Virtual power plant response potential credible domain construction method and system

Abstract

The invention discloses a virtual power plant response potential credible domain construction method and system, relates to the technical field of virtual power plant management, and solves the problem of large calculation result error of virtual power plant response potential. According to the method, the joint probability distribution model is established, the space-time correlation among the wind speed, the irradiance and the load is quantified and is included in calculation of the response potential credible domain, so that the response potential obtained through calculation can cope with the uncertainty of the supply side and the main variable component of the demand side, and more extreme scenes can be covered; multi-dimensional uncertainty is quantified into resource available capacity probability distribution of each node in different weather scenes through double-layer Monte Carlo simulation, and then a contraction credible domain boundary is adjusted based on model output and a simulation result. The problem of large response potential error caused by neglecting dynamic constraint and real-time fluctuation of a traditional static model can be effectively solved, and accurate data support is provided for virtual power plant scheduling under high-proportion new energy access.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant management, and in particular to a method and system for constructing a virtual power plant response potential trusted domain. Background Art

[0002] We must build a clean, low-carbon, safe, and efficient energy system, control the total amount of fossil energy, focus on improving its utilization efficiency, implement renewable energy substitution initiatives, deepen power system reform, and build a new power system dominated by new energy. The emergence of new energy as the dominant power source will pose new challenges to the economic and stable operation of the power system. Renewable energy output exhibits significant randomness and intermittence, with fluctuating and unregulated characteristics. Therefore, tapping into load-side regulation resources, upgrading traditional source-following-load dynamics to source-load interaction, and enhancing the power system's regulation capabilities are new and important approaches to ensuring the safe and economic operation of the new power system.

[0003] With the full entry of all industrial and commercial users into the electricity market, policy mechanisms for user-side participation in grid interaction and transactions are gradually maturing, and several provinces have successively promulgated trading rules related to demand response and virtual power plants. Virtual power plants are currently the most cost-effective way to regulate power system stability. Participating in virtual power plant demand response requires load aggregators to explore and aggregate controllable load resources. Therefore, considering multiple influencing factors and assessing the response potential of controllable loads can fully tap the aggregated response potential of loads, which has significant guiding significance for demand response quantity and price submissions.

[0004] When it comes to calculating the response potential of virtual power plants, current mainstream technologies rely on deterministic parameters to define the potential range. This calculation model builds a model based on fixed, preset parameters, treating renewable energy generation and load demand as stable variables. However, it fails to factor in uncertainties such as renewable energy output fluctuations and load forecast deviations. In actual operation, renewable energy generation sources such as wind and solar power are highly susceptible to natural conditions such as light intensity, wind speed and direction, and temperature fluctuations, resulting in dramatic fluctuations in power generation curves. Furthermore, power load demand is influenced by multiple factors, including residents' daily routines, industrial production plans, and social activities. Even with advanced forecasting algorithms, it is difficult to completely eliminate forecast deviations.

[0005] In special scenarios such as extreme weather and large-scale events, the response potential calculation results based on deterministic parameters often overestimate the actual response capabilities of virtual power plants, resulting in supply gaps in key links such as power dispatching and load balancing, affecting the stability and reliability of the power system, and failing to meet the development needs of smart grids for efficient and flexible allocation of power resources.

[0006] In view of this, a method and system for constructing a trusted domain of virtual power plant response potential is needed. Summary of the Invention

[0007] To address the problem of large errors in the calculation results of virtual power plant response potential in the existing technology, the present invention provides a method and system for constructing a trusted domain of virtual power plant response potential, which can more accurately calculate the virtual power plant response potential. The specific technical solution is as follows:

[0008] In a first aspect, an embodiment of the present application provides a method for constructing a trusted domain of a virtual power plant's response potential, comprising:

[0009] Acquire meteorological data, equipment failure probability data of the power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid; the meteorological data includes wind speed data and irradiance data; establish a joint probability distribution model based on the wind speed data, the irradiance data and the temperature-sensitive load data through a Copula function, and the joint probability distribution model is used to indicate the spatiotemporal correlation of the wind speed data, the irradiance data and the temperature-sensitive load data; perform a double-layer Monte Carlo simulation based on the equipment failure probability data and the joint probability distribution model to obtain a weather scenario set and the temperature-sensitive load data of each node in the power grid. The probability distribution of resource available capacity of the node; wherein, the outer layer of the double-layer Monte Carlo simulation is used to generate the weather scenario set related to the new energy fluctuation based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is used to construct the equipment failure event tree under each weather scenario based on the weather scenario set and the equipment failure probability data, and the equipment failure event tree is used to calculate the probability distribution of resource available capacity of each node under the corresponding weather scenario; based on the meteorological data, the weather scenario set, the probability distribution of resource available capacity of each node, and the node operation data, the response potential credible domain boundary of the virtual power plant is adjusted.

[0010] Preferably, the meteorological data includes real-time meteorological data, and the node operation data includes node measurement data and node prediction data of the node; adjusting the response potential credible domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of resource available capacity of each node, and the node operation data includes: performing similarity matching based on the real-time meteorological data and the weather scenario set to obtain a matching weather scenario subset, the matching weather scenario subset including the top N weather scenarios in the weather scenario set sorted from high to low in similarity to the real-time meteorological data, where N is a positive integer greater than or equal to 1; obtaining the corresponding probability distribution of resource available capacity of each node based on the matching weather scenario subset; adjusting the response potential credible domain boundary of the virtual power plant based on the deviation value between the node measurement data and the node prediction data, and the standard deviation of the capacity distribution; the standard deviation of the capacity distribution is calculated based on the probability distribution of resource available capacity of each node corresponding to the matching weather scenario subset.

[0011] Preferably, the step of adjusting the response potential credible region boundary of the virtual power plant based on the deviation value between the node measurement data and the node prediction data, and the capacity distribution standard deviation, includes: calculating a deviation mean of the deviation value based on a sliding time window, wherein the length of the sliding time window is inversely proportional to the rate of change of the deviation value; when the deviation mean is greater than a preset first threshold, shrinking the response potential credible region boundary based on the capacity distribution standard deviation; the calculation formula for shrinking the response potential credible region boundary includes:

[0012]

[0013] Among them, C L 、C U are the lower and upper bounds of the credible region of the response potential before shrinkage, are the lower and upper bounds of the credible region of the response potential after shrinkage, σ C is the standard deviation of capacity distribution, k1 and k2 are dynamic shrinkage coefficients, is the mean deviation.

[0014] Preferably, the calculation process of the dynamic shrinkage coefficients k1 and k2 includes: when the similarity between the real-time meteorological data and the weather scene is higher than a preset second threshold, obtaining a historical optimal coefficient as the dynamic shrinkage coefficient, and the historical optimal coefficient is calculated based on the historical number of limit violations of the node; when the similarity between the real-time meteorological data and the weather scene is lower than a preset third threshold, updating the dynamic shrinkage coefficient; the calculation formula for updating the dynamic shrinkage coefficient includes:

[0015]

[0016] in, are the i-th dynamic shrinkage coefficient before and after the update, η is the learning rate, ΔRisk is the risk deviation, and sgn() is the sign function.

[0017] Preferably, the joint probability distribution model includes the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data; the double-layer Monte Carlo simulation is performed based on the equipment failure probability data and the joint probability distribution model to obtain the weather scenario set and the probability distribution of resource available capacity of each node in the power grid, including: generating the weather scenario set based on the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data through Latin hypercube sampling; and calculating the probability distribution of resource available capacity of each node based on the weather scenario set and the equipment failure probability data.

[0018] Preferably, the method of calculating the probability distribution of the available capacity of resources of each node based on the weather scenario set and the equipment failure probability data includes: generating a Markov state transfer matrix of the node under each weather scenario in the weather scenario set based on the weather scenario set and the equipment failure probability data, wherein the elements in the Markov state transfer matrix are the transition probabilities of the states of the devices in each node; obtaining the operating temperature of the devices at each node; and dynamically correcting the transition probability based on the operating temperature, wherein the calculation formula for correcting the transition probability includes:

[0019]

[0020] Where γ is the temperature sensitivity coefficient, T ref is the rated operating temperature of the equipment, T is the real-time operating temperature of the equipment, and p ij is the probability of the device transferring from state i to state j after correction, is the basic probability of the device transferring from state i to state j before correction; a device failure event tree is generated based on the corrected Markov state transition matrix; and based on the device failure event tree, the probability distribution of the available capacity of the resources of each node is calculated.

[0021] Preferably, after adjusting the response potential credible domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the available capacity of the resources of each node, and the node operation data, the method further includes: obtaining the spinning reserve rate of the power grid; adjusting the response potential credible domain boundary based on the spinning reserve rate; the calculation formula for adjusting the response potential credible domain boundary includes:

[0022]

[0023] Among them, α and β are the adjustment coefficients corresponding to the power grid security level. are the lower and upper bounds of the credible region of response potential after adjusting the spinning reserve ratio, R spin is the spinning reserve ratio.

[0024] In a second aspect, an embodiment of the present application provides a system for constructing a trusted domain of virtual power plant response potential, which is applied to the method described in the first aspect, and the system includes:

[0025] An acquisition module is used to acquire meteorological data, equipment failure probability data of the power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid; the meteorological data includes wind speed data and irradiance data;

[0026] a modeling module, configured to establish a joint probability distribution model based on the wind speed data, the irradiance data, and the temperature-sensitive load data through a Copula function, wherein the joint probability distribution model is configured to indicate a spatiotemporal correlation among the wind speed data, the irradiance data, and the temperature-sensitive load data;

[0027] a simulation module for performing a double-layer Monte Carlo simulation based on the equipment failure probability data and the joint probability distribution model to obtain a set of weather scenarios and a probability distribution of resource available capacity for each node in the power grid; wherein the outer layer of the double-layer Monte Carlo simulation is used to generate the set of weather scenarios related to renewable energy fluctuations based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is used to construct an equipment failure event tree for each weather scenario based on the weather scenario set and the equipment failure probability data, and the equipment failure event tree is used to calculate the probability distribution of resource available capacity for each node under the corresponding weather scenario;

[0028] An adjustment module is used to adjust the response potential trust domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the resource available capacity of each node, and the node operation data.

[0029] In a third aspect, an embodiment of the present application provides a computing device, comprising: a memory for storing a program; and a processor for loading the program to execute the method described in the first aspect.

[0030] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the method as described in the first aspect.

[0031] Compared with the existing technology, the beneficial effects of the present invention are: by establishing a joint probability distribution model, quantifying the dependency between wind speed, irradiance and load and incorporating it into the calculation of the credible domain of response potential, the calculated response potential can cope with the uncertainty on the supply side and the main variable components on the demand side, thereby being able to cover more extreme scenarios; through double-layer Monte Carlo simulation, the multi-dimensional uncertainty is quantified as the probability distribution of resource available capacity of each node under different weather scenarios, and then the credible domain boundary is adjusted and shrunk based on the model output and simulation results, which can effectively solve the problem of large response potential error caused by traditional static models ignoring dynamic constraints and real-time fluctuations, and provide accurate data support for virtual power plant scheduling under a high proportion of new energy access. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for the specific embodiments or the description of the prior art. Similar elements or parts are generally identified by similar reference numerals throughout the drawings. Elements or parts in the drawings are not necessarily drawn to scale.

[0033] Figure 1 A flowchart of a method for constructing a trusted domain of a virtual power plant's response potential provided in an embodiment of the present application;

[0034] Figure 2 A schematic diagram of the structure of a virtual power plant response potential trusted domain construction system provided in an embodiment of the present application;

[0035] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0037] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0038] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0039] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0040] In order to solve the problem of large errors in the calculation results of virtual power plant response potential in traditional methods, the present invention provides a method and system for constructing a trusted domain of virtual power plant response potential, which can more accurately calculate the response potential of virtual power plants.

[0041] See also Figure 1 , Figure 1 A flowchart of a method for constructing a trusted domain of a virtual power plant response potential is provided for an embodiment of the present application, and the method is applied to a computing device; Figure 1 As shown, the method includes:

[0042] Step 101: A computing device obtains meteorological data, equipment failure probability data of a power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid.

[0043] Among them, the computing device can be a computing node in the power grid, a computing device in the main control room of the power grid, or a computing device that is communicatively connected to the control module and data acquisition module of the power grid; the computing device can specifically be a server, or a smart terminal such as a personal computer or tablet directly operated by power grid management personnel or maintenance personnel.

[0044] Among them, the power grid in the embodiment of the present application is a power grid connected to new energy, and the power grid includes multiple nodes, including load nodes, energy storage, new energy stations and traditional power generation nodes; generator sets, transformers or inverters and other equipment can be deployed on the nodes.

[0045] Among them, computing devices can obtain meteorological data through meteorological detection terminals, and obtain node operation data and load data through smart meters. The load data includes temperature-sensitive load data, such as air conditioners, electric heating equipment, etc.; obtain equipment operation data such as equipment operating temperature and node power through sensors deployed on the equipment; and obtain equipment failure probability data from relevant databases of the power grid. The equipment failure probability data includes the probability of equipment failure under different weather scenarios or factors.

[0046] Specifically, meteorological data includes wind speed data and irradiance data. Specifically, the computing device can obtain wind speed and irradiance grid data from the ECMWF global forecast system, and then use the Kriging interpolation method to map the grid data to grid coordinates to obtain wind speed data and irradiance data.

[0047] Specifically, node operation data includes historical data, real-time data and predicted data; temperature-sensitive load data includes the load curves of temperature-sensitive equipment such as air conditioners and electric heating in the distribution network; equipment operation data also includes energy storage charge and discharge times, photovoltaic inverter fault records, and transformer temperature logs.

[0048] Step 102: The computing device establishes a joint probability distribution model based on the wind speed data, the irradiance data, and the temperature-sensitive load data through a Copula function.

[0049] After acquiring the above data, the computing device may align the wind speed data, the irradiance data, and the temperature-sensitive load data according to the timestamps, and then perform marginal distribution fitting on the wind speed data and the irradiance data, respectively.

[0050] The computing device can then use the Weibull distribution to fit the marginal distribution of the wind speed data, and its probability density function is:

[0051]

[0052] Among them, v is the wind speed, k is the shape parameter used to control the distribution shape, and λ is the scale parameter used to determine the distribution range.

[0053] Then, the computing device can perform marginal distribution fitting on the wind power output data by combining the probability density function with maximum likelihood estimation or moment estimation.

[0054] Among them, the computing device uses Beta distribution to fit the marginal distribution of irradiance data, and its probability density function is:

[0055]

[0056] Where r is the irradiance, α and β are shape parameters used to control the distribution skewness. In sunny weather, α>β; B(α,β) is the Beta function.

[0057] The computing device then uses the EM algorithm to fit the marginal distribution of the irradiance data using this probability density function. The absence of light at night is a key physical characteristic of irradiance data. By using the EM algorithm to process the zero-value data caused by the absence of light at night, the computing device can better reflect this characteristic in the model, enabling the model to mathematically fit the data.

[0058] The computing device can then fit the wind speed data v to the marginal distribution i and irradiance data r i Convert to a uniformly distributed variable:

[0059] u i =F v (v i ),w i =F r (r i );

[0060] Among them, F v 、F r are the cumulative distribution functions of wind speed data and irradiance data respectively; the converted data w i 、u i ∈[0,1] obeys uniform distribution.

[0061] After converting the differently distributed wind speed and irradiance data into a uniform distribution, the computing device can use this uniform variable as input to a vine-structured copula function to establish a joint probability distribution model. This joint probability distribution model is used to indicate the spatiotemporal correlation between the wind speed data, irradiance data, and temperature-sensitive load data.

[0062] Specifically, the vine structure Copula can capture the nonlinear time-varying dependency among wind speed, irradiance and temperature-sensitive loads through a hierarchical structure.

[0063] For example, the first layer of the vine structure Copula function uses the Clayton Copula function to describe the wind speed u i and irradiance w i Binary CopulaC uw , used to describe u i and w i The lower tail dependence of ; Gaussian Copula function is used to describe the wind speed u i Binary CopulaC with temperature-sensitive load L uL , used to describe u i and L linear correlation; given wind power u i In the case of the second layer, the Gumbel Copula function is used to construct the photoelectric w i Condition CopulaC with temperature sensitive load L wLu , describe w i and L at a given u i Conditional upper tail dependence.

[0064] The joint distribution expression of this joint probability distribution model can be:

[0065] f(v,r,l)=f V (v) f R (r)·f L (l)·c uw (F v (v),F r (r))·c uL (F v (v),F L (l))·c wL|u (F w|u (w|u),F L|u (l|u));

[0066] Among them, c uw 、c uL 、c wL|u is the density function of each layer of Copula; F w|u(w|u), F L|u (l|u) is the conditional distribution function, which is calculated through partial derivatives.

[0067] Step 103: The computing device performs a double-layer Monte Carlo simulation based on the device failure probability data and the joint probability distribution model to obtain a weather scenario set and a probability distribution of available capacity of resources of each node in the power grid.

[0068] Among them, the outer layer of the double-layer Monte Carlo simulation is used to generate the weather scenario set related to the new energy fluctuation based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is used to construct the equipment failure event tree under each weather scenario based on the weather scenario set and the equipment failure probability data. The equipment failure event tree is used to calculate the probability distribution of the resource available capacity of each node under the corresponding weather scenario.

[0069] Preferably, the computing device can generate the weather scenario set based on the Weibull probability distribution of the wind speed, the Beta probability distribution of the irradiance and the joint probability distribution model through Latin hypercube sampling; and calculate the probability distribution of the resource available capacity of each node based on the equipment failure probability data.

[0070] The wind speed-Weibull probability distribution and the irradiance-Beta probability distribution have been obtained in step 102 and will not be described again here.

[0071] Latin Hypercube sampling determines the corresponding value range based on the wind speed-Weibull probability distribution and the irradiance-Beta probability distribution. It then divides the value range of each input variable into several equally probable intervals, randomly extracts a sample point from each interval, and combines these sample points to form a weather scenario. For example, a computing device can generate 1,000 weather scenario samples using Latin Hypercube sampling based on the wind speed and irradiance probability distributions, ensuring layered coverage across all dimensions.

[0072] The joint probability distribution model is obtained based on the Copula function, and then the correlation weather scene is generated based on the model, which can ensure that the spatiotemporal correlation between wind speed and irradiance in the weather scene conforms to the actual observation law. In addition, it can also replace the traditional independent sampling to make the probability of extreme weather events (such as windless and cloudy days) more accurate.

[0073] The computing device may quantify the impact of the device failure on the available resource capacity of each node based on the device failure probability data, and obtain a probability distribution of the available resource capacity of each node.

[0074] Preferably, the computing device can generate a Markov state transition matrix based on the weather scenario set and the equipment failure probability data, wherein the elements of the transition matrix are the basic probabilities of the equipment transitioning from the current state to various states under the corresponding weather scenario.

[0075] The state space of the Markov state transition matrix can include {normal, derated, faulty}, and the base probability comes from the mean time between failures (MTBF) data provided by the equipment manufacturer. For example, the transition matrix can include the transition probabilities of three devices under corresponding weather scenarios, as shown below:

[0076]

[0077] It is understandable that, when a node is deployed with multiple devices, the computing device may take the mean or weighted value of the transition probabilities of the multiple devices as the transition probability of the node.

[0078] Then, the computing device can obtain the operating temperature of the device of each node; based on the operating temperature, the transition probability in the Markov state transition matrix is dynamically corrected, and the correction formula is:

[0079]

[0080] Where γ is the temperature sensitivity coefficient, T ref is the rated operating temperature, T is the real-time operating temperature, p ij is the probability of the device transferring from state i to state j after correction, is the basic probability of the device transitioning from state i to state j before correction; for example, T ref The value can be 40℃ and γ can be 0.1. These two data are calibrated through accelerated life tests.

[0081] Then, the computing device may generate a device failure event tree based on the modified Markov state transition matrix; and calculate the probability distribution of the available capacity of resources of each node based on the device failure event tree.

[0082] Each node corresponds to a device failure event tree. The root node of the device failure event tree is the initial state. Each node generates three child nodes (state transitions) in the next period. The edge weights between nodes are transition probabilities. Typically, 3-5 layers are constructed, corresponding to 15-75 minute forecasts, to control the depth of the event tree.

[0083] In the specific calculation process, the single-path probability is equal to the product of all transition probabilities on the path; then all single-path probabilities of the same state are added together to obtain the probability of that state. The computing device can then obtain the maximum resource available capacity C_MAX of each device under the corresponding weather scenario, and then map the state to the resource available capacity C. Specifically, for wind power / photovoltaic power, the maximum resource available capacity C_MAX is the maximum power that can be generated under current weather conditions; for energy storage, the maximum resource available capacity C_MAX is the available capacity after considering the charging and discharging efficiency and the state of charge (SOC); for adjustable loads, the maximum resource available capacity C_MAX is the load power that can be reduced or transferred.

[0084] Exemplarily, C(normal)=100% C_MAX, C(derating)=60% C_MAX, and C(fault)=0% C_MAX.

[0085] The computing device can then combine the capacity and probability corresponding to the state to obtain the probability distribution of the device's resource available capacity. The probability distribution of the node's resource available capacity includes the probability distribution of the available capacity of all devices deployed on it.

[0086] Step 104: The computing device adjusts the response potential trust domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the available capacity of the resources of each node, and the node operation data.

[0087] Among them, the node operation data includes node measurement data and node prediction data; the node measurement data includes the actual output of wind power and photovoltaic power at the new energy site, the current charging and discharging power and charge status of the energy storage system, and the actual power demand of adjustable loads such as temperature-sensitive loads and interruptible loads at the load node; correspondingly, the node prediction data includes the ultra-short-term prediction data of wind power / photovoltaic power at the new energy site driven by numerical weather forecast (NWP), the charging and discharging power curve of the energy storage system based on the day-ahead optimization results, and the power demand forecast of the adjustable load.

[0088] The meteorological data includes real-time meteorological data.

[0089] Preferably, the computing device can perform similarity matching based on the real-time meteorological data and the weather scene set to obtain a matching weather scene subset, which includes the top N weather scenes in the weather scene set ranked from high to low in similarity with the real-time meteorological data, where N is a positive integer greater than or equal to 1.

[0090] Preferably, N may be 5% of the number of weather scenes in the Tianji scene set.

[0091] Then, the computing device can obtain the corresponding probability distribution of available resource capacity based on the matching weather scenario subset; then, based on the deviation value between the node measurement data and the node prediction data, adjust the response potential trust domain boundary of the virtual power plant; the capacity distribution standard deviation is calculated based on the probability distribution of available resource capacity of each node corresponding to the matching weather scenario subset.

[0092] Specifically, the computing device can calculate the corresponding tolerance distribution standard deviation based on the probability distribution of available resource capacity corresponding to each weather scenario in the matching weather scenario subset, and then take the mean as the global tolerance distribution standard deviation; it can also select the weather scene with the highest similarity from the matching weather scenario subset to calculate the corresponding tolerance distribution standard deviation.

[0093] The calculation process of the standard deviation of the tolerance distribution of a single device is as follows:

[0094] First, the computing device can set the minimum and maximum capacities of the device based on historical data or device ratings; then, multiple intervals are divided between the minimum and maximum capacities, and the median capacity of each interval can be marked as C. i , i is the interval index; then for the capacity samples generated by Monte Carlo simulation, count the frequency of falling into each interval and calculate the probability density:

[0095]

[0096] The computing device can use the Monte Carlo simulation output of the resource available capacity probability distribution P prior (C i ) as the prior distribution, C is the resource available capacity of the virtual power plant in the current state, that is, the total power that can be dispatched by the virtual power plant in the current state; and obtain the actual available capacity C at the current moment obs Then assume that the observation error follows a normal distribution Establish the likelihood function:

[0097]

[0098] The likelihood function represents the available capacity C of a given resource. i In the case of obs,i Then update the posterior distribution:

[0099]

[0100] Among them, σ e is the standard deviation of the observation error, calibrated according to the accuracy of the equipment, and N is the total number of intervals.

[0101] The computing device can then calculate the cumulative probability distribution:

[0102]

[0103] Then, the computing device calculates the upper and lower bounds that meet the confidence level α(t) based on the cumulative probability distribution, that is, the upper bound C of the available capacity of the device resources U and the lower bound C L :

[0104]

[0105] Preferably, the confidence level α(t) may vary with the rate of change of the deviation between the node measurement data and the node prediction data:

[0106] α(t)=α0·exp(-λΔε τ );

[0107] Among them, α0 is the preset initial confidence level, which can be set to 0.95; λ is the shrinkage rate coefficient, which can be set to 0.1; Δε τ is the rate of change of the deviation value.

[0108] The computing device can then calculate the expected value of the available capacity of the resource for that device:

[0109]

[0110] Then calculate the variance and standard deviation of the available capacity of the resource of the device:

[0111] Then, the device capacity distribution standard deviation σ of each device can be calculated based on q , calculate the global capacity distribution standard deviation σ total :

[0112]

[0113] Where m is the number of devices. Similarly, the initial response potential credible region of the virtual power plant can also be obtained by accumulating the upper and lower bounds of the resource available capacity of each device, and the global mean deviation can be the cumulative mean of the mean deviations of all devices.

[0114] Preferably, the computing device may calculate a mean deviation of the deviation value based on a sliding time window, wherein the length of the sliding time window is inversely proportional to the rate of change of the deviation value; when the mean deviation is greater than a preset first threshold, the boundary of the response potential credible region is contracted; the calculation formula for contracting the boundary of the response potential credible region includes:

[0115]

[0116] Among them, C L 、C Uare the lower and upper bounds of the credible region of the response potential before shrinkage, are the lower and upper bounds of the credible region of the post-contraction response potential, σ C is the standard deviation of capacity distribution, k1 and k2 are dynamic shrinkage coefficients, It is understandable that the parameters in this preferred implementation are all global parameters.

[0117] Preferably, the calculation process of the dynamic shrinkage coefficients k1 and k2 includes: when the similarity between the real-time meteorological data and the weather scene is higher than a preset second threshold, obtaining a historical optimal coefficient as the dynamic shrinkage coefficient, and the historical optimal coefficient is calculated based on the historical number of limit violations of the node; when the similarity between the real-time meteorological data and the weather scene is lower than a preset third threshold, updating the dynamic shrinkage coefficient; the calculation formula for updating the dynamic shrinkage coefficient includes:

[0118]

[0119] in, are the i-th dynamic shrinkage coefficient before and after the update, η is the learning rate, ΔRisk is the risk deviation, and sgn() is the sign function.

[0120] Specifically, the historical optimal coefficient can be calculated based on the node's historical number of limit violations and the revenue loss caused by conservative estimates. For example, the computing device can minimize the weighted sum of the number of limit violations in historical data and the revenue loss caused by conservative estimates, and find the dynamic contraction coefficient corresponding to this goal as the historical optimal coefficient.

[0121] Preferably, after adjusting the response potential credible domain boundary of the virtual power plant, the computing device may further obtain the spinning reserve rate of the power grid, and then further adjust the response potential credible domain boundary based on the spinning reserve rate; the calculation formula for adjusting the response potential credible domain boundary includes:

[0122]

[0123] Among them, α and β are the adjustment coefficients corresponding to the power grid security level. are the lower and upper bounds of the credible region of response potential after adjusting the spinning reserve ratio, R spin is the spinning reserve ratio.

[0124] The spinning reserve ratio refers to the ratio of a system's spinning reserve capacity to the system's maximum load demand. Spinning reserve capacity refers to the spare capacity of operating generators in a power system, readily available to balance system load fluctuations and respond to sudden failures.

[0125] It is understandable that the higher the grid security level, the smaller the corresponding adjustment coefficients α and β, to prevent the credible domain boundary of the response potential after adjustment from causing excessive power fluctuations and bringing unnecessary risks.

[0126] In an embodiment of the present application, a joint probability distribution model is established to quantify the dependency between wind speed, irradiance and load and incorporate it into the calculation of the credible domain of response potential, so that the calculated response potential can cope with the uncertainty on the supply side and the main variable components on the demand side, thereby being able to cover more extreme scenarios; through a double-layer Monte Carlo simulation, the multi-dimensional uncertainty is quantified as the probability distribution of the available capacity of resources at each node under different weather scenarios, and the dynamic shrinkage mechanism is combined to adaptively adjust the shrinkage credible domain boundary, which can effectively solve the problem of large response potential errors caused by ignoring dynamic constraints and real-time fluctuations in traditional static models, and provide accurate data support for virtual power plant scheduling under a high proportion of new energy access.

[0127] The above describes the method part provided by the embodiment of the present application. The following describes the system part provided by the embodiment of the present application.

[0128] See also Figure 2 , Figure 2 A schematic diagram of a virtual power plant response potential trusted domain construction system provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the system 20 includes:

[0129] Acquisition module 201 is used to acquire meteorological data, equipment failure probability data of the power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid; the meteorological data includes wind speed data and irradiance data;

[0130] A modeling module 202 is configured to establish a joint probability distribution model based on the wind speed data, the irradiance data, and the temperature-sensitive load data using a Copula function, wherein the joint probability distribution model is configured to indicate a spatiotemporal correlation among the wind speed data, the irradiance data, and the temperature-sensitive load data;

[0131] A simulation module 203 is configured to perform a double-layer Monte Carlo simulation based on the equipment failure probability data and the joint probability distribution model to obtain a set of weather scenarios and a probability distribution of available resource capacity for each node in the power grid. The outer layer of the double-layer Monte Carlo simulation is configured to generate the set of weather scenarios associated with renewable energy fluctuations based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is configured to construct a device failure event tree for each weather scenario based on the weather scenario set and the equipment failure probability data. The device failure event tree is configured to calculate the probability distribution of available resource capacity for each node under the corresponding weather scenario.

[0132] The adjustment module 204 is used to adjust the response potential trust domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the resource available capacity of each node, and the node operation data.

[0133] Preferably, the meteorological data includes real-time meteorological data, and the node operation data includes node measurement data and node prediction data of the node; the adjustment module 204 is specifically used to perform similarity matching based on the real-time meteorological data and the weather scene set to obtain a matching weather scene subset, and the matching weather scene subset includes the top N weather scenes in the weather scene set that are sorted from high to low in similarity with the real-time meteorological data, where N is a positive integer greater than or equal to 1; based on the matching weather scene subset, the corresponding probability distribution of the available capacity of the resources of each node is obtained; based on the deviation value between the node measurement data and the node prediction data, and the standard deviation of the capacity distribution, the response potential credible domain boundary of the virtual power plant is adjusted; the standard deviation of the capacity distribution is calculated based on the probability distribution of the available capacity of the resources of each node corresponding to the matching weather scene subset.

[0134] Preferably, the adjustment module 204 is specifically configured to calculate a mean deviation of the deviation value based on a sliding time window, where the length of the sliding time window is inversely proportional to the rate of change of the deviation value; when the mean deviation is greater than a preset first threshold, shrinking the response potential credible region boundary based on the capacity distribution standard deviation; the calculation formula for shrinking the response potential credible region boundary includes:

[0135]

[0136] Among them, C L 、C U are the lower and upper bounds of the credible region of the response potential before shrinkage, are the lower and upper bounds of the credible region of the response potential after shrinkage, σ C is the standard deviation of capacity distribution, k1 and k2 are dynamic shrinkage coefficients, is the mean deviation.

[0137] Preferably, the calculation process of the dynamic shrinkage coefficients k1 and k2 includes: when the similarity between the real-time meteorological data and the weather scene is higher than a preset second threshold, obtaining a historical optimal coefficient as the dynamic shrinkage coefficient, and the historical optimal coefficient is calculated based on the historical number of limit violations of the node; when the similarity between the real-time meteorological data and the weather scene is lower than a preset third threshold, updating the dynamic shrinkage coefficient; the calculation formula for updating the dynamic shrinkage coefficient includes:

[0138]

[0139] in, are the i-th dynamic shrinkage coefficient before and after the update, η is the learning rate, ΔRisk is the risk deviation, and sgn() is the sign function.

[0140] Preferably, the joint probability distribution model includes the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data; the simulation module 203 is specifically used to generate the weather scenario set based on the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data through Latin hypercube sampling; based on the weather scenario set and the equipment failure probability data, calculate the probability distribution of the resource available capacity of each node.

[0141] Preferably, the simulation module 203 is specifically used to generate a Markov state transfer matrix of the node under each weather scenario in the weather scenario set based on the weather scenario set and the equipment failure probability data, wherein the elements in the Markov state transfer matrix are the transition probabilities of the states of the devices in the respective nodes; obtain the operating temperature of the devices at the respective nodes; and dynamically correct the transition probabilities based on the operating temperatures, wherein the calculation formula for correcting the transition probabilities includes:

[0142]

[0143] Where γ is the temperature sensitivity coefficient, T ref is the rated operating temperature of the equipment, T is the real-time operating temperature of the equipment, and p ij is the probability of the device transferring from state i to state j after correction, is the basic probability of the device transferring from state i to state j before correction; a device failure event tree is generated based on the corrected Markov state transition matrix; and based on the device failure event tree, the probability distribution of the available capacity of the resources of each node is calculated.

[0144] Preferably, the acquisition module 201 is further configured to acquire the spinning reserve ratio of the power grid; the adjustment module 204 is further configured to adjust the response potential credible region boundary based on the spinning reserve ratio; the calculation formula for adjusting the response potential credible region boundary includes:

[0145]

[0146] Among them, α and β are the adjustment coefficients corresponding to the power grid security level. are the lower and upper bounds of the credible region of response potential after adjusting the spinning reserve ratio, R spin is the spinning reserve ratio.

[0147] The virtual power plant response potential trusted domain construction system provided in the embodiment of the present application can be understood by referring to the corresponding content of the aforementioned method embodiment part, and will not be repeated here.

[0148] like Figure 3 As shown, Figure 3 A possible logical structure diagram of a computing device provided in an embodiment of the present application. The computing device 30 includes: a processor 301, a communication interface 302, a memory 303, and a bus 304. The processor 301, the communication interface 302, and the memory 303 are interconnected via the bus 304. In the embodiment of the present application, the processor 301 is used to control and manage the actions of the computing device 30. For example, the processor 301 is used to execute Figure 1 The steps in the embodiments and / or other processes for the technology described herein. The communication interface 302 is used to support the computing device 30 to communicate. The memory 303 is used to store program codes and data of the computing device 30.

[0149] Among them, the processor 301 can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute the various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and so on. The bus 304 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0150] In another embodiment of the present application, a computer-readable storage medium is further provided, wherein the computer-readable storage medium includes instructions, which, when executed on a computer, causes the computer to execute the above-mentioned Figure 1 The method described in the embodiment.

[0151] Those skilled in the art will appreciate that the units of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. 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 the present invention.

[0152] Those skilled in the art will 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.

[0153] In the several embodiments provided in the embodiments of the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0154] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the purpose of this embodiment.

[0155] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0156] If the functions 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 this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, and other media that can store program codes.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A method for constructing a trusted domain of virtual power plant response potential, characterized in that: include: Acquiring meteorological data, equipment failure probability data of a power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid; the meteorological data includes wind speed data and irradiance data; Establishing a joint probability distribution model based on the wind speed data, the irradiance data, and the temperature-sensitive load data through a Copula function, wherein the joint probability distribution model is used to indicate the spatiotemporal correlation among the wind speed data, the irradiance data, and the temperature-sensitive load data; A double-layer Monte Carlo simulation is performed based on the equipment failure probability data and the joint probability distribution model to obtain a weather scenario set and a probability distribution of resource available capacity for each node in the power grid; wherein the outer layer of the double-layer Monte Carlo simulation is used to generate the weather scenario set related to renewable energy fluctuations based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is used to construct an equipment failure event tree under each weather scenario based on the weather scenario set and the equipment failure probability data, and the equipment failure event tree is used to calculate the probability distribution of resource available capacity for each node under the corresponding weather scenario; Based on the meteorological data, the weather scenario set, the probability distribution of the available capacity of resources of each node, and the node operation data, the response potential trust domain boundary of the virtual power plant is adjusted.

2. The method according to claim 1, characterized in that The meteorological data includes real-time meteorological data, and the node operation data includes node measurement data and node prediction data of the node; adjusting the response potential trust domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the resource available capacity of each node, and the node operation data includes: Performing similarity matching between the real-time meteorological data and the weather scene set to obtain a matching weather scene subset, wherein the matching weather scene subset includes the top N weather scenes in the weather scene set sorted from high to low similarity to the real-time meteorological data, where N is a positive integer greater than or equal to 1; Obtaining a probability distribution of available resource capacity of each corresponding node based on the matching weather scenario subset; Based on the deviation value between the node measurement data and the node prediction data, and the capacity distribution standard deviation, the response potential credible domain boundary of the virtual power plant is adjusted; the capacity distribution standard deviation is calculated based on the probability distribution of the available capacity of resources of each node corresponding to the matching weather scenario subset.

3. The method according to claim 2, characterized in that The adjusting the response potential credible region boundary of the virtual power plant based on the deviation value between the node measurement data and the node prediction data, and the capacity distribution standard deviation, includes: Calculating a deviation mean of the deviation value based on a sliding time window, wherein the length of the sliding time window is inversely proportional to the rate of change of the deviation value; When the deviation mean is greater than a preset first threshold, the response potential credible region boundary is shrunk based on the capacity distribution standard deviation; the calculation formula for shrinking the response potential credible region boundary includes: Among them, C L 、C U are the lower and upper bounds of the credible region of the response potential before shrinkage, are the lower and upper bounds of the credible region of the response potential after shrinkage, σ C is the standard deviation of capacity distribution, k1 and k2 are dynamic shrinkage coefficients, is the mean deviation.

4. The method according to claim 3, characterized in that The calculation process of the dynamic shrinkage coefficients k1 and k2 includes: When the similarity between the real-time meteorological data and the weather scene is higher than a preset second threshold, obtaining a historical optimal coefficient as the dynamic shrinkage coefficient, where the historical optimal coefficient is calculated based on the historical number of limit violations of the node; When the similarity matching degree between the real-time meteorological data and the weather scene is lower than a preset third threshold, the dynamic shrinkage coefficient is updated; the calculation formula for updating the dynamic shrinkage coefficient includes: in, are the i-th dynamic shrinkage coefficient before and after the update, η is the learning rate, ΔRisk is the risk deviation, and sgn() is the sign function.

5. The method according to any one of claims 1 to 4, characterized in that: The joint probability distribution model includes the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data; the double-layer Monte Carlo simulation based on the equipment failure probability data and the joint probability distribution model is performed to obtain a weather scenario set and a probability distribution of resource available capacity of each node in the power grid, including: Generating the weather scene set based on the Weibull probability distribution of the wind speed data and the Beta probability distribution of the irradiance data through Latin hypercube sampling; Based on the weather scenario set and the equipment failure probability data, a probability distribution of available capacity of resources of each node is calculated.

6. The method according to claim 5, characterized in that The calculating, based on the weather scenario set and the equipment failure probability data, the probability distribution of the available capacity of resources of each node includes: Based on the weather scenario set and the equipment failure probability data, generating a Markov state transition matrix for the node under each weather scenario in the weather scenario set, where the elements in the Markov state transition matrix are transition probabilities of the states of the devices in the respective nodes; Obtaining the operating temperature of the device at each node; The transition probability is dynamically corrected based on the operating temperature. The calculation formula for correcting the transition probability includes: Where γ is the temperature sensitivity coefficient, T ref is the rated operating temperature of the equipment, T is the real-time operating temperature of the equipment, and p ij is the probability of the device transferring from state i to state j after correction, is the basic probability of the device transferring from state i to state j before correction; Generate a device fault event tree based on the modified Markov state transition matrix; Based on the equipment failure event tree, a probability distribution of available capacity of resources of each node is calculated.

7. The method according to any one of claims 1 to 4, characterized in that After adjusting the response potential trust region boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the available capacity of resources of each node, and the node operation data, the method further includes: Obtaining a spinning reserve ratio of the power grid; The response potential credible region boundary is adjusted based on the spinning reserve ratio; the calculation formula for adjusting the response potential credible region boundary includes: Among them, α and β are the adjustment coefficients corresponding to the power grid security level. are the lower and upper bounds of the credible region of response potential after adjusting the spinning reserve ratio, R spin is the spinning reserve ratio.

8. A virtual power plant response potential trusted domain construction system, characterized by: The method according to any one of claims 1 to 7, wherein the system comprises: an acquisition module, configured to acquire meteorological data, equipment failure probability data of a power grid, node operation data of the power grid, and temperature-sensitive load data of load nodes in the power grid; the meteorological data includes wind speed data and irradiance data; a modeling module, configured to establish a joint probability distribution model based on the wind speed data, the irradiance data, and the temperature-sensitive load data through a Copula function, wherein the joint probability distribution model is configured to indicate spatiotemporal correlation among the wind speed data, the irradiance data, and the temperature-sensitive load data; a simulation module for performing a double-layer Monte Carlo simulation based on the equipment failure probability data and the joint probability distribution model to obtain a weather scenario set and a probability distribution of resource available capacity for each node in the power grid; wherein the outer layer of the double-layer Monte Carlo simulation is used to generate the weather scenario set related to renewable energy fluctuations based on the joint probability distribution model, and the inner layer of the double-layer Monte Carlo simulation is used to construct an equipment failure event tree for each weather scenario based on the weather scenario set and the equipment failure probability data, and the equipment failure event tree is used to calculate the probability distribution of resource available capacity for each node under the corresponding weather scenario; An adjustment module is used to adjust the response potential trust domain boundary of the virtual power plant based on the meteorological data, the weather scenario set, the probability distribution of the resource available capacity of each node, and the node operation data.

9. A computing device, characterized in that include: Memory, used to store programs; A processor, configured to load the program to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

Images