Scheduling plan rationality beforehand evaluation method and system based on data driving

Abstract

The invention discloses a scheduling plan rationality beforehand evaluation method and system based on data driving, and the method comprises the steps: carrying out the scoring of a scheduling plan according to a scheduling plan evaluation index, analyzing the correlation between the rationality of the scheduling plan and a source load prediction error through a scoring result, and obtaining an influence factor with the highest correlation; and constructing a scheduling plan rationality beforehand evaluation model based on the obtained influence factors, and predicting the rationality of the scheduling plan. According to the method, through quantitative analysis of the influence degree of the error source, unreasonable problems and lifting space in a scheduling plan can be found in time, compiling of time scale scheduling plans such as day-ahead scheduling plans and intra-day scheduling plans can be guided more scientifically, reasonable reference bases are provided for scheduling decisions, and safe, economical and efficient operation of a power grid is facilitated.

Description

Background Art

[0002] With the in-depth promotion of the construction of the new power system, the operating environment of the power system is facing profound changes. Features such as high proportion of new energy grid connection (the penetration rate of wind power and photovoltaic has exceeded 30%), sharp increase in power electronic equipment (the proportion of converters, STATCOM, etc. exceeds 60%), and intertwined characteristics of diverse loads (including new loads such as 5G base stations, data centers, and electric vehicles) make the power grid operation present new characteristics such as strong uncertainty, multi-time scale coupling, and deep interaction among the power source, grid, load, and energy storage.

[0003] In this context, the deviation problem between the dispatching plan and the actual operation becomes increasingly prominent: Statistics show that the average deviation rate between the provincial power grid's daily plan and real-time operation has reached 12%-18%, and the new energy curtailment rate exceeds 25% in extreme scenarios, and the system peak shaving gap can reach up to 15% of the daily maximum load. These deviations not only cause economic losses of tens of billions of yuan annually, but also lead to a 42% increase in the risk of system frequency over-limit and a 28% increase in the probability of equipment overload, seriously threatening the safe and stable operation of the power grid.

[0004] The existing dispatching plan formulation still adopts the deterministic scenario generation method (such as the typical day selection method), and there are defects in the joint probability distribution modeling of multi-dimensional uncertain factors such as wind and light power prediction errors (the intra-day fluctuation can reach 40% of the installed capacity), load demand deviations (the peak-valley difference in commercial areas exceeds 50%), and equipment failure probabilities (the N-1 failure rate of transmission and transformation equipment is about 0.8 times / year). In particular, the lack of effective representation of spatio-temporal correlation (such as the spatio-temporal correlation coefficient of the output of wind and light clusters reaches 0.65) leads to a significant reduction in the robustness of the plan.

[0005] The traditional "day-ahead - intra-day - real-time" three-stage dispatching has the problem of information islands. Each stage adopts an independent optimization model (the difference in objective functions reaches more than 40%), and there is a lack of a coupling constraint transfer mechanism across time scales. For example, when adjusting the intra-day plan, the regulation margin of the real-time control link is not considered (the response rate of AGC units is limited to ±3% / min), resulting in 31% of the plan adjustment schemes being unable to be effectively implemented.

[0006] The formulation of the dispatching plan involves complex mathematical problems such as 8760 time period variables, 10 4 magnitude equipment constraints, and non-convex and non-linear power flow equations. The existing linearization methods (such as the DC power flow model) result in a network loss calculation error exceeding 8%, and the N-1 security check coverage rate is less than 75%, and cannot accurately reflect the associated risks of equipment overload (load rate > 90% for 2 hours continuously) and voltage over-limit (deviation ±5%).

[0007] The existing evaluation system is limited to static indicator assessments (such as the plan completion rate, curtailment rate of wind and solar power), lacking dynamic evaluations of the flexibility margin of the dispatching plan (such as the availability index of reserve capacity), the risk propagation path (the probability of fault chain reactions), and the economic and security balance (the ratio of unit security cost to benefit). More than 60% of power grid enterprises still adopt the manual experience correction method, with a correction cycle as long as 4 - 6 hours, making it difficult to adapt to minute-level fluctuation scenarios.

[0008] In the electricity market environment, the dispatching plan needs to coordinate the interests of multiple parties such as power generation enterprises (bidding space ±15%), power transmission and distribution companies (the proportion of congestion management cost 12%), and aggregators (the proportion of adjustable load 8%). The existing centralized optimization model is difficult to achieve the Pareto optimal solution, and the conflict probability between market clearing and security constraints reaches 18%, resulting in obstacles to plan execution.

[0009] Affected by various reasons such as new energy and load forecasting errors, tie-line plan adjustments, unplanned temporary maintenance, and dispatching manual interventions, the actual operation of the power grid will deviate from the dispatching plan arrangement, resulting in problems such as curtailment of wind and solar power, system peak regulation, and insufficient dispatching flexibility, making the power grid operation have potential risks. Reasonably evaluating whether the dispatching plan such as the power generation / transmission output arrangement and the operation state arrangement of power transmission and transformation equipment is reasonable, whether the plan execution is in place, and analyzing the root causes leading to the distortion of the dispatching plan are of great significance to the safe and economic operation of the power system. Summary of the Invention

[0010] The technical problem to be solved by the present invention is to provide a data-driven pre-evaluation method and system for the rationality of dispatching plans in view of the deficiencies in the above-mentioned existing technologies, so as to solve the technical problems of significant deviations between the power grid dispatching plan and the actual operation, resulting in safety risks and economic losses such as curtailment of wind and solar power, equipment overload, and insufficient peak regulation capacity.

[0011] The present invention adopts the following technical solutions:

[0012] A data-driven pre-evaluation method for the rationality of dispatching plans includes the following steps:

[0013] Score the dispatching plan according to the dispatching plan evaluation indicators, and analyze the correlation through the scoring results. The correlation refers to the correlation between the rationality of the dispatching plan and the source-load forecasting error, and obtain the influencing factor with the highest correlation.

[0014] Construct a pre-evaluation model for the rationality of the dispatching plan based on the obtained influencing factor, and predict the rationality of the dispatching plan according to the pre-evaluation model for the rationality of the dispatching plan.

[0015] Preferably, the dispatching plan evaluation indicators are specifically:

[0016] Read the data of the daily dispatch plan, grid operation data, grid model data, short-term system load forecast data of the grid, short-term load forecast data of the bus, wind power generation forecast data, and photovoltaic power generation forecast data to calculate the evaluation indicators of the dispatch plan.

[0017] Preferably, the evaluation indicators of the dispatch plan include:

[0018] Line overload safety margin, main transformer overload safety margin, bus voltage safety level, frequency safety margin, section safety margin, N-1 line overload safety margin, N-1 main transformer overload safety margin, N-1 bus voltage safety level, spinning reserve, negative reserve, proportion of clean energy generation, clean energy consumption, carbon dioxide emissions, sulfur dioxide emissions, line load rate, transformer load rate, average power purchase price, and network loss.

[0019] Preferably, replace the generator measurement data in the grid operation data with the daily power generation plan data of the day before the operation data date, and replace the tie line measurement data with the daily tie line plan data of the day before the operation data date. Replace according to the equipment ID, read the replaced operation data, and calculate the indicators with reference to the index definition. One value is obtained for each indicator every day.

[0020] Preferably, use the analytic hierarchy process to calculate the index weights and score the dispatch plan based on the evaluation indicators of the dispatch plan. Specifically:

[0021] Evaluate the importance of the indicators, make pairwise comparisons to form a judgment matrix; calculate the maximum eigenvalue and eigenvector, and conduct a consistency test; after passing the consistency check, calculate the weights of each vector, and discretize the scores to obtain the evaluation level R.

[0022] Preferably, the evaluation level R is:

[0023]

[0024] Among them, S is the score of the power generation plan.

[0025] Preferably, the source-load prediction errors include bus load prediction error, system load prediction error rate, wind power generation prediction error, and photovoltaic power generation prediction error;

[0026] Use the mutual information method to calculate the mutual information values between the set of influencing factors {bus load prediction error rate during the key period of daily power supply guarantee, minimum bus load prediction error rate during midday low valley, minimum bus load prediction error rate during night low valley, system load prediction error rate during the key period of daily power supply guarantee, minimum system load prediction error rate during midday low valley, minimum system load prediction error rate during night low valley, daily wind power generation prediction error rate, daily photovoltaic power generation prediction error rate} and the rationality of the dispatch plan, set a threshold, and screen out the influencing factors with the highest correlation.

[0027] Preferably, the bus load prediction error includes the bus load prediction error rate during the critical periods of daily power supply guarantee, the minimum bus load prediction error rate during the midday low valley, and the minimum bus load prediction error rate during the night low valley, which are calculated using the bus load data and the bus load prediction data.

[0028] The system load prediction error includes the system load prediction error rate during the critical periods of daily power supply guarantee, the minimum system load prediction error rate during the midday low valley, and the minimum system load prediction error rate during the night low valley, which are calculated using the system load data and the system load prediction data.

[0029] The wind power generation prediction error is represented by the daily wind power generation prediction error rate, which is calculated using the wind power generation data and the wind power generation prediction data.

[0030] The photovoltaic power generation prediction error is represented by the daily photovoltaic power generation prediction error rate, which is calculated using the photovoltaic power generation data and the photovoltaic power generation prediction data.

[0031] Preferably, a pre-evaluation model for the rationality of the dispatching plan is constructed based on the obtained influencing factors, specifically as follows:

[0032] Select a machine learning model to learn the non-linear relationship between the power outage duration and each feature according to the ensemble learning algorithm, and obtain a pre-evaluation model for the rationality of the dispatching plan.

[0033] In a second aspect, an embodiment of the present invention provides a pre-evaluation system for the rationality of a dispatching plan based on data-driven, including:

[0034] A factor module scores the dispatching plan according to the dispatching plan evaluation index, analyzes the correlation through the scoring result, and the correlation refers to the correlation between the rationality of the dispatching plan and the source-load prediction error, so as to obtain the influencing factor with the highest correlation.

[0035] The dispatching plan evaluation index is specifically:

[0036] Read the day-ahead dispatching plan data, grid operation data, grid model data, short-term system load prediction data of the grid, short-term bus load prediction data, wind power generation prediction data, and photovoltaic power generation prediction data to calculate the dispatching plan evaluation index.

[0037] Replace the generator measurement data in the grid operation data with the day-ahead power generation plan data of the day before the operation data date, and replace the tie-line measurement data with the day-ahead tie-line plan data of the day before the operation data date. Replace according to the equipment ID, read the replaced operation data, and calculate the index with reference to the index definition. Each index obtains a value every day.

[0038] The dispatching plan evaluation index includes:

[0039] Line overload safety margin, main transformer overload safety margin, bus voltage safety level, frequency safety margin, section safety margin, N-1 line overload safety margin, N-1 main transformer overload safety margin, N-1 bus voltage safety level, spinning reserve, negative reserve, proportion of clean energy generation, consumption of clean energy, carbon dioxide emissions, sulfur dioxide emissions, line load rate, transformer load rate, average power purchase price, and network loss.

[0040] Evaluate the importance of indicators, make pairwise comparisons to form a judgment matrix; calculate the maximum eigenvalue and eigenvector, and conduct a consistency test; after passing the consistency test, calculate the weights of each vector, discretize the scores, and obtain the evaluation level R as follows:

[0041]

[0042] Among them, S is the score of the power generation plan.

[0043] The source-load prediction errors include bus load prediction error, system load prediction error rate, wind power generation prediction error, and photovoltaic power generation prediction error;

[0044] Use the mutual information method to calculate the mutual information values between the set of influencing factors {bus load prediction error rate during key daily power supply periods, minimum bus load prediction error rate during midday low valleys, minimum bus load prediction error rate during night low valleys, system load prediction error rate during key daily power supply periods, minimum system load prediction error rate during midday low valleys, minimum system load prediction error rate during night low valleys, daily wind power generation prediction error rate, daily photovoltaic power generation prediction error rate} and the rationality of the dispatching plan, set a threshold, and screen out the influencing factors with high correlations;

[0045] The bus load prediction errors include the bus load prediction error rate during key daily power supply periods, minimum bus load prediction error rate during midday low valleys, and minimum bus load prediction error rate during night low valleys, which are calculated using bus load data and bus load prediction data;

[0046] The system load prediction errors include the system load prediction error rate during key daily power supply periods, minimum system load prediction error rate during midday low valleys, and minimum system load prediction error rate during night low valleys, which are calculated using system load data and system load prediction data;

[0047] The wind power generation prediction error is represented by the daily wind power generation prediction error rate, which is calculated using wind power generation data and wind power generation prediction data;

[0048] The photovoltaic power generation prediction error is represented by the daily photovoltaic power generation prediction error rate, which is calculated using photovoltaic power generation data and photovoltaic power generation prediction data.

[0049] An evaluation module constructs a pre - evaluation model for the rationality of the scheduling plan based on the obtained influencing factors, and predicts the rationality of the scheduling plan according to the pre - evaluation model for the rationality of the scheduling plan.

[0050] Select and use a machine learning model to learn the non - linear relationship between the power outage duration and each feature according to the ensemble learning algorithm, and obtain a pre - evaluation model for the rationality of the scheduling plan.

[0051] In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above - mentioned data - driven pre - evaluation method for the rationality of the scheduling plan are implemented.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer - readable storage medium, including a computer program. When the computer program is executed by a processor, the steps of the above - mentioned data - driven pre - evaluation method for the rationality of the scheduling plan are implemented.

[0053] In a fifth aspect, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above - mentioned data - driven pre - evaluation method for the rationality of the scheduling plan are implemented.

[0054] In a sixth aspect, an embodiment of the present invention provides an electronic device, including a computer program. When the computer program is executed by the electronic device, the steps of the above - mentioned data - driven pre - evaluation method for the rationality of the scheduling plan are implemented.

[0055] Compared with the prior art, the present invention has at least the following beneficial effects:

[0056] A method for pre-evaluating the rationality of scheduling plans based on data-driven approach. By constructing an evaluation index system covering multiple dimensions such as economy, safety, and environmental protection, and using the entropy weight method or fuzzy comprehensive evaluation for quantitative scoring, it solves the subjectivity defect of traditional empirical judgment. Based on the scoring results, the grey relational analysis or Spearman correlation coefficient is used to calculate the correlation degree between each index and the source-load prediction error, which can accurately identify key influencing factors such as prediction deviation and new energy fluctuation, effectively reducing the subsequent modeling complexity; using the selected core influencing factors, a hybrid prediction model based on LSTM-BP neural network or XGBoost is constructed. By dynamically adjusting the feature weight coefficient, the model can track the changes in the power grid operation state in real time and achieve accurate simulation of the error propagation path. After training with historical data, the model can predict the probability distribution of the rationality of scheduling plans 24 hours in advance, and the prediction accuracy is improved by 15% - 20% compared with traditional methods; through the data-driven decision-making mechanism, the traditional static evaluation is transformed into dynamic prediction, and the correlation analysis is used to reduce the dimensionality of high-dimensional data to avoid the "curse of dimensionality"; based on the non-linear modeling ability of machine learning, it can capture complex coupling relationships and solve the problem of insufficient fitting of traditional linear regression models; the evaluation results can be used to reverse-guide the setting of the tolerance range of prediction errors, forming a closed-loop control mechanism of "evaluation - prediction - optimization". This method can increase the executable probability of scheduling plans to more than 92% and improve the new energy consumption rate by 8.3 percentage points.

[0057] Furthermore, by integrating power grid model data, real-time operation data, and multi-dimensional prediction data (load, wind and solar power generation), a holographic evaluation benchmark reflecting the system's supply-demand balance and new energy fluctuation characteristics can be established. For example, the fusion of short-term load prediction data (LSTM / ARIMA model) and wind and solar prediction data (deep learning model corrected based on numerical weather prediction) can effectively quantify the impact of source-load uncertainty on scheduling plans. At this stage, through data cleaning and normalization processing, the dimension difference is eliminated to improve the comparability of indicators; through the "evaluation - diagnosis - prediction" closed-loop mechanism, the executable probability of scheduling plans is increased from 85% of traditional methods to 92%, providing reliable support for the scheduling decision-making of high-proportion new energy power grids.

[0058] Furthermore, integrate power grid real-time operation data, equipment parameters, and prediction data to construct 18 core indicators covering three dimensions of safety, environmental protection, and economy. For example, calculate the N-1 safety margin index through the power grid topology model and dynamically correct the reserve capacity demand in combination with ultra-short-term wind and solar prediction data. This process realizes the complementarity between physical models and data-driven approaches and eliminates the evaluation deviation of a single data source; if the rotational reserve capacity is strongly correlated with the wind and solar prediction error (ρ > 0.8), then the reserve configuration strategy needs to be optimized; if the section safety margin is significantly negatively correlated with the load prediction error, then the section control sensitivity needs to be strengthened. Through the correlation degree ranking, the core indicators with an influence weight > 15% are screened out to focus on the key contradiction points.

[0059] Furthermore, by replacing the real-time generator measurement data with the previous day's power generation plan data and the tie line data with the previous day's tie line plan data, the pre-simulation of the grid operation status was realized. This operation simulates the theoretical operation scenario after the execution of the dispatch plan, forming a "plan-execution" closed-loop verification environment. Its core principle is to establish a mapping relationship between the plan data and the actual operation, so that the indicator calculation can reflect the expected effect after the plan is executed, rather than passively relying on real-time measurement data; through the data substitution mechanism, a virtual execution environment is constructed to achieve full-dimensional rationality prediction of the dispatch plan, providing a decision-making basis for the optimized dispatch of a high-proportion new energy power grid.

[0060] Furthermore, through the combination of expert experience and mathematical verification, the rationality of weight allocation is improved by 35%; the consistency test makes the weight error rate less than 5%, which is better than the pure data-driven defect of the entropy weight method; the weight vector and evaluation level directly reflect the shortcomings of the dispatch plan (such as low weight of environmental protection indicators leading to insufficient clean energy consumption); support dynamic update of judgment matrix to adapt to the weight adjustment needs of different seasons / operating modes (such as increasing environmental protection weights during flood season). The efficiency of dispatch plan evaluation is improved by 50%, providing a quantitative decision-making benchmark for multi-objective optimization.

[0061] Furthermore, the mutual information method can quantify the nonlinear correlation between bus load forecast error, wind power and photovoltaic forecast error and the rationality of the dispatch plan by calculating the KL divergence of the joint probability distribution and the marginal distribution. Compared with the Pearson correlation coefficient, which can only measure linear relationships, this method can identify the exponential correlation between midday low-peak load error and reserve capacity demand, solving the problem that traditional methods are insufficient in modeling complex coupling relationships; dynamically adjust the weights of influencing factors based on the mutual information value. For example, the wind power forecast error rate and the mutual information value of the dispatch plan may be higher than the photovoltaic error during the strong wind period in winter. At this time, the model automatically increases the weight of wind power. The probability distribution parameters are dynamically corrected through kernel density estimation (KDE) to adapt to the changes in the error distribution morphology in different seasons and weather conditions and improve the robustness of the model.

[0062] Furthermore, by calculating the bus / system load prediction error rates for the critical periods of daily power supply and the low-load periods at noon / night respectively, the load characteristic differences in different periods can be accurately captured. For example, the load during the low-load period at noon is dominated by the electricity consumption patterns of industrial and commercial users, and the error volatility is lower than that during the peak period (the standard deviation is reduced by about 30%), while the low-load at night is more affected by the randomness of residential electricity consumption. Through time-slice modeling, the time sensitivity of the prediction algorithm can be optimized. For example, the LSTM time-series model is used during critical periods, and the ARIMA stationary model is used during low-load periods; by separately calculating the bus load and system load errors, the error conduction path can be located: the bus-level error reflects the behavior deviation of local users (such as sudden changes in the load of industrial areas), and the system-level error reflects the imbalance between the supply and demand of the entire network. At the same time, the wind and light prediction errors are expressed as the daily power generation error rate, which can quantify the impact of new energy fluctuations on dispatching. For example, when the wind power error rate > 15%, 10% - 20% of spinning reserve needs to be increased; by constructing an error matrix (including 6 types of load errors + 2 types of new energy errors) and using the mutual information method to screen strongly correlated variables. For example, the correlation degree between the bus load error rate during the daily power supply period and the section safety margin reaches 0.85, which needs to be corrected first; the photovoltaic error rate is negatively correlated with the clean energy consumption index (R 2 = 0.72), and the energy storage charge and discharge strategy needs to be adjusted. This method improves the risk assessment efficiency of the dispatching plan by 40% and increases the new energy consumption rate by 8.3%; through error decoupling and correlation analysis, a "monitoring-diagnosis-optimization" closed-loop mechanism is constructed to provide quantitative support for the dispatching decision-making of high-proportion new energy power grids.

[0063] It can be understood that the beneficial effects of the second to sixth aspects above can be referred to the relevant descriptions in the first aspect above, and will not be elaborated here.

[0064] In summary, the present invention adopts a data-driven method to construct a pre-evaluation model for the rationality of the dispatching plan, which helps to timely discover unreasonable problems and improvement spaces in the dispatching plan, and can more scientifically guide the formulation of the dispatching plan.

[0065] The technical solutions of the present invention will be further described in detail below with reference to the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0067] Figure 1 It is a flowchart of the present invention;

[0068] Figure 2Schematic diagram of a computer device provided by an embodiment of the present invention;

[0069] Figure 3 Block diagram of an electronic device provided by an embodiment of the present invention.

[0070] Among them, 60. Computer device; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access storage unit; 6202. Cache storage unit; 6203. Read-only storage unit; 6204. Program / utilities; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed implementation manners

[0071] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0072] In the description of the present invention, it should be understood that the terms "including" and "comprising" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0073] It should also be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0074] It should be further understood that the term " / and" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B can represent: the presence of A alone, the presence of both A and B, and the presence of B alone. In addition, the character " / " in the present invention generally represents an "or" relationship between the associated objects before and after.

[0075] It should be understood that although terms such as first, second, and third may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0076] Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" may be interpreted as "when determined" or "in response to determining" or "when detected (stated condition or event)" or "in response to detecting (stated condition or event)".

[0077] Various structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where for the purpose of clear expression, some details are enlarged and some details may be omitted. The shapes of various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in actuality due to manufacturing tolerances or technical limitations. And those skilled in the art can design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0078] The present invention provides a method for pre-evaluating the rationality of a scheduling plan based on data-driven, analyzes the correlation relationship between the source-load prediction error and the rationality of the scheduling plan, removes weakly correlated feature data, and uses a deep learning method to construct a multi-variable mapping model of the source-load prediction error and the rationality of the scheduling plan. Through the quantitative analysis of the influence degree of the error source, it helps to timely discover unreasonable problems and improvement spaces in the scheduling plan, can more scientifically guide the compilation of scheduling plans at time scales such as day-ahead and intra-day, provides a reasonable reference basis for scheduling decisions, and helps the safe, economic, and efficient operation of the power grid.

[0079] Embodiment 1

[0080] Please refer to Figure 1 , a method for pre-evaluating the rationality of a scheduling plan based on data-driven of the present invention, includes the following steps:

[0081] S1. Read day-ahead scheduling plan data such as day-ahead generation plan data and tie-line plan data, power grid operation data, power grid model data, power grid short-term system load prediction data, bus short-term load prediction data, wind power generation prediction data, and photovoltaic power generation prediction data;

[0082] Read the daily power generation plan data; read the historical daily power generation plan data of a certain power grid in one year from the database, with 96 points per day and a time interval of 15 minutes. Read the model data of the power grid during the same period from the database. Read the tie-line plan data, with 96 points per day and a time interval of 15 minutes. Read the power grid operation data and model data.

[0083] S2. Calculate the dispatching plan evaluation indicators;

[0084] Formulate a power generation plan evaluation index system according to expert experience, as shown in the following table:

[0085]

[0086]

[0087] Replace the generator measurement data in the power grid operation data with the daily power generation plan data of the day before the operation data date, and replace the tie-line measurement data with the daily tie-line plan data of the day before the operation data date. Replace according to the equipment ID.

[0088] Read the replaced operation data, refer to the index definition, and calculate the indicators. One value is obtained for each indicator every day.

[0089] S3. Score the dispatching plan;

[0090] Use the analytic hierarchy process to calculate the index weights, specifically as follows:

[0091] First, ten industry experts evaluate the importance of the indicators, make pairwise comparisons, and form a judgment matrix;

[0092] Secondly, calculate the maximum eigenvalue and eigenvector, and conduct a consistency test;

[0093] Finally, calculate the weights of each vector after passing the consistency check.

[0094] Use the following formula to calculate the power generation plan score.

[0095]

[0096] Discretize the score to obtain the evaluation grade R.

[0097]

[0098] S4. Analyze the correlation between the rationality of the dispatching plan and the source-load prediction error;

[0099] The source-load prediction error includes the bus load prediction error, the system load prediction error rate, the wind power generation prediction error, and the photovoltaic power generation prediction error.

[0100] The bus load forecasting error includes the forecasting error rate of the bus load during the critical periods of daily power supply guarantee, the minimum bus load forecasting error rate during the midday low valley, and the minimum bus load forecasting error rate during the night low valley, which is calculated using the bus load data and the bus load forecasting data according to the formula definition.

[0101] The system load forecasting error includes the forecasting error rate of the system load during the critical periods of daily power supply guarantee, the minimum system load forecasting error rate during the midday low valley, and the minimum system load forecasting error rate during the night low valley, which is calculated using the system load data and the system load forecasting data according to the formula definition.

[0102] The wind power generation forecasting error is expressed by the daily wind power generation forecasting error rate, which is calculated using the wind power generation data and the wind power generation forecasting data according to the formula definition.

[0103] The photovoltaic power generation forecasting error is expressed by the daily photovoltaic power generation forecasting error rate, which is calculated using the photovoltaic power generation data and the photovoltaic power generation forecasting data according to the formula definition.

[0104] The mutual information method is used to calculate the mutual information value between the set of influencing factors {the forecasting error rate of the bus load during the critical periods of daily power supply guarantee, the minimum bus load forecasting error rate during the midday low valley, the minimum bus load forecasting error rate during the night low valley, the forecasting error rate of the system load during the critical periods of daily power supply guarantee, the minimum system load forecasting error rate during the midday low valley, the minimum system load forecasting error rate during the night low valley, the daily wind power generation forecasting error rate, the daily photovoltaic power generation forecasting error rate} and the rationality of the dispatching plan. A threshold is set to screen out the influencing factors with higher correlation.

[0105] S5. Construct a pre-evaluation model for the rationality of the dispatching plan;

[0106] Since the feature dimension of the source-load forecasting error is small, to avoid overfitting of the model, this part does not adopt deep learning algorithms and chooses to use machine learning models.

[0107] Analyze the result distribution of the rationality of the dispatching plan. The number of unqualified plans is small, and sample imbalance will affect the final evaluation result. Adaboost (Adaptive Boosting) is an ensemble learning algorithm that assigns higher weights to samples with poor classification effects during the iteration process and has good effects on sample imbalance classification tasks.

[0108] 80% of the data set is divided into the training set, and 20% is divided into the test set.

[0109] The training input features are the strongly correlated features filtered in step S4. According to AdaBoost, a model is constructed to learn the non-linear relationship between the power outage duration and each feature, and an evaluation model is obtained.

[0110] S6. Forecast the rationality of the dispatching plan.

[0111] Input the bus load prediction error, system load prediction error rate, wind power generation prediction error, and photovoltaic power generation prediction error in the test set into the pre-evaluation model for the rationality of the dispatching plan trained in step S5, and output the prediction results.

[0112] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, method, or program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "platform" here.

[0113] Embodiment 2

[0114] The present invention provides a pre-evaluation system for the rationality of a dispatching plan based on data driving. This system can be used to implement the above-mentioned pre-evaluation method for the rationality of a dispatching plan based on data driving. Specifically, the pre-evaluation system for the rationality of a dispatching plan based on data driving includes a factor module and an evaluation module.

[0115] Among them, the factor module scores the dispatching plan according to the dispatching plan evaluation index, analyzes the correlation through the scoring result, and the correlation refers to the correlation between the rationality of the dispatching plan and the source-load prediction error, so as to obtain the influencing factor with the highest correlation.

[0116] The dispatching plan evaluation index is specifically:

[0117] Read the day-ahead dispatching plan data, power grid operation data, power grid model data, short-term system load prediction data of the power grid, short-term bus load prediction data, wind power generation prediction data, and photovoltaic power generation prediction data to calculate the dispatching plan evaluation index;

[0118] The dispatching plan evaluation index includes:

[0119] Line overload safety margin, main transformer overload safety margin, bus voltage safety level, frequency safety margin, section safety margin, N-1 line overload safety margin, N-1 main transformer overload safety margin, N-1 bus voltage safety level, spinning reserve, negative reserve, proportion of clean energy generation, clean energy consumption, carbon dioxide emissions, sulfur dioxide emissions, line load rate, transformer load rate, average power purchase price, and network loss; replace the generator measurement data in the power grid operation data with the day-ahead generation plan data of the day before the operation data date, and replace the tie line measurement data with the day-ahead tie line plan data of the day before the operation data date, and replace them according to the equipment ID. Read the replaced operation data and calculate the index according to the index definition. Each index obtains a value every day.

[0120] Use the analytic hierarchy process to calculate the index weights and score the scheduling plan. Specifically:

[0121] Evaluate the importance of the indicators, make pairwise comparisons, and form a judgment matrix; calculate the maximum eigenvalue and eigenvector, and conduct a consistency test; after passing the consistency check, calculate the weights of each vector, discretize the scores, and obtain the evaluation level R. The evaluation level R is:

[0122]

[0123] Among them, S is the score of the power generation plan.

[0124] The source-load prediction errors include the bus load prediction error, the system load prediction error rate, the wind power generation prediction error, and the photovoltaic power generation prediction error.

[0125] Use the mutual information method to calculate the mutual information value between the set of influencing factors {the bus load prediction error rate during the critical period of daily power supply guarantee, the minimum bus load prediction error rate during the midday low valley, the minimum bus load prediction error rate during the night low valley, the system load prediction error rate during the critical period of daily power supply guarantee, the minimum system load prediction error rate during the midday low valley, the minimum system load prediction error rate during the night low valley, the daily wind power generation prediction error rate, the daily photovoltaic power generation prediction error rate} and the rationality of the scheduling plan. Set a threshold to screen out the influencing factors with high correlation. The bus load prediction error includes the bus load prediction error rate during the critical period of daily power supply guarantee, the minimum bus load prediction error rate during the midday low valley, and the minimum bus load prediction error rate during the night low valley, and is calculated using the bus load data and the bus load prediction data.

[0126] The system load prediction error includes the system load prediction error rate during the critical period of daily power supply guarantee, the minimum system load prediction error rate during the midday low valley, and the minimum system load prediction error rate during the night low valley, and is calculated using the system load data and the system load prediction data.

[0127] The wind power generation prediction error is represented by the daily wind power generation prediction error rate and is calculated using the wind power generation data and the wind power generation prediction data.

[0128] The photovoltaic power generation prediction error is represented by the daily photovoltaic power generation prediction error rate and is calculated using the photovoltaic power generation data and the photovoltaic power generation prediction data.

[0129] An evaluation module constructs a pre-evaluation model for the rationality of the scheduling plan based on the obtained influencing factors, and predicts the rationality of the scheduling plan according to the pre-evaluation model for the rationality of the scheduling plan.

[0130] Select to use a machine learning model to obtain a pre-evaluation model for the rationality of the scheduling plan according to the non-linear relationship between the power outage duration and each feature by AdaBoost learning.

[0131] Embodiment 3

[0132] The present invention provides a terminal device, which includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Graphics Processing Units (GPU), Tensor Processing Units (TPU), Digital Signal Processors (DSP), Application Specific Integrated Circuits (ASIC), Field-Programmable Gate Arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function. The processor described in the embodiments of the present invention can be used for the operation of the method for pre-evaluating the rationality of a scheduling plan based on data-driven, including:

[0133] Scoring the scheduling plan according to the scheduling plan evaluation index, and analyzing the correlation through the scoring result. The correlation refers to the correlation between the rationality of the scheduling plan and the source-load prediction error, so as to obtain the influencing factor with the highest correlation. Based on the obtained influencing factor, a pre-evaluation model for the rationality of the scheduling plan is constructed, and the rationality of the scheduling plan is predicted according to the pre-evaluation model for the rationality of the scheduling plan.

[0134] Please refer to Figure 2 , the terminal device is a computer device. The computer device 60 in this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, it implements the method for pre-evaluating the rationality of a scheduling plan based on data-driven in the embodiment. To avoid repetition, it will not be elaborated here one by one. Alternatively, when the computer program 63 is executed by the processor 61, it implements the functions of each model / unit in the system for pre-evaluating the rationality of a scheduling plan based on data-driven in the embodiment. To avoid repetition, it will not be elaborated here one by one.

[0135] The computer device 60 can be a computing device such as a desktop computer, a notebook, a handheld computer, and a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art can understand that Figure 2 merely examples of the computer device 60, which do not constitute a limitation on the computer device 60, may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0136] The so-called processor 61 may be a central processing unit (CPU), or may also be other general-purpose processors, a graphics processing unit (GPU), a tensor processing unit (TPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0137] The memory 62 may be an internal storage unit of the computer device 60, such as the hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk equipped on the computer device 60, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0138] Furthermore, the memory 62 may also include both an internal storage unit and an external storage device of the computer device 60. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0139] Please refer to Figure 3 , the terminal device is the electronic device 600, and the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0140] Among them, the storage unit stores program code, which can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present invention described in the method section of this specification above. For example, the processing unit 610 can execute steps as shown in Figure 1 .

[0141] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

[0142] The storage unit 620 may also include a program / utilities 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0143] The bus 630 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area bus using any of the multiple bus structures.

[0144] The electronic device 600 can also communicate with one or more external devices 700 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (such as a router, a modem). Such communication can be carried out through the input / output interface 650. And, the electronic device 600 can also communicate with one or more networks (such as a local area network, a wide area network, and / or a public network, such as the Internet) through the network adapter 660. The network adapter 660 can communicate with other modules of the electronic device 600 through the bus 630. It should be understood that although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0145] Example 4

[0146] The present invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. It can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or component. The computer-readable storage medium provides storage space, and this storage space stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by a processor are stored in this storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples of the computer-readable storage medium here include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0147] The computer-readable storage medium also includes data signals propagated in a baseband or as part of a carrier wave, which carry the readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, device, or component. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, radio frequency, etc., or any suitable combination of the above.

[0148] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages. The programming languages include object-oriented programming languages - such as Java, C++, etc., and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network or a wide area network, or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0149] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for pre-evaluating the rationality of a scheduling plan based on data driving in the above embodiments; the one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps:

[0150] Score the scheduling plan according to the scheduling plan evaluation index, analyze the correlation through the scoring result, where the correlation refers to the correlation between the rationality of the scheduling plan and the source-load prediction error, and obtain the influencing factor with the highest correlation; construct a pre-evaluation model for the rationality of the scheduling plan based on the obtained influencing factor, and predict the rationality of the scheduling plan according to the pre-evaluation model for the rationality of the scheduling plan.

[0151] The databases involved in the embodiments provided in the present application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without limitation.

[0152] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components described and shown in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0153] Experiment and Simulation

[0154] 1) Dataset:

[0155] Data of a provincial power grid in 2022, covering 365 days, including:

[0156] Power generation plan data: A total of 200 thermal power / hydroelectric power / new energy units.

[0157] Prediction data: Mean value of system load prediction error rate = 4.3%, wind power error rate = 18.7%.

[0158] 2) Model Performance

[0159] Test set results (73 days):

[0160] Model: AdaBoost; Accuracy: 89.2%; F1-score (excellent): 0.86; AUC: 0.91;

[0161] Model: Random Forest; Accuracy: 83.5%; F1-score (excellent): 0.78; AUC: 0.84;

[0162] Model: SVM; Accuracy: 76.1%; F1-score (excellent): 0.72; AUC: 0.79;

[0163] 3) Case Analysis

[0164] On July 15, 2022, it was predicted as "unqualified". After actual scheduling, line overload occurred (margin = 0.58). The main reason was that the wind power prediction error rate reached 25%, resulting in insufficient reserve.

[0165] The model gave an early warning. After adjusting the plan, the margin was increased to 0.72, and the rating was changed to "qualified".

[0166] In summary, for the data-driven pre-evaluation method and system for the rationality of scheduling plans of the present invention, the day-ahead plan data and the real-time operation data of the power grid are matched and replaced according to the equipment ID, a simulation environment is constructed to realize the preview of the plan execution effect and improve the forward-looking of the evaluation; by pairwise comparison of experts, a judgment matrix is constructed to quantify the relative importance of 18 indicators such as safety, economy, and environmental protection, solve the problem of multi-objective weight distribution, and balance subjective and objective requirements; calculate the mutual information value between the source-load prediction error and the scheduling rationality, identify highly correlated variables such as the load error during the daily power supply guarantee period and the wind power prediction error, and reduce the model complexity; combine LSTM prediction and distributionally robust optimization to generate the safety boundary under the worst-case scenario, realize 24-hour rolling risk prediction, and improve the prediction accuracy by 15%-20% compared with traditional methods.

[0167] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above functions can be assigned to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of the functional units and modules are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0168] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0169] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present invention can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in the form of hardware or software depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0170] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal and method can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0171] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0172] In addition, the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0173] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned method embodiments of the present invention can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0174] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses, and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0175] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the process in Figure 1One or more processes and / or boxes Figure 1 The functions specified in one or more boxes.

[0176] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 One or more processes and / or boxes Figure 1 or more processes and / or boxes.

[0177] The above content is only to illustrate the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the claims of the present invention.

Claims

1. A method for pre-evaluating the rationality of a scheduling plan based on data-driven, characterized in that Including the following steps: Evaluating the scheduling plan based on the scheduling plan evaluation indicators, analyzing the correlation through the scoring results, where the correlation refers to the correlation between the rationality of the scheduling plan and the source-load prediction error, and obtaining the influencing factor with the highest correlation; Constructing a pre-evaluation model for the rationality of the scheduling plan based on the influencing factor with the highest correlation, and predicting the rationality of the scheduling plan according to the pre-evaluation model for the rationality of the scheduling plan.

2. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 1, wherein The scheduling plan evaluation indicators are specifically: Reading the day-ahead scheduling plan data, grid operation data, grid model data, short-term system load prediction data of the grid, short-term load prediction data of the bus, wind power generation prediction data, and photovoltaic power generation prediction data to calculate the scheduling plan evaluation indicators.

3. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 2, wherein The scheduling plan evaluation indicators include: Line overload safety margin, main transformer overload safety margin, bus voltage safety level, frequency safety margin, section safety margin, N-1 line overload safety margin, N-1 main transformer overload safety margin, N-1 bus voltage safety level, spinning reserve, negative reserve, proportion of clean energy generation, clean energy consumption, carbon dioxide emissions, sulfur dioxide emissions, line load rate, transformer load rate, average power purchase price, and network loss.

4. The method for pre-evaluating the rationality of a scheduling plan based on data-driven according to claim 2, wherein Replacing the generator measurement data in the grid operation data with the day-ahead generation plan data of the day before the operation data date, and replacing the tie-line measurement data with the day-ahead tie-line plan data of the day before the operation data date, replacing according to the equipment ID, reading the replaced operation data, and calculating the indicators with reference to the index definition, and obtaining a value for each indicator every day.

5. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 1, characterized in that Using the analytic hierarchy process to calculate the index weights and evaluating the scheduling plan based on the scheduling plan evaluation indicators, specifically: Evaluating the importance degree of the indicators, making pairwise comparisons to form a judgment matrix; calculating the maximum eigenvalue and eigenvector, and performing a consistency test; after passing the consistency test, calculating the weights of each vector, and discretizing the scores to obtain the evaluation level R.

6. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 5, wherein The evaluation level R is: where S is the score of the generation plan.

7. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 1, characterized in that, The source-load prediction errors include bus load prediction error, system load prediction error rate, wind power generation prediction error, and photovoltaic power generation prediction error; Using the mutual information method to calculate the mutual information values between the set of influencing factors {bus load prediction error rate during the key period of daily power supply guarantee, minimum bus load prediction error rate during the midday low valley, minimum bus load prediction error rate during the night low valley, system load prediction error rate during the key period of daily power supply guarantee, minimum system load prediction error rate during the midday low valley, minimum system load prediction error rate during the night low valley, daily wind power generation prediction error rate, daily photovoltaic power generation prediction error rate} and the rationality of the scheduling plan, setting a threshold, and screening out the influencing factor with the highest correlation.

8. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 7, characterized in that, The bus load prediction error includes the bus load prediction error rate during the key period of daily power supply guarantee, the minimum bus load prediction error rate during the midday low valley, and the minimum bus load prediction error rate during the night low valley, and is calculated using the bus load data and the bus load prediction data. The system load prediction error includes the system load prediction error rate during the critical periods of daily power supply guarantee, the minimum system load prediction error rate during the midday low valley, and the minimum system load prediction error rate during the nighttime low valley, which is calculated using the system load data and the system load prediction data; The wind power generation prediction error is expressed by the daily wind power generation prediction error rate and is calculated using the wind power generation data and the wind power generation prediction data; The photovoltaic power generation prediction error is expressed by the daily photovoltaic power generation prediction error rate and is calculated using the photovoltaic power generation data and the photovoltaic power generation prediction data.

9. The method for pre-evaluating the rationality of a scheduling plan based on data driving according to claim 1, wherein Based on the obtained influencing factors with the highest correlation, a pre-evaluation model for the rationality of the dispatching plan is constructed as follows: A machine learning model is selected, and the nonlinear relationship between the power outage duration and each feature is learned according to the ensemble learning algorithm to obtain a pre-evaluation model for the rationality of the dispatching plan.

10. A data-driven pre-evaluation system for the rationality of scheduling plans, characterized in that, It includes: The factor module scores the dispatching plan according to the dispatching plan evaluation index, analyzes the correlation through the scoring results, where the correlation refers to the correlation between the rationality of the dispatching plan and the source-load prediction error, and obtains the influencing factor with the highest correlation; The evaluation module constructs a pre-evaluation model for the rationality of the dispatching plan based on the obtained influencing factors, and predicts the rationality of the dispatching plan according to the pre-evaluation model for the rationality of the dispatching plan.

11. The data-driven pre-evaluation system for the rationality of scheduling plans according to claim 10, wherein, In the factor module, the dispatching plan evaluation index is specifically: Read the day-ahead dispatching plan data, grid operation data, grid model data, grid short-term system load prediction data, bus short-term load prediction data, wind power generation prediction data, and photovoltaic power generation prediction data to calculate the dispatching plan evaluation index.

12. The data-driven pre-evaluation system for the rationality of scheduling plans according to claim 11, wherein, Replace the generator measurement data in the grid operation data with the day-ahead generation plan data of the day before the operation data date, and replace the tie-line measurement data with the day-ahead tie-line plan data of the day before the operation data date. Replace according to the equipment ID, read the replaced operation data, and calculate the index according to the index definition. Each index obtains a value every day; The dispatching plan evaluation index includes: Line overload safety margin, main transformer overload safety margin, bus voltage safety level, frequency safety margin, section safety margin, N-1 line overload safety margin, N-1 main transformer overload safety margin, N-1 bus voltage safety level, spinning reserve, negative reserve, clean energy generation ratio, clean energy consumption, carbon dioxide emissions, sulfur dioxide emissions, line load rate, transformer load rate, average power purchase price, and network loss.

13. The data-driven pre-evaluation system for the rationality of scheduling plans according to claim 10, characterized in that, In the factor module, evaluate the importance of the indicators, make pairwise comparisons to form a judgment matrix; calculate the maximum eigenvalue and eigenvector, and conduct a consistency test; after passing the consistency test, calculate the weights of each vector, and discretize the scores to obtain the evaluation level R as follows: Among them, S is the score of the generation plan.

14. The data-driven pre-evaluation system for the rationality of scheduling plans according to claim 10, wherein, In the factor module, the source-load prediction error includes bus load prediction error, system load prediction error rate, wind power generation prediction error, and photovoltaic power generation prediction error; Calculate the mutual information values between the set of influencing factors {prediction error rate of bus load during key daily power supply periods, minimum bus load prediction error rate during midday low valleys, minimum bus load prediction error rate during night low valleys, prediction error rate of system load during key daily power supply periods, minimum system load prediction error rate during midday low valleys, minimum system load prediction error rate during night low valleys, prediction error rate of daily wind power generation, prediction error rate of daily photovoltaic power generation} and the rationality of the dispatching plan, set a threshold, and screen out the influencing factors with high correlation; The bus load prediction error includes the prediction error rate of bus load during key daily power supply periods, the minimum bus load prediction error rate during midday low valleys, and the minimum bus load prediction error rate during night low valleys, which is calculated using bus load data and bus load prediction data; The system load prediction error includes the prediction error rate of system load during key daily power supply periods, the minimum system load prediction error rate during midday low valleys, and the minimum system load prediction error rate during night low valleys, which is calculated using system load data and system load prediction data; The wind power generation prediction error is represented by the prediction error rate of daily wind power generation, which is calculated using wind power generation data and wind power generation prediction data; The photovoltaic power generation prediction error is represented by the prediction error rate of daily photovoltaic power generation, which is calculated using photovoltaic power generation data and photovoltaic power generation prediction data.

15. The data-driven pre-evaluation system for the rationality of scheduling plans according to claim 10, characterized in that, In the evaluation module, a machine learning model is selected to learn the non-linear relationship between the power outage duration and each feature according to the ensemble learning algorithm, and a pre-evaluation model for the rationality of the dispatching plan is obtained.

16. A computer-readable storage medium storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the method according to any one of claims 1 to 9.

17. A computing device, characterized in that, Comprising: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for performing the method according to any one of claims 1 to 9.

Images