Power load prediction and optimization management method and system based on artificial intelligence

Abstract

The invention relates to a power load prediction and optimization management method and system based on artificial intelligence, and belongs to the technical field of power system management.Various data of a power system are collected, then the data are processed, a machine learning model is trained through historical data, then the existing data are analyzed, and the power load prediction and optimization management method and system based on artificial intelligence are obtained. According to the method, the power load is predicted, an optimization scheme of power management is provided through prediction, and operation can be carried out on a display screen, so that the stability and resource utilization efficiency of a power system are improved, the method has the advantages of accuracy, intelligence, real-time performance, portability, easiness in operation and the like, and the load prediction and management efficiency of the power system can be effectively improved. Wide application prospects are realized.

Description

Technical Field

[0001] The present invention relates to a method and system for power load forecasting and optimization management based on artificial intelligence, belonging to the technical field of power system management. Background Art

[0002] In modern power systems, accurate load forecasting and optimal management are crucial to the stable operation of power systems. Existing load forecasting methods mainly rely on historical data and simple statistical models, which are difficult to cope with complex and changing power demand. At the same time, the optimization of power management often relies on experience and rules, lacks intelligent means, and leads to low efficiency in the utilization of power resources. Therefore, there is an urgent need for an intelligent tool that uses artificial intelligence technology to forecast and optimize power load management in order to improve forecast accuracy and management efficiency. Summary of the invention

[0003] According to the problem described in the background, the problem to be solved by the present invention is: to provide a method and system for power load forecasting and optimization management based on artificial intelligence, which can accurately predict the power load and provide optimized power management solutions, thereby improving the stability of the power system and resource utilization efficiency.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for power load forecasting and optimization management based on artificial intelligence is provided, comprising the following steps:

[0005] (1) The data acquisition software controls the operation of the sensors and transmits the collected data on the operation of the power system to the prediction algorithm;

[0006] (2) Analyze the collected data and predict future power loads through the use of machine learning and deep learning algorithms through artificial intelligence prediction algorithms;

[0007] (3) Providing an optimized power management solution through an optimization management algorithm based on the results of predicting future power loads;

[0008] (4) Through a user-friendly interface, users can view all data in real time, as well as the forecast results of future power loads and the plan for optimizing power management, and make relevant settings;

[0009] (5) Finally, all relevant data are uploaded to the cloud for long-term storage and analysis, generating historical reports and trend analysis to further optimize management.

[0010] The step (2) is further divided into the following steps:

[0011] (2.1) Data preprocessing: cleaning and standardizing the collected data;

[0012] (2.2) Perform feature extraction on the data to extract key features from the cleaned and processed data;

[0013] (2.3) Train machine learning models using historical information and data;

[0014] (2.4) The trained machine learning model is used to predict future loads by extracting key features to form a power load curve; the power load curve is a comparison of the baseline load and the real-time load at different times.

[0015] After the data collection in step (2) is completed, the current real-time load and the current seasonal average load are plotted through the data collection software, and the load curve is further judged, and then decomposed according to different customer types; then, based on whether the customer is a new customer, the small base of the initial load and the large changes, and the industry in which the customer is located, the AI ​​is trained on data learning in combination with the data preprocessing feature type to further complete the load forecast and prepare for early warning and early planning for load scheduling and resource allocation.

[0016] The step (3) is further divided into the following steps:

[0017] (3.1) Load dispatching: reasonably dispatching the power load through the load curve of power load forecasting;

[0018] (3.2) Optimize the allocation of power resources through allocation strategies;

[0019] (3.3) Then, through demand response, intelligent control is performed to meet demand to achieve a balance between supply and demand.

[0020] The present invention also provides a system for power load prediction and optimization management based on artificial intelligence, including a data acquisition module, a data processing module, a communication module, a display module and a power supply module. The data acquisition module collects parameters such as voltage, current, power and load, and transmits various parameters to the data processing module. The data processing module processes and stores the collected data, transmits the data to the display module through the communication module, and can also remotely monitor the data.

[0021] The data acquisition module is specifically a sensor module, including a voltage sensor, a current sensor and a power sensor, and is provided with a sensor probe, through which data is collected.

[0022] The data processing module includes a high-performance processor and a large-capacity storage device to quickly process and store the collected data.

[0023] The communication module includes wireless communication and wired communication. The wireless communication can perform data transmission and remote monitoring through Wi-Fi, 4G and 5G, and the wired communication can perform data transmission and remote monitoring through Ethernet.

[0024] The display module includes a high-definition touch screen for displaying prediction results and adjusting and optimizing management.

[0025] The power module includes a lithium battery and solar charging. The lithium battery supplies power to all other parts and can also be charged through solar energy and a charging port.

[0026] Preferably, the large-capacity storage device is also equipped with a USB interface so that data can be copied.

[0027] The beneficial effects of the present invention are:

[0028] 1. Improved accuracy: Use artificial intelligence technology to improve the accuracy of load forecasting.

[0029] 2. Intelligent process, built-in optimization management algorithm, can provide intelligent power management solutions.

[0030] 3. Real-time content: real-time collection and processing of power system data, and timely provision of prediction and optimization results.

[0031] 4. Improve portability, simple system design, easy to operate and monitor on site.

[0032] 5. Improved operability, providing a simple and clear user interface, which can be operated by ordinary users without professional training.

[0033] 6. Improved work efficiency. Load forecasting and optimal management of the power system improve the efficiency of staff.

[0034] 7. It has broad future prospects and can adapt to a variety of power systems. The results of independent learning are remarkable and the future prospects are broad. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the steps of the method of the present invention;

[0036] Figure 2 It is a schematic diagram of the process of the system of the present invention;

[0037] Figure 3 This is a schematic diagram of step 2;

[0038] Figure 4 This is a schematic diagram of step 3;

[0039] Figure 5 is a schematic diagram of a data acquisition module;

[0040] Figure 6 is a schematic diagram of a data processing module;

[0041] Figure 7 is a schematic diagram of a communication module;

[0042] Figure 8 is a schematic diagram of a display module;

[0043] Fig. 9 is a schematic diagram of a power module; DETAILED DESCRIPTION

[0044] The embodiments of the present invention are further described below in conjunction with the accompanying drawings:

[0045] Example 1

[0046] like Figure 1 As shown, the present invention comprises the following steps:

[0047] (1) The data acquisition software controls the operation of the sensor and transmits the collected data on the operation of the power system to the prediction algorithm. The collected data includes current, voltage, load, power consumption, etc. The data is comprehensive and the prediction is more accurate.

[0048] (2) Through the use of artificial intelligence prediction algorithms on machine learning and deep learning algorithms, the collected data is analyzed to predict future power loads. During the analysis, the data is cleaned and processed in advance to avoid excessive data redundancy, eliminate irrelevant data, reduce duplicate data, form a power load curve, and analyze the data more directly;

[0049] (3) Based on the results of predicting future power load, an optimized power management solution is provided through an optimization management algorithm, and power is reasonably optimized and allocated through the power load curve;

[0050] (4) Through a friendly operation interface, users can view all data in real time, as well as the forecast results of future power loads and the plan for optimizing power management, and make relevant settings. Remote monitoring can also be achieved, and the current problems of the power system and the optimization management results of the treatment plan can be clearly seen at a glance;

[0051] (5) Finally, all relevant data is uploaded to the cloud for long-term storage and analysis, and historical reports and trend analysis are generated to further optimize management and prevent periodic errors. It can also be compared with previous cases and certain measures can be taken manually.

[0052] like Figure 3 As shown, step (2) is further specifically divided into the following steps:

[0053] (2.1) Data preprocessing: cleaning and standardizing the collected data. Do not simply use all the data for prediction and eliminate invalid data;

[0054] (2.2) Extract features from the data, extract key features from the cleaned and processed data, and classify the key features for easy searching and supervision;

[0055] (2.3) Training machine learning models through historical information and data. Using historical information can better target different problems in different power systems and make it easier to provide prediction solutions;

[0056] (2.4) The trained machine learning model is used to predict future loads by extracting key features to form a power load curve for better analysis of existing problems; the power load curve is a comparison of baseline load and real-time load at different times.

[0057] After the data collection in step (2) is completed, the current real-time load and the current seasonal average load are plotted through the data collection software, and the load curve is further judged. Then, it is decomposed according to different types of customers, such as: whether the daily load fluctuation is stable (whether it is a shift continuous production enterprise), whether the annual load is stable (whether it is a year-round production enterprise or an order-based production enterprise), and feature extraction is performed on the three time periods of morning, middle and evening peaks to facilitate further load management; then, according to whether the customer is a new customer, the initial load base is small and the change is large, and the industry (such as: the load of chemical customers is relatively stable; the load fluctuation of customers such as sand mining, construction and building materials is large and the regularity is weak; office, hotel, supermarket and other types of customers have a strong time correlation), combined with the data preprocessing feature type, the AI ​​is trained for data learning to further complete the load forecast, and prepare for early warning and early planning for load scheduling and resource allocation.

[0058] like Figure 4 As shown, step (3) is further specifically divided into the following steps:

[0059] (3.1) Load dispatching: rationally dispatching the power load through the load curve predicted by the power load, so as to maximize the power, adjust the efficiency and improve the power configuration;

[0060] (3.2) Optimize the allocation of power resources through allocation strategies. The power demand in different time periods is also different, which can better optimize supply and demand;

[0061] (3.3) Then, through demand response, intelligent control is used to meet demand to achieve a balance between supply and demand. The balance between supply and demand is not a constant point, but an equilibrium representation of all electricity, a dynamic balance.

[0062] like Figure 2 As shown, the present invention includes a data acquisition module, a data processing module, a communication module, a display module and a power module. The data acquisition module collects parameters such as voltage, current, power and load, and transmits various parameters to the data processing module. The data processing module processes and stores the collected data, transmits the data to the display module through the communication module, and can also remotely monitor the data.

[0063] like Figure 5 As shown, the data acquisition module is specifically a sensor module, including a voltage sensor, a current sensor and a power sensor, and is provided with a sensor probe, through which data is collected.

[0064] like Figure 6 As shown, the data processing module includes a high-performance processor and a large-capacity storage device to quickly process and store the collected data.

[0065] like Figure 7 As shown, the communication module includes wireless communication and wired communication. The wireless communication can perform data transmission and remote monitoring through Wi-Fi, 4G and 5G, and the wired communication can perform data transmission and remote monitoring through Ethernet.

[0066] The large-capacity storage device is also equipped with a USB interface, so that data can be copied.

[0067] like Figure 8 As shown, the display module includes a high-definition touch screen for displaying prediction results and adjusting and optimizing management.

[0068] like Fig. 9 As shown, the power module includes a lithium battery and solar charging. The lithium battery supplies power to all other parts and can also be charged through solar energy and a charging port.

Claims

1. A method for power load forecasting and optimization management based on artificial intelligence, characterized in that: The following steps are involved: (1) The data acquisition software controls the operation of the sensors and transmits the collected data on the operation of the power system to the prediction algorithm; (2) Analyze the collected data and predict future power loads through the use of machine learning and deep learning algorithms through artificial intelligence prediction algorithms; (3) Providing an optimized power management solution through an optimization management algorithm based on the results of predicting future power loads; (4) Through a user-friendly interface, users can view all data in real time, as well as the forecast results of future power loads and the plan for optimizing power management, and make relevant settings; (5) Finally, all relevant data are uploaded to the cloud for long-term storage and analysis, generating historical reports and trend analysis to further optimize management.

2. According to the method of power load forecasting and optimization management based on artificial intelligence according to claim 1, it is characterized in that: The step (2) is further divided into the following steps: (2.1) Data preprocessing: cleaning and standardizing the collected data; (2.2) Perform feature extraction on the data to extract key features from the cleaned and processed data; (2.3) Train machine learning models using historical information and data; (2.4) The trained machine learning model is used to predict future loads by extracting key features to form a power load curve.

3. The method for power load forecasting and optimization management based on artificial intelligence according to claim 1 is characterized in that: The step (3) is further divided into the following steps: (3.1) Load dispatching: reasonably dispatching the power load through the load curve of power load forecasting; (3.2) Optimize the allocation of power resources through allocation strategies; (3.3) Then, through demand response, intelligent control is performed to meet demand to achieve a balance between supply and demand.

4. A system for power load forecasting and optimization management based on artificial intelligence, characterized in that: It includes a data acquisition module, a data processing module, a communication module, a display module and a power supply module. The data acquisition module collects parameters such as voltage, current, power and load, and passes various parameters to the data processing module. The data processing module processes and stores the collected data, transmits the data to the display module through the communication module, and can also remotely monitor the data.

5. The system for power load forecasting and optimization management based on artificial intelligence according to claim 4 is characterized in that: The data acquisition module is specifically a sensor module, including a voltage sensor, a current sensor and a power sensor, and is provided with a sensor probe, through which data is collected.

6. The system for power load forecasting and optimization management based on artificial intelligence according to claim 4 is characterized in that: The data processing module includes a high-performance processor and a large-capacity storage device to quickly process and store the collected data.

7. The system for power load forecasting and optimization management based on artificial intelligence according to claim 4 is characterized in that: The communication module includes wireless communication and wired communication. The wireless communication can perform data transmission and remote monitoring through Wi-Fi, 4G and 5G, and the wired communication can perform data transmission and remote monitoring through Ethernet.

8. The system for power load forecasting and optimization management based on artificial intelligence according to claim 4 is characterized in that: The display module includes a high-definition touch screen for displaying prediction results and adjusting and optimizing management.

9. The system for power load forecasting and optimization management based on artificial intelligence according to claim 4 is characterized in that: The power module includes a lithium battery and solar charging. The lithium battery supplies power to all other parts and can also be charged through solar energy and a charging port.

10. The system for power load forecasting and optimization management based on artificial intelligence according to claim 7, characterized in that: The large-capacity storage device is also equipped with a USB interface, so that data can be copied.

Images