Transformer capacity optimization method and device, equipment and storage medium

Abstract

The invention discloses a transformer capacity optimization method and device, equipment and a storage medium, and the method comprises the steps: determining a heavy load function of a transformer of a power system under the guidance of a regulation strategy through obtaining a time sequence load prediction model of the power system based on the hysteresis effect of the regulation strategy, and carrying out the optimization of the capacity of the transformer according to a time sequence load prediction result and the heavy load function; the method comprises the steps of determining the heavy load penalty cost of a transformer, optimizing the capacity expansion capacity of the transformer with the purpose of minimizing the capacity expansion cost of the transformer of the power system under the constraint of the heavy load penalty cost, obtaining the optimal capacity expansion capacity of the transformer, and optimizing the transformer of the power system through the optimal capacity expansion capacity of the transformer. Therefore, load prediction is carried out through the time sequence load prediction model considering the regulation and control strategy hysteresis effect, load prediction overestimation is avoided, the prediction result more conforming to the actual regulation and control process is used for transformer capacity optimization, and therefore the precision and rationality of transformer capacity optimization are improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to a method, apparatus, device, and storage medium for optimizing transformer capacity. Background Technology

[0002] With the development of smart grids and the energy internet, regional load structures are becoming more diversified and dynamic, and the large-scale integration of electric vehicles and other technologies is posing challenges to load forecasting and grid capacity planning. Demand-side management and price guidance mechanisms have become key to regulation, and traditional load forecasting and capacity optimization technologies are struggling to keep up.

[0003] However, current load forecasting models do not consider the impact of the lag in control strategies on the time-series load distribution, resulting in an overestimation of control effectiveness and crude forecasting results, leading to insufficient transformer capacity planning.

[0004] How to integrate the lag factors of control strategies to optimize transformer capacity, so as to improve the accuracy and rationality of transformer capacity optimization, is an issue that needs attention. Summary of the Invention

[0005] In view of the above problems, this application provides a method, apparatus, equipment and storage medium for optimizing transformer capacity, so as to improve the accuracy and rationality of transformer capacity optimization.

[0006] To achieve the above objectives, the following specific solutions are proposed:

[0007] A method for optimizing transformer capacity includes:

[0008] A time-series load forecasting model based on the lag effect of control strategies is obtained for the power system, and the time-series load forecasting results are obtained by using the time-series load forecasting model;

[0009] Based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined;

[0010] Based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer;

[0011] Under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity;

[0012] The transformers in the power system are optimized by expanding the capacity of the optimal transformer.

[0013] Optionally, the acquisition of the time-series load forecasting model of the power system based on the lag effect of control strategies includes:

[0014] Acquire historical data on various types of loads in the power system;

[0015] Using the historical data, a monthly lag response model and a minute-level lag response model for the power system's control strategy are constructed.

[0016] By coupling the monthly lag response model and the minute lag response model, a time-series load prediction model is obtained.

[0017] Optionally, the monthly hysteresis response model is:

[0018]

[0019] in, For a set of preset control strategies Index of regulatory strategies This represents the total number of price guidance and control strategies. The time independent variable representing the ordinal number of the month is... For regulation strategy The theoretical maximum response rate, For regulation strategy The attenuation coefficient, For regulation strategy The effective month, For indicator functions, Indicates when The value is 1 when the time is right, and 0 otherwise. This refers to the monthly hysteresis response function;

[0020] The minute-level hysteresis response model is as follows:

[0021]

[0022] in, The time independent variable is in the minute range. To incentivize the timing of the implementation of regulatory policies, The lag factor for the users of the power system. The threshold for hysteresis strength. For users with high latency, the minimum response delay time For users with strong lag, the minimum response rate under control strategy k is given. For users with weak hysteresis, this represents the minimum response rate under control strategy k. For users with strong lag, the response rate coefficient under control strategy k is... For users with weak hysteresis, the response rate coefficient under control strategy k is... This refers to the minute-level hysteresis response function;

[0023] The coupling of the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model includes:

[0024] By coupling the monthly lag response model and the minute-level lag response model using the following formula, a time-series load forecasting model is obtained:

[0025]

[0026] in, In order to regulate strategies After it was issued, Year moon Load forecast values ​​for the time period The total amount of steerable load in the power system. For load growth rate, For regulation strategy The initial year of issuance, For regulation strategy The initial year of issuance The baseline load, In order to be in Year relative to the initial year The amount of load growth.

[0027] Optionally, based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined, including:

[0028] Using the following formula, based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined:

[0029]

[0030] in, For the overloaded function, The capacity of the transformer in the power system. , This refers to the capacity of the transformer before the expansion. The capacity expansion for the transformer, For the first A set of date indices for the month. For the first The collection of time periods of the day, This is the transformer overload threshold.

[0031] Optionally, based on the time-series load forecast results and the overload function, the overload penalty cost of the transformer is determined, including:

[0032] Using the following formula, based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer:

[0033]

[0034] in, The overload penalty cost, This represents the daily upper limit of transformer capacity. This is the monthly upper limit for transformer capacity. The ratio of the daily peak value of the transformer's capacity to the transformer's total capacity exceeds [a certain threshold]. The unit loss cost of part The ratio of the monthly peak capacity of the transformer to the total capacity of the transformer is given. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This refers to the unit loss cost of the transformer during a single overload. Under the control strategy k Year Monthly daily peak, Under the control strategy k Annual monthly peak.

[0035] Optionally, under the constraint of the overload penalty cost, with the objective of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity, including:

[0036] Under the constraint of the overload penalty cost, the expansion capacity of the transformer is optimized by using an optimization function that aims to minimize the transformer expansion cost of the power system, thus obtaining the optimal transformer expansion capacity. The optimization function is as follows:

[0037]

[0038] in, The unit capacity expansion cost of the transformer, For regulation strategy Execution costs This is the upper limit of the overload penalty cost. As the first weighting coefficient, This is the second weighting coefficient. This is the third weighting coefficient.

[0039] A transformer capacity optimization device, comprising:

[0040] The time-series load forecasting model acquisition unit is used to acquire the time-series load forecasting model of the power system based on the lag effect of the control strategy, and to predict the time-series load forecasting result through the time-series load forecasting model.

[0041] The reload function determination unit is used to determine the reload function of the transformer in the power system under the guidance of the control strategy based on the time-series load prediction results.

[0042] The overload penalty cost determination unit is used to determine the overload penalty cost of the transformer based on the time-series load prediction results and the overload function.

[0043] The capacity expansion optimization calculation unit is used to optimize the transformer expansion capacity under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, and obtain the optimal transformer expansion capacity.

[0044] An optimization unit is used to optimize the transformers of the power system by means of the optimal transformer expansion capacity.

[0045] Optionally, the time-series load forecasting model acquisition unit includes:

[0046] Historical data acquisition unit, used to acquire historical data of various loads in the power system;

[0047] The lag response model construction unit is used to construct a monthly lag response model and a minute-level lag response model of the power system's control strategy using the historical data.

[0048] The model coupling unit is used to couple the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model.

[0049] Optionally, the monthly hysteresis response model is:

[0050]

[0051] in, For a set of preset control strategies Index of regulatory strategies This represents the total number of price guidance and control strategies. The time independent variable representing the ordinal number of the month is... For regulation strategy The theoretical maximum response rate, For regulation strategy The attenuation coefficient, For regulation strategy The effective month, For indicator functions, Indicates when The value is 1 when the time is right, and 0 otherwise. This refers to the monthly hysteresis response function;

[0052] The minute-level hysteresis response model is as follows:

[0053]

[0054] in, The time independent variable is in the minute range. To incentivize the timing of the implementation of regulatory policies, The lag factor for the users of the power system. The threshold for hysteresis strength. For users with high latency, the minimum response delay time For users with strong lag, the minimum response rate under control strategy k is given. For users with weak hysteresis, this represents the minimum response rate under control strategy k. For users with strong lag, the response rate coefficient under control strategy k is... For users with weak hysteresis, the response rate coefficient under control strategy k is... This refers to the minute-level hysteresis response function;

[0055] The model coupling unit includes:

[0056] The model coupling subunit is used to couple the monthly lag response model and the minute lag response model using the following formula to obtain the time-series load forecasting model:

[0057]

[0058] in, In order to regulate strategies After it was issued, Year moon Load forecast values ​​for the time period The total amount of steerable load in the power system. For load growth rate, For regulation strategy The initial year of issuance, For regulation strategy The initial year of issuance The baseline load, In order to be in Year relative to the initial year The amount of load growth.

[0059] Optionally, the overload function determination unit includes:

[0060] The reload function determination subunit is used to determine the reload function of the transformer in the power system under the guidance of the control strategy, based on the time-series load forecast results, using the following formula:

[0061]

[0062] in, For the overloaded function, The capacity of the transformer in the power system. , This refers to the capacity of the transformer before the expansion. The capacity expansion for the transformer, For the first A set of date indices for the month. For the first The collection of time periods of the day, This is the transformer overload threshold.

[0063] Optionally, the overload penalty cost determination unit includes:

[0064] The overload penalty cost determination subunit is used to determine the overload penalty cost of the transformer using the following formula, based on the time-series load forecast results and the overload function:

[0065]

[0066] in, The overload penalty cost, This represents the daily upper limit of transformer capacity. This is the monthly upper limit for transformer capacity. The ratio of the daily peak value of the transformer's capacity to the transformer's total capacity exceeds [a certain threshold]. The unit loss cost of part The ratio of the monthly peak capacity of the transformer to the total capacity of the transformer is given. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This refers to the unit loss cost of the transformer during a single overload. Under the control strategy k Year Monthly daily peak, Under the control strategy k Annual monthly peak.

[0067] Optionally, the capacity expansion optimization calculation unit includes:

[0068] The capacity expansion optimization calculation subunit is used to optimize the transformer expansion capacity under the constraint of the overload penalty cost by using an optimization function aimed at minimizing the transformer expansion cost of the power system, to obtain the optimal transformer expansion capacity. The optimization function is as follows:

[0069]

[0070] in, The unit capacity expansion cost of the transformer, For regulation strategy Execution costs This is the upper limit of the overload penalty cost. As the first weighting coefficient, This is the second weighting coefficient. This is the third weighting coefficient.

[0071] A transformer capacity optimization device, comprising a memory and a processor;

[0072] The memory is used to store programs;

[0073] The processor is used to execute the program to implement the various steps of the transformer capacity optimization method described above.

[0074] A storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the transformer capacity optimization method as described above.

[0075] By employing the aforementioned technical solution, this application obtains a time-series load forecasting model for the power system based on the lag effect of the control strategy, and uses this model to predict the time-series load forecasting results. Based on these results, it determines the reload function of transformers under the guidance of the control strategy. According to the time-series load forecasting results and the reload function, it determines the reload penalty cost of the transformers. Under the constraint of the reload penalty cost, and with the goal of minimizing the transformer expansion cost of the power system, it optimizes the transformer expansion capacity to obtain the optimal transformer expansion capacity. The optimal transformer expansion capacity is then used to optimize the transformers of the power system. Therefore, by using a time-series load forecasting model that considers the lag effect of the control strategy for load forecasting, overestimation of load forecasts is avoided, allowing forecasts that better reflect the actual control process to be used for transformer capacity optimization, thereby improving the accuracy and rationality of transformer capacity optimization. Attached Figure Description

[0076] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0077] Figure 1 This is a schematic diagram of a process for optimizing transformer capacity provided in an embodiment of this application;

[0078] Figure 2A flowchart illustrating a method for obtaining a time-series load forecasting model, provided in an embodiment of this application;

[0079] Figure 3 A schematic diagram of a device structure for optimizing transformer capacity provided in an embodiment of this application;

[0080] Figure 4 This is a schematic diagram of a device for optimizing transformer capacity, provided as an embodiment of this application. Detailed Implementation

[0081] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0082] The proposed solution can be implemented based on a terminal with data processing capabilities, such as a computer, cloud, or server.

[0083] Next, combined Figure 1 The method for optimizing transformer capacity according to this application may include the following steps:

[0084] Step S110: Obtain the time-series load forecasting model of the power system based on the lag effect of the control strategy, and obtain the time-series load forecasting result through the time-series load forecasting model.

[0085] The time-series load forecasting model can be obtained by fusing a monthly-level lag response model and a minute-level lag response model.

[0086] Understandably, traditional models assume that control strategies take effect instantaneously, leading to an overestimation of load reduction, such as misjudging that real-time electricity prices can immediately alter industrial load. This step, however, avoids prediction bias by coupling a dual-scale model, resulting in more accurate predictions of daily peak-valley variations and annual load trends.

[0087] Specifically, the results of time-series load forecasting can be the load time-series data (such as daily peak values ​​or monthly peak values) output by the time-series load forecasting model, which directly reflects the actual load distribution under the lag effect of the control strategy. For example, considering minute-level lag, the risk of short-term overload caused by equipment start-up and shutdown delays can be accurately calculated, avoiding insufficient capacity planning based solely on historical peak values.

[0088] Step S120: Based on the time-series load forecast results, determine the overload function of the transformer in the power system under the guidance of the control strategy.

[0089] Specifically, the time-series load forecast results can be transformed into evaluation indicators of transformer operating status through the reload function, directly serving the subsequent calculation of reload penalty costs. For example, the monthly reload duration percentage of a transformer under control strategy k can be obtained by accumulating the reload function, thereby calculating the additional loss cost and providing a quantitative basis for capacity expansion decisions.

[0090] The overload function can transform the uncertainty of load forecasting (such as peak fluctuations caused by the lag in control strategies) into a penalty term in the optimization model, establishing a mathematical relationship between expansion costs and overload risks. For example, if the forecast shows that the probability of overload reaches 30% during a certain period, the overload function will increase the corresponding penalty cost, prompting the optimization model to choose a larger expansion capacity to reduce the risk.

[0091] Step S130: Determine the overload penalty cost of the transformer based on the time-series load prediction results and the overload function.

[0092] Understandably, the impact of different control strategies on the cost of heavy load penalties can be directly reflected in the cost of heavy load penalties. For example, if strategy one reduces the cost of heavy load penalties by 40%, while strategy two only reduces it by 15%, then strategy one can be prioritized to improve cost-effectiveness, providing data support for optimizing demand-side management strategies.

[0093] Step S140: Under the constraint of heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, optimize the transformer expansion capacity to obtain the optimal transformer expansion capacity.

[0094] Among them, the optimization function can integrate the direct cost of expansion, the cost of overload penalty, and the cost of implementing the control strategy into a unified objective. By flexibly adjusting the priority through weight coefficients, it can avoid the extreme tendency of traditional methods that favor one side.

[0095] Specifically, based on the time-series load forecast results and the heavy load penalty cost, the optimized model can accurately calculate the required expansion capacity. For example, if the forecast shows that the annual heavy load penalty cost of a certain transformer under a certain control strategy is 80,000 yuan, while the annual equivalent cost (investment amortization + operation and maintenance) of expanding capacity by 500kVA is 60,000 yuan, then the model selects to expand capacity by 500kVA, reducing the total annual cost by 20,000 yuan and realizing on-demand capacity expansion.

[0096] When load forecasts are updated due to policy adjustments, new energy source integration, and other factors, the overload penalty cost changes in real time, and the optimization model can dynamically adjust the expansion plan. For example, if a new electric vehicle charging station is added to a certain area, the peak load forecast increases by 20%. The model automatically calculates the new expansion capacity and assesses whether it needs to be combined with a demand response strategy, avoiding the capacity lag problem under static planning.

[0097] Step S150: Optimize the transformers in the power system by using the optimal transformer expansion capacity.

[0098] Specifically, the terminal can directly interface with the energy management system to send the optimal transformer expansion capacity directly to the energy management system for real-time capacity scheduling.

[0099] The transformer capacity optimization method provided in this embodiment obtains a time-series load forecasting model of the power system based on the lag effect of the control strategy, and predicts the time-series load using this model. Based on the time-series load forecasting results, the reload function of the transformers under the guidance of the control strategy is determined. According to the time-series load forecasting results and the reload function, the reload penalty cost of the transformers is determined. Under the constraint of the reload penalty cost, the transformer expansion capacity is optimized with the goal of minimizing the transformer expansion cost of the power system, resulting in the optimal transformer expansion capacity. The transformers of the power system are then optimized using the optimal transformer expansion capacity. Therefore, by using a time-series load forecasting model that considers the lag effect of the control strategy for load forecasting, overestimation of load forecasts is avoided, allowing forecasts that better reflect the actual control process to be used for transformer capacity optimization, thereby improving the accuracy and rationality of transformer capacity optimization.

[0100] In some embodiments of this application, the process of obtaining the time-series load forecasting model of the power system based on the lag effect of the control strategy, as mentioned in the above embodiments, is described, such as... Figure 2 As shown, the process may include:

[0101] Step S210: Obtain historical data of various loads in the power system.

[0102] The characteristics of the load components can be expressed as follows:

[0103]

[0104] In the above formula, Let i be the i-th type of load component, representing the physical properties of the load. The total number of load components, For ingredient index, The eigenvector contains the peak-to-valley difference of the daily load curve. Fluctuation standard deviation ,power Key parameters, such as those related to the dynamic behavior patterns of the load, are used to define these parameters. The component percentage reflects its weight in the total load and satisfies... .

[0105] Specifically, by defining the characteristics of load components, complex loads are decomposed into analytically independent components, providing a physical basis for subsequently constructing differentiated response models. For example, due to the high adjustability of charging time, the response hysteresis of electric vehicle loads is significantly lower than that of industrial loads in continuous production.

[0106] Step S220: Using historical data, construct a monthly-level lag response model and a minute-level lag response model for the power system's control strategy.

[0107] Understandably, price-incentive strategies (such as time-of-use pricing and charging subsidies) require a period of adjustment in user perception, and their effects gradually saturate over time. A monthly lag response model over a larger time scale can be:

[0108]

[0109] in, For a set of preset control strategies Index of regulatory strategies This represents the total number of price guidance and control strategies. Let the time independent variable represent the month ordinal number (1, 2, ..., 12). For regulation strategy The theoretical maximum response rate depends on the strength of the strategy and user acceptance. >0 indicates a regulatory strategy The attenuation coefficient, It reflects the speed at which the strategy effect spreads over time (the larger the value, the faster the response saturates). For regulation strategy The effective month, For indicator functions, Indicates when The value is 1 when the time is right, and 0 otherwise. This is a monthly lag response function. (Exponential function) This can describe the process by which users gradually improve their response rate from an initial low level to a steady state. For example, the rollout of electric vehicle charging time migration policies requires several months to permeate into user habits. A smaller value can characterize the inertia of this type of long-period behavior.

[0110] A minute-level lag response model at a smaller time scale can be:

[0111]

[0112] in, The time independent variable is in the minute range. To incentivize the timing of the implementation of regulatory policies, For the lag factor of the power system users, The threshold for hysteresis strength. The minimum response delay time for users with high latency can be the cooling / preheating time required for equipment start-up and shutdown. For users with strong lag, the minimum response rate under control strategy k is given. For users with weak hysteresis, this represents the minimum response rate under control strategy k. and The values ​​are all within [0,1]. For users with strong lag, the response rate coefficient under control strategy k is... For users with weak hysteresis, the response rate coefficient under control strategy k is... It is a minute-level lag response function.

[0113] Among them, the strongly lag function ( This indicates that due to production continuity constraints, the initial response rate is low and decays over time, making immediate shutdown impossible; weakly hysteretic functions ( This describes the user's ability to quickly adjust their behavior, such as an air conditioner changing its set temperature within minutes. Weak hysteresis describes... Respond immediately at all times, with response rate based on Rapid rise, strong lag in the delay period No response from inside. Post-response rate Rising slowly.

[0114] Step S230: Couple the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model.

[0115] Specifically, the monthly lag response model and the minute-level lag response model can be coupled using the following formula to obtain the time-series load forecasting model:

[0116]

[0117] in, In order to regulate strategies After it was issued, Year moon Load forecast values ​​for the time period The total amount of guideable load in the power system represents the control strategy. The adjustable load under ideal, lag-free scenarios can be obtained specifically from historical operational data. The load growth rate is used to characterize the rigid increase in load brought about by economic development. For regulation strategy The initial year of issuance, For regulation strategy The initial year of issuance The baseline load, In order to be in Year relative to the initial year The amount of load growth.

[0118] Understandably, time-series load forecasting models incorporate feedback lags at both the monthly and minute scales. and or At that time, the maximum load reduction approaches the theoretical maximum reduction value.

[0119] The transformer capacity optimization method provided in this application constructs a coupled monthly and minute-level response model to address the multi-scale lag effect of price guidance mechanisms. At the monthly scale, it characterizes the gradual process of policy penetration and quantifies the delay in user habit migration. At the minute scale, based on production continuity constraints, it distinguishes between strong and weak lag users, introduces a delay time parameter, and establishes a piecewise response function to accurately describe the time-varying characteristics of load migration under real-time incentives. Coupled with natural growth and load reduction effects, a multi-scale load forecasting model is formed. By adjusting the attenuation coefficient and response rate, it dynamically reflects the policy implementation cycle and user behavioral inertia, making the time-series load forecasting model more consistent with actual expectations and avoiding overestimation of load predicted by the model.

[0120] In some embodiments of this application, the process of determining the reload function of the transformer under the guidance of the control strategy based on the time-series load forecast results in step S120 is described. This process may include:

[0121] Using the following formula, based on the time-series load forecast results, the reload function of transformers in the power system under the guidance of control strategies is determined:

[0122]

[0123] in, For overloaded functions, For the capacity of transformers in the power system, , This refers to the transformer's capacity before expansion. The capacity expansion for the transformer, For the first A set of date indices for the month. For the first The collection of time periods of the day, This is the transformer overload threshold.

[0124] Considering that daily and monthly peak loads of transformers exceeding the prescribed upper limit, and overload conditions exceeding the allowable range, will cause significant losses to the transformers, it is necessary to construct a transformer overload penalty mechanism. Based on this, the process mentioned in the previous embodiment for determining the transformer overload penalty cost based on time-series load forecast results and the overload function is described. This process may include:

[0125] The overload penalty cost of the transformer is determined using the following formula, based on the time-series load forecast results and the overload function:

[0126]

[0127] in, To reduce the cost of heavy load penalties, This represents the daily upper limit of transformer capacity. This is the monthly upper limit for transformer capacity. This is the ratio of the daily peak value of the transformer capacity to the transformer's total capacity. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This is the ratio of the monthly peak value of the transformer capacity to the total transformer capacity. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This refers to the unit loss cost of a transformer during a single overload. Under the control strategy k Year Monthly daily peak, Under the control strategy k Annual monthly peak.

[0128] Understandably, by designing an overload penalty mechanism, the number of transformer overloads at each time scale is calculated and limited to within the equipment tolerance threshold, thus avoiding cumulative overload caused by control lag.

[0129] Based on this, the process mentioned in the foregoing embodiments, which optimizes the transformer expansion capacity under the constraint of heavy load penalty cost and aims to minimize the transformer expansion cost of the power system, to obtain the optimal transformer expansion capacity, is described. This process may include:

[0130] Under the constraint of heavy load penalty cost, the transformer expansion capacity is optimized by using an optimization function that aims to minimize the transformer expansion cost of the power system, and the optimal transformer expansion capacity is obtained.

[0131] The optimization function can be:

[0132]

[0133] in, The unit capacity expansion cost of transformers, For regulation strategy Execution costs This is the upper limit of the overload penalty cost. As the first weighting coefficient, This is the second weighting coefficient. The third weighting coefficient represents the decision-maker's preference for the degree of avoidance of cost and overload risk.

[0134] Understandably, the optimization function guides the combination of strategies by jointly deciding on transformer capacity expansion, and dynamically balances economy and safety through weighting coefficients. By overcoupling the lag parameters in the prediction model, the actual load reduction effect of the control strategy is quantified, avoiding capacity shortages caused by idealized assumptions, and achieving coordinated optimization of power grid investment and operational safety.

[0135] The apparatus for optimizing transformer capacity provided in the embodiments of this application will be described below. The apparatus for optimizing transformer capacity described below can be referred to in correspondence with the method for optimizing transformer capacity described above.

[0136] See Figure 3 , Figure 3 This is a schematic diagram of a device structure for optimizing transformer capacity disclosed in an embodiment of this application.

[0137] like Figure 3 As shown, the device may include:

[0138] The time-series load forecasting model acquisition unit 11 is used to acquire the time-series load forecasting model of the power system based on the lag effect of the control strategy, and to predict the time-series load forecasting result through the time-series load forecasting model.

[0139] The reload function determination unit 12 is used to determine the reload function of the transformer of the power system under the guidance of the control strategy based on the time-series load prediction results.

[0140] The overload penalty cost determination unit 13 is used to determine the overload penalty cost of the transformer based on the time-series load prediction results and the overload function.

[0141] The capacity expansion optimization calculation unit 14 is used to optimize the capacity expansion of the transformer under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, and obtain the optimal transformer expansion capacity.

[0142] The optimization unit 15 is used to optimize the transformers of the power system by means of the optimal transformer expansion capacity.

[0143] Optionally, the time-series load forecasting model acquisition unit includes:

[0144] Historical data acquisition unit, used to acquire historical data of various loads in the power system;

[0145] The lag response model construction unit is used to construct a monthly lag response model and a minute-level lag response model of the power system's control strategy using the historical data.

[0146] The model coupling unit is used to couple the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model.

[0147] Optionally, the monthly hysteresis response model is:

[0148]

[0149] in, For a set of preset control strategies Index of regulatory strategies This represents the total number of price guidance and control strategies. The time independent variable representing the ordinal number of the month is... For regulation strategy The theoretical maximum response rate, For regulation strategy The attenuation coefficient, For regulation strategy The effective month, For indicator functions, Indicates when The value is 1 when the time is right, and 0 otherwise. This refers to the monthly hysteresis response function;

[0150] The minute-level hysteresis response model is as follows:

[0151]

[0152] in, The time independent variable is in the minute range. To incentivize the timing of the implementation of regulatory policies, The lag factor for the users of the power system. The threshold for hysteresis strength. For users with high latency, the minimum response delay time For users with strong lag, the minimum response rate under control strategy k is given. For users with weak hysteresis, this represents the minimum response rate under control strategy k. For users with strong lag, the response rate coefficient under control strategy k is... For users with weak hysteresis, the response rate coefficient under control strategy k is... This refers to the minute-level hysteresis response function;

[0153] The model coupling unit includes:

[0154] The model coupling subunit is used to couple the monthly lag response model and the minute lag response model using the following formula to obtain the time-series load forecasting model:

[0155]

[0156] in, In order to regulate strategies After it was issued, Year moon Load forecast values ​​for the time period The total amount of steerable load in the power system. For load growth rate, For regulation strategy The initial year of issuance, For regulation strategy The initial year of issuance The baseline load, In order to be in Year relative to the initial year The amount of load growth.

[0157] Optionally, the overload function determination unit includes:

[0158] The reload function determination subunit is used to determine the reload function of the transformer in the power system under the guidance of the control strategy, based on the time-series load forecast results, using the following formula:

[0159]

[0160] in, For the overloaded function, The capacity of the transformer in the power system. , This refers to the capacity of the transformer before the expansion. The capacity expansion for the transformer, For the first A set of date indices for the month. For the first The collection of time periods of the day, This is the transformer overload threshold.

[0161] Optionally, the overload penalty cost determination unit includes:

[0162] The overload penalty cost determination subunit is used to determine the overload penalty cost of the transformer using the following formula, based on the time-series load forecast results and the overload function:

[0163]

[0164] in, The overload penalty cost, This represents the daily upper limit of transformer capacity. This is the monthly upper limit for transformer capacity. The ratio of the daily peak value of the transformer's capacity to the transformer's total capacity exceeds [a certain threshold]. The unit loss cost of part The ratio of the monthly peak capacity of the transformer to the total capacity of the transformer is given. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This refers to the unit loss cost of the transformer during a single overload. Under the control strategy k Year Monthly daily peak, Under the control strategy k Annual monthly peak.

[0165] Optionally, the capacity expansion optimization calculation unit includes:

[0166] The capacity expansion optimization calculation subunit is used to optimize the transformer expansion capacity under the constraint of the overload penalty cost by using an optimization function aimed at minimizing the transformer expansion cost of the power system, to obtain the optimal transformer expansion capacity. The optimization function is as follows:

[0167]

[0168] in, The unit capacity expansion cost of the transformer, For regulation strategy Execution costs This is the upper limit of the overload penalty cost. As the first weighting coefficient, This is the second weighting coefficient. This is the third weighting coefficient.

[0169] The transformer capacity optimization device provided in this application embodiment can be applied to equipment for optimizing transformer capacity, such as terminals like mobile phones and computers. Optionally, Figure 4 The hardware structure block diagram of the device for optimizing transformer capacity is shown, with reference to... Figure 4The hardware structure of the equipment for optimizing transformer capacity may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4.

[0170] In this embodiment of the application, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4;

[0171] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0172] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0173] The memory stores a program, which the processor can call. The program is used for:

[0174] A time-series load forecasting model based on the lag effect of control strategies is obtained for the power system, and the time-series load forecasting results are obtained by using the time-series load forecasting model;

[0175] Based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined;

[0176] Based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer;

[0177] Under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity;

[0178] The transformers in the power system are optimized by expanding the capacity of the optimal transformer.

[0179] Optionally, the refined and extended functions of the program can be found in the description above.

[0180] This application embodiment also provides a storage medium that can store a program suitable for execution by a processor, the program being used for:

[0181] A time-series load forecasting model based on the lag effect of control strategies is obtained for the power system, and the time-series load forecasting results are obtained by using the time-series load forecasting model;

[0182] Based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined;

[0183] Based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer;

[0184] Under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity;

[0185] The transformers in the power system are optimized by expanding the capacity of the optimal transformer.

[0186] Optionally, the refined and extended functions of the program can be found in the description above.

[0187] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0188] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0189] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing transformer capacity, characterized in that, include: A time-series load forecasting model based on the lag effect of control strategies is obtained for the power system, and the time-series load forecasting results are obtained by using the time-series load forecasting model; Based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined; Based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer; Under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity; The transformers in the power system are optimized by expanding the capacity of the optimal transformer.

2. The method according to claim 1, characterized in that, The time-series load forecasting model for the power system based on the lag effect of control strategies includes: Acquire historical data on various types of loads in the power system; Using the historical data, a monthly lag response model and a minute-level lag response model for the power system's control strategy are constructed. By coupling the monthly lag response model and the minute lag response model, a time-series load prediction model is obtained.

3. The method according to claim 2, characterized in that, The monthly hysteresis response model is as follows: in, For a set of preset control strategies Index of regulatory strategies This represents the total number of price guidance and control strategies. The time independent variable representing the ordinal number of the month is... For regulation strategy The theoretical maximum response rate, For regulation strategy The attenuation coefficient, For regulation strategy The effective month, For indicator functions, Indicates when The value is 1 when the time is right, and 0 otherwise. This refers to the monthly hysteresis response function; The minute-level hysteresis response model is as follows: in, The time independent variable is in the minute range. To incentivize the timing of the implementation of regulatory policies, The lag factor for the users of the power system. The threshold for hysteresis strength. For users with high latency, the minimum response delay time For users with strong lag, the minimum response rate under control strategy k is given. For users with weak hysteresis, this represents the minimum response rate under control strategy k. For users with strong lag, the response rate coefficient under control strategy k is... For users with weak hysteresis, the response rate coefficient under control strategy k is... This refers to the minute-level hysteresis response function; The coupling of the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model includes: By coupling the monthly lag response model and the minute-level lag response model using the following formula, a time-series load forecasting model is obtained: in, In order to regulate strategies After it was issued, Year moon Load forecast values ​​for the time period The total amount of steerable load in the power system. For load growth rate, For regulation strategy The initial year of issuance, For regulation strategy The initial year of issuance The baseline load, In order to be in Year relative to the initial year The amount of load growth.

4. The method according to claim 3, characterized in that, Based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined, including: Using the following formula, based on the time-series load forecast results, the reload function of the transformers in the power system under the guidance of the control strategy is determined: in, For the overloaded function, The capacity of the transformer in the power system. , This refers to the capacity of the transformer before the expansion. The capacity expansion for the transformer, For the first A set of date indices for the month. For the first The collection of time periods of the day, This is the transformer overload threshold.

5. The method according to claim 4, characterized in that, Based on the time-series load forecast results and the overload function, the overload penalty cost of the transformer is determined, including: Using the following formula, based on the time-series load forecast results and the overload function, determine the overload penalty cost of the transformer: in, The overload penalty cost, This represents the daily upper limit of transformer capacity. This is the monthly upper limit for transformer capacity. The ratio of the daily peak value of the transformer's capacity to the transformer's total capacity exceeds [a certain threshold]. The unit loss cost of part The ratio of the monthly peak capacity of the transformer to the total capacity of the transformer is given. (The last part, "exceeding," appears to be a typo and can be omitted.) The unit loss cost of part This refers to the unit loss cost of the transformer during a single overload. Under the control strategy k Year Monthly daily peak, Under the control strategy k Annual monthly peak.

6. The method according to claim 5, characterized in that, Under the constraint of the overload penalty cost, with the objective of minimizing the transformer expansion cost of the power system, the expansion capacity of the transformer is optimized to obtain the optimal transformer expansion capacity, including: Under the constraint of the overload penalty cost, the expansion capacity of the transformer is optimized by using an optimization function that aims to minimize the transformer expansion cost of the power system, thus obtaining the optimal transformer expansion capacity. The optimization function is as follows: in, The unit capacity expansion cost of the transformer, For regulation strategy Execution costs This is the upper limit of the overload penalty cost. As the first weighting coefficient, This is the second weighting coefficient. This is the third weighting coefficient.

7. A device for optimizing transformer capacity, characterized in that, include: The time-series load forecasting model acquisition unit is used to acquire the time-series load forecasting model of the power system based on the lag effect of the control strategy, and to predict the time-series load forecasting result through the time-series load forecasting model. The reload function determination unit is used to determine the reload function of the transformer in the power system under the guidance of the control strategy based on the time-series load prediction results. The overload penalty cost determination unit is used to determine the overload penalty cost of the transformer based on the time-series load prediction results and the overload function. The capacity expansion optimization calculation unit is used to optimize the transformer expansion capacity under the constraint of the heavy load penalty cost, with the goal of minimizing the transformer expansion cost of the power system, and obtain the optimal transformer expansion capacity. An optimization unit is used to optimize the transformers of the power system by means of the optimal transformer expansion capacity.

8. The apparatus according to claim 7, characterized in that, The time-series load forecasting model acquisition unit includes: Historical data acquisition unit, used to acquire historical data of various loads in the power system; The lag response model construction unit is used to construct a monthly lag response model and a minute-level lag response model of the power system's control strategy using the historical data. The model coupling unit is used to couple the monthly lag response model and the minute lag response model to obtain the time-series load forecasting model.

9. A transformer capacity optimization device, characterized in that, Including memory and processor; The memory is used to store programs; The processor is configured to execute the program to implement the steps of the transformer capacity optimization method as described in any one of claims 1-6.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the transformer capacity optimization method as described in any one of claims 1-6.

Images