2), it is considered that the possibility of mutation occurrence is small and belongs to low risk;

[0134] Medium early warning: When 0.2 ≤ mutation probability < 0.5, it is considered that the possibility of mutation occurrence is medium and belongs to medium risk;

[0135] High early warning: When the mutation probability ≥ 0.5, it is considered that the possibility of mutation occurrence is large and belongs to high risk;

[0136] Adopt different response strategies according to the early warning level. For example, in the case of high risk, the energy storage system can be started or other measures can be taken to stabilize the grid load.

[0137] Refer to Figure 6 As shown, based on the obtained power supply path and load demand, according to the distributed negotiation protocol, each node reaches a scheduling consensus through the priority broadcast mechanism, which specifically includes:

[0138] Generate scheduling data for each node based on the obtained power supply path and load demand;

[0139] Set dynamic weights based on energy abundance and historical packet loss rate to obtain node priorities;

[0140] Based on the obtained node priorities, eliminate conflicts through an iterative negotiation process and broadcast in descending order.

[0141] Furthermore, the method according to the embodiment of the present application can also be implemented with the help of Figure 7 The architecture of the electronic device shown. As Figure 7 shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to the network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, can store a microgrid distribution optimization method and system for wind-solar-storage-charging provided by the present application. The electronic device 500 may also include a terminal interface 508. Of course, Figure 7 The architecture shown is only exemplary. When implementing different devices, according to actual needs, it can be omitted Figure 7One or more components in the illustrated electronic device.

[0142] Figure 8 It is a schematic structural diagram of a computer-readable storage medium provided by an embodiment of the present application. As Figure 8 shown, it is a computer-readable storage medium 600 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are run by a processor, a microgrid distribution optimization method and system for wind-solar-storage-charging according to the embodiments of the present application described with reference to the above drawings can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0143] It should be noted that: the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0144] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

[0145] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A microgrid distribution optimization method for wind-solar-storage-charging, characterized in that Including: Real-time collect the wind and solar power generation data, energy storage state data, charging load demand data, and grid interaction constraint parameters of each node in the microgrid; Based on the collected data, decompose the load demand into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension, construct a load demand prediction model, and automatically adjust the scheduling weight coefficients of the wind-solar-storage-charging equipment according to the weather type; Based on the output results of the load demand prediction model, establish a matching relationship between the line impedance and the load urgency, generate a set of power supply paths, and create a power path selection mechanism based on the dynamic programming algorithm; Calculate the real-time charge and discharge efficiency based on the equivalent resistance parameters of the power supply path, dynamically adjust the charge and discharge rate threshold according to the prediction error, and suppress the power fluctuation by adjusting the virtual impedance; Based on the real-time monitored grid load data, obtain the real-time monitored load curve, calculate the similarity between the load curve and the historical data, and match and identify potential mutation risks; Based on the obtained power supply path and load demand, according to the distributed negotiation protocol, each node reaches a scheduling consensus through the priority broadcast mechanism.

2. The microgrid distribution optimization method for wind-solar-storage-charging according to claim 1, wherein, The specific steps of decomposing the load demand into a priority sequence in the time dimension and a power supply radius topological structure in the space dimension based on the collected data, constructing a load demand prediction model, and automatically adjusting the scheduling weight coefficients of the wind-solar-storage-charging equipment according to the weather type include: Obtain the type characteristics of the charging load, divide the priority according to the tolerance time of the load power supply interruption, and obtain the priority sequence in the time dimension; Based on the microgrid node topology map, use the Dijkstra algorithm to calculate the shortest electrical distance from each load point to the distributed power source, start the energy storage compensation for the edge load, and obtain the power supply radius in the space dimension; Based on the real-time obtained meteorological monitoring data, divide the weather type, preset the basic weight coefficient according to the equipment type, and dynamically adjust based on the weather type; Prevent the weight mutation through a first-order inertia link, trigger the emergency mode under extreme weather, cancel the smooth transition and directly switch to the preset safety weight; Based on the collected data, two-dimensional load demand data, and dynamic weight coefficients, train through the LSTM neural network model to obtain the trained load demand prediction model.

3. A method for optimizing the distribution of a microgrid with wind, light, energy storage and charging, according to claim 1, characterized in that, The specific steps of establishing a matching relationship between the line impedance and the load urgency based on the output results of the load demand prediction model, generating a set of power supply paths, and creating a power path selection mechanism based on the dynamic programming algorithm include: Based on the output results of the load demand prediction model, collect the parameters of each branch of the microgrid, construct a line impedance matrix, and calculate the equivalent impedance covering all possible power supply paths; Quantify the load urgency index through the comprehensive urgency score and perform the normalization processing of the urgency; Abstract the microgrid as a weighted directed graph, with the grid nodes as the vertices of the directed graph and the electrical connection lines as the edges of the directed graph. Generate the initial path through the Prim algorithm and output the candidate power supply path set; Define the state variable based on the current node position and the real-time state of charge of the energy storage, and design the power supply recurrence equation; Through initializing the boundary conditions, perform backward iterative solution to obtain the optimal path and generate the globally optimal power supply path.

4. A method for optimizing the distribution of a microgrid with wind, light, energy storage and charging, according to claim 1, characterized in that, Calculating the real-time charge and discharge efficiency based on the equivalent resistance parameter of the power supply path, dynamically adjusting the charge and discharge rate threshold according to the prediction error, and suppressing power fluctuations by adjusting the virtual impedance specifically includes: Based on the obtained globally optimal power supply path, extracting the resistance of the line, the loss resistance of the transformer, and the contact resistance, and calculating the total equivalent resistance; Obtaining the real-time charge and discharge efficiency through defining the comprehensive efficiency of the energy storage system and updating the efficiency value through synchronized phasor measurement; Based on the prediction error value between the real-time load demand and the predicted load demand, setting an error adaptive adjustment rule, controlling the change rate of the charge and discharge rate through a ramp function, and constraining the upper limit of the charge rate and the lower limit of the discharge rate; Selecting the values of virtual resistance, inductance, and capacitance based on the equivalent resistance of the power supply path, and adjusting the virtual impedance through the feedback state data to respond to power fluctuations.

5. A method for optimizing the distribution of a microgrid with wind, light, storage and charging, according to claim 1, characterized in that The method for obtaining the real-time monitored load curve based on the real-time monitored power grid load data, calculating the similarity between the load curve and the historical data, and matching and identifying potential mutation risks specifically includes: Based on the real-time monitored power grid load data, extracting key time-frequency domain features and constructing a feature vector set; Dividing the mutation load patterns in the historical data and adding risk labels; Calculating the similarity between the real-time grid load data and the historical data through dynamic time warping similarity to obtain the mutation probability distribution; Conducting hierarchical early warning based on the type of mutation.

6. The microgrid distribution optimization method for wind-solar-storage-charging according to claim 1, characterized in that, The method for each node to reach a scheduling consensus through the priority broadcast mechanism according to the distributed negotiation protocol based on the obtained power supply path and load demand specifically includes: Generating the scheduling data of each node based on the obtained power supply path and load demand; Setting dynamic weights based on the energy abundance and historical packet loss rate to obtain the node priorities; Based on the obtained node priorities, eliminating conflicts through an iterative negotiation process and broadcasting in descending order.

7. A microgrid distribution optimization system integrating wind power, photovoltaic power, energy storage and charging is used to implement a microgrid distribution optimization method according to any one of claims 1-6, characterized in that Including: Data acquisition module: The data acquisition module is mainly used to collect the wind and solar power generation power, energy storage status, charging load demand data, and grid interaction constraint parameters of each node in the microgrid in real time, providing basic data support; Load demand prediction module: The load demand prediction module is mainly used to decompose the load demand for the collected data, construct a prediction model in the time and space dimensions, and dynamically adjust the device scheduling weights according to the weather type; Path selection module: The path selection module is mainly used to calculate the equivalent resistance parameter based on the power supply path, adjust the charge and discharge efficiency in real time, and suppress power fluctuations through virtual impedance adjustment; Risk identification module: Matching and identifying potential mutation risks by the similarity between the real-time monitored load curve and the historical data, and conducting hierarchical early warning of the mutation type; Scheduling consensus module: The scheduling consensus module is mainly used to enable each node to reach a scheduling consensus through the priority broadcast mechanism according to the distributed negotiation protocol, ensuring the coordination and efficiency of the microgrid scheduling; Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

8. An electronic device, characterized in that, Including: At least one processor; And a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute a method for optimizing the distribution of a microgrid with wind power, photovoltaics, energy storage, and charging as described in any one of claims 1-6.

9. A computer-readable storage medium storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by a processor, a method for optimizing the distribution of a microgrid with wind power, photovoltaics, energy storage, and charging as described in any one of claims 1-6 is implemented.

Images