New energy electric power and electric quantity balancing method based on energy consumption prediction result

Abstract

The invention discloses a new energy electric power and electric quantity balancing method based on an energy consumption prediction result, relates to the technical field of new energy electric power, and improves the accuracy and integrity of meteorological observation data of a new energy power station. A new energy power station theoretical power generation capability evaluation method is perfected; developing a numerical weather forecast mode system oriented to the weather process; establishing an online interactive power prediction method; and application research of new technologies such as artificial intelligence and the like in power prediction is positively developed. According to the method, the previous power data is detected, power generation, power utilization and power utilization side energy consumption conditions are accurately predicted subsequently according to the data, and a power balance strategy is flexibly formulated and implemented in combination with new energy power generation power prediction; the problems of poor power supply and demand matching, insufficient new energy power utilization and the like in the prior art are effectively solved, and stable and efficient operation of a power system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of new energy power technology, and in particular to a new energy power quantity balancing method based on energy consumption prediction results. Background Art

[0002] With the continuous advancement of the "dual carbon" strategic goals, the proportion of new energy installed capacity will continue to increase. Against the background of the construction of the electricity spot market, there is an urgent need to improve the prediction accuracy and practicality of new energy to ensure the normal operation of the market. In recent years, new energy power prediction technology has developed rapidly, with hundreds of algorithm models, and great improvements in data cleaning, meteorological information utilization, etc. However, new energy power prediction algorithms from different manufacturers and with different strategies have their own strengths, and the matching effects with specific prediction scenarios are not consistent. In actual operation, new energy is relatively fragile in its tolerance to extreme weather. Existing prediction technologies are difficult to achieve accurate predictions in extreme weather, which may further lead to an aggravation of the power supply tension.

[0003] The invention patent with application publication number: CN114186725A discloses a method for balancing electric power for grid-connected operation of new energy sources. The specific steps include power supply area survey, determination of installed capacity of power generation equipment, determination of working capacity of power supply area, determination of spare capacity of power supply area, determination of required capacity of power supply area, installation of storage equipment, zoning of power supply area, laying of external power supply system, power supply connection and installation of master control equipment. The present invention solves the problem that traditional electric power balancing method cannot effectively balance electric power, resulting in frequent power outages during peak hours of electricity consumption, thus affecting the normal life of residents and causing a large amount of electricity waste during the trough period of electricity consumption. By improving and optimizing the structure of the electric power balancing method, the electric power balancing method is applicable to grid-connected operation of new energy sources, so that when the grid-connected operation of new energy sources is carried out, it can ensure that the electricity consumption of residents is not affected during the peak period of operation.

[0004] Although the above-mentioned equipment can ensure that residents' electricity consumption is not affected during peak periods, the above-mentioned power and electricity balancing methods often focus on power regulation on the power generation side or simply rely on conventional means such as energy storage systems and interactions with external power grids. They lack accurate prediction of energy consumption on the power consumption side and the formulation of systematic and comprehensive balancing strategies based on this. For example, some methods do not fully consider the energy consumption trends of different time periods and different user types when performing power balance adjustment, and the electricity generated by equipment in different seasons is different, which leads to oversupply or shortage of electricity during peak or off-peak periods, affecting the stability of the power grid and may even cause energy waste or power supply interruption. It cannot meet the needs of efficient and stable operation of modern power systems and full utilization of new energy electricity. Summary of the invention

[0005] The purpose of the present invention is to provide a new energy power balancing method based on energy consumption prediction results to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a new energy power and quantity balancing method based on energy consumption prediction results, the steps of the power and quantity balancing method are as follows:

[0007] Step 1: Improve the accuracy and completeness of meteorological observation data of new energy power plants, form a standard system for regulating data monitoring, collection, reporting, quality control, and installation, operation, maintenance and calibration of observation equipment, and improve the accuracy and completeness of meteorological observation data;

[0008] Step 2: Improve the theoretical power generation capacity evaluation method of new energy power stations, and reproduce the output power data of new energy power stations during power cuts through theoretical power generation capacity evaluation;

[0009] Based on step 2, considering that the amount of new energy power has random characteristics based on seasons, the power generation power of the power supply area is predicted;

[0010] S1: First, collect past data of the detection area, collect various data on the electricity used in the area, and arrange and integrate the power data provided by multiple regions according to time to ensure that there is complete power consumption at each time point. In this way, the previous power consumption is processed according to day, week, month, and year to form a structured and large-scale database;

[0011] S2: Taking solar power generation as an example, the output power of photovoltaic power stations will vary due to different weather conditions. The key factors that have a significant impact on photovoltaic power generation are accurately screened out from the original data to provide a solid data foundation and analysis basis for the subsequent establishment of a prediction model, thereby measuring the degree of correlation between the influencing factors at each time and the photovoltaic power generation. The specific calculation formula is:

[0012]

[0013] Where R ZU Indicates the degree of correlation between the two. represents the average value in element Z, represents the average value in element U;

[0014] Step 3: Develop a numerical weather forecast model system for weather processes, and continuously improve the ability of numerical weather forecasts to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance;

[0015] Step 4: Establish an online interactive power prediction method, and establish an online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution";

[0016] Step 5: Actively carry out research on the application of new technologies such as artificial intelligence in power prediction, and use computer vision technology based on machine learning and image processing to provide accurate basic information for the device.

[0017] Preferably, based on step 2, taking solar illumination power generation as an example, photovoltaic power generation uses solar panels to convert solar energy into electrical energy. The greater the solar radiation intensity, the more solar energy is received, and thus the more electrical energy is generated, so as to calculate the power generation conversion rate in different seasons. The specific calculation formula is as follows:

[0018] P = gSI[1-0.05(t0+25)]

[0019] Where P represents the conversion rate, g represents the photovoltaic cell power conversion efficiency, S represents the light contact area, I represents the irradiance, and t0 represents the temperature value;

[0020] According to the above-mentioned relevant detection data, the physical model and the statistical model are used simultaneously to predict the new energy power. The prediction results of the physical model are used as the input variables of the statistical model. The physical model is used to predict the long-term meteorological forecast data. On this basis, the statistical model is used to complete the prediction of wind power in the future.

[0021] Preferably, in step one, the system covers multiple links such as standardized data monitoring, collection, reporting, quality control, and installation, operation and maintenance and calibration of observation equipment, which is an important basic data guarantee for improving the accuracy of power forecasting. Through this standard system, the accuracy of meteorological observation data can be significantly improved, so that it can more accurately reflect the actual meteorological conditions, avoid missing or incomplete data, give full play to its application value, and provide strong data support in many fields such as electric power forecasting.

[0022] Preferably, in step 2, the theoretical power generation capacity evaluation method of new energy is improved. Through the theoretical power generation capacity evaluation, the output power data of new energy power stations during the power outage period is reproduced to form a sample set corresponding to the meteorological parameters and the power station output power, thereby providing complete data for prediction model training and model optimization, thereby improving the prediction accuracy.

[0023] Preferably, the step three combines the measured meteorological data of the new energy power station to carry out engineering practical research on the real-time four-dimensional variational assimilation technology, and continuously improves the ability of numerical weather forecasting to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance.

[0024] Preferably, the online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution" established in step 4 is of great significance for improving the responsiveness of the prediction model and improving the prediction accuracy.

[0025] Preferably, the machine learning and database data mining technology in step five improves the accuracy of short-term power prediction, the computer vision technology of machine learning plus image processing can provide accurate basic information for ultra-short-term photovoltaic minute-level prediction, and the pattern recognition technology can accurately identify weather types and power fluctuation types, laying the foundation for the practical application of power prediction technology and ramp prediction technology based on weather type classification.

[0026] Preferably, in s4, a short-term photovoltaic power generation power prediction combined model of a boost (XGBoost) model and a long short-term memory network (LSTM) model is used. According to the short-term photovoltaic power generation characteristics, an XGBoost model and an LSTM model are first established respectively, and then the XGBoost model is used to perform preliminary prediction to add features, and the inverse error method is used to solve the weight coefficients of the two models to calculate the final prediction result.

[0027] Preferably, in S1, an electricity consumption monitor is installed on the transmission line belonging to each power consumption area, and the monitoring function of the electricity consumption monitor is used to effectively monitor the electricity consumption of the corresponding power consumption area, thereby providing data support for subsequent power consumption management and other tasks.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present invention detects past power data and subsequently makes accurate predictions on power generation, power consumption and energy consumption on the power consumption side based on such data. It also combines the prediction of new energy power generation to flexibly formulate and implement power balance strategies, effectively solving the problems of poor power supply and demand matching and insufficient new energy power utilization in the prior art, and ensuring stable and efficient operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 Provides a schematic diagram of the overall system flow for an embodiment of the present invention;

[0031] Figure 2 A schematic diagram of the power and electricity balance of a product query unit provided in an embodiment of the present invention;

[0032] Figure 3 A schematic diagram of power prediction for a product query unit provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] 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.

[0034] See also Figure 1-Figure 3 The present invention provides a technical solution: a new energy power and quantity balancing method based on energy consumption prediction results, the steps of the power and quantity balancing method are:

[0035] Step 1: Improve the accuracy and completeness of meteorological observation data of new energy power plants, form a standard system for regulating data monitoring, collection, reporting, quality control, and installation, operation, maintenance and calibration of observation equipment, and improve the accuracy and completeness of meteorological observation data;

[0036] Step 2: Improve the theoretical power generation capacity evaluation method of new energy power stations, and reproduce the output power data of new energy power stations during power cuts through theoretical power generation capacity evaluation;

[0037] Based on step 2, considering that the amount of new energy power has random characteristics based on seasons, the power generation power of the power supply area is predicted;

[0038] S1: First, collect past data of the detection area, collect various data on the electricity used in the area, and arrange and integrate the power data provided by multiple regions according to time to ensure that there is complete power consumption at each time point. In this way, the previous power consumption is processed according to day, week, month, and year to form a structured and large-scale database;

[0039] S2: Taking solar power generation as an example, the output power of photovoltaic power stations will vary due to different weather conditions. The key factors that have a significant impact on photovoltaic power generation are accurately screened out from the original data to provide a solid data foundation and analysis basis for the subsequent establishment of a prediction model, thereby measuring the degree of correlation between the influencing factors at each time and the photovoltaic power generation. The specific calculation formula is:

[0040]

[0041] Where R ZU Indicates the degree of correlation between the two. represents the average value in element Z, represents the average value in element U;

[0042] Step 3: Develop a numerical weather forecast model system for weather processes, and continuously improve the ability of numerical weather forecasts to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance;

[0043] Step 4: Establish an online interactive power prediction method, and establish an online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution";

[0044] Step 5: Actively carry out research on the application of new technologies such as artificial intelligence in power prediction, and use computer vision technology with machine learning and image processing to provide accurate basic information for the device;

[0045] Based on step 2, taking solar power generation as an example, photovoltaic power generation uses solar panels to convert solar energy into electrical energy. The greater the solar radiation intensity, the more solar energy is received, and the more electrical energy is generated. The power generation conversion rate in different seasons is calculated. The specific calculation formula is as follows:

[0046] P = gSI[1-0.05(t0+25)]

[0047] Where P represents the conversion rate, g represents the photovoltaic cell power conversion efficiency, S represents the light contact area, I represents the irradiance, and t0 represents the temperature value;

[0048] According to the above-mentioned relevant test data, the new energy power is predicted by using both physical models and statistical models. The prediction results of the physical model are used as the input variables of the statistical model. The physical model is used to predict the long-term meteorological forecast data. On this basis, the statistical model is used to complete the prediction of wind power in the future.

[0049] In step 1, the system covers multiple links such as standardized data monitoring, collection, reporting, quality control, and installation, operation, maintenance and calibration of observation equipment. It is an important basic data guarantee for improving the accuracy of power forecasting. Through this standard system, the accuracy of meteorological observation data can be significantly improved, so that it can more accurately reflect the actual meteorological conditions, avoid missing or incomplete data, give full play to its application value, and provide strong data support in many fields such as power forecasting.

[0050] In step 2, the theoretical power generation capacity evaluation method of new energy is improved. Through the theoretical power generation capacity evaluation, the output power data of new energy power stations during the power curtailment period is reproduced to form a sample set corresponding to the meteorological parameters and the power station output power, providing complete data for prediction model training and model optimization, thereby improving the prediction accuracy;

[0051] Step 3: Combine the measured meteorological data of new energy power stations to carry out engineering practical research on real-time four-dimensional variational assimilation technology, and continuously improve the ability of numerical weather forecasting to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance;

[0052] The online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution" established in step 4 is of great significance for improving the responsiveness of the prediction model and improving the prediction accuracy;

[0053] In step 5, the machine learning and database data mining technology can improve the accuracy of short-term power prediction. The computer vision technology of machine learning plus image processing can provide accurate basic information for ultra-short-term photovoltaic minute-level prediction. The pattern recognition technology can accurately identify the weather type and power fluctuation type, laying the foundation for the practical application of power prediction technology and ramp prediction technology based on weather type classification.

[0054] In s4, a short-term photovoltaic power generation power prediction combined model of the boost (XGBoost) model and the long short-term memory network (LSTM) model is used. According to the short-term photovoltaic power generation characteristics, the XGBoost model and the LSTM model are first built respectively, and then the XGBoost model is used to perform preliminary prediction to add features, and the inverse error method is used to solve the weight coefficients of the two models to calculate the final prediction result;

[0055] In S1, a power monitor is installed on the transmission line belonging to each power consumption area. Through the monitoring function of the power monitor, the power consumption of the corresponding power consumption area can be effectively monitored, thereby providing data support for subsequent power consumption management and other work.

[0056] Working principle: First, collect past data of the detection area, and comprehensively collect various data information on the application of electricity in this area to various related electricity consumption. At the same time, the power data provided by multiple areas are carefully arranged and integrated in chronological order to ensure that complete power consumption data can be presented at each time node. In this way, the previous power consumption data is further processed systematically in units of days, weeks, months, and years to convert them into structured and large-scale databases. Taking solar power generation as an example, the output power of photovoltaic power stations will vary significantly due to different weather conditions. Therefore, it is necessary to accurately screen out the key factors that have a significant impact on photovoltaic power generation from the original data, so as to provide an effective data analysis basis for the subsequent establishment of a prediction model, and then use it to measure the degree of correlation between various time influencing factors and photovoltaic power generation. Photovoltaic power generation is the process of converting solar energy into electrical energy with the help of solar panels. In this process, solar radiation The greater the intensity, the more solar energy the solar panels receive, and the more electricity is generated accordingly. Based on this principle, the power conversion rate in different seasons and time periods can be calculated. According to the above-mentioned relevant detection data, the physical model and statistical model are used in combination to carry out the prediction of new energy power. Specifically, the prediction results of the physical model are used as the input variables of the statistical model. The long-term meteorological forecast data is predicted with the help of the physical model, and then the electric power in the future is predicted based on the statistical model. For the short-term photovoltaic power generation prediction, a combined model combining the boost (XGBoost) model and the long short-term memory network (LSTM) model is constructed. According to the characteristics of short-term photovoltaic power generation, the XGBoost model and the LSTM model are first constructed respectively, and then the XGBoost model is used for preliminary prediction to increase the features. Finally, the inverse error method is used to solve the weight coefficients of the two models, and then the final prediction result is calculated.

[0057] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0058] 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 new energy power balance method based on energy consumption prediction results, characterized in that: The steps of the power and quantity balancing method are as follows: Step 1: Improve the accuracy and completeness of meteorological observation data of new energy power plants, form a standard system for regulating data monitoring, collection, reporting, quality control, and installation, operation, maintenance and calibration of observation equipment, and improve the accuracy and completeness of meteorological observation data; Step 2: Improve the theoretical power generation capacity evaluation method of new energy power stations, and reproduce the output power data of new energy power stations during power cuts through theoretical power generation capacity evaluation; Based on step 2, considering that the amount of new energy power has random characteristics based on seasons, the power generation power of the power supply area is predicted; S1: First, collect past data of the detection area, collect various data on the electricity used in the area, and arrange and integrate the power data provided by multiple regions according to time to ensure that there is complete power consumption at each time point. In this way, the previous power consumption is processed according to day, week, month, and year to form a structured and large-scale database; S2: Taking solar power generation as an example, the output power of photovoltaic power stations will vary due to different weather conditions. The key factors that have a significant impact on photovoltaic power generation are accurately screened out from the original data to provide a solid data foundation and analysis basis for the subsequent establishment of a prediction model, thereby measuring the degree of correlation between the influencing factors at each time and photovoltaic power generation; Step 3: Develop a numerical weather forecast model system for weather processes, and continuously improve the ability of numerical weather forecasts to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance; Step 4: Establish an online interactive power prediction method, and establish an online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution"; Step 5: Actively carry out research on the application of new technologies such as artificial intelligence in power prediction, and use computer vision technology based on machine learning and image processing to provide accurate basic information for the device.

2. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: Based on step 2, it also includes: S3: Taking solar power generation as an example, photovoltaic power generation uses solar panels to convert solar energy into electrical energy. The greater the solar radiation intensity, the more solar energy is received, and thus the more electrical energy is generated. The power generation conversion rate in different seasons can be calculated; S4: According to the above-mentioned relevant monitoring data, the physical model and the statistical model are used simultaneously to predict the new energy power. The prediction results of the physical model are used as the input variables of the statistical model. The physical model is used to predict the long-term meteorological forecast data. On this basis, the statistical model is used to complete the prediction of wind power in the future.

3. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: In step 1, the system covers multiple links such as standardized data monitoring, collection, reporting, quality control, and installation, operation, maintenance and calibration of observation equipment. It is an important basic data guarantee for improving the accuracy of power forecasting. Through this standard system, the accuracy of meteorological observation data can be significantly improved, making it more accurately reflect the actual meteorological conditions, avoiding missing or incomplete data, and giving full play to its application value, providing strong data support in many fields such as electric power forecasting.

4. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: In the step 2, the theoretical power generation capacity evaluation method of new energy is improved. Through the theoretical power generation capacity evaluation, the output power data of new energy power stations during the power outage period is reproduced to form a sample set corresponding to the meteorological parameters and the power station output power, providing complete data for prediction model training and model optimization, thereby improving the prediction accuracy.

5. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized in that: The third step combines the measured meteorological data of new energy power stations to carry out engineering practical research on real-time four-dimensional variational assimilation technology, and continuously improves the ability of numerical weather forecasting to accurately predict the changing trends of key meteorological elements such as wind speed, wind direction, and irradiance.

6. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: The online interactive prediction method of "online modeling, real-time adjustment, and centralized distribution" established in step 4 is of great significance for improving the responsiveness of the prediction model and improving the prediction accuracy.

7. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: The machine learning and database data mining technology in step five can improve the accuracy of short-term power prediction, the computer vision technology of machine learning plus image processing can provide accurate basic information for ultra-short-term photovoltaic minute-level prediction, and the pattern recognition technology can accurately identify weather types and power fluctuation types, laying the foundation for the practical application of power prediction technology and ramp prediction technology based on weather type classification.

8. The new energy power balance method based on energy consumption prediction results according to claim 2 is characterized by: In s4, a short-term photovoltaic power generation power prediction combined model of the boost (XGBoost) model and the long short-term memory network (LSTM) model is used. According to the short-term photovoltaic power generation characteristics, the XGBoost model and the LSTM model are first established respectively, and then the XGBoost model is used to perform preliminary prediction to add features, and the inverse error method is used to solve the weight coefficients of the two models to calculate the final prediction result.

9. The new energy power balance method based on energy consumption prediction results according to claim 1 is characterized by: In S1, an electricity consumption monitor is installed on the transmission line belonging to each power consumption area. Through the monitoring function of the electricity consumption monitor, effective monitoring of the electricity consumption in the corresponding power consumption area is achieved, thereby providing data support for subsequent power consumption management and other tasks.

Images