Wind and light energy storage method applied to micro-grid

By collecting data in real time in the microgrid and using machine learning algorithms to build a power demand prediction model, and formulating a scheduling strategy in combination with particle swarm optimization algorithms, the problems of low power demand prediction accuracy and insufficient dynamic scheduling strategy in the existing technology are solved, and the problem of low power demand prediction accuracy and insufficient dynamic scheduling strategy in the existing technology is improved. Energy utilization efficiency and system stability.

CN119051032APending Publication Date: 2024-11-29国网内蒙古东部电力有限公司呼伦贝尔供电公司
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410933901.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The power demand forecast model in the microgrid relies on historical data and simple statistical methods, and fails to make full use of real-time data and environmental variables, resulting in low accuracy of prediction results, lack of dynamicity in scheduling strategies, and unable to effectively respond to rapidly changing power demand. .

Method used

Through sensors and intelligent devices, data from distributed photovoltaics, distributed wind power and energy storage systems are collected in real time, and a machine learning algorithm is used to build a short-term power demand prediction model, and a dynamic energy scheduling strategy is formulated in combination with improved particle swarm optimization algorithms to optimize energy allocation.

Benefits of technology

It improves the accuracy of power demand forecasting and the dynamic nature of scheduling strategies, enhances energy utilization efficiency and microgrid operation stability, and can more effectively deal with load fluctuations and environmental changes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051032A_ABST
    Figure CN119051032A_ABST
Patent Text Reader

Abstract

The invention discloses a wind and light energy storage method applied to a micro-grid, and relates to the technical field of wind and light energy storage applied to the micro-grid, and the method comprises the steps: collecting the state data of a distributed photovoltaic system, a distributed wind power system and an energy storage system in real time through a sensor and an intelligent device; constructing a short-term power demand prediction model through a machine learning algorithm, and predicting the power demand in a certain time in the future; and according to the real-time data and the prediction result, an energy scheduling strategy is formulated, and the energy distribution and scheduling strategy is dynamically adjusted. According to the method, the state data of the distributed photovoltaic system, the distributed wind power system and the energy storage system are collected in real time by combining the sensor and the intelligent equipment, and the short-term power demand prediction model is constructed by using the machine learning algorithm, so that the future power demand can be predicted more accurately. And according to the real-time data and the prediction result, a dynamically adjusted energy scheduling strategy is formulated, so that the energy utilization efficiency and the operation stability of the micro-grid are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of wind-solar energy storage in microgrids, and in particular to a wind-solar energy storage method applied to microgrids. Background Art

[0002] With the rapid development of renewable energy, photovoltaic power generation and wind power generation, as two major clean energy sources, have gradually been widely used around the world. Due to the intermittent and volatile nature of these energy sources, it is difficult to adapt to the stable operation of the power grid. In order to effectively utilize renewable energy and ensure the stability of the power grid, energy storage technology needs to be applied to microgrids. Microgrids are small power systems that integrate distributed generation, energy storage devices, and load management systems, and can achieve self-control and energy management.

[0003] The current microgrid has some shortcomings. Traditional power demand forecasting models mostly rely on historical data and simple statistical methods, ignoring the dynamic changes of real-time data and environmental variables. The prediction results are less accurate and difficult to cope with rapidly changing power demand. In terms of energy scheduling strategies, existing technologies mostly use fixed rules or simple optimization algorithms, and fail to make full use of real-time data and prediction results for dynamic adjustment. Such scheduling strategies show great limitations in dealing with load fluctuations and environmental changes, and often fail to respond to changes in power demand in a timely manner, resulting in unreasonable energy distribution and reducing the overall efficiency of the system. In response to these shortcomings, we have invented a wind and solar energy storage method for microgrids. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems or problems existing in the prior art, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the existing technologies mostly adopt fixed rules or simple optimization algorithms, and fail to fully utilize real-time data and prediction results for dynamic adjustment.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a wind-solar energy storage method applied to a microgrid, which includes collecting status data of distributed photovoltaic, distributed wind power and energy storage systems in real time through sensors and intelligent devices;

[0008] Build a short-term electricity demand forecasting model through machine learning algorithms to predict electricity demand within a certain period of time in the future;

[0009] Based on real-time data and forecast results, energy scheduling strategies are formulated to dynamically adjust energy distribution and scheduling strategies.

[0010] As a preferred solution of the wind-solar energy storage method applied to a microgrid according to the present invention, wherein: the short-term power supply and demand prediction model constructed by a machine learning algorithm includes:

[0011] Obtain historical power demand data, weather forecast information, and social media holiday data and perform pre-processing;

[0012] Use machine learning algorithms for data training and optimization;

[0013] Construct short-term electricity demand and generation forecast models.

[0014] As a preferred solution of the wind-solar energy storage method applied to a microgrid according to the present invention, the construction of a short-term power demand forecasting model includes:

[0015] The input power data is transformed nonlinearly by Gaussian radial basis function, which is expressed as:

[0016]

[0017] in, is the transformed value of the jth feature of the i-th sample, μ j and σ j are the mean and standard deviation of the jth feature respectively;

[0018] The meteorological data is processed by polynomial transformation, which is expressed as:

[0019] ψ k (z i,k )=(a k z i,k +c k ) d

[0020] Among them, ψ k (z i,k ) is the value of the kth meteorological data feature of the ith sample after transformation, a k and c k is the transformation coefficient, d is the order of the polynomial;

[0021] The social media holiday data is processed through the logistic regression function, which is expressed as:

[0022]

[0023] Among them, φ l (w i,l) is the transformed value of the lth social media holiday data feature of the i-th sample, η l is the bias parameter of logistic regression;

[0024] A linear regression model is constructed to capture the basic trends of electricity demand and generation, expressed as:

[0025]

[0026] Among them, β0 is the intercept, β j is the weight of the jth feature, is the power data feature after Gaussian radial basis function transformation;

[0027] The kernel function is used to process high-dimensional data and capture complex feature interactions, which can be expressed as:

[0028]

[0029] Among them, K(x i ,x) is the sample x and the support vector x i The kernel function value between σ and ∥x i -x∥ 2 is the Euclidean distance between samples;

[0030] Combining the kernel function with the linear regression model is expressed as:

[0031]

[0032] Among them, α i is the coefficient of the support vector, b is the bias term;

[0033] Combined with social media holiday data, it is expressed as:

[0034]

[0035] Among them, λ is the adjustment parameter, θ l is the weight of the lth social media holiday data feature, φ l (w i,l ) is the lth social media holiday data feature of the ith sample after logistic regression transformation;

[0036] A complete short-term electricity supply and demand forecasting model is constructed comprehensively, which is expressed as:

[0037]

[0038] in, is the predicted value of power demand at time t, t is the specific time point of acquisition, γ is the adjustment parameter, δ kis the weight of the kth meteorological data feature, ψ k (z i,k ) is the kth meteorological data feature of the ith sample after transformation, N is the number of support vectors, M is the number of features, P is the number of meteorological data features, and Q is the number of social media holiday data features.

[0039] As a preferred solution of the wind-solar energy storage method applied to a microgrid described in the present invention, the dynamic adjustment of energy distribution and scheduling strategy includes:

[0040] According to the real-time prediction results and current load conditions, the energy dispatch strategy is formulated through the improved particle swarm optimization algorithm;

[0041] Dynamically adjust the output power of each energy source through optimization results, and complete real-time adjustment of energy distribution through the energy management system;

[0042] The energy dispatch strategy includes giving priority to energy storage during load trough periods and releasing energy in the energy storage system during load peak periods.

[0043] As a preferred solution of the wind-solar energy storage method applied to a microgrid according to the present invention, the improved particle swarm optimization algorithm includes:

[0044] Initialize the particle swarm optimization algorithm, set the particle position and velocity to random values, and initialize the velocity update formula:

[0045] v i (0) = ωv i (0)+c1r1(p i (0)-x i (0))+c2r2(g(0)-x i (0))

[0046] Among them, v i (0) is the velocity of particle i at the initial moment, ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers ranging from 0 to 1, p i (0) is the individual optimal position of particle i, x i (0) is the position of particle i at the initial moment, and g(0) is the global optimal position;

[0047] The original speed update formula is improved according to the predicted short-term power demand value. The improved speed update formula is expressed as:

[0048]

[0049] Among them, γ is the adjustment factor; v i (t) is the velocity of particle i at time t; ω is the inertia weight; is the predicted short-term power demand value at time t; p i (t) is the individual optimal position of particle i; x i (t) is the position of particle i at time t; g(t) is the global optimal position; v i (t+1) is the velocity of particle i at time t+1;

[0050] The position update formula is improved by smoothing the speed update effect through the Sigmoid function, which is expressed as follows:

[0051]

[0052] Where η is the adjustment factor;

[0053] The objective function is set to minimize the deviation between power generation and predicted demand value, energy loss and smoothing effect of power generation. The formula is as follows:

[0054]

[0055] Among them, P gen (t) is the total power generation at time t, E loss (t) is the energy loss at time t, λ is the penalty factor, μ is the smoothing factor, and T is the total number of time steps within the prediction and scheduling time range.

[0056] As a preferred solution of the wind-solar energy storage method applied to microgrids described in the present invention, the real-time collection includes installing sensors at key nodes of photovoltaic power generation systems, wind power generation systems and energy storage systems; configuring intelligent monitoring equipment to receive and record data information collected by sensors; and the intelligent monitoring equipment transmits data to a central control system through a wireless communication module.

[0057] As a preferred solution of the wind-solar energy storage method applied to microgrids described in the present invention, the use of machine learning algorithms for data training and optimization includes: performing basic predictions through linear regression to capture major trends; using support vector machine kernel functions to process high-dimensional feature interactions to improve prediction accuracy; and dynamically correcting and adjusting prediction results based on meteorological data and social media holiday data.

[0058] Beneficial effects of the present invention: The present invention uses sensors and intelligent devices to collect real-time status data of distributed photovoltaic, distributed wind power and energy storage systems, and uses machine learning algorithms to build a short-term power demand forecasting model, which can more accurately predict future power demand. According to real-time data and forecast results, a dynamically adjusted energy scheduling strategy is formulated to improve energy utilization efficiency and the operating stability of the microgrid. Compared with the existing technology, improvements have been made in the accuracy of data collection, the precision of the forecasting model, and the dynamics of the scheduling strategy, which can more effectively solve the deficiencies in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0060] Figure 1 The figure is an overall flow chart of the wind-solar energy storage method applied to microgrids. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0062] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0063] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0064] Example 1

[0065] Reference Figure 1 , which is an embodiment of the present invention, provides a wind-solar energy storage method applied to a microgrid, comprising:

[0066] S1: Real-time collection of status data of distributed photovoltaic, distributed wind power and energy storage systems through sensors and smart devices.

[0067] Furthermore, the real-time collection includes installing sensors at key nodes of photovoltaic power generation systems, wind power generation systems and energy storage systems; configuring intelligent monitoring equipment to receive and record data information collected by sensors; and the intelligent monitoring equipment transmits data to the central control system through a wireless communication module.

[0068] S2: Build a short-term electricity demand forecasting model through machine learning algorithms to predict electricity demand within a certain period of time in the future.

[0069] Furthermore, the short-term electricity supply and demand forecasting model constructed by machine learning algorithm includes:

[0070] Obtain historical power demand data, weather forecast information, and social media holiday data and perform pre-processing;

[0071] Use machine learning algorithms for data training and optimization;

[0072] Construct short-term electricity demand and generation forecast models.

[0073] Furthermore, the use of machine learning algorithms for data training and optimization includes: performing basic predictions through linear regression to capture major trends; using support vector machine kernel functions to process high-dimensional feature interactions to improve prediction accuracy; and performing dynamic corrections based on meteorological data and social media holiday data to adjust prediction results.

[0074] Furthermore, the construction of the short-term power demand forecasting model includes:

[0075] The input power data is transformed nonlinearly by Gaussian radial basis function, which is expressed as:

[0076]

[0077] in, is the transformed value of the jth feature of the i-th sample, μ j and σ j are the mean and standard deviation of the jth feature respectively;

[0078] The meteorological data is processed by polynomial transformation, which is expressed as:

[0079] ψ k (z i,k )=(a k z i,k +c k ) d

[0080] Among them, ψ k (z i,k ) is the value of the kth meteorological data feature of the ith sample after transformation, ak and c k is the transformation coefficient, d is the order of the polynomial;

[0081] The social media holiday data is processed through the logistic regression function, which is expressed as:

[0082]

[0083] Among them, φ l (w i,l ) is the transformed value of the lth social media holiday data feature of the i-th sample, η l is the bias parameter of logistic regression;

[0084] A linear regression model is constructed to capture the basic trends of electricity demand and generation, expressed as:

[0085]

[0086] Among them, β0 is the intercept, β j is the weight of the jth feature, is the power data feature after Gaussian radial basis function transformation;

[0087] The kernel function is used to process high-dimensional data and capture complex feature interactions, which can be expressed as:

[0088]

[0089] Among them, K(x i ,x) is the sample x and the support vector x i The kernel function value between σ and ∥x i -x∥ 2 is the Euclidean distance between samples;

[0090] Combining the kernel function with the linear regression model is expressed as:

[0091]

[0092] Among them, α i is the coefficient of the support vector, b is the bias term;

[0093] Combined with social media holiday data, it is expressed as:

[0094]

[0095] Among them, λ is the adjustment parameter, θ l is the weight of the lth social media holiday data feature, φ l (w i,l) is the lth social media holiday data feature of the ith sample after logistic regression transformation;

[0096] A complete short-term electricity supply and demand forecasting model is constructed comprehensively, which is expressed as:

[0097]

[0098] in, is the predicted value of power demand at time t, t is the specific time point of acquisition, γ is the adjustment parameter, δ k is the weight of the kth meteorological data feature, ψ k (z i,k ) is the kth meteorological data feature of the ith sample after transformation, N is the number of support vectors, M is the number of features, P is the number of meteorological data features, and Q is the number of social media holiday data features.

[0099] It should be noted that by using sensors and smart devices to collect the status data of distributed photovoltaic, distributed wind power and energy storage systems in real time, and using long short-term memory network (LSTM) for data training, a high-precision short-term electricity demand forecasting model is constructed. The forecasting models in the prior art mostly rely on historical data and simple statistical methods, and fail to make full use of real-time data and environmental variables, resulting in insufficient accuracy and dynamic response capabilities of the forecast results. The present invention improves the accuracy of power demand forecasting by introducing the LSTM model, combining real-time data and environmental variables, and can more accurately predict future power demand, thereby improving the scientific nature of the scheduling strategy.

[0100] S3: Formulate energy dispatch strategies based on real-time data and forecast results, and dynamically adjust energy distribution and dispatch strategies.

[0101] Furthermore, the dynamic adjustment of energy allocation and scheduling strategy includes:

[0102] According to the real-time prediction results and current load conditions, the energy dispatch strategy is formulated through the improved particle swarm optimization algorithm;

[0103] Dynamically adjust the output power of each energy source through optimization results, and complete real-time adjustment of energy distribution through the energy management system;

[0104] The energy dispatch strategy includes giving priority to energy storage during load trough periods and releasing energy in the energy storage system during load peak periods.

[0105] Furthermore, the improved particle swarm optimization algorithm includes:

[0106] Initialize the particle swarm optimization algorithm, set the particle position and velocity to random values, and initialize the velocity update formula:

[0107] v i (0) = ωv i (0)+c1r1(p i (0)-x i (0))+c2r2(g(0)-x i (0))

[0108] Among them, v i (0) is the velocity of particle i at the initial moment, ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers ranging from 0 to 1, p i (0) is the individual optimal position of particle i, x i (0) is the position of particle i at the initial moment, and g(0) is the global optimal position;

[0109] The original speed update formula is improved according to the predicted short-term power demand value. The improved speed update formula is expressed as:

[0110]

[0111] Among them, γ is the adjustment factor, which is used to balance the impact of the predicted demand value on the speed update; v i (t) is the velocity of particle i at time t; ω is the inertia weight; is the predicted short-term power demand value at time t; p i (t) is the individual optimal position of particle i; x i (t) is the position of particle i at time t; g(t) is the global optimal position; v i (t+1) is the velocity of particle i at time t+1;

[0112] The position update formula is improved by smoothing the speed update effect through the Sigmoid function, which is expressed as follows:

[0113]

[0114] Among them, η is the adjustment factor, which is used to adjust the influence of the Sigmoid function to make the speed update smoother;

[0115] The objective function is set to minimize the deviation between power generation and predicted demand value, energy loss and smoothing effect of power generation. The formula is as follows:

[0116]

[0117] Among them, P gen (t) is the total power generation at time t, E loss(t) is the energy loss at time t, λ is the penalty factor, which controls the impact of energy loss on the objective function; μ is the smoothing factor, which controls the smoothing impact of power generation on the objective function; T is the total number of time steps within the prediction and scheduling time range.

[0118] It should be noted that v i The range of (t+1) is [-V max ,V max ], representing the maximum and minimum values ​​of particle velocity. i The value range of (t+1) is the set search space, which represents the position range of the particle; the value range of the objective function is *0,+∞), where the smaller the value, the better the scheduling strategy, the lower the system energy loss, and the smaller the difference between the power generation and the predicted demand value.

[0119] Furthermore, the dynamically adjusting the output power of each energy source through the optimization result includes calculating the total power generation, which is expressed as:

[0120] P gen (t) = P v (t)+P wind (t)+P storage (t)

[0121] Among them, P gen (t) is the total power generation at time t, P pv (t) is the photovoltaic power generation, P wind (t) is wind power generation, P storage (t) is the power generation of the energy storage system;

[0122] The real-time adjustment of energy distribution through the energy management system includes adjusting the charging and discharging power of the energy storage system according to the predicted power demand and the current power generation, which is expressed as:

[0123]

[0124] Among them, α is the adjustment factor, which is used to balance the output of the energy storage system.

[0125] According to the current state of the energy storage system and the adjusted power, the charging state of the energy storage system is updated, which is expressed as:

[0126]

[0127] Where Emax is the maximum energy capacity of the energy storage system, Δt is the time interval, SOC(t) is the battery charge state at time t, expressed as a percentage; E current (t) is the current energy stored in the battery at time t; Emax is the maximum energy capacity of the battery; the SOC value range is 0% to 100%, indicating the state of the battery from fully discharged to fully charged.

[0128] It should be noted that, through the improved particle swarm optimization algorithm (PSO), the output power of each energy source can be dynamically adjusted according to real-time data and prediction results to ensure the dynamic balance between power generation and power demand load. Traditional scheduling strategies mostly use fixed rules or simple optimization algorithms, which cannot fully respond to power demand fluctuations and environmental changes, resulting in unreasonable energy allocation and low system efficiency. The present invention introduces short-term power demand prediction results into the PSO algorithm, optimizes the speed and position update formulas, and makes the scheduling strategy more dynamic and adaptable. Through real-time adjustment of the energy management system (EMS), the output power of each energy source can be flexibly adjusted according to actual needs, maximizing the use of renewable energy, reducing energy waste, and improving the overall efficiency and stability of the system.

[0129] In summary, the present invention uses sensors and intelligent devices to collect real-time status data of distributed photovoltaic, distributed wind power and energy storage systems, and uses machine learning algorithms to build a short-term power demand forecasting model, which can more accurately predict future power demand. According to real-time data and forecast results, a dynamically adjusted energy scheduling strategy is formulated to improve energy utilization efficiency and the operating stability of the microgrid. Compared with the prior art, improvements have been made in the accuracy of data collection, the precision of the forecasting model, and the dynamics of the scheduling strategy, which can more effectively solve the deficiencies in the prior art.

[0130] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0131] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0132] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0133] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A wind and solar energy storage method applied to a microgrid, characterized by: include, Through sensors and smart devices, real-time data collection of distributed photovoltaic, distributed wind power and energy storage system status; Build a short-term electricity demand forecasting model through machine learning algorithms to predict electricity demand within a certain period of time in the future; Based on real-time data and forecast results, energy scheduling strategies are formulated to dynamically adjust energy distribution and scheduling strategies.

2. The wind-solar energy storage method applied to a microgrid as claimed in claim 1, characterized in that: The short-term electricity supply and demand forecasting model constructed by machine learning algorithm includes: Obtain historical power demand data, weather forecast information, and social media holiday data and perform pre-processing; Use machine learning algorithms for data training and optimization; Construct short-term electricity demand and generation forecast models.

3. The wind-solar energy storage method applied to a microgrid as claimed in claim 2, characterized in that: The construction of the short-term power demand forecasting model comprises: The input power data is transformed nonlinearly by Gaussian radial basis function, which is expressed as: in, is the transformed value of the jth feature of the i-th sample, μ j and σ j are the mean and standard deviation of the jth feature respectively; The meteorological data is processed by polynomial transformation, which is expressed as: ψ k (z i,k )=(a k z i,k +c k ) d Among them, ψ k (z i,k ) is the value of the kth meteorological data feature of the ith sample after transformation, a k and c k is the transformation coefficient, d is the order of the polynomial; The social media holiday data is processed through the logistic regression function, which is expressed as: Among them, φ l (w i,l ) is the transformed value of the lth social media holiday data feature of the i-th sample, η l is the bias parameter of logistic regression; A linear regression model is constructed to capture the basic trends of electricity demand and generation, expressed as: Among them, β0 is the intercept, β j is the weight of the jth feature, is the power data feature after Gaussian radial basis function transformation; The kernel function is used to process high-dimensional data and capture complex feature interactions, which can be expressed as: Among them, K(x i ,x) is the sample x and the support vector x i The kernel function value between σ and ∥x i -x∥ 2 is the Euclidean distance between samples; Combining the kernel function with the linear regression model is expressed as: Among them, α i is the coefficient of the support vector, b is the bias term; Combined with social media holiday data, it is expressed as: Among them, λ is the adjustment parameter, θ l is the weight of the lth social media holiday data feature, φ l (w i,l ) is the lth social media holiday data feature of the ith sample after logistic regression transformation; A complete short-term electricity supply and demand forecasting model is constructed comprehensively, which is expressed as: in, is the predicted value of power demand at time t, t is the specific time point of acquisition, γ is the adjustment parameter, δ k is the weight of the kth meteorological data feature, ψ k (z i,k ) is the kth meteorological data feature of the ith sample after transformation, N is the number of support vectors, M is the number of features, P is the number of meteorological data features, and Q is the number of social media holiday data features.

4. The wind-solar energy storage method applied to a microgrid as claimed in claim 3, characterized in that: The dynamic adjustment of energy allocation and scheduling strategy includes: According to the real-time prediction results and current load conditions, the energy dispatch strategy is formulated through the improved particle swarm optimization algorithm; Dynamically adjust the output power of each energy source through optimization results, and complete real-time adjustment of energy distribution through the energy management system; The energy dispatch strategy includes giving priority to energy storage during load trough periods and releasing energy in the energy storage system during load peak periods.

5. The wind-solar energy storage method applied to a microgrid as claimed in claim 4, characterized in that: The improved particle swarm optimization algorithm includes: Initialize the particle swarm optimization algorithm, set the particle position and velocity to random values, and initialize the velocity update formula: v i (0)=ωv i (0)+c1r1(p i (0)-x i (0))+c2r2(g(0)-x i (0)) Among them, v i (0) is the velocity of particle i at the initial moment, ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers ranging from 0 to 1, p i (0) is the individual optimal position of particle i, x i (0) is the position of particle i at the initial moment, and g(0) is the global optimal position; The original speed update formula is improved according to the predicted short-term power demand value. The improved speed update formula is expressed as: Among them, γ is the adjustment factor; v i (t) is the velocity of particle i at time t; ω is the inertia weight; is the predicted short-term power demand value at time t; p i (t) is the individual optimal position of particle i; x i (t) is the position of particle i at time t; g(t) is the global optimal position; v i (t+1) is the velocity of particle i at time t+1; The position update formula is improved by smoothing the speed update effect through the Sigmoid function, which is expressed as follows: Where η is the adjustment factor; The objective function is set to minimize the deviation between power generation and predicted demand value, energy loss and smoothing effect of power generation. The formula is as follows: Among them, P gen (t) is the total power generation at time t, E loss (t) is the energy loss at time t, λ is the penalty factor, μ is the smoothing factor, and T is the total number of time steps within the prediction and scheduling time range.

6. The wind-solar energy storage method applied to a microgrid as claimed in claim 5, characterized in that: The real-time collection includes installing sensors at key nodes of photovoltaic power generation systems, wind power generation systems and energy storage systems; configuring intelligent monitoring equipment to receive and record data information collected by sensors; and the intelligent monitoring equipment transmits data to a central control system via a wireless communication module.

7. The wind-solar energy storage method applied to a microgrid as claimed in claim 6, characterized in that: The data training and optimization using machine learning algorithms include basic prediction through linear regression to capture the main trends; Use support vector machine kernel functions to process high-dimensional feature interactions to improve prediction accuracy; perform dynamic corrections based on meteorological data and social media holiday data to adjust prediction results.

Citation Information

Cited By

  • Natural wind power redundant output intelligent energy storage cabinet discharge configuration method and system

    CN119253625A

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

    CN120377388A