New energy power station electricity abandoning analysis method and system in electricity market environment

Abstract

The invention relates to a new energy power station electricity abandoning analysis method and system in an electricity market environment. The method comprises the following steps of collecting various data and fusing the data into a data set; using the data set and the LSTM model to predict the generating capacity of the new energy power station; determining a power grid constraint, and predicting a power market demand according to the data set; establishing a power abandoning calculation model according to the power grid constraint, the power market demand and the generating capacity, and obtaining the power abandoning quantity of the new energy power station through the power abandoning calculation model; according to the new energy power station optimization strategy making method, a detailed and comprehensive method is adopted in the aspect of data processing, and the method comprises the steps of collecting various types of data (such as new energy power generation data and power grid operation data), and conducting normalization processing and data fusion. The method is helpful for extracting useful information in the data and improving the prediction accuracy of the model.

Description

Technical Field

[0001] The present invention relates to a method and system for analyzing power abandonment in a new energy power station under an electric power market environment, and belongs to the field of data analysis science and technology. Background Art

[0002] With the rapid development of the new energy industry, new energy sources such as wind power and photovoltaics account for an increasingly higher proportion in the power system. However, due to the randomness, intermittency and volatility of new energy power generation, its output is difficult to maintain continuous, stable and balanced, resulting in the increasingly prominent problem of power abandonment in new energy power stations. In the power market environment, power abandonment in new energy power stations will not only affect the economic benefits of new energy power generation companies, but also have an adverse impact on the stable operation of the power market and the optimization of the energy structure.

[0003] Traditional analysis methods are often based on historical data or simulation data, which makes it difficult to reflect the operating status of new energy power stations and the actual situation of the power grid in real time. At the same time, due to the volatility and uncertainty of new energy power generation, traditional analysis methods are also insufficient in accuracy, making it difficult to accurately predict and evaluate the amount of power abandoned by new energy power stations. In the power market environment, the problem of power abandonment by new energy power stations is not only related to power generation technology and power grid scheduling, but also closely related to market factors such as market demand and electricity price mechanism. Traditional analysis methods often ignore the role of these market factors, resulting in deviations between the analysis results and the actual situation.

[0004] The patent document with the patent publication number "CN117595241A" has developed a method for multi-type power supply regulation and new energy coordination based on random forests. The problem with this method is that it may be relatively simple in terms of data processing, and mainly relies on repeated sampling of historical power generation data sets to create different data subsets. Although this method can increase the robustness of the model to a certain extent, it may not make full use of all available data information, thereby affecting the predictive performance of the model. It mainly relies on the random forest model for prediction. Although the random forest model performs well in some cases, it may not be as effective as the LSTM model when processing data with time series characteristics. In addition, when predicting the power generation of new energy, the final power generation is determined only by comparing the predicted value with the theoretical value. This method may be too simplified and cannot fully reflect the complexity of the actual situation. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention proposes a method and system for analyzing power abandonment of a new energy power station in an electricity market environment.

[0006] The technical solution of the present invention is as follows:

[0007] On the one hand, the present invention provides a method for analyzing power abandonment of a new energy power station in an electric power market environment, characterized in that it comprises the following steps:

[0008] Collect new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrate them into a data set;

[0009] Use the data set and LSTM model to predict the power generation of new energy power plants;

[0010] Identify grid constraints and forecast electricity market demand based on data sets;

[0011] Establish a power curtailment calculation model based on grid constraints, power market demand, and power generation, and obtain the power curtailment of new energy power stations through the power curtailment calculation model;

[0012] An optimization strategy for the new energy power station is formulated based on the amount of abandoned electricity.

[0013] As a preferred implementation, the data set fusion method is:

[0014] Normalize new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data:

[0015]

[0016] Among them, x new represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data after normalization, x represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x min represents the minimum value of the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization. max Indicates the maximum value of renewable energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization;

[0017] The normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data are integrated:

[0018]

[0019] Among them, D clean represents the fused data set, α i represents the weight, βi represents the reliability factor, θ i From i=0 to i=n, ​​they represent respectively the normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data.

[0020] As a preferred implementation, the method for predicting the power generation of the new energy power station is:

[0021] According to the LSTM model, n LSTM model prediction values ​​are predicted. The prediction method of the LSTM model is:

[0022] The forget gate of the LSTM model is expressed as:

[0023] f t =σ(W f [h t-1 ,x t ]+b f );

[0024] Among them, f t represents the activation value of the forget gate, W f represents the weight matrix of the forget gate, b f represents the bias term of the forget gate, σ represents the activation function, and h t-1 represents the hidden state at time t-1, x t Represents the data set at time t;

[0025] The LSTM model input gate is represented as:

[0026] i t =σ(W i [h t-1 ,x t ]+b i );

[0027] Among them, W i represents the weight matrix of the input gate, b i represents the bias term of the input gate, i t represents the activation value of the input gate;

[0028] The cell state of the LSTM model at time t is expressed as:

[0029] c t =f t ⊙c t-1 +i t ⊙tanh(W c [h t-1 ,x t ]+b c );

[0030] Among them, c trepresents the cell state at time t, c t-1 represents the cell state at time t-1, tanh represents the hyperbolic tangent activation function, ⊙ represents the element-by-element multiplication, W C represents the weight matrix for updating the cell state, b C represents the bias of the cell state;

[0031] The output gate of the LSTM model is expressed as:

[0032] o t =σ(W o [h t-1 ,x t ]+b o );

[0033] Among them, t represents the activation value of the output gate, W o represents the weight matrix of the input gate, b o is the bias term of the input gate;

[0034] The hidden state of the LSTM model at time t is expressed as:

[0035] h t =o t ⊙tanh(c t );

[0036] Among them, h t represents the hidden state at time t;

[0037] Based on the prediction values ​​of n LSTM models, the average weighted method is used to fuse the final prediction value:

[0038]

[0039] Among them, y represents the final predicted value, y i Represents the predicted value of the i-th LSTM model.

[0040] As a preferred implementation, the method for predicting the power market demand is:

[0041]

[0042] Among them, φ represents the forecast value of electricity market demand, ω F ,ω K represents the preset weight, H(x,x i ) represents the kernel function, x represents the input data set, x i represents the input data set x support vector, T the number of decision trees, f t (x) represents the predicted value of the t-th decision tree in the random forest for the input data set x, N represents the number of support vectors in the support vector machine, and ai , represents the solution of the support vector regression optimization problem, and δ represents the bias term.

[0043] As a preferred implementation, the power abandonment calculation model is expressed as:

[0044] E γ =max(0,G-min(C δ ,D(P θ )));

[0045] Among them, E γ represents the amount of power abandoned, G represents the amount of power generated, C δ represents the grid constraint, P θ represents the electricity price, D(P θ ) represents the electricity market demand, max represents the maximum function, and min represents the minimum function.

[0046] As a preferred implementation, the grid constraints include transmission capacity constraints, voltage stability constraints, frequency stability constraints, power flow constraints, and short-circuit current constraints;

[0047] The method for determining the grid constraints includes calculation of physical characteristics, simulation analysis, analysis of historical data, and reference to grid operation standards and specifications.

[0048] As a preferred implementation, the optimization strategy includes equipment operation and maintenance optimization strategy, power grid dispatch optimization strategy and market mechanism reform strategy.

[0049] On the other hand, the present invention further provides a new energy power station power abandonment analysis system in a power market environment, comprising:

[0050] Data fusion module: collects new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrates them into a data set;

[0051] Power generation prediction module: Use the data set and LSTM model to predict the power generation of new energy power stations;

[0052] Power market demand forecasting module: determines grid constraints and forecasts power market demand based on data sets;

[0053] Abandoned power calculation model building module: establish an abandoned power calculation model based on grid constraints, power market demand and power generation, and obtain the abandoned power of new energy power stations through the abandoned power calculation model;

[0054] Strategy optimization module: formulates an optimization strategy for the new energy power station according to the amount of abandoned electricity.

[0055] The present invention has the following beneficial effects:

[0056] The present invention forms a comprehensive and accurate data set by collecting new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data, and performing normalization processing and data fusion. This data fusion method not only improves the reliability and accuracy of the data, but also provides a solid foundation for the subsequent prediction of new energy power generation and power market demand. The LSTM model is used to predict the power generation of new energy power stations. The LSTM model has the ability to process time series data and can capture the randomness and volatility of new energy power generation, thereby improving the accuracy of the prediction. At the same time, the present invention also uses an average weighted method to fuse multiple LSTM model prediction values, further improving the robustness and accuracy of the prediction. In addition, for the prediction of power market demand, the present invention also uses advanced prediction methods, combined with algorithms such as random forests and support vector machines, to improve the accuracy of the prediction. A power abandonment calculation model is established based on power grid constraints, power market demand and power generation, and the model can accurately calculate the power abandonment of new energy power stations. At the same time, the model also takes into account multiple aspects of power grid constraints, such as transmission capacity constraints, voltage stability constraints, etc., so that the calculation results are more in line with the actual situation. This accuracy and practicality make the present invention have a high application value in the analysis of power abandonment in new energy power stations. The present invention formulates an optimization strategy for new energy power stations based on the amount of abandoned power, including equipment operation and maintenance optimization strategy, power grid dispatch optimization strategy and market mechanism reform strategy. These strategies are highly targeted and can effectively solve the problem of power abandonment in new energy power stations. For example, the equipment operation and maintenance optimization strategy can improve the reliability and stability of the equipment and reduce power abandonment caused by equipment failure; the power grid dispatch optimization strategy can optimize the dispatch strategy of the power grid and improve the absorption capacity of new energy; the market mechanism reform strategy can promote the healthy development of the power market and improve the economic benefits of new energy power generation companies. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The present invention is a flowchart for implementing the method. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0060] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0061] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0062] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0063] Embodiment 1:

[0064] See also Figure 1 The present invention provides a method for analyzing power abandonment of a new energy power station in a power market environment, which is characterized by comprising the following steps:

[0065] Collect new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrate them into a data set;

[0066] Renewable energy power generation data mainly includes data directly related to the renewable energy power generation process, such as:

[0067] Wind speed and direction: For wind power generation, wind speed and direction are key factors in determining power generation. These data are usually collected by meteorological instruments such as anemometers and wind vanes to analyze the availability of wind resources and power generation potential.

[0068] Irradiance: For solar power generation, irradiance (i.e. solar radiation intensity) is an important indicator for evaluating the efficiency of solar power generation. Irradiance data is usually collected by solar radiation monitoring stations and used to predict solar power generation.

[0069] Power generation and electricity: The actual power generation and electricity output of new energy power generation equipment are important bases for evaluating power generation efficiency and equipment operation status. These data are usually collected in real time by the monitoring system of the power generation equipment.

[0070] The power grid operation data mainly reflects the real-time operation status of the power grid, such as:

[0071] Voltage and current: The voltage and current data of each node in the power grid are key indicators for evaluating the stability and operating efficiency of the power grid. These data are usually collected in real time by the power grid monitoring system to monitor the operating status of the power grid.

[0072] Frequency: The frequency of the power grid reflects the balance between the power load and the power generation in the grid. Frequency data is usually collected in real time by the grid dispatch center to guide the dispatch and operation of the grid.

[0073] Grid load: Grid load data reflects the demand of power users in the grid. These data are usually collected and analyzed by the power load forecasting system to predict future power demand and provide reference for grid dispatching and operation.

[0074] Electricity market demand data mainly reflects the supply and demand situation of the electricity market, such as:

[0075] Electricity price: The price data of the electricity market reflects the changes in the relationship between electricity supply and demand. These data are usually released in real time by the power trading center to guide transactions and decision-making in the electricity market.

[0076] Power trading volume: Power trading volume data reflects the activity level and trading scale of the power market. These data are usually collected and released by the power trading center to analyze the operation status and trends of the power market.

[0077] Meteorological data is of great significance for the prediction and dispatch of renewable energy power generation, such as:

[0078] Temperature, humidity: Temperature and humidity data have a certain impact on evaluating the efficiency of solar power generation and the potential of wind power generation. These data are usually collected and analyzed by meteorological monitoring stations.

[0079] Precipitation: Precipitation data has an important impact on the evaluation of the operating status and power generation efficiency of new energy power generation equipment. Especially in wind power generation, excessive precipitation may affect the operation and power generation efficiency of wind turbines.

[0080] Equipment status data reflects the operating status and health of new energy power generation equipment, such as:

[0081] Wind turbine speed: Wind turbine speed data reflects the operating status and power generation efficiency of wind power generation equipment. These data are usually collected and analyzed in real time by the wind turbine monitoring system.

[0082] Inverter efficiency: Inverter efficiency data reflects the conversion efficiency and health status of the inverter in the solar power generation equipment. These data are usually collected and analyzed in real time by the inverter monitoring system.

[0083] Equipment failure and maintenance records: Equipment failure and maintenance record data reflects the reliability and maintenance status of renewable energy power generation equipment. These data are usually recorded and analyzed by the equipment maintenance management system.

[0084] Market policy data has an important impact on the development of renewable energy power generation and the electricity market, such as:

[0085] Electricity price policy: Electricity price policy determines the price level and structure of the electricity market. These data are usually formulated and published by government departments or electricity regulatory agencies.

[0086] Subsidy policy: Subsidy policy plays an important role in promoting the development of renewable energy power generation and the power market. These data are usually formulated and released by government departments or relevant institutions to support the construction and operation of renewable energy power generation and the power market.

[0087] Taxation and environmental protection policies: Taxation and environmental protection policies also have an impact on the development of new energy power generation and the electricity market. These policies are usually formulated and issued by government departments to guide and regulate the behavior and development direction of new energy power generation and the electricity market.

[0088] Use the data set and LSTM model to predict the power generation of new energy power plants;

[0089] Identify grid constraints and forecast electricity market demand based on data sets;

[0090] Establish a power curtailment calculation model based on grid constraints, power market demand, and power generation, and obtain the power curtailment of new energy power stations through the power curtailment calculation model;

[0091] An optimization strategy for the new energy power station is formulated based on the amount of abandoned electricity.

[0092] As a preferred implementation, the data set fusion method is:

[0093] Normalize new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data:

[0094]

[0095] Among them, x new represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data after normalization, x represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x min represents the minimum value of the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization. max Indicates the maximum value of renewable energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization;

[0096] The normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data are integrated:

[0097]

[0098] Among them, D clean represents the fused data set, α i represents the weight, β i represents the reliability factor, θ i From i=0 to i=n, ​​they represent respectively the normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data.

[0099] By collecting new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data, and performing normalization and data fusion, a comprehensive and accurate data set is formed. This data fusion method not only improves the reliability and accuracy of the data, but also provides a solid foundation for subsequent new energy power generation forecasts and power market demand forecasts.

[0100] The LSTM model solves the problem of gradient vanishing or gradient exploding that is prone to occur in traditional RNNs when processing long sequences by introducing a "gate" mechanism (forget gate, input gate, output gate). These "gate" structures can control the flow of information, allowing LSTM to capture long-term dependencies more effectively.

[0101] As a preferred implementation, the method for predicting the power generation of the new energy power station is:

[0102] According to the LSTM model, n LSTM model prediction values ​​are predicted. The prediction method of the LSTM model is:

[0103] The forget gate of the LSTM model is expressed as:

[0104] f t =σ(W f [h t-1 ,x t ]+b f );

[0105] Among them, f t represents the activation value of the forget gate, W f represents the weight matrix of the forget gate, b f represents the bias term of the forget gate, σ represents the activation function, and h t-1 represents the hidden state at time t-1, x t Represents the data set at time t;

[0106] The LSTM model input gate is expressed as:

[0107] i t =σ(W i [h t-1 ,x t ]+b i );

[0108] Among them, W i represents the weight matrix of the input gate, b i represents the bias term of the input gate, i t represents the activation value of the input gate;

[0109] The cell state of the LSTM model at time t is expressed as:

[0110] c t =f t ⊙c t-1 +i t ⊙tanh(W c [h t-1 ,x t ]+b c );

[0111] Among them, c t represents the cell state at time t, c t-1 represents the cell state at time t-1, tanh represents the hyperbolic tangent activation function, ⊙ represents the element-by-element multiplication, W C represents the weight matrix for updating the cell state, b C represents the bias of the cell state;

[0112] The output gate of the LSTM model is expressed as:

[0113] o t =σ(W o [h t-1 ,x t ]+b o );

[0114] Among them, t represents the activation value of the output gate, W o represents the weight matrix of the input gate, b o is the bias term of the input gate;

[0115] The hidden state of the LSTM model at time t is expressed as:

[0116] h t =o t ⊙tanh(c t );

[0117] Among them, h i represents the hidden state at time t;

[0118] Based on the prediction values ​​of n LSTM models, the average weighted method is used to fuse the final prediction value:

[0119]

[0120] Among them, y represents the final predicted value, y i Represents the predicted value of the i-th LSTM model.

[0121] As a preferred implementation, the method for predicting the power market demand is:

[0122]

[0123] in, represents the forecast value of electricity market demand, ω F ,ω K represents the preset weight, H(x,x i ) represents the kernel function, x represents the input data set, x i represents the input data set x support vector, T the number of decision trees, f t (x) represents the predicted value of the t-th decision tree in the random forest for the input data set x, N represents the number of support vectors in the support vector machine, and a i , represents the solution of the support vector regression optimization problem, and δ represents the bias term.

[0124] The LSTM model is used to predict the power generation of new energy power stations. The LSTM model has the ability to process time series data and can capture the randomness and volatility of new energy power generation, thereby improving the accuracy of the prediction. At the same time, the present invention also uses an average weighted method to fuse the predicted values ​​of multiple LSTM models, further improving the robustness and accuracy of the prediction. In addition, for the prediction of power market demand, the present invention also uses advanced prediction methods, combining algorithms such as random forests and support vector machines to improve the accuracy of the prediction.

[0125] As a preferred implementation, the power abandonment calculation model is expressed as:

[0126] E γ =max(0,G-min(C δ ,D(P θ )));

[0127] Among them, E γ represents the amount of power abandoned, G represents the amount of power generated, C δ represents the grid constraint, P θ represents the electricity price, D(P θ) represents the electricity market demand, max represents the maximum function, and min represents the minimum function.

[0128] A power curtailment calculation model was established based on grid constraints, power market demand, and power generation. The model can accurately calculate the amount of power curtailment in new energy power stations. At the same time, the model also takes into account multiple aspects of grid constraints, such as transmission capacity constraints, voltage stability constraints, etc., so that the calculation results are more in line with the actual situation. This accuracy and practicality make the present invention have high application value in the analysis of power curtailment in new energy power stations.

[0129] As a preferred implementation, the grid constraints include transmission capacity constraints, voltage stability constraints, frequency stability constraints, power flow constraints, and short-circuit current constraints;

[0130] Transmission capacity constraints: Each transmission line and transformer in the power grid has its rated transmission capacity. When power transmission exceeds this capacity, it may cause equipment overheating, damage, or even cause power grid failure.

[0131] Voltage stability constraints: The voltage level in the power grid needs to be kept within a certain range to ensure the normal operation of power equipment and the quality of power consumption by users. Voltage that is too high or too low may damage equipment or cause power outages.

[0132] Frequency stability constraint: The frequency of the power grid (usually 50Hz or 60Hz) needs to be stable to ensure the normal operation of equipment such as motors. Frequency fluctuations may be caused by an imbalance between generation and load.

[0133] Flow constraints: The flow of electricity in the power grid needs to meet certain conditions to ensure the safe operation of the power grid. This includes the direction, size and distribution of the flow.

[0134] Short-circuit current constraint: When a short circuit occurs in the power grid, a huge short-circuit current will be generated. This current needs to be limited to a certain range to protect power grid equipment and personal safety.

[0135] The method for determining the grid constraints includes calculation of physical characteristics, simulation analysis, analysis of historical data, and review of grid operation standards and specifications;

[0136] Calculation of physical characteristics: By analyzing the physical characteristics of power grid equipment (such as resistance, inductance, capacitance of conductors, etc.) and the operating parameters of the power grid (such as voltage, current, frequency, etc.), the transmission capacity, voltage stability and other constraints of the power grid can be calculated.

[0137] Simulation analysis: Using power system simulation software (such as PSASP, PSCAD / EMTDC, etc.), the operation of the power grid under various working conditions can be simulated to obtain the power grid's flow constraints, frequency stability constraints, etc.

[0138] Analyze historical data: By analyzing historical data, various parameters and characteristics of the power grid under normal operation and fault conditions can be obtained, thereby determining the constraints of the power grid. This method is usually used for long-term monitoring and analysis of the power grid.

[0139] Consult grid operation standards and specifications: According to the grid operation standards and specifications formulated by the State Grid Corporation of China or other relevant institutions, the constraints of the grid can be determined. These standards and specifications usually take into account the requirements of the safety, reliability and economy of the grid.

[0140] As a preferred implementation, the optimization strategy includes equipment operation and maintenance optimization strategy, power grid dispatch optimization strategy and market mechanism reform strategy;

[0141] Equipment operation and maintenance optimization strategies include preventive maintenance, upgrades and modifications;

[0142] Preventive maintenance: Through continuous monitoring and analysis of equipment status data, potential equipment failures can be predicted, and maintenance and repairs can be performed in advance to reduce power generation interruptions and power abandonment caused by equipment failures.

[0143] Upgrade and transformation: Upgrade or replace old equipment with more efficient and reliable new equipment to improve power generation efficiency and equipment reliability.

[0144] Grid dispatch optimization strategies include flexible dispatch and cross-region dispatch;

[0145] Flexible scheduling: According to the power generation forecast of the new energy power station and the actual operation of the power grid, the scheduling plan is flexibly adjusted to ensure that the power generation of the new energy power station can be accepted by the power grid to the greatest extent possible.

[0146] Cross-regional dispatching: Utilize the cross-regional transmission capacity of the power grid to transmit surplus electricity from new energy power stations to other areas with greater electricity demand, thereby reducing local power abandonment.

[0147] Market mechanism reform strategies include green certificate trading, quota system and penalty mechanism;

[0148] Green certificate trading: Establish a green certificate trading mechanism to encourage renewable energy power generation companies to obtain additional income by selling green certificates, thereby increasing their enthusiasm for power generation.

[0149] Quota system and penalty mechanism: Set up a renewable energy power quota system and punish power companies and users that fail to meet their quotas to promote the consumption of renewable energy.

[0150] Embodiment 2:

[0151] The present invention also provides a new energy power station power abandonment analysis system in a power market environment, comprising:

[0152] Data fusion module: collects new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrates them into a data set;

[0153] Power generation prediction module: Use the data set and LSTM model to predict the power generation of new energy power stations;

[0154] Power market demand forecasting module: determines grid constraints and forecasts power market demand based on data sets;

[0155] Abandoned power calculation model building module: establish an abandoned power calculation model based on grid constraints, power market demand and power generation, and obtain the abandoned power of new energy power stations through the abandoned power calculation model;

[0156] Strategy optimization module: formulates an optimization strategy for the new energy power station according to the amount of abandoned electricity.

[0157] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0158] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0159] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0160] In several embodiments provided in the present application, if any 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 application, 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, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0161] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for analyzing power abandonment of new energy power stations in a power market environment, characterized in that: The following steps are involved: Collect new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrate them into a data set; Use the data set and LSTM model to predict the power generation of new energy power plants; Identify grid constraints and forecast electricity market demand based on data sets; Establish a power curtailment calculation model based on grid constraints, power market demand, and power generation, and obtain the power curtailment of new energy power stations through the power curtailment calculation model; An optimization strategy for the new energy power station is formulated based on the amount of abandoned electricity.

2. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 1 is characterized in that: The fusion method of the data set is: Normalize new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data: Among them, x new represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data after normalization, x represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x min represents the minimum value of the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x max Indicates the maximum value of renewable energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization; The normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data are integrated: Among them, D clean represents the fused data set, α i represents the weight, β i represents the reliability factor, θ i From i=0 to i=n, ​​they represent respectively the normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data.

3. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 1 is characterized in that: The prediction method of the power generation of the new energy power station is: According to the LSTM model, n LSTM model prediction values ​​are predicted. The prediction method of the LSTM model is: The forget gate of the LSTM model is expressed as: f t =σ(W f [h t-1 ,x t ]+b f ); Among them, f t represents the activation value of the forget gate, W f represents the weight matrix of the forget gate, b f represents the bias term of the forget gate, σ represents the activation function, and h t-1 represents the hidden state at time t-1, x t Represents the data set at time t; The LSTM model input gate is expressed as: i t =σ(W i [h t-1 ,x t ]+b i ); Among them, W i represents the weight matrix of the input gate, b i represents the bias term of the input gate, i t represents the activation value of the input gate; The cell state of the LSTM model at time t is expressed as: c t =f t ⊙c t-1 +i t ⊙tanh(W c [h t-1 ,x t ]+b c ); Among them, c t represents the cell state at time t, c t-1 represents the cell state at time t-1, tanh represents the hyperbolic tangent activation function, ⊙ represents the element-by-element multiplication, W c represents the weight matrix for updating the cell state, b c represents the bias of the cell state; The output gate of the LSTM model is expressed as: the t =σ(W o [h t-1 ,x t ]+b o ); Among them, t represents the activation value of the output gate, W o represents the weight matrix of the input gate, b o is the bias term of the input gate; The hidden state of the LSTM model at time t is expressed as: h t =o t ⊙tanh(c t ); Among them, h t represents the hidden state at time t; Based on the prediction values ​​of n LSTM models, the average weighted method is used to fuse the final prediction value: Among them, y represents the final predicted value, y i Represents the predicted value of the i-th LSTM model.

4. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 1 is characterized in that: The forecasting method of the electricity market demand is: in, represents the forecast value of electricity market demand, ω F ,ω K represents the preset weight, H(x, x i ) represents the kernel function, x represents the input data set, x i represents the input data set x support vector, T the number of decision trees, f t (x) represents the predicted value of the t-th decision tree in the random forest for the input data set x, N represents the number of support vectors in the support vector machine, and a i , represents the solution of the support vector regression optimization problem, and δ represents the bias term.

5. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 1 is characterized in that: The power curtailment calculation model is expressed as: E γ =max(0,G-min(C δ ,D(P θ ))); Among them, E γ represents the amount of power abandoned, G represents the amount of power generated, C δ represents the grid constraint, P θ represents the electricity price, D(P θ ) represents the electricity market demand, max represents the maximum function, and min represents the minimum function.

6. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 5 is characterized in that: The grid constraints include transmission capacity constraints, voltage stability constraints, frequency stability constraints, power flow constraints and short-circuit current constraints; The method for determining the grid constraints includes calculation of physical characteristics, simulation analysis, analysis of historical data, and reference to grid operation standards and specifications.

7. The method for analyzing power abandonment of new energy power stations in the power market environment according to claim 1 is characterized in that: The optimization strategies include equipment operation and maintenance optimization strategy, power grid dispatch optimization strategy and market mechanism reform strategy.

8. A new energy power station power abandonment analysis system in the power market environment, characterized in that: include: Data fusion module: collects new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data of new energy power stations, and integrates them into a data set; Power generation prediction module: Use the data set and LSTM model to predict the power generation of new energy power stations; Power market demand forecasting module: determines grid constraints and forecasts power market demand based on data sets; Abandoned power calculation model building module: establish an abandoned power calculation model based on grid constraints, power market demand and power generation, and obtain the abandoned power of new energy power stations through the abandoned power calculation model; Strategy optimization module: formulates an optimization strategy for the new energy power station according to the amount of abandoned electricity.

9. The new energy power station power abandonment analysis system in the power market environment according to claim 8 is characterized in that: The data fusion module uses the data set fusion method as follows: Normalize new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data: Among them, x new represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data after normalization, x represents the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x min represents the minimum value of the new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization, x max Indicates the maximum value of renewable energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data or market policy data before normalization; The normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data are integrated: Among them, D clean represents the fused data set, α i represents the weight, β i represents the reliability factor, θ i From i=0 to i=n, ​​they represent respectively the normalized new energy power generation data, power grid operation data, power market demand data, meteorological data, equipment status data and market policy data.

10. The new energy power station power abandonment analysis system in the power market environment according to claim 8, characterized in that: The power generation prediction module uses a method for predicting the power generation of a new energy power station as follows: According to the LSTM model, n LSTM model prediction values ​​are predicted. The prediction method of the LSTM model is: The forget gate of the LSTM model is expressed as: f t =σ(W f [h t-1 ,x t ]+b f ); Among them, f t represents the activation value of the forget gate, W f represents the weight matrix of the forget gate, b f represents the bias term of the forget gate, σ represents the activation function, and h t-1 represents the hidden state at time t-1, x t Represents the data set at time t; The LSTM model input gate is expressed as: i t =σ(W i [h t-1 ,x t ]+b i ); Among them, W i represents the weight matrix of the input gate, b i represents the bias term of the input gate, i t represents the activation value of the input gate; The cell state of the LSTM model at time t is expressed as: c t =f t ⊙c t-1 +i t ⊙tanh(W c [h t-1 ,x t ]+b c ); Among them, c t represents the cell state at time t, c t-1 represents the cell state at time t-1, tanh represents the hyperbolic tangent activation function, ⊙ represents the element-by-element multiplication, W c represents the weight matrix for updating the cell state, b c represents the bias of the cell state; The output gate of the LSTM model is expressed as: the t =σ(W o [h t-1 ,x t ]+b o ); Among them, t represents the activation value of the output gate, W o represents the weight matrix of the input gate, b o is the bias term of the input gate; The hidden state of the LSTM model at time t is expressed as: h t =o t ⊙tanh(c t ); Among them, h t represents the hidden state at time t; Based on the prediction values ​​of n LSTM models, the average weighted method is used to fuse the final prediction value: Among them, y represents the final predicted value, y i Represents the predicted value of the i-th LSTM model.

Images