Large industrial user demand regulation and control side response method considering multiple time scales

Abstract

The invention discloses a large industrial user demand regulation and control side response method considering multiple time scales, and belongs to the technical field of power system regulation and control. In order to solve the problem that the regulation and control efficiency and precision are difficult to improve by only considering the response of a single time scale, the method comprises the following steps: data segment acquisition, demand quantity prediction, regulation and control quantity calculation, strategy formulation and strategy regulation and control, and through a large industrial user demand regulation and control side response method considering multiple time scales, the user demand can be more accurately predicted, and the regulation and control efficiency and precision can be improved. According to the method, a prediction result is obtained, a corresponding regulation and control strategy is formulated according to the prediction result, so that the requirements of complex and changeable large industrial users are better met, meanwhile, multi-time-scale collection can provide more flexibility for formulating the regulation and control strategy, different regulation and control strategies can be formulated according to different time scales, regulation and control are more flexible and effective, and the regulation and control efficiency is improved. Meanwhile, the regulation and control effect can be monitored and evaluated in real time, and a basis is provided for further optimizing a regulation and control strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of power system regulation, and in particular to a large industrial user demand regulation-side response method considering multiple time scales. Background Art

[0002] Electricity demand response refers to the short-term behavior of electricity users who temporarily change their electricity usage based on price or incentives, reducing (increasing) their electricity use when electricity market prices rise (fall) significantly or when system safety and reliability are at risk. This behavior promotes a balance between electricity supply and demand, ensures stable grid operation, and curbs price increases. Simply put, demand-side response is the behavior of users who respond to the grid's call and temporarily adjust their electricity consumption (including both reductions and increases) in a planned manner, thereby promoting power system stability. With the continuous development of modern energy systems, large industrial users are increasingly demanding the precision and speed of response required to regulate electricity demand.

[0003] Electricity demand is a complex and changeable process, affected by multiple factors, including climate, economy, and population changes. The mode and extent of these influencing factors may vary at different time scales. However, existing demand regulation methods often only consider responses at a single time scale and cannot meet the needs of large industrial users at different time scales, making it difficult to improve regulation efficiency and accuracy. Summary of the Invention

[0004] The purpose of the present invention is to provide a large industrial user demand regulation side response method considering multiple time scales to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a large industrial user demand regulation side response method considering multiple time scales, comprising the following steps:

[0006] Data segment collection: collects power demand data of large industrial users and divides the data into multiple data segments according to different time scales;

[0007] Demand forecasting: perform demand forecasting on each data segment to obtain the predicted demand at each time scale;

[0008] Calculation of control quantity: Calculate the control quantity for each time scale based on the difference between predicted demand and actual demand;

[0009] Strategy formulation: formulate control strategies according to the control amount and different time scales;

[0010] Strategic regulation: regulate the electricity demand of large industrial users according to the regulation strategy.

[0011] Furthermore, the data collection specifically includes the following steps:

[0012] When dividing data segments, divide them by day, month, year or finer time periods;

[0013] Divide the daily data into segments by day, dividing the 24 hours of a day into different time periods, including peak period, flat period and valley period, and count the power demand data in each time period;

[0014] Divide the monthly data into segments by month, dividing a month into different time periods, including the beginning of the month, the middle of the month, and the end of the month, and collect statistics on the power demand data in each time period;

[0015] Divide the annual data into segments by year, dividing the year into different time periods, the annual data segments including the flood season, the normal water season, and the dry season, and collect statistics on the power demand data in each time period;

[0016] Among them, according to actual needs, the data is divided into finer time periods, including dividing the data segments by hour or by week.

[0017] Furthermore, the demand forecasting specifically includes the following steps:

[0018] Data preprocessing, preprocessing the original data in the data segment;

[0019] Feature selection, selecting features related to electricity demand, including weather, season, and holidays;

[0020] Construct a model using a linear regression model based on historical electricity demand data and characteristics in the data segment to build a multi-time scale demand model, including long-term, medium-term, and short-term demand forecast models, and evaluate the model using a test dataset, including error analysis and model optimization.

[0021] Forecasting, using the trained linear regression model, predict the demand for each power data segment in the future.

[0022] Furthermore, the data preprocessing specifically includes the following steps:

[0023] Data import, reading the original data in the data segment, and performing data preprocessing on the original data, wherein the data preprocessing includes format conversion of the original data and unification of data types;

[0024] Data denoising: detect and process outliers in raw data, fill in missing data based on the raw data, remove noise data by setting filtering conditions, and standardize the data to make it fall within a certain range;

[0025] Data cleaning is to correct the erroneous, duplicate or incorrectly formatted data in the original data after denoising, delete the duplicate or irrelevant data in the original data, remove redundancy, reduce the amount of data, and classify the cleaned original data according to the attribute characteristics of the data.

[0026] Furthermore, the control amount calculation specifically includes the following steps:

[0027] Calculate the forecast deviation, select a single time scale, determine the forecast power demand and the actual power demand within the time scale, calculate the difference between the forecast demand and the actual demand, and obtain the budget deviation, which is achieved by the following formula;

[0028] C 偏差 =C 预测 -C 实际 ;

[0029] Among them, C 偏差 represents the budget deviation, C 预测 represents the predicted power demand, C 实际 Indicates the actual power demand;

[0030] Calculate the control amount. The control amount is calculated based on the budget deviation. The calculation is achieved through the following formula:

[0031] Q=C 偏差 / t;

[0032] Among them, Q represents the control quantity, C 偏差 represents the budget deviation, and t represents the time scale.

[0033] Furthermore, the strategy formulation specifically includes the following steps:

[0034] Formulate corresponding regulatory strategies, including long-term stable regulatory strategies, medium-term dynamic regulatory strategies, and short-term rapid regulatory strategies;

[0035] When formulating regulatory strategies, different regulatory strategies should be formulated based on demand models at different time scales, actual operating conditions, policy requirements and other factors, while monitoring, evaluation, adjustment and optimization should be carried out according to actual conditions.

[0036] Furthermore, the long-term stable control strategy specifically includes the following methods:

[0037] Generation capacity planning, which plans future generation capacity and type based on forecasted power demand and load characteristics;

[0038] Optimize the energy structure, optimize the energy structure, and rationally utilize various types of renewable energy and fossil energy;

[0039] Network construction and renovation: construction and renovation of power network facilities based on predicted power demand and load characteristics.

[0040] Furthermore, the mid-term dynamic control strategy specifically includes the following methods:

[0041] Dispatch optimization: optimize power dispatch and balance the output of different regions and types of power generation facilities;

[0042] Demand response management, implementing demand response projects to encourage users to adjust their electricity usage behavior;

[0043] Construction of energy storage facilities: Construction and utilization of energy storage facilities to smooth out power load fluctuations.

[0044] Furthermore, the short-term rapid control strategy specifically includes the following methods:

[0045] Rapid dispatch, optimize power dispatch, ensure stable operation of the power system, and quickly respond to sudden changes in power demand;

[0046] Emergency demand response: In response to sudden changes in electricity demand, emergency demand response measures are initiated, including demand shifting and power rationing;

[0047] Reserve capacity management: Manage and utilize reserve capacity to ensure the stable operation of the power system and respond to sudden changes in power demand.

[0048] Furthermore, the strategy control specifically includes the following steps:

[0049] Communication and negotiation: Communicate and negotiate with large industrial users based on control strategies to understand their production plans and electricity usage arrangements, and jointly adjust control plans;

[0050] Implementing control measures, including adjusting electricity prices, restricting electricity use, and issuing dispatch instructions based on the negotiation results;

[0051] Monitoring and evaluation: monitor and evaluate the effects of regulation and control, and timely adjust and optimize regulation and control strategies.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The multi-time scale collection of the present invention can provide more comprehensive power demand information, making the formulation of control strategies more accurate and comprehensive. At the same time, the multi-time scale collection can provide more flexibility for the formulation of control strategies. By analyzing and processing data at different time scales, more information and rules can be extracted, providing a more scientific and effective basis for the formulation of control strategies. At the same time, it can also provide more references for the evaluation and optimization of strategies, making the control strategies more in line with actual needs.

[0054] 2. By analyzing and calculating budget deviations, this invention can more accurately predict power control quantities, improve the accuracy of power control, help ensure the stable operation of the power system, and reduce operational risks. It can also optimize the allocation of power resources, achieving reasonable allocation and efficient utilization of power resources. This helps improve the utilization efficiency of power resources and reduce electricity costs.

[0055] 3. The present invention formulates power control strategies based on the control quantity and different time scales, which can better respond to changes in power demand and improve the stability of power supply. At the same time, it can better optimize resource allocation, improve the utilization efficiency of power resources, better reduce operating costs, improve the benefits of power companies, and better enhance emergency response capabilities to deal with emergencies such as sudden incidents and natural disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] To address the technical problem that existing demand control methods often only consider the response of a single time scale and cannot meet the needs of large industrial users at different time scales, thus making it difficult to improve control efficiency and accuracy, please refer to Figure 1 , the present invention provides the following technical solutions:

[0059] The large industrial user demand regulation side response method considering multiple time scales includes the following steps:

[0060] Data segment collection: collects power demand data of large industrial users and divides the data into multiple data segments according to different time scales;

[0061] Demand forecasting: perform demand forecasting on each data segment to obtain the predicted demand at each time scale;

[0062] Calculation of control quantity: Calculate the control quantity for each time scale based on the difference between predicted demand and actual demand;

[0063] Strategy formulation: formulate control strategies according to the control amount and different time scales;

[0064] Strategic regulation: regulate the electricity demand of large industrial users according to the regulation strategy.

[0065] Specifically, by considering multiple timescales for large-scale industrial user demand control and response methods, user demand can be more accurately predicted, and corresponding control strategies can be formulated based on the prediction results, thereby better meeting the complex and changing needs of large-scale industrial users. Furthermore, the present invention can also formulate different control strategies based on different timescales, making control more flexible and effective, improving control efficiency and precision, and also allowing for real-time monitoring and evaluation of control effects, providing a basis for further optimizing control strategies.

[0066] Data collection specifically includes the following steps:

[0067] When dividing data segments, divide them by day, month, year or finer time periods;

[0068] Divide the daily data into segments by day, dividing the 24 hours of a day into different time periods, including peak period, flat period and valley period, and count the power demand data in each time period;

[0069] Divide the monthly data into segments by month, dividing a month into different time periods, including the beginning of the month, the middle of the month, and the end of the month, and collect statistics on the power demand data in each time period;

[0070] Divide the annual data into segments by year, dividing the year into different time periods, the annual data segments including the flood season, the normal water season, and the dry season, and collect statistics on the power demand data in each time period;

[0071] Among them, according to actual needs, the data is divided into finer time periods, including dividing the data segments by hour or by week.

[0072] Specifically, data collected at multiple time scales can provide more comprehensive information on electricity demand, making the formulation of regulation strategies more precise and comprehensive. It also provides greater flexibility in formulating regulation strategies. In the short term, regulation can be tailored to real-time electricity demand, ensuring stable power supply. In the medium term, electricity demand can be forecasted and planned for the coming weeks or months to adapt to economic and climate changes. In the long term, future development trends can be forecasted and planned to develop more sustainable electricity development strategies.

[0073] Acquiring data at multiple time scales also helps improve the scientific nature and effectiveness of regulatory strategies. By analyzing and processing data at different time scales, more information and patterns can be extracted, providing a more scientific and effective basis for formulating regulatory strategies. Furthermore, acquiring data at multiple time scales can provide more references for strategy evaluation and optimization, ensuring that regulatory strategies are more aligned with practical needs.

[0074] Demand forecasting specifically includes the following steps:

[0075] Data preprocessing, preprocessing the original data in the data segment;

[0076] Feature selection, selecting features related to electricity demand, including weather, season, and holidays;

[0077] Construct a model using a linear regression model based on historical electricity demand data and characteristics in the data segment to build a multi-time scale demand model, including long-term, medium-term, and short-term demand forecast models, and evaluate the model using a test dataset, including error analysis and model optimization.

[0078] Forecasting, using the trained linear regression model, predict the demand for each power data segment in the future.

[0079] Specifically, using a linear regression model to predict electricity demand data can effectively predict future electricity demand trends and help companies formulate reasonable power dispatch plans. At the same time, the linear regression model can flexibly handle various types of input variables, including continuous variables, discrete variables, etc., to improve prediction accuracy. At the same time, the linear regression model has good interpretability, which can help companies better understand the factors affecting electricity demand.

[0080] Data preprocessing specifically includes the following steps:

[0081] Data import, reading the original data in the data segment, and performing data preprocessing on the original data, wherein the data preprocessing includes format conversion of the original data and unification of data types;

[0082] Data denoising: detect and process outliers in raw data, fill in missing data based on the raw data, remove noise data by setting filtering conditions, and standardize the data to make it fall within a certain range;

[0083] Data cleaning is to correct the erroneous, duplicate or incorrectly formatted data in the original data after denoising, delete the duplicate or irrelevant data in the original data, remove redundancy, reduce the amount of data, and classify the cleaned original data according to the attribute characteristics of the data.

[0084] Specifically, data cleaning of power demand data before analysis can make the data more accurate and complete, reduce errors and deviations, and improve the quality and credibility of the data. At the same time, the cleaned data is more standardized and standardized, which can facilitate subsequent linear regression model prediction work. Moreover, if there are a lot of noise, missing values, and outliers in the data, the data processing cost will increase. Data cleaning can remove these problems, reduce data processing costs, improve data accuracy, and improve the accuracy of data analysis.

[0085] The control amount calculation specifically includes the following steps:

[0086] Calculate the forecast deviation, select a single time scale, determine the forecast power demand and the actual power demand within the time scale, calculate the difference between the forecast demand and the actual demand, and obtain the budget deviation, which is achieved by the following formula;

[0087] C 偏差 =C 预测 -C 实际 ;

[0088] Among them, C 偏差 represents the budget deviation, C 预测 represents the predicted power demand, C 实际 Indicates the actual power demand;

[0089] Calculate the control amount. The control amount is calculated based on the budget deviation. The calculation is achieved through the following formula:

[0090] Q=C 偏差 / t;

[0091] Among them, Q represents the control quantity, C 偏差 represents the budget deviation, and t represents the time scale.

[0092] Specifically, by analyzing and calculating budget deviations, we can more accurately predict the amount of power regulation, improve the accuracy of power regulation, help ensure the stable operation of the power system, reduce operational risks, and optimize the allocation of power resources, achieving reasonable configuration and efficient utilization of power resources. This helps improve the utilization efficiency of power resources and reduce electricity costs.

[0093] Strategy formulation specifically includes the following steps:

[0094] Formulate corresponding regulatory strategies, including long-term stable regulatory strategies, medium-term dynamic regulatory strategies, and short-term rapid regulatory strategies;

[0095] When formulating regulatory strategies, different regulatory strategies should be formulated based on demand models at different time scales, actual operating conditions, policy requirements and other factors, while monitoring, evaluation, adjustment and optimization should be carried out according to actual conditions.

[0096] Among them, the long-term stable regulation strategy also includes the following methods:

[0097] Power generation capacity planning: planning future power generation capacity and types based on predicted power demand and load characteristics to meet long-term power demand; energy structure optimization: optimizing the energy structure, rationally utilizing various types of renewable energy and fossil energy, and reducing energy consumption and carbon emissions; network construction and transformation: constructing and transforming power network facilities based on predicted power demand and load characteristics to improve the stability and reliability of the power system.

[0098] The medium-term dynamic control strategy also includes the following methods:

[0099] Dispatching optimization: optimize power dispatching, balance the output of power generation facilities in different regions and types, and ensure the stable operation of the power system; demand response management: implement demand response projects to encourage users to adjust their electricity usage behavior and reduce power load fluctuations; energy storage facility construction: build and utilize energy storage facilities to smooth out power load fluctuations and improve the stability of the power system.

[0100] The short-term rapid regulation strategy also includes the following methods:

[0101] Rapid dispatch, optimize power dispatch, ensure the stable operation of the power system, and quickly respond to sudden changes in power demand; emergency demand response, initiate emergency demand response measures for sudden changes in power demand, and the response measures include demand transfer and power rationing; spare capacity management, manage and utilize spare capacity to ensure the stable operation of the power system and respond to sudden changes in power demand.

[0102] Specifically, formulating power regulation strategies based on the regulation amount and different time scales can better respond to changes in power demand and improve the stability of power supply. At the same time, it can better optimize resource allocation, improve the utilization efficiency of power resources, better reduce operating costs, improve the efficiency of power companies, and better enhance emergency response capabilities to deal with emergencies such as sudden incidents and natural disasters.

[0103] Strategy control specifically includes the following steps:

[0104] Communication and negotiation: Communicate and negotiate with large industrial users based on control strategies to understand their production plans and electricity usage arrangements, and jointly adjust control plans;

[0105] Implementing control measures, including adjusting electricity prices, restricting electricity use, and issuing dispatch instructions based on the negotiation results;

[0106] Monitoring and evaluation: monitor and evaluate the effects of regulation and control, and timely adjust and optimize regulation and control strategies

[0107] Specifically, through communication and consultation with large industrial users, we can better understand their production plans and electricity usage arrangements, thereby more accurately predicting electricity demand. This helps power companies balance power supply and demand and make scheduling arrangements in advance, avoiding power shortages or surpluses. At the same time, it can better utilize limited power resources and optimize resource allocation. It can also better understand users' actual needs and emergency situations, thereby better responding to emergencies such as sudden incidents and natural disasters.

[0108] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A large industrial user demand regulation side response method considering multiple time scales, characterized in that: The steps include: Data segment collection: collect power demand data of large industrial users and divide the data into multiple data segments according to different time scales; Demand forecasting: forecast demand for each data segment to obtain forecast demand at each time scale; Calculation of control quantity: Calculate the control quantity for each time scale based on the difference between predicted demand and actual demand; Strategy formulation: formulate regulation strategies according to the regulation amount and different time scales; Strategic control: regulate the electricity demand of large industrial users according to the control strategy.

2. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 1 is characterized by: The data collection specifically includes the following steps: When dividing data segments, divide them according to days, months, years or more detailed time periods; Divide the daily data into segments by day, and divide the 24 hours of a day into different time periods. The daily data segments include peak periods, flat periods, and valley periods, and count the power demand data in each time period; Divide the monthly data into segments by month, and divide a month into different time periods, wherein the monthly data segments include the beginning of the month, the middle of the month, and the end of the month, and count the power demand data in each time period; Divide the annual data into segments by year, and divide a year into different time periods, wherein the annual data segments include a flood season, a normal water season, and a dry season, and count the power demand data in each time period; According to actual needs, the data is divided into finer time periods, including dividing the data into hourly segments or weekly segments.

3. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 1 is characterized by: Demand forecasting specifically includes the following steps: Data preprocessing, preprocessing the raw data in the data segment; Feature selection, selecting features related to electricity demand, the features including weather, season, and holidays; Constructing a model, using a linear regression model, based on historical power demand data and characteristics in the data segment, constructing a multi-time scale demand model, including long-term, medium-term and short-term demand forecasting models, and using a test data set to evaluate the model, the evaluation including error analysis and model optimization; Forecasting, using the trained linear regression model, predicts the demand for each power data segment in the future.

4. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 3 is characterized by: The data preprocessing specifically includes the following steps: Data import, reading the original data in the data segment, and performing data preprocessing on the original data, wherein the data preprocessing includes format conversion of the original data and unification of data types; Data denoising: detect and process outliers in raw data, fill in missing data based on raw data, remove noise data by setting filtering conditions, and standardize data to make it fall within a certain range; Data cleaning, correcting the erroneous, duplicate or incorrectly formatted data in the original data after denoising, deleting duplicate or irrelevant data in the original data, removing redundancy, reducing the amount of data, and classifying the cleaned original data according to the attribute characteristics of the data.

5. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 1, characterized in that: The control amount calculation specifically includes the following steps: Calculate the forecast deviation, select a single time scale, determine the forecast power demand and the actual power demand within the time scale, calculate the difference based on the forecast demand and the actual demand, and obtain the budget deviation, which is achieved by the following formula; C 偏差 =C 预测 -C 实际 ; Among them, C 偏差 represents the budget deviation, C 预测 represents the predicted electricity demand, C 实际 Indicates the actual power demand; Calculate the control amount. The control amount is calculated based on the budget deviation. The calculation is implemented by the following formula; Q=C 偏差 / t; Among them, Q represents the control amount, C 偏差 represents the budget deviation, and t represents the time scale.

6. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 1, characterized in that: The strategy formulation specifically includes the following steps: Formulate corresponding control strategies, including long-term stable control strategies, medium-term dynamic control strategies and short-term rapid control strategies; When formulating regulation and control strategies, different regulation and control strategies are formulated based on demand models, actual operating conditions, and policy requirements at different time scales, while monitoring, evaluation, and adjustment and optimization are carried out according to actual conditions.

7. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 6, characterized in that: The long-term stable control strategy specifically also includes the following methods: Generation capacity planning, planning future generation capacity and type based on forecasted electricity demand and load characteristics; Optimize the energy structure, optimize the energy structure, and rationally utilize various types of renewable energy and fossil energy; Network construction and renovation: construction and renovation of power network facilities based on predicted power demand and load characteristics.

8. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 6, characterized in that: The mid-term dynamic control strategy specifically also includes the following methods: Dispatch optimization: optimize power dispatch and balance the output of different regions and types of power generation facilities; Demand response management, implementing demand response projects to encourage users to adjust their electricity usage behavior; Construction of energy storage facilities: Construction and utilization of energy storage facilities to smooth out power load fluctuations.

9. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 6, characterized in that: The short-term rapid control strategy specifically also includes the following methods: Rapid dispatch, optimize power dispatch, ensure stable operation of the power system, and quickly respond to sudden changes in power demand; Emergency demand response: In response to sudden changes in electricity demand, emergency demand response measures are initiated, including demand shifting and power rationing; Reserve capacity management: Manage and utilize reserve capacity to ensure the stable operation of the power system and respond to sudden changes in power demand.

10. The large industrial user demand regulation side response method considering multiple time scales as claimed in claim 1, characterized in that: The strategy control specifically includes the following steps: Communication and negotiation: communicate and negotiate with large industrial users based on control strategies, understand their production plans and electricity usage arrangements, and jointly adjust control plans; Implementing control, and implementing corresponding control measures according to the negotiation results, wherein the control measures include adjusting electricity prices, restricting electricity consumption, and sending dispatch instructions; Monitoring and evaluation: Monitor and evaluate the effects of regulation and control, and timely adjust and optimize regulation and control strategies.

Images