Pumped storage safety operation monitoring management system and method based on digital twinning technology

Abstract

The present application relates to the technical field of energy equipment management, industrial automation control and information physical system, in particular to a pumped storage safety operation monitoring management system and method based on digital twin technology, comprising: S1, collecting unit physical state parameters and market environment economic signals in real time; S2, calculating marginal physical loss cost representing single variable condition operation based on unit physical state parameters; S3, accumulating historical marginal physical loss cost to calculate health depreciation index; S4, correcting marginal physical loss cost by using dynamic penalty factor to generate corrected marginal physical loss cost; S5, calculating corrected risk-adjusted yield by combining corrected marginal physical loss cost; S6, determining risk level of scheduling opportunity and executing graded scheduling control corresponding to risk level. The present application overcomes the defect that economic benefits and long-term health of equipment are separated in traditional decision-making, and realizes accurate evaluation of implicit loss.

Description

Technical Field

[0001] This invention relates to the fields of energy equipment management, industrial automation control, and cyber-physical systems, specifically to a pumped storage safety operation monitoring and management system and method based on digital twin technology. Background Technology

[0002] In modern power systems, pumped storage units are key assets for achieving grid stability and market arbitrage. Their operation involves frequent and rapid changes in operating conditions. Each dispatch operation brings immediate economic benefits, but also causes physical stress and fatigue damage to key components of the unit, resulting in long-term hidden costs.

[0003] Traditional operation and management methods generally suffer from a disconnect between short-term economic scheduling and long-term asset health management; their decision-making models mainly focus on maximizing market returns, but lack a mechanism to accurately quantify the physical losses caused by a single operation and incorporate them into economic assessments, resulting in the neglect of equipment losses as an implicit cost in real-time decision-making.

[0004] Existing technologies rely heavily on offline, experience-based operation and maintenance strategies, failing to establish a dynamic feedback loop that allows for the adaptation of accumulated deterioration in unit health status to current operational decisions. This prevents operational strategies from adaptively adjusting to the actual current health level of the equipment, potentially leading to excessive consumption of unit lifespan in pursuit of short-term high profits, increasing the risk of unplanned downtime, and damaging the full lifecycle value of assets.

[0005] Therefore, how to construct a monitoring and management method that can quantify the physical loss cost of pumped storage units in real time and combine it with market benefits, and ultimately achieve a dynamic balance between unit operation risks and economic benefits, is a technical problem that urgently needs to be solved in this field.

[0006] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention discloses a pumped storage safety operation monitoring and management system and method based on digital twin technology. Specifically, the technical solution of this invention is as follows:

[0008] The pumped storage safety operation monitoring and management method based on digital twin technology includes:

[0009] S1. Real-time acquisition of unit physical status parameters and market environment economic signals;

[0010] S2. Based on the unit's physical state parameters, the marginal physical loss cost characterizing a single variable operating condition operation is calculated.

[0011] S3. Accumulate historical marginal physical loss costs and combine them with the preset total replacement cost of the unit to calculate the health depreciation index.

[0012] S4. Construct a dynamic penalty factor based on the health depreciation index, and use the dynamic penalty factor to correct the marginal physical loss cost to generate the corrected marginal physical loss cost.

[0013] S5. Determine the direct market return based on market environment economic signals, and calculate the adjusted risk-adjusted rate of return by combining the adjusted marginal physical loss cost.

[0014] S6. Compare the revised risk-adjusted rate of return with the preset coordination limit to determine the risk level of the scheduling opportunity, and execute the hierarchical scheduling control corresponding to the risk level.

[0015] Preferably, S1 specifically includes:

[0016] The physical condition parameters of the generating unit include the vibration intensity, temperature, and water flow pressure of key components such as the unit bearings and stator windings; the market environment economic signals include real-time grid connection price, pumping power price, and grid frequency.

[0017] Preferably, S2 specifically includes:

[0018] S21. Use the normalized reference value to perform dimensionless normalization on the physical state parameters of the unit to generate normalized physical quantities;

[0019] S22. Based on the normalized physical quantity, combined with the preset weighting coefficient and damage index, the comprehensive physical damage is calculated.

[0020] S23. Determine the monetization conversion factor for physical losses, wherein the monetization conversion factor is determined by dividing the total replacement cost of the unit by the design total fatigue damage life limit.

[0021] S24. Multiply the total physical damage by the monetization conversion factor to calculate the marginal physical loss cost.

[0022] Preferably, S5 specifically includes:

[0023] S51. Based on the real-time grid connection price, pumping price, regulation power and duration, the direct market revenue is calculated.

[0024] S52. Subtract the adjusted marginal physical loss cost from the direct market revenue, and divide the difference by the direct market revenue to generate the adjusted risk-adjusted return.

[0025] Preferably, S4 specifically includes:

[0026] S41. Determine the health status influencing factors, wherein the health status influencing factors are determined by statistical analysis of the failure rate data of the same type of units at different aging stages.

[0027] S42. Multiply the health depreciation index by the health status impact factor, and add 1 to the product to construct a dynamic penalty factor.

[0028] S43. Using a dynamic penalty factor, multiply the currently calculated marginal physical loss cost to generate a corrected marginal physical loss cost.

[0029] Preferably, S6 specifically includes:

[0030] The revised risk-adjusted return is compared with the pre-set harmonized limit;

[0031] S61. When the corrected risk-adjusted return is greater than or equal to twice the preset coordination limit, it is defined as a value range, and the original scheduling instruction is executed.

[0032] S62. When the corrected risk-adjusted return is greater than or equal to the preset coordination limit but less than twice the preset coordination limit, it is defined as a level one risk arbitrage range. The power adjustment rate in the original scheduling instruction is automatically adjusted to generate a risk mitigation scheme.

[0033] S63. When the corrected risk-adjusted return is greater than or equal to 0 and less than the preset coordination limit, it is defined as a secondary risk arbitrage range. A high physical loss risk warning is submitted to the dispatch center, and an alternative solution to reduce the response intensity is recommended.

[0034] Preferred options also include:

[0035] When the corrected marginal physical loss cost exceeds the preset absolute upper limit of single operation risk, a rejection instruction is sent to the dispatch terminal, and an enhanced monitoring process for key components of the unit is automatically triggered. The enhanced monitoring process includes increasing the data acquisition frequency of sensors and prioritizing the maintenance of marked components.

[0036] The pumped storage safety operation monitoring and management system based on digital twin technology includes the following modules:

[0037] The data acquisition module is used to collect real-time physical status parameters of the unit and economic signals from the market environment.

[0038] The cost calculation module is used to calculate the marginal physical loss cost characterizing a single variable operating condition operation based on the unit's physical state parameters.

[0039] The condition assessment module is used to accumulate historical marginal physical loss costs and, in combination with the preset total replacement cost of the unit, calculate the health depreciation index.

[0040] The closed-loop correction module is used to construct a dynamic penalty factor based on the health depreciation index, and to use the dynamic penalty factor to correct the marginal physical loss cost, thereby generating the corrected marginal physical loss cost.

[0041] The decision analysis module is used to determine direct market returns based on market environment economic signals, and calculate the corrected risk-adjusted rate of return by combining the corrected marginal physical loss cost.

[0042] The scheduling control module is used to compare the corrected risk-adjusted rate of return with the preset coordination limit, determine the risk level of the scheduling opportunity, and execute the hierarchical scheduling control corresponding to the risk level.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. This invention quantifies the abstract physical stress and fatigue damage during unit operation into specific marginal physical loss costs, overcoming the shortcomings of traditional decision-making where economic benefits and long-term equipment health are separated, and achieving accurate assessment of hidden losses.

[0045] 2. This invention constructs a closed-loop feedback mechanism from the long-term health status of the unit to the short-term decision-making cost. Through dynamic penalty factors, the decision-making system can adaptively adjust its risk preference according to the different stages of the unit's life cycle. Units with poor health status will introduce a protective conservative tendency in decision-making.

[0046] 3. This invention establishes a modified risk-adjusted rate of return as a unified decision-making indicator, evaluating both immediate market returns and health-adjusted internal attrition costs within the same framework. This enables each scheduling decision to achieve a dynamic balance between pursuing short-term economic gains and safeguarding long-term asset value.

[0047] 4. This invention realizes intelligent hierarchical scheduling and control based on risk levels. It not only enables differentiated responses according to the risk-reward ratio but also ensures the fundamental safety of the unit by setting an absolute risk ceiling. This significantly improves the safety and sustainability of pumped storage power station operation, effectively reduces the risk of unplanned shutdowns due to excessive losses, and thus maximizes the comprehensive benefits throughout the unit's entire life cycle. Attached Figure Description

[0048] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0049] Figure 1 This is a flowchart of the method of the present invention.

[0050] Figure 2 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0052] Example 1:

[0053] Please see Figure 1 The method for monitoring and managing the safe operation of pumped storage tanks based on digital twin technology includes the following steps:

[0054] S1. Real-time acquisition of unit physical status parameters and market environment economic signals;

[0055] S2. Based on the unit's physical state parameters, the marginal physical loss cost characterizing a single variable operating condition operation is calculated.

[0056] S3. Accumulate historical marginal physical loss costs and combine them with the preset total replacement cost of the unit to calculate the health depreciation index.

[0057] S4. Construct a dynamic penalty factor based on the health depreciation index, and use the dynamic penalty factor to correct the marginal physical loss cost to generate the corrected marginal physical loss cost.

[0058] S5. Determine the direct market return based on market environment economic signals, and calculate the adjusted risk-adjusted rate of return by combining the adjusted marginal physical loss cost.

[0059] S6. Compare the revised risk-adjusted rate of return with the preset coordination limit to determine the risk level of the scheduling opportunity, and execute the hierarchical scheduling control corresponding to the risk level.

[0060] This invention provides a method for monitoring and managing the safe operation of pumped-storage units based on digital twin technology. The aim is to maximize the life-cycle value of pumped-storage units by constructing a closed-loop feedback mechanism between physical losses and economic benefits. This invention relies on a digital twin platform for pumped-storage power stations. This platform fuses a high-fidelity physical model with real-time data collected by sensors to construct a virtual mirror image that operates synchronously with the physical unit. The physical state parameters of the unit and market environment economic signals are supplied in real-time by the data acquisition module of the digital twin platform. Subsequent calculations of key indicators such as marginal physical loss costs and health depreciation indices are all completed within the digital twin model, thereby achieving accurate monitoring, evaluation, and closed-loop control of the physical unit's status. This method creates a complete technical closed loop: from real-time data acquisition to the quantification of single-operation costs, to the assessment of long-term health status, and then to correcting current decisions through a dynamic feedback mechanism. Finally, differentiated control strategies are implemented based on the corrected comprehensive evaluation indicators.

[0061] The specific implementation steps of this method are as follows:

[0062] S1: Real-time acquisition of unit physical state parameters and market environment economic signals; the purpose of this step is to provide real-time and accurate data input for subsequent physical loss quantification and market benefit assessment; in this embodiment, through the data acquisition module deployed on the pumped storage power station digital twin platform, two types of core data are continuously acquired via industrial bus interface or wireless sensor network; the first type is unit physical state parameters, which are the direct basis for judging the unit's operating stress; the second type is market environment economic signals, which are the direct basis for assessing the economic value of operation;

[0063] S2: Based on the unit's physical state parameters, calculate the marginal physical loss cost characterizing a single variable operating condition operation; the purpose of this step is to quantify the abstract, multi-dimensional physical stress into a specific, tangible economic cost, namely the marginal physical loss cost. In this embodiment, the calculation first normalizes parameters with different physical dimensions and integrates them into a dimensionless comprehensive physical damage measure. Then through a monetization conversion factor Map it to economic cost ;

[0064] S3: Accumulate historical marginal physical depreciation costs and combine them with a preset total unit replacement cost to calculate a health depreciation index; the purpose of this step is to track and assess the asset health of the unit from a long-term perspective and generate a health depreciation index. In this embodiment, the system continuously records and accumulates the data of each operation calculated by S2 in the database. Value; by comparing this accumulated total loss cost with the total replacement cost of the unit, which represents the initial total value of the unit. By comparing the values, a percentage is obtained; this index differs from the traditional accounting method of depreciation over a fixed period of time, and is based on the physical losses generated during actual operation, thus more realistically reflecting the degree of consumption of the unit's effective lifespan.

[0065] S4: Construct a dynamic penalty factor based on the health depreciation index, and use the dynamic penalty factor to correct the marginal physical loss cost, generating the corrected marginal physical loss cost; this step is the core of the closed-loop control of this invention, and its purpose is to establish a dynamic feedback mechanism from long-term health status to short-term decision-making costs; in this embodiment, the value obtained in S3 is used... Construct a dynamic penalty factor that increases as health deteriorates; apply this factor to the current condition calculated by S2. To obtain a corrected cost This results in a higher cumulative wear and tear on a single unit. A higher-performing unit will calculate a higher cost of loss than a healthy unit when performing the same operation, thus automatically introducing a protective conservative tendency in decision-making.

[0066] S5: Determine direct market returns based on market environment economic signals, and calculate the adjusted risk-adjusted rate of return by combining the adjusted marginal physical loss cost; the purpose of this step is to conduct a final risk-return assessment within a unified framework and generate the adjusted risk-adjusted rate of return based on the decision indicator. In this embodiment, the direct market revenue of a single dispatch operation is first calculated based on real-time market signals such as the on-grid electricity price and the pumping power price. Then, the result obtained from S4 from Deduct from the net income and combine it with the net income. In comparison, the final rate of return is obtained; this indicator aims to intuitively quantify the extent to which the returns from this operation can cover the core economic relationship of the health-adjusted losses to the body.

[0067] S6: Compare the revised risk-adjusted rate of return with a preset coordination limit to determine the risk level of the scheduling opportunity, and execute hierarchical scheduling control corresponding to the risk level; the purpose of this step is to transform the analysis results of the previous steps into specific, executable control instructions; in this embodiment, a coordination limit is preset. It represents the minimum profit threshold that power plant managers are willing to accept; calculated by S5 and By comparing the risk levels of the opportunities with their multiples, market opportunities are classified into different risk levels, and corresponding automated scheduling strategies are triggered, such as executing as originally planned, executing after parameter optimization, or issuing warnings and recommending rejection.

[0068] Through the aforementioned steps, this invention constructs a complete decision-making system encompassing data perception, cost quantification, long-term evaluation, closed-loop correction, and hierarchical control. It overcomes the shortcomings of existing technologies where operational decisions are separated from long-term equipment health. By costing physical losses and establishing a feedback mechanism dynamically linked to the unit's health status, it ensures that every scheduling decision achieves a dynamic balance between pursuing immediate economic benefits and safeguarding long-term asset value. This method significantly improves the overall economic benefits of pumped-storage units throughout their entire lifecycle, while effectively reducing the risk of unplanned shutdowns due to excessive losses, thus enhancing the safety and sustainability of power plant operation.

[0069] Example 2:

[0070] S1 specifically includes:

[0071] The physical condition parameters of the generating unit include the vibration intensity, temperature, and water flow pressure of key components such as the unit bearings and stator windings; the market environment economic signals include real-time grid connection price, pumping power price, and grid frequency;

[0072] Based on Example 1, this embodiment specifies the real-time acquisition of unit physical state parameters and market environment economic signals in S1 to ensure the relevance and effectiveness of the model input.

[0073] Specifically, the purpose of collecting parameters characterizing the physical state of the unit in S1 is to capture the core physical processes that contribute most to the fatigue damage of the unit; in this embodiment, the physical state parameters of the unit include at least:

[0074] Vibration intensity of key components This refers to the vibration velocity or acceleration values ​​collected in real time by vibration sensors installed at key locations such as the unit's bearings and top cover; its function is to characterize the level of mechanical stress that the unit experiences during operation.

[0075] temperature This refers to the temperature values ​​of key components such as stator windings and bearing bearings monitored by temperature sensors; its function is to characterize the level of thermal stress borne by the unit.

[0076] Water flow pressure This refers to monitoring the water pressure values ​​at hydraulic components such as the spiral casing and tailrace pipe; its function is to characterize the magnitude of the hydraulic load borne by the unit.

[0077] The purpose of collecting market environment economic signals in S1, which characterize the market environment, is to accurately capture the external economic incentives that drive the pumped-storage units to engage in arbitrage operations. In this embodiment, the market environment economic signals mainly include:

[0078] Real-time electricity price This refers to the immediate price at which electricity is sold to the grid when it is generated;

[0079] Pumping electricity price This refers to the immediate price at which electricity is purchased from the grid for water pumping.

[0080] Grid frequency This refers to the real-time operating frequency of the power grid, and its fluctuations are a key indicator for determining whether the power grid requires emergency frequency response services.

[0081] By explicitly limiting the collection of the aforementioned specific physical and economic parameters, this invention ensures the criticality and directness of the model input. The selected physical parameters are the most important factors leading to unit fatigue damage, and the selected market environment economic signals are the core variables determining the unit's profit margin. This precise selection of data sources greatly improves the accuracy of subsequent marginal physical loss cost calculations and the authenticity of market return assessments, thereby making the foundation of the entire decision-making system more solid and the final output dispatch instructions more reliable and targeted.

[0082] Example 3:

[0083] S2 specifically includes:

[0084] S21. Use the normalized reference value to perform dimensionless normalization on the physical state parameters of the unit to generate normalized physical quantities;

[0085] S22. Based on normalized physical quantities, combined with preset weighting coefficients and damage indices, the comprehensive physical damage is calculated; this damage integrates the mechanical stress, thermal stress, hydraulic load and vibration effects brought about by the unit's variable operating conditions.

[0086] S23. Determine the monetization conversion factor for physical losses, wherein the monetization conversion factor is determined by dividing the total replacement cost of the unit by the design total fatigue damage life limit.

[0087] S24. Multiply the total physical damage by the monetization conversion factor to calculate the marginal physical loss cost.

[0088] Based on Example 1, this embodiment calculates the marginal physical loss cost characterizing a single variable operating condition operation based on the unit's physical state parameters. The internal technical process; this process, through a series of rigorous mathematical transformations, realizes the mapping from multidimensional physical signals to one-dimensional economic costs;

[0089] S21: The unit's physical state parameters are normalized using a normalized reference value to generate normalized physical quantities. This step aims to eliminate the incomparability issues caused by differences in dimensions and numerical ranges between different physical quantities. In this embodiment, a normalized reference value is introduced. These refer to the rated or maximum permissible rate of change that the equipment can withstand, as specified in the unit's design specifications; for example, the power regulation rate. By dividing by the reference power adjustment rate Receive dimensionless term ;

[0090] S22: Based on normalized physical quantities, combined with preset weighting coefficients and damage indices, the comprehensive physical damage is calculated. The purpose of this step is to fuse multiple normalized physical stress components into a unified, dimensionless damage metric. In this embodiment, the comprehensive physical damage is calculated using the following formula. :

[0091]

[0092] in, The total physical damage from a single operation, dimensionless;

[0093] These are power change, adjustment time, and temperature change rate, respectively, and all data are real-time monitoring values ​​from the sensors.

[0094] : These are the normalized reference values ​​for power regulation rate and temperature change rate, respectively, from the unit design specification;

[0095] Weighting coefficient, dimensionless; its function is to reflect the different contributions of different types of physical stress to the total damage; it is determined by multivariate regression analysis of historical unit failure data or by finite element simulation analysis.

[0096] Damage index, dimensionless; its function is to reflect the sensitivity of the core component material to a specific stress cycle; its value is derived from the fitting of fatigue test data of the core component material and is an inherent property of the material; to further clarify, parameters... The calibration process is as follows: A series of fatigue tests are conducted on material samples of the unit's core components to obtain data at different stress levels. Fatigue life and number of cycles Data points; based on these experimental data points By employing regression analysis techniques such as the least squares method, the SN curve equation of the material is fitted, thereby determining the damage index that characterizes the fatigue properties of the material. ;

[0097] S23: Determine the monetization conversion factor for physical losses. The purpose of this step is to establish a bridge between dimensionless physical damage and dimensional monetary costs; in this embodiment, the monetization conversion factor... The calibration method is as follows:

[0098]

[0099] in, Total replacement cost of the unit, expressed in monetary units; equipment purchase contract or latest asset appraisal report;

[0100] Total fatigue damage life limit (TGF-β), dimensionless; provided by the equipment manufacturer according to relevant design specifications, represents the total amount of cumulative physical damage that the unit can withstand before it is scrapped.

[0101] S24: Multiply the total physical damage by the monetization conversion factor to calculate the marginal physical loss cost. This is the final step in cost calculation.

[0102]

[0103] The detailed calculation process disclosed in this embodiment provides a scientific method with clear physical meaning and engineering basis for determining marginal physical loss cost; through normalization (S21), introduction of materials science damage models (S22), and calibration combined with asset value (S23), the final calculated cost is... (S24) is no longer an empirical estimate, but a comprehensive quantitative indicator rooted in physics, materials science and economics; this greatly improves the accuracy and credibility of cost assessment and provides a solid foundation for the scientific nature of the entire decision-making system.

[0104] Example 4:

[0105] S5 specifically includes:

[0106] S51. Based on the real-time grid connection price, pumping price, regulation power and duration, the direct market revenue is calculated.

[0107] S52. Subtract the adjusted marginal physical loss cost from the direct market revenue, and divide the difference by the direct market revenue to generate the adjusted risk-adjusted return.

[0108] This embodiment, based on Embodiment 1, calculates the modified risk-adjusted rate of return. The specific mathematical implementation;

[0109] S51: Calculate the direct market revenue based on the real-time grid connection price, pumping price, regulation power, and duration. The purpose of this step is to quantify the gross profit that a single dispatch operation can directly obtain from the electricity market; in this embodiment, direct market revenue... The calculation formula is:

[0110]

[0111] in, These are the real-time grid connection electricity price and the pumping electricity price, respectively, which are obtained in real time by the data acquisition module; to ensure unit matching, The unit is usually currency / energy, such as yuan / kWh. The unit is power, such as kW. The unit is time, such as h, thus ensuring The final unit is currency;

[0112] Adjust power;

[0113] Duration;

[0114] S52: Subtract the adjusted marginal physical loss cost from the direct market return, and divide the difference by the direct market return to generate the adjusted risk-adjusted return. The purpose of this step is to fully account for health-adjusted internal attrition costs when evaluating returns, thereby obtaining a net return rate metric; in this embodiment, the corrected risk-adjusted return rate... The calculation formula is:

[0115]

[0116] To ensure the robustness of the model, it is necessary to... The value of is handled in special cases. When direct market returns... Greater than a preset minimum positive threshold When, the above formula is used to calculate the rate of return; when ,include In situations where free grid ancillary services are provided, the decision-making metric should be switched to absolute net revenue. At this point, the scheduling control module (S6) will compare the absolute net profit value with the preset cost threshold, rather than the rate of return, thereby avoiding the calculation failure caused by the denominator being zero or too small, and ensuring that the model behaves logically under all possible inputs.

[0117] The underlying logic lies in the molecules. This represents the net profit from the operation; compare it with the denominator. Dividing (total revenue) by the total revenue essentially calculates the net profit margin of this operation; this normalization method makes the assessment results more accurate. Regardless of the total scale of the operation, only the profitability and risk-cost ratio of the operation itself are evaluated, which facilitates a fair comparison among different opportunities;

[0118] Defined through this specific mathematical structure This invention provides a standardized decision indicator with clear economic significance; it simplifies the complex relationship between market returns and internal attrition costs into an intuitive dimensionless ratio; this makes the decision-making process simple, efficient and highly consistent, greatly improving the robustness and reliability of automated decision-making systems.

[0119] Example 5:

[0120] S4 specifically includes:

[0121] S41. Determine the health status influencing factors, wherein the health status influencing factors are determined by statistical analysis of the failure rate data of the same type of units at different aging stages.

[0122] S42. Multiply the health depreciation index by the health status impact factor, and add 1 to the product to construct a dynamic penalty factor.

[0123] S43. Using a dynamic penalty factor, multiply the currently calculated marginal physical loss cost to generate a corrected marginal physical loss cost.

[0124] This embodiment, based on Embodiment 1, details the technical process for dynamically correcting marginal physical loss costs; this is a key step in achieving system adaptability and long-term health management.

[0125] S41: Identify factors influencing health status The purpose of this step is to set a suitable sensitivity for the system's feedback loop; health status influencing factors. It is a dimensionless parameter whose function is to regulate cumulative damage. The intensity of the impact on the current cost calculation; in this embodiment, The calibration of values ​​is determined through statistical analysis of historical operating data of this power station or similar units; specifically, this calibration process can collect data on different units at different cumulative losses. Failure rate data below Through analysis and By fitting the statistical correlation between the two factors, a health status influencing factor that reflects the strength of the association can be obtained. ;

[0126] S42: Multiply the health depreciation index by the health status impact factor, and then add 1 to the product to construct a dynamic penalty factor; the purpose of this step is to generate a correction coefficient that monotonically increases as health deteriorates; in this embodiment, the mathematical expression of the dynamic penalty factor is:

[0127]

[0128] This 1+x structural design ensures that when the unit is in a brand new state, The penalty factor is close to 0, and the penalty factor is close to 1, without affecting the original cost. This will have an impact; and as the unit ages, As the penalty factor increases, the cost of loss increases linearly; this linear model is used in this embodiment. In more refined models, nonlinear structures can also be used to better reflect the accelerated aging risk at the end of the lifespan, for example: or ,in ;

[0129] S43: Employ a dynamic penalty factor to multiply the currently calculated marginal physical loss cost, generating a corrected marginal physical loss cost. This is the final step in performing the correction.

[0130]

[0131] This embodiment introduces health status influencing factors calibrated through statistical analysis. Furthermore, a specific dynamic penalty factor mathematical structure was constructed, realizing a highly adaptive cost adjustment mechanism; this mechanism establishes a long-term health indicator representing historical accumulation. The quantitative feedback path between the current instantaneous operating cost assessment and the decision-making system enables the system to automatically adjust its risk preference according to the different life stages of the unit, thereby more effectively delaying equipment aging, avoiding high maintenance costs and failure risks in the later stages, and realizing the long-term preservation of asset value.

[0132] Example 6:

[0133] S6 specifically includes:

[0134] The revised risk-adjusted return is compared with the pre-set harmonized limit;

[0135] S61. When the corrected risk-adjusted return is greater than or equal to twice the preset coordination limit, it is defined as a value range, and the original scheduling instruction is executed.

[0136] S62. When the corrected risk-adjusted return is greater than or equal to the preset coordination limit but less than twice the preset coordination limit, it is defined as a level one risk arbitrage range. The power adjustment rate in the original scheduling instruction is automatically adjusted to generate a risk mitigation scheme.

[0137] S63. When the corrected risk-adjusted return is greater than or equal to 0 and less than the preset coordination limit, it is defined as a secondary risk arbitrage range. A high physical loss risk warning is submitted to the dispatch center, and an alternative solution to reduce the response intensity is recommended.

[0138] Also includes:

[0139] When the corrected marginal physical loss cost exceeds the preset absolute upper limit of single operation risk, a rejection instruction is sent to the dispatching terminal, and an enhanced monitoring process for key components of the unit is automatically triggered.

[0140] Based on Example 1, this embodiment further specifies the hierarchical scheduling control logic of S6 and adds a hard constraint mechanism based on absolute risk, forming a comprehensive control strategy that combines flexibility and a safety bottom line.

[0141] The risk level determined by S6 is based on With harmonization limits The relationships can be specifically divided into:

[0142] Value range: This range indicates that the net benefit of the operation far exceeds the physical losses it incurs, demonstrating significant economic value.

[0143] Level 1 risk arbitrage range: This range indicates that although the operation has considerable net profit, the physical loss cost is also high, and there is a certain risk-reward trade-off.

[0144] Secondary risk arbitrage range: This range indicates that the net profit from the operation is low, almost equal to or slightly lower than the cost of physical losses, and the risk is high.

[0145] To ensure feasibility, harmonization limits are set. The settings are explained; for example, based on the preset risk strategy of the power plant. The basis for setting this value is that the management requires that any market arbitrage behavior that comes at the cost of shortening the lifespan of equipment must have a net profit margin of no less than 20% in order to ensure that the risk and return are basically equal.

[0146] Based on the above risk levels, the hierarchical scheduling control implemented is as follows:

[0147] S61: When the adjusted risk-adjusted return is in the value range When the time frame indicates that the risk-adjusted return of this operation is high and its economic value is significant, the original scheduling instruction should be executed; when it is in the rejection interval... When this occurs, it indicates that the operation results in a net loss after accounting for physical losses, and the system defaults to not responding or suggests not executing the original scheduling instruction;

[0148] S62: When the adjusted risk-adjusted return is within the Level 1 risk arbitrage range In such cases, the system automatically adjusts the power regulation rate in the original dispatching command to generate a risk mitigation plan; for example, while meeting the most basic response requirements of the power grid dispatching agency, the system adjusts the power regulation rate in the command. Proactively reduce costs by 10%-30% to significantly lower physical wear and tear costs while sacrificing a small amount of revenue. ;

[0149] S63: When the adjusted risk-adjusted return is within the secondary risk arbitrage range When this happens, a high physical loss risk warning should be submitted to the dispatch center, and alternative solutions to reduce the response intensity should be recommended; for example, it should be recommended to limit the power regulation rate to within 50% of the design safety threshold.

[0150] Furthermore, this embodiment also introduces the absolute risk hard constraint of Embodiment 6:

[0151] When the corrected marginal physical loss cost When the risk of a single operation exceeds the preset absolute limit, a rejection instruction is sent to the dispatching terminal, and an enhanced monitoring process for key components of the unit is automatically triggered. The enhanced monitoring process includes increasing the data acquisition frequency of sensors and prioritizing the maintenance of marked components.

[0152] To further clarify, the absolute upper limit of risk for a single operation is a preset cost value, representing the maximum physical loss allowed by a single operation that cannot be exceeded under any circumstances. The basis for setting this threshold is to prevent irreversible and severe damage to critical components. For example, this upper limit can be set according to material mechanics analysis as the equivalent damage of a single operation corresponding to the crack propagation entering an unstable stage. Strengthening the monitoring process may specifically include: increasing the data acquisition and uploading frequency of relevant unit critical components, such as vibration and temperature sensors, when the upper limit is triggered; conducting in-depth mining of historical data to identify abnormal patterns; and automatically marking the component in the observation list for priority inspection in subsequent maintenance plans.

[0153] By combining this refined hierarchical control with absolute risk constraints, the present invention achieves highly intelligent scheduling. It can not only respond proportionally and differently based on the risk-reward ratio, but also ensure that profits are not pursued at the expense of the fundamental safety of the generating units under any extreme market conditions by setting an absolute safety bottom line. This makes the power plant's operation strategy take into account both economic benefits and operational safety.

[0154] Example 8:

[0155] Please see Figure 2 The data acquisition module is used to collect real-time physical status parameters of the unit and economic signals from the market environment.

[0156] The cost calculation module is used to calculate the marginal physical loss cost characterizing a single variable operating condition operation based on the unit's physical state parameters.

[0157] The condition assessment module is used to accumulate historical marginal physical loss costs and, in combination with the preset total replacement cost of the unit, calculate the health depreciation index.

[0158] The closed-loop correction module is used to construct a dynamic penalty factor based on the health depreciation index, and to use the dynamic penalty factor to correct the marginal physical loss cost, thereby generating the corrected marginal physical loss cost.

[0159] The decision analysis module is used to determine direct market returns based on market environment economic signals, and calculate the corrected risk-adjusted rate of return by combining the corrected marginal physical loss cost.

[0160] The scheduling control module is used to compare the corrected risk-adjusted rate of return with the preset coordination limit, determine the risk level of the scheduling opportunity, and execute the hierarchical scheduling control corresponding to the risk level.

[0161] This invention also provides a pumped storage safety operation monitoring and management system based on digital twin technology. This system is designed to execute the methods of any of the foregoing embodiments. The system is based on a modular design concept, and its structure and functions are as follows:

[0162] The system includes:

[0163] A data acquisition module is used to collect unit physical status parameters and market environment economic signals in real time. This module is connected to the power plant's Supervisory Control and Data Acquisition (SCADA) system, Distributed Control System (DCS), and external power market information release platform via application programming interfaces (APIs), corresponding to S1 in the method.

[0164] A cost calculation module is used to calculate the marginal physical loss cost characterizing a single variable operating condition operation based on the unit's physical state parameters. This module contains the calculation logic and formulas of Example 3, corresponding to S2 in the method;

[0165] A condition assessment module is used to accumulate historical marginal physical loss costs. In conjunction with the preset total replacement cost of the unit The final calculation yielded the health depreciation index. This module contains a time-series database, corresponding to method S3;

[0166] A closed-loop correction module is used to construct a dynamic penalty factor based on the health depreciation index, and to use this factor to correct the marginal physical loss cost, generating the corrected marginal physical loss cost. This module incorporates the dynamic penalty factor model from Example 5, corresponding to S4 in the method.

[0167] A decision analysis module for determining direct market returns. and combined The corrected risk-adjusted return was calculated. This module incorporates the calculation logic of Example 4, corresponding to S5 in the method.

[0168] A scheduling control module is used to... The risk level of the scheduling opportunity is determined by comparing it with the preset coordination limit, and the hierarchical scheduling control corresponding to the risk level is executed; according to the rules defined in Example 6, the module sends the adjusted instructions to the automatic generation control (AGC) system of the power plant, or pops up the warning information on the human-machine interface, corresponding to S6 in the method;

[0169] Through this clear modular architecture, the present invention materializes complex methodologies into a well-structured and functionally cohesive system. Each module has clearly defined responsibilities and works collaboratively, ensuring the smooth and reliable flow of information and decision-making from data input to control output. This system design not only facilitates implementation and deployment but also provides great convenience for future functional upgrades and maintenance, thereby enabling the innovative methods proposed in this invention to be stably and efficiently applied to the safe operation and economic management of pumped storage power stations.

[0170] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0171] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A pumped storage safety operation monitoring and management method based on digital twin technology, characterized in that, The method comprises the following steps: S1, real-time collection of unit physical state parameters and market environment economic signals; S2, calculation of marginal physical loss cost representing single variable operating condition based on unit physical state parameters; S3, accumulation of marginal physical loss cost in history and calculation of health depreciation index combined with preset total unit replacement cost; S4, construction of dynamic penalty factor based on health depreciation index and correction of marginal physical loss cost by using dynamic penalty factor to generate corrected marginal physical loss cost; S5, determination of direct market income based on market environment economic signals and calculation of corrected risk-adjusted yield rate combined with corrected marginal physical loss cost; S6, comparison of corrected risk-adjusted yield rate with preset coordination limit value to determine risk level of dispatching opportunity and execution of graded dispatching control corresponding to risk level.

2. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, S1 specifically comprises: Unit physical state parameters include vibration intensity, temperature and water flow pressure; market environment economic signals include real-time grid access price, pumped storage power price and grid frequency.

3. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, S2 specifically comprises: S21, non-dimensional normalization processing of unit physical state parameters by using normalized reference value to generate normalized physical quantity; S22, calculation of comprehensive physical damage based on normalized physical quantity combined with preset weight coefficient and damage index; S23, determination of monetary conversion coefficient of physical loss, wherein the monetary conversion coefficient is calibrated by total unit replacement cost divided by design total fatigue damage life limit; S24, multiplication of comprehensive physical damage by monetary conversion coefficient to solve marginal physical loss cost.

4. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, S5 specifically comprises: S51, calculation of direct market income based on real-time grid access price, pumped storage power price, regulation power and duration; S52, subtraction of corrected marginal physical loss cost from direct market income and generation of corrected risk-adjusted yield rate by dividing difference by direct market income.

5. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, S4 specifically comprises: S41, determination of health state influence factor, wherein the health state influence factor is calibrated by statistical analysis of failure rate data of same type units at different aging stages; S42, multiplication of health depreciation index by health state influence factor and addition of 1 to the product to construct dynamic penalty factor; S43, multiplication of current calculated marginal physical loss cost by dynamic penalty factor to generate corrected marginal physical loss cost.

6. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, S6 specifically comprises: Comparison of corrected risk-adjusted yield rate with preset coordination limit value; S61, when corrected risk-adjusted yield rate is greater than or equal to twice preset coordination limit value, it is defined as value interval and original dispatching instruction is executed; S62, when corrected risk-adjusted yield rate is greater than or equal to preset coordination limit value and less than twice preset coordination limit value, it is defined as first-level risk arbitrage interval and power regulation rate in original dispatching instruction is automatically adjusted to generate risk mitigation scheme. S63, when the modified risk-adjusted yield is greater than or equal to 0 and less than the preset coordination limit, it is defined as a secondary risk arbitrage interval, a high physical loss risk warning is submitted to the dispatch center, and an alternative solution of reducing the response strength is recommended.

7. The pumped storage safety operation monitoring management method based on digital twin technology according to claim 1, characterized in that, Also includes: When the modified marginal physical loss cost exceeds the preset single operation risk absolute upper limit, a suggestion rejection instruction is sent to the dispatch end, and a strengthened monitoring process for key components of the unit is automatically triggered; The strengthened monitoring process includes increasing the data acquisition frequency of the sensor and marking the priority maintenance of the component.

8. A pumped storage safety operation monitoring and management system based on digital twin technology, applying the pumped storage safety operation monitoring and management method based on digital twin technology of any one of claims 1-7, characterized in that, Comprise the following modules: A data acquisition module for real-time acquisition of unit physical state parameters and market environment economic signals; A cost calculation module for calculating the marginal physical loss cost representing single variable operation based on the unit physical state parameters; A state evaluation module for accumulating the marginal physical loss cost in history and calculating the health depreciation index in combination with the preset total unit replacement cost; A closed-loop correction module for constructing a dynamic penalty factor based on the health depreciation index and correcting the marginal physical loss cost using the dynamic penalty factor to generate a modified marginal physical loss cost; A decision analysis module for determining the direct market yield based on the market environment economic signal and calculating the modified risk-adjusted yield in combination with the modified marginal physical loss cost; A dispatch control module for comparing the modified risk-adjusted yield with the preset coordination limit, determining the risk level of the dispatch opportunity, and performing graded dispatch control corresponding to the risk level.

Images