Power transmission network expansion planning method and system considering renewable energy extreme scenarios

Abstract

The application discloses a power transmission network expansion planning method and system considering renewable energy extreme scenarios, generates massive renewable energy random scenarios based on GAN, provides reliable data information for extreme scenario analysis, gives the definition and determination condition of the renewable energy extreme scenario from the actual influence degree and the occurrence probability of the power grid operation, proposes a data-driven power transmission network expansion planning framework considering the extreme scenario analysis, and realizes the balance of planning economy and robustness.

Description

Technical Field

[0001] This invention belongs to the field of power transmission network expansion planning, specifically relating to a method and system for power transmission network expansion planning that considers extreme scenarios of renewable energy. Background Technology

[0002] Against the backdrop of global warming, low-probability extreme weather events such as extreme cold and heat are becoming more frequent and intense, causing abnormal fluctuations in the output of renewable energy sources such as wind and solar power, and severely impacting the power system's supply and demand balance. Considering the future pivotal role of the power grid in promoting the inter-regional and inter-provincial consumption of renewable energy, it is of great significance to consider extreme renewable energy output in the TEP (Transmission-Oriented Power) problem.

[0003] Extreme renewable energy output exhibits the characteristics of low-probability, high-impact (LPHI) scenarios. If grid planning fully accommodates low-probability extreme situations, it will significantly increase planning costs and lead to wasted investment. Conversely, if only high-probability, low-impact conventional output scenarios are considered, power shortages will occur in extreme situations where renewable energy output is severely insufficient. Therefore, grid planning that considers extreme scenarios not only needs to balance the economics and robustness of decision-making but also requires accurate assessment of the negative effects of extreme scenarios to provide a basis for developing extreme situation mitigation measures.

[0004] Existing technology 1: A robust optimization-based power grid planning method, which provides a planning scheme that meets the worst-case scenario. The specific form is as follows:

[0005] The range of fluctuations in renewable energy output is described by a closed and bounded uncertainty set, which typically includes box uncertainty set, polyhedral uncertainty set, ellipsoidal uncertainty set or cardinal uncertainty set. The worst-case scenario within the set is sought for power grid planning decisions. The planning scheme is feasible for all scenarios in the uncertainty set. It usually has a min-max-min model structure, which is difficult to solve.

[0006] Disadvantage 1: The setting of parameters for the uncertain set is influenced by human experience and is subjective;

[0007] Disadvantage 2: It does not require understanding the specific probability distribution of renewable energy output, and usually ignores or simplifies the spatiotemporal correlation between multiple random variables, making it difficult for the worst-case scenario to occur in reality, and the planning is too conservative.

[0008] Existing technology two: A grid planning method based on stochastic optimization, establishing a probabilistic model to describe renewable energy output. The specific form is as follows:

[0009] This includes the expected value model and the opportunity constraint model. The expected value model uses discrete probability scenarios to describe renewable energy output, which can easily handle the correlation characteristics between uncertain factors, and the planning decision meets the safety constraints under all discrete scenarios. The opportunity constraint model defines risk indicators to measure the impact of extreme cases of uncertain factors. The planning scheme allows for the occurrence of situations where safety constraints are not met at a certain confidence level, sacrificing some robustness to ensure economic efficiency.

[0010] Disadvantage 1: The quality of expected value model decision depends on the selection of typical scenario sets, which usually excludes low-probability extreme scenarios, resulting in poor planning robustness.

[0011] Disadvantage 2: The chance constraint is a non-convex probabilistic constraint, which is difficult to solve when dealing with multiple couplings of random variables and massive random scenarios. Summary of the Invention

[0012] To address the above issues, this patent presents a method and system for transmission network expansion planning that considers extreme scenarios of renewable energy. Based on GAN generation, massive amounts of random renewable energy scenarios are generated, providing reliable data for extreme scenario analysis. The definition and judgment criteria of extreme renewable energy scenarios are given from two aspects: the actual impact on grid operation and the probability of occurrence. A data-driven transmission network expansion planning framework considering extreme scenario analysis is proposed, achieving a balance between planning economy and robustness.

[0013] To solve the technical problem, the technical solution of the present invention is as follows:

[0014] A transmission grid expansion planning method considering extreme scenarios for renewable energy, the method comprising:

[0015] A scene generation model using a generative adversarial network (GAN) is used to generalize the historical dataset, generating tens of thousands of spatiotemporally correlated random renewable energy scenes. Scene simplification is then performed to obtain a typical scene set.

[0016] A two-stage transmission network expansion planning TEP expectation value model is established and solved based on a typical scenario set to obtain the proposed transmission network planning scheme for the kth iteration.

[0017] To obtain the actual impact and probability of occurrence of extreme situations, and to determine extreme scenarios and criteria for renewable energy;

[0018] Using the aforementioned proposed planning scheme for the power transmission network, a risk assessment model REM is established. Based on the REM model, random scenarios of renewable energy are screened and processed, and extreme scenario analysis is conducted based on the determined extreme scenarios and judgment conditions of renewable energy.

[0019] The typical scenario set is adaptively updated based on the results of extreme scenario analysis, and the next round of loop is started until the low probability condition is met, so as to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions.

[0020] Furthermore, a two-stage TEP expected value model is established based on a set of typical scenarios, including:

[0021] With the minimum annual cost as the objective function, and considering the power system security constraints under typical scenarios, a two-stage TEP (Transmission Line Expectation Value) model is established. The objective function for minimizing annual cost includes: annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost. The power system security constraints include: existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and maximum renewable energy curtailment ratio.

[0022] Furthermore, the determination of extreme renewable energy scenarios specifically includes:

[0023] The actual impact of extreme situations is measured using the load loss risk index; the system load loss at node n in time period t is set as follows under the renewable energy time-series scenario ξ. The hourly load loss factor and the daily load loss factor are used to describe the maximum load loss risk per unit time period and the cumulative daily load loss risk of the power system under this scenario, respectively. The calculation formula is as follows:

[0024]

[0025]

[0026] Where D t Let ξ be the total load demand of the power system during time period t; the subscript ξ|O represents the scenario ξ acting on planning scheme O; acceptable unit time period and daily cumulative load shedding thresholds are set based on manual experience or specific circumstances, when HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D When the scenario is in the high-impact scenario for planning scheme O, it is considered a low-impact scenario; otherwise, it is considered a low-impact scenario.

[0027] The probability of extreme cases is described using a significance level α. When the proportion of LPHI scene samples in the generated scene set is less than the significance level α, the occurrence of an extreme scene is a low-probability event. In this case, the probability of the LPHI scene occurring is estimated using the number of HI samples, satisfying the following:

[0028]

[0029] Where S LPHIFor LPHI scene set, S RES To generate the scene set, N(·) is the number of scene samples;

[0030] The determination of the extreme scenarios for renewable energy includes: determining the first determination condition for extreme scenarios for renewable energy and the second determination condition for extreme scenarios for renewable energy;

[0031] The first criterion for determining extreme scenarios for renewable energy: HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D ;

[0032] Second criterion for extreme scenarios of renewable energy: P(ξ∈S) LPHI )≤α.

[0033] Furthermore, after screening and processing random scenarios for renewable energy, extreme scenario analysis is conducted based on the identified extreme renewable energy scenarios and judgment conditions; specifically including:

[0034] Based on the REM model, the stochastic scenario of renewable energy is solved to obtain the load shedding information of the system at each time period under the renewable energy time series scenario ξ. Then, according to the above-defined extreme scenario of renewable energy, the load shedding risk index HLOLF is calculated. ξ|O and DLOLF ξ|O The LPHI scenario is determined based on the first criterion for extreme renewable energy scenarios, and P(ξ∈S) is calculated. LPHI ).

[0035] Furthermore, the typical scenario set is adaptively updated based on the extreme scenario analysis results, and the next round of iteration is initiated until the low-probability condition is met; specifically including:

[0036] When the second criterion for extreme renewable energy scenarios is not met, the scenario with the least load loss is selected from the LPHI scenarios and added to the typical scenario set to update the typical scenario set, and the next cycle is started; until the second criterion for extreme renewable energy scenarios is met, the LPHI scenario set that meets the criterion and the final planning scheme are obtained.

[0037] A power grid expansion planning system considering extreme scenarios for renewable energy, the system comprising:

[0038] The processing module is used to generalize the historical dataset using a scene generation model based on a generative adversarial network (GAN), generate tens of thousands of spatiotemporally correlated random renewable energy scenes, and perform scene simplification to obtain a typical scene set.

[0039] The solution module is used to establish and solve the expected value model of the two-stage transmission network expansion plan (TEP) based on a set of typical scenarios, and obtain the proposed planning scheme of the transmission network in the kth iteration.

[0040] The determination module is used to obtain the actual impact and probability of occurrence of extreme situations, and to determine the extreme scenarios and judgment conditions for renewable energy.

[0041] The analysis and processing module is used to establish a risk assessment model REM based on the above-mentioned proposed planning scheme of the transmission network, and to conduct extreme scenario analysis based on the determined extreme scenarios and judgment conditions of renewable energy after screening and processing the random scenarios of renewable energy based on REM.

[0042] The update loop module is used to adaptively update the typical scenario set based on the extreme scenario analysis results and start the next loop until the low probability condition is met, that is, to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions.

[0043] Furthermore, the solution module includes a solution unit; the solution unit is used to establish a two-stage TEP expected value model with the minimum annual cost as the objective function, considering the power system security constraints under typical scenario sets; the minimum annual cost as the objective function includes: annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost; the power system security constraints include: existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and maximum renewable energy curtailment ratio.

[0044] Furthermore, the determining module includes a first determining unit and a second determining unit; the first determining unit is used to determine extreme renewable energy scenarios, specifically including:

[0045] The actual impact of extreme situations is measured using the load loss risk index; the system load loss at node n in time period t is set as follows under the renewable energy time-series scenario ξ. The hourly load loss factor and the daily load loss factor are used to describe the maximum load loss risk per unit time period and the cumulative daily load loss risk of the power system under this scenario, respectively. The calculation formula is as follows:

[0046]

[0047]

[0048] Where D t Let ξ be the total load demand of the power system during time period t; the subscript ξ|O represents the scenario ξ acting on planning scheme O; acceptable unit time period and daily cumulative load shedding thresholds are set based on manual experience or specific circumstances, when HLOLF ξ|O ≥ε HOr DLOLF ξ|O ≥ε D When the scenario is in the high-impact scenario for planning scheme O, it is considered a low-impact scenario; otherwise, it is considered a low-impact scenario.

[0049] The probability of extreme cases is described using a significance level α. When the proportion of LPHI scene samples in the generated scene set is less than the significance level α, the occurrence of an extreme scene is a low-probability event. In this case, the probability of the LPHI scene occurring is estimated using the number of HI samples, satisfying the following:

[0050]

[0051] Where S LPHI For LPHI scene set, S RES To generate the scene set, N(·) is the number of scene samples;

[0052] The second determining unit is used to determine the conditions for determining extreme scenarios of renewable energy, including: determining the first condition for determining extreme scenarios of renewable energy and the second condition for determining extreme scenarios of renewable energy;

[0053] The first criterion for determining extreme scenarios for renewable energy: HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D ;

[0054] Second criterion for extreme scenarios of renewable energy: P(ξ∈S) LPHI )≤α.

[0055] Furthermore, the analysis and processing module includes: an analysis and processing unit; the analysis and processing unit is used to screen and process random renewable energy scenarios, and then perform extreme scenario analysis based on the determined extreme renewable energy scenarios and judgment conditions; specifically including:

[0056] Based on the REM model, the stochastic scenario of renewable energy is solved to obtain the load shedding information of the system at each time period under the renewable energy time series scenario ξ. Based on the above-established extreme scenario of renewable energy, the load shedding risk index HLOLF is calculated. ξ|O and DLOLF ξ|O The LPHI scenario is determined based on the first criterion for extreme renewable energy scenarios, and P(ξ∈S) is calculated. LPHI );

[0057] The update loop module includes an update loop unit; the update loop unit is used to adaptively update the typical scenario set based on the extreme scenario analysis results, and start the next loop until the low probability condition is met; specifically, it includes:

[0058] When the second criterion for extreme renewable energy scenarios is not met, the scenario with the least load loss is selected from the LPHI scenarios and added to the typical scenario set to update the typical scenario set, and the next cycle is started; until the second criterion for extreme renewable energy scenarios is met, the LPHI scenario set that meets the criterion and the final planning scheme are obtained.

[0059] Furthermore, taking the minimum cost of load shedding penalty as the objective function, and considering the satisfaction of power system security constraints under a single scenario, a REM model is established.

[0060] A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in any one of the preceding descriptions.

[0061] Compared with the prior art, the advantages of the present invention are as follows:

[0062] A method and system for transmission network expansion planning considering extreme scenarios of renewable energy. The planning method of this invention includes four steps: massive random scenario generation, TEP planning of typical scenario sets, risk assessment and extreme scenario analysis, and adaptive updating of typical scenario sets.

[0063] 1) Generate a massive amount of temporally correlated random scenarios of renewable energy based on GAN generalization of historical datasets;

[0064] 2) A proposed planning scheme is generated using a TEP expected value model based on a typical scenario set, and the planning scheme satisfies the feasibility of the HPLI scenario;

[0065] 3) Define extreme scenarios for renewable energy and conduct rapid LPHI scenario screening based on the REM model;

[0066] 4) Conduct extreme scenario analysis and adaptively update the typical scenario set based on the analysis results as input for the next stage of the TEP model.

[0067] The optimization of planning schemes is an interactive iterative process of extreme scenario analysis and planning decision correction. The LPHI and HPLI scenarios are dynamically transformed as the planning scheme is adjusted. The final planning scheme is obtained by verifying the LPHI extreme scenario judgment conditions.

[0068] The definition of extreme renewable energy scenarios proposed in this invention considers both "low probability" and "high impact" factors, providing a comprehensive and accurate assessment of the actual operating state of the power system. The data-driven iterative planning method for transmission networks that considers extreme renewable energy scenarios proposed in this invention achieves a balance between planning economy and robustness. Attached Figure Description

[0069] Figure 1A power grid expansion planning framework considering extreme scenarios;

[0070] Figure 2 Framework for generating and reducing massive random scenes;

[0071] Figure 3 Garver-6 node system. Detailed Implementation

[0072] The specific implementation of the present invention is described below with reference to embodiments:

[0073] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0074] Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity of description and are not intended to limit the scope of the invention. Any changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

[0075] Example 1:

[0076] Definitions of abbreviations and key terms:

[0077] Transmission expansion planning (TEP);

[0078] Low-probability, high-impact (LPHI)

[0079] High-probability, low-impact (HPLI)

[0080] Risk evaluation model (REM);

[0081] Generative adversarial network (GAN);

[0082] To address the shortcomings of technique one, a scenario generation model is employed. A GAN-based scenario generation model generalizes historical operational data to generate massive amounts of spatiotemporally correlated random renewable energy scenarios. Compared to explicit models of renewable energy uncertainty, this method, benefiting from the high-dimensional nonlinear mapping capabilities of deep neural networks, can implicitly extract the temporal and spatial features of multi-source-load relationships. The probability distribution of the generated scenarios shows high consistency with historical data and includes a small number of LPHI extreme scenarios, making it effective input information for power system expansion planning.

[0083] To address the shortcomings of Technique 2, this paper proposes criteria for determining extreme scenarios in renewable energy and a risk assessment model. A set of typical renewable energy scenarios represents the characteristics of HPLI scenarios and includes a small number of correlated multi-source-load time-series scenarios. The TEP model based on this typical scenario set is a convex mixed-integer linear programming problem, which can be easily solved using a solver. Unlike existing methods that define extreme cases based on the renewable energy output space, this paper provides criteria for determining extreme scenarios in LPHI from both the actual impact and probability of occurrence. A REM model is used to screen a massive number of renewable energy scenarios for extreme scenarios, and the typical scenario set is adaptively updated based on the extreme scenario analysis results to balance the economy and robustness of the TEP planning results. The REM model is a convex linear programming problem that can be solved in parallel for a large number of scenarios.

[0084] like Figure 1 As shown, the method includes:

[0085] The planning method of this invention includes four steps: massive random scene generation, TEP planning of typical scene sets, risk assessment and extreme scene analysis, and adaptive updating of typical scene sets.

[0086] 1) Generate a massive amount of temporally correlated random scenarios of renewable energy based on GAN generalization of historical datasets;

[0087] 2) A proposed planning scheme is generated using a TEP expected value model based on a typical scenario set, and the planning scheme satisfies the feasibility of the HPLI scenario;

[0088] 3) Define extreme scenarios for renewable energy and conduct rapid LPHI scenario screening based on the REM model;

[0089] 4) Conduct extreme scenario analysis and adaptively update the typical scenario set based on the analysis results as input for the next stage of the TEP model.

[0090] The optimization of the planning scheme is an interactive iterative process of extreme scenario analysis and planning decision correction. The LPHI and HPLI scenarios dynamically transform as the planning scheme is adjusted. The final planning scheme is obtained by verifying the LPHI extreme scenario judgment conditions. The solution framework is as follows: Figure 1As shown.

[0091] The specific implementation of each part is as follows:

[0092] (1) Generation of massive random scenes

[0093] Wind and solar power plants have relatively short operational histories and limited data storage capacity, making it difficult to effectively analyze future extreme scenarios using only a small amount of historical data. This paper employs a GAN-based scenario generation model to generalize historical operational data and generate a massive number of spatiotemporally correlated random renewable energy scenarios. The probability distribution of the generated scenarios shows a high degree of consistency with the historical data. The generated massive number of scenarios include a large number of HPLI scenarios. A scenario reduction network is used to extract representative typical scenarios from this massive dataset. These typical scenarios can represent numerous HPLI scenarios. The framework for massive random scenario generation and scenario reduction is as follows: Figure 2 As shown.

[0094] (2) TEP model based on typical scenario set

[0095] A two-stage TEP (Transmission Power Expectation) model is established, with the objective function being the minimum annual cost. This model includes the annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost. All typical scenarios satisfy power system security constraints, and the load demand at each node is strictly met. Model constraints include existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and the maximum renewable energy curtailment ratio. Solving this model yields the proposed transmission network planning scheme O for the k-th iteration. k .

[0096] (3) Risk assessment

[0097] i) Definition of extreme scenarios for renewable energy

[0098] Extreme scenarios for renewable energy have the characteristics of LPHI. Their definition takes into account both the actual impact and the probability of occurrence of extreme situations. It is a comprehensive examination of the actual operating state of the power system, and its boundary is constantly adjusted as the source-grid-load state changes.

[0099] The actual impact of extreme situations is measured using the load loss risk index. Assuming the system load loss at node n in time period t under the renewable energy time-series scenario ξ is... The hourly loss of load factor (HLOLF) and daily loss of load factor (DLOLF) are defined to describe the maximum loss of load risk per unit time period and the cumulative daily loss of load risk of the power system under this scenario, respectively. The calculation formulas are as follows:

[0100]

[0101]

[0102] Where D t Let be the total load demand of the power system during time period t; the subscript ξ|O indicates the scenario ξ acting on planning scheme O. Acceptable unit time periods and daily cumulative load shedding thresholds are set based on manual experience or specific circumstances, when HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D When the impact is high, the scenario is considered a high-impact (HI) scenario for planning scheme O; otherwise, it is considered a low-impact scenario.

[0103] The probability of extreme cases is described using a significance level α. When the number of samples in the generated scenario set is sufficiently large, the occurrence of the LPHI scenario becomes a low-probability event. In this case, the probability of the LPHI scenario occurring is estimated using the number of HI samples, satisfying the following condition:

[0104]

[0105] Where S LPHI For LPHI scene set, S RES To generate a scene set, N(·) represents the number of scene samples.

[0106] The criteria for determining "high impact" and "low probability" in extreme scenarios for renewable energy LPHI are as follows:

[0107] i)HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D ;

[0108] ii)P(ξ∈S LPHI )≤α.

[0109] ii) Risk assessment model

[0110] The proposed planning scheme O obtained by solving the TEP model k This provides identification targets for extreme scenario analysis of the generated scenario set. A system is established based on the proposed planning scheme O. k The risk assessment model, with the objective function of minimizing the cost of load shedding penalty, examines the satisfaction of power system security constraints under a single scenario. The REM and TEP models have the same security constraints, and the network topology is O. k Given known information and allowing for node load deficits, REM is guaranteed to have a feasible solution in any scenario. REM is a linear programming problem with relatively low computational complexity, allowing for parallel computation.

[0111] (4) Extreme scenario analysis and adaptive update of typical scenario sets

[0112] Solving REM yields the system load loss information for each time period under scenario ξ. The load loss risk index HLOLF can be calculated using formulas (1)-(2). ξ|O and DLOLF ξ|O Based on the extreme scenario determination condition i), determine the LPHI scenario and calculate P(ξ∈S). LPHI When the extreme scenario determination condition ii) is not met, the scenario with the minimum load loss is selected from the LPHI scenarios and added to the typical scenario set, and the next round of looping begins. This continues until the "low probability" condition is met, that is, the LPHI scenario set that meets the determination condition and the final planning scheme are obtained.

[0113] In summary, the solution steps for the transmission network expansion planning method considering extreme scenarios of renewable energy proposed in this invention are as follows:

[0114]

[0115]

[0116] Example 2:

[0117] The definition of extreme renewable energy scenarios proposed in this invention considers both "low probability" and "high impact" factors, providing a comprehensive and accurate assessment of the actual operating state of the power system. 2) The proposed data-driven iterative planning method for transmission networks, which considers extreme renewable energy scenarios, achieves a balance between planning economy and robustness.

[0118] A case study analysis is conducted on the improved Garver-6 node system, with the system topology as follows: Figure 3 As shown. Renewable energy data uses historical measured power output from a wind-solar power station in Northwest China from January 1, 2019 to October 31, 2020, with a data interval of 1 hour, totaling 670 historical scenarios. A GAN was used to generalize 10,000 scenarios to generate new ones; for example... Figure 3 As shown.

[0119] The power transmission network planning problem is solved using existing methods based on the expected value model and the method of this patent, respectively. The planning results are analyzed in extreme scenarios, and the comparison results are shown in Table 1.

[0120] Table 1 Comparison of Garver-6 Node System Planning Results

[0121]

[0122] Existing methods based on the expected value model (EVM) plan require fewer new transmission lines and offer good economic efficiency. However, extreme scenario analysis shows that the probability of the LPHI extreme scenario is 4.43%, which cannot be considered a low-probability event and could potentially lead to severe power grid outages, indicating poor robustness. The method of this invention, by constructing an additional transmission line, reduces the probability of the LPHI extreme scenario to 0.28% (<1%), which can be considered a low-probability event, achieving a balance between economic efficiency and robustness.

[0123] Example 3:

[0124] To better implement the above methods, this embodiment provides a power grid expansion planning system that takes into account extreme scenarios of renewable energy.

[0125] For example, a transmission grid expansion planning system considering extreme renewable energy scenarios, the system includes:

[0126] The determination module, the first determination unit, the second determination unit, and the processing module are used to generalize the historical dataset using the scene generation model of Generative Adversarial Network (GAN), generate tens of thousands of spatiotemporally correlated random renewable energy scenes, and perform scene simplification processing to obtain a typical scene set.

[0127] The solution module is used to establish and solve the expected value model of the two-stage transmission network expansion plan (TEP) based on a set of typical scenarios, and obtain the proposed planning scheme of the transmission network in the kth iteration.

[0128] The determination module is used to obtain the actual impact and probability of occurrence of extreme situations, and to determine the extreme scenarios and judgment conditions for renewable energy.

[0129] The analysis and processing module is used to establish a risk assessment model REM based on the above-mentioned proposed planning scheme of the transmission network, and to conduct extreme scenario analysis based on the determined extreme scenarios and judgment conditions of renewable energy after screening and processing the random scenarios of renewable energy based on REM.

[0130] The update loop module is used to adaptively update the typical scenario set based on the extreme scenario analysis results and start the next loop until the low probability condition is met, that is, to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions.

[0131] Furthermore, the solution module includes a solution unit; the solution unit is used to establish a two-stage TEP expected value model with the minimum annual cost as the objective function, considering the power system security constraints under typical scenario sets; the minimum annual cost as the objective function includes: annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost; the power system security constraints include: existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and maximum renewable energy curtailment ratio.

[0132] Furthermore, the determining module includes a first determining unit and a second determining unit; the first determining unit is used to determine extreme renewable energy scenarios, specifically including:

[0133] The actual impact of extreme situations is measured using the load loss risk index; the system load loss at node n in time period t is set as follows under the renewable energy time-series scenario ξ. The hourly load loss factor and the daily load loss factor are used to describe the maximum load loss risk per unit time period and the cumulative daily load loss risk of the power system under this scenario, respectively. The calculation formula is as follows:

[0134]

[0135]

[0136] Where D t Let ξ be the total load demand of the power system during time period t; the subscript ξ|O represents the scenario ξ acting on planning scheme O; acceptable unit time period and daily cumulative load shedding thresholds are set based on manual experience or specific circumstances, when HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D When the scenario is in the high-impact scenario for planning scheme O, it is considered a low-impact scenario; otherwise, it is considered a low-impact scenario.

[0137] The probability of extreme cases is described using a significance level α. When the proportion of LPHI scene samples in the generated scene set is less than the significance level α, the occurrence of an extreme scene is a low-probability event. In this case, the probability of the LPHI scene occurring is estimated using the number of HI samples, satisfying the following:

[0138]

[0139] Where S LPHI For LPHI scene set, S RES To generate the scene set, N(·) is the number of scene samples;

[0140] The second determining unit is used to determine the conditions for determining extreme scenarios of renewable energy, including: determining the first condition for determining extreme scenarios of renewable energy and the second condition for determining extreme scenarios of renewable energy;

[0141] The first criterion for determining extreme scenarios for renewable energy: HLOLF ξ|O ≥ε H Or DLOLF ξ|O ≥ε D ;

[0142] Second criterion for extreme scenarios of renewable energy: P(ξ∈S) LPHI )≤α.

[0143] Furthermore, the analysis and processing module includes: an analysis and processing unit; the analysis and processing unit is used to screen and process random renewable energy scenarios, and then perform extreme scenario analysis based on the determined extreme renewable energy scenarios and judgment conditions; specifically including:

[0144] Based on the REM model, the stochastic scenario of renewable energy is solved to obtain the load shedding information of the system at each time period under the renewable energy time series scenario ξ. Based on the above-established extreme scenario of renewable energy, the load shedding risk index HLOLF is calculated. ξ|O and DLOLF ξ|O The LPHI scenario is determined based on the first criterion for extreme renewable energy scenarios, and P(ξ∈S) is calculated. LPHI );

[0145] The update loop module includes an update loop unit; the update loop unit is used to adaptively update the typical scenario set based on the extreme scenario analysis results, and start the next loop until the low probability condition is met; specifically, it includes:

[0146] When the second criterion for extreme renewable energy scenarios is not met, the scenario with the least load loss is selected from the LPHI scenarios and added to the typical scenario set to update the typical scenario set, and the next cycle is started; until the second criterion for extreme renewable energy scenarios is met, the LPHI scenario set that meets the criterion and the final planning scheme are obtained.

[0147] Example 4:

[0148] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0149] To this end, embodiments of the present invention provide a storage medium storing a plurality of instructions which can be loaded by a processor to execute steps in any of the transmission grid expansion planning methods considering extreme scenarios of renewable energy provided in embodiments of the present invention.

[0150] For example, this instruction can perform the following steps:

[0151] A transmission grid expansion planning method considering extreme scenarios for renewable energy, the method comprising:

[0152] A scene generation model using a generative adversarial network (GAN) is used to generalize the historical dataset, generating tens of thousands of spatiotemporally correlated random renewable energy scenes. Scene simplification is then performed to obtain a typical scene set.

[0153] A two-stage transmission network expansion planning TEP expectation value model is established and solved based on a typical scenario set to obtain the proposed transmission network planning scheme for the kth iteration.

[0154] To obtain the actual impact and probability of occurrence of extreme situations, and to determine extreme scenarios and criteria for renewable energy;

[0155] Using the aforementioned proposed planning scheme for the power transmission network, a risk assessment model REM is established. Based on the REM model, random scenarios of renewable energy are screened and processed, and extreme scenario analysis is conducted based on the determined extreme scenarios and judgment conditions of renewable energy.

[0156] The typical scenario set is adaptively updated based on the results of extreme scenario analysis, and the next round of loop is started until the low probability condition is met, so as to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions.

[0157] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0158] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0159] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0160] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

[0161] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. A transmission grid expansion planning method considering extreme scenarios of renewable energy, characterized in that, The method includes: A scene generation model using a generative adversarial network (GAN) is used to generalize the historical dataset, generating tens of thousands of spatiotemporally correlated random renewable energy scenes. Scene simplification is then performed to obtain a typical scene set. A two-stage transmission network expansion planning TEP expectation value model is established and solved based on a set of typical scenarios to obtain the proposed transmission network planning scheme for the kth iteration. To obtain the actual impact and probability of occurrence of extreme situations, and to determine extreme scenarios and criteria for renewable energy; Using the aforementioned proposed planning scheme for the power transmission network, a risk assessment model REM is established. Based on the REM model, random scenarios of renewable energy are screened and processed, and extreme scenario analysis is conducted based on the determined extreme scenarios and judgment conditions of renewable energy. The typical scenario set is adaptively updated based on the extreme scenario analysis results, and the next round of loop is started until the low probability condition is met, so as to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions. The determination of extreme scenarios for renewable energy specifically includes: The actual impact of extreme situations is measured using load shedding risk indicators; renewable energy time-series scenarios are determined. The system load loss at node n in time period t is The hourly load loss factor and daily load loss factor are used to describe the maximum load loss risk per unit time period and the daily cumulative load loss risk of the power system under this scenario, respectively. The calculation formula is as follows: ；（1） ；（2） in Let t represent the total load demand of the power system during time period t; subscript This represents the scenario acting on planning scheme O. Based on human experience or specific circumstances, set acceptable unit time period and daily cumulative load loss thresholds. or When the scenario is in the high-impact scenario for planning scheme O, the scenario is in the low-impact scenario. The significance level α is used to describe the probability of extreme cases. When the proportion of LPHI scene samples in the generated scene set is less than the significance level α, the occurrence of the extreme scene is determined to be a low-probability event. In this case, the probability of the LPHI scene occurring is estimated by the number of HI samples, satisfying the following: ；（3） in For LPHI scene set, To generate scene sets, Number of scene samples; Establish the first and second criteria for determining extreme scenarios of renewable energy. The first criterion for determining extreme scenarios for renewable energy: or ; Second criterion for determining extreme scenarios of renewable energy: .

2. The transmission network expansion planning method considering extreme scenarios of renewable energy as described in claim 1, characterized in that, A two-stage TEP (Temperature, Expectation, and Value) model is established based on a set of typical scenarios, including: With the minimum annual cost as the objective function, and considering the power system security constraints under typical scenarios, a two-stage TEP (Transmission Line Expectation Value) model is established. The objective function for minimizing annual cost includes: annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost. The power system security constraints include: existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and maximum renewable energy curtailment ratio.

3. The transmission network expansion planning method considering extreme scenarios of renewable energy as described in claim 1, characterized in that, After screening and processing random scenarios for renewable energy, extreme scenario analysis is conducted based on the identified extreme renewable energy scenarios and judgment conditions; specifically including: Solving the stochastic scenario of renewable energy based on the REM model yields the time series scenario of renewable energy. The system's load loss information for each time period is analyzed, and load loss risk indicators are calculated based on the aforementioned extreme renewable energy scenarios. and The LPHI scenario is determined based on the first criterion for extreme renewable energy scenarios, and calculations are performed. .

4. The transmission network expansion planning method considering extreme scenarios of renewable energy as described in claim 3, characterized in that, The typical scenario set is adaptively updated based on the results of extreme scenario analysis, and the next round of iteration is started until the low-probability condition is met; specifically including: When the second criterion for extreme renewable energy scenarios is not met, the scenario with the least load loss is selected from the LPHI scenarios and added to the typical scenario set to update the typical scenario set, and the next cycle is started; until the second criterion for extreme renewable energy scenarios is met, the LPHI scenario set that meets the criterion and the final planning scheme are obtained.

5. A power grid expansion planning system considering extreme scenarios of renewable energy, characterized in that the system comprises: The processing module is used to generalize the historical dataset using a scene generation model based on a generative adversarial network (GAN), generate tens of thousands of spatiotemporally correlated random renewable energy scenes, and perform scene simplification to obtain a typical scene set. The solution module is used to establish and solve the expected value model of the two-stage transmission network expansion plan (TEP) based on a set of typical scenarios, and obtain the proposed planning scheme of the transmission network in the kth iteration. The determination module is used to obtain the actual impact and probability of occurrence of extreme situations, and to determine the extreme scenarios and judgment conditions for renewable energy. The analysis and processing module is used to establish a risk assessment model REM based on the above-mentioned proposed planning scheme of the transmission network, and to conduct extreme scenario analysis based on the determined extreme scenarios and judgment conditions of renewable energy after screening and processing the random scenarios of renewable energy based on REM. The update loop module is used to adaptively update the typical scenario set based on the extreme scenario analysis results and start the next loop until the low probability condition is met, that is, to obtain the low probability high impact LPHI scenario set and the final planning scheme that meet the judgment conditions. The determining module includes a first determining unit and a second determining unit; the first determining unit is used to determine extreme renewable energy scenarios, specifically including: The actual impact of extreme situations is measured using a load shedding risk index; renewable energy time-series scenarios are defined. The system load loss at node n in time period t is The hourly load loss factor and daily load loss factor are used to describe the maximum load loss risk per unit time period and the daily cumulative load loss risk of the power system under this scenario, respectively. The calculation formula is as follows: ；（1） ；（2） in Let t represent the total load demand of the power system during time period t; subscript This represents the scenario acting on planning scheme O. Based on human experience or specific circumstances, set acceptable unit time period and daily cumulative load loss thresholds. or When the scenario is in the high-impact scenario for planning scheme O, the scenario is in the low-impact scenario. The probability of extreme cases is described using a significance level α. When the proportion of LPHI scene samples in the generated scene set is less than the significance level α, the occurrence of an extreme scene is a low-probability event. In this case, the probability of the LPHI scene occurring is estimated using the number of HI samples, satisfying the following: ；（3） in For LPHI scene set, To generate scene sets, Number of scene samples; The second determining unit is used to determine the conditions for determining extreme scenarios of renewable energy, including: determining the first condition for determining extreme scenarios of renewable energy and the second condition for determining extreme scenarios of renewable energy; The first criterion for determining extreme scenarios for renewable energy: or ; Second criterion for determining extreme scenarios of renewable energy: .

6. The power grid expansion planning system considering extreme scenarios of renewable energy according to claim 5, characterized in that, The solution module includes a solution unit; the solution unit is used to establish a two-stage TEP expected value model with the minimum annual cost as the objective function and considering the power system security constraints under typical scenario sets. The objective function is to minimize annual costs, including: annual value of transmission line investment, annual generation cost, and renewable energy curtailment penalty cost; power system security constraints include: existing line status constraints, node power balance constraints, generator output constraints, line power flow constraints, system reserve capacity constraints, node voltage phase angle constraints, and maximum renewable energy curtailment ratio.

7. The power grid expansion planning system considering extreme scenarios of renewable energy according to claim 6, characterized in that, The analysis and processing module includes: an analysis and processing unit; the analysis and processing unit is used to screen and process random renewable energy scenarios, and then perform extreme scenario analysis based on the determined extreme renewable energy scenarios and judgment conditions; specifically including: Solving the stochastic scenario of renewable energy based on the REM model yields the time series scenario of renewable energy. The system's load loss information for each time period is analyzed, and load loss risk indicators are calculated based on the aforementioned established extreme renewable energy scenarios. and The LPHI scenario is determined based on the first criterion for extreme renewable energy scenarios, and calculations are performed. ; The update loop module includes an update loop unit; the update loop unit is used to adaptively update the typical scenario set based on the extreme scenario analysis results, and start the next loop until the low probability condition is met; specifically, it includes: When the second criterion for extreme renewable energy scenarios is not met, the scenario with the least load loss is selected from the LPHI scenarios and added to the typical scenario set to update the typical scenario set, and the next cycle is started; until the second criterion for extreme renewable energy scenarios is met, the LPHI scenario set that meets the criterion and the final planning scheme are obtained.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method of any one of claims 1-4.

Images