Virtual power plant scheduling device and method

Abstract

The invention discloses a virtual power plant scheduling device and method, and belongs to the technical field of load control, which combines the analysis of a basic mode and scheduling parameters, constructs a security analysis mode, achieves the expansion of a mode learning set by means of previous parameters, and greatly improves the information sent by the final mode. The method also achieves internal iteration, has good adaptability to long-time application deployment or large-batch parameter scheduling, and effectively avoids the defects that parameter verification in the prior art depends on a principle mode and manual verification, is insufficient in efficiency and is often unfavorable due to the effect of human factors.

Description

Technical Field

[0001] The present invention belongs to the technical field of load control, and in particular relates to a virtual power plant scheduling device and method. Background Art

[0002] The core of the concept of virtual power plant can be summarized as "communication" and "aggregation". The key technologies of virtual power plant mainly include coordinated control technology, smart metering technology and information and communication technology. The most attractive function of virtual power plant is that it can aggregate DER to participate in the operation of electricity market and auxiliary service market, and provide management and auxiliary services for distribution network and transmission network. The solution of "virtual power plant" has great market potential in my country. It is undoubtedly a good choice for China, which is facing the contradiction between "electricity shortage and low energy efficiency".

[0003] In terms of virtual power plant scheduling, the existing technical solution with patent publication number "CN114444851A" is often used to execute virtual power plant scheduling, which obtains scheduling parameters related to virtual power plant scheduling and sends the scheduling parameters to the virtual power plant robust optimization scheduling model to perform virtual power plant scheduling analysis.

[0004] Before sending the dispatch parameters to the virtual power plant robust optimization dispatch model to perform virtual power plant dispatch analysis, the dispatch parameters must be verified. Parameter verification is an important stage to ensure the quality of sending the dispatch parameters to the virtual power plant robust optimization dispatch model to perform virtual power plant dispatch analysis. The current parameter verification relies on principle models and manual verification, which is not efficient enough and often suffers from the influence of human factors.

[0005] Invention parameters

[0006] In order to solve the defects in the prior art, the present invention proposes a virtual power plant scheduling device and method, which combines the analysis of the basic mode and the scheduling parameters to construct a safe analysis mode, and relies on the expansion of the mode learning set by previous parameters, which greatly improves the information sent by the final mode. The test iteration process of the present invention also achieves internal iteration, and has good adaptability to long-term deployment applications or large-scale scheduling parameters, effectively avoiding the defects of the prior art that parameter verification relies on principle models and manual verification, lacks efficiency and is often adversely affected by human factors.

[0007] The present invention uses the following technical solutions.

[0008] A method for information management of load control, comprising:

[0009] Obtain dispatch parameters related to virtual power plant dispatch to perform verification, and send the verified dispatch parameters to the virtual power plant robust optimization dispatch model to perform virtual power plant dispatch analysis;

[0010] Methods for performing verification include:

[0011] Step 1: Use the scheduling parameter to obtain the parameter attributes of the scheduling parameter;

[0012] Step 2: construct a scheduling parameter grouping selection mode according to the basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters;

[0013] Step 3: Perform regression selection according to the scheduling parameter grouping selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters.

[0014] Furthermore, step 1 is:

[0015] Step 1-1: Obtain the parameter type of the scheduling parameter, the parameter source host ID of the scheduling parameter, and the value of the scheduling parameter as an instant value;

[0016] Step 1-2: Use the instantaneous value to obtain the parameter attribute of the scheduling parameter.

[0017] Further, steps 1-2 are:

[0018] Step 1-2-1: Determine whether the parameter type of the scheduling parameter meets the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-2. Otherwise, exit the verification.

[0019] Step 1-2-2: Determine whether the parameter source host ID of the scheduling parameter satisfies the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-3. Otherwise, exit verification.

[0020] Step 1-2-3: using the value of the scheduling parameter as the first verification parameter in the scheduling parameter set, and using the parameter type of the scheduling parameter corresponding to the value of the scheduling parameter and the parameter source host ID of the scheduling parameter as the second verification parameter in the scheduling parameter set;

[0021] Step 1-2-4: Use the scheduling parameter centralized verification parameter 1 and the scheduling parameter centralized verification parameter 2 as parameter attributes of the scheduling parameters.

[0022] Furthermore, step 2 is:

[0023] Step 2-1: Obtain the previous parameter attributes corresponding to the values ​​of the scheduling parameters of the parameter attributes of the scheduling parameters to form a learning set 1;

[0024] Step 2-2: Obtain the parameter type of the scheduling parameter corresponding to the learning set 1 and the parameter source host ID of the scheduling parameter to form a test set 1;

[0025] Step 2-3: Use the learning set 1 and the test set 1 to construct a scheduling parameter centralized verification starting selection mode according to the basic mode;

[0026] Step 2-4: Use the parameter attributes of the scheduling parameters to obtain the scheduling parameter grouping selection mode according to the scheduling parameter centralized verification starting selection mode.

[0027] Further, steps 2-3 are:

[0028] Step 2-3-1: Use the learning set 1 as the input parameter, and the scheduling parameters of the learning set 1 corresponding to the scheduling parameters after centralized verification as the output parameters, and perform learning according to the basic mode to obtain the scheduling parameter centralized verification basic selection mode;

[0029] Step 2-3-2: Send the test set 1 to the scheduling parameter centralized verification basic selection mode to obtain the scheduling parameter centralized verification basic selection information;

[0030] Step 2-3-3: Determine whether the basic selection information for centralized verification of scheduling parameters and the scheduling parameters after the corresponding centralized verification of scheduling parameters in test set 1 are all consistent. If so, send out the basic selection mode for centralized verification of scheduling parameters as the starting selection mode for centralized verification of scheduling parameters. Otherwise, use the subset of the test set 1 that does not have the corresponding basic selection information for centralized verification of scheduling parameters to add to the learning set 1, and then execute step 2-3-1.

[0031] Furthermore, all consistency means that each subset in the test set 1 has corresponding scheduling parameters to centrally verify the basic selection information.

[0032] Furthermore, the basic model is a support vector machine SVM.

[0033] Further, steps 2-4 are:

[0034] Step 2-4-1: sending the dispatch parameter centralized verification parameter 1 of the parameter attribute of the dispatch parameter to the dispatch parameter centralized verification starting selection mode to obtain the dispatch parameter centralized verification starting selection information;

[0035] Step 2-4-2: Determine whether the scheduling parameter centralized verification parameter 1 has a corresponding past parameter attribute. If yes, execute step 2-4-3. Otherwise, send the scheduling parameter centralized verification starting selection mode as the scheduling parameter grouping selection mode.

[0036] Step 2-4-3: Determine whether the dispatching parameter centralized verification start selection information is consistent with the dispatching parameter centralized verification start selection information corresponding to the previous parameter attribute. If yes, send the dispatching parameter centralized verification start selection mode as the dispatching parameter group selection mode. Otherwise, execute step 2-4-4;

[0037] Step 2-4-4: Determine whether parameter 2 of the scheduling parameter set of the parameter attributes of the scheduling parameter is in a reasonable state. If so, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-4-1. Otherwise, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-1.

[0038] Furthermore, the reasonable condition is that the parameter type of the scheduling parameter of the second scheduling parameter set verification parameter and the parameter source host ID of the scheduling parameter both correspond to the current parameter scenario.

[0039] Furthermore, step 3 is:

[0040] Step 3-1: Get the current time point as the instant starting time point u;

[0041] Step 3-2: Determine whether the second scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter at the instant selection starting time point u is in a reasonable state. If so, send the scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter into the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection reserved information, and execute step 3-3. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing;

[0042] Step 3-3: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+1 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification immediate selection and retain information, and execute step 3-4. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing;

[0043] Step 3-4: determine whether the information reserved before the centralized verification of the dispatching parameters and the information reserved after the centralized verification of the dispatching parameters are the same. If so, send the information reserved before the centralized verification of the dispatching parameters as the dispatching parameters after the centralized verification of the dispatching parameters at time u. Otherwise, execute step 3-5;

[0044] Step 3-5: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+2 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection additional information, and execute step 3-6. Otherwise, exit the verification;

[0045] Step 3-6: Determine whether the additional information selected in the centralized verification of the scheduling parameters is the same as the information retained after the centralized verification of the scheduling parameters. If so, send the information retained before the centralized verification of the scheduling parameters as the scheduling parameters after the centralized verification of the scheduling parameters at time u. Otherwise, exit the verification and send the parameter attributes of the scheduling parameters at time points u, u+1 and u+2 respectively.

[0046] Further, step 3-3 is:

[0047] Step 3-3-1: Determine whether the current centralized verification parameter 2 of the scheduling parameter and the centralized verification parameter 2 of the previous scheduling parameter have matching parameters. If so, the test is reasonable and the verification is terminated. Otherwise, obtain the centralized verification parameter 1 corresponding to the centralized verification parameter 2 of the scheduling parameter;

[0048] Step 3-3-2: Use the scheduling parameter centralized verification parameter one and the scheduling parameter centralized verification parameter two to form a refresh group to perform refresh processing.

[0049] Further, step 3-3-2 is:

[0050] Step 3-3-2-1: Use the scheduling parameter set to verify parameter one to form a refresh learning set;

[0051] Step 3-3-2-2: Use the scheduling parameter set to verify parameter 2 to form a refresh test set;

[0052] Step 3-3-2-3: Combine the refreshed learning set and learning set 1 to form current learning set 1, and execute step 2-3-1;

[0053] Step 3-3-2-4: Use the refresh test set to replace test set one, and execute step 2-3-2.

[0054] A virtual power plant dispatching device, comprising:

[0055] An attribute module, which is used to obtain parameter attributes of the scheduling parameters using the scheduling parameters;

[0056] A selection module, which is used to construct a scheduling parameter group selection mode according to a basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters;

[0057] The verification module is used to perform regression selection according to the scheduling parameter group selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters.

[0058] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0059] The analysis of the basic mode and the scheduling parameters are combined to construct a safe analysis mode, and the expansion of the mode learning set is achieved by relying on the previous parameters, which greatly improves the information sent by the final mode. The test iteration process of the present invention also achieves internal iteration, and has good adaptability to long-term deployment of applications or large-scale scheduling parameters, effectively avoiding the defects of the prior art that parameter verification relies on principle models and manual verification, lacks efficiency and is often adversely affected by human factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a partial flow chart of the virtual power plant dispatching method described in the present invention;

[0061] Figure 2 It is a partial structural diagram of the virtual power plant dispatching device described in the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely expressed in combination with the drawings in the embodiments of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. According to the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0063] like Figure 1 As shown, a virtual power plant scheduling device according to the present invention includes:

[0064] Obtain dispatch parameters related to virtual power plant dispatch to perform verification, and send the verified dispatch parameters to the virtual power plant robust optimization dispatch model to perform virtual power plant dispatch analysis;

[0065] Methods for performing verification include:

[0066] Step 1: Use the scheduling parameter to obtain the parameter attributes of the scheduling parameter;

[0067] Step 2: construct a scheduling parameter grouping selection mode according to the basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters;

[0068] Step 3: Perform regression selection according to the scheduling parameter grouping selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters.

[0069] In a preferred but non-limiting embodiment of the present invention, step 1 is:

[0070] Step 1-1: Get the parameter type of the scheduling parameter, the parameter source host ID of the scheduling parameter and the value of the scheduling parameter as an instant value; the parameter source is often a computer.

[0071] Step 1-2: Use the instantaneous value to obtain the parameter attribute of the scheduling parameter.

[0072] In a preferred but non-limiting embodiment of the present invention, steps 1-2 are:

[0073] Step 1-2-1: Determine whether the parameter type of the scheduling parameter meets the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-2. Otherwise, exit the verification.

[0074] Step 1-2-2: Determine whether the parameter source host ID of the scheduling parameter satisfies the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-3. Otherwise, exit verification.

[0075] Step 1-2-3: using the value of the scheduling parameter as the first verification parameter in the scheduling parameter set, and using the parameter type of the scheduling parameter corresponding to the value of the scheduling parameter and the parameter source host ID of the scheduling parameter as the second verification parameter in the scheduling parameter set;

[0076] Step 1-2-4: Use the scheduling parameter centralized verification parameter 1 and the scheduling parameter centralized verification parameter 2 as parameter attributes of the scheduling parameters.

[0077] In a preferred but non-limiting embodiment of the present invention, step 2 is:

[0078] Step 2-1: Obtain the past parameter attributes corresponding to the values ​​of the scheduling parameters of the parameter attributes of the scheduling parameters to form a learning set one; a past parameter attribute is a subset of the learning set one.

[0079] Step 2-2: Obtain the parameter type of the scheduling parameter corresponding to the learning set one and the parameter source host ID of the scheduling parameter to form a test set one; a parameter source host ID is a subset of the test set one.

[0080] Step 2-3: Use the learning set 1 and the test set 1 to construct a scheduling parameter centralized verification starting selection mode according to the basic mode;

[0081] Step 2-4: Use the parameter attributes of the scheduling parameters to obtain the scheduling parameter grouping selection mode according to the scheduling parameter centralized verification starting selection mode.

[0082] In a preferred but non-limiting embodiment of the present invention, steps 2-3 are:

[0083] Step 2-3-1: Use the learning set 1 as the input parameter, and the scheduling parameters of the learning set 1 corresponding to the scheduling parameters after centralized verification as the output parameters, and perform learning according to the basic mode to obtain the scheduling parameter centralized verification basic selection mode;

[0084] Step 2-3-2: Send the test set 1 to the scheduling parameter centralized verification basic selection mode to obtain the scheduling parameter centralized verification basic selection information;

[0085] Step 2-3-3: Determine whether the basic selection information for centralized verification of scheduling parameters and the scheduling parameters after the corresponding centralized verification of scheduling parameters in test set 1 are all consistent. If so, send out the basic selection mode for centralized verification of scheduling parameters as the starting selection mode for centralized verification of scheduling parameters. Otherwise, use the subset of the test set 1 that does not have the corresponding basic selection information for centralized verification of scheduling parameters to add to the learning set 1, and then execute step 2-3-1.

[0086] In a preferred but non-limiting embodiment of the present invention, all consistency is that each subset in the test set 1 has corresponding scheduling parameter centralized verification basic selection information.

[0087] In a preferred but non-limiting embodiment of the present invention, the basic model is a support vector machine SVM.

[0088] In a preferred but non-limiting embodiment of the present invention, steps 2-4 are:

[0089] Step 2-4-1: sending the dispatch parameter centralized verification parameter 1 of the parameter attribute of the dispatch parameter to the dispatch parameter centralized verification starting selection mode to obtain the dispatch parameter centralized verification starting selection information;

[0090] Step 2-4-2: Determine whether the scheduling parameter centralized verification parameter 1 has a corresponding past parameter attribute. If yes, execute step 2-4-3. Otherwise, send the scheduling parameter centralized verification starting selection mode as the scheduling parameter grouping selection mode.

[0091] Step 2-4-3: Determine whether the dispatching parameter centralized verification start selection information is consistent with the dispatching parameter centralized verification start selection information corresponding to the previous parameter attribute. If yes, send the dispatching parameter centralized verification start selection mode as the dispatching parameter group selection mode. Otherwise, execute step 2-4-4;

[0092] Step 2-4-4: Determine whether parameter 2 of the scheduling parameter set of the parameter attributes of the scheduling parameter is in a reasonable state. If so, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-4-1. Otherwise, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-1.

[0093] In a preferred but non-limiting embodiment of the present invention, the reasonable condition is that the parameter type of the scheduling parameter of the second scheduling parameter set verification parameter and the parameter source host ID of the scheduling parameter both correspond to the current parameter scenario.

[0094] The parameter scenario is an overall architecture that contains the resources and format specifications required for parameter storage, scheduling, and transmission.

[0095] The present invention flexibly adds the various attributes of the scheduling parameters into the initial selection mode of the centralized verification to obtain the initial verification value. This stage reflects the detailed analysis and comparison of the parameters to ensure that each point is consistent with the pre-standard of the mode, and further checks whether the parameters are correctly adapted to the attributes of the previous parameters. The analysis at this stage requires high accuracy and extreme control of the previous parameters to ensure the security of the verification value. If there is an accurate adaptation place, the consistency of the initial selection information and the attributes of the previous parameters is compared. The comparison process must retain sensitivity to even the smallest differences in the parameters to ensure that any inconsistencies can be accurately identified.

[0096] If they are all consistent, the initial selection model is used as the accurate benchmark model for scheduling parameter grouping. It is based on a high-baseline estimate of parameter consistency to ensure that the model can accurately reflect the actual status of the current parameters. If they are inconsistent, it is determined whether the parameter set of the scheduling parameters is reasonable. The determination process must focus on the type and source of the parameters and the synchronization with the current parameter scenario to ensure that all factors are fully considered.

[0097] In a preferred but non-limiting embodiment of the present invention, step 3 is:

[0098] Step 3-1: Get the current time point as the instant starting time point u;

[0099] Step 3-2: Determine whether the second scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter at the instant selection starting time point u is in a reasonable state. If so, send the scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter into the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection reserved information, and execute step 3-3. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing;

[0100] Step 3-3: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+1 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification immediate selection and retain information, and execute step 3-4. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing;

[0101] Step 3-4: determine whether the information reserved before the centralized verification of the dispatching parameters and the information reserved after the centralized verification of the dispatching parameters are the same. If so, send the information reserved before the centralized verification of the dispatching parameters as the dispatching parameters after the centralized verification of the dispatching parameters at time u. Otherwise, execute step 3-5;

[0102] Step 3-5: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+2 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection additional information, and execute step 3-6. Otherwise, exit the verification;

[0103] Step 3-6: Determine whether the additional information selected in the centralized verification of the scheduling parameters is the same as the information retained after the centralized verification of the scheduling parameters. If so, send the information retained before the centralized verification of the scheduling parameters as the scheduling parameters after the centralized verification of the scheduling parameters at time u. Otherwise, exit the verification and send the parameter attributes of the scheduling parameters at time points u, u+1 and u+2 respectively.

[0104] Here, the reasonable condition is that the parameter type of the scheduling parameter of the second scheduling parameter set verification parameter and the parameter source host ID of the scheduling parameter both correspond to the current parameter scenario.

[0105] In a preferred but non-limiting embodiment of the present invention, step 3-3 is:

[0106] Step 3-3-1: Determine whether the current centralized verification parameter 2 of the scheduling parameter and the centralized verification parameter 2 of the previous scheduling parameter have matching parameters. If so, the test is reasonable and the verification is terminated. Otherwise, obtain the centralized verification parameter 1 corresponding to the centralized verification parameter 2 of the scheduling parameter;

[0107] Step 3-3-2: Use the scheduling parameter centralized verification parameter one and the scheduling parameter centralized verification parameter two to form a refresh group to perform refresh processing.

[0108] In a preferred but non-limiting embodiment of the present invention, step 3-3-2 is:

[0109] Step 3-3-2-1: Use the scheduling parameter set to verify parameter one to form a refresh learning set;

[0110] Step 3-3-2-2: Use the scheduling parameter set to verify parameter 2 to form a refresh test set;

[0111] Step 3-3-2-3: Combine the refreshed learning set and learning set 1 to form current learning set 1, and execute step 2-3-1;

[0112] Step 3-3-2-4: Use the refresh test set to replace test set one, and execute step 2-3-2.

[0113] like Figure 2 As shown, a virtual power plant scheduling device according to the present invention includes:

[0114] An attribute module, which is used to obtain parameter attributes of the scheduling parameters using the scheduling parameters;

[0115] A selection module, which is used to construct a scheduling parameter group selection mode according to a basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters;

[0116] A verification module is used to perform regression selection according to the scheduling parameter grouping selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters. The scheduling parameters include energy market electricity prices, rotating reserve market electricity prices, gas turbine fuel costs, gas turbine fixed costs, gas turbine startup costs, gas turbine shutdown costs, load interruption compensation prices, gas turbine power generation costs, the upper limit of the capacity of the cold storage tank, the energy conversion efficiency in the cold storage tank during the release process, the gas turbine downward ramp rate, the gas turbine upward ramp rate, the minimum gas turbine startup time, the minimum gas turbine shutdown time, the minimum storage capacity of the energy storage power station, the maximum storage capacity of the energy storage power station, the charging efficiency, the discharge efficiency and the total interruption load, the photovoltaic power generation prediction power, and the maximum photovoltaic power generation prediction deviation.

[0117] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0118] The analysis of the basic mode and the scheduling parameters are combined to construct a safe analysis mode, and the expansion of the mode learning set is achieved by relying on the previous parameters, which greatly improves the information sent by the final mode. The test iteration process of the present invention also achieves internal iteration, and has good adaptability to long-term deployment of applications or large-scale scheduling parameters, effectively avoiding the defects of the prior art that parameter verification relies on principle models and manual verification, lacks efficiency and is often adversely affected by human factors.

[0119] The present disclosure can be a system, method and / or computer program product. The computer program product can include a computer-readable backup medium carrying computer-readable program instructions for causing a processor to implement various aspects of the disclosure.

[0120] The computer readable backup medium can be a tangible network capable of retaining and backing up the instructions used by the instruction execution network. The computer readable backup medium can be, but is not limited to, an electrical backup network, a magnetic backup network, an optical backup network, an electromagnetic backup network, a semiconductor backup network, or any suitable combination of the above. Further examples of computer readable backup media (non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (HD-ROM), digital versatile disk (DXD), memory stick, floppy disk, mechanically encoded network, such as punch cards or protrusions in grooves with instructions backed up thereon, and any suitable combination of the above. The computer readable backup medium used herein is not to be construed as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a power line cable), or an electrical signal transmitted through wires.

[0121] The computer readable program instructions expressed herein can be downloaded from the computer readable backup medium to each inference / processing power grid line, or downloaded to an external computer or external backup power grid line through a wireless network, such as the Internet, a local area network, a wide area network and / or a wireless network. The wireless network can include copper transmission cables, power transmission lines, wireless transmission, routers, firewalls, switches, gateway computers and / or edge service devices. The wireless network adapter card or wireless network interface in each inference / processing power grid line receives the computer readable program instructions from the wireless network and forwards the computer readable program instructions for storage in the computer readable backup medium in each inference / processing power grid line.

[0122] The computer program instructions for performing the operations of the present disclosure can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, conditional setting values, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Sdaletal A, H++, etc., and ordinary procedural programming languages ​​such as "H" language or similar programming languages. The computer-readable program instructions can be executed completely on the client computer, partially on the client computer, as a separate software package, partially on the client computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the client computer through any other wireless network, including a local area network (LAb) or a wide area network (WAb), or can be connected to an external computer (just like using an Internet service provider to connect through the Internet). In some embodiments, the electronic circuit is customized by using the status values ​​of computer-readable program instructions, such as a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA), and the electronic circuit can execute computer-readable program instructions to achieve various aspects of the overhead disclosure.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A virtual power plant scheduling method, characterized in that: include: Obtain dispatch parameters related to virtual power plant dispatch to perform verification, and send the verified dispatch parameters to the virtual power plant robust optimization dispatch model to perform virtual power plant dispatch analysis; Methods for performing verification include: Step 1: Use the scheduling parameter to obtain the parameter attributes of the scheduling parameter; Step 2: construct a scheduling parameter grouping selection mode according to the basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters; Step 3: Perform regression selection according to the scheduling parameter grouping selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters.

2. The virtual power plant scheduling method according to claim 1, characterized in that: Step 1 is: Step 1-1: Obtain the parameter type of the scheduling parameter, the parameter source host ID of the scheduling parameter, and the value of the scheduling parameter as an instant value; Step 1-2: using the instantaneous value to obtain parameter attributes of the scheduling parameter; Steps 1-2 are: Step 1-2-1: Determine whether the parameter type of the scheduling parameter meets the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-2. Otherwise, exit the verification. Step 1-2-2: Determine whether the parameter source host ID of the scheduling parameter satisfies the parameter scenario corresponding to the current scheduling parameter. If yes, execute step 1-2-3. Otherwise, exit verification. Step 1-2-3: using the value of the scheduling parameter as the first verification parameter in the scheduling parameter set, and using the parameter type of the scheduling parameter corresponding to the value of the scheduling parameter and the parameter source host ID of the scheduling parameter as the second verification parameter in the scheduling parameter set; Step 1-2-4: Use the scheduling parameter centralized verification parameter 1 and the scheduling parameter centralized verification parameter 2 as parameter attributes of the scheduling parameters.

3. The virtual power plant scheduling method according to claim 2, characterized in that: Step 2 is: Step 2-1: Obtain the previous parameter attributes corresponding to the values ​​of the scheduling parameters of the parameter attributes of the scheduling parameters to form a learning set 1; Step 2-2: Obtain the parameter type of the scheduling parameter corresponding to the learning set 1 and the parameter source host ID of the scheduling parameter to form a test set 1; Step 2-3: Use the learning set 1 and the test set 1 to construct a scheduling parameter centralized verification starting selection mode according to the basic mode; Step 2-4: using the parameter attributes of the scheduling parameters to obtain the scheduling parameter grouping selection mode according to the scheduling parameter centralized verification starting selection mode; Steps 2-3 are: Step 2-3-1: Use the learning set 1 as the input parameter, and the scheduling parameters of the learning set 1 corresponding to the scheduling parameters after centralized verification as the output parameters, and perform learning according to the basic mode to obtain the scheduling parameter centralized verification basic selection mode; Step 2-3-2: Send the test set 1 to the scheduling parameter centralized verification basic selection mode to obtain the scheduling parameter centralized verification basic selection information; Step 2-3-3: Determine whether the scheduling parameter centralized verification basic selection information and the scheduling parameters after the corresponding scheduling parameter centralized verification of the test set 1 are all consistent. If so, send the scheduling parameter centralized verification basic selection mode as the scheduling parameter centralized verification starting selection mode. Otherwise, use the subset of the scheduling parameter centralized verification basic selection information in the test set 1 that does not have the corresponding scheduling parameter centralized verification to add to the learning set 1, and then execute step 2-3-1; All consistent means that each subset in the test set has corresponding scheduling parameters to centrally verify the basic selection information.

4. The virtual power plant scheduling method according to claim 3, characterized in that: The basic model is support vector machine SVM.

5. The virtual power plant scheduling method according to claim 4, characterized in that: Steps 2-4 are: Step 2-4-1: sending the dispatch parameter centralized verification parameter 1 of the parameter attribute of the dispatch parameter to the dispatch parameter centralized verification starting selection mode to obtain the dispatch parameter centralized verification starting selection information; Step 2-4-2: Determine whether the scheduling parameter centralized verification parameter 1 has a corresponding past parameter attribute. If yes, execute step 2-4-3. Otherwise, send the scheduling parameter centralized verification starting selection mode as the scheduling parameter grouping selection mode. Step 2-4-3: Determine whether the dispatching parameter centralized verification start selection information is consistent with the dispatching parameter centralized verification start selection information corresponding to the previous parameter attribute. If yes, send the dispatching parameter centralized verification start selection mode as the dispatching parameter group selection mode. Otherwise, execute step 2-4-4; Step 2-4-4: Determine whether parameter 2 of the scheduling parameter set of the parameter attributes of the scheduling parameter is in a reasonable state. If so, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-4-1. Otherwise, obtain the parameter attributes of the scheduling parameter at the next subsequent time point as the parameter attributes of the current scheduling parameter, and execute step 2-1.

6. The virtual power plant scheduling method according to claim 5, characterized in that: The reasonable condition is that the parameter type of the scheduling parameter of the second scheduling parameter set verification parameter and the parameter source host ID of the scheduling parameter both correspond to the current parameter scenario.

7. The virtual power plant scheduling method according to claim 6, characterized in that: Step 3 is: Step 3-1: Get the current time point as the instant starting time point u; Step 3-2: Determine whether the second scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter at the instant selection starting time point u is in a reasonable state. If so, send the scheduling parameter centralized verification parameter of the parameter attribute of the scheduling parameter into the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection reserved information, and execute step 3-3. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing; Step 3-3: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+1 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification immediate selection and retain information, and execute step 3-4. Otherwise, use the scheduling parameter centralized verification parameter 2 to perform test refresh processing; Step 3-4: determine whether the information reserved before the centralized verification of the dispatching parameters and the information reserved after the centralized verification of the dispatching parameters are the same. If so, send the information reserved before the centralized verification of the dispatching parameters as the dispatching parameters after the centralized verification of the dispatching parameters at time u. Otherwise, execute step 3-5; Step 3-5: Determine whether the scheduling parameter centralized verification parameter 2 of the parameter attribute of the scheduling parameter at time u+2 is in a reasonable state. If so, send the scheduling parameter centralized verification parameter 1 of the parameter attribute of the scheduling parameter to the scheduling parameter group selection mode to obtain the scheduling parameter centralized verification instant selection additional information, and execute step 3-6. Otherwise, exit the verification; Step 3-6: Determine whether the additional information selected in the centralized verification of the scheduling parameters is the same as the information retained after the centralized verification of the scheduling parameters. If so, send the information retained before the centralized verification of the scheduling parameters as the scheduling parameters after the centralized verification of the scheduling parameters at time u. Otherwise, exit the verification and send the parameter attributes of the scheduling parameters at time points u, u+1 and u+2 respectively.

8. The virtual power plant scheduling method according to claim 7, characterized in that: Step 3-3 is: Step 3-3-1: Determine whether the current centralized verification parameter 2 of the scheduling parameter and the centralized verification parameter 2 of the previous scheduling parameter have matching parameters. If so, the test is reasonable and the verification is terminated. Otherwise, obtain the centralized verification parameter 1 corresponding to the centralized verification parameter 2 of the scheduling parameter; Step 3-3-2: Use the scheduling parameter centralized verification parameter one and the scheduling parameter centralized verification parameter two to form a refresh group to perform refresh processing.

9. The virtual power plant scheduling method according to claim 8, characterized in that: Step 3-3-2 is: Step 3-3-2-1: Use the scheduling parameter set to verify parameter one to form a refresh learning set; Step 3-3-2-2: Use the scheduling parameter set to verify parameter 2 to form a refresh test set; Step 3-3-2-3: Combine the refreshed learning set and learning set 1 to form current learning set 1, and execute step 2-3-1; Step 3-3-2-4: Use the refresh test set to replace test set one, and execute step 2-3-2.

10. A virtual power plant dispatching device, characterized in that: include: An attribute module, which is used to obtain parameter attributes of the scheduling parameters using the scheduling parameters; A selection module, which is used to construct a scheduling parameter group selection mode according to a basic mode by using the previous parameter attributes corresponding to the parameter attributes of the scheduling parameters; The verification module is used to perform regression selection according to the scheduling parameter group selection mode to obtain the scheduling parameters after centralized verification of the scheduling parameters.

Images