Energy management system of multi-energy complementary smart energy based on source network load storage

Abstract

The invention discloses an energy management system of multi-energy complementary smart energy based on source network load storage, which relates to the technical field of energy scheduling management and comprises an intelligent sensing unit, an intelligent control unit and a data management unit. According to the invention, energy supply data of an energy source system and various energy control systems in a park and an enterprise are obtained through an intelligent sensing unit, and energy medium production, consumption and cost data are collected in real time, so that an energy plan scheduling scheme is obtained; the system is used for realizing digitization, informatization, visualization, predication and intelligentization of the whole process of enterprise energy supply, production, conveying, conversion and consumption, realizing energy management of whole-plant energy data acquisition full coverage and enterprise main energy consumption system analysis, optimization and auxiliary decision making, improving enterprise energy utilization efficiency and improving enterprise energy utilization efficiency. The enterprise energy efficiency maximization, the energy efficiency online optimization and the important information online pushing can be realized, and the goals of energy conservation, emission reduction, cost reduction, efficiency improvement and management and control level improvement are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of energy dispatching and management, and in particular to an energy management system based on multi-energy complementary smart energy of source, grid, load and storage. Background Art

[0002] Optimizing and adjusting the energy structure is not only a critical task for my country's energy development but also a crucial component in ensuring energy security. Optimizing energy requires regulating and coordinating the distribution and generation plans of various energy sources to improve energy utilization. However, due to the limitations of traditional energy analysis techniques, regional energy complementarity optimization is low, individual power plants lack refined energy analysis, and carbon emissions and cost considerations are not considered, making sustainable energy development difficult to achieve.

[0003] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0004] The purpose of this invention is to realize the digitization, informatization, visualization, prediction and intelligence of the entire process of enterprise energy supply, production, transportation, conversion and consumption, and at the same time realize full coverage of energy data collection in the whole plant and energy management of the main energy consumption systems of the enterprise that integrates analysis, optimization and decision-making assistance, improve the energy utilization efficiency of the enterprise, maximize the energy efficiency of the enterprise, optimize energy efficiency online, and push important information online, so as to achieve the goals of energy conservation and emission reduction, cost reduction and efficiency improvement, and improve the level of management and control.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions: an energy management system based on multi-energy complementary smart energy of source, grid, load and storage, comprising an intelligent sensing unit, an intelligent control unit and a data management unit;

[0006] The intelligent sensing unit uses the Industrial Internet of Things to build energy consumption systems, production systems, energy storage systems, detection systems, and monitoring systems in various areas of the park and enterprise site, achieving comprehensive perception of the factory's energy flow data;

[0007] Based on data from the intelligent perception layer, the intelligent control unit constructs a centralized energy control system for each production and auxiliary area of ​​the park or enterprise, centrally and uniformly controls, dispatches, and directs energy flows, creating an integrated centralized control center. Specifically, it includes a lighting energy dispatching module, a power energy dispatching module, a heating energy dispatching module, and other energy dispatching modules.

[0008] The intelligent perception layer includes lighting equipment, heating equipment, power equipment, energy-consuming equipment, substations, heat exchange stations, photovoltaic and air energy production capacity, as well as detection and monitoring systems;

[0009] The data management unit includes an energy consumption control module and a production capacity and energy storage control module;

[0010] The energy consumption control module is used to divide the various areas of the park and enterprise site into modules to obtain several power supply demand areas, obtain power supply data within each power supply demand area, draw a power supply demand table based on the power consumption cycle, and obtain corresponding carbon emission data and electricity cost data to improve the power supply demand table. Based on the convolutional neural network, a power consumption prediction model is constructed to obtain total power supply forecast data within the power supply demand area;

[0011] The production capacity and energy storage management and control module obtains energy supply data from energy source systems (substations, heat exchange stations, photovoltaic capacity and air energy capacity) and various energy control systems within the park and the enterprise (lighting equipment, heating equipment, power equipment, energy-consuming equipment, detection systems and monitoring systems) through the intelligent sensing unit, and collects energy medium production, consumption and cost data in real time, calculates the energy consumption coefficient, imports power supply demand characteristics, energy supply data and energy consumption coefficient into the energy analysis model to perform energy complementary optimization calculation, obtains the energy plan scheduling plan and sends it to the intelligent control unit;

[0012] Furthermore, the data management unit also includes an IaaS layer, a PaaS layer, and an energy data platform, where:

[0013] The IaaS layer is used to achieve hardware resource integration and virtualization, forming an elastic resource pool, performing resource segmentation, resource allocation, and resource integration. It provides servers, storage, backup, network, and other resources for the energy management system. Based on actual business needs, the energy management system can flexibly dispatch pooled resources.

[0014] The PaaS layer serves as a unified support and integration platform for applications. It integrates the resources provided by the IaaS layer downward and supports application development and operation upward. Based on a containerized structure, it provides automated tools for development, testing, deployment, and operation and maintenance for top-level API and app development. It also provides highly available, clustered databases and various middleware resources, empowering project management platforms with application deployment and development as well as middleware management capabilities.

[0015] The energy data platform is used to collect and store real-time data scattered across various energy application systems and real-time database systems, and to perform standardized management of real-time data. By extracting, cleaning, converting, and storing it, it forms a data warehouse based on master data. Relying on the distributed computing and big data analysis capabilities of the industrial data platform, it provides external data analysis services to support real-time computing and scientific decision-making in smart factories.

[0016] Furthermore, the energy data platform collects energy data from energy flow-related PLCs, DCSs, intelligent equipment, sensors, and other underlying devices into a real-time database through OPC, MODBUS, IEC101 / 104 protocols, RS232 / 485 interface standards, and proprietary protocol customized parsing methods. The energy data is integrated and processed in the real-time database, and the energy data is transmitted to a relational database for logical calculation, matching association, and business integration, and provided to the factory production end and various energy-consuming application ends. Specifically, the energy data platform includes a data acquisition layer, a data processing layer, a data storage layer, a data analysis layer, a data sharing layer, and a data visualization layer, wherein:

[0017] The data collection layer is used to comprehensively collect data from park or production site equipment, instruments, control systems, energy sources, energy storage systems, and information systems through platform interfaces. It provides a rich set of industrial communication interfaces for the collection of structured and unstructured data, such as OPC, Modus, RS232 / 485, IEC01 / 103 / 104, DLT645, MQTT, Restful API, and Web Service.

[0018] The data processing layer is used to aggregate, clean, convert, and encapsulate the underlying standardized data sources and then provide them to upper-layer applications.

[0019] The data storage layer includes real-time database systems, relational data, and NoSQL databases to form a data base storage platform, responsible for storing data in accordance with data standards and specifications to meet analysis and sharing needs;

[0020] The data analysis layer is used to provide data analysis tools with reporting, query, analysis, dashboard, APP, and 3D visualization functions to meet various data analysis application needs;

[0021] The data sharing layer is used to provide applications for core scenarios including resource directory generation, data service publishing, data service consumption, and data service monitoring.

[0022] The data visualization layer is used to provide various visualization methods such as mobile applications, self-service analysis, real-time analysis, large-screen display, smart dashboards, and statistical reports based on the understanding of data and indicators.

[0023] Furthermore, the specific process of obtaining the power supply demand table is as follows:

[0024] S101. Obtain a planar distribution map of each area of ​​the park and the enterprise site, divide each area of ​​the park and the enterprise site into unit areas according to a preset unit area, and obtain a number of power supply demand areas;

[0025] S102. Acquire, through the intelligent sensing unit, several power demand points within the power demand area, and obtain historical power consumption characteristic information, multi-energy data information, power consumption levels, and power consumption time periods corresponding to the power demand points, and standardize and classify the data for storage.

[0026] S103: Establish a time axis based on a preset power consumption cycle, set the vertical axis using the standardized data average value, calibrate the power consumption time period on the time axis, and then mark the historical power consumption characteristic information, multi-energy data information, and power consumption level one by one in the corresponding power consumption time period. Connect each data node with a smooth curve to obtain a power supply demand table;

[0027] S104: Obtain carbon emission data corresponding to the electricity consumption time period, and also mark the carbon emission data in the power supply demand table to obtain an optimized power supply demand table.

[0028] Furthermore, the specific process of obtaining the total power supply forecast data in the power demand area is as follows:

[0029] S201, obtaining power supply demand tables corresponding to several power supply demand points, integrating the power supply demand tables as training samples, and splitting the training samples into a training set, a test set, and a validation set in a ratio of 6:2:2;

[0030] S202: Build a power prediction model based on deep learning, download the weight file and load it onto the corresponding network to initialize the migration network parameters;

[0031] S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the node power supply prediction data, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters, retrain the entire network, and obtain the power prediction model;

[0032] S204, during the training process, randomly and non-repeatedly extract a small batch of power demand tables from the training set for training. The extraction of all power demand tables constitutes one training cycle. The training is completed after a certain number of iterations to obtain a power prediction model;

[0033] S205. Obtain the real-time power consumption characteristic information, multi-energy data information and power consumption level of the power supply demand point and import them into the power forecasting model to perform power demand forecasting, obtain the node power supply forecast data of the power supply demand point, and add the node power supply forecast data of several power supply demand points in the power supply demand area to obtain the total power supply forecast data.

[0034] Furthermore, the specific process of calculating the energy consumption coefficient is as follows:

[0035] S301. The energy supply data includes the equipment operating time t and equipment power P of various energy control systems within the park and the enterprise;

[0036] S302: Collect energy medium production Q1, consumption Q2, and cost data m, and calculate the energy consumption coefficient Lk according to the following formula: , where n is the energy output rate, and the energy consumption coefficient is used to reflect the energy utilization rate at the power supply demand point. The larger the energy consumption coefficient, the lower the energy utilization rate at the power supply demand point. Conversely, the larger the energy consumption coefficient, the higher the energy utilization rate at the power supply demand point.

[0037] Furthermore, the specific process of obtaining the energy plan scheduling scheme is as follows:

[0038] S401. Obtain input, output, loss, and calorific value data of the energy source system, and establish an energy analysis model based on the production plan and unit energy consumption of the product;

[0039] S402: Importing power demand characteristics, energy supply data, and energy consumption coefficients into an energy analysis model to perform energy complementarity optimization calculations to obtain an energy scheduling plan, wherein the energy scheduling plan includes energy supply time periods, energy supply sources and energy supply costs, and predicted carbon emissions;

[0040] S403: After normalizing and de-dimensionalizing the energy supply cost D and the predicted carbon emissions Qc corresponding to the energy supply time period, calculate the energy availability coefficient Hk according to the following formula: , where e1 and e2 are preset proportional coefficients, and the energy availability coefficient is used to evaluate the feasibility of the energy planning and scheduling scheme;

[0041] S404: Obtain a preset energy availability judgment threshold. If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy scheduling plan is qualified and sent to the intelligent control unit.

[0042] If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy plan scheduling scheme is unqualified and a new energy plan scheduling scheme is regenerated.

[0043] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0044] This energy management system based on multi-energy complementary smart energy of source, grid, load and storage obtains energy supply data of energy source system, park and various energy control systems within the enterprise through intelligent sensing units, and collects energy medium production, consumption and cost data in real time to obtain energy planning and scheduling solutions, which are used to realize the digitization, informatization, visualization, predictive and intelligentization of the entire process of enterprise energy supply, production, transportation, conversion and consumption. At the same time, it realizes full coverage of energy data collection in the whole plant and energy management of the main energy consumption systems of the enterprise that integrates analysis, optimization and decision-making assistance, improves the energy utilization efficiency of the enterprise, maximizes the energy efficiency of the enterprise, optimizes energy efficiency online, and pushes important information online, so as to achieve the goals of energy conservation and emission reduction, cost reduction and efficiency improvement, and improve the level of management and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Shows a schematic diagram of the overall structure of the present invention;

[0046] Figure 2 A schematic structural diagram of the data management unit of the present invention is shown. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0048] Example:

[0049] like Figure 1-2 As shown, the energy management system based on multi-energy complementary smart energy of source, grid, load and storage includes an intelligent sensing unit, an intelligent control unit and a data management unit;

[0050] The intelligent sensing unit uses the Industrial Internet of Things to build energy consumption systems, production systems, energy storage systems, detection systems, and monitoring systems in various areas of the park and enterprise site, achieving comprehensive perception of the factory's energy flow data;

[0051] Based on data from the intelligent perception layer, the intelligent control unit constructs a centralized energy control system for each production and auxiliary area of ​​the park or enterprise, centrally and uniformly controls, dispatches, and directs energy flows, creating an integrated centralized control center. Specifically, it includes a lighting energy dispatching module, a power energy dispatching module, a heating energy dispatching module, and other energy dispatching modules.

[0052] The intelligent perception layer includes lighting equipment, heating equipment, power equipment, energy-consuming equipment, substations, heat exchange stations, photovoltaic and air energy production capacity, as well as detection and monitoring systems;

[0053] The data management unit includes an energy consumption control module and a production capacity and energy storage control module;

[0054] The energy management and control module is used to divide the various areas of the park and enterprise site into modules to obtain several power supply demand areas, obtain power supply data within each power supply demand area, draw a power supply demand table based on the power consumption cycle, and obtain corresponding carbon emission data and electricity cost data to improve the power supply demand table. Based on the convolutional neural network, a power consumption prediction model is constructed to obtain the total power supply forecast data within the power supply demand area;

[0055] The specific process of obtaining the power supply demand table is as follows:

[0056] S101. Obtain a planar distribution map of each area of ​​the park and the enterprise site, divide each area of ​​the park and the enterprise site into unit areas according to a preset unit area, and obtain a number of power supply demand areas;

[0057] S102. Acquire, through the intelligent sensing unit, several power demand points within the power demand area, and obtain historical power consumption characteristic information, multi-energy data information, power consumption levels, and power consumption time periods corresponding to the power demand points, and standardize and classify the data for storage.

[0058] S103: Establish a time axis based on a preset power consumption cycle, set the vertical axis using the standardized data average value, calibrate the power consumption time period on the time axis, and then mark the historical power consumption characteristic information, multi-energy data information, and power consumption level one by one in the corresponding power consumption time period. Connect each data node with a smooth curve to obtain a power supply demand table;

[0059] S104: Obtain carbon emission data corresponding to the electricity consumption time period, and also mark the carbon emission data in the power supply demand table to obtain an optimized power supply demand table.

[0060] The specific process of obtaining the total power supply forecast data in the power demand area is as follows:

[0061] S201, obtaining power supply demand tables corresponding to several power supply demand points, integrating the power supply demand tables as training samples, and splitting the training samples into a training set, a test set, and a validation set in a ratio of 6:2:2;

[0062] S202: Build a power prediction model based on deep learning, download the weight file and load it onto the corresponding network to initialize the migration network parameters;

[0063] S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the node power supply prediction data, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters, retrain the entire network, and obtain the power prediction model;

[0064] S204, during the training process, randomly and non-repeatedly extract a small batch of power demand tables from the training set for training. The extraction of all power demand tables constitutes one training cycle. The training is completed after a certain number of iterations to obtain a power prediction model;

[0065] S205. Obtain the real-time power consumption characteristic information, multi-energy data information and power consumption level of the power supply demand point and import them into the power forecasting model to perform power demand forecasting, obtain the node power supply forecast data of the power supply demand point, and add the node power supply forecast data of several power supply demand points in the power supply demand area to obtain the total power supply forecast data.

[0066] The capacity and energy storage management and control module uses intelligent sensing units to obtain energy supply data from energy source systems (substations, heat exchange stations, photovoltaic capacity, and air energy capacity) and various energy control systems within the park and the enterprise (lighting equipment, heating equipment, power equipment, energy-consuming equipment, detection systems, and monitoring systems). It also collects real-time energy medium production, consumption, and cost data, calculates the energy consumption coefficient, and imports power supply demand characteristics, energy supply data, and energy consumption coefficient into the energy analysis model to perform energy complementarity optimization calculations, obtain an energy plan and scheduling plan, and send it to the intelligent control unit.

[0067] The specific process of calculating the energy consumption coefficient is as follows:

[0068] S301. Energy supply data includes the equipment operating time t and equipment power P of various energy control systems within the park and the enterprise;

[0069] S302: Collect energy medium production Q1, consumption Q2, and cost data m, and calculate the energy consumption coefficient Lk according to the following formula: , where n is the energy output rate, and the energy consumption coefficient is used to reflect the energy utilization rate at the power supply demand point. The larger the energy consumption coefficient, the lower the energy utilization rate at the power supply demand point. Conversely, the larger the energy consumption coefficient, the higher the energy utilization rate at the power supply demand point.

[0070] The specific process of obtaining the energy plan scheduling scheme is as follows:

[0071] S401. Obtain input, output, loss, and calorific value data of the energy source system, and establish an energy analysis model based on the production plan and unit energy consumption of the product;

[0072] S402: Importing power demand characteristics, energy supply data, and energy consumption coefficients into an energy analysis model to perform energy complementarity optimization calculations to obtain an energy scheduling plan. The energy scheduling plan includes energy supply time periods, energy supply sources, energy supply costs, and predicted carbon emissions.

[0073] S403: After normalizing and de-dimensionalizing the energy supply cost D and the predicted carbon emissions Qc corresponding to the energy supply time period, calculate the energy availability coefficient Hk according to the following formula: , where e1 and e2 are preset proportional coefficients, and the energy availability coefficient is used to evaluate the feasibility of the energy planning and scheduling scheme;

[0074] S404: Obtain a preset energy availability judgment threshold. If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy scheduling plan is qualified and sent to the intelligent control unit.

[0075] If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy plan scheduling scheme is unqualified and a new energy plan scheduling scheme is regenerated.

[0076] The data management unit also includes the IaaS layer, PaaS layer, and energy data platform, including:

[0077] The IaaS layer is used to achieve hardware resource integration and virtualization, forming an elastic resource pool, performing resource segmentation, resource allocation, and resource integration. It provides servers, storage, backup, network, and other resources for the energy management system. Based on actual business needs, the energy management system can flexibly dispatch pooled resources.

[0078] The PaaS layer serves as a unified support and integration platform for applications. It integrates the resources provided by the IaaS layer downward and supports application development and operation upward. Based on a containerized structure, it provides automated tools for development, testing, deployment, and operation and maintenance for top-level API and app development. It also provides highly available, clustered databases and various middleware resources, empowering project management platforms with application deployment and development as well as middleware management capabilities.

[0079] The energy data platform is used to collect and store real-time data scattered across various energy application systems and real-time database systems, and to perform standardized management of real-time data. By extracting, cleaning, converting, and storing it, it forms a data warehouse based on master data. Relying on the distributed computing and big data analysis capabilities of the industrial data platform, it provides external data analysis services to support real-time computing and scientific decision-making in smart factories.

[0080] The energy data platform collects energy data from energy flow-related PLCs, DCSs, intelligent equipment, sensors, and other underlying devices into a real-time database through OPC, MODBUS, IEC101 / 104 protocols, RS232 / 485 interface standards, and proprietary protocol customized parsing methods. The energy data is then integrated and processed within the real-time database and transmitted to a relational database for logical calculation, matching, association, and business integration. The energy data is then provided to the factory's production capacity and various energy-consuming applications. Specifically, the platform includes a data acquisition layer, a data processing layer, a data storage layer, a data analysis layer, a data sharing layer, and a data visualization layer, including:

[0081] The data collection layer is used to comprehensively collect data from park or production site equipment, instruments, control systems, energy sources, energy storage systems, and information systems through platform interfaces. It provides a rich set of industrial communication interfaces for the collection of structured and unstructured data, such as OPC, Modus, RS232 / 485, IEC01 / 103 / 104, DLT645, MQTT, Restful API, and Web Service.

[0082] The data processing layer is used to aggregate, clean, convert, and encapsulate the underlying standardized data sources and then provide them to upper-layer applications.

[0083] The data storage layer includes real-time database systems, relational data, and NoSQL databases to form a data base storage platform, responsible for storing data in accordance with data standards and specifications to meet analysis and sharing needs;

[0084] The data analysis layer is used to provide data analysis tools with reporting, query, analysis, dashboard, APP, and 3D visualization functions to meet various data analysis application needs;

[0085] The data sharing layer is used to provide applications for core scenarios including resource directory generation, data service publishing, data service consumption, and data service monitoring.

[0086] The data visualization layer is used to provide various visualization methods such as mobile applications, self-service analysis, real-time analysis, large-screen display, smart dashboards, and statistical reports based on the understanding of data and indicators.

[0087] The present invention obtains energy supply data of energy source systems, parks and various energy control systems within enterprises through intelligent sensing units, and collects energy medium production, consumption and cost data in real time to obtain energy planning and scheduling solutions, which are used to realize the digitization, informatization, visualization, prediction and intelligence of the entire process of enterprise energy supply, production, transportation, conversion and consumption. At the same time, it realizes full coverage of energy data collection in the whole plant and energy management of the main energy consumption systems of the enterprise that integrates analysis, optimization and decision-making assistance, improves the energy utilization efficiency of the enterprise, maximizes the energy efficiency of the enterprise, optimizes energy efficiency online, and pushes important information online, so as to achieve the goals of energy conservation and emission reduction, cost reduction and efficiency improvement, and improvement of management and control level.

[0088] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0089] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by those skilled in the art according to actual conditions.

[0090] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An energy management system based on multi-energy complementary smart energy of source, grid, load and storage, characterized by: Including intelligent sensing unit, intelligent control unit and data management unit; The intelligent sensing unit uses the Industrial Internet of Things to build energy consumption systems, production systems, energy storage systems, detection systems, and monitoring systems in various areas of the park and enterprise site, achieving comprehensive perception of the factory's energy flow data; Based on data from the intelligent perception layer, the intelligent control unit constructs a centralized energy control system for each production and auxiliary area of ​​the park or enterprise, centrally and uniformly controls, dispatches, and directs energy flows, creating an integrated centralized control center. Specifically, it includes a lighting energy dispatching module, a power energy dispatching module, a heating energy dispatching module, and other energy dispatching modules. The data management unit includes an energy consumption control module and a production capacity and energy storage control module; The energy consumption control module is used to divide the various areas of the park and enterprise site into modules to obtain several power supply demand areas, obtain power supply data within each power supply demand area, draw a power supply demand table based on the power consumption cycle, and obtain corresponding carbon emission data and electricity cost data to improve the power supply demand table. Based on the convolutional neural network, a power consumption prediction model is constructed to obtain total power supply forecast data within the power supply demand area; The production capacity and energy storage management and control module obtains energy supply data from the energy source system, the park, and various energy control systems within the enterprise through the intelligent sensing unit, and collects energy medium production, consumption, and cost data in real time, calculates the energy consumption coefficient, and imports the power supply demand characteristics, energy supply data, and energy consumption coefficient into the energy analysis model to perform energy complementary optimization calculations, obtain an energy planning and scheduling plan, and send it to the intelligent control unit.

2. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 1 is characterized in that: The data management unit also includes the IaaS layer, PaaS layer, and energy data platform, including: The IaaS layer is used to realize the integration and virtualization of hardware resources, form an elastic resource pool, perform resource segmentation, resource allocation and resource integration, and provide servers, storage, backup, network and other resources for the energy management system; The PaaS layer serves as a unified support and integration platform for applications. It integrates the resources provided by the IaaS layer downward and supports application development and operation upward. Based on a containerized structure, it provides automated tools for development, testing, deployment, and operation and maintenance for top-level API and app development. It also provides highly available, clustered databases and various middleware resources. The energy data platform is used to collect and store real-time data scattered across various energy application systems and real-time database systems, perform standardized management of real-time data, and form a data warehouse based on master data through extraction, cleaning, conversion, and storage. Relying on the distributed computing and big data analysis capabilities of the industrial data platform, it provides external data analysis services.

3. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 2 is characterized in that: The energy data platform collects energy data from energy flow-related PLCs, DCSs, intelligent equipment, sensors and other underlying devices into a real-time database through OPC, MODBUS, IEC101 / 104 protocols, RS232 / 485 interface standards and proprietary protocol customized analysis methods, integrates and processes the energy data in the real-time database, and transmits the energy data to a relational database for logical calculation, matching association, and business integration, and provides it to the factory production capacity end and various energy-consuming application ends. It specifically includes a data acquisition layer, a data processing layer, a data storage layer, a data analysis layer, a data sharing layer and a data visualization layer.

4. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 1 is characterized in that: The specific process of obtaining the power supply demand table is as follows: S101. Obtain a planar distribution map of each area of ​​the park and the enterprise site, divide each area of ​​the park and the enterprise site into unit areas according to a preset unit area, and obtain a number of power supply demand areas; S102. Acquire, through the intelligent sensing unit, several power demand points within the power demand area, and obtain historical power consumption characteristic information, multi-energy data information, power consumption levels, and power consumption time periods corresponding to the power demand points, and standardize and classify the data for storage. S103: Establish a time axis based on a preset power consumption cycle, set the vertical axis using the standardized data average value, calibrate the power consumption time period on the time axis, and then mark the historical power consumption characteristic information, multi-energy data information, and power consumption level one by one in the corresponding power consumption time period. Connect each data node with a smooth curve to obtain a power supply demand table; S104: Obtain carbon emission data corresponding to the electricity consumption time period, and also mark the carbon emission data in the power supply demand table to obtain an optimized power supply demand table.

5. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 1 is characterized in that: The specific process of obtaining the total power supply forecast data in the power demand area is as follows: S201, obtaining power supply demand tables corresponding to several power supply demand points, integrating the power supply demand tables as training samples, and splitting the training samples into a training set, a test set, and a validation set in a ratio of 6:2:2; S202: Build a power prediction model based on deep learning, download the weight file and load it onto the corresponding network to initialize the migration network parameters; S203. Modify the last fully connected layer of the network, keep the input unchanged, set the output to the node power supply prediction data, initialize the weights of the last layer, use the gradient descent algorithm for learning, and use fixed step size decay to optimize the training parameters, retrain the entire network, and obtain the power prediction model; S204, during the training process, randomly and non-repeatedly extract a small batch of power demand tables from the training set for training. The extraction of all power demand tables constitutes one training cycle. The training is completed after a certain number of iterations to obtain a power prediction model; S205. Obtain the real-time power consumption characteristic information, multi-energy data information and power consumption level of the power supply demand point and import them into the power forecasting model to perform power demand forecasting, obtain the node power supply forecast data of the power supply demand point, and add the node power supply forecast data of several power supply demand points in the power supply demand area to obtain the total power supply forecast data.

6. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 1 is characterized in that: The specific process of calculating the energy consumption coefficient is as follows: S301. The energy supply data includes the equipment operating time t and equipment power P of various energy control systems within the park and the enterprise; S302: Collect energy medium production Q1, consumption Q2, and cost data m, and calculate the energy consumption coefficient Lk according to the following formula: , where n is the energy output rate, and the energy consumption coefficient is used to reflect the energy utilization rate at the power supply demand point. The larger the energy consumption coefficient, the lower the energy utilization rate at the power supply demand point. Conversely, the larger the energy consumption coefficient, the higher the energy utilization rate at the power supply demand point.

7. The energy management system based on multi-energy complementary smart energy of source, grid, load and storage according to claim 1 is characterized in that: The specific process of obtaining the energy plan scheduling scheme is as follows: S401. Obtain input, output, loss, and calorific value data of the energy source system, and establish an energy analysis model based on the production plan and unit energy consumption of the product; S402: Importing power demand characteristics, energy supply data, and energy consumption coefficients into an energy analysis model to perform energy complementarity optimization calculations to obtain an energy scheduling plan, wherein the energy scheduling plan includes energy supply time periods, energy supply sources and energy supply costs, and predicted carbon emissions; S403: After normalizing and de-dimensionalizing the energy supply cost D and the predicted carbon emissions Qc corresponding to the energy supply time period, calculate the energy availability coefficient Hk according to the following formula: , where e1 and e2 are preset proportional coefficients, and the energy availability coefficient is used to evaluate the feasibility of the energy planning and scheduling scheme; S404: Obtain a preset energy availability judgment threshold. If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy scheduling plan is qualified and sent to the intelligent control unit. If the energy availability coefficient is greater than or equal to the energy availability judgment threshold, the energy plan scheduling scheme is unqualified and a new energy plan scheduling scheme is regenerated.

Images