Big data intelligent green energy adjusting system

Abstract

The invention discloses a big data intelligent green energy adjusting system. The system belongs to the technical field of green energy device power generation prediction and algorithm optimization, and comprises a big data meteorological sensing module, a power supply module, a central processing module and an output adjustment module. The big data meteorological sensing module collects temperature, wind speed, visibility, weather types and the like in a severe environment, a VMD-LSTM prediction model in the central processing module predicts the green energy generating capacity according to data collected by the data collection module, a solar array panel in the power supply module is used for achieving photovoltaic power generation, and a wind driven generator achieves wind power generation. The storage battery stores energy, and the power grid is used for realizing complementation and mutual backup among different types of power supplies; the output adjusting module adjusts and controls the output voltage or current in the circuit and ensures stable and reliable output. In combination with green energy device power generation prediction, an accurate energy supply demand can be obtained, and effective utilization of new energy and effective supply of required electric energy in a severe environment area can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of virtual power plant optimal dispatching, and relates to a coal power plant resource optimal dispatching system; specifically, it relates to a coal power plant optimal dispatching system cooperating with a virtual power plant; particularly, it relates to a big data intelligent green energy regulation system. Background Art

[0002] With global climate change, frequent occurrence of severe weather such as haze, sand and dust, heavy rainfall, etc., increasing natural disasters, increasing energy waste and loss, the multi-level output characteristics of the energy system are affected, the utilization rate of energy is reduced, and the existing situation can no longer meet the demand for energy. A corresponding matching and dispatching system and method aiming at energy conservation and emission reduction will replace the original system and method for long-term development. Existing methods and technologies include early warning systems, energy storage and dispatching systems, diversified energy structures, smart grid technologies, etc. However, these technologies and systems all have disadvantages such as being greatly affected by natural weather conditions, poor stability, large errors in meteorological disaster prediction, and inability to consider multiple severe meteorological systems, and cannot effectively improve and enhance the utilization rate of natural energy. Based on this background, it is necessary to develop a corresponding matching and dispatching system and method to weaken the impact of these severe meteorological systems on the multi-level output of the energy system, improve the stability and economy of the energy system, and achieve the effect of energy conservation and emission reduction. Summary of the Invention

[0003] In view of the above problems, the object of the present invention is to propose a big data intelligent green energy regulation system.

[0004] The technical solution of the present invention is: A big data intelligent green energy regulation system described in the present invention includes a big data meteorological perception module, a central processing module, an output regulation device, and a power supply module that are connected to each other;

[0005] The big data meteorological perception module includes a wind speed sensor, a temperature sensor, a photosensitive sensor, etc., and is mainly responsible for collecting temperature, wind speed, light, PM2.5, and weather type in a harsh environment, and performing statistics and processing;

[0006] The big data meteorological perception module realizes unified monitoring and management of local weather through a local platform and a centralized control platform;

[0007] The central processing module includes a data processing system and a VMD-LSTM prediction model;

[0008] The VMD-LSTM prediction model in the central processing module predicts the green energy power generation according to the data collected by the big data meteorological perception module and the data acquisition module, and generates a prediction model through algorithm optimization; and transmits the prediction data to the data processing system for further processing;

[0009] Among them, the data processing system processes and classifies the data collected by the meteorological perception module and the data collection module based on big data collection, and then transmits the processed data to the power supply regulation equipment.

[0010] The power supply module includes a solar array panel, a wind turbine, a storage battery, and a power grid;

[0011] In the power supply module, the solar array panel is used to realize photovoltaic power generation, convert solar energy into electrical energy, and provide clean and renewable energy; the wind turbine captures wind energy through the wind turbine blades, converts it into mechanical energy, and then converts it into electrical energy through the generator; the storage battery is used to receive the electrical energy generated by the solar array panel and the wind turbine, and release it when needed; the main function of the power grid is to transmit, control, and distribute electrical energy, which is composed of a substation and transmission lines of different voltage levels, and transmits the electrical energy from the storage battery to each output regulation equipment;

[0012] The output regulation equipment adjusts and controls the voltage or current in the circuit and ensures stable and reliable output; it can adjust the input voltage or current to keep the output constant within a set range, and monitor and adjust the output in real time through a feedback loop to cope with load changes or environmental interference, ensuring the normal operation of the circuit system.

[0013] Furthermore, this technology conducts a data output process. First, the obtained relevant environmental data is preprocessed, a feature extraction algorithm is used, a VMD-LSTM prediction model and an objective function are established, and the optimal improvement method is obtained, so as to reduce the impact of bad weather on the output of the energy system and achieve the purpose of energy conservation.

[0014] Furthermore, according to its harsh environment, a prediction model for the device needs to be established. The establishment process of the VMD-LSTM prediction model is as follows:

[0015] 1): Use the VMD algorithm to decompose the environmental data signal into several modal components. The process is as follows:

[0016] VMD decomposes various environmental data signals, such as temperature, wind speed, power, light intensity and other data signals, into k modal components u k (t), and the formula is as follows: u k (t) = A k (t)cos[φ k (t)];

[0017] In the formula, u k (t) is the modal component after the decomposition of the environmental data signal; A k (t) is the instantaneous amplitude; φ k (t) is the phase function; k is the number of modal components; t is the time variable;

[0018] The VMD algorithm can be divided into two parts: the construction and solution of the variational problem;

[0019] Among them, the construction of the variational problem:

[0020] Apply the Hilbert transform to each of the decomposed modal components to obtain the analytic signal and the single-sided spectrum:

[0021] Add an exponential term Modulate the spectrum of each modal component to the corresponding base frequency band:

[0022] The bandwidth of each modal component is estimated by the norm of the gradient L of the environmental data signal 2 of, and the constrained variational problem is as follows:

[0023]

[0024] In the formula, {u k} is the set of all modal components of the environmental data signal; {ω k} is the set of the center frequencies of all modal components of the environmental data signal; f(t) is the environmental data signal;

[0025] The solution of the variational problem:

[0026] Transform the above constrained variational problem into an unconstrained variational problem, introduce a quadratic penalty factor and a Lagrange multiplier operator, and the extended Lagrangian expression is:

[0027]

[0028] In the formula, λ is the Lagrange multiplier; α is the quadratic penalty coefficient;

[0029] 2): Input the modal component u k (t) after decomposing the environmental data signal, and use the LSTM model to model each modal component separately;

[0030] The LSTM model is mainly composed of a forget gate, an input gate, a memory gate, and an output gate; the forget gate is used to selectively forget the power data input by the previous node, the input gate is used to input the power data, the memory gate is used to selectively remember the input power data, and the output gate is used to output the power data; its operation formula is as follows:

[0031] f t =σ(W f ·[x t ,h t-1 +b f )

[0032] i t =σ(W i·[x t ,h t-1 +b i )

[0033] C t =tanh(W c ·[x t ,h t-1 +b c )

[0034] C t =f t ·C t-1 +i t ·C t

[0035] O t =σ(W o ·[x t ,h t-1 +b o )

[0036] h t =O t ·tanh(C t )

[0037] In the formula, f t , i t , C t , O t are the information update processes of the forget gate, input gate, memory gate, and output gate, W f , W i , W c , W o are their corresponding weight matrices, b f , b i , b c , b o are their corresponding bias constants; C t is the temporary state; σ is the sigmoid function; tanh is the hyperbolic tangent function; x t is the power data of the current state; h t-1 is the power data of the previous node; h t is the output power data.

[0038] Through the VMD-LSTM model, an accurate prediction of the power generation of green energy devices within a certain period of time is obtained.

[0039] Furthermore, to find the optimal improvement plan for the device, the EVO algorithm can be used to optimize the decision variables in the objective function: Define the number of operating photovoltaic modules N PV , the number of operating wind turbines N WT , the number of operating batteries Nbat The process is as follows:

[0040] 1): Initialize the positions and parameters of the particles: N PV , N WT and N bat , and the position formula of the particle is as follows:

[0041]

[0042] In the formula, is the j-th decision variable for determining the initial position of the i-th candidate; rand is an arbitrary vector with a range of [0, 1]; are the upper and lower bounds for solving the problem respectively; n is the total number of particles (candidate solutions) in the universe (search space); d is the dimension of the problem to be solved; this step initializes the data;

[0043] 2): Substitute the randomly generated particle positions into the objective function for evaluation, so that each particle generates a corresponding set of equipment operation quantities N PV , N WT , N bat , and the objective function formula is as follows:

[0044] minC = C 1 + C 2

[0045] C 1 = N PV ·p PV + N WT ·p WT + N bat ·p bat

[0046]

[0047] In this formula, constraint (1) ensures the stability of the energy supply of the green device in a harsh environment; constraints (2)-(4) limit the maximum installation number; constraint (5) limits the maximum power traded with the power grid;

[0048] In the formula, C is the objective function value, representing the lowest cost; C 1 is the equipment operation cost; C 2 is the power purchase cost; N PV is the operation quantity of the photovoltaic modules; p PV is the operation cost of each group of photovoltaic modules; N WT is the operation quantity of the wind turbines; p WT is the operation cost of each group of wind turbines; N bat is the operation quantity of the storage batteries; p bat is the operation cost of each group of storage batteries; uE and u S are the unit power purchase price and the unit power selling price respectively; is a 0-1 variable. When it is 0, power is purchased from the power grid. When it is 1, power is sold to the power grid; are the power purchase power and the power selling power at time t respectively; is the output power of a unit photovoltaic module at time t; is the output power of a unit wind turbine at time t; are the charging and discharging powers of the battery at time t respectively; D t is the required electrical energy, D t = h t q t h t is traffic flow data; q t is the required electrical energy at time t; are the maximum installation quantities of photovoltaic modules, wind turbines, and batteries respectively; P max is the maximum power for trading with the power grid;

[0049] 3): Determine the enrichment boundary of the particles. The formula is as follows:

[0050]

[0051] In the formula, NEL i is the neutron enrichment level of the i-th particle;

[0052] 4): Obtain a more suitable operation plan for the multi-layer output characteristic matching and scheduling system based on the severe weather process by updating the positions of the equipment operation quantities, so as to obtain the lowest cost;

[0053] Specifically, when the neutron enrichment level is higher than the enrichment boundary,

[0054] when the stability level of the particle is higher than the stability boundary, emit α rays to improve the stability of the product in the physical reaction:

[0055]

[0056] when the stability level of the particle is lower than the stability boundary, the particle releases β rays to improve the stability level of the particle:

[0057]

[0058] When the neutron enrichment level is lower than the enrichment boundary, the particle moves towards the stable band through electron capture or positron emission:

[0059]

[0060] 5): Determine whether the termination condition is satisfied; if the termination condition is not satisfied, return to step 2); if it is satisfied, go to step 6);

[0061] 6): Output the optimal operation plan of the multi-layer output feature matching and scheduling system based on the severe weather process.

[0062] Furthermore, the improved part of the algorithm lies in the improvement of 5), introducing an adaptive weight factor to improve the position of the newly generated particles and enhancing the local optimization ability of the IEVO algorithm; the improved formula is as follows:

[0063]

[0064] By combining the improved algorithm with the relevant data collected by meteorology, the possibility of finding the optimal solution can be greatly increased, making the cost required by the device the lowest, and enabling stable operation under severe weather conditions.

[0065] Natural energy such as light energy and wind energy is converted into electrical energy and output after passing through the processing and regulating equipment; in the algorithm, the ability of the power supply module to convert electrical energy is optimized, the stability level of electrons is optimized, the operation efficiency of the power grid is improved, and the energy utilization rate is increased.

[0066] The beneficial effects of the present invention are as follows: The present invention will weaken the influence of various severe weather on the energy utilization rate according to the actual situation of different regions, improve the stability and economy of the energy system, promote the large-scale application of renewable energy under severe weather, and promote the transformation and upgrading of the energy structure; it can effectively make a prediction model according to the actual situation of each region, aiming at the meteorological system and its characteristics in this region, and establish a VMD-LSTM prediction model and an objective function according to the feature extraction algorithm to obtain the optimal solution for the utilization of natural energy in this region, improve the timeliness and utilization efficiency, thereby reducing the influence of severe weather on the output of the energy system and achieving the purpose of energy conservation. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic structural diagram of each device in the present invention;

[0068] Figure 2 It is a system flowchart related to the algorithm in the present invention;

[0069] Figure 3 It is a schematic diagram of the principle of a multi-layer output feature matching and scheduling system based on a severe weather process in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0070] The following further elaborates on the specific technical solutions of the present invention with specific examples.

[0071] As shown in the figure, a big data intelligent green energy regulation system according to the present invention includes a big data meteorological perception module, a power supply module, a central processing module, and an output regulation module;

[0072] The big data meteorological perception module is responsible for collecting temperature, wind speed, PM2.5, weather type in harsh environments, and collecting environmental data;

[0073] The VMD-LSTM prediction model in the central processing module predicts the green energy power generation according to the data collected by the data acquisition module; the algorithm is used to optimize the model parameters;

[0074] The power supply module includes a solar array panel, a wind turbine, a storage battery, and a power grid. The solar array panel is used to realize photovoltaic power generation, the wind turbine is used to realize wind power generation, the storage battery is used to store energy, and the power grid is used to realize the complementarity and mutual backup between different types of power sources;

[0075] The output regulation device regulates and controls the output voltage or current in the circuit and ensures stable and reliable output; it can regulate the input voltage or current to keep a constant output within a set range, and monitor and adjust the output in real time through a feedback loop to cope with load changes or environmental interference to ensure the normal operation of the circuit system.

[0076] Main issues considered in the design:

[0077] 1. Unstable weather: Clean energy devices (such as solar panels, wind turbines, etc.) are greatly affected by weather factors. For example, cloudy days, overcast days, and unstable wind will all affect the power generation efficiency, making regulation more difficult;

[0078] 2. Load fluctuations: In harsh environments, the load fluctuates greatly, which may cause the output voltage or current of clean energy devices to be unstable, and timely regulation is required to ensure stable power output;

[0079] 3. Pollution of clean energy devices: In harsh environments, clean energy devices may be affected by pollutants such as dust, sand, and haze, resulting in a decrease in power generation efficiency and an increase in regulation difficulty;

[0080] 4. Temperature influence: In extreme temperature environments, the performance of clean energy devices may be affected, and appropriate regulation is required to ensure normal power generation.

[0081] This technology conducts a data processing process. First, the obtained relevant environmental data is preprocessed, a feature extraction algorithm is used, a VMD-LSTM prediction model and an objective function are established, and the optimal improvement means are obtained, so as to reduce the impact of bad weather on the output of the energy system.

[0082] Furthermore, according to the influence of its harsh environment, a prediction model for the device needs to be established. The establishment process of the VMD-LSTM prediction model is as follows:

[0083] 1). Use the VMD algorithm to decompose the environmental data signal into several modal components. The process is as follows:

[0084] VMD decomposes various environmental data signals, such as temperature, wind speed, power, light, etc., into k modal components u k (t), and the formula is as follows: u k (t) = A k (t)cos[φ k (t)];

[0085] In the formula, u k (t) is the modal component after decomposing the environmental data signal; A k (t) is the instantaneous amplitude; φ k (t) is the phase function; k is the number of modal components; t is the time variable;

[0086] The VMD algorithm can be divided into two parts: the construction and solution of the variational problem;

[0087] Construction of the variational problem:

[0088] Apply the Hilbert transform to each modal component obtained by decomposition to obtain the analytic signal and the single-sided spectrum:

[0089] Add an exponential term Modulate the spectrum of each modal component to the corresponding base frequency band:

[0090] The bandwidth of each modal component is estimated by the norm of the gradient L 2 of the environmental data signal. The constrained variational problem is as follows:

[0091]

[0092] In the formula, {u k} is the set of all modal components of the environmental data signal; {ω k} is the set of the center frequencies of all modal components of the environmental data signal; f(t) is the environmental data signal;

[0093] Solution of the variational problem:

[0094] Convert the above constrained variational problem into an unconstrained variational problem, introduce a quadratic penalty factor and a Lagrange multiplier operator, and the extended Lagrangian expression is:

[0095]

[0096] where λ is the Lagrange multiplier; α is the quadratic penalty coefficient;

[0097] 2), input the modal component u k (t) of the decomposed environmental data signal, and use the LSTM model to model each modal component respectively.

[0098] The LSTM model is mainly composed of a forget gate, an input gate, a memory gate, and an output gate; the forget gate is used to selectively forget the power data input by the previous node, the input gate is used to input the power data, the memory gate is used to selectively remember the input power data, and the output gate is used to output the power data; the operation formulas are as follows:

[0099] f t = σ(W f · [x t , h t-1 + b f )

[0100] i t = σ(W i · [x t , h t-1 + b i )

[0101] C t = tanh(W c · [x t , h t-1 + b c )

[0102] C t = f t · C t-1 + i t · C t

[0103] O t = σ(W o · [x t , h t-1 + b o )

[0104] h t = O t · tanh(C t )

[0105] where f t , i t , C t , O t are the information update processes of the forget gate, input gate, memory gate, and output gate, and W f , W i , W c , Wo is its corresponding weight matrix, b f , b i , b c , b o is its corresponding bias constant; C t is the temporary state; σ is the sigmoid function; tanh is the hyperbolic tangent function; x t is the power data of the current state; h t-1 is the power data of the previous node; h t is the output power data;

[0106] Through the VMD-LSTM model, an accurate prediction of the power generation of green energy devices within a certain period of time is obtained.

[0107] Furthermore, it is necessary to find the optimal improvement plan for the device. The decision variables in the objective function can be optimized using the EVO algorithm: Define the operating quantity N of photovoltaic modules PV , the operating quantity N of wind turbines WT , and the operating quantity N of storage batteries bat The process is as follows:

[0108] 1): Initialize the position and parameters of the particles: N PV , N WT , and N bat . The position formula of the particles is as follows:

[0109]

[0110] In the formula, is the jth decision variable used to determine the initial position of the ith candidate; rand is an arbitrary vector with a range of [0,1]; are the upper and lower boundaries of the problem to be solved respectively; n is the total number of particles (candidate solutions) in the universe (search space); d is the dimension of the problem to be solved. This step initializes the data;

[0111] 2): Substitute the randomly generated particle positions into the objective function for evaluation, so that each particle generates a corresponding set of device operating quantities N PV , N WT , N bat . The objective function formula is as follows:

[0112] minC = C 1 + C 2

[0113] C 1 = N PV ·p PV + N WT ·p WT + N bat·p bat

[0114]

[0115] In this formula, constraint (1) ensures the stability of the energy supply of the green device in harsh environments; constraints (2)-(4) limit the maximum installation number; constraint (5) limits the maximum power traded with the power grid.

[0116] In the formula, C is the objective function value, representing the lowest cost; C 1 is the equipment operation cost; C 2 is the electricity purchase cost; N PV is the number of operating photovoltaic modules; p PV is the operation cost of each group of photovoltaic modules; N WT is the number of operating wind turbines; p WT is the operation cost of each group of wind turbines; N bat is the number of operating batteries; p bat is the operation cost of each group of batteries; u E and u S are the unit electricity purchase price and the unit electricity selling price respectively; is a 0-1 variable, which means purchasing electricity from the power grid when it is 0 and selling electricity to the power grid when it is 1; are the electricity purchase power and the electricity selling power at time t respectively; is the output power of a unit photovoltaic module at time t; is the output power of a unit wind turbine at time t; are the charging and discharging powers of the battery at time t respectively; D t is the required electrical energy, D t = h t q t h t is traffic flow data; q t is the required electrical energy at time t; are the maximum installation numbers of photovoltaic modules, wind turbines, and batteries respectively; P max is the maximum power traded with the power grid;

[0117] 3): Determine the enrichment boundary of the particles, and the formula is as follows:

[0118]

[0119] In the formula, NEL i is the neutron enrichment level of the i-th particle, representing the density of possible solution occurrences; the greater the density, the more likely an optimal solution, i.e., the lowest cost, will appear.

[0120] 4): Obtain an operation plan that is more suitable for the multi-layer output characteristic matching and scheduling system based on severe meteorological processes of the present invention by updating the position of the number of operating devices, thereby obtaining the lowest cost.

[0121] When the enrichment level of neutrons is higher than the enrichment boundary,

[0122] When the stability level of the particle is higher than the stability boundary, emit α rays to improve the stability of the product in the physical reaction:

[0123]

[0124] When the stability level of the particle is lower than the stability boundary, the particle releases β rays to improve the stability level of the particle:

[0125]

[0126] When the neutron enrichment level is lower than the enrichment boundary, the particle moves towards the stable band through electron capture or positron emission:

[0127]

[0128] 5): Determine whether the termination condition is satisfied; if the termination condition is not satisfied, return to step 2); if satisfied, go to step 6);

[0129] 6): Output the optimal operation plan of the multi-layer output characteristic matching and scheduling system based on severe meteorological processes.

[0130] Furthermore, the improved part of the algorithm lies in the improvement of 5), introducing an adaptive weight factor to improve the position of the newly generated particles and enhancing the local optimization ability of the IEVO algorithm; the improvement formula is as follows:

[0131]

[0132] By combining the improved algorithm with the relevant data collected by meteorological collection, the possibility of finding the optimal solution can be greatly increased, making the cost required by the device the lowest, and enabling stable operation under severe weather conditions.

[0133] The working principle of the present invention: Establish a big data meteorological perception module to collect data such as visibility, PM2.5, wind speed, temperature, and weather type, and transmit it to the central control module. In the central control module, a data processing system is established to preprocess the obtained data, perform feature extraction algorithms, establish a VMD-LSTM prediction model and an objective function, obtain the optimal scheduling means, and the power supply module provides power to the output adjustment device, thereby reducing the impact of severe weather on the output of the energy system, improving the utilization rate of energy, promoting the large-scale application of renewable energy under severe weather conditions, reducing energy loss, and ensuring the stability of power supply.

[0134] Performance analysis:

[0135] First of all, the advantage of the system lies in its comprehensive consideration of the impact of environmental factors on energy production and supply; the data collected by the big data meteorological perception module provides accurate environmental data for the system. These data help the VMD-LSTM prediction model in the central processing module to accurately predict the power generation of green energy. Thus, the system can more effectively allocate power supply modules such as solar array panels, wind turbines, storage batteries, and power grids, and achieve optimal scheduling of green energy production and supply under adverse weather conditions. This advantage enables the system to maintain high energy supply stability and reliability under adverse weather conditions;

[0136] Secondly, the output regulation module adopted by the system can accurately regulate the output voltage or current in the circuit to ensure stable and reliable output; by real-time monitoring and adjustment of the output, the system can timely respond to load changes or environmental disturbances and effectively maintain the normal operation of the circuit system; this feature makes the system have high adaptability and adjustability and can flexibly meet the requirements of different working environments and application scenarios.

[0137] However, the system also faces some challenges in the design and implementation process; first of all, for the VMD-LSTM prediction model in the central processing module, its accuracy and stability are crucial to the overall performance of the system; it is necessary to fully consider the quality and timeliness of meteorological data, as well as the algorithm design and parameter optimization of the prediction model to improve the prediction accuracy and reliability; secondly, the design and configuration of the power supply module need to fully consider the characteristics of energy production and supply under different environmental conditions, as well as the energy efficiency and cost-effectiveness of the system; especially under adverse weather conditions, the performance of solar array panels and wind turbines is affected to a certain extent, and it is necessary to reasonably allocate and configure energy storage devices and backup power supplies to ensure the stable operation of the system.

[0138] In summary, this big data intelligent green energy regulation technology has significant advantages in improving the utilization rate of green energy and supply stability, but still needs to be further optimized and improved in terms of prediction model accuracy, power supply module design and configuration, etc. to meet the actual application requirements under different environmental conditions.

[0139] Features and Applications of the Invention: (I) Considering the impacts of multiple meteorological factors: In the scheduling of traditional green energy systems, there is often a lack of comprehensive consideration of multiple adverse weather factors; usually, only a single meteorological condition, such as wind speed or light intensity, is concerned, while the impacts of other key factors such as visibility and humidity on the system output are ignored; this technology breaks through the consideration of a single adverse weather factor and innovatively incorporates multiple adverse weather factors such as haze, sand and dust, reduced visibility, and weakened solar radiation into the system evaluation scope; by comprehensively evaluating the impacts of these factors on the system output and processing the collected data, it can accurately predict and respond to the energy production fluctuations under adverse weather conditions, and can more accurately predict and respond to the changes in the system output of the energy system, thereby improving the stability of the system;

[0140] (II) Applying advanced comprehensive output situation awareness technology: There are technical bottlenecks in the traditional energy system for real-time monitoring and awareness of the output of green energy systems. Relying on limited data collection devices and means, it is difficult to achieve comprehensive and accurate awareness of the system output. The project introduces advanced cloud platforms and data collection technologies to build an efficient comprehensive output situation awareness system for large-scale wind-solar-storage base groups; this system can real-time monitor and regulate the energy output of the base groups to ensure that the system can still maintain stable and efficient energy supply under adverse weather conditions; the application of this situation awareness technology not only improves the system's response ability but also enhances its operation safety and reliability;

[0141] (III) Optimizing the energy scheduling strategies for different climate zones within the base group: There was a lack of in-depth research on the characteristics of different climate zones within the base group and the flexibility of energy storage before. This technology deeply studies the unique characteristics of different climate zones within the base group and combines the flexible characteristics of energy storage technology to propose targeted energy scheduling optimization strategies, coordinating green energy power generation, traditional hydropower stations, and thermal power stations to achieve complementarity and mutual backup between different types of power sources; by making full use of the resource advantages of different climate zones and the flexible adjustment ability of energy storage technology, the project aims to maximize the utilization of energy output and ensure the stable operation of the system, and can maintain efficient energy supply even under adverse weather conditions;

[0142] (IV) Exploring the energy scheduling compensation mechanism under adverse weather conditions: Facing the adverse impacts of adverse weather on energy output, this technology explores and establishes an effective energy scheduling compensation mechanism; through scientific scheduling and reasonable compensation, it aims to make up for the shortage of energy output under adverse weather conditions, reduce the operation risks of the system, and ensure the continuity and stability of energy supply; the establishment of this mechanism is of great significance for improving the response ability and operation efficiency of green energy systems.

Claims

1. A big data intelligent green energy regulation system, characterized in that: It includes interconnected big data meteorological sensing modules, central processing modules, output regulation equipment and power supply modules; The big data meteorological perception module includes a wind speed sensor, a temperature sensor and a photosensor, which are used to collect temperature, wind speed, light, PM2.5 and weather type in harsh environments and perform statistics and processing; The central processing module includes a data processing system and a VMD-LSTM prediction model; The power supply module includes a solar array panel, a wind turbine, a battery and a power grid; The output regulating device regulates and controls the voltage or current in the circuit, and regulates the input voltage or current to maintain a constant output within a set range.

2. A big data intelligent green energy regulation system according to claim 1, characterized in that: The VMD-LSTM prediction model predicts green energy power generation based on the data collected by the big data meteorological perception module and the data acquisition module, generates a prediction model through algorithm optimization, and transmits the predicted data to the data processing system.

3. The big data intelligent green energy regulation system according to claim 1 is characterized in that: The data processing system is based on big data collection, and processes and classifies the data collected by the big data meteorological perception module and the data collection module and transmits them to the power supply regulation equipment.

4. The big data intelligent green energy regulation system according to claim 1 is characterized in that: The solar array panel is used to realize photovoltaic power generation, converting solar energy into electrical energy; The wind turbine captures wind energy through the wind blades, converts it into mechanical energy, and then converts it into electrical energy through the generator; The storage battery is used to receive the electric energy generated by the solar array panel and the wind turbine generator and release it when needed; The power grid is used to transmit, control and distribute electric energy, and is composed of a substation and transmission lines of different voltage levels, which transmit electric energy from the battery to various output regulation devices.

5. The big data intelligent green energy regulation system according to claim 2 is characterized in that: The establishment process of the VMD-LSTM prediction model is as follows: 1): Use the VMD algorithm to decompose the environmental data signal into several modal components. The process is as follows: VMD decomposes the data signals of various environmental data signals into k modal components u k (t), the formula is as follows: k (t) = A k (t)cos[φ k (t)]; In the formula, u k (t) is the modal component after decomposition of the environmental data signal; A k (t) is the instantaneous amplitude; φ k (t) is the phase function; k is the number of modal components; t is the time variable; 2): The modal component u after decomposition of the input environmental data signal k (t), each modal component is modeled separately using the LSTM model.

6. The big data intelligent green energy regulation system according to claim 5 is characterized in that: In step 1), the VMD algorithm is divided into two parts: construction and solution of variational problem; Among them, the construction of the variational problem is: Apply the Hibert transform to each modal component obtained by decomposition to obtain the analytical signal and the single-sided spectrum: Add index item Modulate the spectrum of each modal component to the corresponding baseband: The bandwidth of each modal component is determined by the environmental data signal gradient L 2 The norm of is estimated, and the constrained variational problem is as follows: In the formula, {u k } is the set of all ambient data signal modal components; {ω k } is the set of central frequencies of all ambient data signal modal components; f(t) is the ambient data signal; Solution of variational problem: The above constrained variational problem is transformed into an unconstrained variational problem, and the quadratic penalty factor and Lagrangian multiplication operator are introduced. The extended Lagrangian expression is: Where λ is the Lagrange multiplier and α is the quadratic penalty coefficient.

7. The big data intelligent green energy regulation system according to claim 5 is characterized in that: In step 2), the LSTM model is composed of a forget gate, an input gate, a memory gate and an output gate; the forget gate is used to selectively forget the power data input by the previous node, the input gate is used to input the power data, the memory gate is used to selectively memorize the input power data, and the output gate is used to output the power data; its calculation formula is as follows: f t =σ(W f ·[x t ,h t-1 ]+b f ) i t =σ(W i ·[x t ,h t-1 ]+b i ) C t =tanh(W c ·[x t ,h t-1 ]+b c ) C t =f t ·C t-1 +i t ·C t The t =σ(W o ·[x t ,h t-1 ]+b o ) h t =O t ·tanh(C t ) In the formula, f t 、i t , C t , O t is the information update process of the forget gate, input gate, memory gate, and output gate, W f , W i , W c , W o Its corresponding weight matrix, b f 、b i 、b c 、b o is the corresponding bias constant; C t is a temporary state; σ is the sigmoid function; tanh is the hyperbolic tangent function; x t is the power data of the current state; h t-1 is the power data of the previous node; h t The output power data.

8. The big data intelligent green energy regulation system according to claim 1 is characterized in that: To find the optimal improvement plan for the device, the EVO algorithm is used to optimize the decision variables in the objective function: define the number of photovoltaic modules in operation N PV 、N number of wind turbines in operation WT 、N number of running batteries bat The process is as follows: 1): Initialize the position and parameters of the particle: N PV 、N WT and N bat , the particle position formula is as follows: In the formula, is the jth decision variable used to determine the initial position of the i-th candidate; rand is an arbitrary vector in the range of [0,1]; are the upper and lower boundaries of the problem to be solved; n is the total number of particles in the universe; d is the dimension of the problem to be solved; 2): The randomly generated particle positions are brought into the objective function for evaluation, so that each particle generates a set of corresponding device operation numbers N PV 、N WT 、N bat , the objective function formula is as follows: min C=C1+C2 C1=N PV ·p PV +N WT ·p WT +N bat ·p bat In the above formula, constraint (1) ensures the stability of green device energy supply in harsh environments; constraints (2)-(4) limit the maximum number of installations; constraint (5) limits the maximum power traded with the grid; In the formula, C is the objective function value, representing the lowest cost; C1 is the equipment operating cost; C2 is the electricity purchase cost; N PV is the number of PV panels in operation; p PV is the operating cost of each set of photovoltaic modules; N WT is the number of wind turbines in operation; p WT is the operating cost of each wind turbine generator set; N bat is the number of battery operations; p bat is the operating cost of each battery pack; u E and u S They are the unit electricity purchase price and the unit electricity sales price respectively; It is a 0-1 variable. When it is 0, it buys electricity from the power grid, and when it is 1, it sells electricity to the power grid; P t E , P t S are the purchased power and sold power at time t respectively; P t PV is the output power of the unit photovoltaic module at time t; P t WT is the output power of the unit wind turbine at time t; P t di , P t ch are the charging and discharging power of the battery at time t; D t is the required electrical energy, D t =h t q t ,h t is the traffic flow data; q t The power demand at time t; are the maximum number of photovoltaic modules, wind turbines, and batteries installed; P max is the maximum power traded with the grid; 3): Determine the enrichment boundary of the particles, the formula is as follows: Where NEL i is the neutron enrichment level of the ith particle; 4): Obtain the operation plan of the multi-layer output characteristic matching and dispatching system by updating the position of the equipment operation quantity; 5): Determine whether the termination condition is met; If the termination condition is not met, return to step 2); if it is met, go to step 6); 6): Output the optimal multi-layer output characteristic matching and dispatching system operation plan based on severe meteorological processes.

9. The big data intelligent green energy regulation system according to claim 8 is characterized in that: In step 4), when the neutron enrichment level is higher than the enrichment limit, When the stability level of the particle is higher than the stability limit, the emission of α rays increases the stability of the product in physical reactions: When the stability level of a particle is lower than the stability limit, the particle releases beta rays to improve the stability level of the particle: When the neutron enrichment level is below the enrichment boundary, the particles move toward the stability band by electron capture or positron emission:

10. The big data intelligent green energy regulation technology according to claim 8 is characterized in that: The improved part of the algorithm is to improve 5), introduce an adaptive weight factor to improve the position of the newly generated particles, and improve the local optimization ability of the IEVO algorithm; the improved formula is as follows: By combining improved algorithms with data collected by meteorology, the possibility of finding the optimal solution is increased.

Images