A cloud edge collaboration-based load-side virtual power plant optimization regulation method and system

Abstract

The application provides a load side virtual power plant optimization regulation method and system based on cloud edge cooperation, deploys an edge-end cooperation system at a user side, and draws a user energy consumption portrait based on the edge-end cooperation system; the edge-end cooperation system sends the user portrait to a cloud end contract management system through the Internet; the cloud end contract management system gathers user regulation capabilities, and formulates a contract; a running management system formulates a regulation invitation quantity according to a power grid regulation demand and an effective contract in the contract management system, and publishes the demand to each edge-end cooperation system through a network layer protocol; the edge-end cooperation system is subjected to demand bidding; the clearing result is sent to the edge-end system; a control plan is executed through the running management system; online equipment load is disassembled, and a regulation sequence is subjected to load reduction from large to small according to current load until the load is greater than or equal to a required regulation load. The application solves the problem of how to effectively aggregate low load user side resources to participate in power grid interaction for a load type virtual power plant.

Description

Technical Field

[0001] This invention relates to the field of load-side virtual power plant technology, and in particular to a method and system for optimizing and controlling load-side virtual power plants based on cloud-edge collaboration. Background Technology

[0002] With the continuous development of the economy and society, the electricity demand from industry, commerce, and low-load users has been increasing year by year, resulting in a rapid increase in total electricity consumption. This has led to a continuous increase in the load on the power system, which has greatly affected the stability of power grid operation, and power system dispatch is facing severe challenges. Traditional power grid dispatch systems lack effective information communication support and intelligent processing methods, and can usually only reduce peak loads through orderly power consumption methods such as power rationing. However, this method is extremely unfriendly to the demand side and provides a poor user experience. If the peak-valley difference during peak hours is alleviated through power supply-side investments such as building power plants and supporting power grids, it will result in a waste of social resources. Against this backdrop, the vast amount of demand-side resources in the power grid has gradually gained attention, giving rise to demand response. Demand response uses preferential electricity prices or electricity fee refunds as incentives to guide users to reduce or shift some of their peak-hour electricity consumption, thereby ensuring the stability of the power system. In the demand response process, issues such as how to increase the participation willingness of the vast number of low-load users, how to predict the control potential, and how to flexibly dispatch demand-side resources have received considerable attention.

[0003] Currently, low-load user participation in grid interaction mainly falls into two categories: price-based and incentive-based. Price-based methods are more common, adjusting the time-of-use price through time-of-use pricing and other means, with users responding by adjusting their load accordingly. However, this approach fails to accurately reflect the uncertainty and flexibility of the dynamic electricity market, resulting in low user participation. Incentive-based methods involve users proactively reducing their load to receive compensation when the system requires it. This approach is less frequently used in practice, primarily involving high-energy-consuming users such as industrial enterprises and buildings. Ordinary low-load users often cannot participate because their load does not meet the control requirements. However, with the overall improvement in residential living standards, residential customer load is growing rapidly. Incorporating this resource into ancillary services will yield significant benefits. Therefore, this paper proposes a cloud-edge collaborative ancillary service optimization control method and system.

[0004] In the existing scheme, "A Method for Rapid Load Regulation of Household Users Oriented to Demand Response," the demand response organization first signs an agreement with the user, clarifying that the user will voluntarily cut off the upper limit of load in the event of a power grid emergency. Then, the user's commonly used appliances are classified according to usage time and importance, and their priorities are determined. When a regulation demand arises, a load regulation command is sent to the household user via power line carrier. The controller sends regulation commands to low-priority household appliances via short-range wireless WIFI, enabling direct communication between low-priority appliances and the transmission of mutual control signals to achieve primary regulation. The controller then sends regulation commands to high-priority household appliances via short-range wireless WIFI, enabling direct communication between high-priority appliances and the transmission of mutual control signals to achieve secondary regulation.

[0005] In the aforementioned scheme, the demand response organization failed to conduct a prior assessment of user response capabilities. When urgent control needs arise, it cannot perform a control potential analysis, potentially leading to an ineffective response that fails to meet grid control demands. Furthermore, using power line carrier communication results in significant signal attenuation when power line loads are high, as line impedance can drop below 1 ohm. In practical applications, point-to-point carrier signals can transmit over several kilometers when power lines are unloaded. However, when power lines are heavily loaded, transmission is limited to tens of meters, restricting the affected area and significantly reducing control potential. Summary of the Invention

[0006] This invention provides a load-side virtual power plant optimization and control method based on cloud-edge collaboration, which can expand participating users to the low-load user side. The method introduces a cloud-edge-device collaborative system, guides demand-side users to participate in grid interaction through a bidding model, and can optimize control strategies based on users' energy consumption habits, while avoiding the problem of system computational overload caused by too many participating users.

[0007] The methods include: Step 1: Deploy the edge-device collaboration system on the user side and create a user energy consumption profile based on the edge-device collaboration system; Step 2: After the user profile is generated, the edge-to-edge collaboration system sends the user profile to the cloud contract management system via the Internet; the cloud contract management system gathers the user's control capabilities, response time period, response duration, and response speed, and formulates contracts. Step 3: When the power grid issues a control request, the operation management system formulates the control invitation quantity based on the power grid control request and the effective contracts in the contract management system, and then publishes the request to each edge collaboration system through the network layer protocol. Step 4: The edge-to-edge collaborative system conducts demand bidding; Step 5: Conduct a competitive bidding process based on the operation management system and send the clearing results to the edge system; Step 6: Execute the control plan through the operation management system; Step 7: The edge-end collaboration system receives the execution command sent by the operation management system, and according to the current operating status of each user's device, it breaks down the load of online devices and adjusts the load in descending order of the current load until it is greater than or equal to the required adjustment load.

[0008] It should be further explained that in step 1, the user energy profile is obtained by using a neural network algorithm with time, date and power consumption as features, and using a time series model as shown in formula (1) to predict the power load W of various household appliances during different time periods, and combining the user's usage habits. To satisfy user comfort, the adjustable load of each user's electrical equipment is calculated by weighting the loads of the user during that time period, as shown in formula (2): (1) (2) In formula (1), x is the input layer; h is the hidden state layer; and y is the output layer. The meaning is: Output layer weights; The meaning is: the weights from the input layer to the hidden state layer; The meaning is: the weights between hidden state layers; Let p represent the electrical equipment, and let x represent the input features, W represent the model configuration parameters, g and f represent the current model activation functions, h represent the hidden layer output, t represent the current time, and t-1 represent the previous time. In formula 2, This represents the total load adjustment for the current user at time t. Let p be the energy load of equipment at time t. The threshold for device p to participate in load regulation while meeting user comfort requirements.

[0009] It should be further explained that in step 2, The contract drafting process is as follows; (3) (4) In formula (3), This refers to the user's basic subsidy price within a certain time period. To aggregate the predicted response volume for all users, Electricity price is based on time of day; Formula 4 is the weighted subsidy price based on the base subsidy price, taking into account response speed, duration, and response volume; where For response speed weights, For duration weights, Response weights, The weighted average price is the subsidy price.

[0010] It should be further explained that the calculation method for each weight in formula (4) is as follows: (5) Indicates the user's actual speed; S Agree This represents the ratio of the actual speed to the agreed-upon speed. The meaning is: the agreed response speed when the user participates in the response; (6) (7) Formula (5) is the response speed weight. Calculation method, in which This represents the maximum actual response load. To aggregate the baseline load, the weight is 0 when the actual load is less than the baseline load * 0.8, and 1.2 when the response speed is greater than the agreed speed. The duration weights are given in formula (6), where, This is the actual response time. To define the agreed response time, the weight is 0 when the duration is less than 0.8 of the agreed duration; the weight is the ratio of the duration to the agreed time when the duration is between 0.8 and 1 of the agreed duration; and the weight is 1.2 when the response time is greater than the agreed time. In formula (7), e represents the actual response load, e t The integral represents the actual response load at time t, and the integral represents the actual response amount during that time period. The response weights are as shown in formula (7), when Actual response quantity To define the agreed-upon response amount, when the actual response amount is less than 0.8 of the agreed-upon response amount, the current weight is 0; When the response quantity is greater than 0.8 times the agreed quantity but less than or equal to the agreed quantity, the weight is the ratio of the actual response quantity to the agreed quantity. When the response quantity is greater than the agreed response quantity, the weight is 1.2.

[0011] It should be further noted that after the contract is generated, it is published via SMS and WeChat official account, and users sign the contract electronically. At the same time, it provides a reference for the virtual power plant operation and control system to formulate control tasks.

[0012] It should be further noted that step 3 also includes: determining the amount of regulatory invitations through formula (8). To meet the planned demand, the number of invitations should not be less than 130% of the demand. (8) The meaning is: the planned response volume for the i-th user; n The meaning is that the plan is to invite n users.

[0013] It should be further explained that in step 4, after receiving the control plan issued by the operation management system, the edge-end collaboration system estimates the user's adjustable load and adjustment duration within the execution plan by using the user load control profile and configuring historical bidding strategies, formulates the current bidding strategy, and submits the bidding strategy to the cloud operation management system.

[0014] It should be further explained that in step 6, when the execution time is reached, the operation management system divides the time dimension according to the plan content, and dynamically selects the edge collaboration system that needs to participate in the control in each time period. Simultaneously, it receives feedback from the edge system in real time. If the control requirements are met, it waits for the next time period to repeat the resource selection and execution. If the regulatory requirements are not met, emergency resources will be mobilized to supplement them until the corresponding requirements are met in the current time period.

[0015] It should be further explained that in step 7, the load of each device is adjusted using formula (9); in, For the p-th device, the adjustable load is... Adjust demand for current users; (9).

[0016] The present invention also provides a load-side virtual power plant optimization and control system based on cloud-edge collaboration, the system comprising: a user energy consumption acquisition module, a data processing module, a bidding module, and a coordination and control module; The user energy consumption acquisition module is used to deploy the edge-end collaborative system on the user side and to draw user energy consumption profiles based on the edge-end collaborative system; The data processing module is used to send the user profile to the cloud contract management system via the Internet after the user profile is generated. The cloud contract management system gathers the user's control capabilities, response time period, response duration, and response speed, and formulates contracts. It is also used when the power grid issues a control request, the operation management system formulates the control invitation quantity based on the power grid control request and the effective contracts in the contract management system, and then publishes the request to each edge coordination system through the network layer protocol. The bidding module conducts demand bidding for the edge-end collaborative system, then clears the bids based on the operation management system, and sends the clearing results down to the edge-end system; The coordination and control module is used to execute the control plan through the operation and management system. The edge coordination system receives the execution command sent by the operation and management system, and according to the current operating status of each user's equipment, it breaks down the load of online equipment and reduces the load in the order of current load from large to small until it is greater than or equal to the required control load.

[0017] As can be seen from the above technical solutions, the present invention has the following advantages: This invention proposes a load-side virtual power plant optimization and control method based on cloud-edge collaboration. It constructs user energy consumption profiles at the edge using neural network algorithms, analyzes the control potential of each user, builds an overall load-side control capability curve, and accurately assesses the aggregated resource response capability at different time periods. Simultaneously, it uses an edge-end bidding model to maximize user participation in control. Using an internet-based network layer protocol for remote communication avoids problems such as severe power line carrier signal attenuation and limited communication distance. Through cloud-edge collaboration technology, it analyzes user energy consumption characteristics, generates user energy consumption profiles, and flexibly performs load control based on these profiles, maximizing user participation in load control while ensuring user comfort. This patent introduces a cloud-edge-end collaborative system, which can expand participating users to ordinary low-load users while avoiding system overload due to excessive participation.

[0018] The method also utilizes a user energy consumption acquisition module, a data processing module, a bidding module, and a coordination and control module to collect data on users' daily electricity consumption, create user load control profiles, analyze control potential, and conduct proactive bidding. It can extend ancillary service participation to any user side equipped with an edge-end collaboration system, effectively lowering the barrier to entry for ancillary services. Furthermore, it allows for flexible demand publishing through feedback from the user-side edge-end collaboration system, while using aggregated bidding to guide user-side response, effectively improving response completion rates. Detailed Implementation

[0019] The load-side virtual power plant optimization and control method based on cloud-edge collaboration provided by this invention addresses the problem that without prior assessment of user response capabilities, the ability to analyze regulation potential when emergency control needs arise may fail to meet grid control requirements, resulting in ineffective responses.

[0020] The cloud-edge collaboration technology of this invention addresses the increasing data volume generated by terminal devices as IoT and information communication technologies continue to develop. To accelerate data analysis and reduce system pressure, data processing is gradually shifting from a centralized architecture to a more global distributed architecture. Simultaneously, with the increasing intelligence of demand-side energy-consuming equipment, edge data processing capabilities have been initially acquired. Therefore, the application of cloud-edge collaboration technology in the field of power grid operation and management is gradually maturing.

[0021] The Demand Response (DR) method of this invention allows electricity users to temporarily change their inherent habitual electricity consumption patterns based on dynamic changes in electricity prices and policies, thereby altering or shifting their electricity load at a certain moment to respond to the power supply and thus ensuring the stability of the power grid system.

[0022] Load forecasting: Load forecasting is the process of determining load data at a specific point in the future, based on factors such as system operating characteristics, capacity expansion decisions, natural conditions, and social impacts, while meeting certain accuracy requirements. Load refers to the amount of electricity demand (power) or electricity consumption.

[0023] Baseline Load: Baseline load refers to a load curve estimated based on the user's historical load data. It reflects the user's electricity demand when not participating in a demand response project and is an important basis for the demand response project implementing agency to compensate users.

[0024] The load-side virtual power plant optimization and control method based on cloud-edge collaboration includes: Step 1: The edge-end collaborative system draws user energy consumption profiles.

[0025] The edge-end collaborative system is deployed on the user side. After the system is installed on the user side, the system's data acquisition module and user energy consumption acquisition module start collecting user energy consumption data through smart home devices or smart sockets and store the user energy consumption data in the edge system. When the acquisition module accumulates more than one month's worth of data, the data processing module combines the user's energy consumption information with information such as weekdays and months, analyzes the user's energy consumption curve based on neural networks, and draws a user load control profile based on the user's energy consumption habits.

[0026] The user load control profile is generated using a neural network algorithm, with time, date, and power consumption as features. A time series model, as shown in Formula 1, is used to predict the electricity load W of various household appliances during different time periods. Combined with user habits, and while ensuring user comfort, the adjustable load of each appliance is calculated by weighting the loads. The specific formula is shown in Formula 2 below. (1) (2) In formula (1), x is the input layer; h is the hidden state layer; and y is the output layer. The meaning is: Output layer weights; The meaning is: the weights from the input layer to the hidden state layer; The meaning is: the weights between hidden state layers; Let represent the electrical load of device p at time t, and x represent the input features (temperature, date, time, etc., which may vary depending on the actual data). W represents the model configuration parameters. g and f represent the current model activation functions. h represents the hidden layer output. t represents the current time, and t-1 represents the previous time. In Equation 2, This represents the total load adjustment for the current user at time t. Let p be the energy load of equipment at time t. The threshold for device p to participate in load regulation while meeting user comfort requirements.

[0027] Step 2. Adjust the contract formulation of the resource management system.

[0028] After the user profile is generated, the edge-to-edge collaboration system transmits the user profile to the cloud-based contract management system via the internet. The cloud-based contract management system aggregates user control capabilities, response time periods, response durations, response speeds, etc., and formulates contracts. The contract formulation process is as follows.

[0029] (3) (4) In formula (3), This refers to the user's basic subsidy price within a certain time period. To aggregate the predicted response volume for all users, This refers to the time-of-use electricity price. Formula 4 is the weighted subsidy price based on the base subsidy price, taking into account response speed, duration, and response volume. Wherein, For weighted average subsidy prices, For response speed weights, For duration weights, Response weights. The calculation methods for each weight are as follows.

[0030] (5) Indicates the user's actual speed; S Agree This represents the ratio of the actual speed to the agreed-upon speed. The meaning is: the agreed response speed when the user participates in the response; (6) (7) Formula (5) is the response speed weight. Calculation method, in which This represents the maximum actual response load. To aggregate the baseline load, when the actual load is less than the baseline load * 0.8, the weight is 0, meaning the response is invalid. When the response speed is less than or equal to the agreed speed, the weight is the ratio of the actual speed to the agreed speed; when the response speed is greater than the agreed speed, the weight is 1.2.

[0031] The duration weights are shown in formula (6), where, This is the actual response time. To define the agreed response time, the weight is 0 when the duration is less than 0.8 of the agreed duration; the weight is the ratio of the duration to the agreed time when the duration is between 0.8 and 1 of the agreed duration; and the weight is 1.2 when the response time is greater than the agreed time.

[0032] In formula (7), e represents the actual response load, e t The integral represents the actual response load at time t, and the integral represents the actual response amount during that time period. The response weights are shown in formula (7). Actual response quantity The weight is 0 when the actual response is less than 0.8 times the agreed response; when the response is greater than 0.8 times the agreed response, but less than or equal to the agreed response, the weight is the ratio of the actual response to the agreed response; when the response is greater than the agreed response, the weight is 1.2.

[0033] After the contract is generated, it is published via SMS, WeChat official account, etc. Users sign the contract electronically, which is considered as agreeing to the subsequent control operations, ensuring the rationality of the control, and providing a reference for the virtual power plant operation control system to formulate control tasks.

[0034] Step 3. Run the management system to release the plan.

[0035] When the power grid issues a control request, the operation management system formulates the control invitation quantity for that event based on the power grid control request and the effective contracts in the contract management system, as shown in Formula 8. To meet the planned demand, the number of invitations should be no less than 130% of the total demand. After review, the demand will be published to the collaborative systems of each edge through the network layer protocol.

[0036] (8).

[0037] The meaning is: the planned response volume for the i-th user; n The meaning is that the plan is to invite n users.

[0038] Step 4. The edge-end collaborative system conducts demand bidding. After receiving the control plan issued by the operation management system, the edge-end collaborative system estimates the adjustable load and adjustment duration of the user within the execution plan by using the user load control profile and historical bidding strategies, formulates the bidding strategy for this time, and submits the bidding strategy to the cloud operation management system.

[0039] Step 5. The operation management system performs a bid clearing process. After the bidding period ends, the operation management system clears the bids reported by the edge system and sends the clearing results back to the edge system. For successful bidders, the edge system records the execution time and control requirements; for unsuccessful bidders, the edge system records the content of this bidding process for reference in future bidding.

[0040] Step 6. The operation management system coordinates and controls the execution of the plan. When the execution time arrives, the operation management system first divides the time dimension according to the plan content. In each time period, it dynamically selects the edge collaborative systems that need to participate in the control for the current time period. At the same time, it receives feedback from the edge systems in real time. If the control requirements are met, it waits for the next time period to repeat the resource selection and issue the execution. If the control requirements are not met, it calls emergency resources to supplement them until the corresponding requirements are met in the current time period.

[0041] Step 7. The edge-end collaborative system executes the control command. Upon receiving the execution command from the cloud-based operation management system, the edge-end collaborative system breaks down the online device load based on the current operating status of each user's device. The control sequence is to reduce the load from highest to lowest according to the current load, until it is greater than or equal to the required control load. The load control principle for each device is shown in Formula 9 below, where... The adjustable load is for the p-th device; Adjust the current user demand.

[0042] (9).

[0043] Based on the above method, this invention constructs an edge-to-edge collaborative system on the user side. By connecting to the cloud platform of the equipment manufacturer or smart socket, it can realize functions such as user energy consumption collection, data processing, bidding, and coordinated control. Through user profiling, it proposes the optimal participation strategy for users while ensuring user comfort, and maximizes the response potential of low-load users.

[0044] The load-side virtual power plant optimization and control method uses a contract-based approach. Contracts are formulated based on users' electricity consumption habits and adjustable load, thereby maximizing user participation and fully tapping into user response potential.

[0045] The load-side virtual power plant optimization and control method adopts a cloud-edge collaborative strategy based on internet communication protocols. On the one hand, it decentralizes user data processing to the user side, avoiding the pressure of large-scale data transmission and analysis; on the other hand, it effectively protects user privacy and prevents data leakage. Simultaneously, it uses a bidding mechanism to guide participation, with the cloud handling the bidding clearing and the edge developing bidding strategies based on user energy consumption habits to participate in the aggregate bidding. The response process is controlled holistically by the cloud, and the edge, based on cloud instructions, flexibly controls the load of home appliances while ensuring user comfort.

[0046] The following are embodiments of the load-side virtual power plant optimization and control system based on cloud-edge collaboration provided in this disclosure. This system and the load-side virtual power plant optimization and control method based on cloud-edge collaboration in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the load-side virtual power plant optimization and control system based on cloud-edge collaboration, please refer to the embodiments of the load-side virtual power plant optimization and control method based on cloud-edge collaboration described above.

[0047] The system includes: a user energy consumption acquisition module, a data processing module, a bidding module, and a coordination and control module; The user energy consumption acquisition module is used to deploy the edge-end collaborative system on the user side and to draw user energy consumption profiles based on the edge-end collaborative system; The data processing module is used to send the user profile to the cloud contract management system via the Internet after the user profile is generated. The cloud contract management system gathers the user's control capabilities, response time period, response duration, and response speed, and formulates contracts. It is also used when the power grid issues a control request, the operation management system formulates the control invitation quantity based on the power grid control request and the effective contracts in the contract management system, and then publishes the request to each edge coordination system through the network layer protocol. The bidding module conducts demand bidding for the edge-end collaborative system, then clears the bids based on the operation management system, and sends the clearing results down to the edge-end system; The coordination and control module is used to execute the control plan through the operation and management system. The edge coordination system receives the execution command sent by the operation and management system, and according to the current operating status of each user's equipment, it breaks down the load of online equipment and reduces the load in the order of current load from large to small until it is greater than or equal to the required control load.

[0048] In one embodiment of the present invention, a possible implementation of a cloud-edge collaborative load-side virtual power plant optimization and control system will be provided, and its specific implementation will be described in a non-limiting manner.

[0049] The user energy consumption acquisition module collects users' daily electricity consumption data, records users' daily electricity consumption habits, and accumulates users' historical energy consumption data. The data processing module preprocesses the user's energy consumption data and uses the preprocessed data in combination with other dimensions of data obtained from weekdays to create a user energy consumption profile using a neural network algorithm. The bidding module is a crucial module for user-side participation in grid interaction. Upon receiving the control plan from the virtual power plant operation control system, the bidding module plots the user's energy consumption curve within the control demand period based on the user's energy consumption profile, analyzes the user's control potential, and participates in the bidding according to the bidding rules (Note: the bidding rules can be optimized based on previous bidding results). If the bid is successful, the coordination control module, based on the bidding content, coordinates and controls devices such as air conditioners and water heaters through the connection to the smart home cloud platform or smart sockets within the control demand period, adjusting the energy load of the devices to meet the agreed content.

[0050] Secondly, in the cloud, the control resource management system formulates contracts based on the load-side response data provided by the edge system during the contract period, and sends them to the user side via SMS, WeChat official account, etc. After signing the contract, the user is deemed to have agreed to participate in subsequent operations, ensuring the rationality of control and providing a reference for the formulation of control tasks for the virtual power plant operation control system.

[0051] Finally, a virtual power plant operation control system is built in the cloud for overall operation control. This system includes a demand publishing module, a centralized bidding module, and a coordinated control module. The demand publishing module formulates a single execution plan based on the grid regulation requirements and the controllable load on the energy consumption side, and publishes the demand to each edge collaborative system through a network layer protocol. The centralized bidding module conducts overall bidding and clearing based on the bidding information fed back by each edge collaborative system, and sends the clearing results to each edge collaborative system. When the regulation task execution time arrives, the coordinated control module, based on the pre-defined task and participants, and referring to the task completion status, sends regulation information to each edge collaborative system in real time, and obtains feedback results from the edge collaborative systems in real time.

[0052] Through the above system, the participating resources can be extended to any user side with an edge-to-edge collaborative system installed, mobilizing a large number of low-load users to participate in regulation and control, increasing the participation rate of ordinary low-load users, and effectively reducing the participation threshold.

[0053] The units and algorithm steps described in the embodiments of the cloud-edge collaborative load-side virtual power plant optimization and control method provided by this invention can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0054] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0055] The cloud-edge collaborative load-side virtual power plant optimization and control method of this invention can be written in one or more programming languages ​​or a combination thereof to perform the operations disclosed herein. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or power server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (exemplarily using an Internet service provider for Internet connection).

[0056] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A load-side virtual power plant optimization and control method based on cloud-edge collaboration, characterized in that, The methods include: Step 1: Deploy the edge-device collaboration system on the user side and create a user energy consumption profile based on the edge-device collaboration system; Step 2: After the user profile is generated, the edge-to-edge collaboration system sends the user profile to the cloud contract management system via the Internet; the cloud contract management system gathers the user's control capabilities, response time period, response duration, and response speed, and formulates contracts. In step 2, The contract drafting process is as follows; (3) (4) In formula (3), This refers to the user's basic subsidy price within a certain time period. To aggregate the predicted response volume for all users, Electricity price is based on time of day; Formula 4 is the weighted subsidy price based on the base subsidy price, taking into account response speed, duration, and response volume; where For response speed weights, For duration weights, Response weights, The weighted average subsidy price; Step 3: When the power grid issues a control request, the operation management system formulates the control invitation quantity based on the power grid control request and the effective contracts in the contract management system, and then publishes the request to each edge collaboration system through the network layer protocol. Step 3 also includes: determining the amount of regulatory invitations using formula (8), To meet the planned demand, the number of invitations should not be less than 130% of the demand. (8) The meaning is: the planned response volume for the i-th user; n The meaning is that the plan is to invite n users; Step 4: The edge-to-edge collaborative system conducts demand bidding; Step 5: Conduct a competitive bidding process based on the operation management system and send the clearing results to the edge system; Step 6: Execute the control plan through the operation management system; Step 7: The edge-end collaboration system receives the execution command sent by the operation management system, and according to the current operating status of each user's device, it breaks down the load of the online device, and the control order reduces the load from large to small according to the current load until it is greater than or equal to the required control load.

2. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, In step 1, the user energy profile is obtained by using a neural network algorithm with time, date and power consumption as features, and using a time series model as shown in formula (1) to predict the power load W of various home appliances during different time periods, and combining the user's usage habits. To satisfy user comfort, the adjustable load of each user's electrical equipment is calculated by weighting the loads of the user during that time period, as shown in formula (2): (1) (2) In formula (1), x is the input layer; h is the hidden state layer; and y is the output layer. The meaning is: Output layer weights; The meaning is: the weights from the input layer to the hidden state layer; The meaning is: the weights between hidden state layers; Let p represent the electrical equipment, and let x represent the input features, W represent the model configuration parameters, g and f represent the current model activation functions, h represent the hidden layer output, t represent the current time, and t-1 represent the previous time. In formula 2, This represents the total load adjustment for the current user at time t. Let p be the energy load of equipment at time t. The threshold for device p to participate in load regulation while meeting user comfort requirements.

3. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, The calculation methods for each weight in formula (4) are as follows; (5) Indicates the user's actual speed; S Agree This represents the ratio of the actual speed to the agreed-upon speed. The meaning is: the agreed response speed when the user participates in the response; (6) (7) Formula (5) is the response speed weight. Calculation method, in which This represents the maximum actual response load. To aggregate the baseline load, when the actual load is less than the baseline load... When the value is 0.8, the weight is 0; when the response speed is greater than the agreed speed, the weight is 1.

2. The duration weights are given in formula (6), where, This is the actual response time. To define the agreed response time, the weight is 0 when the duration is less than 0.8 of the agreed duration; the weight is the ratio of the duration to the agreed time when the duration is between 0.8 and 1 of the agreed duration; and the weight is 1.2 when the response time is greater than the agreed time. In formula (7), e represents the actual response load, e t The integral represents the actual response load at time t, and the integral represents the actual response amount during that time period. The response weights are as shown in formula (7), when Actual response quantity To define the agreed-upon response amount, when the actual response amount is less than 0.8 of the agreed-upon response amount, the current weight is 0; When the response quantity is greater than 0.8 times the agreed quantity but less than or equal to the agreed quantity, the weight is the ratio of the actual response quantity to the agreed quantity. When the response quantity is greater than the agreed response quantity, the weight is 1.

2.

4. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, After the contract is generated, it is published via SMS and WeChat official account, and users sign the contract electronically. At the same time, it provides a reference for the virtual power plant operation and control system to formulate control tasks.

5. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, In step 4, after receiving the control plan issued by the operation management system, the edge-end collaboration system estimates the user's adjustable load and adjustment duration within the execution plan by using the user load control profile and configuring historical bidding strategies, formulates the current bidding strategy, and submits the bidding strategy to the cloud operation management system.

6. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, In step 6, when the execution time arrives, the operation management system divides the time dimension according to the plan content, and dynamically selects the edge collaboration system that needs to participate in the control in each time period. Simultaneously, it receives feedback from the edge system in real time. If the control requirements are met, it waits for the next time period to repeat the resource selection and execution. If the regulatory requirements are not met, emergency resources will be mobilized to supplement them until the corresponding requirements are met in the current time period.

7. The load-side virtual power plant optimization and control method based on cloud-edge collaboration according to claim 1, characterized in that, In step 7, the load of each device is adjusted using formula (9); in, For the p-th device, the adjustable load is... Adjust demand for current users; (9)。 8. A load-side virtual power plant optimization and control system based on cloud-edge collaboration, characterized in that, The system adopts the load-side virtual power plant optimization and control method based on cloud-edge collaboration as described in any one of claims 1 to 7; The system includes: a user energy consumption acquisition module, a data processing module, a bidding module, and a coordination and control module; The user energy consumption acquisition module is used to deploy the edge-end collaborative system on the user side and to draw user energy consumption profiles based on the edge-end collaborative system; The data processing module is used to send the user profile to the cloud contract management system via the Internet after the user profile is generated. The cloud contract management system gathers the user's control capabilities, response time period, response duration, and response speed, and formulates contracts. It is also used when the power grid issues a control request, the operation management system formulates the control invitation quantity based on the power grid control request and the effective contracts in the contract management system, and then publishes the request to each edge coordination system through the network layer protocol. The bidding module conducts demand bidding for the edge-end collaborative system, then clears the bids based on the operation management system, and sends the clearing results down to the edge-end system; The coordination and control module is used to execute the control plan through the operation and management system. The edge coordination system receives the execution command sent by the operation and management system, and according to the current operating status of each user's equipment, it breaks down the load of online equipment and reduces the load in the order of current load from large to small until it is greater than or equal to the required control load.

Images