A market-oriented method for assessing the risk of hydropower station revenue

Abstract

The present invention discloses a market-oriented hydropower station revenue risk assessment method, comprising the following steps: S1. Acquiring real-time hydrological data for the hydropower station basin, constructing a dynamic water energy revenue assessment model, establishing a dynamic mapping relationship between water energy value and power generation revenue, and calculating the water energy value in the current and future periods; S2. Identifying multidimensional risk factors that affect the hydropower station's revenue; S3. Determining the probability distribution of each risk factor, establishing a multi-source risk factor coupling model using a Copula function, and quantifying the comprehensive impact of the interaction between risk factors on the hydropower station's revenue; S4. Randomly simulating the hydropower station's power generation revenue over a period of time in the future to generate a large number of revenue simulation samples; S5. Based on the revenue simulation samples, calculating the risk assessment index of the hydropower station's revenue, and completing the hydropower station revenue risk assessment of the hydropower value. This method achieves dynamic quantitative assessment of hydropower station revenue, multidimensional risk interaction analysis, and full-scenario risk prediction, improving the timeliness and accuracy of revenue assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower station revenue risk assessment, and in particular to a market-oriented hydropower station revenue risk assessment method. Background Art

[0002] In the electricity market, hydropower station revenue is influenced by multiple dynamic factors, necessitating an accurate assessment of revenue risk to optimize operational decisions. Traditional hydropower station revenue assessments often focus on a single risk factor or static data, making them ill-suited to the dynamic, interactive nature of multidimensional risks in the market. These assessments only consider the impact of hydrological runoff fluctuations on power generation, while ignoring the coupling of factors like electricity price fluctuations and equipment failures with hydrological risk. Consequently, these assessments fail to fully reflect the actual risk landscape and meet the demands for revenue stability and risk controllability in market-based operations. Furthermore, traditional methods lack the dynamic integration of real-time data and refined analysis across spatiotemporal scales, making them incapable of forward-looking forecasts of future revenue. This leaves hydropower stations without a scientific basis for responding to market fluctuations and unexpected risks.

[0003] Traditional technical solutions mainly use single-factor analysis models to evaluate the revenue risk of hydropower stations. For example, they predict power generation based on historical runoff data, calculate revenue in combination with fixed electricity prices, and then separately analyze the impact of equipment failure rates on revenue. The advantages of this type of solution are simple model structure and low computational cost, which are suitable for scenarios with small amounts of data or single risk factors. However, its disadvantages are significant: on the one hand, it ignores the interaction between multi-dimensional risk factors such as hydrology, market, and equipment, and cannot quantify the comprehensive impact of risk coupling on revenue, resulting in one-sided evaluation results; on the other hand, it relies on static historical data and lacks the ability to integrate dynamic data such as real-time hydrology and market electricity prices. It is difficult to capture the impact of sudden risk events such as short-term extreme weather and sudden changes in electricity prices. The timeliness and accuracy of the evaluation are insufficient, making it difficult to meet the dynamic management needs of hydropower stations for revenue risks in the power market environment.

[0004] While existing technologies have introduced some multifactor analysis methods, they still have limitations in the dynamic modeling, data fusion, and comprehensive assessment of risk factors. Some studies have used multivariate statistical methods to analyze the correlation of risk factors, but have failed to construct spatiotemporal dynamic coupling models, unable to reflect the evolution of risk factors at different time scales and spatial ranges. Some studies have conducted multivariate joint distribution modeling, but at the data fusion level, they simply overlay historical data with real-time monitoring data, lacking spatiotemporal alignment and feature extraction for multimodal data. This results in insufficient consistency and reliability of model input data. Furthermore, existing technologies often use simple Monte Carlo sampling in the benefit simulation phase, failing to consider the spatiotemporal stratification characteristics of risk factors. The resulting simulation samples are difficult to cover complex risk scenario combinations, limiting the accuracy of risk assessment indicators and failing to provide refined risk management strategies for hydropower stations.

[0005] This technical solution solves the technical problems of insufficient multi-dimensional risk dynamic modeling, insufficient data fusion and incomplete risk scenario coverage in existing technologies by constructing a dynamic assessment model for hydropower benefits and a coupling model for multi-source risk factors. Summary of the Invention

[0006] Based on the above technical problems, this application discloses a market-oriented hydropower station revenue risk assessment method, including:

[0007] S1. Acquire real-time hydrological data for the hydropower station basin, including real-time flow, real-time water level, and real-time hydraulic head. Combined with historical hydrological data and meteorological forecast data, calculate the available amount of water resources at different time scales using a hydrodynamic model, construct a dynamic water energy revenue assessment model, establish a dynamic mapping relationship between water energy value and power generation revenue based on real-time electricity market prices and hydropower station unit power generation efficiency, and calculate the water energy value for the current and future time periods.

[0008] S2. Identify the multi-dimensional risk factors that affect the profitability of hydropower stations, including hydrological risk factors, market risk factors, and equipment risk factors. Hydrological risk factors include runoff fluctuations and abnormal water inflow caused by extreme weather; market risk factors include electricity price fluctuations and changes in electricity demand; and equipment risk factors include unit failures and equipment aging.

[0009] S3. Using historical data and real-time monitoring data, determine the probability distribution of each risk factor, establish a multi-source risk factor coupling model through Copula functions, and quantify the comprehensive impact of the interaction between each risk factor on the hydropower station's revenue;

[0010] S4. Based on the hydropower revenue dynamic assessment model and the multi-source risk factor coupling model, stochastically simulate the power generation revenue of the hydropower station in the future period to generate a large number of revenue simulation samples;

[0011] S5. Based on the revenue simulation sample, calculate the risk assessment indicators of the hydropower station's revenue, wherein the risk assessment indicators include value at risk (VaR), conditional value at risk (CVaR), and risk contribution of water energy value loss, thereby completing the risk assessment of the hydropower station's revenue based on water energy value.

[0012] Preferably, the available amount of water resources at different time scales is calculated by the hydrodynamic model in S1, specifically: the real-time flow , real-time water level , real-time water head As the current moment Real-time hydrological parameters, combined with historical hydrological data and rainfall in meteorological forecast data , evaporation As input parameters, substituting into the hydrodynamic model, the formula is: in, for The amount of water resources available at any given moment, is the dynamic correction coefficient of unit efficiency, is the historical data weight factor, is the meteorological-hydrological coupling influence coefficient, is the discharge in historical hydrological data, is the water head in the historical hydrological data, is the amount of historical data, and the available amount of water resources at different time scales is obtained.

[0013] Preferably, the dynamic mapping relationship between water energy value and power generation income is established in S1, by obtaining Available amount of water resources at any time , combined with the real-time electricity price in the power market , Real-time power generation efficiency of hydropower station units and dynamic loss coefficient ; Construct a time-space weighted benefit-value coupling model to obtain The value of water energy at any moment , the formula is: ,in is the time weight of the peak and valley characteristics of power supply and demand, is the number of time periods divided into within the time period, The first The time weight of each time period is used to obtain the dynamic quantification of the water energy value in the current and future time periods.

[0014] Preferably, the probability distribution of each risk factor is determined by using historical data and real-time monitoring data in S3, specifically: by constructing a multimodal data fusion framework, historical hydrological data, market transaction data, equipment operation and maintenance records are aligned with real-time monitored runoff data, electricity price fluctuation data, and unit operating parameters in time and space; for hydrological risk factors, the potential characteristic distribution of historical runoff sequences is extracted, and dynamic correction is performed in combination with real-time rainfall data; for market risk factors, the time-dependent pattern and abnormal characteristics of electricity price fluctuations are extracted to generate the conditional probability distribution of electricity price fluctuations; for equipment risk factors, a dynamic Bayesian network is constructed through equipment sensor data, and the multidimensional parameters of equipment operating temperature and vibration frequency are used as nodes. The probability transfer relationship between nodes is updated through the particle filtering algorithm to obtain the dynamic probability distribution of risk factors at different time scales.

[0015] Preferably, a multi-source risk factor coupling model is established in S3, specifically: a spatiotemporal Copula-Bayesian fusion framework is constructed to integrate the hydrological risk factors , market risk factors , Equipment Risk Factors Dynamic probability distribution function 、 、 Enter the Copula-Bayesian fusion framework, the formula is: ,in, , 、 For the corresponding spatiotemporal parameters, the joint probability density function of multi-source risk factors is constructed, where is the Copula density function, which models the spatiotemporal dynamic coupling of the interaction of multidimensional risk factors.

[0016] Preferably, a multi-source risk factor coupling model established by a Copula function is used to quantify the comprehensive impact of the interaction between various risk factors on the revenue of the hydropower station, and the marginal distribution functions of hydrological risk factors, market risk factors, and equipment risk factors are substituted into the multi-source risk factor coupling model to generate a joint distribution of each risk factor. By calculating the difference between the joint distribution and the product of the marginal distribution, the nonlinear dependency relationship between the risk factors is obtained; using the conditional probability distribution, the degree of change in the impact of the remaining risk factors on the revenue of the hydropower station when a certain risk factor is in a specific state is analyzed; the joint distribution of each risk factor is combined with the revenue function of the hydropower station, and the revenue distribution under different risk factor combinations is calculated by numerical integration. The variance and skewness of the revenue distribution are used as comprehensive impact quantitative indicators to achieve the quantification of the comprehensive impact of the revenue of the hydropower station under the interaction of multi-source risk factors.

[0017] Preferably, the power generation income of the hydropower station in the future is randomly simulated in S4, specifically: obtaining the water energy value sequence calculated by the dynamic evaluation model of water energy income and the joint distribution function generated by the multi-source risk factor coupling model , generating a multidimensional random variable sequence ,in are random variables representing hydrological, market, and equipment risk factors, and the random variable sequence is substituted into the water energy-benefit mapping model, and the formula is: ,in, for Simulated value of power generation income at each moment, is the real-time scheduling correction coefficient, is the main effect coefficient of the risk factor, is the risk factor interaction effect coefficient, Respectively Risk factors in Random variables at time t, generated by Monte Carlo simulation Independent random paths , constituting the random simulation sample space of power generation revenue.

[0018] Preferably, a large number of revenue simulation samples are generated in the random simulation sample space of power generation revenue in said S4, specifically: a spatiotemporal stratified sampling framework is constructed, the historical fluctuation range of hydrological risk factors is divided into multiple subspaces according to frequency characteristics, market risk factors are divided into time periods according to the electricity price fluctuation cycle, and equipment risk factors are classified according to the operating status level; initial sample points are generated in each risk factor subspace, and the sample points are screened and reorganized through spatiotemporal correlation constraints. For each sample point, the hydropower value is calculated through the hydropower revenue dynamic evaluation model, and the corresponding power generation revenue simulation value is generated in combination with the joint distribution generated by the multi-source risk factor coupling model. K-means clustering analysis is performed on the generated initial revenue simulation samples to obtain a large number of revenue simulation samples covering different risk scenario combinations.

[0019] Preferably, the risk assessment index of the hydropower station income is calculated in S5, specifically: the income simulation samples are sorted from small to large, and for the risk value VaR, according to the preset confidence level , calculate the corresponding quantile value by linear interpolation, if the and The sample values ​​are and ,but , is the total number of samples; for the conditional value at risk CVaR, Based on this, select all values ​​less than or equal to The return simulation sample is used to calculate the mean of the sample by weighted average. The weight is determined according to the probability density of the sample in the Copula joint distribution. ,in For the The probability value of the simulated return sample in the joint distribution, For the The return simulation sample values ​​are used to obtain risk assessment indicators.

[0020] Preferably, the risk contribution of water energy value loss is calculated in S5, specifically: each income simulation sample Compared with the benchmark hydropower value under ideal conditions Compare and calculate the water energy value loss of a single sample , calculate the contribution of each risk factor to the loss of water energy value, the formula is: ,in Indicates the The risk factors for The impact strength of the income simulation sample, is the sensitivity coefficient of water energy value loss, For the The risk contribution of each risk factor to the loss of water energy value, is the total number of return simulation samples, Corresponding to the three risk factors of hydrology, market and equipment respectively, the contribution of each risk factor to the loss of water energy value is obtained.

[0021] Compared with the prior art, the technical solution of this application has the following technical effects:

[0022] The present invention integrates real-time hydrological data (flow, water level, head), historical data and meteorological forecast information (precipitation, evaporation) through a hydrodynamic model, thereby realizing the dynamic calculation of the available amount of water resources at different time scales. At the same time, combined with the real-time electricity price in the power market, the power generation efficiency of the unit and the loss coefficient, a time-space weighted benefit-value coupling model is constructed to quantify the value of water energy in the current and future time periods in real time. This breaks through the limitation of traditional static evaluation that relies on historical data, and can dynamically capture the impact of real-time information such as changes in hydrological conditions and fluctuations in electricity prices on benefits, significantly improving the timeliness and accuracy of benefit evaluation, and providing real-time data support for hydropower stations to cope with short-term market fluctuations and weather changes.

[0023] This invention incorporates hydrological risks (runoff fluctuations, extreme weather), market risks (electricity price fluctuations, demand changes), and equipment risks (unit failures, aging) into a unified analysis framework. Through the Copula function, a multi-source risk factor coupling model is constructed. This model can not only determine the dynamic probability distribution of each risk factor, but also quantify the nonlinear dependencies between risks. This breaks the limitations of independent risk assessment and reveals the spatiotemporal dynamic characteristics of multi-dimensional risk interactions. This makes the assessment results closer to the complex risk scenarios in actual operations and provides a scientific basis for power plants to formulate comprehensive risk prevention and control strategies.

[0024] In the revenue simulation link, the present invention uses a spatiotemporal stratified sampling framework to divide the hydrological fluctuation range, electricity price cycle, and equipment operating status into multiple dimensions, combines Monte Carlo simulation to generate a large number of independent random paths, and uses K-means cluster analysis to cover different risk combination scenarios, avoiding the omission of extreme situations and complex risk combinations in traditional simple sampling. It can simulate revenue fluctuations in all scenarios from routine operation to extreme disasters, allowing power stations to predict the revenue performance under various risk combinations in advance, thereby enhancing the foresight and comprehensiveness of risk response.

[0025] This invention constructs an evaluation indicator system that includes value at risk (VaR), conditional value at risk (CVaR) and risk contribution. VaR and CVaR can quantify the potential maximum loss and average loss under different confidence levels, helping power plants set risk thresholds. The risk contribution clarifies the specific impact of various factors such as hydrology, market, and equipment on value loss by comparing actual returns with ideal operating conditions. It provides power plants with full-chain data support from overall risk measurement to single-factor attribution analysis, promotes risk management from empirical judgment to data-driven, and improves the scientific nature and effectiveness of operational decisions.

[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiment of the present application in conjunction with the accompanying drawings.

[0027] Based on the detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0029] Figure 1 This is a flow chart of a market-oriented hydropower station revenue risk assessment method according to the present invention;

[0030] Figure 2 This is a comparison curve of the hydropower station benefit assessment accuracy in the embodiment;

[0031] Figure 3 A comparison chart of risk assessment indicators for hydropower stations in the embodiment;

[0032] Figure 4 2 is a comparison chart of risk contribution error trends in the embodiment. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. In the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the present application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, for clarity and brevity, the description of known functions and structures has been omitted in the embodiments.

[0034] It should be understood that references throughout this specification to "one embodiment" or "this embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Therefore, the appearance of "one embodiment" or "this embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0035] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0036] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" in this article describes another type of association object relationship, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the previous and subsequent associated objects are in an "or" relationship.

[0037] The term "at least one" in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, at least one of A and B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0038] It should also be noted that, in this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprises," or any other variations thereof are intended to cover non-exclusive inclusion.

[0039] Example 1

[0040] This embodiment mainly describes a market-oriented method for assessing the risk of hydropower station revenue. Figure 1 As shown, specifically including:

[0041] S1. Obtain real-time hydrological data for the hydropower station basin, including real-time flow, real-time water level, and real-time hydraulic head. Combined with historical hydrological data and meteorological forecast data, the available amount of water resources at different time scales is calculated using a hydrodynamic model. A dynamic evaluation model for water energy benefits is constructed. Based on the real-time electricity market price and the power generation efficiency of the hydropower station units, a dynamic mapping relationship between water energy value and power generation benefits is established to calculate the water energy value in the current and future time periods.

[0042] S2. Identify the multi-dimensional risk factors that affect hydropower station revenue. These multi-dimensional risk factors include hydrological risk factors, market risk factors, and equipment risk factors. Hydrological risk factors include runoff fluctuations and abnormal water inflow caused by extreme weather. Market risk factors include electricity price fluctuations and changes in electricity demand. Equipment risk factors include unit failures and equipment aging.

[0043] S3. Using historical data and real-time monitoring data, determine the probability distribution of each risk factor, establish a multi-source risk factor coupling model through Copula functions, and quantify the comprehensive impact of the interaction between each risk factor on the hydropower station's revenue;

[0044] S4. Based on the hydropower revenue dynamic assessment model and the multi-source risk factor coupling model, stochastically simulate the power generation revenue of the hydropower station in the future period to generate a large number of revenue simulation samples;

[0045] S5. Based on the income simulation sample, calculate the risk assessment indicators of the hydropower station income. The risk assessment indicators include risk value VaR, conditional risk value CVaR, and the risk contribution of water energy value loss, and complete the risk assessment of the hydropower station income based on water energy value.

[0046] Furthermore, S1 uses the hydrodynamic model to calculate the available amount of water resources at different time scales, specifically: , real-time water level , real-time water head As the current moment Real-time hydrological parameters, combined with historical hydrological data and rainfall in meteorological forecast data , evaporation As input parameters, substituting into the hydrodynamic model, the formula is: in, for The amount of water resources available at any given moment, is the dynamic correction coefficient of unit efficiency, is the historical data weight factor, is the meteorological-hydrological coupling influence coefficient, is the discharge in historical hydrological data, is the water head in the historical hydrological data, is the amount of historical data, and the available amount of water resources at different time scales is obtained.

[0047] Furthermore, a dynamic mapping relationship between water energy value and power generation income is established in S1, specifically: by obtaining Available amount of water resources at any time , combined with the real-time electricity price in the power market , Real-time power generation efficiency of hydropower station units and dynamic loss coefficient ; Construct a time-space weighted benefit-value coupling model to obtain The value of water energy at any moment , the formula is: ,in is the time weight of the peak and valley characteristics of power supply and demand, is the number of time periods divided into within the time period, The first The time weight of each time period is used to obtain the dynamic quantification of the water energy value in the current and future time periods.

[0048] Furthermore, S3 uses historical data and real-time monitoring data to determine the probability distribution of each risk factor. Specifically, by constructing a multimodal data fusion framework, historical hydrological data, market transaction data, equipment operation and maintenance records are aligned in time and space with real-time monitored runoff data, electricity price fluctuation data, and unit operating parameters; for hydrological risk factors, the potential characteristic distribution of historical runoff sequences is extracted, and dynamic corrections are made in combination with real-time rainfall data; for market risk factors, the time-dependent patterns and abnormal characteristics of electricity price fluctuations are extracted to generate the conditional probability distribution of electricity price fluctuations; for equipment risk factors, a dynamic Bayesian network is constructed through equipment sensor data, and the multidimensional parameters of equipment operating temperature and vibration frequency are used as nodes. The probability transfer relationship between nodes is updated through the particle filtering algorithm to obtain the dynamic probability distribution of risk factors at different time scales.

[0049] Furthermore, a multi-source risk factor coupling model is established in S3, specifically: a spatiotemporal Copula-Bayesian fusion framework is constructed to integrate hydrological risk factors , market risk factors , Equipment Risk Factors Dynamic probability distribution function 、 、 Enter the Copula-Bayesian fusion framework, the formula is: ,in, , 、 For the corresponding spatiotemporal parameters, the joint probability density function of multi-source risk factors is constructed, where is the Copula density function, which models the spatiotemporal dynamic coupling of the interaction of multidimensional risk factors.

[0050] Furthermore, a multi-source risk factor coupling model established through Copula function is used to quantify the comprehensive impact of the interaction between various risk factors on the revenue of the hydropower station. The marginal distribution functions of hydrological risk factors, market risk factors, and equipment risk factors are substituted into the multi-source risk factor coupling model to generate the joint distribution of each risk factor. By calculating the difference between the joint distribution and the product of the marginal distribution, the nonlinear dependence relationship between the risk factors is obtained; the conditional probability distribution is used to analyze the degree of change in the impact of the remaining risk factors on the revenue of the hydropower station when a certain risk factor is in a specific state; the joint distribution of each risk factor is combined with the revenue function of the hydropower station, and the revenue distribution under different risk factor combinations is calculated by numerical integration. The variance and skewness of the revenue distribution are used as comprehensive impact quantitative indicators to achieve the quantification of the comprehensive impact of the hydropower station revenue under the interaction of multi-source risk factors.

[0051] Furthermore, S4 performs a random simulation of the power generation income of the hydropower station in the future, specifically: obtaining the hydropower value sequence calculated by the hydropower income dynamic evaluation model and the joint distribution function generated by the multi-source risk factor coupling model , generating a multidimensional random variable sequence ,in are random variables representing hydrological, market, and equipment risk factors, and the random variable sequence is substituted into the water energy-benefit mapping model, and the formula is: ,in, for Simulated value of power generation income at each moment, is the real-time scheduling correction coefficient, is the main effect coefficient of the risk factor, is the risk factor interaction effect coefficient, Respectively Risk factors in Random variables at time t, generated by Monte Carlo simulation Independent random paths , constituting the random simulation sample space of power generation revenue.

[0052] Furthermore, a large number of revenue simulation samples are generated in the random simulation sample space of power generation revenue in S4, specifically: a spatiotemporal stratified sampling framework is constructed, the historical fluctuation range of hydrological risk factors is divided into multiple subspaces according to frequency characteristics, market risk factors are divided into time periods according to the electricity price fluctuation cycle, and equipment risk factors are classified according to the operating status level; initial sample points are generated in each risk factor subspace, and the sample points are screened and reorganized through spatiotemporal correlation constraints. For each sample point, the hydropower value is calculated through the hydropower revenue dynamic evaluation model, and the corresponding power generation revenue simulation value is generated by combining the joint distribution generated by the multi-source risk factor coupling model. K-means cluster analysis is performed on the generated initial revenue simulation samples to obtain a large number of revenue simulation samples covering different risk scenario combinations.

[0053] Furthermore, the risk assessment index of the hydropower station income is calculated in S5, specifically: the risk assessment index of the hydropower station income is calculated in S5, specifically: the income simulation samples are sorted from small to large, and for the risk value VaR, according to the preset confidence level , calculate the corresponding quantile value by linear interpolation, if the and The sample values ​​are and ,but , is the total number of samples; for the conditional value at risk CVaR, Based on this, select all values ​​less than or equal to The return simulation sample is used to calculate the mean of the sample by weighted average. The weight is determined according to the probability density of the sample in the Copula joint distribution. ,in For the The probability value of the simulated return sample in the joint distribution, For the The return simulation sample values ​​are used to obtain risk assessment indicators.

[0054] Furthermore, the risk contribution of water energy value loss is calculated in S5, specifically: each income simulation sample Compared with the benchmark hydropower value under ideal conditions Compare and calculate the water energy value loss of a single sample , calculate the contribution of each risk factor to the loss of water energy value, the formula is: ,in Indicates the The risk factors for The impact strength of the income simulation sample, is the sensitivity coefficient of water energy value loss, For the The risk contribution of each risk factor to the loss of water energy value, is the total number of return simulation samples, Corresponding to the three risk factors of hydrology, market and equipment respectively, the contribution of each risk factor to the loss of water energy value is obtained.

[0055] This embodiment describes in detail that the present application makes up for the shortcomings of traditional static evaluation by constructing a dynamic evaluation model of hydropower revenue and a coupling model of multi-source risk factors, and establishing a dynamic mapping relationship between revenue and value in combination with real-time electricity prices and unit efficiency; utilizes a multimodal data fusion framework and a Copula-Bayesian fusion model to perform spatiotemporal coupling modeling of the dynamic probability distribution and interaction of hydrological, market, and equipment risk factors, breaking through the limitations of independent analysis of risk factors in the existing technology; generates revenue samples covering multiple risk scenarios through spatiotemporal stratified sampling and Monte Carlo simulation, and combines VaR, CVaR, and risk contribution indicators to achieve a refined quantitative evaluation of revenue risk.

[0056] Based on Example 1, this example describes in detail the specific implementation effects of this application, specifically:

[0057] A medium-sized hydropower station in Southwest China (with an installed capacity of 302.5MW and an average annual power generation of 1.217 billion kWh) was selected as the experimental object. The experimental period was a full year to ensure coverage of the entire hydrological cycle. The data collection frequency was: real-time monitoring of hydrological data (5 minutes / time), minute-level updates of market data, and second-level collection of equipment data. The data collection covered multi-dimensional dynamic information, as follows:

[0058] Hydrological data: Real-time flow: Sensor measured value, the peak flow on August 12, 2023 reached 289.47m³ / s, the minimum flow in the dry season was 0.12m³ / s, and the daily average fluctuation range was 123.5-228.7m³ / s; Water level and head: The average water level in the flood season was 912.43m (head 72.3-81.5m), and the average water level in the dry season was 865.21m (head 21.7-38.4m). Historical runoff data was fitted with P-Ⅲ distribution (mean 150.3m³ / s, potential coefficient Market data: Real-time electricity prices: The average price during peak hours (07:00-11:00, 19:00-23:00) is 0.612 yuan / kWh (standard deviation 0.034), the average price during off-peak hours (23:00-07:00) is 0.281 yuan / kWh (standard deviation 0.021), and the average price during flat hours is 0.423 yuan / kWh; Power demand: Daily load peak is 187.6-245.3MW, the maximum load in summer 2023 is 242.8MW, and the minimum load in winter is 98.5MW.

[0059] Equipment data: Unit vibration frequency: 12.3-35.7Hz under normal operating conditions, fault warning threshold 45Hz, a total of 27 warnings were triggered in 2023 (average duration 4.2 hours); bearing temperature: 42.7-78.5℃ under normal operating conditions, fault threshold 85℃, 82% of historical unplanned shutdown events were accompanied by bearing temperatures >80℃.

[0060] A dynamic assessment model for hydropower benefits was constructed to calculate the available hydropower resources. Taking a typical day, July 15, 2023 (peak load period during the flood season), as an example, the measured flow rate was 278.43 m³ / s and the water level was 915.21 m (head 74.8 m). Combined with the monthly weather forecast (rainfall 178.6 mm, evaporation 119.3 mm), the model calculated the available hydropower resources for that day to be 12,387,450.2 kWh.

[0061] Full-cycle statistics: The daily mean is 10,215,340 kWh (standard deviation 2,118,670 kWh), and the daily mean in the flood season is 41.6% higher than that in the dry season; monthly scale: the highest is in August 2023 (3,256,700 kWh), and the lowest is in January 2024 (8,921,300 kWh). The annual average is 10,234,500 kWh×365=37.356 billion kWh (matching the installed capacity).

[0062] Mapping hydropower value to power generation revenue, calculating real-time revenue: On July 15, 2023, the peak electricity price is 0.623 yuan / kWh, the unit efficiency is 92.37%, the loss coefficient is 3.48%, and the time weight is 1.217. The calculated hydropower value on that day is 6,521,890 yuan. Formula: =12,387,500 kWh × 0.623 yuan / kWh × 92.37% ÷ (1 + 3.48%) × 1.217. After considering the risk factor (equipment mild vibration warning, correction factor 0.98), the power generation income is 6,391,560 yuan;

[0063] Annual revenue distribution: Mean: 321,456,700 yuan (standard deviation: 65,234,800 yuan), deviating 2.3% from the hydropower station's annual budget; Extreme values: 45,876,200 yuan on August 18, 2023 (flood season + peak electricity price + full load), and 12,158,300 yuan on January 5, 2024 (dry season + off-peak electricity price + unit maintenance);

[0064] Multi-source risk factor coupling modeling and risk factor probability distribution construction; Hydrological risk: runoff fluctuation probability density function: In 2023, the measured runoff is 15.68% higher than normal, corresponding to a rightward shift of the probability density peak by 0.12 standard deviations. Market risk: electricity price fluctuations. GARCH model parameters: , the electricity price volatility in Q3 2023 is 25.7%, and the conditional variance mean is 0.015; Equipment risk: The relationship between bearing temperature and failure probability: , when T>70℃, the failure probability at 80℃ is 5.63% (91% consistent with historical data).

[0065] Copula-Bayesian coupled model output, using ClaytonCopula to fit three-dimensional risk, parameter estimation: =1.572 tail dependence strength), =0.889 (time lag coefficient), joint distribution log likelihood value -2345.6 (goodness of fit R²=0.89);

[0066] Correlation between risk factors: The correlation coefficient between hydrology and market risk is 0.681 (electricity prices are often low during the flood season due to excess electricity, with a negative correlation of -0.32), and the correlation coefficient between hydrology and equipment risk is 0.342 (high-load operation increases equipment loss).

[0067] Return simulation and risk assessment, the Monte Carlo simulation process generates N=100,000 samples, and the stratified sampling rules are:

[0068] Hydrology: The runoff is divided into three levels according to the degree of deviation from the mean (>+10%, ±10%, <-10%), accounting for 32%, 41% and 27% of the samples.

[0069] Market: Based on the combination of electricity price period and volatility, high volatility in peak period (>20%) accounts for 15%, and low volatility in valley period (<10%) accounts for 30%.

[0070] Equipment: Based on the combination of vibration frequency and temperature, normal (<35Hz and <75℃) accounted for 65%, warning (35-45Hz or 75-85℃) accounted for 28%, and fault (>45Hz or >85℃) accounted for 7%.

[0071] Simulated extreme scenario revenue: Scenario X (dry season -12% + off-peak electricity price -8% + bearing temperature 88°C): 8,921,500 yuan (a 62% decrease from the baseline value). Scenario Y (flood season +8% + peak electricity price + normal equipment): 48,723,400 yuan (close to the theoretical maximum).

[0072] Risk assessment indicators, 95% confidence level: VaR = 15,623,800 yuan (i.e., the return is less than 15,623,800 yuan under a 5% probability), corresponding to the 5,000th sample percentile of 15,619,400 yuan; CVaR = 12,358,700 yuan (mean of extreme losses, calculated as the weighted average of 5,000 samples).

[0073] Risk contribution: Market risk: 45.7% (peak-valley electricity price difference results in an average annual loss of 14,689,000 yuan); Hydrological risk: 38.8% (dry season income is 11,234,000 yuan lower than the average); Equipment risk: 15.5% (average loss per failure is 215,000 yuan).

[0074] Existing technologies such as Figure 2-4 As shown, experimental comparison is carried out:

[0075] Traditional single-factor model (hydrological assessment only): Power generation is predicted based on runoff data from 2013 to 2022, assuming an electricity price of 0.42 yuan / kWh (average price during flat seasons) and 100% equipment availability. Annual mean revenue: 289,567,000 yuan (9.9% lower than the proposed figure), with a deviation of -10.2% from actual revenue. Extreme value errors: 12.4% underestimation of flood season peak revenue (measured 45.8762 million yuan vs. modeled 40.1230 million yuan); 25.8% overestimation of dry season valley revenue (measured 12.1583 million yuan vs. modeled 15.2890 million yuan). VaR 95% =18,923,400 yuan (overestimated by 21.1%, not taking into account market decline risk).

[0076] The existing multi-factor model (static Copula) uses historical data from 2022 to fit the Copula, and does not update the real-time data for 2023. The standard deviation of the return is RMB 78,956,000 (21.0% higher than this application), because the widening of the peak-to-valley difference after the market-oriented reform of electricity prices in 2023 is ignored; the risk contribution is misjudged: market risk is 30.2% (underestimated by 15.5%), and equipment risk is 22.4% (overestimated by 6.9%).

[0077] The Monte Carlo model (uniform sampling) randomly generates samples and does not distinguish between hydrological cycles and electricity price periods. The scenario coverage rate is only 68.9% (omitting scenarios such as "flood + off-peak electricity prices + equipment warnings"). The average extreme loss simulation value is 18.7% higher. CVaR 95% = 15,421,300 yuan (overestimated by 24.8%) due to insufficient sampling of low-probability events.

[0078] Through the technology of this application: market risk (45.7%) > hydrological risk (38.8%) > equipment risk (15.5%); traditional model: only hydrological risk (100%), completely ignoring the impact of market and equipment; existing multi-factor model: market risk is underestimated, equipment risk is overestimated, reflecting the defects of static modeling.

[0079] This embodiment describes in detail the technology of the present application, which controls the error of profit assessment within ±3% through real-time data updating and spatiotemporal modeling, which is significantly better than the deviation of more than ±10% of traditional models; multi-dimensional risk coupling analysis makes the risk contribution error less than 5%, accurately identifies market risk as the primary factor (45.7%), guides power stations to prioritize the optimization of power trading strategies, and spatiotemporal stratified sampling covers 98.8% of actual risk scenarios. The accuracy of extreme loss assessment is improved by 30%, providing data support for the formulation of emergency plans.

[0080] The above are only preferred embodiments of the present invention, which do not limit the scope of protection of the present invention. For those skilled in the art, the present invention can be modified and varied in various ways. Any changes, modifications, replacements, integrations and parameter changes to these embodiments through conventional substitutions or that can achieve the same functions without departing from the principles and spirit of the present invention fall within the scope of protection of the present invention.

Claims

1. A market-oriented method for assessing the risk of hydropower station revenue, characterized in that: The following steps are involved: S1. Acquire real-time hydrological data for the hydropower station basin, including real-time flow, real-time water level, and real-time hydraulic head. Combined with historical hydrological data and meteorological forecast data, calculate the available amount of water resources at different time scales using a hydrodynamic model, construct a dynamic water energy revenue assessment model, establish a dynamic mapping relationship between water energy value and power generation revenue based on real-time electricity market prices and hydropower station unit power generation efficiency, and calculate the water energy value for the current and future time periods. S2. Identify the multi-dimensional risk factors that affect the profitability of hydropower stations, including hydrological risk factors, market risk factors, and equipment risk factors. Hydrological risk factors include runoff fluctuations and abnormal water inflow caused by extreme weather; market risk factors include electricity price fluctuations and changes in electricity demand; and equipment risk factors include unit failures and equipment aging. S3. Using historical data and real-time monitoring data, determine the probability distribution of each risk factor, establish a multi-source risk factor coupling model through Copula functions, and quantify the comprehensive impact of the interaction between each risk factor on the hydropower station's revenue; S4. Based on the hydropower revenue dynamic assessment model and the multi-source risk factor coupling model, stochastically simulate the power generation revenue of the hydropower station in the future period to generate a large number of revenue simulation samples; S5. Calculate risk assessment indicators for hydropower station revenue based on the revenue simulation sample. The risk assessment indicators include value at risk (VaR), conditional value at risk (CVaR), and risk contribution of water energy value loss, thereby completing the risk assessment of hydropower station revenue based on water energy value. In S1, the available amount of water resources at different time scales is calculated by the hydrodynamic model, and the real-time flow , real-time water level , real-time water head As the current moment Real-time hydrological parameters, combined with historical hydrological data and rainfall in meteorological forecast data , evaporation As input parameters, substituting into the hydrodynamic model, the formula is: in, for The amount of water resources available at any given moment, is the dynamic correction coefficient of unit efficiency, is the historical data weight factor, is the meteorological-hydrological coupling influence coefficient, is the discharge in historical hydrological data, is the water head in the historical hydrological data, is the amount of historical data, and the available amount of water resources at different time scales is obtained; In S1, a dynamic mapping relationship between water energy value and power generation income is established, and by obtaining Available amount of water resources at any time , combined with the real-time electricity price in the power market , Real-time power generation efficiency of hydropower station units and dynamic loss coefficient ; Construct a time-space weighted benefit-value coupling model to obtain The value of water energy at any moment , the formula is: ,in is the time weight of the peak and valley characteristics of power supply and demand, is the number of time periods divided into within the time period, The first The time weight of each time period is used to obtain the dynamic quantification of the water energy value in the current and future time periods.

2. A market-oriented hydropower station revenue risk assessment method according to claim 1, characterized in that: In the aforementioned S3, historical data and real-time monitoring data are used to determine the probability distribution of each risk factor. Specifically, by constructing a multimodal data fusion framework, historical hydrological data, market transaction data, equipment operation and maintenance records are aligned in time and space with real-time monitored runoff data, electricity price fluctuation data, and unit operating parameters; for hydrological risk factors, the potential characteristic distribution of historical runoff sequences is extracted, and dynamic correction is performed in combination with real-time rainfall data; for market risk factors, the time-dependent patterns and abnormal characteristics of electricity price fluctuations are extracted to generate the conditional probability distribution of electricity price fluctuations; for equipment risk factors, a dynamic Bayesian network is constructed through equipment sensor data, and the multidimensional parameters of equipment operating temperature and vibration frequency are used as nodes. The probability transfer relationship between nodes is updated through the particle filtering algorithm to obtain the dynamic probability distribution of risk factors at different time scales.

3. A market-oriented hydropower station revenue risk assessment method according to claim 1 or 2, characterized in that: The multi-source risk factor coupling model is established in S3, specifically: a spatiotemporal Copula-Bayesian fusion framework is constructed to integrate the hydrological risk factors , market risk factors , Equipment Risk Factors Dynamic probability distribution function 、 、 Enter the Copula-Bayesian fusion framework, the formula is: ,in, , 、 For the corresponding spatiotemporal parameters, the joint probability density function of multi-source risk factors is constructed, where is the Copula density function, which models the spatiotemporal dynamic coupling of the interaction of multidimensional risk factors.

4. A market-oriented hydropower station revenue risk assessment method according to claim 3, characterized in that: A multi-source risk factor coupling model established through Copula function is used to quantify the comprehensive impact of the interaction between various risk factors on the hydropower station's revenue. The marginal distribution functions of hydrological risk factors, market risk factors, and equipment risk factors are substituted into the multi-source risk factor coupling model to generate the joint distribution of each risk factor. By calculating the difference between the joint distribution and the product of the marginal distribution, the nonlinear dependence between the risk factors is obtained. The conditional probability distribution is used to analyze the degree of change in the impact of the remaining risk factors on the hydropower station's revenue when a certain risk factor is in a specific state. The joint distribution of each risk factor is combined with the hydropower station's revenue function, and the revenue distribution under different risk factor combinations is calculated through numerical integration. The variance and skewness of the revenue distribution are used as comprehensive impact quantitative indicators to quantify the comprehensive impact of the hydropower station's revenue under the interaction of multi-source risk factors.

5. The market-oriented hydropower station revenue risk assessment method according to claim 1 is characterized in that: In S4, the power generation income of the hydropower station in the future is randomly simulated, specifically: obtaining the water energy value sequence calculated by the dynamic evaluation model of water energy income and the joint distribution function generated by the multi-source risk factor coupling model , generating a multidimensional random variable sequence ,in are random variables representing hydrological, market, and equipment risk factors, and the random variable sequence is substituted into the water energy-benefit mapping model, and the formula is: ,in, for Simulated value of power generation income at each moment, is the real-time scheduling correction coefficient, is the main effect coefficient of the risk factor, is the risk factor interaction effect coefficient, Respectively Risk factors in Random variables at time t, generated by Monte Carlo simulation Independent random paths , constituting the random simulation sample space of power generation revenue.

6. A market-oriented hydropower station revenue risk assessment method according to claim 5, characterized in that: In the S4, a large number of revenue simulation samples are generated in the random simulation sample space of power generation revenue, specifically: a spatiotemporal stratified sampling framework is constructed, the historical fluctuation range of hydrological risk factors is divided into multiple subspaces according to frequency characteristics, market risk factors are divided into time periods according to the electricity price fluctuation cycle, and equipment risk factors are classified according to the operating status level; initial sample points are generated in each risk factor subspace, and the sample points are screened and reorganized through spatiotemporal correlation constraints. For each sample point, the hydropower value is calculated through the hydropower revenue dynamic evaluation model, and the corresponding power generation revenue simulation value is generated in combination with the joint distribution generated by the multi-source risk factor coupling model. K-means cluster analysis is performed on the generated initial revenue simulation samples to obtain a large number of revenue simulation samples covering different risk scenario combinations.

7. The market-oriented hydropower station revenue risk assessment method according to claim 1 is characterized in that: The risk assessment index of the hydropower station income is calculated in S5, specifically: the income simulation samples are sorted from small to large, and for the risk value VaR, according to the preset confidence level , calculate the corresponding quantile value by linear interpolation, if the and The sample values ​​are and ,but , is the total number of samples; for the conditional value at risk CVaR, Based on this, select all values ​​less than or equal to The return simulation sample is used to calculate the mean of the sample by weighted average. The weight is determined according to the probability density of the sample in the Copula joint distribution. ,in For the The probability value of the simulated return sample in the joint distribution, For the A return simulation sample is used to obtain risk assessment indicators.

8. A market-oriented hydropower station revenue risk assessment method according to claim 1 or 7, characterized in that: The risk contribution of water energy value loss is calculated in S5 as follows: Compared with the benchmark hydropower value under ideal conditions Compare and calculate the water energy value loss of a single sample , calculate the contribution of each risk factor to the loss of water energy value, the formula is: ,in Indicates the The risk factors for The impact strength of the income simulation sample, is the sensitivity coefficient of water energy value loss, For the The risk contribution of each risk factor to the loss of water energy value, is the total number of return simulation samples, Corresponding to the three risk factors of hydrology, market and equipment respectively, the contribution of each risk factor to the loss of water energy value is obtained.

Images