Quota management optimization method and system for low-voltage power protection project in distribution network

Abstract

The invention discloses a quota management optimization method and system for medium and low voltage power protection projects in a distribution network, and relates to the technical field of quota management of medium and low voltage distribution networks, and the method comprises the following steps: calculating an initial quota base price of a target project sample according to a reference quota; based on the initial quota base price, the target project scene feature vector and historical project data, generating a corrected quota price through scene dynamic parameter learning; based on the target project scene feature vector and historical project data, generating a deviation estimation value through a scene deviation estimation model; determining a deviation compensation quota price according to the corrected quota price and the deviation estimation value; and taking the initial quota base price as a reference, carrying out weighted fusion on the increment between the enterprise quota price and the deviation compensation quota price according to confidence, and generating a final optimized quota price. The method is used for solving the problems that the difference between the quota price and the actual settlement price is too large in the quota management of the low-voltage power protection project, and the adaptability of the traditional quota to the complex construction scene is insufficient.

Description

Technical Field

[0001] This invention relates to the field of quota management technology for medium and low voltage distribution networks, and more specifically, to an optimization method and system for quota management of medium and low voltage power supply protection projects in distribution networks. Background Technology

[0002] As a crucial component of the power system, the power distribution network directly impacts the reliability and quality of power supply for end users. Medium and low-voltage distribution network projects are extensive and widespread, encompassing both routine new construction and renovation projects as well as emergency repairs and power supply protection tasks in unforeseen circumstances. Power supply protection projects often require the installation and commissioning of distribution lines, switchgear, ring main units, and prefabricated substations within a limited timeframe to ensure continuous power supply to key areas and critical users. Quotas, serving as a basis for measuring the consumption of labor, materials, and machinery, provide a unified reference for project budgeting, bidding, and final settlement. Traditional engineering quota and pricing systems are primarily based on basic quota texts issued by relevant authorities or internal enterprise quotas. These basic texts provide a relatively uniform pricing benchmark under typical construction scenarios.

[0003] However, the traditional quota system is mainly designed for routine construction scenarios, and has the following problems in emergency power supply projects: First, the response speed requirement is high, and the routine quota fails to fully consider the resource allocation and construction organization characteristics under emergency conditions, resulting in large budget deviations; Second, the adaptability to scenarios is poor. Medium and low voltage distribution network power supply projects often involve special scenarios such as live-line work, night construction, and operation in confined spaces, and the traditional quota is difficult to accurately reflect actual consumption; Third, there is insufficient sample of rare scenarios. Power supply is an event-driven and highly unpredictable type of project, and there are few historical project samples for some high-risk or extreme scenarios. Directly applying data-driven methods can easily lead to high variance.

[0004] To address the above problems, this invention proposes a solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method and system for optimizing quota management of low-voltage power supply protection projects in distribution networks. By learning dynamic parameters of scenarios, combining scenario deviation estimation models, and deviation compensation based on sample quality, the method solves the problems of excessively large discrepancies between quota prices and actual settlement prices in quota management of low-voltage power supply protection projects in distribution networks, and the insufficient adaptability of traditional quotas to complex construction scenarios.

[0006] To achieve the above objectives, the present invention provides the following technical solution: An optimization method for quota management of low-voltage power supply protection projects in distribution networks includes the following steps: calculating the initial quota base price of the target project sample based on reference quotas; generating a corrected quota price through scenario dynamic parameter learning based on the initial quota base price, the target project scenario feature vector, and historical project data; generating a deviation estimate through a scenario deviation estimation model based on the target project scenario feature vector and historical project data; determining the deviation compensation quota price based on the corrected quota price and the deviation estimate; and generating the final optimized quota price by weighting the increment between the enterprise quota price and the deviation compensation quota price according to the confidence level, using the initial quota base price as a benchmark.

[0007] In a preferred embodiment, the step of calculating the initial quota base price of the target project sample based on the reference quota specifically involves: obtaining the quota consumption of the target project sample sub-items based on the reference quota; and calculating the initial quota base price based on the quota consumption and the market unit price of the corresponding sub-items.

[0008] In a preferred embodiment, the scenario dynamic parameter learning includes establishing a parameterized function, parameter fitting, causal logic constraints, and constraint penalties; the historical project data includes historical scenario feature vectors, historical project initial quota base prices, and historical project actual settlement prices.

[0009] In a preferred embodiment, the step of generating a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data through scenario dynamic parameter learning specifically involves: obtaining the base price difference of historical project samples, where the base price difference is the difference between the actual settlement price of historical projects and the initial quota base price of historical projects; establishing a set of parameterized functions based on the target project scenario feature vector; using the base price difference of historical project samples as the optimization objective, calculating the basic fitting objective by minimizing a loss function with a regularization term based on the set of parameterized functions; introducing causal logic constraints into the basic fitting objective, constructing a comprehensive objective function through constraint penalty terms, and solving it to obtain the optimal parameters and weights; and generating the corrected quota price based on the initial quota base price of the target project using the optimal parameters and weights.

[0010] In a preferred embodiment, causal logic constraints are introduced into the basic fitting objective, and a comprehensive objective function is constructed by correcting parameters through constraint penalty terms. Specifically, causal logic constraints are introduced to correct the fitting parameters, including nonnegativity constraints, monotonicity constraints, boundary constraints, and interaction logic constraints. When the fitting parameters violate the causal logic constraints, the degree of constraint violation is quantified through constraint penalty to obtain a constraint violation quantification index. A comprehensive objective function is constructed based on the basic fitting objective and the constraint violation quantification index.

[0011] In a preferred embodiment, the scene bias estimation model includes the calculation of scene rarity scores, hierarchical sampling of rare scenes, and the calculation of bias estimates based on Bayesian contraction.

[0012] In a preferred embodiment, the step of generating a deviation estimate based on the target project scene feature vector and historical project data through a scene deviation estimation model specifically involves: obtaining the scene rarity score of each sample by comprehensively calculating the inverse frequency of discrete features and the distribution of continuous features based on the historical scene feature vector and the target project scene feature vector; stratifying the historical project samples according to the scene rarity score, and filtering out historical project samples similar to the target project sample within the same stratum by similarity calculation to form a similar sample set; calculating the weighted mean of the base price difference and the effective sample size in the filtered historical project samples using similarity as the weight; and obtaining the deviation estimate by Bayesian contraction based on the weighted mean and the effective sample size.

[0013] In a preferred embodiment, determining the deviation compensation quota price based on the corrected quota price and the deviation estimate specifically involves: when the effective sample size reaches a preset threshold, calculating the deviation compensation quota price for the target project sample using the deviation estimate as the compensation amount.

[0014] In a preferred embodiment, the step of using the initial quota base price as a benchmark and weighting and fusing the increments between the enterprise quota price and the deviation compensation quota price according to confidence level to generate the final optimized quota price specifically involves: using the initial quota base price as a unified benchmark, calculating the first increment and the second increment of the deviation compensation quota price and the enterprise quota price relative to the benchmark respectively; determining the first confidence level and the second confidence level corresponding to the deviation compensation quota price and the enterprise quota price respectively based on a preset confidence level assessment rule; calculating the fusion weight based on the first confidence level and the second confidence level; calculating the comprehensive increment correction value based on the first increment, the second increment, and the fusion weight; and adding the initial quota base price and the comprehensive increment correction value to generate the final optimized quota price.

[0015] A system for optimizing quota management in low-voltage power supply protection projects in distribution networks includes: a base price calculation module for calculating the initial quota base price of a target project sample based on reference quotas; a learning correction module for generating a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data through scenario dynamic parameter learning; a deviation estimation module for generating a deviation estimate value based on the target project scenario feature vector and historical project data through a scenario deviation estimation model; a compensation price generation module for determining the deviation compensation quota price based on the corrected quota price and the deviation estimate value; and a quota price optimization module for generating the final optimized quota price by weighting and fusing the increments between the enterprise quota price and the deviation compensation quota price according to confidence level, using the initial quota base price as a benchmark.

[0016] The technical effects and advantages of the optimization method and system for quota management of low-voltage power supply protection projects in distribution networks according to this invention are as follows: This invention, through dynamic parameter learning and correction for different construction scenarios, reflects the actual consumption characteristics of various construction scenarios, thereby reducing the systematic deviation between the applied quota and the actual settlement. Simultaneously, through a scenario deviation estimation model, it estimates and corrects deviations for scenarios with scarce or abnormal samples, ensuring that reasonable quota adjustments can be generated even with insufficient historical data. This effectively solves the problems of instability and large deviations in traditional experience-based methods in rare scenarios. Using the reference quota base price as an anchor point, it comprehensively considers the enterprise quota price and the deviation compensation quota price based on learning compensation, achieving fusion through a confidence-weighted mechanism. This ensures that the final optimized quota price, while adhering to the reference standard, reflects the differences between the enterprise's internal management practices and actual engineering projects, resulting in more reasonable and reliable results. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a method for optimizing quota management of low-voltage power supply protection projects in distribution networks, provided as an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of the system structure of an optimization method for quota management of low-voltage power supply protection projects in distribution networks, provided in an embodiment of the present invention.

[0019] Figure 3 The present invention provides a logic diagram for an optimization method of quota management for low-voltage power supply protection projects in distribution networks. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, Figure 1 This invention presents an optimization method for quota management of low-voltage power supply protection projects in distribution networks, comprising the following steps: S1, Calculate the initial quota base price of the target project sample based on the reference quota; S2, based on the initial fixed quota base price, the target project scenario feature vector and historical project data, generates a corrected fixed quota price through scenario dynamic parameter learning; S3, based on the target project scene feature vector and historical project data, generates deviation estimates through a scene deviation estimation model; S4. Determine the deviation compensation quota price based on the corrected quota price and the deviation estimate; S5 uses the initial quota base price as a benchmark, and weights and merges the increments between the enterprise quota price and the deviation compensation quota price according to the confidence level to generate the final optimized quota price.

[0022] This embodiment, through dynamic parameter learning and correction for different construction scenarios, reflects the actual consumption characteristics of various construction scenarios, thereby reducing the systematic deviation between the applied quota and the actual settlement. Simultaneously, through a scenario deviation estimation model, it estimates and corrects deviations for scenarios with scarce or abnormal samples, ensuring that reasonable quota adjustments can be generated even with insufficient historical data. This effectively solves the problem of instability and large deviations in traditional experience-based methods for rare scenarios. Using the reference quota base price as an anchor, it comprehensively considers the enterprise quota price and the deviation compensation quota price based on learning compensation, achieving fusion through a confidence-weighted mechanism. This ensures that the final optimized quota price, while adhering to the reference standard, reflects the differences between the enterprise's internal management practices and actual engineering projects, resulting in a more reasonable and reliable outcome.

[0023] S1, calculate the initial quota base price of the target project sample based on the reference quota.

[0024] In this embodiment, the calculation of the initial quota base price of the target project sample based on the reference quota specifically involves: Based on the reference quota, obtain the quota consumption of the sample sub-items of the target project; The initial quota base price is calculated based on the quota consumption and the market unit price of the corresponding sub-item.

[0025] In this embodiment, the reference quota standard required to obtain the target project sample is obtained, including the project code and the material consumption, man-days and machine shifts required for each project sub-item, and the market unit price corresponding to each project sub-item is obtained, including the material unit price, labor unit price and machine shift unit price. The initial quota base price is calculated based on the quota consumption and the market unit price.

[0026] The specific formula for calculating the initial fixed base price is as follows:

[0027] In the formula, This is the initial fixed base price.

[0028] S2 generates a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data, through scenario dynamic parameter learning.

[0029] In this embodiment, the scenario dynamic parameter learning includes establishing parameterized functions, parameter fitting, causal logic constraints, and constraint penalties; the historical project data includes historical scenario feature vectors, historical project initial quota base prices, and historical project actual settlement prices.

[0030] In this embodiment, the step of generating a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data through scenario dynamic parameter learning specifically involves: Obtain the base price difference for historical project samples, whereby the base price difference is the difference between the actual settlement price of the historical project and the initial quota base price of the historical project; Based on the feature vector of the target project scenario, establish a set of parameterized functions; Using the historical project sample base price difference as the optimization objective, and based on a set of parameterized functions, the basic fitting objective is obtained by minimizing the loss function with a regularization term. A causal logic constraint is introduced into the basic fitting objective, and a comprehensive objective function is constructed by parameter correction through constraint penalty terms and then solved to obtain the optimal parameters and weights. Based on the initial quota base price of the target project, the corrected quota price is generated by applying the optimal parameters and weights.

[0031] In this embodiment, the feature vector of the target project scene is obtained. Historical project data And record the historical project sample base price difference as .

[0032] in, For transportation distance, For construction time, This is the level for live-line work. For emergency response purposes, For prefabrication rate, The first in the historical project sample Scene feature vectors of each project The first in the historical project sample The initial fixed base price for each project The first in the historical project sample The actual settlement price of each project For the first Historical project sample base price difference for each project This represents the total number of historical projects.

[0033] In this embodiment, a set of parameterized functions is established based on the scene feature vector as a fitting benchmark. This set of parameterized functions is denoted as... The parameter set is denoted as , "in "" represents a specific scene feature vector.

[0034] The parameterized functions specifically include transportation parameterized functions, construction period parameterized functions, live-line work parameterized functions, and prefabrication rate parameterized functions.

[0035] The specific formula for the transportation parameterization function is as follows:

[0036] In the formula, For transportation parameters, This is the upper limit coefficient for transportation costs. For the sensitivity coefficient of transportation distance, For transportation distance.

[0037] The specific formula for the parameterized function during the construction period is as follows:

[0038] In the formula, For construction period parameters, During the construction period, These are the parameter values ​​for nighttime construction. These are parameter values ​​for peak construction periods. Parameter values ​​for construction during holidays.

[0039] It should be noted that the parameterized functions for the construction period are defined according to mutually exclusive categories, assuming... The construction period category is determined according to the following priority: if it is a statutory holiday... Otherwise, if it is during the daytime peak hours... Otherwise, if it is nighttime... ;otherwise The method for determining peak hours is the time window method, specifically: during non-holiday periods, 18:00–21:00 is considered the peak period.

[0040] The specific formula for the parameterized function of live-line working is as follows:

[0041] In the formula, For live-line working parameters, This is the sensitivity coefficient for live-line working. The difficulty level of live-line work.

[0042] The specific formula for the emergency parameterization function is as follows:

[0043] In the formula, For emergency parameters, These are the adjustment coefficients for the emergency parameterization function.

[0044] The specific formula for the prefabrication rate parameterization function is as follows:

[0045] In the formula, For prefabrication rate parameter, The adjustment coefficient for the prefabrication rate parameterization function. Prefabrication rate.

[0046] It should be noted that the prefabrication rate refers to the proportion of components, structures, or parts required for construction in low-voltage power distribution network protection projects that are prefabricated in a factory, rather than being temporarily fabricated or cast on-site. The purpose of the prefabrication rate is to reflect the "difference between construction workload and efficiency" in the project by quantifying the ratio of on-site construction to factory prefabrication, thereby adjusting the quota price and interacting with other scenario parameters to ensure that the quota management of low-voltage power protection projects is more in line with the actual situation.

[0047] In this embodiment, the basic fitting target is obtained by minimizing the loss function with regularization term based on the base price difference of historical project samples, using a set of parameterized functions.

[0048] It should be noted that the basic fit objective means minimizing the error between the predicted base price difference with regularization and the actual base price difference. By adding a regularization term to constrain the parameters, the optimal values ​​of the parameterized function and weights are obtained.

[0049] The formula for minimizing the loss function with regularization is as follows:

[0050] In the formula, Based on the fitting target, For the first Historical project sample base price difference for each project The number of historical project samples. For the first The weights of each parameterized function, For the first A parameterized function, The first in the historical project sample The scene feature vector of each sample, For the first in the parameter set One parameter, The regularization coefficient is . This is a regularization term.

[0051] In this embodiment, causal logic constraints are introduced into the basic fitting objective, and a comprehensive objective function is constructed by parameter correction through constraint penalty terms, specifically as follows: The fitting parameters are corrected by introducing causal logic constraints, which include nonnegativity constraints, monotonicity constraints, boundary constraints, and interaction logic constraints. When the fitted parameters violate the causal logic constraints, the degree of constraint violation is quantified by constraint penalty, and a constraint violation quantification index is obtained. Based on the basic fitting objective and the constraint violation quantification index, a comprehensive objective function is constructed.

[0052] In this embodiment, the causal logic constraints are specifically as follows: Non-negativity constraint: Enforcing resource consumption-related parameters ; Monotonicity constraint: Apply the property of non-decreasing monotonicity to the transport distance and the charge level; Boundary constraints: Applicable to emergency parameters and construction period parameters, an upper limit needs to be set, and the upper limit should be set according to the actual situation of the project and industry standards; Interaction logic constraints: Applicable to the interaction relationship between transportation distance and prefabrication rate in transportation parameters, as detailed below: The interaction between transportation distance and prefabrication rate in transportation parameters is as follows: as the prefabrication rate increases, the marginal sensitivity of transportation parameters to distance should decrease. This logic is reflected by adjusting the form of transportation parameters. The specific formula for the interaction logic between transportation distance and prefabrication rate is as follows:

[0053] In the formula, A coefficient used to control the relationship between transportation distance and prefabrication rate.

[0054] In this embodiment, for all parameters, when a causal logic constraint is violated, a constraint penalty is introduced to quantify the part that violates the constraint, resulting in a penalty term. ; Define iterative weights According to the penalty items Construct a comprehensive objective function by combining the basic fitting objective; Extract the corrected parameters from the comprehensive objective function. and parameter weights ; According to the corrected parameters and parameter weights Calculate the revised quota price.

[0055] The formula for the comprehensive objective function is as follows:

[0056] In the formula, For the comprehensive objective function, Based on the fitting target, This is a penalty term function for the parameters and parameter weights.

[0057] The revised fixed price formula is as follows:

[0058] In the formula, To revise the fixed price, This is the initial fixed base price. For the target project scene feature vector, The final parameterized function value is generated after completing parameter fitting and correction by causal logic constraints.

[0059] It should be noted that, The meaning is

[0060] S3 generates deviation estimates based on the target project scene feature vector and historical project data through a scene deviation estimation model.

[0061] In this embodiment, the scene bias estimation model includes the calculation of scene rarity score, hierarchical sampling of rare scenes, and the calculation of bias estimate based on Bayesian contraction.

[0062] In this embodiment, the step of generating a deviation estimate based on the target project scene feature vector and historical project data using a scene deviation estimation model specifically involves: Based on the feature vectors of historical scenes and the feature vectors of the target project scene, the scene rarity score of each sample is obtained by comprehensively calculating the inverse frequency of discrete features and the distribution of continuous features. Historical project samples are stratified based on scene rarity scores, and within the stratum to which the target project sample belongs, historical project samples similar to the target project sample are selected through similarity calculation to form a similar sample set. Using similarity as the weight, calculate the weighted mean of the base price difference and the effective sample size in the historical project samples after screening; The bias estimate is obtained by Bayesian shrinkage calculation based on the weighted mean and the effective sample size.

[0063] In this embodiment, the scene feature vector is based on each historical project sample. and target project scene feature vector Based on the inverse frequency of discrete features and the tail parameters of continuous features, the rarity scores of discrete features and continuous features for historical project samples and target project samples are calculated, respectively. Then, based on the discrete feature rarity scores and continuous feature rarity scores, a scene rarity score is calculated for each historical project sample. Scene rarity score of the target project sample ; Based on rarity score The historical project samples are stratified, and samples are retrieved only from the same stratum as the target rarity stratum to construct a candidate set. satisfy and in The similarity between the target project sample and the historical project samples is calculated based on the similarity. ,reserve The samples are used to form a set of historical project samples that are similar to the target project sample, denoted as the similar sample set. ,in For similarity threshold, The threshold for the candidate set; Using similarity as sample weight, the sample weight is defined. Calculate the set of similar samples The weighted mean and effective sample size of the base price spread; Based on the weighted mean and the effective sample size, the bias estimate is obtained through Bayesian shrinkage calculation.

[0064] The specific formula for calculating the discrete feature rarity score is as follows:

[0065] In the formula, For the first Each discrete feature value Rarity score, It is a smoothing constant. For the first In a discrete feature, the value is... The frequency.

[0066] The specific formula for calculating the rarity score of continuous features is as follows:

[0067] In the formula, For the first Continuous feature values Rarity score, It is a smoothing constant. The empirical distribution function for the feature is denoted as less than or equal to . The proportion.

[0068] The formula for calculating the overall rarity score is as follows:

[0069] In the formula, Score for rarity, Let be the number of dimensions of the sample features. For the first The rarity score of each feature in a single dimension.

[0070] The formula for calculating the weighted average is as follows:

[0071] In the formula, For weighted average, For sample weights, For the first Historical project sample base price difference for each project It is a set of similar samples.

[0072] The formula for calculating the effective sample size is as follows:

[0073] In the formula, This is the effective sample size.

[0074] The specific formula for calculating Bayesian shrinkage is as follows:

[0075] like Then the formula simplifies to:

[0076] In the formula, This is the bias estimate after Bayesian shrinkage. For shrinkage strength parameters, This is the prior mean.

[0077] S4. Determine the deviation compensation quota price based on the corrected quota price and the deviation estimate.

[0078] In this embodiment, determining the deviation compensation quota price based on the corrected quota price and the deviation estimate specifically involves: When the effective sample size reaches a preset threshold, the deviation compensation quota price for the target project sample is calculated using the deviation estimate as the compensation amount.

[0079] In this embodiment, when the effective sample size At that time, the deviation estimate As the compensation amount, the deviation compensation quota price for the target project sample is calculated, where, This is a preset threshold.

[0080] The specific formula for calculating the deviation compensation quota price is as follows:

[0081] In the formula, The deviation compensation quota price, To revise the fixed price.

[0082] S5 uses the initial quota base price as a benchmark, and weights and merges the increments between the enterprise quota price and the deviation compensation quota price according to the confidence level to generate the final optimized quota price.

[0083] In this embodiment, the step of using the initial quota base price as a benchmark and weighting and fusing the increments between the enterprise quota price and the deviation compensation quota price according to confidence level to generate the final optimized quota price specifically involves: Using the initial quota base price as a unified benchmark, calculate the first and second increments of the deviation compensation quota price and the enterprise quota price relative to the benchmark, respectively. Based on the preset confidence level assessment rules, the first confidence level and the second confidence level corresponding to the deviation compensation quota price and the enterprise quota price are determined respectively; The fusion weights are calculated based on the first confidence level and the second confidence level. Based on the first increment, the second increment, and the fusion weight, the comprehensive increment correction value is calculated; The initial fixed base price is added to the comprehensive incremental correction value to generate the final optimized fixed base price.

[0084] In this embodiment, the initial quota base price calculated by the reference quota is used as a unified anchor point, and the first increment and the second increment of the deviation compensation quota price and the enterprise quota price relative to the initial quota base price are calculated respectively. Obtain three historical data points: historical error, effective sample size, and timeliness. Map each of these three data points to a 0-1 score to form an error score, a sample size score, and a timeliness score. The confidence levels of the reference quota, deviation compensation quota price, and internal enterprise quota are calculated based on the error score, sample size score, and timeliness score, respectively, to obtain the confidence level of the reference quota. First confidence level Second confidence level ; Based on the first confidence level Second confidence level Calculate fusion weights ; Based on the fusion weight, and combining the first and second increments, the comprehensive increment correction value is calculated. The initial quota base price is added to the comprehensive incremental correction value to generate the final optimized quota price; Based on the optimized quota price, a revision instruction document is generated, listing the revision parameters and amounts for each step, and the revision process is recorded in the operation log chain, generating a compliance audit report and a revision suggestion package that can be submitted.

[0085] It should be noted that the confidence level of the calculated reference quota is only used for archiving and auditing reference, and is not involved in the calculation of the final optimized quota price.

[0086] The specific formula for calculating confidence level is as follows:

[0087] In the formula, For confidence level, For error fractions, For sample size fractions, This is a timeliness score.

[0088] Furthermore, the specific formula for calculating the fusion weight is as follows:

[0089] In the formula, To integrate weights, As the first confidence level, This represents the second confidence level.

[0090] Furthermore, the specific formula for calculating the comprehensive incremental correction value is as follows:

[0091] In the formula, This is a comprehensive incremental correction value. This is the initial fixed base price. The deviation compensation quota price, As the first increment, This is the second increment.

[0092] Furthermore, the final optimized formula for calculating the fixed price is as follows:

[0093] In the formula, To ultimately optimize the fixed price.

[0094] Example 2, Figure 2 The present invention provides a system for optimizing quota management of low-voltage power supply protection projects in distribution networks, comprising: The base price calculation module is used to calculate the initial base price of the target project sample based on the reference quota; The learning correction module is used to generate a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data, by learning dynamic parameters of the scenario. The deviation estimation module is used to generate deviation estimates based on the target project scene feature vector and historical project data through a scene deviation estimation model. The compensation price generation module is used to determine the deviation compensation quota price based on the corrected quota price and the deviation estimate. The quota price optimization module is used to generate the final optimized quota price by weighting the increments between the enterprise quota price and the deviation compensation quota price according to the confidence level, based on the initial quota base price.

[0095] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0096] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0097] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0099] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0100] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing quota management of low-voltage power supply protection projects in distribution networks, characterized in that, Includes the following steps: Calculate the initial quota base price for the target project sample based on the reference quota; Based on the initial fixed base price, the target project scenario feature vector, and historical project data, a corrected fixed base price is generated through scenario dynamic parameter learning. Based on the target project scene feature vector and historical project data, a deviation estimate is generated through a scene deviation estimation model. The deviation compensation quota price is determined based on the revised quota price and the deviation estimate; Using the initial quota base price as a benchmark, the increments between the enterprise quota price and the deviation compensation quota price are weighted and merged according to the confidence level to generate the final optimized quota price.

2. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 1, characterized in that, The calculation of the initial quota base price for the target project sample based on the reference quota is specifically as follows: Based on the reference quota, obtain the quota consumption of the sample sub-items of the target project; The initial quota base price is calculated based on the quota consumption and the market unit price of the corresponding sub-item.

3. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 2, characterized in that, The scenario dynamic parameter learning includes establishing parameterized functions, parameter fitting, causal logic constraints, and constraint penalties; the historical project data includes historical scenario feature vectors, historical project initial quota base prices, and historical project actual settlement prices.

4. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 3, characterized in that, The process of generating a revised quota price based on the initial quota base price, the target project scenario feature vector, and historical project data through scenario dynamic parameter learning is as follows: Obtain the base price difference for historical project samples, whereby the base price difference is the difference between the actual settlement price of the historical project and the initial quota base price of the historical project; Based on the feature vector of the target project scenario, establish a set of parameterized functions; Using the historical project sample base price difference as the optimization objective, and based on a set of parameterized functions, the basic fitting objective is obtained by minimizing the loss function with a regularization term. A causal logic constraint is introduced into the basic fitting objective, and a comprehensive objective function is constructed by parameter correction through constraint penalty terms and then solved to obtain the optimal parameters and weights. Based on the initial quota base price of the target project, the corrected quota price is generated by applying the optimal parameters and weights.

5. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 4, characterized in that, The basic fitting objective introduces causal logic constraints, and constructs a comprehensive objective function through parameter correction using constraint penalty terms, specifically: The fitting parameters are corrected by introducing causal logic constraints, which include nonnegativity constraints, monotonicity constraints, boundary constraints, and interaction logic constraints. When the fitted parameters violate the causal logic constraints, the degree of constraint violation is quantified by constraint penalty, and a constraint violation quantification index is obtained. Based on the basic fitting objective and the constraint violation quantification index, a comprehensive objective function is constructed.

6. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 5, characterized in that, The scene bias estimation model includes the calculation of scene rarity score, hierarchical sampling of rare scenes, and calculation of bias estimate based on Bayesian contraction.

7. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 6, characterized in that, The process of generating deviation estimates based on the target project scene feature vector and historical project data using a scene deviation estimation model is as follows: Based on the feature vectors of historical scenes and the feature vectors of the target project scene, the scene rarity score of each sample is obtained by comprehensively calculating the inverse frequency of discrete features and the distribution of continuous features. Historical project samples are stratified based on scene rarity scores, and within the stratum to which the target project sample belongs, historical project samples similar to the target project sample are selected through similarity calculation to form a similar sample set. Using similarity as the weight, calculate the weighted mean of the base price difference and the effective sample size in the historical project samples after screening; The bias estimate is obtained by Bayesian shrinkage calculation based on the weighted mean and the effective sample size.

8. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 7, characterized in that, The determination of the deviation compensation quota price based on the corrected quota price and the deviation estimate is as follows: When the effective sample size reaches a preset threshold, the deviation compensation quota price for the target project sample is calculated using the deviation estimate as the compensation amount.

9. The method for optimizing quota management of low-voltage power supply protection projects in distribution networks according to claim 8, characterized in that, The process involves using the initial quota base price as a benchmark, weighting and fusing the increments between the enterprise quota price and the deviation compensation quota price according to confidence level to generate the final optimized quota price, specifically as follows: Using the initial quota base price as a unified benchmark, calculate the first and second increments of the deviation compensation quota price and the enterprise quota price relative to the benchmark, respectively. Based on the preset confidence level assessment rules, the first confidence level and the second confidence level corresponding to the deviation compensation quota price and the enterprise quota price are determined respectively; The fusion weights are calculated based on the first confidence level and the second confidence level. Based on the first increment, the second increment, and the fusion weight, the comprehensive increment correction value is calculated; The initial fixed base price is added to the comprehensive incremental correction value to generate the final optimized fixed base price.

10. A system using the optimization method for quota management of low-voltage power supply protection projects in distribution networks as described in any one of claims 1-9, comprising: The base price calculation module is used to calculate the initial base price of the target project sample based on the reference quota; The learning correction module is used to generate a corrected quota price based on the initial quota base price, the target project scenario feature vector, and historical project data, by learning dynamic parameters of the scenario. The deviation estimation module is used to generate deviation estimates based on the target project scene feature vector and historical project data through a scene deviation estimation model. The compensation price generation module is used to determine the deviation compensation quota price based on the corrected quota price and the deviation estimate. The quota price optimization module is used to generate the final optimized quota price by weighting the increments between the enterprise quota price and the deviation compensation quota price according to the confidence level, based on the initial quota base price.

Images