Remote control method for virtual power plant platform

Abstract

The invention relates to the technical field of virtual power plant remote control, and discloses a virtual power plant platform remote control method, which comprises the following steps: acquiring real-time key operation data of equipment based on a preset sensor, transmitting the real-time key operation data to a virtual power plant platform, and obtaining a real-time operation evaluation value; judging whether remote control operation is needed or not according to the real-time operation evaluation value, and if yes, determining an initial control operation strategy and generating an initial control operation instruction; determining operation influence factors and data change characteristics, constructing an influence factor prediction model according to the operation influence factors, and constructing a data prediction model according to the data change characteristics; and generating a predicted operation evaluation value according to the data prediction model, generating a predicted operable level according to the influence factor prediction model, and judging whether to issue an initial control operation instruction according to the predicted operation evaluation value and the predicted operable level, thereby improving the management and control efficiency of the virtual power plant and the system stability.

Description

Technical Field

[0001] The present application relates to the technical field of virtual power plant remote control, and in particular to a remote control method for a virtual power plant platform. Background Art

[0002] The virtual power plant platform has the ability to remotely control equipment in integrated photovoltaic, storage and charging projects. Through remote control of the virtual power plant platform, the operating parameters of the equipment can be adjusted to improve the reliability of power supply to the power grid.

[0003] In the existing technology, the remote control method of the virtual power plant platform generally evaluates the operating status of each device and selects the corresponding control strategy to optimize the control of the entire virtual power plant. However, it is difficult to ensure the accuracy of the overall operating status assessment and cannot accurately predict the control effect, thereby reducing the management and control efficiency of the virtual power plant and the system stability. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a remote control method for a virtual power plant platform, which evaluates real-time key operating data through the virtual power plant, obtains real-time operation evaluation values and determines whether to perform remote operation. If so, it generates initial control operation instructions, constructs an influencing factor prediction model and a data prediction model, generates predicted operation evaluation values and predicted operability levels respectively, and determines whether to issue initial control operation instructions, accurately predicts the regulation effect and user operation risks, and improves the management and control safety and efficiency of the virtual power plant.

[0005] In some embodiments of the present application, a remote control method for a virtual power plant platform is provided, comprising: Building a virtual power plant platform, collecting real-time key operating data of the equipment based on preset sensors and transmitting it to the virtual power plant platform, wherein the virtual power plant platform evaluates the real-time key operating data to obtain a real-time operating evaluation value; Determine whether remote control operation is required based on the real-time operation evaluation value, and if so, determine the initial control operation strategy based on the preset control operation reference library and generate the initial control operation instruction; Analyze historical control operation logs to determine operation impact factors and data change characteristics, build an impact factor prediction model based on operation impact factors, and build a data prediction model based on data change characteristics; According to the data prediction model, the predicted key operation data is obtained and the predicted operation evaluation value is generated. According to the influence factor prediction model, the predicted impact factor is obtained and the predicted operability level is generated. According to the predicted operation evaluation value and the predicted operability level, it is determined whether to issue the initial control operation instruction.

[0006] In some embodiments of the present application, the virtual power plant platform evaluates key operating data to obtain a real-time operating evaluation value, including: Pre-set a number of preset operation evaluation indicators, and pre-set the preset monitoring data for each preset operation evaluation indicator; Obtain historical monitoring logs stored in the virtual power plant platform, and set analysis nodes based on the historical monitoring duration and preset time intervals in each historical monitoring log; Generate a historical first operation sub-evaluation value of the corresponding preset operation evaluation indicator according to the historical preset monitoring data of each preset operation evaluation indicator at each analysis node, and construct a historical initial operation sub-evaluation value change curve of each preset operation evaluation indicator; According to each analysis node, other historical monitoring data in the corresponding historical monitoring log is collected, and a historical monitoring data change curve is constructed based on the other historical monitoring data at multiple analysis nodes, and mapped to the historical first operation sub-evaluation indicator change curve corresponding to each preset operation evaluation indicator; Perform curve correlation analysis on each historical monitoring data change curve and the historical first operation sub-evaluation indicator change curve to obtain the curve correlation degree; The historical monitoring data corresponding to the historical monitoring data change curve whose curve correlation degree is greater than the preset correlation degree threshold and the preset monitoring data are set as the historical key operation data corresponding to the preset operation evaluation index, and the weight coefficient of the corresponding historical key operation data is set according to the curve correlation degree; Generate a historical second operation sub-evaluation value of the corresponding preset evaluation indicator based on the historical key operation data of the same preset operation evaluation indicator at the same analysis node and the corresponding weight coefficient; Allocate the historical second operation sub-evaluation value at each analysis node according to the weight ratio of the weight coefficient, and obtain the historical third operation sub-evaluation value allocated to each historical key operation data at the corresponding analysis node; Constructing an indicator-data-evaluation value mapping table based on multiple historical key operating data of the same preset operating evaluation indicator and the corresponding historical third operating sub-evaluation value; Analyze the real-time key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the real-time third operation sub-evaluation value of the real-time key operation data; Generate a real-time second operation sub-evaluation value corresponding to the preset operation evaluation indicator according to the real-time third operation sub-evaluation value of all real-time key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A real-time operation evaluation value is generated according to the real-time second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

[0007] In some embodiments of the present application, determining whether a remote control operation is required based on the real-time operation evaluation value includes: Presetting an operation evaluation value threshold and a second operation sub-evaluation value threshold for each key operation data; If the real-time operation evaluation value is less than the operation evaluation value threshold, remote control operation is required; If the real-time operation evaluation value is greater than the operation evaluation value threshold, no remote control operation is required.

[0008] In some embodiments of the present application, determining an initial control operation strategy and generating an initial control operation instruction based on a preset control operation reference library includes: When a remote control operation is required, the second operation sub-evaluation value threshold of each real-time key operation data is compared with the corresponding second operation sub-evaluation value threshold, the data to be regulated is screened out according to the comparison result, and the required regulation value of the data to be regulated is calculated; Setting the priority coefficient of the data to be regulated according to the weight coefficient corresponding to the data to be regulated and the preset operation evaluation index to which it belongs; Sort all the data that need to be regulated according to their priority coefficients to obtain a sequence of data that need to be regulated; Determine a corresponding preset control operation reference library based on the preset operation evaluation index corresponding to each data to be regulated in the sequence of data to be regulated, wherein the preset control operation reference library includes a plurality of preset control values of key operation data corresponding to the preset operation evaluation index, and each preset control value is associated with a corresponding preset control operation strategy; Determine the control operation sub-strategy corresponding to each data to be regulated in the data sequence to be regulated based on the preset control operation reference library, and construct a control operation sub-strategy sequence; Perform conflict analysis on the control operation sub-strategy sequence. If there is a conflict, optimize the conflicting control operation sub-strategy based on the priority principle and the principle of maximizing operating efficiency, and generate the initial control operation strategy based on the optimized control operation sub-strategy sequence; Generate initial control operation instructions according to the initial control operation strategy and the operation user level.

[0009] In some embodiments of the present application, historical control operation logs are parsed to determine operation impact factors and data change characteristics, including: Parsing the historical control operation log, determining a number of historical control operation sub-strategies, obtaining corresponding historical control operation nodes, and setting the left adjacent preset time period and the right adjacent preset time period of the historical control operation node of each historical control operation sub-strategy as a first historical focus time period and a second historical focus time period, respectively; Obtaining a mean of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the first historical focus period and a maximum of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the second historical focus period for each historical control operation sub-strategy; Calculate the absolute value of the historical difference between the mean of the historical second sub-evaluation value and the maximum value of the historical second sub-evaluation value for the same preset operation evaluation indicator; The preset operation evaluation index whose absolute value of the historical difference is greater than the preset absolute value of the difference is set as the pending operation impact factor of the corresponding historical control operation sub-strategy; Compare the pending operation impact factors of the same historical control operation sub-strategy in different historical control operation logs to obtain multiple impact degrees and impact probabilities of each pending operation impact factor; Generate an influence coefficient based on multiple influence degrees and influence probabilities of the same pending operation influence factor; The calculation formula of the influence coefficient is: ; Where Y is the influence coefficient, a1 is the influence conversion coefficient, w0 / w is the operation probability, w is the total number of the same historical control operation sub-strategy in different historical control operation logs, w0 is the number of times the corresponding pending operation influence factor appears in the pending operation influence factors of the same historical control operation sub-strategy in different historical control operation logs, Zi is the absolute value of the historical difference between the pending operation influence factor and the corresponding historical control operation sub-strategy in the i-th historical control operation log, and Z0 is the preset difference absolute value threshold; Setting the pending operation impact factor whose impact coefficient is greater than the preset impact coefficient threshold as the operation impact factor of the corresponding historical control operation sub-strategy; The historical data change characteristics of each key operating data in the second historical focus period of each historical control operation sub-strategy are obtained, where the historical data change characteristics include a historical change trend, a historical change rate, and a historical change value.

[0010] In some embodiments of the present application, an impact factor prediction model is constructed based on the operation impact factor, and a data prediction model is constructed based on the data change characteristics, including: Each historical control operation sub-strategy is used as training input data, and the operation influence factor and the corresponding influence coefficient of each historical control operation sub-strategy are used as training output data to perform neural network training to obtain an influence factor prediction model; Comparing several historical data change features of the same key operation data of the same historical control operation sub-strategy in different historical control operation logs to obtain similarities of the several historical data change features of the same key operation data, wherein the similarity includes several sub-similarity; If the similarity is greater than a preset similarity threshold, the historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; If the similarity is less than the preset similarity threshold, the historical data change characteristics of the corresponding key operation data are fine-tuned according to the sub-similarity until the similarity is greater than the preset similarity threshold, and the fine-tuned historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; Combine each historical control operation sub-strategy as training input data, perform neural network training, and obtain a data prediction model.

[0011] In some embodiments of the present application, obtaining predicted key operation data and generating a predicted operation evaluation value according to a data prediction model includes: Inputting several control operation sub-strategies in the initial control operation instruction into the data prediction model, obtaining the predicted data change characteristics of each key operation data, and combining them with the current real-time key operation data to obtain the predicted key operation data; Analyze the predicted key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the predicted third operation sub-evaluation value of the predicted key operation data; Generate a predicted second operation sub-evaluation value corresponding to the preset operation evaluation indicator based on the predicted third operation sub-evaluation values of all predicted key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A predicted operation evaluation value is generated according to the predicted second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

[0012] In some embodiments of the present application, obtaining a predicted impact factor and generating a predicted actionable level according to an impact factor prediction model includes: Inputting several control operation sub-strategies in the initial control operation instruction into the impact factor prediction model to obtain the predicted impact factor of each control operation sub-strategy; Obtaining an initial operational level for each predicted impact factor based on an operational impact factor-operational level mapping table; A predicted actionable level is generated according to the influence coefficient of each predicted influencing factor and the initial actionable level.

[0013] In some embodiments of the present application, determining whether to issue an initial control operation instruction based on the predicted operation evaluation value and the predicted operability level includes: The initial control operation instruction includes an initial control operation strategy and an operation user level; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is issued; If the predicted operation evaluation value is less than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is not issued and a control operation adjustment instruction is generated; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is less than the predicted operable level, the initial control operation instruction will not be issued and a user operation risk warning instruction will be generated.

[0014] Compared with the prior art, the remote control method for a virtual power plant platform in the embodiment of the present application has the following advantages: Through the virtual power plant, real-time key operating data is evaluated to obtain real-time operating evaluation values and determine whether remote operation is to be performed. If so, initial control operation instructions are generated, and an influencing factor prediction model and a data prediction model are constructed to generate predicted operating evaluation values and predicted operability levels respectively. It is also determined whether to issue initial control operation instructions, accurately predict the regulation effect and user operation risks, and improve the management and control safety and efficiency of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a remote control method for a virtual power plant platform in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0017] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0020] like Figure 1 As shown, a remote control method for a virtual power plant platform in an embodiment of the present application includes: Step S101: constructing a virtual power plant platform, collecting real-time key operating data of the equipment based on preset sensors and transmitting the data to the virtual power plant platform, wherein the virtual power plant platform evaluates the real-time key operating data to obtain a real-time operating evaluation value; Step S102: determining whether a remote control operation is required based on the real-time operation evaluation value; if so, determining an initial control operation strategy based on a preset control operation reference library and generating an initial control operation instruction; Step S103: parsing historical control operation logs to determine operation impact factors and data change characteristics, building an impact factor prediction model based on the operation impact factors, and building a data prediction model based on the data change characteristics; Step S104: Determine key operational data based on the data prediction model and generate a predicted operational evaluation value. Determine the predicted impact factor based on the impact factor prediction model and generate a predicted operability level. Determine whether to issue initial control operation instructions based on the predicted operational evaluation value and the predicted operability level. In this embodiment, the virtual power plant platform has the ability to remotely control equipment in the integrated photovoltaic, energy storage, and charging project. Through the virtual power plant platform's remote control interface, users can remotely start or stop equipment such as photovoltaic inverters, energy storage system charging and discharging equipment, and charging piles. They can also adjust key operating parameters of the equipment (such as photovoltaic power generation setpoints, energy storage charging and discharging power limits, and charging pile output power). During remote control operations, it is necessary to ensure that only authorized users can perform operations to ensure operational safety and reliability.

[0021] In this embodiment, users include three levels: administrators, operation and maintenance personnel, and ordinary users. Different levels have different operating permissions. For example, administrators can modify all key operating parameters, operation and maintenance personnel can only adjust some non-key parameters, and ordinary users can only view the device status.

[0022] In some embodiments of the present application, the virtual power plant platform evaluates key operating data to obtain a real-time operating evaluation value, including: Pre-set a number of preset operation evaluation indicators, and pre-set the preset monitoring data for each preset operation evaluation indicator; Obtain historical monitoring logs stored in the virtual power plant platform, and set analysis nodes based on the historical monitoring duration and preset time intervals in each historical monitoring log; Generate a historical first operation sub-evaluation value of the corresponding preset operation evaluation indicator according to the historical preset monitoring data of each preset operation evaluation indicator at each analysis node, and construct a historical initial operation sub-evaluation value change curve of each preset operation evaluation indicator; According to each analysis node, other historical monitoring data in the corresponding historical monitoring log is collected, and a historical monitoring data change curve is constructed based on the other historical monitoring data at multiple analysis nodes, and mapped to the historical first operation sub-evaluation indicator change curve corresponding to each preset operation evaluation indicator; Perform curve correlation analysis on each historical monitoring data change curve and the historical first operation sub-evaluation indicator change curve to obtain the curve correlation degree; The historical monitoring data corresponding to the historical monitoring data change curve whose curve correlation degree is greater than the preset correlation degree threshold and the preset monitoring data are set as the historical key operation data corresponding to the preset operation evaluation index, and the weight coefficient of the corresponding historical key operation data is set according to the curve correlation degree; Generate a historical second operation sub-evaluation value of the corresponding preset evaluation indicator based on the historical key operation data of the same preset operation evaluation indicator at the same analysis node and the corresponding weight coefficient; Allocate the historical second operation sub-evaluation value at each analysis node according to the weight ratio of the weight coefficient, and obtain the historical third operation sub-evaluation value allocated to each historical key operation data at the corresponding analysis node; Constructing an indicator-data-evaluation value mapping table based on multiple historical key operating data of the same preset operating evaluation indicator and the corresponding historical third operating sub-evaluation value; Analyze the real-time key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the real-time third operation sub-evaluation value of the real-time key operation data; Generate a real-time second operation sub-evaluation value corresponding to the preset operation evaluation indicator according to the real-time third operation sub-evaluation value of all real-time key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A real-time operation evaluation value is generated according to the real-time second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

[0023] In this embodiment, the preset operation evaluation indicators include the equipment status, power generation power, power generation, inverter efficiency, etc. related to photovoltaic equipment, the charging and discharging power, battery health status, charging and discharging power, etc. related to energy storage equipment, and the charging pile operation status, charging power, charging time, etc. related to charging equipment.

[0024] In this embodiment, the preset monitoring data of each preset operation evaluation indicator is set in advance, and the historical first operation evaluation value is calculated based on the preset monitoring data corresponding to the preset operation evaluation indicator.

[0025] In this embodiment, the curve correlation degree refers to the curve correlation index between the historical monitoring data change curve and the historical first operation sub-evaluation indicator change curve, that is, the consistency of the change of the historical first operation sub-evaluation indicator change curve with the curve periodicity, trend, mutation and other characteristics of the historical monitoring data change curve. The greater the change consistency, the greater the corresponding curve correlation degree, and vice versa.

[0026] In this embodiment, the real-time third operation sub-evaluation value is obtained by dividing the real-time second operation sub-evaluation value according to the weight coefficient of each real-time key operation data and the corresponding preset operation evaluation index, and the real-time third operation sub-evaluation value is used to evaluate the operation status of each real-time key operation data. The larger the real-time third operation sub-evaluation value is, the better the operation status of the corresponding real-time key operation data, and vice versa.

[0027] In this embodiment, by calculating the key operating data of each preset operating evaluation indicator, the subsequent data analysis and processing volume is reduced, the evaluation efficiency and accuracy of each preset operating evaluation indicator are improved, and a comprehensive and accurate evaluation of the overall operating status of the virtual power plant is achieved, laying the foundation for subsequent remote control and improving the timeliness and accuracy of remote control.

[0028] In some embodiments of the present application, determining whether a remote control operation is required based on the real-time operation evaluation value includes: Presetting an operation evaluation value threshold and a second operation sub-evaluation value threshold for each key operation data; If the real-time operation evaluation value is less than the operation evaluation value threshold, remote control operation is required; If the real-time operation evaluation value is greater than the operation evaluation value threshold, no remote control operation is required.

[0029] In some embodiments of the present application, determining an initial control operation strategy and generating an initial control operation instruction based on a preset control operation reference library includes: When a remote control operation is required, the second operation sub-evaluation value threshold of each real-time key operation data is compared with the corresponding second operation sub-evaluation value threshold, the data to be regulated is screened out according to the comparison result, and the required regulation value of the data to be regulated is calculated; Setting the priority coefficient of the data to be regulated according to the weight coefficient corresponding to the data to be regulated and the preset operation evaluation index to which it belongs; Sort all the data that need to be regulated according to their priority coefficients to obtain a sequence of data that need to be regulated; Determine a corresponding preset control operation reference library based on the preset operation evaluation index corresponding to each data to be regulated in the sequence of data to be regulated, wherein the preset control operation reference library includes a plurality of preset control values of key operation data corresponding to the preset operation evaluation index, and each preset control value is associated with a corresponding preset control operation strategy; Determine the control operation sub-strategy corresponding to each data to be regulated in the data sequence to be regulated based on the preset control operation reference library, and construct a control operation sub-strategy sequence; Perform conflict analysis on the control operation sub-strategy sequence. If there is a conflict, optimize the conflicting control operation sub-strategy based on the priority principle and the principle of maximizing operating efficiency, and generate the initial control operation strategy based on the optimized control operation sub-strategy sequence; Generate initial control operation instructions according to the initial control operation strategy and the operation user level.

[0030] In this embodiment, the second operation sub-evaluation value threshold is the minimum second operation sub-evaluation value threshold under the normal operation state of each key operation data. When the second operation sub-evaluation value threshold is greater than the second operation sub-evaluation value threshold, the corresponding key operation data is set as the data to be regulated, and the preset value corresponding to the key operation data at the second operation sub-evaluation value threshold is compared with the current real-time key operation data to obtain the value to be regulated.

[0031] In this embodiment, the priority coefficient is calculated based on the weight coefficient corresponding to the data to be regulated and the weight coefficient of the preset operation evaluation index to which it belongs. When the weight coefficients are larger, the corresponding priority coefficient is larger.

[0032] In this embodiment, the preset control operation reference library of the preset operation evaluation index is constructed based on the historical control value of each key operation data and the corresponding historical control operation sub-strategy.

[0033] In this embodiment, the priority principle means that the control operation sub-strategy corresponding to the data requiring regulation with the smaller priority coefficient is optimized or corrected first, and the operating efficiency maximization principle means that the overall operating evaluation value of the virtual power plant is maximized.

[0034] In this embodiment, by determining the initial control operation instructions, formulating several control operation sub-strategies for the current data that needs to be regulated, accurately evaluating the overall operating status and timely formulating corresponding control strategies, the management and control efficiency and accuracy of the virtual power plant are improved.

[0035] In some embodiments of the present application, historical control operation logs are parsed to determine operation impact factors and data change characteristics, including: Parsing the historical control operation log, determining a number of historical control operation sub-strategies, obtaining corresponding historical control operation nodes, and setting the left adjacent preset time period and the right adjacent preset time period of the historical control operation node of each historical control operation sub-strategy as a first historical focus time period and a second historical focus time period, respectively; Obtaining a mean of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the first historical focus period and a maximum of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the second historical focus period for each historical control operation sub-strategy; Calculate the absolute value of the historical difference between the mean of the historical second sub-evaluation value and the maximum value of the historical second sub-evaluation value for the same preset operation evaluation indicator; The preset operation evaluation index whose absolute value of the historical difference is greater than the preset absolute value of the difference is set as the pending operation impact factor of the corresponding historical control operation sub-strategy; Compare the pending operation impact factors of the same historical control operation sub-strategy in different historical control operation logs to obtain multiple impact degrees and impact probabilities of each pending operation impact factor; Generate an influence coefficient based on multiple influence degrees and influence probabilities of the same pending operation influence factor; The calculation formula of the influence coefficient is: ; Where Y is the influence coefficient, a1 is the influence conversion coefficient, w0 / w is the operation probability, w is the total number of the same historical control operation sub-strategy in different historical control operation logs, w0 is the number of times the corresponding pending operation influence factor appears in the pending operation influence factors of the same historical control operation sub-strategy in different historical control operation logs, Zi is the absolute value of the historical difference between the pending operation influence factor and the corresponding historical control operation sub-strategy in the i-th historical control operation log, and Z0 is the preset difference absolute value threshold; Setting the pending operation impact factor whose impact coefficient is greater than the preset impact coefficient threshold as the operation impact factor of the corresponding historical control operation sub-strategy; The historical data change characteristics of each key operating data in the second historical focus period of each historical control operation sub-strategy are obtained, where the historical data change characteristics include a historical change trend, a historical change rate, and a historical change value.

[0036] In this embodiment, the first historical attention period and the second historical attention period refer to the front preset period and the back preset period at the historical control operation node of the control operation sub-strategy, respectively, and the preset period is 20 minutes.

[0037] In this embodiment, the mean value of the historical second operation sub-evaluation value refers to the value obtained by averaging multiple historical second operation sub-evaluation values in the first historical focus period, and the maximum value of the historical second operation sub-evaluation value refers to the maximum historical second operation sub-evaluation value or the minimum historical second historical operation sub-evaluation value in the second historical focus period. If the historical second operation sub-evaluation value in the second historical focus period is a downward trend, the maximum value of the historical second operation sub-evaluation value is the minimum historical second historical operation sub-evaluation value; if it is an upward trend, the maximum value of the historical second operation sub-evaluation value is the maximum historical second historical operation sub-evaluation value.

[0038] In this embodiment, the impact conversion coefficient refers to a coefficient that converts a numerical value whose absolute value of the historical difference is greater than the absolute value of the preset difference into an impact degree. When the numerical value is larger, the corresponding impact degree is greater. When the impact degree is greater and the impact probability is greater, the corresponding impact coefficient is greater.

[0039] In this embodiment, by calculating the influence coefficient, the degree of change of each preset operation evaluation indicator under the influence of the second operation sub-evaluation value of the corresponding historical control operation sub-strategy is accurately evaluated, and the corresponding operation influence factor is set. By determining the data change characteristics, the change characteristics of each key operation data after being controlled are accurately predicted, laying the foundation for subsequent judgment of whether the initial control operation strategy in the initial control operation instruction and the operation user level meet the control requirements and whether they have the corresponding operation authority, thereby ensuring the safety and reliability of the operation.

[0040] In some embodiments of the present application, an impact factor prediction model is constructed based on the operation impact factor, and a data prediction model is constructed based on the data change characteristics, including: Each historical control operation sub-strategy is used as training input data, and the operation influence factor and the corresponding influence coefficient of each historical control operation sub-strategy are used as training output data to perform neural network training to obtain an influence factor prediction model; Comparing several historical data change features of the same key operation data of the same historical control operation sub-strategy in different historical control operation logs to obtain similarities of the several historical data change features of the same key operation data, wherein the similarity includes several sub-similarity; If the similarity is greater than a preset similarity threshold, the historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; If the similarity is less than the preset similarity threshold, the historical data change characteristics of the corresponding key operation data are fine-tuned according to the sub-similarity until the similarity is greater than the preset similarity threshold, and the fine-tuned historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; Combine each historical control operation sub-strategy as training input data, perform neural network training, and obtain a data prediction model.

[0041] In this embodiment, the similarity is calculated based on the sub-similarity of the historical change trend, historical change rate and historical change value in different historical data change characteristics. Fine-tuning the historical data change characteristics refers to determining the change trend, change rate and change value with the greatest credibility based on the sub-similarity and combining them to obtain a new data change characteristic.

[0042] In this embodiment, the initial control operation instructions are predicted by the data prediction model and the operation factor prediction model to obtain the predicted operation evaluation value and the predicted operability level after control, thereby predicting the control effect and operation safety of the initial control operation instructions.

[0043] In some embodiments of the present application, obtaining predicted key operation data and generating a predicted operation evaluation value according to a data prediction model includes: Inputting several control operation sub-strategies in the initial control operation instruction into the data prediction model, obtaining the predicted data change characteristics of each key operation data, and combining them with the current real-time key operation data to obtain the predicted key operation data; Analyze the predicted key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the predicted third operation sub-evaluation value of the predicted key operation data; Generate a predicted second operation sub-evaluation value corresponding to the preset operation evaluation indicator based on the predicted third operation sub-evaluation values of all predicted key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A predicted operation evaluation value is generated according to the predicted second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

[0044] In some embodiments of the present application, obtaining a predicted impact factor and generating a predicted actionable level according to an impact factor prediction model includes: Inputting several control operation sub-strategies in the initial control operation instruction into the impact factor prediction model to obtain the predicted impact factor of each control operation sub-strategy; Obtaining an initial operational level for each predicted impact factor based on an operational impact factor-operational level mapping table; A predicted actionable level is generated according to the influence coefficient of each predicted influencing factor and the initial actionable level.

[0045] In this embodiment, the operation impact factor-operability level means that each operation impact factor is mapped to a corresponding operability level, that is, the initial operability level. The predicted operability level is generated based on the initial operability level and the impact coefficient to improve the accuracy of the predicted operability level.

[0046] In some embodiments of the present application, determining whether to issue an initial control operation instruction based on the predicted operation evaluation value and the predicted operability level includes: The initial control operation instruction includes an initial control operation strategy and an operation user level; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is issued; If the predicted operation evaluation value is less than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is not issued and a control operation adjustment instruction is generated; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is less than the predicted operable level, the initial control operation instruction will not be issued and a user operation risk warning instruction will be generated.

[0047] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and replacements can be made without departing from the technical principles of the present application. These improvements and replacements should also be regarded as the scope of protection of the present application.

Claims

1. A remote control method for a virtual power plant platform, characterized in that: include: Building a virtual power plant platform, collecting real-time key operating data of the equipment based on preset sensors and transmitting it to the virtual power plant platform, wherein the virtual power plant platform evaluates the real-time key operating data to obtain a real-time operating evaluation value; Determine whether remote control operation is required based on the real-time operation evaluation value, and if so, determine the initial control operation strategy based on the preset control operation reference library and generate the initial control operation instruction; Analyze historical control operation logs to determine operation impact factors and data change characteristics, build an impact factor prediction model based on operation impact factors, and build a data prediction model based on data change characteristics; According to the data prediction model, the predicted key operation data is obtained and the predicted operation evaluation value is generated. According to the influence factor prediction model, the predicted impact factor is obtained and the predicted operability level is generated. According to the predicted operation evaluation value and the predicted operability level, it is determined whether to issue the initial control operation instruction.

2. The remote control method for a virtual power plant platform according to claim 1, characterized in that: The virtual power plant platform evaluates key operating data to obtain real-time operating evaluation values, including: Pre-set a number of preset operation evaluation indicators, and pre-set the preset monitoring data for each preset operation evaluation indicator; Obtain historical monitoring logs stored in the virtual power plant platform, and set analysis nodes based on the historical monitoring duration and preset time intervals in each historical monitoring log; Generate a historical first operation sub-evaluation value of the corresponding preset operation evaluation indicator according to the historical preset monitoring data of each preset operation evaluation indicator at each analysis node, and construct a historical initial operation sub-evaluation value change curve of each preset operation evaluation indicator; According to each analysis node, other historical monitoring data in the corresponding historical monitoring log is collected, and a historical monitoring data change curve is constructed based on the other historical monitoring data at multiple analysis nodes, and mapped to the historical first operation sub-evaluation indicator change curve corresponding to each preset operation evaluation indicator; Perform curve correlation analysis on each historical monitoring data change curve and the historical first operation sub-evaluation indicator change curve to obtain the curve correlation degree; The historical monitoring data corresponding to the historical monitoring data change curve whose curve correlation degree is greater than the preset correlation degree threshold and the preset monitoring data are set as the historical key operation data corresponding to the preset operation evaluation index, and the weight coefficient of the corresponding historical key operation data is set according to the curve correlation degree; Generate a historical second operation sub-evaluation value of the corresponding preset evaluation indicator based on the historical key operation data of the same preset operation evaluation indicator at the same analysis node and the corresponding weight coefficient; Allocate the historical second operation sub-evaluation value at each analysis node according to the weight ratio of the weight coefficient, and obtain the historical third operation sub-evaluation value allocated to each historical key operation data at the corresponding analysis node; Constructing an indicator-data-evaluation value mapping table based on multiple historical key operating data of the same preset operating evaluation indicator and the corresponding historical third operating sub-evaluation value; Analyze the real-time key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the real-time third operation sub-evaluation value of the real-time key operation data; Generate a real-time second operation sub-evaluation value corresponding to the preset operation evaluation indicator according to the real-time third operation sub-evaluation value of all real-time key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A real-time operation evaluation value is generated according to the real-time second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

3. The remote control method for a virtual power plant platform according to claim 2, characterized in that: Determine whether remote control operations are required based on real-time operation evaluation values, including: Presetting an operation evaluation value threshold and a second operation sub-evaluation value threshold for each key operation data; If the real-time operation evaluation value is less than the operation evaluation value threshold, remote control operation is required; If the real-time operation evaluation value is greater than the operation evaluation value threshold, no remote control operation is required.

4. The remote control method for a virtual power plant platform according to claim 3, characterized in that: Determine the initial control operation strategy and generate initial control operation instructions based on the preset control operation reference library, including: When a remote control operation is required, the second operation sub-evaluation value threshold of each real-time key operation data is compared with the corresponding second operation sub-evaluation value threshold, the data to be regulated is screened out according to the comparison result, and the required regulation value of the data to be regulated is calculated; Setting the priority coefficient of the data to be regulated according to the weight coefficient corresponding to the data to be regulated and the preset operation evaluation index to which it belongs; Sort all the data that need to be regulated according to their priority coefficients to obtain a sequence of data that need to be regulated; Determine a corresponding preset control operation reference library based on the preset operation evaluation index corresponding to each data to be regulated in the sequence of data to be regulated, wherein the preset control operation reference library includes a plurality of preset control values of key operation data corresponding to the preset operation evaluation index, and each preset control value is associated with a corresponding preset control operation strategy; Determine the control operation sub-strategy corresponding to each data to be regulated in the data sequence to be regulated based on the preset control operation reference library, and construct a control operation sub-strategy sequence; Perform conflict analysis on the control operation sub-strategy sequence. If there is a conflict, optimize the conflicting control operation sub-strategy based on the priority principle and the principle of maximizing operating efficiency, and generate the initial control operation strategy based on the optimized control operation sub-strategy sequence; Generate initial control operation instructions according to the initial control operation strategy and the operation user level.

5. The remote control method for a virtual power plant platform according to claim 4, characterized in that: Analyze historical control operation logs to determine operation influencing factors and data change characteristics, including: Parsing the historical control operation log, determining a number of historical control operation sub-strategies, obtaining corresponding historical control operation nodes, and setting the left adjacent preset time period and the right adjacent preset time period of the historical control operation node of each historical control operation sub-strategy as a first historical focus time period and a second historical focus time period, respectively; Obtaining a mean of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the first historical focus period and a maximum of the historical second operation sub-evaluation values of each preset operation evaluation indicator in the second historical focus period for each historical control operation sub-strategy; Calculate the absolute value of the historical difference between the mean of the historical second sub-evaluation value and the maximum value of the historical second sub-evaluation value for the same preset operation evaluation indicator; The preset operation evaluation index whose absolute value of the historical difference is greater than the preset absolute value of the difference is set as the pending operation impact factor of the corresponding historical control operation sub-strategy; Compare the pending operation impact factors of the same historical control operation sub-strategy in different historical control operation logs to obtain multiple impact degrees and impact probabilities of each pending operation impact factor; Generate an influence coefficient based on multiple influence degrees and influence probabilities of the same pending operation influence factor; The calculation formula of the influence coefficient is: ； Where Y is the influence coefficient, a1 is the influence conversion coefficient, w0 / w is the operation probability, w is the total number of the same historical control operation sub-strategy in different historical control operation logs, w0 is the number of times the corresponding pending operation influence factor appears in the pending operation influence factors of the same historical control operation sub-strategy in different historical control operation logs, Zi is the absolute value of the historical difference between the pending operation influence factor and the corresponding historical control operation sub-strategy in the i-th historical control operation log, and Z0 is the preset difference absolute value threshold; Setting the pending operation impact factor whose impact coefficient is greater than the preset impact coefficient threshold as the operation impact factor of the corresponding historical control operation sub-strategy; The historical data change characteristics of each key operating data in the second historical focus period of each historical control operation sub-strategy are obtained, where the historical data change characteristics include a historical change trend, a historical change rate, and a historical change value.

6. The remote control method for a virtual power plant platform according to claim 5, characterized in that: Build an impact factor prediction model based on operational impact factors, and build a data prediction model based on data change characteristics, including: Each historical control operation sub-strategy is used as training input data, and the operation influence factor and the corresponding influence coefficient of each historical control operation sub-strategy are used as training output data to perform neural network training to obtain an influence factor prediction model; Comparing several historical data change features of the same key operation data of the same historical control operation sub-strategy in different historical control operation logs to obtain similarities of the several historical data change features of the same key operation data, wherein the similarity includes several sub-similarity; If the similarity is greater than a preset similarity threshold, the historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; If the similarity is less than the preset similarity threshold, the historical data change characteristics of the corresponding key operation data are fine-tuned according to the sub-similarity until the similarity is greater than the preset similarity threshold, and the fine-tuned historical data change characteristics of the corresponding key operation data are set as the training output data of the corresponding historical control operation sub-strategy; Combine each historical control operation sub-strategy as training input data, perform neural network training, and obtain a data prediction model.

7. The remote control method for a virtual power plant platform according to claim 6, characterized in that: According to the data prediction model, key operation data is predicted and the prediction operation evaluation value is generated, including: Inputting several control operation sub-strategies in the initial control operation instruction into the data prediction model, obtaining the predicted data change characteristics of each key operation data, and combining them with the current real-time key operation data to obtain the predicted key operation data; Analyze the predicted key operation data and the corresponding preset operation evaluation indicators based on the indicator-data-evaluation value mapping table to obtain the predicted third operation sub-evaluation value of the predicted key operation data; Generate a predicted second operation sub-evaluation value corresponding to the preset operation evaluation indicator based on the predicted third operation sub-evaluation values of all predicted key operation data of the same preset operation evaluation indicator and the corresponding weight coefficient; A predicted operation evaluation value is generated according to the predicted second operation sub-evaluation values of a plurality of preset operation evaluation indicators and weight coefficients corresponding to the preset operation evaluation indicators.

8. The remote control method for a virtual power plant platform according to claim 7, characterized in that: The impact factor prediction model is used to obtain the predicted impact factor and generate the predicted actionable level, including: Inputting several control operation sub-strategies in the initial control operation instruction into the impact factor prediction model to obtain the predicted impact factor of each control operation sub-strategy; Obtaining an initial operational level for each predicted impact factor based on an operational impact factor-operational level mapping table; A predicted actionable level is generated according to the influence coefficient of each predicted influencing factor and the initial actionable level.

9. The remote control method for a virtual power plant platform according to claim 8, characterized in that: Determine whether to issue initial control operation instructions based on the predicted operation evaluation value and the predicted operability level, including: The initial control operation instruction includes an initial control operation strategy and an operation user level; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is issued; If the predicted operation evaluation value is less than the operation evaluation value threshold and the operating user level is greater than the predicted operable level, the initial control operation instruction is not issued and a control operation adjustment instruction is generated; If the predicted operation evaluation value is greater than the operation evaluation value threshold and the operating user level is less than the predicted operable level, the initial control operation instruction will not be issued and a user operation risk warning instruction will be generated.

Images