Content authentication system for big data electric energy management
Through the conversion of deep convolutional inverse graphical network into an intelligent identification model, combined with the dual input single output formula, the problem of difficult management of the power generation system data combined with multiple power sources is solved, and the intelligent identification and controllability of future power generation data is achieved.
Patent Information
- Application Number
- CN202411200975.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The combined power generation system of multiple sources of electricity is difficult to predict and manage due to the complex power generation environment, and there is a lack of reliable basic data.
A content identification system for big data power management was designed, and the combined data of the target light energy and thermal power generation network was converted into an intelligent identification model through deep convolutional inverse graphics networks, and a data conversion formula with dual input and single output was used. It represents the relationship between the number of windsurfaces of the photoelectric power generation network and the rated power of the thermal power generation network and the number of days in the past, and realizes the targeted design of the intelligent identification model.
It provides intelligent identification of power generation data in the future period for the power generation network combination, and improves the controllability and management stability of the power generation network.
Abstract
Description
Technical Field
[0001] The present invention relates to the field of electric energy management, and more specifically, to a content identification system for big data electric energy management. Background Art
[0002] EMS (Energy Management System) is a set of power management systems suitable for low-voltage power distribution systems, which are developed according to user needs and in accordance with the standard specifications of power distribution systems. It has the characteristics of strong professionalism, high degree of automation, ease of use, high performance, and high reliability. Through telemetry and remote control, loads can be reasonably allocated, optimized operation can be achieved, and electricity can be effectively saved. There are records of peak and valley electricity consumption, which provides the necessary conditions for energy management. At the same time, electricity is measured by item according to the electricity consumption of open sockets, power, air conditioning, and special electricity, providing a basis for energy energy conservation audits of enterprises and institutions.
[0003] Generally speaking, for an electric energy management system, its electric energy may come from multiple aspects, such as solar power generation and thermal power generation. However, for a combined power generation system with multiple electric energy sources, the power generation environment of each power generation method is relatively complex, resulting in a more complex power generation environment for multiple power generation methods. The power generation data for future periods is even more unpredictable, which makes the power generation management of the combined power generation system lack reliable basic data. Summary of the invention
[0004] In order to solve the technical problems in the related fields, the present invention provides a content identification system for big data power management, which detects various combination network data of the power generation network combination for a power generation network combination consisting of a target solar power generation network and a target thermal power generation network, wherein the various combination network data of the power generation network combination are multiple network data of the target solar power generation network of the power generation network combination and multiple network data of the target thermal power generation network, the multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination of the preset time period of the target solar power generation network of the power generation network combination, and the multiple network data of the target thermal power generation network of the power generation network combination. The network data includes the power factor, rated power, fan opening, fan active power and fan operating speed of the target thermal power generation network of the power generation network combination. At the same time, a corresponding intelligent identification model is designed for the intelligent identification of the power generation data of the power generation network combination in the future period. Specifically, each learning operation is performed on the deep convolutional inverse graphics network to complete the conversion from the deep convolutional inverse graphics network to the intelligent identification model, and a dual-input and single-output data conversion formula is used to represent the joint conversion relationship between the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network and the number of days in the past before the current day, thereby completing the targeted design of the model structure of the intelligent identification model.
[0005] According to the present invention, a content identification system for big data power management is provided, the system comprising:
[0006] A model conversion device, used for converting a deep convolutional inverse graphics network into an intelligent identification model, wherein converting the deep convolutional inverse graphics network into the intelligent identification model comprises: performing each learning operation on the deep convolutional inverse graphics network to complete the conversion of the deep convolutional inverse graphics network into the intelligent identification model;
[0007] A combination detection device is used to detect various combination network data of a power generation network combination consisting of a target solar power generation network and a target thermal power generation network, wherein the various combination network data of the power generation network combination are multiple network data of the target solar power generation network and multiple network data of the target thermal power generation network of the power generation network combination;
[0008] At the same time, an extraction device is used to obtain the power generation values of the power generation network combination corresponding to the preset time periods in the past days before the current day, and the number of days in the past days before the current day is positively correlated with the number of sailboards in the target solar power generation network and positively correlated with the rated power of the target thermal power generation network;
[0009] A content identification mechanism is connected to the model conversion device, the combination detection device and the simultaneous extraction device respectively, and is used to use the intelligent identification model to intelligently identify the power generation value corresponding to the power generation combination in the preset time period of the day based on the various combination network data of the power generation network combination and the power generation values corresponding to the power generation combination in the past days before the day;
[0010] A wireless notification mechanism, connected to the content identification mechanism, is used to send the obtained generated electric energy value corresponding to the preset time period of the power generation network combination on the day together with the preset time period on the day to the big data management network element that performs electric energy dispatch management;
[0011] The simultaneous extraction device is used to obtain the respective generated electric energy values corresponding to the preset time periods of the past days of the power generation network combination before the current day, and the number of days before the current day is positively correlated with the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network, including: using a double-input single-output data conversion formula to represent the joint conversion relationship between the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network and the number of days before the current day;
[0012] Among them, for the power generation network combination composed of the target solar power generation network and the target thermal power generation network, various combination network data of the power generation network combination are detected, and the various combination network data of the power generation network combination are multiple network data of the target solar power generation network of the power generation network combination and multiple network data of the target thermal power generation network, including: the multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination in a preset period of time of the target solar power generation network of the power generation network combination; the multiple network data of the target thermal power generation network of the power generation network combination are the power factor, rated power, fan opening, fan active power and fan operating speed of the target thermal power generation network of the power generation network combination.
[0013] Therefore, the present invention has at least the following three beneficial technical effects:
[0014] First, for a power generation network combination consisting of a target solar power generation network and a target thermal power generation network, various combination network data of the power generation network combination are detected, wherein the various combination network data of the power generation network combination include multiple network data of the target solar power generation network and multiple network data of the target thermal power generation network of the power generation network combination, thereby providing effective basic data for intelligent identification of power generation data of the power generation network combination in future time periods;
[0015] Second place: Specifically, the multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination of the preset time period of the target solar power generation network of the power generation network combination, and the multiple network data of the target thermal power generation network of the power generation network combination are the power factor, rated power, fan opening, fan active power and fan operating speed of the target thermal power generation network of the power generation network combination;
[0016] Third: a corresponding intelligent identification model is designed for the intelligent identification of the power generation data of the power generation network combination in the future period. Specifically, each learning operation is performed on the deep convolutional inverse graphics network to complete the conversion from the deep convolutional inverse graphics network to the intelligent identification model, and a dual-input and single-output data conversion formula is used to represent the joint conversion relationship between the number of panels of the target solar power generation network and the rated power of the target thermal power generation network and the number of days before the current day, thereby completing the targeted design of the model structure of the intelligent identification model.
[0017] The content identification system for big data power management of the present invention has stable operation and intelligent control. Due to the power generation network combination consisting of the target solar power generation network and the target thermal power generation network, a targeted intelligent identification model is used based on the various combined network data of the power generation network combination to intelligently identify the power generation data of the power generation network combination in the future period, thereby improving the controllability of the power generation network combination. DETAILED DESCRIPTION
[0018] The implementation scheme of the content identification system for big data power management of the present invention will be described in detail below.
[0019] A content identification system for big data power management according to a first embodiment of the present invention includes:
[0020] A model conversion device, used for converting a deep convolutional inverse graphics network into an intelligent identification model, wherein converting the deep convolutional inverse graphics network into the intelligent identification model comprises: performing each learning operation on the deep convolutional inverse graphics network to complete the conversion of the deep convolutional inverse graphics network into the intelligent identification model;
[0021] Specifically, the model conversion device is used to convert the deep convolutional inverse graphics network into the intelligent identification model, wherein the conversion of the deep convolutional inverse graphics network into the intelligent identification model includes: performing each learning operation on the deep convolutional inverse graphics network to complete the conversion of the deep convolutional inverse graphics network to the intelligent identification model includes: selecting a numerical simulation mode to implement the test and simulation of the conversion process of the deep convolutional inverse graphics network into the intelligent identification model;
[0022] A combination detection device is used to detect various combination network data of a power generation network combination consisting of a target solar power generation network and a target thermal power generation network, wherein the various combination network data of the power generation network combination are multiple network data of the target solar power generation network and multiple network data of the target thermal power generation network of the power generation network combination;
[0023] At the same time, an extraction device is used to obtain the power generation values of the power generation network combination corresponding to the preset time periods in the past days before the current day, and the number of days in the past days before the current day is positively correlated with the number of sailboards in the target solar power generation network and positively correlated with the rated power of the target thermal power generation network;
[0024] A content identification mechanism is connected to the model conversion device, the combination detection device and the simultaneous extraction device respectively, and is used to use the intelligent identification model to intelligently identify the power generation value corresponding to the power generation combination in the preset time period of the day based on the various combination network data of the power generation network combination and the power generation values corresponding to the power generation combination in the past days before the day;
[0025] A wireless notification mechanism, connected to the content identification mechanism, is used to send the obtained generated electric energy value corresponding to the preset time period of the power generation network combination on the day together with the preset time period on the day to the big data management network element that performs electric energy dispatch management;
[0026] The simultaneous extraction device is used to obtain the respective generated electric energy values corresponding to the preset time periods of the past days of the power generation network combination before the current day, and the number of days before the current day is positively correlated with the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network, including: using a double-input single-output data conversion formula to represent the joint conversion relationship between the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network and the number of days before the current day;
[0027] Among them, for the power generation network combination composed of the target solar power generation network and the target thermal power generation network, various combination network data of the power generation network combination are detected, and the various combination network data of the power generation network combination are multiple network data of the target solar power generation network of the power generation network combination and multiple network data of the target thermal power generation network, including: multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination of the target solar power generation network of the power generation network combination during a preset period, and multiple network data of the target thermal power generation network of the power generation network combination are the power factor, rated power, fan opening, fan active power and fan operating speed of the target thermal power generation network of the power generation network combination;
[0028] Among them, the multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination of the preset time period of the target solar power generation network of the power generation network combination, including: the overall illumination of the preset time period is the arithmetic mean of the remaining multiple light illuminations after removing the maximum and minimum values of the light illuminations corresponding to each moment evenly spaced within the preset time period.
[0029] Different from the first embodiment of the present invention, the content identification system for big data power management in the second embodiment of the present invention may further include the following components:
[0030] An operation measurement mechanism, connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, respectively, for measuring the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively;
[0031] Wherein, the operation measurement mechanism is respectively connected with the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, and is used to respectively measure the number of unit time operations of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, including: the operation measurement mechanism includes a plurality of operation measurement units, which are respectively connected with the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, so as to complete the measurement of the number of unit time operations of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device;
[0032] Wherein, the operation measurement mechanism includes a plurality of operation measurement units, which are used to be connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively, so as to complete the measurement of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively, including: the plurality of operation measurement units are a plurality of operation sensor circuits, which are used to be connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively, so as to complete the measurement of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively;
[0033] The plurality of operation measurement units are a plurality of operation sensing circuits, which are respectively connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device to respectively measure the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, including: the plurality of operation sensing circuits have the same structure;
[0034] And wherein, the multiple operation measurement units are multiple operation sensor circuits, which are used to respectively connect with the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device to complete the respective measurements of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, and also include: the multiple operation sensor circuits have the same operation measurement upper limit value and operation measurement lower limit value.
[0035] Different from the first embodiment of the present invention, the content identification system for big data power management in the third embodiment of the present invention may further include the following components:
[0036] A base structure, arranged below the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device, for providing bottom support services for the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device respectively;
[0037] Wherein, the base structure is arranged below the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device, and is used to provide bottom support services for the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device respectively, including: a plurality of slots are arranged on the base structure to respectively accommodate the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device;
[0038] And wherein, a plurality of slots are arranged on the base structure to respectively accommodate the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, including: the opening cross-sectional areas of the plurality of slots are the same to respectively accommodate the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device.
[0039] In addition, in the content identification system for big data power management, a dual-input single-output data conversion formula is used to represent the joint conversion relationship between the number of sailboards in the target solar power generation network and the rated power of the target thermal power generation network and the number of days in the past before the current day, including: in the dual-input single-output data conversion formula, the number of sailboards in the target solar power generation network and the rated power of the target thermal power generation network are two input data of the dual-input single-output data conversion formula, and the number of days in the past before the current day is a single output data of the dual-input single-output data conversion formula.
[0040] While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the following claims is to be accorded the broadest interpretation so as to encompass all modifications and equivalent structures and functions.
Claims
1. A content identification system for big data power management, characterized in that: The system comprises: A model conversion device, used for converting a deep convolutional inverse graphics network into an intelligent identification model, wherein converting the deep convolutional inverse graphics network into the intelligent identification model comprises: performing each learning operation on the deep convolutional inverse graphics network to complete the conversion of the deep convolutional inverse graphics network into the intelligent identification model; A combination detection device is used to detect various combination network data of a power generation network combination consisting of a target solar power generation network and a target thermal power generation network, wherein the various combination network data of the power generation network combination are multiple network data of the target solar power generation network and multiple network data of the target thermal power generation network of the power generation network combination; At the same time, an extraction device is used to obtain the power generation values of the power generation network combination corresponding to the preset time periods in the past days before the current day, and the number of days in the past days before the current day is positively correlated with the number of sailboards in the target solar power generation network and positively correlated with the rated power of the target thermal power generation network; A content identification mechanism is connected to the model conversion device, the combination detection device and the simultaneous extraction device respectively, and is used to use the intelligent identification model to intelligently identify the power generation value corresponding to the power generation combination in the preset time period of the day based on the various combination network data of the power generation network combination and the power generation values corresponding to the power generation combination in the past days before the day; A wireless notification mechanism, connected to the content identification mechanism, is used to send the obtained generated electric energy value corresponding to the preset time period of the power generation network combination on the day together with the preset time period on the day to the big data management network element that performs electric energy dispatch management; The simultaneous extraction device is used to obtain the respective generated electric energy values corresponding to the preset time periods of the past days of the power generation network combination before the current day, and the number of days before the current day is positively correlated with the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network, including: using a double-input single-output data conversion formula to represent the joint conversion relationship between the number of sailboards of the target solar power generation network and the rated power of the target thermal power generation network and the number of days before the current day; Among them, for the power generation network combination composed of the target solar power generation network and the target thermal power generation network, various combination network data of the power generation network combination are detected, and the various combination network data of the power generation network combination are multiple network data of the target solar power generation network of the power generation network combination and multiple network data of the target thermal power generation network, including: the multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination in a preset period of time of the target solar power generation network of the power generation network combination; the multiple network data of the target thermal power generation network of the power generation network combination are the power factor, rated power, fan opening, fan active power and fan operating speed of the target thermal power generation network of the power generation network combination.
2. The content identification system for big data power management according to claim 1, characterized in that: The multiple network data of the target solar power generation network of the power generation network combination are the number of sailboards, the occupied area and the overall illumination of the preset time period of the target solar power generation network of the power generation network combination, including: the overall illumination of the preset time period is the arithmetic mean of the remaining multiple light illuminations after removing the maximum and minimum values of the light illuminations corresponding to each moment evenly spaced within the preset time period.
3. The content identification system for big data power management according to claim 2, characterized in that: The system further comprises: An operation measurement mechanism, connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, respectively, for measuring the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively; Among them, the operation measurement mechanism is respectively connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, and is used to respectively measure the number of unit time operations of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device. The operation measurement mechanism includes multiple operation measurement units, which are used to respectively connect to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device to complete the respective measurement of the number of unit time operations of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device.
4. The content identification system for big data power management according to claim 3, characterized in that: The operation measurement mechanism includes a plurality of operation measurement units, which are used to respectively connect with the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device to complete the respective measurement of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device. The plurality of operation measurement units are a plurality of operation sensor circuits, which are used to respectively connect with the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device to complete the respective measurement of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device.
5. The content identification system for big data power management according to claim 4, characterized in that: The multiple operation measurement units are multiple operation sensing circuits, which are used to be connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively, so as to complete the respective measurements of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, including: the structures of the multiple operation sensing circuits are the same.
6. The content identification system for big data power management according to claim 5, characterized in that: The multiple operation measurement units are multiple operation sensor circuits, which are used to be connected to the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device respectively, so as to complete the respective measurements of the number of operations per unit time of the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device. It also includes: the multiple operation sensor circuits have the same operation measurement upper limit value and operation measurement lower limit value.
7. The content identification system for big data power management according to claim 2, characterized in that: The system further comprises: The base structure is arranged below the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device, and is used to provide bottom support services for the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device respectively.
8. The content identification system for big data power management according to claim 7, characterized in that: The base structure is arranged below the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device, and is used to provide bottom support services for the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device respectively, including: a plurality of slots are arranged on the base structure to respectively accommodate the content identification mechanism, the model conversion device, the combined detection device and the simultaneous extraction device.
9. The content identification system for big data power management according to claim 8, characterized in that: A plurality of slots are arranged on the base structure to respectively accommodate the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device, including: the opening cross-sectional areas of the plurality of slots are the same to respectively accommodate the content identification mechanism, the model conversion device, the combination detection device and the simultaneous extraction device.
Citation Information
Patent Citations
Solar panel dust detection and power influence evaluation method and device
CN117710342A
Multi-source power dispatching method and system based on big data
CN117767446A
Management method for carbon emission data of power distribution network
CN118195094A
