Power load prediction method and system based on big data

Abstract

The invention relates to the technical field of power load prediction, and discloses a power load prediction method and system based on big data. According to the method and the system, historical power consumption data, environmental data, meteorological data, social and economic data and power generation equipment state data are acquired, repeated data are identified and eliminated, data standardization is performed, data from different data sources are combined to form an integrated data set, and according to the integrated data, the data of the power generation equipment is acquired. Calculating daily power consumption, temperature change rate, regional population density, power load peak value, load mean value and power supply capacity, calculating a power load predicted value by integrating the numerical values, generating a power load trend chart according to the power load predicted value, and outputting the power load predicted value and the power load trend chart to a display terminal. The electric power related data is obtained in real time for analysis and calculation, the accurate electric power load prediction value is obtained, the change of the electric power demand is responded in time, and therefore the electric power dispatching and supplying efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load forecasting, and in particular to a power load forecasting method and system based on big data. Background Art

[0002] Electricity load refers to the amount of electricity actually used by consumers in the power system within a specific period of time. It is a measure of electricity demand and is usually expressed in kilowatts or megawatts. Electricity load is affected by many factors, including weather conditions, time, social activities, and economic activities. Forecasting electricity load can help power companies arrange power generation and transmission reasonably to ensure adequate power supply during high load periods. By accurately forecasting load, the power system can effectively respond to emergencies and reduce fluctuations in power supply. At the same time, as the proportion of renewable energy, such as wind and solar energy, gradually increases, load forecasting can help balance the output uncertainty of renewable energy and achieve more efficient energy utilization.

[0003] Currently, most power load forecasting technologies are unable to perform real-time or near real-time forecasts, resulting in the inability to respond promptly to changes in power demand, thereby reducing the efficiency of power dispatch and supply. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for predicting power load based on big data, which has the advantages of obtaining historical power consumption data, environmental data, meteorological data, socio-economic data and power generation equipment status data, identifying and eliminating duplicate data, performing data standardization, combining data from different data sources to form an integrated data set, and calculating daily power consumption, temperature change rate, regional population density, power load peak, load mean and power supply capacity based on the integrated data; based on the above calculated values, comprehensively calculating the power load prediction value based on the weighted average method, generating a power load trend graph based on the power load prediction value, outputting the power load prediction value and the power load trend graph to a display terminal, obtaining accurate power load prediction values ​​by analyzing and calculating power-related data in real time, and responding to changes in power demand in a timely manner, thereby improving the efficiency of power dispatching and supply, etc., and solving the above problems.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: a method for power load forecasting based on big data, comprising the following steps:

[0008] S1. Obtain historical power consumption data, environmental data, meteorological data, socio-economic data, and power generation equipment status data;

[0009] S2, identify and eliminate duplicate data, perform data standardization, and combine data from different data sources to form an integrated data set;

[0010] S3. Calculate daily electricity consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data;

[0011] S4, based on the value calculated in S3, comprehensively calculate the power load forecast value based on the weighted average method;

[0012] S5. Generate a power load trend graph according to the power load forecast value, and output the power load forecast value and the power load trend graph to a display terminal.

[0013] Preferably, the formula for removing duplicate data in S2 is as follows:

[0014] Dx={x∈S|H[x]is not defined}

[0015] In the formula, Dx represents the data set after deduplication, and H represents the created empty hash table. The formula means traversing the data set S, for each element x, checking whether x exists in H, if so, skip it, if not, add x to the result set and store it in H.

[0016] Preferably, the formula for calculating the daily power consumption according to the integrated data in S3 is as follows:

[0017]

[0018] In the formula, Dlxh represents the daily electricity consumption, C h It represents the electricity consumption in the hth hour, and the summation formula represents the electricity consumption from the first hour to the 24th hour, and the accumulation is performed.

[0019] Preferably, the formula for calculating the temperature change rate in S3 based on the integrated data is as follows:

[0020]

[0021] In the formula, Wdrt represents the temperature change rate, T2 represents the average temperature of the second day, and T1 represents the average temperature of the first day.

[0022] Preferably, the formula for calculating the regional population density in S3 based on the integrated data is as follows:

[0023]

[0024] In the formula, Rkmd represents the regional population density, Pr represents the total population in the region, and Qymj represents the total area of ​​the region.

[0025] Preferably, the formula for calculating the peak power load according to the integrated data in S3 is as follows:

[0026] Peld=max{L1, L2, L3,…,L n}

[0027] In the formula, Peld represents the peak value of power load, L1, L2, L3, ..., L n represents the power load at different time points, and n represents the total number of data points in the time period.

[0028] Preferably, the formula for calculating the load mean value in S3 based on the integrated data is as follows:

[0029]

[0030] In the formula, Fhjz represents the load mean, n represents the total number of data points in the time period, and L i represents the power load at each time point, It means calculating from the first power data to the nth power data and summing them.

[0031] Preferably, the formula for calculating the power supply capacity according to the integrated data in S3 is as follows:

[0032] Gdnl=Zdnl-Sjdh

[0033] In the formula, Gdnl represents the power supply capacity, Zdnl represents the total power generation capacity of the power grid, and Sjdh represents the current actual power load.

[0034] Preferably, the formula for comprehensively calculating the power load forecast value based on the weighted average method in S4 is as follows:

[0035] Ycfh=0.2*Dlxh+0.1*Wdrt+0.2*Rkmd+0.3*Peld+0.1*Fhjz+0.1

[0036] *Gdnl

[0037] In the formula, Dlxh represents daily electricity consumption, Wdrt represents the temperature change rate, Rkmd represents the regional population density, Peld represents the peak power load, Fhjz represents the average load, Gdnl represents the power supply capacity, and 0.1, 0.2 and 0.3 represent the weight values ​​represented by each parameter.

[0038] A power load forecasting system based on big data, comprising a data acquisition module, a data preprocessing module, a data analysis module, a power forecasting module and an output module;

[0039] The data acquisition module acquires historical power consumption data, environmental data, meteorological data, social and economic data, and power generation equipment status data, and transmits the acquired data to the data preprocessing module;

[0040] The data preprocessing module identifies and removes duplicate data, performs data standardization, and combines data from different data sources to form an integrated data set;

[0041] The data analysis module calculates the daily power consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data;

[0042] The power prediction module comprehensively calculates the power load prediction value based on the weighted average method according to the value calculated by the data analysis module;

[0043] The output module generates an electric load trend graph according to the electric load prediction value, and outputs the electric load prediction value and the electric load trend graph to a display terminal.

[0044] Compared with the prior art, the present invention provides a method and system for power load forecasting based on big data, which has the following beneficial effects:

[0045] The present invention obtains historical power consumption data, environmental data, meteorological data, socio-economic data and power generation equipment status data, identifies and eliminates duplicate data, performs data standardization, combines data from different data sources to form an integrated data set, and calculates daily power consumption, temperature change rate, regional population density, power load peak, load mean and power supply capacity based on the integrated data; based on the above calculated values, a power load forecast value is comprehensively calculated based on a weighted average method; a power load trend graph is generated based on the power load forecast value; the power load forecast value and the power load trend graph are output to a display terminal; accurate power load forecast values ​​are obtained by acquiring power-related data in real time for analysis and calculation, and changes in power demand are responded to in a timely manner, thereby improving the efficiency of power dispatch and supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the steps of the method of the present invention;

[0047] Figure 2 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0049] Most of the current power load forecasting technologies are unable to make real-time or near-real-time forecasts, which results in an inability to respond to changes in power demand in a timely manner, thereby reducing the efficiency of power dispatch and supply. To this end, a power load forecasting method based on big data is proposed. Figure 1 , the method comprises the following steps:

[0050] S1. Obtain historical power consumption data, environmental data, meteorological data, socio-economic data, and power generation equipment status data;

[0051] Obtaining historical power consumption data, environmental data, meteorological data, socio-economic data, and power generation equipment status data is an important basis for power load forecasting and power system optimization. In terms of data collection, the use of advanced data collection systems and smart meters can monitor and record power consumption data in real time to ensure the accuracy and timeliness of the data. At the same time, the deployment of Internet of Things sensors can obtain environmental and meteorological data, such as temperature, humidity, wind speed, and rainfall, to help analyze the impact of external factors on power load. In addition, socio-economic data can be obtained through cooperation with statistical departments and market research institutions, including information such as population growth, industry development, and economic indicators, to provide background support for load forecasting. In terms of obtaining power generation equipment status data, it can be achieved through the equipment status monitoring system. After the data is collected, it needs to be stored and analyzed using a big data processing framework. Through data fusion and cleaning methods, various types of data can be integrated into a unified data platform to provide high-quality input for subsequent load forecasting models.

[0052] S2, identify and eliminate duplicate data, perform data standardization, and combine data from different data sources to form an integrated data set;

[0053] The formula for removing duplicate data is as follows:

[0054] Dx={x∈S|H[x]is notdefined}

[0055] Eliminating duplicate data can purify the data set and ensure the accuracy and reliability of the analysis. High-quality data is crucial to the effectiveness of modeling and prediction. In the formula, Dx represents the data set after deduplication, and H represents the created empty hash table. The formula means traversing the data set S, for each element x, checking whether x exists in H. If so, skip it. If not, add x to the result set and store it in H. Data redundancy will take up a lot of storage space and computing resources. By eliminating duplicate data, the complexity of data processing can be reduced and computing efficiency can be improved.

[0056] S3. Calculate daily electricity consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data;

[0057] The formula for calculating daily electricity consumption is as follows:

[0058]

[0059] Through the statistics of daily electricity consumption, the fluctuation and change of demand can be clearly observed, providing instant feedback for power dispatch. In the formula, Dlxh represents the daily electricity consumption, C h represents the electricity consumption in the hth hour. The summation formula represents the electricity consumption from the first hour to the 24th hour and the accumulation. The daily consumption information can be used to predict future electricity demand and provide reference for market analysis, pricing strategies and supply plans.

[0060] The formula for calculating the temperature change rate is as follows:

[0061]

[0062] Temperature changes directly affect electricity demand. Calculating the temperature change rate can help the model more accurately capture the impact of climate factors on load, especially in summer and winter. In the formula, Wdrt represents the temperature change rate, T2 represents the average temperature of the second day, and T1 represents the average temperature of the first day. The calculation of the temperature change rate enables power companies to better predict peak loads under extreme weather conditions and prepare for power generation and power dispatch in advance.

[0063] The formula for calculating regional population density is as follows:

[0064]

[0065] Population density is usually positively correlated with electricity demand. Understanding this indicator can help power companies assess the electricity consumption potential of various regions. In the formula, Rkmd represents the regional population density, Pr represents the total population in the region, and Qymj represents the total area of ​​the region. With population density information, power companies can more effectively promote products and services to consumers in specific areas.

[0066] The formula for calculating the peak power load is as follows:

[0067] Peld=max{L1, L2, L3,..., L n}

[0068] Understanding the load peak can help the power dispatch center prepare for power generation during high-demand periods, balance the load, and improve system stability. In the formula, Peld represents the power load peak, L1, L2, L3, ..., L n represents the power load at different time points, and n represents the total number of data points in the time period. By accurately predicting the load peak, power companies can avoid or reduce power outages caused by insufficient supply and improve the satisfaction of all users;

[0069] The formula for calculating the load mean is as follows:

[0070]

[0071] The load mean provides baseline information for the normal operation of the power system and helps to detect potential abnormal conditions. In the formula, Fhjz represents the load mean, n represents the total number of data points in the time period, and L i represents the power load at each time point, It means that the power data from the first to the nth are calculated and summed. The stable load mean value helps the power company to carry out dynamic scheduling and load forecasting, thus reducing operating costs.

[0072] The formula for calculating power supply capacity is as follows:

[0073] Gdnl=Zdnl-Sjdh

[0074] Understanding power supply capacity can help power companies determine whether the existing system can meet the expected power demand and avoid power shortages. In the formula, Gdnl represents the power supply capacity, Zdnl represents the total power generation capacity of the power grid, and Sjdh represents the current actual power load. Accurate power supply capacity calculation can provide support for emergency plans and ensure rapid response in the event of an emergency.

[0075] S4, based on the value calculated in S3, comprehensively calculate the power load forecast value based on the weighted average method;

[0076] The formula for calculating the power load forecast value is as follows:

[0077] Ycfh=0.2*Dlxh+0.1*Wdrt+0.2*Rkmd+0.3*Peld+0.1*Fhjz+0.1

[0078] *Gdnl

[0079] Through accurate load forecasting, power companies can more effectively plan and allocate power generation resources and infrastructure to ensure sufficient power supply during high-demand periods. In the formula, Dlxh represents daily power consumption, Wdrt represents the temperature change rate, Rkmd represents the regional population density, Peld represents the peak power load, Fhjz represents the average load, Gdnl represents the power supply capacity, and 0.1, 0.2, and 0.3 represent the weight values ​​represented by each parameter. Forecasting power load in advance can effectively prevent the power grid from being in short supply during peak hours and reduce the risk of power outages.

[0080] S5. Generate a power load trend graph according to the power load forecast value, and output the power load forecast value and the power load trend graph to a display terminal;

[0081] The process of generating a power load trend chart based on power load forecast values ​​mainly involves data collection, analysis and visualization technology. The system collects power load data in real time through advanced monitoring equipment and sensors, and cleans and preprocesses it in combination with historical data to ensure the accuracy and reliability of the input data. Next, the sorted data is analyzed and calculated to generate power load forecast values ​​for a period of time in the future. Then, these forecast values ​​are converted into trend charts and graphically displayed using the data visualization tool Matplotlib, which facilitates the intuitive presentation of load change trends and patterns on the display terminal. Finally, through the interactive dashboard, the power load forecast values ​​and trend charts are updated and output in real time to help decision makers quickly grasp the trend of power demand and optimize power scheduling and resource allocation.

[0082] See also Figure 2 , a power load forecasting system based on big data, including a data acquisition module, a data preprocessing module, a data analysis module, a power forecasting module and an output module;

[0083] The data acquisition module acquires historical power consumption data, environmental data, meteorological data, social and economic data, and power generation equipment status data, and transmits the acquired data to the data preprocessing module;

[0084] The data preprocessing module identifies and removes duplicate data, performs data standardization, and combines data from different data sources to form an integrated data set;

[0085] The data analysis module calculates the daily power consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data;

[0086] The power prediction module comprehensively calculates the power load prediction value based on the weighted average method according to the value calculated by the data analysis module;

[0087] The output module generates an electric load trend graph according to the electric load prediction value, and outputs the electric load prediction value and the electric load trend graph to a display terminal.

[0088] Combined with the application of the above method and system, by acquiring power-related data in real time for analysis and calculation, accurate power load forecast values ​​can be obtained, and changes in power demand can be responded to in a timely manner, thereby improving the efficiency of power dispatch and supply.

[0089] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for power load forecasting based on big data, characterized in that: The following steps are involved: S1. Obtain historical power consumption data, environmental data, meteorological data, socio-economic data, and power generation equipment status data; S2, identify and eliminate duplicate data, perform data standardization, and combine data from different data sources to form an integrated data set; S3. Calculate daily electricity consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data; S4, based on the value calculated in S3, comprehensively calculate the power load forecast value based on the weighted average method; S5. Generate a power load trend graph according to the power load forecast value, and output the power load forecast value and the power load trend graph to a display terminal.

2. The method for predicting power load based on big data according to claim 1, characterized in that: The formula for removing duplicate data in S2 is as follows: Dx={x∈S|H[x]is not defined} In the formula, Dx represents the data set after deduplication, and H represents the created empty hash table. The formula means traversing the data set S, for each element x, checking whether x exists in H, if so, skip it, if not, add x to the result set and store it in H.

3. The method for predicting power load based on big data according to claim 2, characterized in that: The formula for calculating the daily electricity consumption according to the integrated data in S3 is as follows: In the formula, Dlxh represents the daily electricity consumption, C h It represents the electricity consumption in the hth hour, and the summation formula represents the electricity consumption from the first hour to the 24th hour, and the accumulation is performed.

4. The method for predicting power load based on big data according to claim 3 is characterized in that: The formula for calculating the temperature change rate according to the integrated data in S3 is as follows: In the formula, Wdrt represents the temperature change rate, T2 represents the average temperature of the second day, and T1 represents the average temperature of the first day.

5. The method for predicting power load based on big data according to claim 4 is characterized in that: The formula for calculating regional population density in S3 based on the integrated data is as follows: In the formula, Rkmd represents the regional population density, Pr represents the total population in the region, and Qymj represents the total area of ​​the region.

6. The method for predicting power load based on big data according to claim 5 is characterized in that: The formula for calculating the peak power load according to the integrated data in S3 is as follows: Peld=max{L1,L2,L3,…,L n } In the formula, Peld represents the peak value of power load, L1, L2, L3, ..., L n represents the power load at different time points, and n represents the total number of data points in the time period.

7. The method for predicting power load based on big data according to claim 6 is characterized by: The formula for calculating the load mean value according to the integrated data in S3 is as follows: In the formula, Fhjz represents the load mean, n represents the total number of data points in the time period, and L i represents the power load at each time point, It means calculating from the first power data to the nth power data and summing them.

8. The method for predicting power load based on big data according to claim 7, characterized in that: The formula for calculating the power supply capacity of S3 based on the integrated data is as follows: Gdnl=Zdnl-Sjdh In the formula, Gdnl represents the power supply capacity, Zdnl represents the total power generation capacity of the power grid, and Sjdh represents the current actual power load.

9. The method for predicting power load based on big data according to claim 8, characterized in that: The formula for comprehensively calculating the power load forecast value based on the weighted average method in S4 is as follows: Ycfh=0.2*Dlxh+0.1*Wdrt+0.2*Rkmd+0.3*Peld+0.1*Fhjz+0.1*Gdnl In the formula, Dlxh represents daily electricity consumption, Wdrt represents the temperature change rate, Rkmd represents the regional population density, Peld represents the peak power load, Fhjz represents the average load, Gdnl represents the power supply capacity, and 0.1, 0.2 and 0.3 represent the weight values ​​represented by each parameter.

10. A power load forecasting system based on big data, characterized by: It includes data acquisition module, data preprocessing module, data analysis module, power prediction module and output module; The data acquisition module acquires historical power consumption data, environmental data, meteorological data, social and economic data, and power generation equipment status data, and transmits the acquired data to the data preprocessing module; The data preprocessing module identifies and removes duplicate data, performs data standardization, and combines data from different data sources to form an integrated data set; The data analysis module calculates the daily power consumption, temperature change rate, regional population density, power load peak, load average and power supply capacity based on the integrated data; The power prediction module comprehensively calculates the power load prediction value based on the weighted average method according to the value calculated by the data analysis module; The output module generates an electric load trend graph according to the electric load prediction value, and outputs the electric load prediction value and the electric load trend graph to a display terminal.

Images