A collaborative optimization method for integrated energy monitoring operation based on edge autonomy

Through the integrated energy monitoring and operation collaborative optimization method of edge autonomy, combined with cloud prediction and edge autonomy optimization, the problem of cloud systems being susceptible to interruptions and insufficient computing power in local systems is solved, real-time energy management and efficient utilization are achieved, and costs are reduced and improved Reliability.

CN117408470BActive Publication Date: 2024-11-22STATE GRID JIANGSU INTEGRATED ENERGY SERVICE CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202311407041.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-27
Publication Date
2024-11-22
Estimated Expiration
2043-10-27

AI Technical Summary

Technical Problem

The existing comprehensive energy control system is prone to network interruption when implemented in the cloud, making it difficult to operate normally; due to limited computing power, locally deployed systems are difficult to improve energy optimization and operation and maintenance efficiency, and lack the optimization method of combining cloud and edge.

Method used

Adopt the integrated energy monitoring operation collaborative optimization method based on edge autonomy, collect equipment data through the edge monitoring system, build optimization models and perform strategy calculations, load prediction and calibration adjustments are performed in the cloud, and edge systems are optimized and calibration and corrections are established based on The monitoring network of an ad-organized network is used to realize real-time data collection and processing.

Benefits of technology

Real-time monitoring and autonomous optimization of integrated energy equipment have been achieved, energy management efficiency and system reliability have been improved, operational costs have been reduced, and energy utilization efficiency and sustainability have been enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117408470B_ABST
    Figure CN117408470B_ABST
Patent Text Reader

Abstract

The present invention discloses a collaborative optimization method for integrated energy monitoring operation based on edge autonomy, which relates to the field of integrated energy management and control technology. The method includes the following steps: S1, establishing a monitoring network of integrated energy equipment-gateway equipment-edge monitoring system, and collecting the operation data of integrated energy equipment through the edge monitoring system; S2, constructing an optimization model for integrated energy equipment, and inputting the real-time collected operation data into the optimization model for optimization strategy calculation; S3, obtaining the historical monitoring data of each edge monitoring system in the cloud, and constructing an equipment load prediction model; S4, the edge monitoring system uploads the optimization strategy results to the cloud; S5, the cloud feeds back the verification and strategy adjustment results to the edge monitoring system. The present invention collects the operation data of the enterprise's integrated energy equipment through the edge monitoring system, and realizes the real-time monitoring, autonomous optimization and verification and correction of the integrated energy equipment by the edge system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of integrated energy management and control technology, and in particular to a collaborative optimization method for integrated energy monitoring operation based on edge autonomy. Background Art

[0002] At present, the development of comprehensive energy is in full swing. Combining new energy with energy-saving equipment transformation can reduce carbon emissions while reducing corporate energy costs. The development and use of renewable energy is becoming more and more extensive, and distributed photovoltaic power generation is its most important use method, which has the characteristics of convenient construction, low price, and fast commissioning. Distributed photovoltaic power generation is an energy production method based on solar photovoltaic technology. It converts solar radiation into electrical energy and is usually distributed in multiple locations, such as building roofs, solar cell fields, and distributed energy systems. This energy production method is different from traditional centralized power generation. Distributed photovoltaic power generation is usually distributed in multiple locations, reducing losses in the process of power transmission and distribution, and improving the stability and reliability of the power grid.

[0003] While developing energy construction, it is necessary to introduce a more intelligent and comprehensive management and control system. Integrated energy management and control is a comprehensive energy management method that aims to optimize and coordinate different types of energy resources to achieve efficient use of energy, cost reduction and environmental sustainability. Integrated energy management and control systems usually combine multiple technologies and strategies to meet energy needs and achieve various energy management goals.

[0004] However, if integrated energy management and control is implemented based on the cloud, there is an obvious problem that it will be difficult to operate normally in the event of a network interruption; if only a locally deployed integrated energy monitoring system is used, it will be difficult to improve energy optimization, energy operation and maintenance, etc. due to limited local computing power. In other words, the optimization of the integrated energy operation control strategy requires local real-time operation status data, a large amount of historical operation data and model references of other similar operating environments. Otherwise, relying solely on partial equipment data is difficult to meet the needs of actual production operation strategy formulation.

[0005] At present, many scholars have studied the comprehensive energy management and optimization methods. Some are based on local system calculations, and some are optimized and analyzed through the cloud, but there is a lack of optimization methods that combine cloud and edge computing. For example, Chinese patent CN202310519078.3 discloses a comprehensive energy management and control method and system based on a smart cloud gateway. The system has multiple control modules. The gateway collects the energy consumption data of the client, and then implements the central real-time control and dispatching method by providing real-time scheduling strategies and frequency modulation scheduling according to the energy demand time and the power supply demand. The system still uses distributed monitoring and collection, and uses central processing, analysis and scheduling to solve energy management problems. There are still technical defects in cloud management.

[0006] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0007] In response to the problems in the related technology, the present invention proposes a collaborative optimization method for integrated energy monitoring operation based on edge autonomy to overcome the above-mentioned technical problems existing in the existing related technology.

[0008] To this end, the specific technical solution adopted by the present invention is as follows:

[0009] A collaborative optimization method for integrated energy monitoring operation based on edge autonomy, the method comprising the following steps:

[0010] S1. Establish a monitoring network of integrated energy equipment, gateway equipment and edge monitoring system, collect the operation data of integrated energy equipment through the edge monitoring system, and then store it in the database after pre-processing;

[0011] S2. Build an optimization model for integrated energy equipment, input the real-time collected operation data into the optimization model to calculate the optimization strategy, and output the optimization strategy results;

[0012] S3. The cloud obtains historical monitoring data of each edge monitoring system, builds an equipment load prediction model, and uses the equipment load prediction model to output load prediction results;

[0013] S4. The edge monitoring system uploads the optimization strategy results to the cloud, and the cloud uses the load forecast results to verify the optimization strategy results and adjust the strategy;

[0014] S5. The cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system performs autonomous optimization and verification correction on the integrated energy equipment.

[0015] Furthermore, a monitoring network of integrated energy equipment, gateway equipment and edge monitoring system is established, and the operation data of the integrated energy equipment is collected through the edge monitoring system and then stored in the database after preprocessing, including the following steps:

[0016] S11. Configure a gateway device between the integrated energy device and the edge monitoring system to establish a monitoring network based on a self-organizing network that coordinates the integrated energy device, the gateway device, and the edge monitoring system;

[0017] S12, the gateway device plays the role of a communication bridge in the monitoring network, receives the operation data of the integrated energy equipment in real time, and transmits it to the edge monitoring system;

[0018] S13, the edge monitoring system performs data analysis, information extraction and format conversion on the received original operation data, and converts it into operation data that meets the data processing format;

[0019] S14. The edge monitoring system performs rationality verification and outlier identification on the converted operating data, marks the outliers as abnormal quality codes, and stores the data in the database after processing.

[0020] Furthermore, configuring a gateway device between the integrated energy device and the edge monitoring system to establish a monitoring network based on a self-organizing network of integrated energy devices, gateway devices, and edge monitoring systems includes the following steps:

[0021] S111, all integrated energy devices connected in the local area network where the gateway device is located are regarded as a cluster, and the gateway device is used as the cluster head of the cluster, and the cluster head generates a public key pair and a unique identifier of the cluster, which are then sent to the edge monitoring system through the gateway device;

[0022] S112, select m integrated energy devices in the cluster as leaf nodes, and the cluster head starts to assign serial numbers to the leaf nodes according to the order of the Hamiltonian cycle. The leaf nodes then use a pseudo-random generator to generate a pseudo-random sequence, and use a hash calculation method to obtain m leaf keys;

[0023] S113, the edge monitoring system randomly selects one from all leaf nodes as an authentication node, the authentication node negotiates with the cluster head to generate a shared key, the cluster head then uses the shared key to encrypt a leaf key and sends it to the authentication node, and the cluster head sends a request packet to the edge monitoring system, the edge monitoring system uses the public key pair to decrypt and verify, after the authentication node and the edge monitoring system complete the decryption and verification, the cluster head generates a session key and distributes it to all integrated energy devices to establish a trust relationship between the devices;

[0024] S114. A monitoring network of an integrated energy device-gateway device-edge monitoring system is formed by all integrated energy devices, gateway devices and edge monitoring systems that have established a trust relationship.

[0025] Furthermore, an optimization model of integrated energy equipment is constructed, and the real-time collected operation data is input into the optimization model for optimization strategy calculation. Outputting the optimization strategy results includes the following steps:

[0026] S21. Clean the operation data and align the data according to time synchronization;

[0027] S22. According to the electricity consumption and power generation demand and energy supply mode, the comprehensive energy equipment is divided into three types of equipment: electricity consumption equipment, power generation equipment and equipment that both consumes and generates electricity;

[0028] S23, construct physical models according to the working principles of different types of integrated energy equipment;

[0029] S24, constructing empirical formulas according to the data relationships of different types of comprehensive energy equipment;

[0030] S25. Obtain historical operation data of a large amount of comprehensive energy equipment, solve the empirical parameters in the empirical equation by fitting calculation, and construct optimization models for different types of equipment;

[0031] S26. The edge monitoring system inputs the equipment type information and real-time operation data of the integrated energy equipment into the corresponding optimization model to calculate the operating power of the integrated energy equipment;

[0032] S27. Optimize the operation strategy of the comprehensive energy equipment based on the operation power calculation result. If the operation power calculation result is higher than the current power, the equipment power will be increased. If the operation power calculation result is lower than the current power, the equipment power will be decreased. The adjusted operation strategy will be used as the optimization strategy result.

[0033] Furthermore, constructing physical models according to the working principles of different types of integrated energy equipment includes the following steps:

[0034] S231. Based on the circuit theory and electrical characteristics of the electrical equipment consuming electric energy, a physical model of the electrical equipment is established. The expression of the physical model of the electrical equipment is:

[0035] P Y =V×I×cos(φ);

[0036] Where P Y represents power consumption; V represents voltage; I represents current; cos(φ) represents power factor;

[0037] S232. According to the electric energy conversion characteristics of the power generation equipment, a physical model of the power generation equipment is established. The expression of the physical model of the power generation equipment is:

[0038]

[0039] Where P F represents the power generation; η represents the conversion efficiency;

[0040] S233. Based on the power consumption and generation characteristics of the power generation equipment, a physical model of the power generation equipment is established. The expression of the physical model of the power generation equipment is:

[0041]

[0042] Where P J Represents net energy consumption.

[0043] Furthermore, constructing empirical formulas according to the data relationships of different types of integrated energy equipment includes the following steps:

[0044] S241. Based on the relationship between the total voltage and total current of the electrical equipment, an empirical formula for the electrical equipment is established. The expression of the empirical formula for the electrical equipment is:

[0045] P nF =a×V total +b×I total ;

[0046] Where P nF Indicates the power consumption of electrical equipment; a and b are both empirical parameters; V total Indicates the total voltage of all electrical equipment; I total Indicates the total current of all electrical equipment;

[0047] S242. Establish an empirical formula for power generation equipment based on the total system power of the power generation equipment. The empirical formula for power generation equipment is expressed as:

[0048] P nY =a×P total +b;

[0049] Where P nY Indicates the power generation of the power generation equipment; P total Indicates the total system power;

[0050] S243. Based on the relationship between the system power and the remaining power in the device that both uses and generates electricity, an empirical formula for the device that both uses and generates electricity is established. The empirical formula for the device that both uses and generates electricity is expressed as follows:

[0051] P nJ =a×P total +b×Q n ;

[0052] Where P nJ Indicates the power consumption or power generation of the equipment that both uses and generates electricity; Q n Indicates the remaining power of the device that is both in use and in power.

[0053] Furthermore, the cloud obtains historical monitoring data of each edge monitoring system, builds a device load prediction model, and uses the device load prediction model to output load prediction results, including the following steps:

[0054] S31. Obtain the entity information of the monitoring network where all edge monitoring systems are located, and use digital twin technology to build a virtual network and a visual virtual management platform in the cloud;

[0055] S32, the cloud obtains historical monitoring data of the edge monitoring system and maps it to the virtual network, and extracts time characteristics, weather data and historical load data from the historical monitoring data;

[0056] S33, cleaning and standardizing the extracted historical monitoring data;

[0057] S34, establish a distributed gradient enhancement library model, and use the optimization target as the minimization loss function, iteratively use historical monitoring data to train the distributed gradient enhancement library model, and after the training is completed, the final equipment load prediction model is obtained. The expression of the minimization loss function is:

[0058]

[0059] In the formula, y i represents the target variable; x i represents each prediction training vector; F represents the prediction result vector set; N represents the number of decision trees; F(x i ) represents the prediction result vector of the i-th decision tree; f represents the prediction value of the decision tree;

[0060] S35. The equipment load prediction model performs prediction calculations, outputs load prediction results, and calculates the load curve for the next 24 hours, which is visualized on the virtual management platform.

[0061] Furthermore, obtaining the entity information of the monitoring network where all edge monitoring systems are located, using digital twin technology, building a virtual network and a visualized virtual management platform in the cloud includes the following steps:

[0062] S311, obtaining topological information of integrated energy equipment, gateway equipment and edge monitoring systems constituting the monitoring network, and sending it to the cloud as entity information;

[0063] S312, according to the number and model of the devices in the physical information, establish the corresponding number and model of virtual devices, send a query instruction, receive the online information and task information of the virtual devices for verification, and after the verification is passed, perform parameter status mapping between the physical device and the virtual device;

[0064] S313, converting the entity information into a virtual topology structure, sending it to the corresponding virtual device, establishing a virtual network of the monitoring network digital twin, and synchronously updating the energy monitoring data;

[0065] S314. Build a visual virtual management platform to display the monitoring data of the monitoring network and the virtual network in real time, as well as the data information to be executed and predicted.

[0066] Furthermore, the edge monitoring system uploads the optimization strategy results to the cloud, and the cloud uses the load forecast results to verify the optimization strategy results and adjust the strategy, including the following steps:

[0067] S41, the edge monitoring system uploads the optimization strategy results to the cloud;

[0068] S42, the cloud determines whether there is a new combination in the optimization strategy result. If there is a new combination, the new combination is recorded, and then step S43 is executed to perform verification and adjustment, and the verification result is returned to the edge monitoring system. If there is no new combination, step S44 is executed;

[0069] S43, extract the power information in the optimization strategy result uploaded by the edge monitoring system, and substitute the load prediction result calculated by the equipment load prediction model into the energy conservation equation to check whether the equation is satisfied. If it is satisfied, the optimization strategy result is not adjusted. If it is not satisfied, the optimization strategy result is adjusted. The expression of the energy conservation equation is:

[0070] Total input power + total power generation = total power consumption;

[0071] S44. Determine whether there is new historical operation data of the equipment in the optimization strategy result. If there is new historical operation data of the equipment, execute step S43 to perform verification and adjustment, and return the verification result to the edge monitoring system. If there is no new historical operation data of the equipment, return the optimization strategy result uploaded by the edge monitoring system.

[0072] Furthermore, the cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system autonomously optimizes and verifies the integrated energy equipment, including the following steps:

[0073] S51, the edge monitoring system receives the verification and strategy adjustment results fed back from the cloud, and the edge monitoring system decomposes the target data that needs to be adjusted for the integrated energy equipment into multiple sub-targets, and each sub-target is adjusted according to a specific parameter of the rated capacity;

[0074] S52. After completing the adjustment of all sub-goals, the operation data of the current integrated energy equipment is input into the optimization model, the operation power is recalculated, and all the data are substituted into the energy conservation equation. Combined with the pre-configured parameters for preventing transformer overload and preventing reverse power transmission to the power grid, the actual operation power of the integrated energy equipment is adjusted. If there is an overload risk, the power consumption is reduced; if there is a reverse power transmission risk, the power generation is reduced;

[0075] S53. Continuously monitor the actual operating status of integrated energy equipment and implement real-time dynamic adjustments.

[0076] The beneficial effects of the present invention are:

[0077] 1. The operating data of the enterprise's integrated energy equipment is collected through the edge monitoring system, and the data is processed and stored; then the edge optimization model is built using historical data, and the real-time operating data is input into the optimization model to calculate the optimization strategy; the cloud uses the XGBoost algorithm to build a prediction model based on the historical data of each system, and calculates the prediction data; the edge system sends the calculation results to the cloud for verification. If there are new combination changes or equipment historical data updates in the operating mode, the cloud will calculate and provide verification and correction results, and send them to the edge monitoring system; the edge system makes corrections based on the verification results to achieve real-time monitoring, autonomous optimization and verification and correction of integrated energy equipment by the edge system.

[0078] 2. By establishing a monitoring network of comprehensive energy equipment-gateway equipment-edge monitoring system based on self-organizing network, it can realize real-time data collection, preprocessing and storage, improve the efficiency of energy equipment, monitor its performance, accurately identify abnormal situations, and provide a reliable data basis for decision-making, thereby reducing operating costs, increasing the reliability of energy systems, and improving the efficiency of energy management.

[0079] 3. By integrating historical monitoring data and weather information in the cloud and based on digital twin technology, efficient equipment load forecasting can be achieved, helping enterprises to better plan energy supply and demand management, improve energy utilization efficiency, reduce waste, and provide key decision support through visual display, thereby reducing energy costs, improving sustainability, and improving energy management. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0081] Figure 1 It is a flow chart of a collaborative optimization method for integrated energy monitoring operation based on edge autonomy according to an embodiment of the present invention. DETAILED DESCRIPTION

[0082] According to an embodiment of the present invention, a collaborative optimization method for integrated energy monitoring operation based on edge autonomy is provided.

[0083] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the collaborative optimization method of integrated energy monitoring operation based on edge autonomy according to an embodiment of the present invention, the method includes the following steps:

[0084] S1. Establish a monitoring network of integrated energy equipment, gateway equipment and edge monitoring system, collect the operation data of integrated energy equipment through the edge monitoring system, and then store it in the database after pre-processing.

[0085] Among them, integrated energy equipment is equipment used to generate, store, distribute and manage various forms of energy (such as electricity, heat, and cold energy) to meet the needs of integrated energy systems, including generators, transformers, switchgear, sensors and instruments, etc.

[0086] In the description of the present invention, the monitoring network of integrated energy equipment-gateway equipment-edge monitoring system is established, and the operation data of the integrated energy equipment is collected by the edge monitoring system and then stored in the database after preprocessing, which includes the following steps:

[0087] S11. Configure a gateway device between the integrated energy equipment and the edge monitoring system to establish a monitoring network based on a self-organizing network that coordinates the integrated energy equipment, gateway device, and edge monitoring system.

[0088] In the description of the present invention, a gateway device is configured between the integrated energy device and the edge monitoring system, and a monitoring network based on a self-organizing network of integrated energy device-gateway device-edge monitoring system is established, which includes the following steps:

[0089] S111. All integrated energy devices connected within the local area network where the gateway device is located are regarded as a cluster, and the gateway device is used as the cluster head of the cluster. The cluster head generates a public key pair and a unique identifier of the cluster, which are then sent to the edge monitoring system through the gateway device.

[0090] S112. Select m integrated energy devices in the cluster as leaf nodes. The cluster head starts to assign serial numbers to the leaf nodes according to the order of the Hamiltonian cycle. The leaf nodes then use a pseudo-random generator to generate a pseudo-random sequence and use a hash calculation method to obtain m leaf keys.

[0091] S113. The edge monitoring system randomly selects one from all leaf nodes as the authentication node. The authentication node negotiates with the cluster head to generate a shared key. The cluster head then uses the shared key to encrypt a leaf key and sends it to the authentication node. At the same time, the cluster head sends a request packet to the edge monitoring system. The edge monitoring system uses the public key pair to decrypt and verify. After the authentication node and the edge monitoring system complete the decryption and verification, the cluster head generates a session key and distributes it to all integrated energy devices to establish a trust relationship between the devices.

[0092] S114. A monitoring network of an integrated energy device-gateway device-edge monitoring system is formed by all integrated energy devices, gateway devices and edge monitoring systems that have established a trust relationship.

[0093] S12. The gateway device plays the role of a communication bridge in the monitoring network, receives the operating data of the integrated energy equipment in real time, and transmits it to the edge monitoring system.

[0094] The gateway device supports the IEC101 protocol. In addition, the integrated energy device needs to be configured to ensure that the device can establish communication with the gateway and support the IEC104 protocol. The gateway (IP address, port number, communication mode, etc.) is configured to ensure that it can establish a connection with the device and the edge monitoring system, and send and receive data in the format of the IEC104 protocol. Finally, the operating data of the integrated energy device is mapped to the data point of the IEC101 protocol and transmitted to the edge monitoring system by the gateway device.

[0095] S13. The edge monitoring system performs data analysis, information extraction, and format conversion on the received original operation data to convert it into operation data that meets the data processing format.

[0096] S14. The edge monitoring system performs rationality verification and outlier identification on the converted operating data, marks the outliers as abnormal quality codes, and stores the data in the database after processing.

[0097] S2. Build an optimization model for integrated energy equipment, input the real-time collected operation data into the optimization model to calculate the optimization strategy, and output the optimization strategy results.

[0098] In the description of the present invention, an optimization model of a comprehensive energy device is constructed, and the operation data collected in real time is input into the optimization model for optimization strategy calculation, and the optimization strategy result is output, which includes the following steps:

[0099] S21. Clean the operation data and align the data according to time synchronization.

[0100] S22. According to the electricity consumption and power generation demand and the energy supply mode, comprehensive energy equipment is divided into three types of equipment: electricity-consuming equipment (air conditioning, lighting), power generation equipment (photovoltaic, wind turbines) and equipment that both consumes and generates electricity (energy storage).

[0101] S23. Construct physical models according to the working principles of different types of integrated energy equipment.

[0102] In the description of the present invention, constructing a physical model according to the working principles of different types of integrated energy equipment includes the following steps:

[0103] S231. Based on the circuit theory and electrical characteristics of the electrical equipment consuming electric energy, a physical model of the electrical equipment is established. The expression of the physical model of the electrical equipment is:

[0104] P Y =V×I×cos(φ);

[0105] Where P Y represents power consumption, V represents voltage, I represents current, and cos(φ) represents power factor.

[0106] S232. According to the electric energy conversion characteristics of the power generation equipment, a physical model of the power generation equipment is established. The expression of the physical model of the power generation equipment is:

[0107]

[0108] Where P F represents the generated power, and η represents the conversion efficiency.

[0109] S233. Based on the power consumption and generation characteristics of the power generation equipment, a physical model of the power generation equipment is established. The expression of the physical model of the power generation equipment is:

[0110]

[0111] Where P J Represents net energy consumption.

[0112] S24. Construct empirical formulas according to the data relationships of different types of integrated energy equipment.

[0113] In the description of the present invention, constructing empirical formulas according to the data relationships of different types of integrated energy equipment includes the following steps:

[0114] S241. Based on the relationship between the total voltage and total current of the electrical equipment, an empirical formula for the electrical equipment is established. The expression of the empirical formula for the electrical equipment is:

[0115] P nF =a×V total +b×I total ;

[0116] Where P nF Indicates the power consumption of electrical equipment, a and b are both empirical parameters, V total Indicates the total voltage of all electrical equipment, I total Indicates the total current of all electrical devices.

[0117] S242. Establish an empirical formula for power generation equipment based on the total system power of the power generation equipment. The empirical formula for power generation equipment is expressed as:

[0118] P nY =a×P total +b;

[0119] Where P nY Indicates the power generation of the power generation equipment, P total Indicates the total system power.

[0120] S243. Based on the relationship between the system power and the remaining power in the device that both uses and generates electricity, an empirical formula for the device that both uses and generates electricity is established. The empirical formula for the device that both uses and generates electricity is expressed as follows:

[0121] P nJ =a×P total +b×Q n ;

[0122] Where P nJ Indicates the power consumption or power generation of the equipment that both uses and generates electricity, Q n Indicates the remaining power of the device that is both in use and in power.

[0123] S25. Obtain a large amount of historical operating data of comprehensive energy equipment, use fitting calculation to solve the empirical parameters in the empirical equation, and construct optimization models for different types of equipment.

[0124] S26. The edge monitoring system inputs the equipment type information and real-time operation data of the integrated energy equipment into the corresponding optimization model to calculate the operating power of the integrated energy equipment.

[0125] S27. Optimize the operation strategy of the comprehensive energy equipment based on the operation power calculation result. If the operation power calculation result is higher than the current power, the equipment power will be increased. If the operation power calculation result is lower than the current power, the equipment power will be decreased. The adjusted operation strategy will be used as the optimization strategy result.

[0126] S3. The cloud obtains historical monitoring data of each edge monitoring system, builds an equipment load prediction model, and uses the equipment load prediction model to output the load prediction results.

[0127] In the description of the present invention, the cloud obtains historical monitoring data of each edge monitoring system, builds a device load prediction model, and outputs a load prediction result using the device load prediction model, including the following steps:

[0128] S31. Obtain the entity information of the monitoring network where all edge monitoring systems are located, and use digital twin technology to build a virtual network and a visual virtual management platform in the cloud.

[0129] In the description of the present invention, obtaining the entity information of the monitoring network where all edge monitoring systems are located, using digital twin technology, building a virtual network and a visualized virtual management platform in the cloud includes the following steps:

[0130] S311. Obtain the topological information of the integrated energy equipment, gateway equipment and edge monitoring system that constitute the monitoring network, and send it to the cloud as entity information.

[0131] S312. According to the number and model of the devices in the physical information, establish virtual devices of corresponding number and model, send a query instruction, receive the online information and task information of the virtual devices for verification, and after the verification is passed, perform parameter status mapping between the physical device and the virtual device.

[0132] S313. Convert the entity information into a virtual topology structure, send it to the corresponding virtual device, establish a virtual network of the monitoring network digital twin, and synchronously update the energy monitoring data.

[0133] S314. Build a visual virtual management platform to display the monitoring data of the monitoring network and the virtual network in real time, as well as the data information to be executed and predicted.

[0134] S32. The cloud obtains historical monitoring data of the edge monitoring system and maps it to the virtual network, and extracts time characteristics, weather data and historical load data from the historical monitoring data.

[0135] Time characteristics: day of the week, hour;

[0136] Weather data: sunny, cloudy, rainy;

[0137] Historical load data: historical load data for the past 7 days.

[0138] S33. Clean and standardize the extracted historical monitoring data.

[0139] Among them, it is necessary to clean the historical monitoring data of each edge monitoring system, deal with missing values, outliers and noise, and standardize the features to ensure that they have similar scales.

[0140] S34. Establish a distributed gradient boosting library (XGBoost) model, and use the optimization objective as the minimization of the loss function. Iteratively use historical monitoring data to train the distributed gradient boosting library model. After the training is completed, the final equipment load prediction model is obtained.

[0141] Among them, the XGBoost model algorithm formula is:

[0142]

[0143] In the formula, F(x) represents the prediction result vector; f(x i ) represents the predicted value of the i-th decision tree.

[0144] The expression for minimizing the loss function is:

[0145]

[0146] In the formula, y irepresents the target variable (i.e., actual load), x i represents each prediction training vector, F represents the prediction result vector set, N represents the number of decision trees, F(x i ) represents the prediction result vector (i.e., prediction load) of the i-th decision tree, and f represents the prediction value of the decision tree.

[0147] S35. The equipment load prediction model performs prediction calculations, outputs load prediction results, and calculates the load curve for the next 24 hours, which is visualized on the virtual management platform.

[0148] Among them, the trained load forecasting model is used for forecasting calculation, and the predicted value (load forecasting result) can be expressed as: y new =F(X new );

[0149] In the formula, X new represents the new feature vector, including the new time feature, weather data and the historical load data of the past 7 days, y new Indicates the forecast result, i.e. the load forecast result.

[0150] S4. The edge monitoring system uploads the optimization strategy results to the cloud, and the cloud uses the load forecast results to verify the optimization strategy results and adjust the strategy.

[0151] In the description of the present invention, the edge monitoring system uploads the optimization strategy results to the cloud, and the cloud uses the load forecast results to verify the optimization strategy results and adjust the strategy, including the following steps:

[0152] S41. The edge monitoring system uploads the optimization strategy results to the cloud.

[0153] S42. The cloud determines whether there is a new combination in the optimization strategy results. If there is a new combination, the new combination is recorded, and then step S43 is executed to perform verification and adjustment, and the verification result is returned to the edge monitoring system. If there is no new combination, step S44 is executed.

[0154] S43, extract the power information in the optimization strategy result uploaded by the edge monitoring system, and substitute the load prediction result calculated by the equipment load prediction model into the energy conservation equation to check whether the equation is satisfied. If it is satisfied, the optimization strategy result is not adjusted. If it is not satisfied, the optimization strategy result is adjusted. The expression of the energy conservation equation is:

[0155] Total input power + total generated power = total power consumption.

[0156] S44. Determine whether there is new historical operation data of the equipment in the optimization strategy result. If there is new historical operation data of the equipment, execute step S43 to perform verification and adjustment, and return the verification result to the edge monitoring system. If there is no new historical operation data of the equipment, return the optimization strategy result uploaded by the edge monitoring system.

[0157] S5. The cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system performs autonomous optimization and verification correction on the integrated energy equipment.

[0158] In the description of the present invention, the cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system performs autonomous optimization and verification correction on the integrated energy equipment, including the following steps:

[0159] S51. The edge monitoring system receives verification and strategy adjustment results from the cloud. The edge monitoring system decomposes the target data that needs to be adjusted for the integrated energy equipment into multiple sub-targets. Each sub-target is adjusted according to specific parameters of the rated capacity (for example, the system goal is to set the energy storage to discharge at 500kW, the first step is set to discharge at 50kW, the second step is set to discharge at 200kW, and the third step is set to discharge at 500kW).

[0160] S52. After completing the adjustment of all sub-goals, input the operating data of the current integrated energy equipment into the optimization model, recalculate the operating power, and substitute all the data into the energy conservation equation. Combined with the pre-configured parameters for preventing transformer overload and backfeeding to the grid, adjust the actual operating power of the integrated energy equipment. If there is an overload risk, reduce the power consumption; if there is a backfeed risk, reduce the power generation power.

[0161] S53. Continuously monitor the actual operating status of integrated energy equipment and implement real-time dynamic adjustments.

[0162] In summary, with the help of the above technical solution of the present invention, the operation data of the enterprise's comprehensive energy equipment is collected through the edge monitoring system, and the data is processed and stored; then the edge optimization model is constructed using historical data, and the real-time operation data is input into the optimization model to calculate the optimization strategy; the cloud uses the XGBoost algorithm to construct a prediction model based on the historical data of each system, and calculates the prediction data; the edge system sends the calculation results to the cloud for verification. If a new combination mode changes or the equipment historical data is updated in the operation mode, the cloud will calculate and provide the verification correction results, and send them to the edge monitoring system; the edge system makes corrections based on the verification results to realize the real-time monitoring, autonomous optimization and verification correction of the comprehensive energy equipment by the edge system. By establishing a monitoring network of comprehensive energy equipment-gateway equipment-edge monitoring system based on a self-organizing network, real-time data collection, preprocessing and storage can be achieved, the efficiency of energy equipment can be improved, its performance can be monitored, abnormal conditions can be accurately identified, and a reliable data basis can be provided for decision-making, thereby reducing operating costs, increasing the reliability of energy systems, and improving the efficiency of energy management. By integrating historical monitoring data and weather information in the cloud and based on digital twin technology, efficient equipment load forecasting can be achieved, helping companies better plan energy supply and demand management, improve energy utilization efficiency, reduce waste, and provide key decision support through visual display, thereby reducing energy costs, improving sustainability, and improving energy management.

[0163] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A collaborative optimization method for integrated energy monitoring operation based on edge autonomy, characterized in that: The method comprises the following steps: S1. Establish a monitoring network of integrated energy equipment, gateway equipment and edge monitoring system, collect the operation data of integrated energy equipment through the edge monitoring system, and then store it in the database after pre-processing; S2, constructing an optimization model for the integrated energy equipment, inputting the real-time collected operation data into the optimization model to perform optimization strategy calculation, and outputting the optimization strategy result; S3, the cloud obtains the historical monitoring data of each edge monitoring system, builds an equipment load prediction model, and uses the equipment load prediction model to output a load prediction result; S4, the edge monitoring system uploads the optimization strategy result to the cloud, and the cloud uses the load forecast result to verify the optimization strategy result and adjust the strategy; S5. The cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system performs autonomous optimization and verification correction on the integrated energy equipment; The step of constructing the optimization model of the integrated energy equipment, inputting the real-time collected operation data into the optimization model to calculate the optimization strategy, and outputting the optimization strategy results comprises the following steps: S21, cleaning the operation data and aligning the data according to time synchronization; S22, classifying the comprehensive energy equipment into three types of equipment: power-consuming equipment, power-generating equipment, and both power-consuming and power-generating equipment according to power-generating demand and energy supply mode; S23, constructing physical models according to the working principles of different types of comprehensive energy equipment; S24, constructing empirical formulas according to the data relationships of different types of comprehensive energy equipment; S25, obtaining a large amount of historical operation data of the integrated energy equipment, solving the empirical parameters in the empirical equation by fitting calculation, and constructing optimization models for different equipment types; S26, the edge monitoring system inputs the equipment type information and real-time operation data of the integrated energy equipment into the corresponding optimization model to calculate the operating power of the integrated energy equipment; S27, optimizing the operation strategy of the comprehensive energy equipment according to the operation power calculation result, if the operation power calculation result is higher than the current power, the equipment power is increased, if the operation power calculation result is lower than the current power, the equipment power is decreased, and the adjusted operation strategy is used as the optimization strategy result; The edge monitoring system uploads the optimization strategy result to the cloud, and the cloud verifies the optimization strategy result and adjusts the strategy using the load forecast result, including the following steps: S41, the edge monitoring system uploads the optimization strategy result to the cloud; S42, the cloud determines whether there is a new combination in the optimization strategy result. If there is a new combination, the new combination is recorded, and then step S43 is executed to perform verification and adjustment, and the verification result is returned to the edge monitoring system. If there is no new combination, step S44 is executed; S43, extracting the power information in the optimization strategy result uploaded by the edge monitoring system, and substituting the power information and the load prediction result calculated by the equipment load prediction model into the energy conservation equation, and checking whether the equation is satisfied. If it is satisfied, the optimization strategy result is not adjusted; if it is not satisfied, the optimization strategy result is adjusted. The expression of the energy conservation equation is: Total input power + total power generation = total power consumption; S44, judging whether there is new historical operation data of the equipment in the optimization strategy result, if there is new historical operation data of the equipment, executing step S43 to perform verification and adjustment, and returning the verification result to the edge monitoring system, if there is no new historical operation data of the equipment, returning the optimization strategy result uploaded by the edge monitoring system; The cloud feeds back the verification and strategy adjustment results to the edge monitoring system, and the edge monitoring system autonomously optimizes and verifies the integrated energy equipment, including the following steps: S51, the edge monitoring system receives the verification and strategy adjustment results fed back from the cloud, and the edge monitoring system decomposes the target data that needs to be adjusted for the integrated energy equipment into multiple sub-targets, and each sub-target is adjusted according to a specific parameter of the rated capacity; S52, after completing the adjustment of all sub-goals, input the current operating data of the integrated energy equipment into the optimization model, recalculate the operating power, and substitute all the data into the energy conservation equation, and adjust the actual operating power of the integrated energy equipment in combination with the pre-configured transformer overload protection and power backfeed protection parameters. If there is an overload risk, reduce the power consumption; if there is a power backfeed risk, reduce the power generation; S53. Continuously monitor the actual operating status of the integrated energy equipment to achieve real-time dynamic adjustment.

2. According to claim 1, a collaborative optimization method for integrated energy monitoring operation based on edge autonomy is characterized in that: The establishment of a monitoring network of integrated energy equipment-gateway equipment-edge monitoring system, collecting the operation data of the integrated energy equipment through the edge monitoring system, and then storing it in the database after preprocessing includes the following steps: S11. Configure a gateway device between the integrated energy device and the edge monitoring system to establish a monitoring network based on a self-organizing network that coordinates the integrated energy device, the gateway device, and the edge monitoring system; S12, the gateway device plays the role of a communication bridge in the monitoring network, receives the operation data of the integrated energy equipment in real time, and transmits it to the edge monitoring system; S13, the edge monitoring system performs data analysis, information extraction and format conversion on the received original operation data to convert it into operation data that meets the data processing format; S14, the edge monitoring system performs rationality check and outlier identification on the converted operating data, marks the outliers as outlier quality codes, and stores the data in a database after processing.

3. According to claim 2, a collaborative optimization method for integrated energy monitoring operation based on edge autonomy is characterized in that: The configuration of a gateway device between the integrated energy device and the edge monitoring system to establish a monitoring network based on a self-organizing network of integrated energy devices, gateway devices, and edge monitoring systems includes the following steps: S111, taking all the integrated energy devices connected in the local area network where the gateway device is located as a cluster, and taking the gateway device as the cluster head of the cluster, generating a public key pair and a unique identifier of the cluster by the cluster head, and then sending them to the edge monitoring system through the gateway device; S112, selecting m integrated energy devices in the cluster as leaf nodes, and starting from the cluster head, assigning serial numbers to the leaf nodes according to the order of the Hamiltonian cycle, and the leaf nodes then use a pseudo-random generator to generate a pseudo-random sequence, and use a hash calculation method to obtain m leaf keys; S113, the edge monitoring system randomly selects one from all the leaf nodes as an authentication node, the authentication node negotiates with the cluster head to generate a shared key, the cluster head then uses the shared key to encrypt a leaf key and sends it to the authentication node, and at the same time the cluster head sends a request packet to the edge monitoring system, the edge monitoring system uses the public key pair to decrypt and verify, after the authentication node and the edge monitoring system complete the decryption and verification, the cluster head generates a session key and distributes it to all the integrated energy devices to establish a trust relationship between the devices; S114. A monitoring network of an integrated energy device-gateway device-edge monitoring system is formed by all the integrated energy devices, the gateway devices and the edge monitoring system that have established a trust relationship.

4. The collaborative optimization method for integrated energy monitoring operation based on edge autonomy according to claim 3 is characterized in that: The physical model is constructed according to the working principle of the different types of comprehensive energy equipment, including the following steps: S231. According to the circuit theory and electrical characteristics of the electrical equipment consuming electric energy, a physical model of the electrical equipment is established. The expression of the physical model of the electrical equipment is: P Y =V×I×cos(φ); Where P Y Indicates power consumption; V represents voltage; I represents current; cos(φ) represents the power factor; S232. Establish a physical model of the power generation equipment according to the power conversion characteristics of the power generation equipment. The expression of the physical model of the power generation equipment is: Where P F Indicates the power generated; η represents the conversion efficiency; S233, establishing a physical model of the device for both use and generation according to the power consumption and generation characteristics of the device for both use and generation, wherein the expression of the physical model of the device for both use and generation is: Where P J Represents net energy consumption.

5. The collaborative optimization method for integrated energy monitoring operation based on edge autonomy according to claim 4 is characterized in that: The constructing of empirical formulas according to the data relationships of different types of comprehensive energy equipment comprises the following steps: S241. Establish an empirical formula for the electrical equipment based on the relationship between the total voltage and the total current of the electrical equipment. The empirical formula for the electrical equipment is expressed as follows: P nF =a×V total +b×I total ; Where P nF Indicates the power consumption of electrical equipment; a and b both represent empirical parameters; V total Indicates the total voltage of all electrical equipment; I total Indicates the total current of all electrical equipment; S242. Establish an empirical formula for the power generation equipment according to the total system power of the power generation equipment. The empirical formula for the power generation equipment is expressed as follows: P nY =a×P total +b; Where P nY Indicates the power generation capacity of the power generation equipment; P total Indicates the total system power; S243, establishing an empirical formula for both use and generation equipment based on the relationship between the system power and the remaining power in the both use and generation equipment, wherein the empirical formula for both use and generation equipment is expressed as: P nJ =a×P total +b×Q n ; Where P nJ Indicates the power consumption or power generation of equipment that both consumes and generates electricity; Q n Indicates the remaining power of the device that is both in use and in power.

6. The collaborative optimization method for integrated energy monitoring operation based on edge autonomy according to claim 1 is characterized in that: The cloud obtains the historical monitoring data of each edge monitoring system, constructs an equipment load prediction model, and outputs a load prediction result using the equipment load prediction model, including the following steps: S31, obtaining entity information of the monitoring network where all the edge monitoring systems are located, and using digital twin technology to build a virtual network and a visual virtual management platform in the cloud; S32, the cloud obtains historical monitoring data of the edge monitoring system and maps it to the virtual network, and extracts time characteristics, weather data and historical load data from the historical monitoring data; S33, cleaning and standardizing the extracted historical monitoring data; S34, establish a distributed gradient enhancement library model, and use the optimization target as the minimization loss function, iteratively use the historical monitoring data to train the distributed gradient enhancement library model, and obtain the final equipment load prediction model after the training is completed. The expression of the minimization loss function is: In the formula, y i represents the target variable; x i Represents each prediction training vector; F represents the set of prediction result vectors; N represents the number of decision trees; F(x i ) represents the prediction result vector of the i-th decision tree; f represents the predicted value of the decision tree; S35. The equipment load prediction model performs prediction calculation, outputs the load prediction result, and calculates the load curve for the next 24 hours, which is visualized on the virtual management platform.

7. The collaborative optimization method for integrated energy monitoring operation based on edge autonomy according to claim 6 is characterized in that: The step of obtaining entity information of the monitoring network where all edge monitoring systems are located, and using digital twin technology to build a virtual network and a visualized virtual management platform in the cloud includes the following steps: S311, obtaining topology information of the integrated energy device, the gateway device and the edge monitoring system constituting the monitoring network, and sending the topology information to the cloud as entity information; S312, according to the number and model of the devices in the physical information, establish virtual devices of corresponding number and model, send a query instruction, receive the online information and task information of the virtual devices for verification, and perform parameter status mapping between the physical device and the virtual device after the verification is passed; S313, converting the entity information into a virtual topology structure, sending it to the corresponding virtual device, establishing a virtual network of the monitoring network digital twin, and synchronously updating the energy monitoring data; S314: Build a visual virtual management platform to display the monitoring data of the monitoring network and the virtual network in real time, as well as data information to be executed and predicted.

Citation Information

Patent Citations

  • Comprehensive energy management and control method and system based on smart cloud gateway

    CN116227757A

  • Comprehensive energy application service system

    CN109672723A

  • Comprehensive energy micro-grid optimization scheduling method and system based on cloud-edge collaboration

    CN114358504A

  • Method and device for comprehensive energy multi-station collaborative edge computing and scheduling

    CN116800745A

  • Systems and methods for real-time forecasting and predicting of electrical peaks and managing the energy, health, reliability, and performance of electrical power systems based on an artificial adaptive neural network

    US20090113049A1