A power consumption analysis and load forecasting method and system

Abstract

The application discloses a power consumption analysis and load prediction method and system, which obtains current user quantity and power consumption load type; analyzes power consumption according to the current user quantity and the power consumption load type to obtain power consumption analysis results; and predicts power consumption and maximum load based on the power consumption analysis results and by using a combined prediction model, so as to configure the capacity of a renewable energy comprehensive utilization system according to the prediction results. The application can determine the load of the renewable energy comprehensive utilization system and effectively predict the power consumption of the load.

Description

Technical Field

[0001] This invention relates to the field of load forecasting technology, and in particular to a method and system for electricity consumption analysis and load forecasting. Background Technology

[0002] Electricity load is influenced by a variety of factors, such as weather, economic activity, and population growth. The uncertainty of these factors poses challenges to load forecasting. Researchers are developing more advanced algorithms, such as deep learning-based models, to improve forecast accuracy. With the development of big data technology, researchers are beginning to utilize historical load data, weather data, and other relevant data to train forecasting models. These models include traditional time series models (such as ARIMA), machine learning models (such as random forests and support vector machines), and deep learning models (such as convolutional neural networks). Load forecasting needs to cover different time scales, from short-term (such as daily and weekly) to long-term (such as monthly and yearly). Forecasts at different time scales require consideration of different influencing factors and models. For example, short-term forecasts may rely more on weather data, while long-term forecasts may need to consider factors such as economic growth and population changes. Electricity demand typically exhibits obvious seasonal and cyclical characteristics. For example, electricity consumption typically increases in winter and summer due to heating and cooling needs. Forecasting models need to be able to capture these patterns and incorporate them into forecasts. Grid electricity data may contain missing values, outliers, or noise. Effective data cleaning and preprocessing are key steps in improving forecast accuracy. Researchers have developed various algorithms to address these problems, including outlier detection and recursive imputation algorithms based on the improved KNN algorithm. Power systems are complex, requiring consideration of multiple variables (such as temperature, humidity, and wind speed) and the interactions of multiple regions. This necessitates predictive models capable of handling multivariate data and capturing the interdependencies between different regions. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] Therefore, this invention proposes a method for electricity consumption analysis and load forecasting, which can improve the accuracy of load forecasting. This is of great significance for electricity management, energy consumption control, and the economic consumption and social benefits of the power system.

[0005] To achieve the above objectives, another aspect of the present invention proposes a power consumption analysis and load forecasting system.

[0006] To achieve the above objectives, the present invention provides a method for electricity consumption analysis and load forecasting, comprising:

[0007] Get the current number of users and the type of electricity load;

[0008] The electricity consumption analysis results are obtained by analyzing the current number of users and the type of electricity load.

[0009] Based on the electricity consumption analysis results, and using a combined prediction model to predict electricity consumption and maximum load, the capacity of the renewable energy integrated utilization system is configured according to the prediction results.

[0010] The power consumption analysis and load forecasting method of this invention may also have the following additional technical features:

[0011] In one embodiment of the present invention, the types of electricity load include: electricity consumption by villagers, electricity consumption for seawater desalination, electricity consumption for production, and electricity consumption for tourism.

[0012] In one embodiment of the present invention, it further includes:

[0013] Electricity consumption and maximum load are predicted according to the prediction time type; wherein, the prediction time type includes ultra-short-term, short-term, medium-term and long-term.

[0014] Based on different conditions and environments, the appropriate forecast time type is selected to forecast electricity consumption and maximum load.

[0015] In one embodiment of the present invention, the combined prediction model includes multiple methods selected from regression analysis, time series analysis, trend analysis, artificial neural network analysis, fuzzy prediction, and grey prediction; the method further includes:

[0016] Obtain multiple prediction results corresponding to multiple combined prediction models;

[0017] The multiple prediction results are summarized to obtain the summarized prediction result;

[0018] The summarized prediction results are compared and analyzed to select the optimal combination prediction model based on the results of the comparative analysis.

[0019] In one embodiment of the present invention, the relationship between total power consumption and total load is shown in the following formula:

[0020]

[0021] Where L is the total load, A is the total electricity consumption, and T is the total load. MAx This represents the number of hours the system is used at its maximum load.

[0022] To achieve the above objectives, a second aspect of this application provides a power consumption analysis and load forecasting system, comprising:

[0023] The load type determination module is used to obtain the current number of users and the type of electricity load.

[0024] The electricity consumption analysis module is used to analyze electricity consumption based on the current number of users and the type of electricity load to obtain electricity consumption analysis results;

[0025] The load forecasting module is used to forecast the electricity consumption and maximum load based on the electricity consumption analysis results and using a combined forecasting model, so as to configure the capacity of the renewable energy integrated utilization system according to the forecast results.

[0026] The electricity consumption analysis and load forecasting method and system of this invention enable the modernization and intelligentization of power system management, becoming an important research direction in the field of smart grid security. Load forecasting technology can cover different time spans, including ultra-short-term, short-term, medium-term, and long-term load forecasting. It is applicable to different scales, including grid-level and user-level load forecasting. This contributes to the stable operation and reasonable improvement of the power system, and is also closely related to the relevant businesses of electricity sales companies.

[0027] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0028] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0029] Figure 1 This is a flowchart of a power consumption analysis and load forecasting method according to an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram of a power consumption analysis and load prediction system according to an embodiment of the present invention. Detailed Implementation

[0031] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0032] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] The following describes a method for electricity consumption analysis and load forecasting according to an embodiment of the present invention, with reference to the accompanying drawings.

[0034] Figure 1 This is a flowchart of a power consumption analysis and load forecasting method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes:

[0035] S1, obtain the current number of users and the type of electricity load;

[0036] S2, Analyze the electricity consumption based on the current number of users and the type of electricity load to obtain the electricity consumption analysis results;

[0037] S3. Based on the electricity consumption analysis results, and using a combined prediction model to predict electricity consumption and maximum load, the capacity of the renewable energy integrated utilization system is configured according to the prediction results.

[0038] In one embodiment of the present invention, since there is currently no official data or information summarizing and statistically analyzing the annual electricity consumption of Daguan Island, the present invention analyzes the electricity consumption of Daguan Island based on the current number of villagers and their electricity usage. The electricity load of Daguan Island is mainly divided into four types: electricity consumption by villagers, electricity consumption for seawater desalination, electricity consumption for production, and electricity consumption for tourism.

[0039] (1) Electricity consumption for villagers' daily life

[0040] Currently, Daguan Island Village has 32 households. Based on an average monthly electricity consumption of 100 kWh per household, the annual electricity consumption would be:

[0041] 32 × 100 × 12 = 38400 (kW·h)

[0042] (2) Electricity consumption for seawater desalination

[0043] Currently, Daguan Island has an existing seawater desalination plant, but its desalination capacity is only sufficient to meet the daily water needs of the island's residents. Therefore, the newly added seawater desalination plant will be used to supply water to tourists. Assuming an average daily tourist volume of 80 people per year, and each person consumes 50 kg of freshwater per day, the average daily water consumption for tourists is 4000 kg. Taking a seawater desalination plant with a daily output of 5000 kg as an example, the electricity consumption is 5 kW·h per 1000 kg of freshwater produced. Therefore, the annual electricity consumption of the seawater desalination system is:

[0044] 5 × 5 × 365 = 9125 (kW·h)

[0045] (3) Electricity consumption for production

[0046] Due to its natural conditions, Daguan Island's main industries are fishing and aquaculture, with abalone and sea cucumber farming ponds built around the island. Therefore, the island's electricity consumption primarily comes from the pumps and aerators used for abalone and sea cucumber farming. Assuming the island currently uses one 10kW pump and one 5kW aerator, and that the suitable farming days are approximately 90 days per year, with pumping and aerating the ponds 24 hours a day during the farming season, then Daguan Island's annual electricity consumption would be:

[0047] (10+5)×24×90=32400(kW·h)

[0048] (4) Electricity consumption in the tourism industry

[0049] With the government's vigorous promotion of tourism on Daguan Island, there are currently more than ten hotels in Daguan Island Village, with a total of about 40 rooms. Each room consumes approximately 100 kWh of electricity per month. Therefore, the annual electricity consumption of tourism on Daguan Island is:

[0050] 40 × 100 × 12 = 48000 (kW·h)

[0051] (5) Total load and electricity consumption

[0052] Based on the above analysis of electricity consumption in four areas—domestic electricity, seawater desalination electricity, industrial electricity, and tourism electricity—the average annual electricity consumption of Daguan Island is approximately:

[0053] 38400+9125+48000+32400=127925(kW·h)

[0054] It is known that, theoretically, the average annual load of Daguan Island is about 70kW. However, considering factors such as season and passenger flow, and estimating it based on a simultaneity rate of 0.8, the annual load of Daguan Island is about 56kW.

[0055] Furthermore, forecasting the electricity consumption and load of independent islands is an important step in the planning of a comprehensive renewable energy utilization system platform. Accurate forecasting can effectively ensure the reliability and economy of the system.

[0056] This invention provides an embodiment for predicting electricity consumption and maximum load according to prediction time type; wherein, the prediction time type includes ultra-short-term, short-term, medium-term, and long-term; and the corresponding prediction time type is selected according to different conditions and environments to predict electricity consumption and maximum load.

[0057] The combined prediction model of this invention includes multiple methods such as regression analysis, time series analysis, trend analysis, artificial neural network, fuzzy prediction, and grey prediction. It also includes: obtaining multiple prediction results corresponding to multiple combined prediction models; summarizing the multiple prediction results to obtain a summarized prediction result; and performing comparative analysis on the summarized prediction results to select the optimal combined prediction model based on the comparative analysis results.

[0058] Specifically, there is currently a great deal of research on power load forecasting. Classified by forecast timeframe, it can be divided into four categories: ultra-short-term, short-term, medium-term, and long-term. Long-term forecasting generally refers to forecasts with a forecast period of 10 years or more, while short-term forecasts have a forecast period of approximately 5 years. Short-term forecasts mainly predict load for no more than one year, and ultra-short-term forecasts have even shorter timeframes. In practical applications, the appropriate type of forecasting is selected based on different conditions and environments. Furthermore, with the deepening of research, there are increasingly more methods for load forecasting, no longer limited to traditional regression analysis, time series analysis, and trend analysis. It can also be seen that experts and scholars are incorporating modern technologies into their research methods, such as artificial neural networks, fuzzy forecasting, and grey forecasting. Since each forecasting method has its own applicable range, a single method is unlikely to meet all conditions. Therefore, researchers have proposed combined forecasting models that summarize and analyze the forecasting results of multiple methods. Compared to the results obtained from a single forecasting method, combined forecasting results are more accurate and comprehensive, and can meet most operating conditions.

[0059] Furthermore, the electricity consumption and maximum load of the main island are predicted to configure the capacity of the renewable energy integrated utilization system. The relationship between total electricity consumption and total load is shown in the following equation:

[0060]

[0061] Where L is the total load, A is the total electricity consumption, and T is the total load. MAx This represents the number of hours the system is used at its maximum load.

[0062] Since there is currently no specific historical data or research on the actual electricity consumption and load of Daguan Island, a long-term forecast is made based on the aforementioned electricity consumption characteristics of Daguan Island and relevant national and governmental support policies. According to general experience, the service life of offshore power generation equipment is typically 15 years; therefore, a 15-year load forecast is conducted for Daguan Island. It is assumed that during these 15 years, the number of villagers on the island remains constant, and the increase in electricity load for villagers' daily life is mainly due to the increase in electrical equipment; the electricity load for seawater desalination depends primarily on the development of tourism, with its growth rate matching that of tourism; and the electricity load for production remains relatively constant. Based on the literature, assuming that the load on Daguan Island increases at a rate of 1.8% per year, and the average electricity consumption on the island increases at a rate of 2% per year, the forecast results are as follows:

[0063] 56 × (1 + 1.8%) 15 ≈73.18 (kW)

[0064] 127925×(1+2%) 5 ≈172170.21 (kW·h)

[0065] Therefore, it is estimated that by 2036, the average annual load of Daguan Island will be 73.18 kW and the average annual electricity consumption will be 172,170.21 kWh.

[0066] Furthermore, Daguan Island is located in the Yellow Sea at 120°46'00"E, 36°13'44"N, surrounded by the sea and possessing abundant marine renewable resources. The following analysis examines the wind, solar, tidal, and salinity gradient energy resources in the waters near Daguan Island.

[0067] 1. Wind energy

[0068] According to relevant data, Daguan Island has relatively abundant wind energy resources, with wind speeds in the surrounding waters typically reaching level 3 or higher. Statistics show that the average sea wind speed is approximately 5 m / s, with the maximum wind speed in the area reaching 34 m / s. The annual effective wind hours (3 m / s-20 m / s) are approximately 7000 hours, and the annual effective wind energy is approximately 3578 kWh / m². 2 .

[0069] 2. Solar energy

[0070] According to data from the "Jimo City Marine and Fisheries Resources Database," Daguan Island has a warm temperate monsoon climate, with occasional foggy weather. The average number of foggy days per year is approximately 29, with the remaining days being sunny. The intensity of sunlight varies throughout the year. Generally, May and June have the longest sunshine hours, with an average monthly sunshine duration of 240-300 hours. December to February of the following year have the shortest sunshine hours, with an average monthly sunshine duration of approximately 142-193 hours. The average annual sunshine duration is approximately 2000 hours. The annual radiation can exceed 500,000 J / cm². 2 .

[0071] 3. Tidal Energy

[0072] According to reference

[26] , the average wave height of the ocean currents near Daguan Island is 1.7-2m, the average water depth is over 40m, the current velocity is approximately 1.5-4.2m / s, and the average power density can reach 1.2-3.7kW / m². 2 Since the direction of ocean currents is the same as the direction of waves, the direction of ocean currents can be inferred from the direction of waves. Specific data such as the direction of currents are shown in Table 1 below.

[0073] Table 1

[0074]

[0075] 4. Salinity gradient energy

[0076] The average annual seawater temperature near Daguan Island is approximately 20.52℃, and the average salinity is 29.37%. The average spring seawater runoff is 1403 m³ / h. 3 / s, the average summer seawater runoff is 2008m³ / s 3 / s, the average autumn seawater runoff is 914m³ 3 / s, the average winter seawater runoff is 622m³ 3 / s.

[0077] According to commonly used marine tables, when the seawater temperature is 20℃, the osmotic pressure values ​​of seawater with different concentrations are shown in Table 2.

[0078] Table 2

[0079]

[0080] Based on the sea temperature and salinity data in the table above, the osmotic pressure near Daguan Island is approximately 2.465 MPa. The salinity gradient energy potential is the product of seawater runoff and osmotic pressure, thus the annual average salinity gradient energy potential near Daguan Island is calculated to be approximately 2.81 × 10⁶ kW.

[0081] The electricity consumption analysis and load forecasting method according to embodiments of the present invention can improve the accuracy of load forecasting. This is of great significance for electricity management, energy consumption control, and the economic consumption and social benefits of the power system. It makes the modernization and intelligentization of power system management possible and has become an important research direction in the field of smart grid security. The application of intelligent technology to the field of smart grid load forecasting has yielded a series of research results. Load forecasting technology can cover different time spans, including ultra-short-term, short-term, medium-term, and long-term load forecasting. This has a significant impact on the formulation of reasonable power generation plans, real-time electricity prices, and load adjustment by power companies. Load forecasting technology can be applied to different scales, including grid-level load forecasting and user-level load forecasting. This contributes to the stable operation and reasonable improvement of the power system and is also closely related to the relevant businesses of electricity sales companies. Since load is affected by many factors, such as weather and holidays, load forecasting technology needs to be able to handle these uncertainties and randomness. This requires the forecasting model to be able to adapt to changes under different conditions and provide multiple forecasting schemes. This makes it possible to combine modern technologies such as big data and deep learning for load forecasting. The application of these technologies provides new solutions for load forecasting and improves the accuracy and efficiency of forecasting. With accurate load forecasting, the power system can more effectively dispatch power and allocate resources, thereby reducing unnecessary energy waste and improving the economic efficiency of the power system.

[0082] like Figure 2 As shown, the present invention also proposes a power consumption analysis and load forecasting system 10, comprising:

[0083] The load type determination module 100 is used to obtain the current number of users and the type of electrical load.

[0084] The power consumption analysis module 200 is used to analyze power consumption based on the current number of users and the type of power load to obtain power consumption analysis results;

[0085] The load forecasting module 300 is used to forecast the electricity consumption and maximum load based on the electricity consumption analysis results and using a combined forecasting model, so as to configure the capacity of the renewable energy integrated utilization system according to the forecast results.

[0086] Furthermore, the types of electricity load include: electricity consumption by villagers, electricity consumption for seawater desalination, electricity consumption for production, and electricity consumption for tourism.

[0087] Furthermore, it also includes:

[0088] Electricity consumption and maximum load are predicted according to the prediction time type; wherein, the prediction time type includes ultra-short-term, short-term, medium-term and long-term.

[0089] Based on different conditions and environments, the appropriate forecast time type is selected to forecast electricity consumption and maximum load.

[0090] Furthermore, the combined prediction model includes multiple methods such as regression analysis, time series analysis, trend analysis, artificial neural network analysis, fuzzy prediction, and grey prediction; the method also includes:

[0091] Obtain multiple prediction results corresponding to multiple combined prediction models;

[0092] The multiple prediction results are summarized to obtain the summarized prediction result;

[0093] The summarized prediction results are compared and analyzed to select the optimal combination prediction model based on the results of the comparative analysis.

[0094] Furthermore, the relationship between total electricity consumption and total load is shown in the following formula:

[0095]

[0096] Where L is the total load, A is the total electricity consumption, and T is the total load. MAx This represents the number of hours the system is used at its maximum load.

[0097] The power consumption analysis and load forecasting system according to embodiments of the present invention can improve the accuracy of load forecasting. This is of great significance for power management, energy consumption control, and the economic consumption and social benefits of the power system. It makes the modernization and intelligentization of power system management possible and has become an important research direction in the field of smart grid security. The application of intelligent technology to the field of smart grid load forecasting has yielded a series of research results. Load forecasting technology can cover different time spans, including ultra-short-term, short-term, medium-term, and long-term load forecasting. This has a significant impact on the formulation of reasonable power generation plans, real-time electricity prices, and load adjustment by power companies. Load forecasting technology can be applied to different scales, including grid-level load forecasting and user-level load forecasting. This contributes to the stable operation and reasonable improvement of the power system and is also closely related to the relevant business of electricity sales companies. Since load is affected by many factors, such as weather and holidays, load forecasting technology needs to be able to handle these uncertainties and randomness. This requires the forecasting model to be able to adapt to changes under different conditions and provide multiple forecasting schemes. This makes it possible to combine modern technologies such as big data and deep learning for load forecasting. The application of these technologies provides new solutions for load forecasting, improving the accuracy and efficiency of forecasting. With accurate load forecasting, the power system can more effectively dispatch power and allocate resources, thereby reducing unnecessary energy waste and improving the economic efficiency of the power system.

[0098] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0099] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for electricity consumption analysis and load forecasting, characterized in that, include: Get the current number of users and the type of electricity load; The electricity consumption analysis results are obtained by analyzing the current number of users and the type of electricity load. Based on the electricity consumption analysis results, and using a combined prediction model to predict electricity consumption and maximum load, the capacity of the renewable energy integrated utilization system is configured according to the prediction results.

2. The method according to claim 1, characterized in that, The types of electrical loads include: Electricity consumption by villagers, electricity consumption for seawater desalination, electricity consumption for production, and electricity consumption for tourism.

3. The method according to claim 2, characterized in that, Also includes: Electricity consumption and maximum load are predicted according to the prediction time type; wherein, the prediction time type includes ultra-short-term, short-term, medium-term and long-term. Based on different conditions and environments, the appropriate forecast time type is selected to forecast electricity consumption and maximum load.

4. The method according to claim 1, characterized in that, The combined prediction model includes multiple methods such as regression analysis, time series analysis, trend analysis, artificial neural network analysis, fuzzy prediction, and grey prediction. The method further includes: Obtain multiple prediction results corresponding to multiple combined prediction models; The multiple prediction results are summarized to obtain the summarized prediction result; The summarized prediction results are compared and analyzed to select the optimal combination prediction model based on the results of the comparative analysis.

5. The method according to claim 1, characterized in that, The relationship between total electricity consumption and total load is shown in the following formula: Where L is the total load, A is the total electricity consumption, and T is the total load. MAx This represents the number of hours the system is used at its maximum load.

6. A power consumption analysis and load forecasting system, characterized in that, include: The load type determination module is used to obtain the current number of users and the type of electricity load. The electricity consumption analysis module is used to analyze electricity consumption based on the current number of users and the type of electricity load to obtain electricity consumption analysis results; The load forecasting module is used to forecast the electricity consumption and maximum load based on the electricity consumption analysis results and using a combined forecasting model, so as to configure the capacity of the renewable energy integrated utilization system according to the forecast results.

7. The system according to claim 6, characterized in that, The types of electricity loads include: electricity consumption by villagers, electricity consumption for seawater desalination, electricity consumption for production, and electricity consumption for tourism.

8. The system according to claim 7, characterized in that, Also includes: Electricity consumption and maximum load are predicted according to the prediction time type; wherein, the prediction time type includes ultra-short-term, short-term, medium-term and long-term. Based on different conditions and environments, the appropriate forecast time type is selected to forecast electricity consumption and maximum load.

9. The system according to claim 6, characterized in that, The combined prediction model includes multiple methods such as regression analysis, time series analysis, trend analysis, artificial neural network analysis, fuzzy prediction, and grey prediction. The method further includes: Obtain multiple prediction results corresponding to multiple combined prediction models; The multiple prediction results are summarized to obtain the summarized prediction result; The summarized prediction results are compared and analyzed to select the optimal combination prediction model based on the results of the comparative analysis.

10. The system according to claim 6, characterized in that, The relationship between total electricity consumption and total load is shown in the following formula: Where L is the total load, A is the total electricity consumption, and T is the total load. MAx This represents the number of hours the system is used at its maximum load.

Images