Electric power resource scheduling method and device suitable for extreme weather, terminal equipment and storage medium

Abstract

The invention discloses an electric power resource scheduling method and device suitable for extreme weather, terminal equipment and a storage medium, and belongs to the technical field of electric power scheduling, and the method adopts nonparametric kernel density estimation to construct a wind and light output joint probability distribution function representing output characteristics of wind power output and photovoltaic output in extreme weather. When flexibility demand analysis is carried out, the advantages of nonlinear correlation between variables and tail risks can be analyzed by means of a joint distribution probability function, wind power prediction errors and photovoltaic prediction errors are accurately analyzed, and then adjustment demand risks of net load flexibility demands are quantified in combination with convolution operation. The problems that the joint probability distribution characteristics of wind and light output in extreme weather cannot be accurately described and the regulation demand risk cannot be ignored in the current flexibility demand evaluation method depending on the normal distribution independent assumption are solved. And the power supply reliability and the operation stability of the power system in extreme weather can be ensured by the formulated power dispatching plan.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching, and particularly to a power resource dispatching method, device, terminal device and storage medium applicable to extreme weather. Background Art

[0002] In modern power systems, with the increase in the grid connection rate of new wind and solar energy generating units, the power supply of the power system is affected by natural environmental factors. Due to the occasional and random nature of natural environmental factors, there will always be a certain error in predicting the output of new wind and solar energy generating units. Therefore, when formulating the system resource dispatching plan at present, the method of flexibility demand assessment is usually adopted. Flexibility demand assessment is a process of quantitatively analyzing and planning the ability of the power system to cope with supply-demand fluctuations and prediction errors. Its core goal is to ensure the safe, reliable and economic operation of the power system under the conditions of high proportion of renewable energy access, load fluctuations and equipment failures by evaluating the adequacy of system flexibility resources (such as power generation side, grid side, energy storage and demand side resources).

[0003] When formulating the system resource dispatching process at present, the method of combining wind power output prediction error and load prediction error is usually adopted for flexibility demand assessment, and then the power system resource dispatching plan is formulated according to the assessment results to ensure the stability of power supply and the stable operation of the power system.

[0004] The current conventional method of flexibility demand assessment is based on the independent assumption of the normal distribution of wind power output prediction error and load prediction error, and calculates the net load fluctuation range through linear superposition. However, in extreme weather, this flexibility demand assessment method relying on the independent assumption of normal distribution cannot accurately describe the joint probability distribution characteristics of wind power output in extreme weather, and ignores the regulation demand risk in extreme scenarios, ultimately resulting in the formulated power dispatching plan being unable to ensure the power supply reliability and operation stability of the power system in extreme weather. Summary of the Invention

[0005] The present invention provides a power resource dispatching method, device, terminal device and storage medium applicable to extreme weather, which overcomes the problems that the current flexibility demand assessment method relying on the independent assumption of normal distribution cannot accurately describe the joint probability distribution characteristics of wind power output in extreme weather and ignores the regulation demand risk.

[0006] An embodiment of the present invention provides a power resource dispatching method applicable to extreme weather, including:

[0007] Obtain the predicted wind power output value, predicted photovoltaic output value, predicted load value, actual wind power output value, actual photovoltaic output value, and actual load value of each moment of the power system in extreme weather;

[0008] Construct a net load prediction curve of the power system according to the load prediction values, the wind power output prediction values, and the photovoltaic power output prediction values at each moment;

[0009] Calculate the load prediction error at each moment in extreme weather according to the load prediction value and the actual load value, and calculate the wind power prediction error and the photovoltaic power prediction error at each moment in extreme weather based on the pre-constructed joint probability distribution function of wind and photovoltaic power outputs, according to the wind power output prediction value, the photovoltaic power output prediction value, the actual wind power output value, and the actual photovoltaic power output value; wherein, the joint probability distribution function of wind and photovoltaic power outputs is a probability distribution function constructed by pre-using non-parametric kernel density estimation to fit the wind power output and the photovoltaic power output in extreme weather, and is used to characterize the output characteristics of the wind power output and the photovoltaic power output in extreme weather;

[0010] Adopt convolution operation to generate the net load flexibility demand and the corresponding regulation demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error;

[0011] Generate the target resource scheduling plan at each moment of the power system in extreme weather according to the net load flexibility demand and the corresponding regulation demand risk value.

[0012] Further, the obtaining of the wind power output prediction value and the photovoltaic power output prediction value at each moment of the power system in extreme weather includes:

[0013] Obtain several types of meteorological forecast data groups at each moment in extreme weather; wherein, each type of meteorological forecast data group is composed of meteorological forecast data provided by several different meteorological data sources;

[0014] Based on the meteorological data sources corresponding to the meteorological forecast data, determine the weights of the meteorological forecast data in each meteorological forecast data group;

[0015] According to the weights, perform weighted fusion on the meteorological forecast data in each meteorological forecast data group to generate the meteorological characteristic value corresponding to each meteorological forecast data group;

[0016] Input the meteorological characteristic value into a preset wind and photovoltaic power output prediction model, so that the wind and photovoltaic power output prediction model outputs the wind power output prediction value and the photovoltaic power output prediction value at each moment of the power system in extreme weather; wherein, the wind and photovoltaic power output prediction model is a deep learning model trained by using the historical meteorological characteristics of historical extreme weather, the historical wind power output values, and the historical photovoltaic power output values.

[0017] Further, the construction of the joint probability distribution function of wind and photovoltaic power outputs includes:

[0018] Obtain the historical meteorological characteristics, historical wind power output values, and historical photovoltaic output values of the power system under historical extreme weather;

[0019] Adopt the non-parametric kernel density estimation method to construct the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic processing according to the historical meteorological characteristics and the corresponding historical wind power output values and historical photovoltaic output values;

[0020] Adopt the Frank-Copula function to construct the cumulative probability distribution function of wind-solar power output according to the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic processing;

[0021] Construct the joint probability distribution function of wind-solar power output according to the cumulative probability distribution function of wind-solar power output.

[0022] Further, based on the pre-constructed joint probability distribution function of wind-solar power output, calculate the wind power prediction error and the photovoltaic prediction error at each moment under extreme weather according to the wind power output prediction value, the photovoltaic output prediction value, the actual wind power output value, and the actual photovoltaic output value, including:

[0023] Adopt the following formula to calculate the wind power prediction error and the photovoltaic prediction error at each moment under extreme weather according to the wind power output prediction value, the photovoltaic output prediction value, the actual wind power output value, and the actual photovoltaic output value:

[0024]

[0025]

[0026] where, ΔP w,t is the wind power prediction error at time t, is the actual wind power output value at time t, is the wind power output prediction value at time t, F w,s (p w , p s ) is the joint probability distribution function of wind-solar power output, ΔP s,t is the photovoltaic prediction error at time t, is the actual photovoltaic output value at time t, is the photovoltaic output prediction value at time t.

[0027] Further, perform convolution operation, and generate the net load flexibility demand and the corresponding target probability density function at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic prediction error, including:

[0028] Obtain the first error probability density corresponding to the load prediction error;

[0029] According to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic prediction error, synthesize the net load flexibility regulation demand of the power system;

[0030] According to the joint probability distribution function of wind and solar power output, calculate the second error probability density corresponding to the negative value of the sum of the wind power prediction error and the photovoltaic prediction error;

[0031] Adopt convolution operation to construct the target probability density function according to the net load flexibility regulation demand, the first error probability density, and the second error probability density;

[0032] According to the net load flexibility demand and the target probability density function, determine the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather.

[0033] Further, according to the net load flexibility demand and the corresponding regulation demand risk value, generate the target resource scheduling plan at each moment of the power system under extreme weather, including:

[0034] Obtain the supply adjustment intervals of several scheduling resources used to participate in power scheduling in the power system;

[0035] According to the preset demand response flexibility supply capacity model, calculate the adjustable transfer load and the adjustable curtailment load at each moment of the power system;

[0036] According to the supply adjustment interval, the adjustable transfer load, the adjustable curtailment load, and the regulation demand risk value, construct a resource scheduling model with the goal of minimizing the operating cost and the flexibility regulation demand expectation;

[0037] Under the preset constraint conditions for ensuring the stable operation of the power system, solve the resource scheduling model to generate the target resource scheduling plan at each moment of the power system under extreme weather.

[0038] Further, the regulation demand risk value includes: value at risk and conditional value at risk;

[0039] The determining the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather according to the net load flexibility demand and the target probability density function includes:

[0040] Discretize the net load flexibility adjustment demand, and calculate the target probability that the net load flexibility adjustment demand at each moment of the power system is equal to the corresponding discretized value;

[0041] Generate the flexibility adjustment demand expectation according to the target probability, the net load flexibility adjustment demand, and the probability density function;

[0042] Calculate the value at risk of the net load flexibility adjustment demand at each moment of the power system under a preset confidence level according to the target probability;

[0043] Calculate the conditional value at risk of the expected loss of the tail extreme risk corresponding to each of the values at risk according to the value at risk and the net load flexibility adjustment demand.

[0044] Another embodiment of the present invention also provides a power resource scheduling device suitable for extreme weather, including:

[0045] An output data acquisition module for acquiring the predicted values of wind power output, the predicted values of photovoltaic power output, the predicted values of load, the actual values of wind power output, the actual values of photovoltaic power output, and the actual values of load at each moment of the power system under extreme weather;

[0046] A net load calculation module for constructing a net load prediction curve of the power system according to the predicted value of load, the predicted value of wind power output, and the predicted value of photovoltaic power output at each moment;

[0047] A prediction error calculation module for calculating the load prediction error at each moment under extreme weather according to the predicted value of load and the actual value of load, and calculating the wind power prediction error and the photovoltaic prediction error at each moment under extreme weather based on a pre-constructed joint probability distribution function of wind and light output according to the predicted value of wind power output, the predicted value of photovoltaic power output, the actual value of wind power output, and the actual value of photovoltaic power output; wherein, the joint probability distribution function of wind and light output is a probability distribution function constructed by pre-using non-parametric kernel density estimation to fit the wind power output and the photovoltaic power output under extreme weather, and is used to characterize the output characteristics of the wind power output and the photovoltaic power output under extreme weather;

[0048] A demand risk analysis module for performing convolution operation and generating the net load flexibility demand and the corresponding adjustment demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic prediction error;

[0049] A scheduling plan formulation module for generating a target resource scheduling plan at each moment of the power system under extreme weather according to the net load flexibility demand and the corresponding adjustment demand risk value.

[0050] Another embodiment of the present invention further provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of a power resource scheduling method applicable to extreme weather as described in any one of the above embodiments of the present invention are implemented.

[0051] Another embodiment of the present invention further provides a computer-readable storage medium item, including: a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the steps of a power resource scheduling method applicable to extreme weather as described in any one of the above embodiments of the present invention.

[0052] By implementing the present invention, the following beneficial effects are achieved:

[0053] The present invention discloses a power resource scheduling method, device, terminal device, and storage medium applicable to extreme weather, belonging to the technical field of power dispatching. The method constructs a joint probability distribution function of wind and light output characteristics representing wind power output and photovoltaic power output under extreme weather by using non-parametric kernel density estimation. When performing flexibility demand analysis, by virtue of the advantage of the joint distribution probability function in analyzing the non-linear correlation and tail risk between variables, the wind power prediction error and the photovoltaic prediction error are accurately analyzed, and then combined with convolution operation to quantify the adjustment demand risk of the net load flexibility demand, overcoming the problem that the current flexibility demand assessment method relying on the independent assumption of normal distribution cannot accurately characterize the joint probability distribution characteristics of wind and light output in extreme weather and ignores the adjustment demand risk, so that the formulated power dispatching plan can ensure the power supply reliability and operation stability of the power system under extreme weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the present application, the drawings required for implementation will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0055] Figure 1 is a flowchart of a power resource scheduling method applicable to extreme weather provided by an embodiment of the present invention;

[0056] Figure 2 is a structural diagram of a power resource scheduling device applicable to extreme weather provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following will clearly and completely describe the technical solutions in this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the description of the specification, claims, and above-mentioned drawings of this application are intended to cover non-exclusive inclusion.

[0059] In the description of the embodiments of this application, technical terms such as "first" and "second" are only used to distinguish different objects and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order, or primary-secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "a plurality of" is more than two, unless otherwise specifically defined.

[0060] Referring to "embodiments" herein means that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of this application. The phrase appearing at various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0061] In the description of the embodiments of this application, the term "and / or" is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0062] In the description of the embodiments of this application, the term "a plurality of" means more than two (including two). Similarly, "a plurality of groups" means more than two groups (including two groups), and "a plurality of pieces" means more than two pieces (including two pieces).

[0063] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installation", "connection", "connection", "fixation", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can also be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and can be the communication inside two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific situations.

[0064] See Figure 1 , to solve the problem that the current flexibility demand assessment method relying on the independent assumption of normal distribution cannot accurately describe the joint probability distribution characteristics of wind and light output in extreme weather and ignores the risk of regulation demand, an embodiment of the present invention provides a power resource scheduling method applicable to extreme weather, including:

[0065] S1. Obtain the predicted values of wind power output, photovoltaic power output, load prediction, actual wind power output, actual photovoltaic power output, and actual load of the power system at each moment in extreme weather;

[0066] Preferably, the obtaining of the predicted values of wind power output and photovoltaic power output of the power system at each moment in extreme weather includes:

[0067] S11. Obtain several types of meteorological forecast data groups at each moment in extreme weather; wherein, each type of meteorological forecast data group is composed of meteorological forecast data provided by several different meteorological data sources;

[0068] S12. Based on the meteorological data sources corresponding to each meteorological forecast data, determine the weights of each meteorological forecast data in each meteorological forecast data group;

[0069] S13. According to the weights, perform weighted fusion on the meteorological forecast data in each meteorological forecast data group to generate the meteorological characteristic value corresponding to each meteorological forecast data group;

[0070] S14. Input the meteorological characteristic value into a preset wind and light output prediction model, so that the wind and light output prediction model outputs the predicted values of wind power output and photovoltaic power output of the power system at each moment in extreme weather; wherein, the wind and light output prediction model is a deep learning model trained using the historical meteorological characteristics, historical wind power output values, and historical photovoltaic power output values of historical extreme weather.

[0071] In a preferred embodiment of the present invention, through multi-source meteorological data weighted fusion and constructing a LightGBM-LSTM hybrid model, the non-linear coupling relationship between wind-solar power output and meteorological elements under extreme weather is deeply mined, breaking through the limitations of traditional single data source and shallow model, and effectively reducing the prediction error.

[0072] Specifically, calculate the Pearson correlation coefficient between historical real data and forecast data according to meteorological monitoring data, and then determine the correlation weighting coefficient, that is, the weight, between the same type of meteorological forecast data among different data sources:

[0073]

[0074] In the formula: r ij is the Pearson correlation coefficient between the i-th meteorological data source and the j-th meteorological forecast data, represents the forecast value of the j-th meteorological forecast data of the i-th meteorological data source at the t-th moment, the average forecast value of the j-th meteorological forecast data of the i-th meteorological data source at all moments, the actual observed value of the j-th meteorological forecast data at the t-th moment, represents the average actual observed value of the j-th meteorological forecast data at all moments, n represents the total number of data sources, and W ij is the correlation weighting coefficient, that is, the weight, where i represents the meteorological data source and j represents the feature.

[0075] It should be noted that in this embodiment, the meteorological forecast data for photovoltaic power output prediction includes the following types: solar irradiance, cloud cover index, atmospheric transmittance, module temperature, ambient temperature; and the meteorological forecast data for wind power output prediction includes the following types: wind speed and direction, air density, vertical pressure gradient.

[0076] Furthermore, based on the physical meaning of wind-solar power generation and meteorological background knowledge, important features affecting power prediction are constructed. The features are input into LightGBM-LSTM for modeling, and finally the power prediction results are output. LightGBM uses a depth-first strategy for decision tree growth, and the LSTM deep learning model can be represented by the following formula. In this embodiment, to improve the prediction accuracy of the model, the input data of the model also includes physical features for predicting photovoltaic power output, photovoltaic panel attenuation coefficient, shading loss, inverter efficiency curve, and historical power sequence; physical features for predicting wind power output, yaw system response delay, and historical wind turbine power.

[0077] For the wind-solar power output prediction model based on the LightGBM-LSTM architecture, in the LSTM input layer, the input data x is constructed using the correlation weighting coefficient t :

[0078]

[0079] Among them: x t Input data of the fused feature vector, Are various important feature items, such as: Is the equivalent wind speed correction term, Is the air pressure correction.

[0080]

[0081] In the formula: f t Is the forget gate (controlling the retention ratio of historical information), i t Is the input gate (adjusting the adoption degree of new information), C t Is the candidate cell state (temporarily storing new feature information), C t Is the updated cell state (fusing historical and current information), o t Is the output gate (controlling the information output intensity), h t Is the hidden state (abstract feature passed to the next moment), W f W i W c W o Are weight matrices, b f b i b c b o Are bias terms.

[0082] It can be understood that the present invention constructs a wind-solar power output prediction model by weighted fusion of multi-source meteorological data and adopting a LightGBM-LSTM hybrid model architecture, deeply excavates the non-linear coupling relationship between wind-solar power output and meteorological elements under extreme weather, breaks through the limitations of traditional single data source and shallow model, and effectively reduces the prediction error.

[0083] S2. Construct a net load prediction curve of the power system according to the load prediction value, the wind power output prediction value, and the photovoltaic power output prediction value at each moment;

[0084] In a preferred embodiment of the present invention, a net load prediction curve is constructed based on the following formula:

[0085]

[0086] Among them: Is the net load, Is the load prediction value at time t, Are the wind power output prediction value and the photovoltaic power output prediction value under extreme weather respectively. Further, the load prediction value can be obtained through a classical GRU model.

[0087] S3. Calculate the load prediction error at each moment in extreme weather according to the load prediction value and the actual load value, and calculate the wind power prediction error and the photovoltaic power prediction error at each moment in extreme weather based on the pre-constructed joint probability distribution function of wind and photovoltaic power outputs. The joint probability distribution function of wind and photovoltaic power outputs is constructed by using non-parametric kernel density estimation to fit the wind power output and the photovoltaic power output in extreme weather, and is a probability distribution function used to characterize the output characteristics of wind power output and photovoltaic power output in extreme weather.

[0088] In a preferred embodiment of the present invention, the load prediction error follows a normal distribution with a mean of 0 and a variance of Therefore, the load prediction error at each moment in extreme weather is calculated by the following formula:

[0089]

[0090] In the formula: is the load prediction error at time t, is the actual load value at time t, is the load prediction value at time t, is the normal distribution.

[0091] Preferably, the construction of the joint probability distribution function of wind and photovoltaic power outputs includes:

[0092] S01. Obtain the historical meteorological characteristics, historical wind power output values, and historical photovoltaic power output values of the power system in historical extreme weather;

[0093] S02. Use the non-parametric kernel density estimation method to construct the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic power output according to the historical meteorological characteristics and the corresponding historical wind power output values and historical photovoltaic power output values;

[0094] S03. Use the Frank-Copula function to construct the cumulative probability distribution function of wind and photovoltaic power outputs according to the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic power output;

[0095] S04. Construct the joint probability distribution function of wind and photovoltaic power outputs according to the cumulative probability distribution function of wind and photovoltaic power outputs.

[0096] In a preferred embodiment of the present invention, to quantify the spatio-temporal correlation of wind and photovoltaic power outputs in extreme weather, a joint probability distribution function of wind and photovoltaic power outputs is established, and non-parametric kernel density estimation is used to fit the output characteristics in extreme weather.

[0097] Specifically, first, the non-parametric kernel density estimation method is used to establish the marginal probability distributions for wind power and photovoltaic power outputs respectively:

[0098] F w (p w ) = P(P w ≤ p w |X t );

[0099] F s (p s ) = P(P s ≤ p s |X t );

[0100] In the formula: F w is the wind power marginal probability distribution, p w is the specific value of the wind power output random variable, P w is the output sequence of the wind farm under historical extreme weather, P() is the non-parametric kernel density estimation calculation formula, X t is the meteorological feature vector (the meteorological feature value obtained in step S13), F s is the photovoltaic marginal probability distribution, p s is the specific value of the photovoltaic power output random variable, P s is the output sequence of the photovoltaic farm under historical extreme weather.

[0101] Secondly, based on the Frank-Copula function, the following cumulative probability distribution function for wind and photovoltaic power outputs is constructed:

[0102] F w,s (p w , p s ) = C θ (F w (p w ), F s (p s ));

[0103] Among them: F w,s is the cumulative probability distribution function for wind and photovoltaic power outputs, C θ is the Frank-Copula function.

[0104] Finally, the following joint probability distribution function for wind and photovoltaic power outputs is constructed:

[0105] f w,s (p w , p s ) = c θ (F w (p w ), F s (p s ))·fw (p w )·f s (p s );

[0106] Preferably, based on the pre - constructed combined probability distribution function of wind and photovoltaic power outputs, calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather according to the wind power prediction value, the photovoltaic power prediction value, the actual wind power value, and the actual photovoltaic power value, including:

[0107] S31. Use the following formula to calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather according to the wind power prediction value, the photovoltaic power prediction value, the actual wind power value, and the actual photovoltaic power value:

[0108]

[0109] where, ΔP w,t is the wind power prediction error at time t, is the actual wind power value at time t, is the wind power prediction value at time t, F w,s (p w ,p s ) is the combined probability distribution function of wind and photovoltaic power outputs, ΔP s,t is the photovoltaic power prediction error at time t, is the actual photovoltaic power output value at time t, is the photovoltaic power prediction value at time t.

[0110] S4. Use convolution operation to generate the net load flexibility demand and the corresponding regulation demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error;

[0111] Preferably, the use of convolution operation to generate the net load flexibility demand and the corresponding target probability density function at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error includes:

[0112] S41. Obtain the first error probability density corresponding to the load prediction error;

[0113] S42. Synthesize the net load flexibility regulation demand of the power system according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error;

[0114] In a preferred embodiment of the present invention, the net load flexibility regulation demand of the power system is synthesized by the following formula:

[0115]

[0116] S43. Calculate the second error probability density corresponding to the negative value of the sum of the wind power prediction error and the photovoltaic power prediction error according to the joint probability distribution function of the wind and light output.

[0117] S44. Use convolution operation to construct the target probability density function according to the net load flexibility regulation demand, the first error probability density, and the second error probability density.

[0118] In a preferred embodiment of the present invention, the target probability density function is synthesized by convolution operation:

[0119]

[0120] where: f D The target probability density function of the flexibility regulation demand is d represents the specific value of the net load flexibility regulation demand, and f ΔL is the first error probability density of the load prediction error determined according to the normal distribution in step S3, is the second error probability density of the negative value of the combined wind and light output prediction error (derived from the Copula joint distribution in step S04).

[0121] S45. Determine the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather according to the net load flexibility demand and the target probability density function.

[0122] Preferably, the regulation demand risk value includes: value at risk and conditional value at risk;

[0123] The determining the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather according to the net load flexibility demand and the target probability density function includes:

[0124] S451. Discretize the net load flexibility regulation demand and calculate the target probability that the net load flexibility regulation demand at each moment of the power system is equal to the corresponding discretized value;

[0125] S452. Generate the flexibility regulation demand expectation according to the target probability, the net load flexibility regulation demand, and the probability density function;

[0126] S453. Calculate the value at risk of the net load flexibility regulation demand at each moment of the power system under the preset confidence level according to the target probability;

[0127] S454. Calculate the conditional value at risk of the expected loss of the tail extreme risk corresponding to each of the risk values according to the risk value and the net load flexibility regulation requirement.

[0128] In a preferred embodiment of the present invention, in step S451, the target probability that the net load flexibility regulation requirement at each moment of the power system is equal to the corresponding discretized value is calculated by the following formula:

[0129]

[0130] Where: P r is the target probability that the flexibility regulation requirement is equal to the discretized value of the flexibility regulation requirement, d i,t is the discretized value of the flexibility regulation requirement, ΔP i,t is the discrete value of the wind and light output prediction error, and N is the total number of discretization intervals of the wind and light output prediction error.

[0131] Furthermore, the flexibility regulation requirement expectation can be expressed by the following formula:

[0132]

[0133] The regulation requirement risk value is quantified as the risk value and the conditional value at risk, which can be calculated by the following formula:

[0134]

[0135] In the formula: VaR α is the risk value, representing the maximum possible value of the regulation requirement at the confidence level α, d represents the specific value of the net load flexibility regulation requirement, α is the confidence level, and CVaR α is the conditional value at risk, reflecting the expected loss of the tail extreme risk.

[0136] It can be understood that the present invention accurately captures the peak and heavy tail characteristics of the wind and light output and the load prediction error through the non-parametric kernel density estimation and discretized convolution technology, improving the robustness of the flexibility demand probability distribution modeling, and is particularly applicable to extreme weather scenarios such as typhoons and blizzards.

[0137] S5. Generate the target resource scheduling plan for each moment of the power system under extreme weather according to the net load flexibility requirement and the corresponding regulation requirement risk value.

[0138] Preferably, generating the target resource scheduling plan for each moment of the power system under extreme weather according to the net load flexibility requirement and the corresponding regulation requirement risk value includes:

[0139] S51. Obtain the supply regulation intervals of several scheduling resources used to participate in power scheduling in the power system;

[0140] In a preferred embodiment of the present invention, the dispatching resources include: thermal power units, photovoltaic units, hydroelectric units, electric vehicles, energy storage, and the external transmission channels for connecting the power system with other power systems;

[0141] Specifically, the supply adjustment intervals of each target unit are obtained through the following supply capacity calculation model:

[0142] According to the flexibility supply capacity ΔP of the following thermal power unit g,t calculation model, determine the supply adjustment interval of the thermal power unit:

[0143]

[0144] In the formula: ΔP g,k,t is the flexibility supply capacity of the kth thermal power unit, N g is the number of thermal power units, P g,k,b 、P g,k,max are respectively the upper and lower limits of the output of the kth thermal power unit, P g,k,t are respectively the output of the kth thermal power unit at time t, are respectively the upper and lower limits of the ramp of the kth thermal power unit.

[0145] According to the following flexibility supply capacity calculation model of the photovoltaic unit, determine the adjustable power ΔP of the photovoltaic unit at time t pv,t , as the supply adjustment interval of the photovoltaic unit:

[0146]

[0147] In the formula ΔP pv,m,t is the flexibility supply capacity of the mth photovoltaic unit, N pv is the number of photovoltaic units, P pv,m,t is the optimized output of the mth photovoltaic unit at time t, is the upper limit of the output of the photovoltaic unit, which is equivalent to the predicted value of the photovoltaic output in this embodiment;

[0148] According to the following flexibility supply capacity calculation model of the hydroelectric unit, determine the adjustable power ΔP of the hydroelectric unit at time t hy,t , as the supply adjustment interval of the photovoltaic unit:

[0149]

[0150] In the formula: ΔP hy,n,t is the flexibility supply capacity of the nth hydroelectric unit, N hy is the number of hydroelectric units, P hy,n,t is the optimized output of the nth hydroelectric unit at time t, is the upper limit of hydropower output, are the upper and lower ramping limits of the nth hydropower unit respectively.

[0151] According to the following calculation model of the flexibility supply capacity of the electric vehicle cluster, determine the regulation power ΔP of the electric vehicle cluster at time t ev,t , as the supply regulation interval of the electric vehicle cluster:

[0152]

[0153] According to the following calculation model of the flexibility supply capacity of the energy storage, determine the regulation power ΔP of the energy storage at time t es,t , as the supply regulation interval of the energy storage:

[0154]

[0155] In the formula: ΔP es,j,t is the flexibility supply capacity of the jth energy storage, N es is the number of energy storages, are the charging and discharging powers of the ith energy storage at time t respectively, P es,j,max is the maximum charging and discharging power of the ith energy storage, U es,j,t is a 0-1 variable representing the charging and discharging state of the jth energy storage unit at time t, U es,j,t =0 means the energy storage is in the charging state, U es,j,t =1 means the energy storage is in the discharging state, N es is the number of energy storages, are the charging and discharging efficiencies of the jth energy storage unit respectively, SOC j,t represents the state of charge of the jth energy storage unit at time t, SOC es,min 、SOC es,max are the minimum and maximum battery levels of the electric vehicle respectively.

[0156] According to the following calculation model of the flexibility supply capacity of the transmission channel, determine the supply regulation interval of the transmission channel:

[0157]

[0158] Where: ΔP e,out,t is the defined regulation amount of the transmission power at time t, P e,out,t is the transmission channel power, is the up and down ramping rate of the transmission channel, are the upper and lower limits of transmission at each moment.

[0159] S52. Calculate the adjustable load regulation amount and the load shedding regulation amount of the power system at each moment according to the preset demand response flexibility supply capacity model;

[0160] S53. Construct a resource scheduling model with the objective of minimizing the operating cost and the expected flexibility regulation demand, based on the supply regulation interval, the adjustable transfer load, the adjustable curtailable load, and the regulation demand risk value;

[0161] S54. Solve the resource scheduling model under the preset constraint conditions for ensuring the stable operation of the power system, and generate the target resource scheduling plan for each moment of the power system under extreme weather.

[0162] In a preferred embodiment of the present invention, by embedding conditional value at risk into the optimization objective function, the regulation capabilities and operation constraints of regulation resources such as thermal power, energy storage, and electric vehicles participating in power dispatching are coordinated to achieve the balance of system flexibility supply and demand under extreme weather. The expression of the resource scheduling model with the objective of minimizing the operating cost and the expected flexibility regulation demand is as follows:

[0163]

[0164] In the formula: c LOSS is the preset flexibility deficiency penalty, CVaR α conditional value at risk, reflecting the expected loss of tail extreme risk (given by step S454) N T is a dispatching period, c e,buy is the unit power purchase cost, P e,buy,t is the power purchase power at time t, P e,out,t is the power transmitted out at time t, P e,cut,t is the curtailed power at time t, c e,cut is the unit curtailed power cost. And the curtailed power is calculated based on the supply regulation interval, the adjustable transfer load, and the adjustable curtailable load.

[0165] Furthermore, according to several dispatching resources participating in power dispatching in step S51, the preset constraint conditions are as follows:

[0166] Power balance constraint:

[0167]

[0168] In the formula: P load,t is the load after demand response at time t, N ev , N es , N pv , N g , N hy are the total numbers of electric vehicles, energy storage, photovoltaic, thermal power, and hydropower respectively; are the charging and discharging powers of the i-th electric vehicle at time t respectively, are the charging and discharging powers of the i-th energy storage at time t, P g,k,t are the powers of the k-th thermal power unit at time t, P pv,m,t 、P hy,n,t are the outputs of the m-th photovoltaic unit and the n-th hydropower unit at time t respectively.

[0169] Operating constraints of the i-th electric vehicle:

[0170]

[0171] In the formula: are 0-1 variables, representing the charging and discharging states of the i-th electric vehicle at time t respectively, represents that the vehicle is in the discharging state, represents that the vehicle is in the charging state. When the vehicle is off-grid are the charging and discharging powers of the i-th electric vehicle at time t, P ev,i,max is the maximum charging and discharging power of the i-th electric vehicle, are the charging and discharging efficiencies of the i-th electric vehicle, SOC i,t-1 represents the SOC of the i-th electric vehicle at time t-1, SOC min 、SOC max are the minimum and maximum battery levels of the electric vehicle respectively.

[0172] Operating constraints of the thermal power unit:

[0173] P g,k,b ≤P g,k,t ≤P g,k,max ;

[0174]

[0175] In the formula: P g,k,b 、P g,k,max are the upper and lower limits of the output of the k-th thermal power unit respectively, P g,k,t are the outputs of the k-th thermal power unit at time t respectively, are the upper and lower ramping limits of the k-th thermal power unit respectively.

[0176] Output constraints of the photovoltaic unit:

[0177]

[0178] In the formula P pv,m,t is the optimized output of the m-th photovoltaic unit at time t, is the predicted value of the photovoltaic output, i.e., the upper limit.

[0179] Output constraints of the hydropower unit:

[0180]

[0181]

[0182] Where P hy,n,t is the optimized output of the nth hydro-optical power generation unit at time t, is the predicted value of hydropower output, i.e., the upper limit, are the ramp-up upper limit and ramp-down lower limit of the nth hydroelectric unit, respectively.

[0183] Energy storage operation constraints:

[0184]

[0185] Where: are the charging and discharging powers of the ith energy storage at time t, P es,j,max is the maximum charging and discharging power of the ith energy storage, U es,j,t is a 0-1 variable representing the charging and discharging state of the jth energy storage unit at time t, U es,j,t = 0 means the energy storage is in the charging state, U es,j,t = 1 means the energy storage is in the discharging state, N es is the number of energy storages, are the charging and discharging efficiencies of the jth energy storage unit, SOC j,t represents the state of charge of the jth energy storage unit at time t, SOC es,min 、SOC es,max are the minimum and maximum battery levels of the electric vehicle, respectively.

[0186] Power flow constraints:

[0187]

[0188] Where: V i,t is the voltage magnitude of node i at time t, V j,t is the voltage magnitude of node j at time t, N bus is the number of branches, n Node is the number of nodes P i,t and Q i,t are the active power injection and reactive power injection of node i at time t; G ij,t 、B ij,t are the real part and imaginary part of the element in the ith row and jth column of the nodal admittance matrix; θ ij,t is the phase angle difference between the two ends of branch ij at time t; are the minimum and maximum active powers allowed to flow through branch ij; V i max 、V i min are the upper and lower limits of the allowable voltage magnitude of node i; P ij,tIt is the active power flowing from node i to node j at time t.

[0189] The demand response model considers the transferable and curtailment-type demand-side responses of electrical loads, and the corresponding constraints are as follows:

[0190]

[0191] In the formula: is the original electrical load; P load,t is the electrical load after response P al,t is the electrical load after response; P al,t is the curtailment-type electrical load; P sl,t is the transferable-type electrical load; P al,t,0 、P sl,t,0 are the maximum electrical loads of the curtailment and transferable-type responses respectively, where T day is the number of time instants in a typical day.

[0192] In summary, this embodiment discloses a power resource scheduling method applicable to extreme weather. By using non-parametric kernel density estimation, a joint probability distribution function of wind power output and photovoltaic power output, which characterizes the output characteristics of wind power and photovoltaic power under extreme weather, is constructed. When conducting flexibility demand analysis, by virtue of the advantage of the joint distribution probability function in analyzing the non-linear correlation between variables and tail risks, the wind power prediction error and the photovoltaic prediction error are accurately analyzed. Then, combined with convolution operation, the regulation demand risk of the net load flexibility demand is quantified, overcoming the problems of the current flexibility demand assessment method that relies on the independent assumption of normal distribution, cannot accurately characterize the joint probability distribution characteristics of wind power and photovoltaic power output in extreme weather, and ignores the regulation demand risk, so that the formulated power scheduling plan can ensure the power supply reliability and operation stability of the power system under extreme weather.

[0193] As Figure 2 shown, on the basis of the above method item embodiment, a corresponding device item embodiment is provided;

[0194] An embodiment of the present invention provides a power resource scheduling device applicable to extreme weather, including:

[0195] An output data acquisition module, configured to acquire the wind power output prediction value, photovoltaic power output prediction value, load prediction value, wind power output actual value, photovoltaic power output actual value, and load actual value of each moment of the power system under extreme weather;

[0196] A net load calculation module, configured to construct a net load prediction curve of the power system according to the load prediction value, the wind power output prediction value, and the photovoltaic power output prediction value at each moment;

[0197] A prediction error calculation module, configured to calculate the load prediction error at each moment under extreme weather according to the load prediction value and the actual load value, and calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather based on a pre-constructed joint probability distribution function of wind and photovoltaic power outputs. Wherein, the joint probability distribution function of wind and photovoltaic power outputs is constructed by using non-parametric kernel density estimation to fit the wind power output and the photovoltaic power output under extreme weather, and is a probability distribution function used to characterize the output characteristics of wind power output and photovoltaic power output under extreme weather;

[0198] A demand risk analysis module, configured to perform convolution operation, and generate the net load flexibility demand and the corresponding regulation demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error;

[0199] A scheduling plan formulation module, configured to generate the target resource scheduling plan at each moment of the power system under extreme weather according to the net load flexibility demand and the corresponding regulation demand risk value.

[0200] It can be understood that the above device item embodiments correspond to the method item embodiments of the present invention, and can implement any one of the method item embodiments of the present invention to provide a power resource scheduling method applicable to extreme weather.

[0201] It should be noted that the device embodiments described above are only illustrative, and some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement without creative work.

[0202] Based on the above embodiments of a power resource scheduling method applicable to extreme weather, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a power resource scheduling method applicable to any embodiment of the present invention.

[0203] Exemplarily, in this embodiment, the computer program may be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to implement the present invention. The one or more module elements may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the terminal device.

[0204] The terminal device may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0205] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal device and connects various parts of the entire terminal device through various interfaces and lines.

[0206] Based on the above method item embodiment, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for scheduling power resources applicable to extreme weather according to any one of the above method item embodiments of the present invention.

[0207] Among them, for the modules / units integrated in the device / terminal device, if they are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0208] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art of the present technology, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A power resource scheduling method applicable to extreme weather, characterized in that, Including: Obtaining the predicted values of wind power output, photovoltaic power output, load, actual wind power output, actual photovoltaic power output, and actual load of the power system at each moment under extreme weather; Constructing a net load prediction curve of the power system according to the load prediction value, the wind power output prediction value, and the photovoltaic power output prediction value at each moment; Calculating the load prediction error at each moment under extreme weather according to the load prediction value and the actual load value, and calculating the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather based on a pre-constructed joint probability distribution function of wind and photovoltaic power output. Wherein, the joint probability distribution function of wind and photovoltaic power output is a probability distribution function constructed by using non-parametric kernel density estimation to fit the wind power output and photovoltaic power output under extreme weather, and is used to characterize the output characteristics of wind power output and photovoltaic power output under extreme weather; Performing convolution operation, and generating the net load flexibility demand and the corresponding regulation demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error; Generating a target resource scheduling plan for each moment of the power system under extreme weather according to the net load flexibility demand and the corresponding regulation demand risk value; 2. The power resource scheduling method applicable to extreme weather according to claim 1, characterized in that The obtaining of the predicted values of wind power output and photovoltaic power output of the power system at each moment under extreme weather includes: Obtaining several types of meteorological forecast data groups at each moment under extreme weather; wherein, each type of meteorological forecast data group is composed of meteorological forecast data provided by several different meteorological data sources; Determining the weight of each meteorological forecast data in each meteorological forecast data group based on the meteorological data source corresponding to each meteorological forecast data; Performing weighted fusion on the meteorological forecast data in each meteorological forecast data group according to the weight to generate a meteorological characteristic value corresponding to each meteorological forecast data group; Inputting the meteorological characteristic value into a preset wind and photovoltaic power output prediction model, so that the wind and photovoltaic power output prediction model outputs the predicted values of wind power output and photovoltaic power output of the power system at each moment under extreme weather; wherein, the wind and photovoltaic power output prediction model is a deep learning model trained by using the historical meteorological characteristics, historical wind power output values, and historical photovoltaic power output values of historical extreme weather; 3. The power resource scheduling method applicable to extreme weather according to claim 2, wherein The construction of the joint probability distribution function of wind and photovoltaic power output includes: Obtaining the historical meteorological characteristics, historical wind power output values, and historical photovoltaic power output values of the power system under historical extreme weather; Using the non-parametric kernel density estimation method to construct the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic power according to the historical meteorological characteristics and the corresponding historical wind power output values and historical photovoltaic power output values; Using the Frank-Copula function to construct a cumulative probability distribution function of wind and photovoltaic power output according to the marginal probability distribution of wind power output and the marginal probability distribution of photovoltaic power; Construct the joint probability distribution function of the wind and solar power output according to the cumulative probability distribution function of the wind and solar power output.

4. The power resource scheduling method applicable to extreme weather according to claim 3, characterized in that, Based on the pre-constructed joint probability distribution function of the wind and solar power output, calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather according to the wind power output prediction value, the photovoltaic power output prediction value, the actual wind power output value, and the actual photovoltaic power output value, including: Use the following formula to calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather according to the wind power output prediction value, the photovoltaic power output prediction value, the actual wind power output value, and the actual photovoltaic power output value: where, ΔP w,t is the wind power prediction error at time t, is the actual wind power output value at time t, is the predicted wind power output value at time t, F w,s (p w , p s ) is the joint probability distribution function of wind and solar power output, ΔP s,t is the photovoltaic prediction error at time t, is the actual photovoltaic output value at time t, is the predicted photovoltaic output value at time t.

5. The power resource scheduling method applicable to extreme weather according to claim 4, characterized in that Use convolution operation to generate the net load flexibility demand and the corresponding target probability density function at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error, including: Obtain the first error probability density corresponding to the load prediction error; Synthesize the net load flexibility regulation demand of the power system according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error; Calculate the second error probability density corresponding to the negative value of the sum of the wind power prediction error and the photovoltaic power prediction error according to the joint probability distribution function of the wind and solar power output; Use convolution operation to construct the target probability density function according to the net load flexibility regulation demand, the first error probability density, and the second error probability density; Determine the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather according to the net load flexibility demand and the target probability density function.

6. The power resource scheduling method applicable to extreme weather according to claim 5, characterized in that, Generate the target resource scheduling plan at each moment of the power system under extreme weather according to the net load flexibility demand and the corresponding regulation demand risk value, including: Obtain the supply regulation interval of several scheduling resources used to participate in power scheduling in the power system; Calculate the adjustable transfer load and the adjustable curtailable load of the power system at each moment according to the preset demand response flexibility supply capacity model; Construct a resource scheduling model with the goal of minimizing the operating cost and the flexibility regulation demand expectation according to the supply regulation interval, the adjustable transfer load, the adjustable curtailable load, and the regulation demand risk value; Solve the resource scheduling model under the preset constraint conditions for ensuring the stable operation of the power system to generate the target resource scheduling plan at each moment of the power system under extreme weather.

7. The power resource scheduling method applicable to extreme weather according to claim 6, wherein, The regulation demand risk value includes: value at risk and conditional value at risk; Determine the flexibility regulation demand expectation and the corresponding regulation demand risk value at each moment of the power system under extreme weather according to the net load flexibility demand and the target probability density function, including: Discretize the net load flexibility regulation demand and calculate the target probability that the net load flexibility regulation demand of the power system at each moment is equal to the corresponding discretized value. Generate the expected value of the flexibility regulation demand according to the target probability, the net load flexibility regulation demand, and the probability density function. Calculate the value at risk of the net load flexibility regulation demand at each moment of the power system under a preset confidence level according to the target probability. Calculate the conditional value at risk of the expected loss of the tail extreme risk corresponding to each of the values at risk according to the value at risk and the net load flexibility regulation demand.

8. An electric power resource scheduling device applicable to extreme weather, characterized in that, Comprising: An output data acquisition module, configured to acquire the predicted values of wind power output, the predicted values of photovoltaic power output, the predicted values of load, the actual values of wind power output, the actual values of photovoltaic power output, and the actual values of load at each moment of the power system under extreme weather. A net load calculation module, configured to construct a net load prediction curve of the power system according to the predicted values of load, the predicted values of wind power output, and the predicted values of photovoltaic power output at each moment. A prediction error calculation module, configured to calculate the load prediction error at each moment under extreme weather according to the predicted value of load and the actual value of load, and calculate the wind power prediction error and the photovoltaic power prediction error at each moment under extreme weather based on a pre-constructed joint probability distribution function of wind and photovoltaic power outputs according to the predicted values of wind power output, the predicted values of photovoltaic power output, the actual values of wind power output, and the actual values of photovoltaic power output; wherein, the joint probability distribution function of wind and photovoltaic power outputs is constructed by pre-using non-parametric kernel density estimation to fit the wind power output and the photovoltaic power output under extreme weather, and is a probability distribution function used to characterize the output characteristics of the wind power output and the photovoltaic power output under extreme weather. A demand risk analysis module, configured to perform convolution operation, and generate the net load flexibility demand and the corresponding regulation demand risk value at each moment according to the net load prediction curve, the load prediction error, the wind power prediction error, and the photovoltaic power prediction error. A scheduling plan formulation module, configured to generate a target resource scheduling plan at each moment of the power system under extreme weather according to the net load flexibility demand and the corresponding regulation demand risk value.

9. A terminal device, characterized in that, Comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, when the processor executes the computer program, implementing a power resource scheduling method applicable to extreme weather according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Comprising: A stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute a power resource scheduling method applicable to extreme weather according to any one of claims 1-7.

Images