Electricity price deduction method and system based on simulation environment

Abstract

The invention belongs to the technical field of electricity marketing, and provides an electricity price deduction method and system based on a simulation environment, and the method comprises the steps: obtaining the power utilization data of a user; constructing a deduction data model based on the acquired power consumption data; dynamically adjusting parameters of the constructed deduction data model to obtain a power regulation and control parameter library; according to the obtained power regulation and control parameter library, under different deduction data model parameter scenes, carrying out simulation calculation on the load electricity price, and generating a dynamic evaluation report; according to the generated dynamic evaluation report, considering multi-dimensional parameter boundaries of a user side, a power grid side and a social side, and constructing an electricity price deduction multi-objective optimization model; and solving the constructed electricity price deduction multi-objective optimization model, performing multi-objective balance optimization in combination with dynamic weight distribution and a constraint punishment mechanism, generating a deduction data parameter configuration scheme, and completing electricity price deduction based on the simulation environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of power marketing, and in particular relates to an electricity price deduction method and system based on a simulation environment. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the advancement of electricity market reform, electricity tariff and price policies are frequently introduced. The implementation of various time-of-use electricity price policies, the entry of new settlement entities into the market, and diversified market transaction forms have a bearing on electricity marketing. Analytical models for topics such as electricity sales situation forecasting, purchase and sales settlement quality and efficiency, and dynamic electricity price evaluation have emerged.

[0004] At present, the cost management mechanism for various regulatory and supporting resources in the power market needs to be improved, and the reform of transmission and distribution prices, on-grid electricity prices, and sales electricity prices still needs to be further deepened. Deep-seated contradictions in the power market mechanism continue to emerge, and the market mechanism and regulation strategy for flexible, efficient, and portable load-side resources urgently need to be improved. The current electricity price policy deduction has the following main drawbacks:

[0005] (1) Computational deduction is highly dependent on professional and technical personnel; the deduction of policy effects involves modeling, analysis, and calculation of multiple complex factors, which is highly professional and difficult, and highly dependent on professional and technical personnel;

[0006] (2) The amount of business data is large, and it is difficult to guarantee the efficiency and accuracy of the deduction work; traditional computational deduction is completed manually offline, and manual data extraction, analysis and comparison, data verification, and business calculation and fee policy simulation work are carried out offline, which has problems such as low efficiency, slow response, and data distortion;

[0007] 3) Due to the high complexity of the business, the traditional manual model can only carry out the deduction of some scenarios; based on the offline manual deduction method, it is difficult to complete the deduction of complex scenarios such as batch adjustment of data tables, correlation adjustment of power factor assessment coefficients, and addition of deep valleys, and the application scenarios are limited. Summary of the Invention

[0008] To solve the above problems, the present invention proposes a method and system for electricity price deduction based on a simulation environment. By conducting relevant research on price policy deduction based on an electricity price simulation environment, the analysis and optimization capabilities of power market operation and policy research and judgment are improved through pre-implementation rehearsal of electricity price policy, post-implementation evaluation of electricity price policy, and deduction of electricity price policy optimization space.

[0009] According to some embodiments, a first solution of the present invention provides an electricity price deduction method based on a simulation environment, which adopts the following technical solutions:

[0010] A method for estimating electricity prices based on a simulation environment, comprising:

[0011] Obtain user electricity consumption data;

[0012] Build a deduction data model based on the acquired electricity consumption data;

[0013] Dynamically adjust the parameters of the constructed deduction data model to obtain the power control parameter library;

[0014] Based on the obtained power control parameter library, simulate the load electricity price under different deduction data model parameter scenarios and generate a dynamic evaluation report;

[0015] Based on the generated dynamic assessment report, a multi-objective optimization model for electricity price deduction is constructed by considering the multi-dimensional parameter boundaries of the user side, the grid side, and the social side;

[0016] Solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate deduction data parameter configuration plan, and complete electricity price deduction based on simulation environment.

[0017] As a further technical limitation, in the process of the multi-objective balance optimization, the differential evolution method is combined to perform global search and parameter optimization on the constructed electricity price deduction multi-objective optimization model, and self-learning updates of dynamic weight changes are performed according to the individual fitness feedback results of the differential evolution method. Alternating local search and global jumps are used to avoid falling into local optimality, and dynamic weight allocation is used to perform rolling monitoring of the changing trend of the objective function of the multi-objective optimization model to obtain a deduction data parameter configuration plan.

[0018] Furthermore, in the process of rolling monitoring of the changing trend of the objective function of the multi-objective optimization model using dynamic weight allocation, the penalty cost of the infeasible solution in the objective function is embedded into the objective function in combination with the constraint penalty mechanism, and the constructed electricity price deduction multi-objective optimization model is dynamically solved to obtain the optimal solution area of ​​the multi-objective optimization model.

[0019] As a further technical limitation, during the load electricity price simulation calculation process, the obtained power control parameters are dynamically evaluated under the adjustment methods of fixed value adjustment, proportional adjustment, time period adjustment and step adjustment, and the impact of the adjustment on the grid load distribution, user electricity costs and the competitive landscape of the power market is determined.

[0020] As a further technical limitation, the deduction data model parameters include at least electricity price adjustment parameters, electricity price policy parameters, demand response subsidy parameters and distribution network investment parameters.

[0021] As a further technical limitation, the user electricity consumption data obtained includes at least the monthly electricity consumption and electricity bill information of the metering point, the monthly electricity consumption and electricity bill data of classified users, the 96-point collected electricity data, the electricity price model data and the power factor assessment coefficient; before constructing the deduction data model, the obtained user electricity consumption data is subjected to data cleaning and data format conversion in sequence.

[0022] According to some embodiments, a second solution of the present invention provides an electricity price deduction system based on a simulation environment, which adopts the following technical solutions:

[0023] An electricity price deduction system based on a simulation environment includes:

[0024] An acquisition module configured to acquire user electricity usage data;

[0025] A construction module configured to construct a deduction data model based on the acquired electricity consumption data;

[0026] an adjustment module configured to dynamically adjust the parameters of the constructed deduction data model to obtain a power regulation parameter library;

[0027] A simulation module is configured to perform simulation calculations of load electricity prices under different deduction data model parameter scenarios based on the obtained power control parameter library and generate a dynamic evaluation report;

[0028] The deduction module is configured to construct a multi-objective optimization model for electricity price deduction based on the generated dynamic evaluation report, taking into account the multi-dimensional parameter boundaries of the user side, the grid side and the social side; solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate a deduction data parameter configuration plan, and complete the electricity price deduction based on the simulation environment.

[0029] According to some embodiments, a third solution of the present invention provides a computer-readable storage medium, which adopts the following technical solution:

[0030] A computer-readable storage medium stores a program, which, when executed by a processor, implements the steps of the electricity price deduction method based on a simulation environment as described in the first solution of the present invention.

[0031] According to some embodiments, a fourth solution of the present invention provides an electronic device, which adopts the following technical solution:

[0032] An electronic device comprises a memory, a processor and a program stored in the memory and running on the processor. When the processor executes the program, the steps of the electricity price deduction method based on the simulation environment as described in the first embodiment of the present invention are implemented.

[0033] According to some embodiments, a fifth solution of the present invention provides a computer program product, which adopts the following technical solution:

[0034] A computer program product includes software code, wherein the program in the software code executes the steps of the electricity price deduction method based on a simulation environment as described in the first embodiment of the present invention.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] This application conducts relevant research on price policy deduction based on electricity price simulation environment, and improves the analysis and optimization capabilities of power market operation and policy research and judgment through pre-implementation rehearsal of electricity price policy, post-implementation evaluation of electricity price policy and deduction of electricity price policy optimization space. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings constituting a part of the specification of this embodiment are used to provide a further understanding of this embodiment. The schematic embodiments and descriptions of this embodiment are used to explain this embodiment and do not constitute an improper limitation on this embodiment.

[0038] Figure 1 This is a flow chart of a method for estimating electricity prices based on a simulation environment in Embodiment 1 of the present invention;

[0039] Figure 2 This is a schematic diagram of an application architecture for electricity price policy deduction in a simulation environment in the first embodiment of the present invention;

[0040] Figure 3 This is a schematic diagram of the technical architecture for electricity price policy deduction in a simulation environment in the first embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of the process of constructing the deduction data model in the first embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the process of constructing a multi-factor influencing factor map in the first embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram of the application process of the correlation influencing factor deduction model in Example 1 of the present invention;

[0044] Figure 7 This is a schematic diagram of the process of constructing a quantifiable correlation influence network in the first embodiment of the present invention;

[0045] Figure 8 This is a schematic diagram of the correlation impact report and the correlation impact star map in Example 1 of the present invention;

[0046] Figure 9 This is a schematic diagram of adjustment of estimation and deduction parameters in the first embodiment of the present invention;

[0047] Figure 10 This is a schematic diagram of editing the estimation deduction formula in the first embodiment of the present invention;

[0048] Figure 11 This is a schematic diagram of the estimation calculation and deduction results in Example 1 of the present invention;

[0049] Figure 12 This is a schematic diagram of adjusting actuarial deduction parameters in the first embodiment of the present invention;

[0050] Figure 13 This is a schematic diagram of editing the actuarial deduction formula in the first embodiment of the present invention;

[0051] Figure 14 This is a schematic diagram of the actuarial deduction calculation results in Example 1 of the present invention;

[0052] Figure 15 This is a schematic diagram of the calculation and deduction diagram for adding valley electricity prices in Example 1 of the present invention;

[0053] Figure 16 This is a structural block diagram of an electricity price deduction system based on a simulation environment in Example 2 of the present invention. DETAILED DESCRIPTION

[0054] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0055] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0056] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0057] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0058] Example 1

[0059] The first embodiment of the present invention introduces a method for estimating electricity prices based on a simulation environment.

[0060] like Figure 1 The electricity price deduction method based on the simulation environment shown includes:

[0061] Obtain user electricity consumption data;

[0062] Build a deduction data model based on the acquired electricity consumption data;

[0063] Dynamically adjust the parameters of the constructed deduction data model to obtain the power control parameter library;

[0064] Based on the obtained power control parameter library, simulate the load electricity price under different deduction data model parameter scenarios and generate a dynamic evaluation report;

[0065] Based on the generated dynamic assessment report, a multi-objective optimization model for electricity price deduction is constructed by considering the multi-dimensional parameter boundaries of the user side, the grid side, and the social side;

[0066] Solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate deduction data parameter configuration plan, and complete electricity price deduction based on simulation environment.

[0067] According to the electricity price policy, the demand is judged. Figure 2 The electricity price policy simulation application architecture and Figure 3 The technical architecture of electricity price policy deduction in the simulation environment is shown; among them, the application architecture of electricity price policy deduction in the simulation environment includes three parts: pre-policy implementation preview, post-policy implementation evaluation and policy optimization space deduction; the technical architecture of electricity price policy deduction in the simulation environment includes four layers: data layer, technology layer, background processing layer and front-end display layer.

[0068] like Figure 3As shown in the figure, the data layer is responsible for storing and managing the system's raw data. It includes monthly electricity consumption and electricity bill information for metering points, monthly electricity consumption and electricity bill data for classified users, electricity data collected at 96 points, electricity price model data, and power factor assessment coefficients. This data is the basis for system operation and measurement, providing necessary information support for the upper layer. The technical layer is divided into two parts: deduction preparation configuration and calculation deduction engine. Deduction preparation configuration includes grouping condition configuration, adjustable parameter field configuration, algorithm formula configuration, and auxiliary parameter management, which are used to set the various parameters and conditions required for deduction. The calculation deduction engine is responsible for executing specific deduction calculations, including deduction data preparation, execution of deduction calculations, generation of deduction data, data comparison and analysis, and generation of deduction conclusions. The background processing layer is responsible for the system's logical processing and business process control. It includes five steps: deduction plan formulation, sample user identification, deduction parameter adjustment, calculation and deduction, and deduction report. Through these steps, the system can perform deduction calculations based on preset plans and parameters and generate corresponding reports; the front-end display layer is responsible for presenting the processed information to the user in a visual manner, including preview graphic reports, mind map analysis windows, correlation impact reports, correlation impact star maps, and three-dimensional scatter plots, etc., to help users better understand and analyze the deduction results.

[0069] This embodiment builds an intelligent deduction system covering the entire life cycle of power policy, forming a closed-loop management process of "preview-evaluation-optimization"; in the policy preview stage, through environmental modeling, parameter setting and simulation calculation, the impact of adjustment scenarios such as electricity prices and electricity consumption on user electricity charges and grid load is simulated, and a multi-dimensional quantitative analysis report is generated to provide a scientific basis for policy formulation; in the policy evaluation stage, the system integrates various types of data, extracts key features, builds a knowledge graph based on causal structure recognition, and combines the XGBoost algorithm for causal deduction; through association rules, the policy effects are attributed and their direct and indirect effects are quantified; at the same time, Monte Carlo simulation is introduced to evaluate different scenarios The confidence interval and risk distribution under the above conditions can improve the scientificity and robustness of the assessment results; in the policy optimization stage, the system integrates the boundary parameters of multiple parties such as users, power grids, and society, adopts the differential evolution algorithm to solve the multi-objective Pareto optimal solution set, and generates differentiated solutions such as radical and robust ones; combines three-dimensional visualization to display parameter associations, and embeds dynamic adjustment to participate in the risk monitoring mechanism to ensure that the optimization solution has good adaptability and security; this embodiment integrates machine learning algorithms, data modeling and visualization technologies to realize full-process digital decision support from policy simulation and deduction, effect attribution analysis to parameter adaptive optimization, and comprehensively improves the scientificity, accuracy and foresight of power policy making.

[0070] This embodiment uses machine learning algorithms to build prediction and deduction models for electricity consumption, load, electricity price, etc., obtains high-quality data from the power industry through the data middle platform, and uses flexible parameter settings to simulate policy effects; through precise simulation calculations and optimization evaluations, the results are visualized and intelligent analysis reports are automatically generated, which can help decision makers comprehensively rehearse the specific impact of policy adjustments on user electricity consumption before implementing the policy, thereby providing a scientific and reliable basis for policy making and ensuring the effectiveness and operability of decision-making.

[0071] As one or more implementation methods, the user electricity consumption data obtained in this embodiment includes at least monthly electricity consumption and electricity fee information of the metering point, monthly electricity consumption and electricity fee data of classified users, 96-point collected electricity data, electricity price model data, power factor assessment coefficient, etc.; among them, the monthly electricity consumption and electricity fee information of the metering point and the monthly electricity consumption and electricity fee data of classified users are derived from the Marketing 2.0 system, which is used to calculate the changes in electricity charges of different categories of users under different electricity price policies; the 96-point collected electricity data is provided by the metering automation system, which is used to analyze the user's electricity load characteristics and support policy simulations such as load transfer and demand response; the electricity price model data is derived from the electricity price management system, including various electricity price policies, time-of-use electricity prices, peak and valley period divisions, etc., which are used to construct electricity price parameters for simulation calculations; the power factor assessment coefficient is provided by the assessment system, which is used to simulate the impact of power factor assessment on the electricity charges of corporate users.

[0072] This embodiment relies on the data center to process, process and aggregate data through technologies such as OGG and Datahub, and forms the following data according to the needs of the computing and deduction platform: Figure 4 The calculation and deduction data model shown covers dimensions including basic customer attributes, settlement volume and fees, basic electricity charges and power regulation, involving more than 240 fields to ensure data integrity and multi-dimensional support.

[0073] In this embodiment, in order to ensure data quality and calculation accuracy, a series of standardized processing measures need to be taken: data cleaning, elimination of outliers and missing data, such as sudden changes in electricity consumption, deformed load curves and other problems. Secondly, data format conversion is performed to ensure that the encoding, timestamp and unit system of data from different systems are consistent. For the 96-point collected data, interpolation, smoothing filtering and other methods are used to process missing points to ensure the continuity of time series data. For classified user data, it is necessary to unify the caliber and classify and summarize to ensure that the statistical indicators of different categories of users are consistent. The electricity price model data needs to be version checked to ensure that the electricity price parameters used in the simulation are consistent with the actual policy.

[0074] The construction of the simulated business operating environment must ensure high consistency with the marketing production environment to ensure the authenticity and reliability of policy simulation. To this end, the application services, business parameters, billing models, process permissions, etc. of the simulation environment must be strictly aligned. The following process is used to achieve efficient synchronization:

[0075] (1) Baseline generation: Perform hash verification on the programs, databases, and configuration files of the production environment to generate a unique identifier as a simulation benchmark.

[0076] (2) Difference comparison: Automatically scan the simulation environment to identify programs, parameters, and business configurations that are inconsistent with the production environment benchmark, and form a difference list.

[0077] (3) Incremental synchronization: Based on the difference list, only abnormal items are transmitted, and the correct version of files and data parameters are extracted and synchronized to avoid full coverage and improve synchronization efficiency.

[0078] (4) Result verification: After synchronization is completed, the simulation environment is hashed again and compared with the production environment to ensure that key parameters, program files, and billing models are completely consistent. A verification report is generated and archived for future reference.

[0079] To build a scientific and systematic policy simulation system, this embodiment needs to widely collect various power market policies and regulatory strategies, use artificial intelligence technology to parse policy documents, break them down into structured data, and compile them into a policy regulation parameter library. This library covers key policy variables such as electricity price adjustments, basic electricity rate policies, demand response subsidies, and distribution network investment, and supports flexible settings and dynamic adjustments to ensure that simulation calculations can timely and accurately simulate policy implementation effects and provide reliable data support for decision-making.

[0080] Regarding electricity price adjustments, the parameter library must include core indicators such as the base price, kilowatt-hour price, tiered price, time-of-use price, peak-valley price, and agricultural price, to facilitate analysis of the impact of price changes on electricity consumption by different user groups. For example, in a time-of-use price adjustment scenario, peak-hour prices could be increased by 5%-10% and off-hour prices could be decreased by 3%-8% to evaluate the effectiveness of different adjustment ranges on load curve optimization. Regarding basic electricity price policies, the parameter library can define different billing methods and support adjustment of assessment cycles to analyze the impact of various plans on enterprise electricity costs. Regarding demand response subsidies, the parameter library can define incentive mechanisms such as direct subsidies and tiered subsidies, and allow for adjustment of subsidy amounts and incentive methods to quantify user response willingness and load regulation potential. Regarding distribution network investment policies, the parameter library can define key variables such as the distribution network investment cost sharing ratio and equipment connection fees to evaluate the impact of different investment strategies on renewable energy access and user economic benefits.

[0081] The adjustment methods of policy parameters mainly include fixed value adjustment, proportional adjustment, time-based adjustment and step-by-step adjustment to meet different policy regulation needs:

[0082] (1) Fixed value adjustment: Applicable to direct adjustments to a single policy variable, such as an electricity price increase of 0.05 yuan / kWh.

[0083] (2) Proportional adjustment: Applicable to relative changes in policy variables, such as a 20% increase in demand response subsidies.

[0084] (3) Time-based adjustment: Applicable to time-dependent policy adjustments, such as a 10% increase in peak-time electricity prices and a 5% decrease in off-peak-time electricity prices.

[0085] (4) Step adjustment: Applicable to policies set for different user categories or electricity consumption levels, for example, the electricity price for large industrial users will increase by 5%, while the electricity price for residential users will remain unchanged.

[0086] These adjustment methods can be applied individually or in combination to ensure the flexibility and accuracy of policy simulations. Based on the power policy control parameter library and its adjustable parameter setting system, simulation calculations can dynamically evaluate the impact of policy adjustments on grid load distribution, user electricity costs, and the competitive landscape of the power market under different policy scenarios. This provides a scientific basis for policy formulation and implementation, and supports policy optimization and implementation.

[0087] The core task of the simulation calculation in this embodiment is to evaluate the changes in electricity consumption and electricity prices under different policy scenarios and analyze the influencing factors. Specifically, using the parameters set in the aforementioned policy simulation, combined with the collected user electricity consumption data, electricity price parameters, demand response and other information, the simulation system can simulate the specific impact of various policy adjustments on the power market and user behavior. For example, adjustments to electricity prices, subsidy policies or grid investment strategies will be directly reflected in changes in electricity consumption, load and electricity prices, thereby providing a quantitative basis for policy optimization.

[0088] In order to evaluate the uncertainty and risks brought about by policy adjustments, this embodiment uses the Monte Carlo simulation method to perform multi-scenario comparative calculations on different policy scenarios; by performing multiple random sampling of uncertain factors (such as electricity price fluctuations, changes in user responses, etc.), the simulation system can calculate the risk range of policy adjustments and provide decision makers with the best and worst possible results under different scenarios. At the same time, the system introduces a dual calculation mechanism of estimation mode and actuarial mode: the estimation mode is suitable for rapid prediction of macro trends, and uses parameter default values ​​and neutral assumptions to quickly give preliminary evaluation results; while the actuarial mode relies on more detailed data input and rule settings to perform high-precision simulation and in-depth tuning, which is suitable for refined calculations in key decision-making scenarios. This process helps to reveal the potential risks of policy implementation and provide data support for adjustment and optimization.

[0089] To achieve optimization under different policy objectives, this embodiment uses a differential evolution algorithm to solve policy solutions. The differential evolution algorithm can efficiently search for optimal solutions in complex multidimensional policy spaces. For example, under strategies such as electricity price adjustments and demand response subsidies, it can solve the optimal policy solutions for different goals such as maximizing benefits and minimizing impacts.

[0090] To adjust the impact on electricity consumption, load, and electricity costs, this embodiment constructs an electricity consumption prediction model, a load response model, and an electricity cost impact model based on collected user electricity consumption data and policy parameters. It also combines machine learning methods such as time series forecasting, reinforcement learning, and regression analysis to achieve quantitative evaluation under different policy scenarios and provide accurate data support. Before the policy is officially implemented, it predicts its possible market impact and changes in user behavior, assisting policymakers in optimizing and adjusting strategies and reducing implementation risks.

[0091] The core objective of the electricity forecasting model is to simulate the changing trends in user electricity consumption after policy adjustments. This embodiment uses two time series analysis methods, long short-term memory (LSTM) and autoregressive integrated moving average (ARIMA), combined with historical user electricity usage data to build a forecasting framework. LSTM is suitable for capturing long-term trends, such as changes in user electricity usage habits after electricity price adjustments, while ARIMA excels at accurately fitting short-term data and can model short-term fluctuations in the early stages of policy adjustments. Based on the characteristics of different users, the model integrates variables such as the price adjustment range, electricity elasticity coefficient, and seasonal load characteristics to analyze the impact of policy implementation on electricity consumption. For example, in the scenario of peak and valley electricity price adjustments, the system can predict whether users will actively shift their electricity load and adjust their electricity consumption structure, thereby optimizing their electricity expenses. For users who require precise load management, such as manufacturing enterprises, the model can simulate the arrangement of different production shifts to assess the impact of electricity price changes on business operations.

[0092] The load response model is used to simulate the user's adjustment of electricity consumption strategy after the electricity price is adjusted. This embodiment adopts reinforcement learning (RL) and game analysis methods. Reinforcement learning can train users' optimal electricity consumption strategies under different electricity price levels based on historical data and reward mechanisms, while game analysis is used to simulate the interaction between multiple user groups after electricity price changes. According to the load characteristics of different users, the system can construct response models at different levels, such as short-term adjustment strategies (load transfer) and long-term optimization strategies (equipment transformation, production adjustment). For example, in the demand response policy scenario, the system can simulate whether users will actively participate in load reduction plans to obtain economic compensation, or reduce power load during peak electricity price periods to reduce electricity costs. For industrial users, the model can analyze whether electricity price fluctuations will prompt enterprises to adopt energy storage equipment or optimize production processes to cope with long-term policy adjustments.

[0093] The electricity cost impact model is used to quantify the changes in electricity costs for different user groups due to policy adjustments. This embodiment uses multiple regression analysis and decision tree regression methods, combined with electricity forecast results and policy parameters, to evaluate the economic impact of policy implementation. The model can set a variety of billing methods, including billing based on maximum demand, billing based on transformer capacity, time-of-use electricity price billing, etc., to analyze the trend of electricity price changes after policy adjustments. For example, in the case of peak and valley electricity price adjustments, the system can predict whether the company will reduce its overall electricity expenditure due to load curve adjustments, or reduce electricity expenditure through peak shaving and valley filling strategies under demand response subsidy policies, thereby increasing policy benefits. In addition, the model can also evaluate long-term policy impacts, such as whether electricity price adjustments will prompt users to invest in energy-saving equipment or distributed energy, thereby changing electricity consumption patterns.

[0094] This embodiment accurately previews the changing trends of electricity consumption, load, and electricity prices before policy implementation, helping policymakers optimize policy parameters, improve implementation effectiveness, and reduce market uncertainty. Furthermore, the system can provide appropriate response strategies for different user groups, helping them optimize their electricity usage structure, improve energy efficiency, and achieve collaborative optimization of the electricity market.

[0095] This embodiment clearly presents the results of simulation calculations and intelligent deductions through multi-dimensional data visualization functions, intuitively demonstrating the impact of policy adjustments. The power consumption, electricity price change trends and load curve comparison charts can accurately reflect the changes in grid load and user electricity price fluctuations under different policy scenarios. Using efficient graphics engines such as Tableau and Power BI, the system converts large-scale data into easy-to-understand interactive charts, allowing users to dynamically view and compare the effects of different policy options. The user impact distribution map combines geographic information systems (GIS) and heat map technology, based on power data and regional characteristics, to show the specific impact of policy adjustments on different user groups, helping decision makers to make accurate assessments between different regions and user categories.

[0096] This embodiment uses natural language generation (NLG) technology and an automated report generation platform to automatically extract key data and conclusions from simulation results and generate a structured intelligent report.

[0097] In order to deeply evaluate the comprehensive impact of power grid policies on load characteristics, user behavior and system operation efficiency, this embodiment adopts a full-process analysis framework covering data fusion, feature modeling, causal deduction and effect attribution. Starting from multi-source heterogeneous data, this framework constructs a standardized data set, extracts key features and identifies multi-factor causal structures, and combines knowledge graph construction and machine learning modeling to quantify the action paths and response strengths of various influencing factors. At the same time, the XGBoost model is integrated to realize multi-step cascade deduction and interpretation of causal paths, and with the help of association rule mining technology, the key effect attribution after policy changes is revealed, providing an accurate and traceable decision support foundation for power grid business optimization, policy evaluation and dispatch strategy adjustment. The system has opened up the entire chain from data perception, mechanism modeling to intelligent deduction and feedback optimization, effectively improving the responsiveness and governance level of the power system under complex business environments.

[0098] The implementation and optimization of various services within power grid companies are often influenced by multiple factors. Key multi-source influencing factors include the amount of grid-connected photovoltaic power generation, changes in volume-pricing and fee policies, the progress of business expansion projects, and the frequency and duration of power outages. These factors not only act independently but also interact through complex relationships to influence the operational efficiency and economic viability of the power system. To better understand the interrelationships between these factors, it is necessary to extract key variables from large-scale data and identify their causal relationships. For example, grid-connected photovoltaic power generation can be affected by weather, seasonal variations, and electricity pricing policies. Changes in electricity pricing policies are closely related to user load demand, government policies, and market supply and demand conditions. Through data fusion and feature extraction, combined with statistical learning methods (such as causal inference and structural equation modeling), the causal chains between these factors can be identified.

[0099] This embodiment uses a causal inference algorithm to model the causal structure of multi-source data: using tools such as DoWhy or CausalImpact, causal relationships are identified in existing data to determine which factors directly or indirectly affect other factors; using Bayesian networks, Granger causality, or dynamic causal inference models, causal chains between various grid business factors are constructed; for example, changes in quantity, price, and fee policies may directly affect user load demand, and changes in load may in turn affect grid stability and the optimization of photovoltaic grid power generation; the identified causal structure will help understand the interaction between various factors in grid business and provide a scientific basis for subsequent decision-making.

[0100] Based on the results of causal structure identification, a knowledge graph of multiple influencing factors is constructed. The complex multi-source influencing factors and their interactions in the power grid business are presented in the form of a graph structure. A graph database (such as Neo4j) is used to construct the knowledge graph. In this graph, each influencing factor is used as a node, the causal relationship as an edge, and the nodes are connected by edges to form a complete graph. Figure 5 The associated impact map shown.

[0101] The knowledge graph formed in this embodiment will show the multi-dimensional impact path from photovoltaic grid-connected power generation, user electricity demand to electricity price policy and business expansion projects. The constructed graph can help power companies and policy makers to more intuitively understand the interaction and influence of various factors, support rapid decision-making and optimize business processes. In addition, by integrating real-time data streams, such as sensor data and market data, the knowledge graph can dynamically update the graph structure to ensure that the optimization decision of the power grid business responds to external environment and policy changes in a timely manner. Through graph query and analysis, decision makers can quickly simulate the impact of changes in various factors on the power grid business under different scenarios, and provide data support and decision-making basis for the dispatching, planning and policy making of the power market.

[0102] To simulate and deduce the dynamic response of key business variables in the power grid under the influence of complex multiple factors, this embodiment integrates the XGBoost model to construct a causal deduction model of influencing factors. This model uses the identified causal relationship path as prior knowledge, characterizes the nonlinear response mechanism between variables through machine learning modeling, and implements multi-step cascade calculation of the causal path and result traceability. The specific modeling process includes:

[0103] (1) Causal path deconstruction: The causal chain in the knowledge graph is split into a set of directed local path units. Each path is represented as "factor A → factor B", where A is the upstream driving factor (such as policy adjustment, load fluctuation, etc.) and B is the downstream response variable (such as grid stability, failure rate, etc.);

[0104] (2) Sample construction and feature preparation: Combine historical business data to extract time series samples for each causal path node, and construct derivative features such as control variables, lag terms, and interaction terms to capture the dynamics and nonlinear structures in the causal mechanism;

[0105] (3) Model training and evaluation: The XGBoost model is used to train the causal mapping relationship on each path, and cross-validation is used to evaluate the accuracy and adjust the parameters to obtain a stable and reliable function model;

[0106] (4) Importance attribution and path weighting: Apply the SHAP algorithm to interpret the model output, measure the marginal contribution of each input variable to the target variable, and achieve quantitative analysis of the causal path strength;

[0107] (5) Multi-step cascade deduction: Cascade single-path models to construct the overall reasoning logic on the causal graph, supporting multi-level response prediction under any upstream variable disturbance.

[0108] This embodiment effectively simulates the cascading impacts of policy changes (e.g., a 5% electricity price increase, the elimination of time-of-use pricing, and changes in peak and off-peak periods) on residential load, grid stability, and power outages, outputting the magnitude of change and response latency for each variable. Combined with a graph database interface, it dynamically updates variable states and inference paths, supporting application scenarios such as fault tracing, strategy evaluation, and operational optimization, significantly improving the real-time, accuracy, and interpretability of grid decision-making.

[0109] In order to accurately identify the actual effects of policy implementation and clarify the impact of multiple factors on key business indicators, such as Figure 6 As shown, this embodiment constructs an attributable and quantifiable analysis mechanism and an effect attribution and quantification model based on association rule mining; focusing on key outcome variables in power grid business, such as changes in user electricity consumption behavior, changes in grid loss after photovoltaic access, and fluctuations in electricity recovery rates under differentiated electricity price policies, by systematically mining potential variable combination patterns in large-scale power grid business data, extracting association rules of "variable combination-business results", revealing the impact path, and supporting the interpretable evaluation of policy effects.

[0110] The effect attribution and quantification model constructed based on association rule mining includes:

[0111] (1) Sorting out multi-source influencing factors: Focusing on specific policy implementation scenarios (such as time-of-use electricity price adjustment, load transfer guidance, business expansion process optimization, etc.), the system collects influencing variables that may have an effect, including multiple dimensions such as grid operation status, user behavior, market price, and environmental conditions, to form structured input samples. Each sample records the status information before and after policy implementation, laying the foundation for subsequent mining.

[0112] (2) Feature clustering and variable reduction: Clustering algorithms (such as K-means and hierarchical clustering) are used to cluster sample features and extract representative behavioral patterns. At the same time, variables are reduced with the help of information gain, variance screening and other methods to eliminate redundant and invalid features, ensuring the effectiveness and computational efficiency of the model when mining rules.

[0113] (3) Association rule mining and rule construction: Using the Apriori or FP-Growth algorithm, strong association rules between variable combinations and outcomes that meet the minimum support and minimum confidence conditions are mined from the sample set. Each rule is presented in the form of "if {variable combination}, then {result changes}". The significance and directionality of the rule are evaluated by combining indicators such as Lift and Conviction, and ultimately a rule library is constructed to attribution the policy execution results.

[0114] (4) Impact quantification and attribution evaluation: Model the attribution path of the rule base and quantify the marginal impact of each variable combination on the result variable. Introduce an attribution scoring mechanism (such as standardized impact factor, relative effect ratio) to evaluate the impact of each indicator. Figure 7 The quantitative scoring shown clarifies the degree of influence of key influencing factors on the change in results, and can generate multi-path attribution diagrams based on specific business scenarios to assist in reviewing the effectiveness of policy implementation and generating optimization suggestions.

[0115] To enhance the application value of causal deduction models and effect attribution results, this embodiment builds an integrated visualization analysis module. Using graph computing, visualization rendering, and interactive analysis technologies, this module provides business users with clear and intuitive analysis and presentation capabilities. It graphically presents the causal links and effect transmission paths between multiple influencing factors, helping decision makers quickly identify the impact of policy implementation on key business indicators and their propagation mechanisms. Through relationship visualization, indicator tracking, and contextual comparison, this module improves the interpretability and operability of causal modeling results in power grid operations. Specifically, it includes:

[0116] (1) Causal chain display: Based on the constructed knowledge graph and causal structure, a graph database (Neo4j) is combined with a front-end visualization framework (such as D3.js or Cytoscape.js) to generate a correlation impact star map, highlighting the direct / indirect impact paths between variables, and supporting node expansion, path highlighting, and impact tracing operations;

[0117] (2) Presentation of attribution results: Using association rule mining and attribution model analysis results, generate a ranking chart of the contribution rate of key factors to changes in business indicators and a heat map of the impact intensity, and use components such as Echarts or Plotly to achieve dynamic chart switching;

[0118] (3) Multi-dimensional indicator evolution trend analysis: Focusing on the evolution of key grid business indicators (such as line loss rate, photovoltaic power curtailment rate, user satisfaction, etc.) in the time dimension before and after policy implementation, combining time series analysis charts, comparative bar charts, etc. to present changes in business effects;

[0119] (4) Strategy simulation and scenario deduction: Combined with the results of the causal deduction model, the predicted output under the change of specific input factors is visualized, showing the expected range and risk distribution of the target indicators under each scenario, supporting uncertainty presentation and sensitivity analysis.

[0120] Dynamic evaluation report generation relies on the aforementioned models and visualization components, and uses templated document building technology to achieve one-click report output. The system integrates a rule engine and a report rendering engine, and automatically summarizes the causal structure, attribution analysis, impact quantification, and key visualization charts based on the evaluation topics selected by the user (such as policy performance, electricity price adjustment, etc.), fills in the preset report template (based on the Jinja2 template engine), and generates structured PDF reports through tools such as WeasyPrint or PDFKit. The report content includes: execution background, data analysis summary, causal chain structure diagram, attribution result ranking, strategy deduction prediction results, dynamic trend analysis charts and comprehensive recommendations, etc. The system supports three modes of periodic scheduling generation, threshold trigger generation, and manual trigger generation to ensure that power grid companies can quickly obtain such information during actual operation management and policy implementation feedback. Figure 8 The structured, professional and intelligent analysis report shown.

[0121] This embodiment uses quantitative modeling and intelligent algorithms to explore the optimal configuration range of policy parameters, integrate the multi-dimensional parameter boundaries on the user side, grid side, and social side, construct a multi-objective optimization model, and use the differential evolution algorithm to globally search for the Pareto optimal solution set; combines dynamic weight allocation and constraint penalty mechanisms to balance conflicting objectives such as investment cycle, grid revenue, and social benefits; ensures the reliability of the solution through sensitivity analysis, extreme scenario simulation, and economic verification, visualizes the multi-dimensional parameter associations with three-dimensional heat maps and parallel coordinate graphs, generates three types of strategic solutions: aggressive / robust / transitional, and embeds elastic adjustment and risk monitoring mechanisms to achieve scientific tuning and dynamic adaptation of policy parameters.

[0122] Integrating multi-dimensional parameters of power grid business is the basis for conducting spatial deductions for policy optimization. The integrated multi-dimensional parameters of power grid business mainly come from three aspects: the user side, the grid side, and the social side. Among them, the user side mainly includes electricity load, electricity consumption behavior characteristics, demand response potential, etc., which help to accurately capture the user's electricity consumption pattern and response strategy; the grid side involves factors such as grid operation parameters, grid load, supply and demand balance, and electricity market price fluctuations, reflecting the economic and stability of the grid; the social side involves macroeconomic indicators, policy adjustment background, social benefit indicators, etc., which are crucial for the long-term impact and comprehensive evaluation of policies. Various data sources are diverse and complex. The use of automated scripts and standardized data frameworks can uniformly classify, convert, and match data from different sources, ensuring that data of different dimensions can be processed and analyzed on the same platform.

[0123] After completing the data integration, in order to ensure the availability and consistency of the data, this embodiment uses an automated process to preprocess the data; the original data is denoised, deduplicated and missing values ​​are filled through data cleaning technology to solve the quality differences of data from different sources; secondly, interpolation, mean filling and other technologies are used to fill in the missing data to ensure the integrity and reliability of the data. Then, for multi-dimensional data, standardization processing methods are applied, including Min-Max standardization and Z-score standardization, to ensure that data of different dimensions can be effectively compared and integrated in the same model. In addition, this embodiment uses data conversion technology (such as discretization, classification coding, etc.) to convert unstructured data (such as text, images, etc.) into structured data, thereby improving the efficiency and accuracy of data analysis. To ensure the efficiency of the data preprocessing process, the solution utilizes the Pandas and NumPy libraries in Python, and combines machine learning platforms such as TensorFlow and Scikit-learn to ensure that the preprocessed data can seamlessly connect to the subsequent optimization model, providing a stable and reliable analysis basis.

[0124] In order to find the optimal policy parameter configuration scheme in a complex decision space, this embodiment constructs a multi-objective optimization model, covering key conflicting objectives such as the power grid investment recovery period, power system revenue, and social and economic benefits. Each objective function is based on the integrated multi-dimensional data-driven modeling of users, power grids, and social sides to ensure that the optimization objectives are business representative and have realistic constraints. Due to the significant conflicts between objectives, traditional optimization methods are prone to fall into local optimality and it is difficult to fully weigh the demands of all parties. Therefore, this embodiment adopts the differential evolution algorithm (DE) as the core optimization tool, which has strong global search capabilities and parameter optimization efficiency. It can effectively construct a Pareto optimal solution set covering a wide solution space and support the feasibility analysis of multiple strategy paths.

[0125] The differential evolution algorithm is based on the population and iteratively generates new solutions. It dynamically explores the optimal solution area in the solution space through the "mutation-crossover-selection" mechanism. This embodiment introduces a dynamic weight allocation strategy in the model, and adjusts the weight according to the relative importance of the actual operation goals, thereby enhancing the flexibility and business relevance of the strategy search. At the same time, in order to ensure that the obtained solution meets various boundary conditions and actual constraints, the model integrates a constraint penalty mechanism (penalty function method), embeds the penalty cost of infeasible solutions into the fitness function, and guides the algorithm away from non-compliant areas; during the optimization process, the DEAP library and SciPy optimization module in Python are combined to implement algorithm scheduling, and the iteration efficiency is improved through GPU acceleration; the Pareto front covering three types of strategies, namely radical, robust and transitional, is automatically generated as the core input for subsequent decision-making layers to conduct strategy selection, sensitivity analysis and scenario evaluation; through the introduction of the differential evolution algorithm, the depth and breadth of policy optimization space exploration are significantly improved, laying a solid foundation for multi-objective coordination and elastic tuning.

[0126] In the multi-objective optimization modeling process, maximizing grid revenue, minimizing investment payback period, and enhancing social benefits often conflict with each other, making it difficult to achieve overall optimization through static target setting. Therefore, this embodiment introduces a dynamic weight allocation mechanism to construct an adjustable and perceptible objective function system to adapt to different decision preferences and operational scenarios, achieving dynamic balance and optimized coordination among multiple objectives. Based on the Weighted Sum Model (WSM) framework, multiple conflicting objective functions are integrated into a comprehensive evaluation function with adjustable weights. This embodiment adjusts the priority of each objective in real time based on business feedback, policy guidance, or operational scenario requirements. For example, during periods of resource shortage, the weights of "grid economic benefits" and "load peak shaving capability" can be automatically increased; while in the medium- and long-term planning stages, the weights of "investment payback period" and "social benefits" are more emphasized. In addition, this embodiment embeds the entropy weight method and the improved Analytic Hierarchy Process (AHP) to automatically generate weight factors in a data-driven manner in the absence of subjective preferences, improving the objectivity and adaptability of the optimization process.

[0127] At the same time, in order to enhance the global adaptability of the algorithm, the dynamic weight change process is combined with the individual fitness feedback results of the differential evolution algorithm for self-learning update. This embodiment adopts local search and global jump strategies to run alternately to avoid falling into local optimality; the multi-objective optimization tool library in Python (such as Platypus and PyMOO) is used to realize dynamic weight configuration, and the changing trend of the objective function is monitored in a rolling manner, and the migration path of the Pareto frontier surface as the weight changes in real time is displayed through the visualization component.

[0128] By introducing a dynamic weighting mechanism, this embodiment can maintain the stability, flexibility, and business coupling of strategic solutions in a complex power policy environment with multiple conflicting objectives, provide flexible support for generating three types of regulatory strategies: radical, robust, and transitional, and improve the practicality and feasibility of policy deduction.

[0129] After completing the multi-objective optimization, this embodiment introduces two methods: extreme scenario simulation and sensitivity analysis to perform system verification and robustness testing on the generated policy parameter configuration scheme: Specifically,

[0130] Extreme scenario simulation tests the performance boundaries of various optimization schemes under extreme conditions by constructing a variety of high-risk and high-uncertainty operating scenarios. The simulation scenarios cover typical risk events such as severe load fluctuations, sudden drops in renewable energy output, and abnormal jumps in electricity prices. Test inputs are generated by combining historical boundary data with forward-looking risk assumptions. This embodiment uses the Monte Carlo method and Bootstrap resampling technology to construct multiple groups of extreme sample inputs and conduct stress testing in the full parameter space. By evaluating the response changes of each strategy scheme in indicators such as grid revenue and responsiveness, high-risk fragile solutions and resilient robust solutions are identified, and based on this, a strategy effectiveness classification standard is established to verify the feasibility and shock resistance of the recommended scheme in a complex real-world environment.

[0131] After completing the extreme scenario simulation, this embodiment conducts a sensitivity analysis to quantitatively identify and attribute the mapping relationship between the main input parameters and the key objective function; a combination of single-factor disturbance and global disturbance is used to evaluate the impact of core variables such as demand response participation rate, distributed output ratio, electricity price cap, and investment subsidy coefficient on the system target output under small disturbances; the Sobol sensitivity index and variable contribution analysis are used to establish a parameter impact priority ranking, identify the core parameter set that drives the policy tuning effect, and assist policy makers in focusing on optimization priorities.

[0132] Through the dual verification mechanism of extreme scenarios and sensitivity, this embodiment realizes the robustness test of the policy optimization strategy scheme under the "worst case" and reveals the sensitive path under "boundary disturbance". It can effectively verify the scheme and ensure that the final output policy parameters are operational, risk-resistant and have regulatory space, laying a solid foundation for subsequent policy classification output and elastic mechanism embedding.

[0133] To achieve an intuitive understanding of the high-dimensional policy parameter space and optimization results, this embodiment constructs a multidimensional data visualization module to present the model output results in a structured manner: First, a three-dimensional heat map is used to display the nonlinear response relationship between key policy variables and the objective function, depicting the impact path of policy parameter changes on indicators such as returns, cycles, and response rates; second, a parallel coordinate map is used to map the high-dimensional Pareto optimal solution set, revealing the mutual checks and balances between multiple objectives and facilitating the analysis of the relative trade-offs between strategies. The system also integrates a dynamic interactive interface that supports users to conditionally filter, conduct comparative analysis, and observe trends within specific parameter intervals, improving the efficiency and accuracy of strategy insights.

[0134] Based on the optimization results and visual analysis, this embodiment constructs a strategy clustering model and divides the Pareto solution set into three categories according to target preference and stability performance: radical (emphasizing profit maximization), robust (taking into account multi-objective balance) and transitional (flexible adjustment between different objectives); under each type of strategy, representative parameter configurations are extracted to generate structured policy solutions, and applicable boundaries, expected benefits and implementation suggestions are given to form a policy recommendation list that is comparative, practical and adaptable, providing a quantitative basis and decision-making support for grid-side policy optimization.

[0135] This example uses the calculation and deduction of the impact of increasing valley electricity prices on "peak shaving and valley filling" as an example to provide a detailed introduction to electricity price deduction:

[0136] By scientifically dividing peak and valley periods and rationally setting price differentials, users can effectively optimize their electricity usage habits, adjust production plans, and improve electricity efficiency. Previewing the implementation of valley electricity pricing policies is an important basis for evaluating the effectiveness of policy applications before implementation. This requires detailed period division and electricity price setting based on multiple factors, such as specific electricity usage and load characteristics, as well as simulation and calculation deduction, to ensure that electricity pricing policies are scientific and effective and achieve the desired results.

[0137] A systematic data collection and processing process is established around the data resources required for deep valley electricity price deduction. Hourly load data, electricity price structure, electricity bills, metering methods, voltage levels and other key indicator information of typical residential and industrial and commercial users are extracted from the data middle platform and load management platform, and data is pulled by combining interface calls with batch scripts. After the data is stored in the database, unified format processing and standardized cleaning operations are carried out based on the Python and SQL tool chains, including filling in missing time series data, unit conversion, outlier removal, electricity price formula standardization, etc., to ensure the consistency and reliability of subsequent simulation inputs. In order to keep the analysis results consistent with the existing business environment, a business logic mirror model is built on the basis of the simulation test platform, covering modules such as user identification, electricity price strategy call, deduction formula calculation and response output, to ensure that the test output has a foundation for engineering implementation.

[0138] This embodiment sets three simulation scenarios around the goal of "increasing deep valley periods" and converts the policy of "increasing deep valley electricity prices" into structured data. Based on the existing peak-valley and flat-valley electricity prices, a new deep valley period (0:00-4:00) is added, and the electricity prices fluctuate by 10%, 20%, and 30% respectively on the basis of the existing valley period. To evaluate the effectiveness of the policy, a group of representative users are selected as the analysis objects, including small and medium-sized industrial and commercial users with a high proportion of load during peak hours, users in substations with adjustable load characteristics, and residential users who enter the second-tier electricity price. By setting filtering rules and data label screening, the analysis objects are clarified to ensure the applicability of the evaluation coverage and deduction effect.

[0139] During the simulation implementation process, two model paths, "estimation simulation" and "actuarial simulation", were constructed to support the multi-dimensional analysis of the impact of valley electricity price adjustment. In the estimation path, a parameter adjustment module was first built to support users to flexibly set the ratio of valley electricity consumption to valley electricity consumption, such as Figure 9 As shown, the load transfer behavior under different scenarios is simulated; through the deduction formula editing interface, a simplified power migration calculation logic is set, such as sinking a certain proportion of valley load to the deep valley, such as Figure 10 As shown, calculate the corresponding electricity price changes.

[0140] This embodiment automatically outputs the estimation results, such as Figure 11 Load shifting and customer bill impacts under different adjustment scenarios are shown.

[0141] The actuarial path is similar to the estimation in structure, but it emphasizes multi-parameter and refined modeling. First, in the parameter adjustment module, the peak, peak, flat peak, valley, deep valley and other full-time electricity factors are introduced. Users can flexibly configure the proportion of each time period according to the business scenario, simulating Figure 12 In the deduction formula editing, the bill-level actuarial logic is constructed, based on the actual electricity price and user time-of-use electricity data, and refined to Figure 13 The electricity charges for each time period are calculated item by item as shown in the figure. This embodiment combines the calculation engine output as shown in the figure. Figure 14 The actuarial results shown can fully reflect the changes in electricity charges for users with different electricity composition structures caused by the deep-valley electricity price adjustment, and support more targeted policy response analysis.

[0142] In terms of processing the deduction results, an intelligent analysis algorithm is introduced to conduct a systematic analysis of the simulation data. First, the electricity charge changes and load adjustment amplitudes of each user are input into the analysis model to generate a three-dimensional feature matrix of "electricity price impact-load response-electricity charge changes"; then, a clustering algorithm is used to group users by load response behavior characteristics, and association rule mining is combined to identify high-response willingness groups; finally, based on the actuarial results, users who are directly affected by the valley electricity price adjustment are identified, and characteristic hierarchical analysis is carried out in combination with electricity consumption data, including grouping by the amount of electricity charge increase, classification by user type, and mining load response characteristics through association analysis. On this basis, differentiated guidance strategies are proposed: such as recommending that key production enterprises optimize their scheduling plans, residential users adjust their electricity usage habits, and promote orderly electricity consumption points reward mechanisms. The deduction results show that after the implementation of the strategy, the peak load can be reduced by an average of about 20%, and the valley power consumption can be increased by an average of about 40%, effectively supporting such as Figure 15 The regulatory role of electricity price policy in "peak shaving and valley filling" is shown.

[0143] This embodiment integrates an advanced visualization engine and a dynamic analysis template, supports real-time rendering of the deduction results in the form of "user distribution map + response behavior trajectory + strategy recommendation list", and automatically generates a user behavior adjustment report and a regional overall peak-saving effect evaluation report. This provides decision support for policy makers and grid operators, further improving the implementation effect of deep valley electricity price policies and the level of user-side interaction; effectively quantifies the impact of different deep valley electricity price strategies on user load patterns and electricity bill expenditures, and verifies the actual value of introducing a lower deep valley electricity price mechanism in stimulating user load transfer potential. By constructing a deduction system that mirrors the real environment, a standardized and replicable measurement and analysis process has been formed, providing data support and technical paths for subsequent refined optimization of electricity price mechanisms, grouped user response guidance, and dynamic balancing of peak and valley loads.

[0144] The electricity bill settlement simulation application proposed by Su in this embodiment helps power companies predict risks in advance, avoid price fluctuations, and ensure revenue stability. Simulations based on real electricity data will further enhance the accuracy and authority of policy simulations, adapt to simulation calculations in various complex scenarios, and significantly reduce the workload of manual offline analysis. Accurate price and fee policy simulations can promote the rational allocation of power resources, improve energy efficiency, reduce resource waste, promote energy conservation and emission reduction, and make positive contributions to environmental protection. Furthermore, transparent electricity price information enhances public trust in the electricity market and helps build a harmonious and transparent electricity consumption environment.

[0145] Example 2

[0146] The second embodiment of the present invention introduces an electricity price deduction system based on a simulation environment.

[0147] like Figure 16 The electricity price deduction system based on the simulation environment shown includes:

[0148] An acquisition module configured to acquire user electricity usage data;

[0149] A construction module configured to construct a deduction data model based on the acquired electricity consumption data;

[0150] an adjustment module configured to dynamically adjust the parameters of the constructed deduction data model to obtain a power regulation parameter library;

[0151] A simulation module is configured to perform simulation calculations of load electricity prices under different deduction data model parameter scenarios based on the obtained power control parameter library and generate a dynamic evaluation report;

[0152] The deduction module is configured to construct a multi-objective optimization model for electricity price deduction based on the generated dynamic evaluation report, taking into account the multi-dimensional parameter boundaries of the user side, the grid side and the social side; solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate a deduction data parameter configuration plan, and complete the electricity price deduction based on the simulation environment.

[0153] The detailed steps are the same as the electricity price deduction method based on the simulation environment provided in Example 1, and will not be repeated here.

[0154] Example 3

[0155] A third embodiment of the present invention provides a computer-readable storage medium.

[0156] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps of the electricity price deduction method based on a simulation environment as described in the first embodiment of the present invention.

[0157] The detailed steps are the same as the electricity price deduction method based on the simulation environment provided in Example 1, and will not be repeated here.

[0158] Example 4

[0159] A fourth embodiment of the present invention provides an electronic device.

[0160] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, the steps of the electricity price deduction method based on a simulation environment as described in Example 1 of the present invention are implemented.

[0161] The detailed steps are the same as the electricity price deduction method based on the simulation environment provided in Example 1, and will not be repeated here.

[0162] Example 5

[0163] A fifth embodiment of the present invention provides a computer program product.

[0164] A computer program product includes software code, wherein the program in the software code executes the steps of the electricity price deduction method based on a simulation environment as described in the first embodiment of the present invention.

[0165] The detailed steps are the same as the electricity price deduction method based on the simulation environment provided in Example 1, and will not be repeated here.

[0166] It will be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0167] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0168] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0169] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0170] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0171] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

[0172] The above description is merely a preferred embodiment of this embodiment and is not intended to limit this embodiment. Those skilled in the art will readily appreciate that this embodiment may be modified and varied in various ways. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this embodiment shall be within the scope of protection of this embodiment.

Claims

1. A method for estimating electricity prices based on a simulation environment, characterized in that: include: Obtain user electricity consumption data; Build a deduction data model based on the acquired electricity consumption data; Dynamically adjust the parameters of the constructed deduction data model to obtain the power control parameter library; Based on the obtained power control parameter library, simulate the load electricity price under different deduction data model parameter scenarios and generate a dynamic evaluation report; Based on the generated dynamic assessment report, a multi-objective optimization model for electricity price deduction is constructed by considering the multi-dimensional parameter boundaries of the user side, the grid side, and the social side; Solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate deduction data parameter configuration plan, and complete electricity price deduction based on simulation environment.

2. The method for estimating electricity prices based on a simulation environment as claimed in claim 1, characterized in that: In the process of the multi-objective balance optimization, the differential evolution method is combined to perform global search and parameter optimization on the constructed electricity price deduction multi-objective optimization model, and self-learning updates of dynamic weight changes are performed according to the individual fitness feedback results of the differential evolution method. Alternating local search and global jumps are used to avoid falling into local optimality, and dynamic weight allocation is used to perform rolling monitoring of the change trend of the objective function of the multi-objective optimization model to obtain a deduction data parameter configuration plan.

3. The method for estimating electricity prices based on a simulation environment as claimed in claim 2, characterized in that: In the process of rolling monitoring of the changing trend of the objective function of the multi-objective optimization model using dynamic weight allocation, the penalty cost of the infeasible solution in the objective function is embedded into the objective function in combination with the constraint penalty mechanism, and the constructed electricity price deduction multi-objective optimization model is dynamically solved to obtain the optimal solution area of ​​the multi-objective optimization model.

4. The method for estimating electricity prices based on a simulation environment as claimed in claim 1, characterized in that: During the load electricity price simulation calculation process, the obtained power control parameters are dynamically evaluated under the adjustment methods of fixed value adjustment, proportional adjustment, time period adjustment and step adjustment, and the impact of the adjustment on the grid load distribution, user electricity costs and the competitive landscape of the power market is determined.

5. The method for estimating electricity prices based on a simulation environment as claimed in claim 1, characterized in that: The deduction data model parameters include at least electricity price adjustment parameters, electricity rate policy parameters, demand response subsidy parameters and distribution network investment parameters.

6. The method for estimating electricity prices based on a simulation environment as claimed in claim 1, characterized in that: The user electricity consumption data obtained includes at least the monthly electricity consumption and electricity fee information of the metering point, the monthly electricity consumption and electricity fee data of classified users, the 96-point collected electricity data, the electricity price model data and the power factor assessment coefficient; Before building the deduction data model, the acquired user electricity consumption data is sequentially cleaned and converted into a data format.

7. An electricity price deduction system based on a simulation environment, characterized in that: include: An acquisition module configured to acquire user electricity usage data; A construction module configured to construct a deduction data model based on the acquired electricity consumption data; an adjustment module configured to dynamically adjust the parameters of the constructed deduction data model to obtain a power regulation parameter library; A simulation module is configured to perform simulation calculations of load electricity prices under different deduction data model parameter scenarios based on the obtained power control parameter library and generate a dynamic evaluation report; The deduction module is configured to construct a multi-objective optimization model for electricity price deduction based on the generated dynamic evaluation report, taking into account the multi-dimensional parameter boundaries of the user side, the grid side and the social side; solve the constructed multi-objective optimization model for electricity price deduction, combine dynamic weight allocation and constraint penalty mechanism to perform multi-objective balance optimization, generate a deduction data parameter configuration plan, and complete the electricity price deduction based on the simulation environment.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method for deducing electricity price based on a simulation environment as described in any one of claims 1 to 6 are implemented.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the steps of the electricity price deduction method based on a simulation environment as described in any one of claims 1 to 6 are implemented.

10. A computer program product comprising software code, characterized in that The program in the software code executes the steps of the electricity price deduction method based on a simulation environment as described in any one of claims 1 to 6.

Images