Fire power unit retirement path planning method and device based on meteorological hydrological risk, equipment, medium and product

Abstract

The application discloses a thermal power unit retirement path planning method and device based on meteorological hydrological risk, equipment, medium and product, and relates to the field of power system decision optimization. The method comprises the following steps: obtaining meteorological hydrological risk data and retirement index of a thermal power unit; performing screening and constraint progressive correction processing by using a progressive hierarchical screening logic method according to the retirement index and the meteorological hydrological risk data under climate change; performing mapping and nonlinear correlation interaction analysis by using Monte Carlo simulation and gradient boosting decision tree in combination with multiple sets of parameter combination data, such as water temperature and runoff in meteorological hydrological variables; performing collaborative optimization processing based on an adaptive strategy model and a set analysis function to obtain an optimal strategy, and then performing mapping and analysis processing in combination with the analysis result to obtain a targeted strategy list information, which is used for planning the retirement path of the thermal power unit. The application aims to realize the planning of the retirement path of the thermal power unit.

Description

Technical Field

[0001] The present application relates to the field of power system decision optimization, and in particular to a method, device, equipment, medium and product for planning the decommissioning path of thermal power units based on meteorological and hydrological risks. Background Art

[0002] Currently, thermal power units are facing retirement and transformation. However, the current retirement paths are mostly based on indicators such as carbon emission intensity and economic costs, ignoring the current and increasingly severe meteorological and hydrological risks under climate change. This may lead to high-risk units being retained first and low-risk units being eliminated first, threatening the security of the energy system and affecting the stable power generation of the power system.

[0003] The current decommissioning path planning for thermal power systems mainly focuses on the following directions:

[0004] Extensive decommissioning strategy: Traditional decommissioning decisions rely on unit age, carbon emission intensity, and air pollution, and lack scientific assessment of meteorological and hydrological risk exposure.

[0005] Local adaptation strategies: Existing adaptation measures (such as cooling tower renovation) often focus on optimizing a single technology and lack quantitative verification of systematic water resource-regulatory synergy strategies.

[0006] However, the following core issues remain:

[0007] Decommissioning decisions are disconnected from meteorological and hydrological risks: Age- or capacity-prioritized retirement mechanisms may mistakenly retain units with high climate vulnerability, exacerbating system vulnerability.

[0008] Insufficient uncertainty management: Existing methods do not jointly quantify the uncertainties of climate models, policy regulations, and technical parameters, resulting in low confidence in strategy recommendations or inability to select efficient targeted adaptation strategies for risk basins. Summary of the Invention

[0009] The purpose of this application is to provide a method, device, equipment, medium and product for planning the decommissioning path of thermal power units based on meteorological and hydrological risks, which can efficiently realize the planning of the decommissioning path of thermal power units.

[0010] To achieve the above objectives, this application provides the following solutions:

[0011] In a first aspect, the present application provides a method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks, comprising:

[0012] Obtaining meteorological and hydrological risk data and retirement indexes for thermal power units; the meteorological and hydrological risk data are determined based on characteristic parameters and operating parameters of the units, and the retirement index is determined based on the operating age and installed capacity of the units;

[0013] According to the retirement index and the meteorological and hydrological risk data, a progressive hierarchical screening logic method is used to perform screening and constraint progressive correction processing to obtain a processing result;

[0014] Monte Carlo simulation and gradient boosting decision tree are used to perform mapping and nonlinear correlation interaction analysis based on the processing results and multiple sets of parameter combination data to obtain analysis results; the parameter combination data include: meteorological and hydrological variables, environmental regulation parameters and design parameters;

[0015] Based on the analysis results, a collaborative optimization process is performed based on an adaptive strategy model and a set analysis function to obtain an optimal strategy; the adaptive strategy model includes: an inter-basin water transfer strategy model, a sewage reuse strategy model, and a flexible environmental regulation strategy model; the optimal strategy is determined based on the highest value calculated by the set analysis function;

[0016] According to the optimal strategy and the analysis results, strategy-factor mapping and parsing processing are performed to obtain targeted strategy list information; the targeted strategy list information is used to plan the retirement path of the thermal power unit.

[0017] In a second aspect, the present application provides a thermal power unit decommissioning path planning device based on meteorological and hydrological risks, comprising:

[0018] A data acquisition module is used to obtain meteorological and hydrological risk data and a retirement index of a thermal power unit; the meteorological and hydrological risk data is determined based on characteristic parameters and operating parameters of the unit;

[0019] A processing module, configured to perform screening and constraint progressive correction processing based on the retirement index and the meteorological and hydrological risk data using a progressive hierarchical screening logic method to obtain a processing result;

[0020] An analysis module, configured to use Monte Carlo simulation and gradient boosting decision tree to perform mapping and nonlinear correlation interaction analysis based on the processing results and multiple sets of parameter combination data to obtain analysis results; the parameter combination data includes: meteorological and hydrological variables, environmental regulation parameters, and design parameters;

[0021] an optimization processing module for performing collaborative optimization processing based on the analysis results, an adaptive strategy model, and a set analysis function to obtain an optimal strategy; the adaptive strategy model includes: an inter-basin water transfer strategy model, a sewage reuse strategy model, and a flexible environmental regulation strategy model; the optimal strategy is determined based on the highest value calculated by the set analysis function;

[0022] The mapping and parsing module is used to perform strategy-factor mapping and parsing processing based on the optimal strategy and the analysis results to obtain targeted strategy list information; the targeted strategy list information is used to plan the retirement path of the thermal power unit.

[0023] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks.

[0024] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks.

[0025] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned method for planning the decommissioning path of thermal power units based on meteorological and hydrological risks.

[0026] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0027] This application provides a method, apparatus, equipment, medium, and product for planning a thermal power plant retirement path based on meteorological and hydrological risk. The method obtains meteorological and hydrological risk data and a retirement index for the thermal power plant. Based on the retirement index and meteorological and hydrological risk data, a progressive hierarchical screening logic method is used to screen and constrain progressive corrections to obtain a result. This hierarchical screening and constrained progressive correction ensures a scientific retirement path and effectively mitigates meteorological and hydrological risks. Furthermore, based on this progressive hierarchical screening logic method, it ensures that retained units are young, high-capacity, and climate-resilient, while eliminating high-risk, inefficient units. Monte Carlo simulation and gradient boosting decision trees are used to map the processing results and multiple sets of parameter combination data, performing nonlinear correlation interaction analysis to obtain analysis results. Based on the analysis results, a collaborative optimization process is performed using an adaptive strategy model and a set analysis function to obtain the optimal strategy. This process quantifies the robustness of the strategy and avoids local optimal solutions. Finally, based on the optimal strategy and analysis results, a mapping and parsing process is performed to obtain a targeted strategy list, enabling a quantifiable, traceable, and optimizable full-chain analysis, thereby enabling planning for the thermal power plant retirement path. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0029] Figure 1 A flow chart of a thermal power unit decommissioning path planning method based on meteorological and hydrological risks;

[0030] Figure 2 A structural diagram of a thermal power unit decommissioning path planning device based on meteorological and hydrological risks;

[0031] Figure 3 A structural schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0033] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0034] In one exemplary embodiment, as shown in Figure 1 A thermal power unit decommissioning path planning method based on meteorological and hydrological risks is provided, comprising:

[0035] Step 100: Obtain meteorological and hydrological risk data and decommissioning index of the thermal power unit. The meteorological and hydrological risk data is determined according to the characteristic parameters and operation parameters of the unit, and the decommissioning index is determined according to the operation age and installed capacity of the unit.

[0036] Step 200: According to the decommissioning index and the meteorological and hydrological risk data, a progressive hierarchical screening logic method is used for screening and constraint progressive correction processing to obtain a processing result.

[0037] Step 300: According to the processing result and a plurality of parameter combination data, a mapping and nonlinear correlation interaction analysis is performed by using Monte Carlo simulation and gradient boosting decision tree to obtain an analysis result. The parameter combination data includes meteorological and hydrological variables, environmental regulation parameters and design parameters.

[0038] Step 400: Based on the analysis results, a collaborative optimization process is performed based on the adaptive strategy model and the set analysis function to obtain the optimal strategy. The adaptive strategy model includes: an inter-basin water transfer strategy model, a wastewater reuse strategy model, and a flexible environmental regulation strategy model. The optimal strategy is determined based on the highest value calculated by the set analysis function.

[0039] Step 500: Based on the optimal strategy and analysis results, strategy-factor mapping and parsing are performed to obtain targeted strategy list information. The targeted strategy list information is used to plan the retirement path of the thermal power unit.

[0040] Among them, according to the retirement index and meteorological and hydrological risk data, a progressive hierarchical screening logic method is used for screening and constrained progressive correction processing to obtain the processing results, including:

[0041] According to the retirement index, each component unit in the thermal power unit is divided according to the sorting order from high to low values ​​to obtain an initial pool; the initial pool includes: a priority elimination pool, a priority retention pool and an intermediate pool; the priority elimination pool is the units in the first range interval determined based on the sorting order; the priority retention pool is the units in the second range interval determined based on the sorting order; the intermediate pool is the corresponding units after excluding the first range interval and the second range interval based on the sorting order; the units in the first range interval are units with high values ​​selected according to a set ratio based on the sorting order of the retirement index; the units in the second range interval are units with low values ​​selected according to a set ratio based on the sorting order of the retirement index.

[0042] Based on the preset meteorological and hydrological risk constraint strategy, the units corresponding to the priority elimination pool and the priority retention pool are progressively corrected according to the constraint level to obtain the processing results; the meteorological and hydrological risk constraint strategy includes: capacity threshold constraint, extreme risk constraint, continuous exposure constraint and compound constraint strategy.

[0043] The inter-basin water transfer strategy model is a mathematical model that includes water transfer capacity and water transfer risk mitigation analysis.

[0044] The calculation formula for water transfer capacity is:

[0045] .

[0046] The calculation formula corresponding to the water diversion risk mitigation analysis is:

[0047] .

[0048] in, It is the available runoff volume of the unit after inter-basin water transfer; is the current available runoff of the unit; Theoretical water supply for water transfer project Actual water supply proportion for water transfer project Water transfer capacity Increment of available capacity for inter-basin water transfer Available capacity for inter-basin water transfer Benchmark available capacity Increment of risk operation days proportion for inter-basin water transfer Risk operation days proportion for inter-basin water transfer Benchmark risk operation days proportion Increment of unsafe operation days proportion for inter-basin water transfer Unsafe operation days proportion for inter-basin water transfer Benchmark unsafe operation days proportion

[0049] The elastic environmental regulation strategy model is a mathematical model for dynamic adjustment of water temperature regulation, adjustment of environmental flow, and risk mitigation analysis of elastic environmental regulation.

[0050] The calculation formula corresponding to the dynamic adjustment of water temperature regulation is:

[0051] .

[0052] The calculation formula corresponding to the adjustment of environmental flow is:

[0053] .

[0054] The calculation formula corresponding to the risk mitigation analysis of elastic environmental regulation is:

[0055] .

[0056] Wherein, is the adjusted environmental flow; is the current environmental flow proportion; is the runoff change amount in the historical period; is the runoff in the historical period; is the adjusted maximum allowable discharge water temperature regulation; is the current maximum allowable discharge water temperature regulation; is the water temperature rise amplitude in the historical period; is the increment of available capacity for elastic environmental regulation; is the available capacity for elastic environmental regulation; is the benchmark available capacity; is the increment of risk operation days proportion for elastic environmental regulation; is the risk operation days proportion for elastic environmental regulation; is the benchmark risk operation days proportion; The proportional increase in the number of unsafe operating days regulated for a resilient environment; The proportion of unsafe operating days regulated for a resilient environment; The ratio of days with unsafe operation as the benchmark.

[0057] The wastewater reuse strategy model is a mathematical model that includes water availability correction and risk mitigation analysis of wastewater reuse.

[0058] The calculation formula corresponding to the available water correction is:

[0059] .

[0060] The corresponding calculation formula for risk mitigation analysis of wastewater reuse is:

[0061] .

[0062] in, It is the available runoff volume of the unit after sewage reuse; is the current available runoff of the unit; is the amount of recyclable wastewater within the grid; Wastewater reuse efficiency; The increase in available capacity for wastewater reuse; Available capacity for wastewater reuse; is the baseline available capacity; The percentage increase of risk operation days for wastewater reuse; The proportion of risky operation days for wastewater reuse; The ratio of days with baseline risk operation; The percentage increase of unsafe operation days for wastewater reuse; The proportion of unsafe operation days for wastewater reuse; The ratio of days with unsafe operation as the benchmark.

[0063] Set the expression of the analysis function to:

[0064] .

[0065] in, To set the analysis function; is the first weight; is the increment of available capacity; is the second weight; The proportional increment of risk operation days; is the third weight; It is the proportional increment of unsafe operation days.

[0066] The research and development logic of this application is as follows:

[0067] Scientific decommissioning path: Build a coordinated optimization mechanism of age, capacity and meteorological and hydrological risks to avoid the problem of "inadvertent retention of high meteorological and hydrological risk units".

[0068] Uncertainty Optimization: Using a Monte Carlo-machine learning framework to analyze dominant variables and drive targeted matching of adaptive strategies.

[0069] Under the dual pressures of the climate crisis and energy transition, this application integrates meteorological and hydrological risks into traditional decommissioning pathways to effectively ensure climate resilience and future energy security. Data-driven machine learning models are used to identify key drivers and evaluate diverse adaptation strategies, supporting scientific decision-making and optimization. This application provides a technical solution for future thermal power system pathway planning, decommissioning screening, and adaptation strategy implementation.

[0070] Step 1: Optimize the retirement path of thermal power units considering meteorological and hydrological risks.

[0071] Execution body: Decommissioning path optimization server (Python code).

[0072] 1.1 Input parameters.

[0073] The meteorological and hydrological risk data (including the average annual available capacity factor, the proportion of high-risk operating days, and the proportion of unsafe operating days) and the retirement index for each thermal power unit are calculated based on the unit's characteristic parameters and operating parameters, namely, the operating parameter database, the power grid data platform, and the meteorological and hydrological conditions at the unit's location. Each thermal power unit includes the four independent parameters listed in Table 1.

[0074] Table 1 Meteorological and hydrological risk data

[0075]

[0076] 1.2 Calculation process (progressive hierarchical screening logic).

[0077] Phase 1: Baseline decommissioning strategy (initial economic screening).

[0078] 1. Sorting rules: All thermal power units are sorted from high to low according to their retirement index (high index = high retirement priority).

[0079] 2. Initial pool generation:

[0080] Priority retirement pool: select the top 20% of the retirement index units (capacity share ≈ 20%).

[0081] Priority retention pool: select the bottom 20% of units in the retirement index ranking (capacity share ≈ 20%).

[0082] Intermediate pool: The remaining 60% of units enter Phase 2 screening.

[0083] 3. Ratio adjustability: The capacity ratio can be customized (such as 10%, 30%) and dynamically adjusted through external parameters such as the ambient flow ratio γ (default γ = 20%).

[0084] Phase 2: Progressive revision of meteorological and hydrological risk constraints.

[0085] Screening objectives: Ensure that pool units are eliminated first and meet both high economic efficiency (high retirement index) and high risk (high climate vulnerability), and prioritize the retention of pool units that meet both low economic efficiency (low retirement index) and low risk (high climate resilience).

[0086] Based on the priority elimination pool and priority retention pool obtained above, four meteorological and hydrological risk constraint strategies (capacity threshold constraint, extreme risk constraint, continuous exposure constraint and composite constraint strategy) are set. The specific constraint strategies are shown in Table 2.

[0087] Table 2 Meteorological and hydrological risk constraint strategies

[0088]

[0089] Phase 3: Capacity balancing and dynamic replacement (fault tolerance mechanism).

[0090] 1. Priority order: Execute progressively according to the constraint hierarchy (capacity threshold → extreme risk → sustained exposure). Different meteorological and hydrological risk constraint strategies can be prioritized based on the risk constraint requirements. For example, if you want to ensure the operation of units in extreme weather conditions, you can choose the extreme risk constraint strategy; if you want to focus on daily continuous and stable operation, you can choose the sustained exposure constraint strategy; if you want to focus on long-term high output levels, you can choose the capacity threshold constraint strategy.

[0091] 2. Replacement logic: If a certain level constraint causes the retention / elimination pool capacity to be less than γ%, then:

[0092] Relaxing constraints: Allow units that meet any two constraints in the composite constraint strategy to enter the candidate pool; or relax the thresholds of each indicator in the constraint conditions. For example, in the capacity threshold constraint strategy, the CRII can be relaxed to no less than 0.6 as a correction condition for the priority retention pool.

[0093] Replacement rule: Sort by retirement index in the candidate pool and replenish to the target capacity.

[0094] 1.3 Output results (get processing results). As shown in Table 3.

[0095] Table 3 Processing results

[0096]

[0097] Through the risk-age-capacity collaborative decision-making mechanism, meteorological and hydrological risk data (indicators) (CRII, HRFR, UORF) are dynamically coupled to the traditional decommissioning standards, namely the decommissioning index (service life, installed capacity), breaking the decision-making model that solely relies on the age or capacity of the unit; based on the progressive screening logic, capacity attenuation, extreme risk and continuous exposure constraints are layered and superimposed, and units that are old, inefficient and have prominent climate vulnerability are eliminated first, while retaining high-quality units with high climate resilience, achieving the dual goals of minimizing stranded risks and maximizing power system resilience, and providing a scientific decision-making paradigm for the orderly withdrawal of thermal power.

[0098] Step 2: Uncertainty analysis of meteorological and hydrological risk data and identification of dominant variables.

[0099] Execution body: Risk simulation and machine learning computing server (Python code).

[0100] 2.1 Input parameters. The input parameter data are shown in Table 4.

[0101] Table 4 Input parameter data table

[0102]

[0103] 2.2 Calculation process.

[0104] For the thermal power units in the priority retention pool and intermediate pool screened in step 1, in order to ensure their operational safety, the dominant driving factors of their meteorological and hydrological risks are identified based on Monte Carlo simulation and LightGBM decision tree, and corresponding adaptation strategies are implemented accordingly.

[0105] Phase 1: Dynamically coupled sampling of multi-source parameters.

[0106] Dynamically couple meteorological and hydrological variables with regulatory / technical parameters to construct a multidimensional probability distribution and quantify the impact of uncertainty on the available capacity of the units.

[0107] Sampling of meteorological and hydrological variables: Based on historical observations and climate model data (daily data for more than 10 years), key quantiles (5th / 25th / 50th / 75th / 95th) are extracted to characterize different climate sensitivity scenarios (such as extreme drought and wet normal water year).

[0108] Regulatory and technical parameter sampling: Extract parameter extreme value intervals from public literature and industry standards, and use uniform sampling to simulate policy and technical uncertainties.

[0109] Dynamic coupling rules: Meteorological and hydrological variables are sampled simultaneously with regulatory parameters to generate 5,000 sets of parameter combinations, covering the joint uncertainty scenarios of "climate-policy-technology".

[0110] Phase 2: Modular computational architecture for risk communication.

[0111] Technical features: Parameter uncertainty is transferred through a modular calculation chain to generate a probability distribution of the unit's average annual available capacity (CRII), avoiding reliance on a single formula.

[0112] Input layer: input sampling parameters (for example, Tw in meteorological and hydrological variables, Tdismax in environmental regulation parameters, ncc in technical design parameters, etc.).

[0113] Processing layer: Capacity calculation module (calculates the available capacity under the current parameter combination based on thermodynamic equilibrium equations and regulatory constraints).

[0114] Output layer: Generates the probability distribution curve of the unit's CRII and the risk exposure probability (for example, the probability of CRII < 0.8 is 23%).

[0115] Phase 3: Analysis of dominant variables.

[0116] ①Data preparation and modeling process.

[0117] Input data: Monte Carlo simulation data set: contains multiple sets of parameter combinations (meteorological and hydrological variables, regulatory parameters, technical parameters) and their corresponding CRII.

[0118] Modeling objective: Through the LightGBM decision tree model, quantify the relative contribution of each parameter to the CRII risk value and identify the core driving factors of meteorological and hydrological risks.

[0119] ②LightGBM decision tree modeling.

[0120] Existing technologies usually use linear regression or univariate analysis, which cannot capture the nonlinear interaction effects between multiple parameters, leading to misjudgment of driving factors.

[0121] Dynamic parameter coupling training: The parameter combination generated by Monte Carlo and the CRII risk value are used as input to construct a "parameter→risk" mapping relationship; through the distributed gradient boosting framework of LightGBM, nonlinear correlations between parameters (such as the synergistic effect of high temperature and low runoff) are automatically learned.

[0122] Feature importance calculation: Based on the feature split gain, the contribution of each decision tree parameter to the reduction of CRII prediction error is counted; the sum of the split gains of all parameters is normalized to generate a relative importance score (0%~100%).

[0123] Technical advantages: Efficiency: LightGBM supports fast training of millions of data (5,000 Monte Carlo simulations take <5 minutes); Interpretability: The importance score intuitively reflects the dominance of parameters on risk.

[0124] ③Rules for determining the dominant driving factors.

[0125] Importance Sort: Sort the parameters by their relative importance scores from high to low.

[0126] Core driving factor screening: Single dominant factor: If the importance score of a certain parameter is significantly higher than that of other parameters (such as >50%), it is judged as a single dominant driving factor; Multi-factor coupling: If the cumulative importance score of the first 2-3 parameters is >70%, it is judged as a joint dominant factor.

[0127] The results of the driving factor analysis of a certain inland thermal power unit in the priority retention pool are shown in Table 5.

[0128] Table 5 Analysis results display table

[0129]

[0130] ④Application of driving factor analysis results.

[0131] Short-term mitigation strategies: If meteorological and hydrological variables (such as Tw and Q) dominate, implement physical control measures (such as cooling tower spray cooling and inter-basin water transfer); if the regulatory parameter (Tdismax) dominates, recommend policy adaptation (such as adjusting environmentally regulated water temperature). For scenarios where multiple coupled factors are dominant (such as Tw and Q jointly dominate), design composite resilience enhancement solutions (such as seawater desalination combined with air cooling retrofits).

[0132] 2.3 Output results.

[0133] Priority is given to the ranking of the importance of driving factors and the core driving factors of individual units in the retention pool and the intermediate pool, providing a reference for the subsequent application of adaptive strategies.

[0134] Quantification of multi-source uncertainty: combining climate models and site observation data, regulatory thresholds, and random sampling of technical parameters to comprehensively cover natural-social system uncertainties.

[0135] Machine learning-driven attribution: Analyze nonlinear interaction effects through the LightGBM decision tree, breaking through the limitations of traditional regression models.

[0136] Through the Monte Carlo-machine learning joint framework, the multi-dimensional uncertainty of meteorological and hydrological risk data (natural variability, policy fluctuations, technological differences) is quantified, the dominant risk drivers of different regions / units (such as water temperature, runoff, and emission regulations) are identified, and the "common constraints of high-risk units" are revealed, providing targeted decision-making basis for differentiated risk prevention and control (such as prioritizing the transformation of cooling systems in high-water temperature areas).

[0137] Step 3: Targeted assessment and optimization of meteorological and hydrological risk adaptation strategies.

[0138] Performing body: adaptive policy simulation calculation server (python code).

[0139] 3.1 Input parameters. Table 6 is the input parameter table.

[0140] Table 6 Input parameter table

[0141]

[0142] 3.2 Adaptive policy modeling.

[0143] 1. Water diversion strategy (WD).

[0144] Water diversion capacity calculation:

[0145] .

[0146] Risk mitigation assessment (water diversion risk mitigation analysis):

[0147] .

[0148] Water diversion increases available runoff, i.e. , directly increases available capacity, and reduces extreme risk and sustained instability due to water shortage.

[0149] 2. Wastewater reuse strategy (WR).

[0150] Available water quantity correction:

[0151] .

[0152] Risk mitigation assessment (risk mitigation analysis of wastewater reuse):

[0153] .

[0154] Wastewater reuse supplements local water quantity, improves unit available capacity, reduces extreme water shortage risk and operation interruption frequency.

[0155] 3. Adaptive regulation strategy (AR).

[0156] Dynamic adjustment of water temperature regulation:

[0157] .

[0158] Environmental flow adjustment:

[0159] .

[0160] Constraints: The adjusted environmental flow must meet the minimum ecological water demand threshold ( ≥γ ecology).

[0161] Risk Mitigation Assessment (Risk Mitigation Analysis for Flexible Environmental Regulation):

[0162] .

[0163] Relax water temperature regulations or optimize environmental flows to alleviate capacity loss caused by high temperature or low flow, and reduce the frequency and duration of exposure to extreme risks.

[0164] 3.3 Strategy optimization and scenario analysis.

[0165] Single strategy effect verification:

[0166] Based on Monte Carlo simulations and LightGBM's dominant driver analysis (Step 2), the improvement in CRII, HRFR, and UORF for each strategy is quantified for units in the priority retention pool and intermediate pool (Step 1). For example, if the dominant factor in a region is water temperature, the ΔUORF of the AR strategy is prioritized.

[0167] Comprehensive scoring function setting (setting analysis function):

[0168] .

[0169] 、 、 It can be set according to regional risk priority (default equal weight). The comprehensive score Score is calculated for the risk mitigation effect of each strategy of each unit, that is, the calculation The value of is used to compare the scores of different strategies, thereby selecting the strategy with the highest comprehensive score as the optimal strategy under the single strategy of the unit.

[0170] Multi-strategy collaborative optimization:

[0171] Randomly combining the three strategies WD, WR, and AR yields the following scenarios: WD+WR, WD+AR, WR+AR, and WD+WR+AR. The composite score is calculated by summing the corresponding mitigation levels under the combined strategies. The composite scores for the combined strategies are compared with those for the individual strategies, and the synergies or antagonisms between the strategies are analyzed to identify the optimal strategy combination, with the goal of maximizing risk mitigation. For example, if the highest-scoring combination for a particular unit is WD+WR, this combination is the optimal strategy. AR and WD+WR exhibit antagonism, while WD and WR exhibit synergy. For a unit with the highest score for WD, the single strategy WD is the optimal strategy, and AR, WD, and WR exhibit antagonism.

[0172] Generation of a recommended list of targeted strategies:

[0173] Following these steps, the optimal strategy for the unit can be identified. The correspondence between the core driving factors output in step 2 and the optimal strategy can be verified to generate a strategy library (strategy-factor mapping table). Based on this strategy library, the correspondence between different core driving factors and optimal mitigation strategies can be analyzed to form a universal list of recommended targeted strategies. An example of a recommended targeted strategy list is shown in Table 7.

[0174] Table 7 Recommended list of targeted strategies

[0175]

[0176] Geospatial Adaptation: This approach combines climate zoning (e.g., the Köppen-Geiger classification) to generate regional strategy priority maps, avoiding one-size-fits-all decisions. It also supports the dynamic addition of new strategies (e.g., desalination DS) and enables continuous iteration of the strategy library through interface design. The output is shown in Table 8.

[0177] Table 8 Output result table

[0178]

[0179] Through driving factor-strategy coupling analysis, the dominant variables identified in step 2 (such as water temperature and runoff) are accurately matched with adaptation strategies (such as water diversion and regulatory adjustments), and a strategy combination that can maximize the mitigation of meteorological and hydrological risks is screened. This achieves refined risk management of "focused intervention on highly sensitive units and dynamic regulation of low-risk units," providing data-driven decision-making support for power system climate adaptation planning.

[0180] In addition, the technical solutions mentioned in this application can be replaced by policy-driven solutions. The specific steps are as follows:

[0181] Step 1: Extensive ratio setting: Directly use administrative orders to force the elimination of a fixed ratio of units (such as a "one-size-fits-all" shutdown of 20% of coal-fired power), without establishing a meteorological and hydrological risk assessment model, and relying only on single indicators such as unit age or capacity.

[0182] Step 2: Static strategy deployment: uniformly implement rigid strategies (such as equal allocation of water transfer quotas), regardless of regional climate differences (such as differences in demand between arid and humid areas).

[0183] Step 3: Optimization without Feedback: This strategy lacks dynamic monitoring and effectiveness evaluation after implementation, making it impossible to adjust the path based on actual risk changes. However, this strategy has its flaws, as analyzed in Table 9.

[0184] Table 9 Defect Analysis Table

[0185]

[0186] Alternatively, a traditional model-driven approach can be adopted, with the following steps.

[0187] Step 1: Static risk assessment: Use the historical mean method to assess climate risk (such as taking the average runoff value over the past 10 years), without considering climate change trends and extreme events.

[0188] Step 2: Linear optimization model: linear programming based on fixed thresholds, ignoring nonlinear interaction effects between parameters.

[0189] Step 3: Experience-Based Strategy Library: This relies on experience to develop general strategies (e.g., installing air cooling systems on all units), lacking data-driven adaptation. Table 10 shows an analysis of its shortcomings.

[0190] Table 10 Analysis table for defects

[0191]

[0192] This application forms a progressive screening logic based on dynamic data fusion and multi-level risk constraints, breaking through the difficulty of "economy-climate" dual-objective optimization; the Monte Carlo-LightGBM joint framework realizes the analysis of multi-factor nonlinear contribution; the strategy mitigation screening mechanism and optimal strategy analysis provide a feasible solution for climate risk mitigation.

[0193] The solution proposed in this application belongs to the optimization of the exit decision of high climate-sensitive units in the low-carbon transformation of the power system.

[0194] (1) By constructing a retirement priority index based on unit age and installed capacity; developing a two-way screening mechanism to screen units for priority retention and priority elimination respectively; and on this basis, introducing multi-dimensional meteorological and hydrological risk constraint decision-making to screen the retirement and retention cluster priorities of units that comprehensively consider carbon emissions, economic costs and meteorological and hydrological risks.

[0195] (2) Through Monte Carlo simulation and LightGBM (Light Gradient Boosting Machine) distributed gradient boosting framework based on decision tree algorithm, the core driving factors of meteorological and hydrological risks at the unit scale (meteorological and hydrological factors, environmental regulatory factors, technical characteristics factors, etc.) for priority elimination and priority retention are identified; based on the main risk-constraining variables, water temperature control and alternative water supply strategies are adopted to mitigate meteorological and hydrological risks.

[0196] This application solves the pain points of the fragmentation of meteorological and hydrological risk factors and insufficient dynamic adaptability in traditional decommissioning decisions, and provides technical support for building a safe and orderly thermal power unit exit mechanism.

[0197] Compared with existing technologies, this application achieves core advantages such as enhanced dynamic adaptability and optimized risk response efficiency in the optimization of thermal power unit decommissioning decisions. The specific technical advantages and sources are as follows:

[0198] 1. The dynamic adaptability of risk decision-making is enhanced.

[0199] Real-time response capability: Combining long-term meteorological and hydrological data with Monte Carlo simulation enables dynamic updates of risk forecasts, significantly improving response speed compared to traditional annual-scale assessments.

[0200] Strategy Target Matching: Adaptive strategy optimization dynamically adjusts water transfer, wastewater reuse and regulatory parameters based on dominant variables, significantly increasing strategy effectiveness.

[0201] Leading variable identification: LightGBM-driven sensitivity analysis locates key driving factors (such as water temperature and environmental flow) to support precise strategy adaptation.

[0202] Targeted strategy evaluation combines different adaptive strategies to match core driving factors and screen the optimal adaptive strategy.

[0203] 2. Optimize risk prevention and control efficiency.

[0204] Scientific decommissioning path: The decommissioning strategy coupled with meteorological and hydrological risks avoids the power supply gap problem caused by traditional "one-size-fits-all" decommissioning through multi-objective optimization of age, capacity and risk, and effectively alleviates meteorological and hydrological risks.

[0205] Resource utilization efficiency: Adaptation strategy evaluation and screening of optimal combinations (such as water diversion + flexible regulation) significantly improve the risk mitigation efficiency of water resources and flexible environmental control.

[0206] Progressive retirement mechanism: Tiered constraint screening (capacity threshold → extreme risk → continuous exposure) ensures that the retained units are young, high-capacity and climate-resilient, while eliminating high-risk and inefficient units.

[0207] Monte Carlo-machine learning joint framework: Through 5,000 scenario simulations per unit and strategy effect distribution analysis, it quantifies strategy robustness and avoids local optimal solutions.

[0208] 3. Breakthrough in uncertainty management capabilities.

[0209] Reliability of risk quantification: Monte Carlo simulation combines the uncertainties of climate models, regulatory parameters and technical variables, and outputs the CRII probability distribution (such as the 5th-95th quantile interval), which significantly reduces the risk misjudgment rate compared with traditional deterministic models.

[0210] Attribution explainability: The LightGBM decision tree analyzes variable contributions (for example, water temperature has a 45% impact on CRII), supporting transparent decision-making.

[0211] Multivariate joint sampling: quantile-extreme value mixed sampling of meteorological and hydrological variables (normal distribution) and regulatory parameters (uniform distribution) to cover the uncertainty of the natural-social system.

[0212] Machine learning attribution: Variable importance ranking based on feature splitting gains, revealing nonlinear interaction effects.

[0213] Therefore, this application systematically solves the bottleneck problems of "static experience bias, single decision-making basis, and weak uncertainty management" in traditional thermal power meteorological and hydrological risk assessment through three major technological breakthroughs: multi-level risk constraints, machine learning attribution, and multi-dimensional comparison of adaptation strategies. It provides quantifiable, traceable, and optimizable full-chain technical support for the low-carbon transformation and climate resilience improvement of the power industry.

[0214] The present application also provides an application scenario, which applies the above-mentioned thermal power unit retirement path planning method. Specifically: the thermal power unit retirement path planning method provided in this embodiment can be applied to the low-carbon transformation planning scenario of the power system. The scenario includes the following core links: 1) power supply and demand load forecasting and renewable energy installed capacity calculation; 2) service life and economic evaluation of existing thermal power units; 3) retirement timing optimization under grid reliability constraints; 4) dynamic simulation verification of multi-energy complementary systems. The method provided in this embodiment belongs to the third link (retirement timing optimization). By coupling the meteorological and hydrological risks and retirement index of the unit, combining multi-level risk constraints, machine learning attribution and multi-dimensional comparison of adaptation strategies, a progressive retirement path plan that meets the triple goals of safety, economy and low carbon is generated, providing quantitative decision support for the transformation path of the power system.

[0215] Specifically, in the processing of this link, we can perform screening and constrained progressive correction based on the progressive hierarchical screening logic method, and use Monte Carlo simulation and gradient boosting decision trees to perform mapping and nonlinear correlation interaction analysis based on the processing results and multiple sets of parameter combination data, thereby realizing the planning of the retirement path of thermal power units.

[0216] In an exemplary embodiment, Figure 2 As shown, a thermal power unit decommissioning path planning device based on meteorological and hydrological risks is provided, comprising:

[0217] The data acquisition module is used to obtain the meteorological and hydrological risk data and retirement index of the thermal power unit; the meteorological and hydrological risk data is determined based on the characteristic parameters and operating parameters of the unit.

[0218] The processing module is used to perform screening and constraint progressive correction processing based on the retirement index and meteorological and hydrological risk data using a progressive hierarchical screening logic method to obtain processing results.

[0219] The analysis module is configured to perform mapping and nonlinear correlation interaction analysis on the processing result and a plurality of parameter combination data according to a Monte Carlo simulation and a gradient boosting decision tree, to obtain an analysis result; the parameter combination data includes meteorological and hydrological variables, environmental regulation parameters, and design parameters.

[0220] The optimization processing module is configured to perform collaborative optimization processing based on an adaptive strategy model and a set analysis function according to the analysis result, to obtain an optimal strategy; the adaptive strategy model includes a cross-basin water transfer strategy model, a sewage reuse strategy model, and an elastic environmental regulation strategy model; the optimal strategy is determined based on a highest value calculated by the set analysis function.

[0221] The mapping and analysis module is configured to perform strategy-factor mapping and analysis processing according to the optimal strategy and the analysis result, to obtain target strategy list information; the target strategy list information is used to plan a decommissioning path of the thermal power unit.

[0222] In an exemplary embodiment, a computer device can be provided, which can be a server or a terminal, and an internal structure diagram thereof can be as shown in Figure 3 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is configured to store thermal power unit decommissioning path planning data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a thermal power unit decommissioning path planning method based on meteorological and hydrological risks.

[0223] Those skilled in the art can understand that Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0224] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0225] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0226] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0227] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0228] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0229] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0230] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0231] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for planning the decommissioning path of thermal power units based on meteorological and hydrological risks, characterized in that: include: Obtain meteorological and hydrological risk data and retirement index for thermal power units; The meteorological and hydrological risk data are determined based on the characteristic parameters and operating parameters of the unit; The retirement index is determined based on the operating age and installed capacity of the unit; According to the retirement index and the meteorological and hydrological risk data, a progressive hierarchical screening logic method is used to perform screening and constraint progressive correction processing to obtain a processing result; Monte Carlo simulation and gradient boosting decision tree are used to perform mapping and nonlinear correlation interaction analysis based on the processing results and multiple sets of parameter combination data to obtain analysis results; The parameter combination data includes: meteorological and hydrological variables, environmental regulation parameters and design parameters; Based on the analysis results, a collaborative optimization process is performed based on an adaptive strategy model and a set analysis function to obtain an optimal strategy; the adaptive strategy model includes: an inter-basin water transfer strategy model, a sewage reuse strategy model, and a flexible environmental regulation strategy model; the optimal strategy is determined based on the highest value calculated by the set analysis function; According to the optimal strategy and the analysis results, strategy-factor mapping and parsing processing are performed to obtain targeted strategy list information; the targeted strategy list information is used to plan the retirement path of the thermal power unit.

2. The method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks according to claim 1, characterized in that: According to the retirement index and the meteorological and hydrological risk data, a progressive hierarchical screening logic method is used to perform screening and constraint progressive correction processing to obtain processing results, specifically including: According to the retirement index, each component unit in the thermal power unit is divided according to the sorting order from high to low values ​​to obtain an initial pool; the initial pool includes: a priority elimination pool, a priority retention pool and an intermediate pool; the priority elimination pool is the units in a first range interval determined based on the sorting order; the priority retention pool is the units in a second range interval determined based on the sorting order; the intermediate pool is the units corresponding to the sorting order after excluding the first range interval and the second range interval; the units in the first range interval are units with high values ​​selected according to a set ratio based on the sorting order of the retirement index; the units in the second range interval are units with low values ​​selected according to a set ratio based on the sorting order of the retirement index; Based on the preset meteorological and hydrological risk constraint strategy, the units corresponding to the priority elimination pool and the priority retention pool are progressively corrected according to the constraint hierarchy to obtain the processing results; the meteorological and hydrological risk constraint strategy includes: capacity threshold constraint, extreme risk constraint, continuous exposure constraint and composite constraint strategy.

3. The method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks according to claim 1, characterized in that: The inter-basin water transfer strategy model is a mathematical model that includes water transfer capacity and water transfer risk mitigation analysis; The calculation formula corresponding to the water transfer capacity is: ； The calculation formula corresponding to the water diversion risk mitigation analysis is: ； in, It is the available runoff volume of the unit after inter-basin water transfer; is the current available runoff of the unit; The theoretical water supply of the water diversion project; The actual water supply ratio of the water diversion project; To adjust the water capacity; Increase in available capacity for inter-basin water transfers; the available capacity for inter-basin water transfers; is the baseline available capacity; The percentage increase of risk operation days for inter-basin water transfer; The proportion of days with risk of inter-basin water transfer operation; The ratio of days with baseline risk operation; The percentage increase in the number of unsafe operation days for inter-basin water transfer; The proportion of unsafe operation days for inter-basin water transfer; The ratio of days with unsafe operation as the benchmark.

4. The method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks according to claim 1, characterized in that: The flexible environmental regulation strategy model is a mathematical model that includes dynamic adjustment of water temperature regulation, environmental flow adjustment, and risk mitigation analysis of flexible environmental regulation; Among them, the calculation formula corresponding to the dynamic adjustment of water temperature regulation is: ； The calculation formula corresponding to the environmental flow adjustment is: ； The corresponding calculation formula for risk mitigation analysis of flexible environmental regulation is: ； in, is the adjusted environmental flow; is the current environmental flow ratio; is the change in runoff during the historical period; is the runoff volume in the historical period; The maximum allowable discharge water temperature regulation after adjustment; The current maximum allowable discharge water temperature regulation; is the increase in water temperature during the historical period; Increases in available capacity regulated for resilient environments; Available capacity regulated for a resilient environment; is the baseline available capacity; The proportional increase in the number of risky operating days regulated for a resilient environment; The proportion of risky operating days regulated for a resilient environment; The ratio of days with baseline risk operation; The proportional increase in the number of unsafe operating days regulated for a resilient environment; The proportion of unsafe operating days regulated for a resilient environment; The ratio of days with unsafe operation as the benchmark.

5. The method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks according to claim 1, characterized in that: The wastewater reuse strategy model is a mathematical model that includes water availability correction and risk mitigation analysis of wastewater reuse; The calculation formula corresponding to the available water correction is: ； The corresponding calculation formula for risk mitigation analysis of wastewater reuse is: ； in, It is the available runoff volume of the unit after sewage reuse; is the current available runoff of the unit; is the amount of recyclable wastewater within the grid; Wastewater reuse efficiency; The increase in available capacity for wastewater reuse; Available capacity for wastewater reuse; is the baseline available capacity; The percentage increase of risk operation days for wastewater reuse; The proportion of risky operation days for wastewater reuse; The ratio of days with baseline risk operation; The percentage increase of unsafe operation days for wastewater reuse; The proportion of unsafe operation days for wastewater reuse; The ratio of days with unsafe operation as the benchmark.

6. The method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks according to claim 1, characterized in that: The expression of the setting analysis function is: ； in, To set the analysis function; is the first weight; is the increment of available capacity; is the second weight; The proportional increment of risk operation days; is the third weight; It is the proportional increment of unsafe operation days.

7. A thermal power unit decommissioning path planning device based on meteorological and hydrological risks, characterized in that: include: Data acquisition module, used to obtain meteorological and hydrological risk data and retirement index of thermal power units; The meteorological and hydrological risk data are determined based on the characteristic parameters and operating parameters of the unit; The retirement index is determined based on the operating age and installed capacity of the unit; A processing module, configured to perform screening and constraint progressive correction processing based on the retirement index and the meteorological and hydrological risk data using a progressive hierarchical screening logic method to obtain a processing result; An analysis module is used to use Monte Carlo simulation and gradient boosting decision tree to perform mapping and nonlinear correlation interaction analysis based on the processing results and multiple sets of parameter combination data to obtain analysis results; The parameter combination data includes: meteorological and hydrological variables, environmental regulation parameters and design parameters; an optimization processing module for performing collaborative optimization processing based on the analysis results, an adaptive strategy model, and a set analysis function to obtain an optimal strategy; the adaptive strategy model includes: an inter-basin water transfer strategy model, a sewage reuse strategy model, and a flexible environmental regulation strategy model; the optimal strategy is determined based on the highest value calculated by the set analysis function; The mapping and parsing module is used to perform strategy-factor mapping and parsing processing based on the optimal strategy and the analysis results to obtain targeted strategy list information; the targeted strategy list information is used to plan the retirement path of the thermal power unit.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for planning the decommissioning path of a thermal power unit based on meteorological and hydrological risks described in any one of claims 1 to 6 is implemented.

Images