New energy power station concentrated power prediction management method, system, equipment and medium

Abstract

The invention discloses a new energy power station concentrated power prediction management method, system, device and medium, and belongs to the technical field of new energy power station concentrated power prediction, and the method comprises the steps: receiving prediction data packets uploaded by a plurality of stations through a forward isolation device; based on the historical prediction precision of each prediction system, calculating the adaptation degree score of each prediction system for each station, and selecting corresponding prediction power data from the plurality of prediction systems of each station as to-be-corrected data according to the adaptation degree score; responding to a correction operation of the user on the to-be-corrected data; and issuing the corrected predicted power data to a gateway machine of each station through a reverse isolation device. According to the method, personalized matching of the prediction systems and the stations is realized, and the stations with different characteristics can automatically select the most suitable prediction systems by calculating the adaptation degree scores of the prediction systems for the stations and establishing the mapping relation table, so that the overall prediction accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of centralized power prediction of new energy power stations, in particular to a centralized power prediction management method, system, device and medium for new energy power stations. BACKGROUND

[0002] With the rapid growth of new energy installed capacity, power prediction has become an important basis for power grid dispatching. In the prior art, each station usually deploys a single or multiple prediction systems, but lacks an effective optimization mechanism, and cannot be individually matched according to the actual performance of each prediction system in different stations, resulting in long-term low prediction accuracy in some stations. In terms of power prediction correction, pure automatic correction relies on the flexibility of preset rules, and pure manual correction is inefficient in point-by-point adjustment, which is difficult to meet the timeliness requirements of simultaneous processing of multiple stations in a group management scenario. In addition, the prior art lacks a prediction accuracy feedback mechanism, cannot form an optimization closed loop, and the technical solution of data cross-security zone transmission under the centralized management architecture is not clear, which has safety hazards. SUMMARY

[0003] In view of the above problems, the present application provides a centralized power prediction management method, system, device and medium for new energy power stations.

[0004] Therefore, the technical problem solved by the present application is that after deploying multiple prediction systems in each station, there is a lack of evaluation mechanism based on historical accuracy, which cannot be individually matched according to the geographical location and climate characteristics of each station, resulting in unreasonable selection of prediction systems, affecting the overall prediction accuracy, and the automatic correction rule is not comprehensive, the manual correction is inefficient in point-by-point operation, the two methods cannot work together, and there is a lack of decision-making mechanism for intelligent selection of correction methods according to the prediction deviation characteristics.

[0005] To solve the above technical problems, the present application provides the following technical solutions: a centralized power prediction management method for new energy power stations, comprising, receiving prediction data packets uploaded by multiple stations through a forward isolation device, the prediction data packets containing prediction power data generated by multiple prediction systems of each station; based on the historical prediction accuracy of each prediction system, calculating the adaptation score of each prediction system for each station, and selecting corresponding prediction power data from the multiple prediction systems of each station as the to-be-corrected data according to the adaptation score; in response to the correction operation of the user on the to-be-corrected data, the correction operation includes applying a uniform correction parameter to the to-be-corrected data of the selected time period, or dragging and adjusting the specified point of the to-be-corrected data curve; downloading the corrected prediction power data to the gateway machine of each station through a reverse isolation device.

[0006] As a preferred embodiment of the centralized power prediction management method for new energy power plants described in this invention, the step of receiving prediction data packets uploaded by multiple power plants through a forward isolation device includes collecting prediction data packets from the gateway of each power plant. The prediction data packets are transferred to the receiving buffer of the centralized prediction center via a one-way optical shutter mechanism of the forward isolation device. Perform integrity checks on the predicted data packets in the receive buffer, and write the predicted data packets to the historical database once the check passes.

[0007] As a preferred embodiment of the centralized power prediction management method for new energy power plants described in this invention, the calculation of the adaptability score of each prediction system for each power plant includes extracting the predicted power data and corresponding measured power data of each prediction system within a set time window from the historical database. Calculate the deviation statistics between the predicted power data and the measured power data of each prediction system; Based on the aforementioned deviation statistics, each prediction system generates an adaptation score for each site.

[0008] As a preferred embodiment of the centralized power prediction management method for new energy power plants described in this invention, the step of selecting the corresponding predicted power data as the data to be corrected from multiple prediction systems of each power plant according to the adaptability score includes comparing the adaptability score values ​​of each prediction system for each power plant. The predicted power data of the prediction system with the highest fit score is determined as the data to be corrected for this site; Establish a mapping table between each site and its corresponding prediction system, and record the identifier of the prediction system currently selected by each site.

[0009] The beneficial effects of this preferred technical solution are as follows: by independently scoring each site and establishing a mapping relationship table, personalized matching between the prediction system and the site is achieved, avoiding the problem of poor prediction accuracy for some sites caused by the global uniform selection of a prediction manufacturer in the existing technology. This enables sites with different geographical locations and different climate characteristics to automatically match the most suitable prediction system for themselves, thereby improving the overall prediction accuracy and reducing scheduling assessment risks.

[0010] As a preferred embodiment of the centralized power prediction and management method for new energy power plants described in this invention, the step of applying a unified correction parameter to the data to be corrected in the selected time period includes receiving the target time period selected by the user through an interactive interface. Obtain the uniform correction parameters input by the user for the target time period; The data to be corrected at each time point within the target time period is calculated with the unified correction parameter to generate the corrected data to be corrected. The uncorrected data curve before correction and the uncorrected data curve after correction are displayed simultaneously on the interactive interface, and the correction is completed after receiving the confirmation instruction of the user.

[0011] As a preferred scheme of the new energy power station centralized power prediction management method, the method comprises the following steps: the interactive interface displays the uncorrected data curve; the user drags the key point of the uncorrected data curve on the interactive interface; the target point coordinate after the dragging is recorded; the uncorrected data curve is calculated according to the target point coordinate to generate an adjusted uncorrected data curve; and the adjusted uncorrected data curve is updated and displayed in real time on the interactive interface. The operation of the user dragging the key point on the uncorrected data curve is captured, and the target point coordinate after the dragging is recorded. The uncorrected data curve is calculated according to the target point coordinate to generate an adjusted uncorrected data curve. The adjusted uncorrected data curve is updated and displayed in real time on the interactive interface.

[0012] As a preferred scheme of the new energy power station centralized power prediction management method, the method comprises the following steps: the interactive interface displays the uncorrected data curve; the user drags the key point of the uncorrected data curve on the interactive interface; the target point coordinate after the dragging is recorded; the uncorrected data curve is calculated according to the target point coordinate to generate an adjusted uncorrected data curve; and the adjusted uncorrected data curve is updated and displayed in real time on the interactive interface. When the overall deviation rate exceeds the set threshold, a correction operation of applying a uniform correction parameter to the uncorrected data of the selected time period is adopted. When the overall deviation rate does not exceed the set threshold but the uncorrected data curve has a local mutation point, a correction operation of dragging and adjusting the specified point of the uncorrected data curve is adopted.

[0013] The beneficial effects of the preferred technical scheme are as follows: by calculating the overall deviation rate and automatically selecting the correction mode according to the deviation characteristics, intelligent decision of the correction strategy is realized, and the problems of low efficiency and easy errors in manual judgment of the correction mode in the prior art are solved. When the overall prediction curve deviates, batch correction is adopted to improve the processing efficiency, and when there is a local anomaly, fine dragging adjustment is adopted to ensure the correction accuracy, so that the correction efficiency and correction quality are considered.

[0014] The application provides a new energy power station centralized power prediction management system.

[0015] To solve the above technical problems, the application provides the following technical scheme: a new energy power station centralized power prediction management system, comprising: a plurality of prediction sub-stations, each prediction sub-station comprising a plurality of prediction systems and a gateway machine, the plurality of prediction systems being used to generate prediction power data, and the gateway machine being used to collect the prediction power data of each prediction system and encapsulate the prediction power data into a prediction data packet; A forward isolation device is arranged between each prediction sub-station and the centralized prediction center, and is used to upload the prediction data packet to the centralized prediction center through a unidirectional transmission channel. The centralized prediction center comprises a data receiving module configured to receive prediction data packets uploaded through a forward isolation device; The precision evaluation module is configured to calculate an adaptation score of each prediction system for each station based on historical prediction precision of each prediction system; The prediction tuning module is configured to select corresponding prediction power data as to-be-corrected data from the plurality of prediction systems of each station according to the adaptation score; The correction processing module is configured to respond to a correction operation of a user on the to-be-corrected data, the correction operation comprising applying a uniform correction parameter to the to-be-corrected data of a selected time period, or performing drag adjustment on a specified point of the to-be-corrected data curve; The data issuing module is configured to generate corrected prediction power data; The reverse isolation device is arranged between the centralized prediction center and each prediction substation, and is configured to issue the corrected prediction power data to a gateway machine of a corresponding prediction substation.

[0016] The application provides a computer device comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps of the new energy power station centralized power prediction management method when executing the computer program.

[0017] The application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the new energy power station centralized power prediction management method when executed by a processor.

[0018] The application has the beneficial effects that the prediction system and the station are individually matched, different characteristics of the station can automatically select the most suitable prediction system by calculating the adaptation score of each prediction system for each station and establishing a mapping relationship table, and the overall prediction accuracy is improved.

[0019] The correction efficiency and the correction quality are considered, a double-mode scheme of batch correction parameter and curve drag adjustment is combined, and the correction mode is intelligently selected according to the overall deviation rate, so that the overall deviation can be quickly processed and the local anomaly can be finely adjusted, and the artificial operation burden is reduced.

[0020] An accuracy feedback optimization closed loop mechanism is constructed, the adaptation score is continuously updated, the system can dynamically adjust the optimization strategy according to the actual operation effect, and the prediction performance is continuously improved.

[0021] The safety and reliability of data transmission are ensured, the forward and reverse isolation devices are used to construct a bidirectional safe transmission channel, the network security protection requirement of the power industry is met, and the external intrusion risk is effectively prevented. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0023] Figure 1 A general flowchart of a centralized power prediction management method for a new energy power station according to an embodiment of the present application is provided. DETAILED DESCRIPTION

[0024] In order to make the present application more apparent and understandable, the specific embodiments of the present application will be described in detail in conjunction with the drawings of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort should fall within the protection scope of the present application.

[0025] Embodiment 1, refer to Figure 1 According to an embodiment of the present application, the embodiment provides a centralized power prediction management method for a new energy power station, which comprises the following steps: Step 1: receiving a prediction data packet uploaded by a plurality of stations through a forward isolation device, wherein the prediction data packet contains prediction power data generated by a plurality of prediction systems of each station; Step 2: calculating an adaptation score of each prediction system for each station based on the historical prediction accuracy of each prediction system, and selecting corresponding prediction power data from the plurality of prediction systems of each station as to-be-corrected data according to the adaptation score; Step 3: responding to a correction operation of a user on the to-be-corrected data, wherein the correction operation comprises applying a uniform correction parameter to the to-be-corrected data of a selected time period, or performing drag adjustment on a specified point of the to-be-corrected data curve; Step 4: issuing the corrected prediction power data to the gateway machine of each station through a reverse isolation device.

[0026] In the embodiment, the gateway machine of each station collects the predicted power data generated by multiple prediction systems in the station and encapsulates the predicted power data as a prediction data packet, the prediction data packet including short-term predicted power data, measured power data and data generation time identifier. The gateway machine sends the prediction data packet to the forward isolation device, and the forward isolation device transmits the prediction data packet to the receiving buffer of the centralized prediction center through the one-way optical gate mechanism. The centralized prediction center performs integrity check on the prediction data packet in the receiving buffer, and after the check passes, the prediction data packet is parsed and written into the historical database. The precision evaluation module extracts the predicted power data and the corresponding measured power data of each prediction system in a set time window from the historical database, calculates the deviation statistics between the predicted power data and the measured power data of each prediction system, the deviation statistics including accuracy, root mean square error and correlation coefficient, and generates an adaptation score value for each prediction system for each station. The prediction optimization module compares the adaptation score values of the prediction systems for each station, determines the predicted power data of the prediction system with the highest adaptation score value as the to-be-corrected data of the station, and establishes a mapping relationship table to record the prediction system identifier currently selected by each station.

[0027] The correction processing module shows the to-be-corrected data curve to the user through the interactive interface. When the user selects batch correction, the target time period is selected through the interactive interface and a uniform correction parameter is input, the correction processing module operates the to-be-corrected data at each time in the target time period with the uniform correction parameter to generate corrected to-be-corrected data, and the to-be-corrected data curve before and after correction is displayed on the interactive interface at the same time for the user to confirm. When the user selects drag adjustment, the correction processing module captures the operation of the user dragging the key point on the to-be-corrected data curve, records the coordinates of the target point after dragging, and performs interpolation calculation on the to-be-corrected data curve according to the coordinates of the target point to generate adjusted to-be-corrected data curve and update the display in real time. After the correction is completed, the data delivery module sends the corrected predicted power data to the reverse isolation device, the reverse isolation device delivers the corrected predicted power data to the gateway machine of the corresponding station through the safe channel, and the gateway machine receives and reports to the dispatching system after receiving.

[0028] Embodiment 2 is an embodiment of the present application, which provides a new energy power station centralized power prediction management method based on the above embodiment, comprising: In step 1, receive the prediction data packet uploaded by multiple stations through the forward isolation device, the prediction data packet containing the predicted power data generated by multiple prediction systems of each station, including steps A1-A3: A1: collect the prediction data packet from the gateway machine of each station; A2: transmit the prediction data packet to the receiving buffer of the centralized prediction center through the one-way optical gate mechanism of the forward isolation device; A3: performing integrity check on the prediction data packet in the receiving buffer, and writing the prediction data packet into the historical database when the check passes.

[0029] In the step A2, the forward isolation device can achieve physical isolation by using a unidirectional optical shutter device, which includes a sending end and a receiving end. The sending end is connected with the gateway machine of each field station, and the receiving end is connected with the receiving buffer of the centralized prediction center. The unidirectional optical signal transmission is used to achieve the safe transfer of the prediction data packet.

[0030] In an optional embodiment, in the step A2, the forward isolation device can achieve data transmission by using a transfer file system. The transfer file system sets a temporary storage area in the isolation device. The gateway machine writes the prediction data packet into the temporary storage area. The isolation device periodically scans the temporary storage area and transfers the prediction data packet to the receiving buffer of the centralized prediction center.

[0031] In another optional embodiment, in the step A2, the forward isolation device can also achieve unidirectional transmission after encrypting the prediction data packet by using a hardware encryption module. The receiving buffer of the centralized prediction center decrypts and restores the encrypted prediction data packet after receiving it.

[0032] In the step A3, the integrity check can be performed by checking whether the file format of the prediction data packet conforms to the preset data protocol specification and whether the necessary fields in the prediction data packet are complete. The necessary fields include the field station identifier, the prediction system identifier, the data generation time identifier, and the prediction power data.

[0033] In an optional embodiment, in the step A3, the integrity check can be performed by calculating the checksum of the prediction data packet and comparing it with the checksum carried in the prediction data packet. When the two checksums are consistent, it is determined that the prediction data packet has not been damaged in the transmission process.

[0034] In another optional embodiment, in the step A3, the integrity check can also be performed by checking whether the timestamp in the prediction data packet is within a reasonable range. When the deviation of the timestamp from the current time exceeds a set threshold, it is determined that the check fails.

[0035] In the step A3, after the check passes, the prediction data packet is written into the historical database by parsing the prediction data packet to extract the prediction power data and the measured power data of each prediction system, establishing an index structure according to the field station identifier and the prediction system identifier, and writing the parsed data into the corresponding data table of the historical database.

[0036] In an optional embodiment, after the verification passes in step A3, the predicted data packet is written into the historical database by: storing the original file of the predicted data packet into the file system, and extracting the key information in the predicted data packet and writing it into the relational database, the key information including the station identification, the prediction system identification, the data generation time and the predicted power value.

[0037] In another optional embodiment, after the verification passes in step A3, the predicted data packet is written into the historical database by: storing the predicted data packet in the distributed database, and according to the station identification, the data is sharded, the predicted data packets of different stations are distributed and stored in different data nodes, and the data read-write performance is improved.

[0038] In step 2: based on the historical prediction accuracy of each prediction system, the adaptation score of each prediction system for each station is calculated, and the corresponding predicted power data is selected from the multiple prediction systems of each station as the to-be-corrected data according to the adaptation score, including steps B1-B6: B1: extract the predicted power data and the corresponding measured power data of each prediction system in the historical database within a set time window; B2: calculate the deviation statistics between the predicted power data and the measured power data of each prediction system; B3: generate an adaptation score value for each prediction system for each station according to the deviation statistics.

[0039] B4: compare the adaptation score values of each prediction system for each station; B5: determine the predicted power data of the prediction system with the highest adaptation score value as the to-be-corrected data of the station; B6: establish a mapping relationship table of each station and the corresponding prediction system, and record the prediction system identification currently selected by each station.

[0040] Specifically, in step B1, the accuracy evaluation module extracts the predicted power data and the corresponding measured power data of each prediction system within a set time window from the historical database. The set time window is configured according to the accuracy evaluation requirements, which can be selected as the last 7 days, the last 15 days, or the last 30 days. During the extraction process, the accuracy evaluation module queries all data records within the specified time window from the corresponding data table of the historical database according to the station identifier and the prediction system identifier. Each data record includes a timestamp, a predicted power value, and a measured power value. To ensure the effectiveness of the data, the accuracy evaluation module filters out data records with missing values or marked as abnormal when extracting the data. In step B2, the accuracy evaluation module calculates the deviation statistics between the predicted power data and the measured power data of each prediction system, including the accuracy rate, the root mean square error, and the correlation coefficient. The accuracy rate is calculated by counting the proportion of time points within the time window where the deviation between the predicted power and the measured power is within the allowed range. The root mean square error is calculated by squaring, summing, and averaging the difference between the predicted power and the measured power, and then taking the square root. The correlation coefficient is calculated by using the Pearson correlation coefficient algorithm to evaluate the linear correlation degree between the predicted power data sequence and the measured power data sequence. In step B3, the accuracy evaluation module generates an adaptation score value for each prediction system for each station based on the deviation statistics. The adaptation score value is calculated using a weighted sum method, which assigns weights to the accuracy rate, the inverse of the root mean square error, and the correlation coefficient, and then performs a weighted sum. The weight coefficients are configured according to the emphasis of the dispatching assessment rules. In step B4, the prediction optimization module compares the adaptation score values of each prediction system for each station. It reads the latest adaptation score values of each prediction system for the station from the historical database and performs numerical comparison. In step B5, the prediction optimization module determines the predicted power data of the prediction system with the highest adaptation score value as the to-be-corrected data for the station. When the adaptation score values of multiple prediction systems are the same, the prediction system with a higher historical reporting success rate is preferred. In step B6, the prediction optimization module establishes a mapping relationship table for each station and the corresponding prediction system. The mapping relationship table takes the station identifier as the primary key and records the prediction system identifier currently used by each station, the adaptation score value, the selection time, and the historical reporting success number. This mapping relationship table is stored in the historical database for subsequent query and statistical analysis.

[0041] In step 3: in response to the user's correction operation on the to-be-corrected data, the correction operation includes applying a uniform correction parameter to the selected time period of to-be-corrected data, or adjusting the specified point of the to-be-corrected data curve by dragging, which includes the following steps C1-C3: C1: calculating the overall deviation rate of the to-be-corrected data and the historical same-period measured power data; C2: when the overall deviation rate exceeds a set threshold, a correction operation of applying a uniform correction parameter to the selected time period of to-be-corrected data is adopted; C3: When the overall deviation rate does not exceed the set threshold but the to-be-corrected data curve has a local mutation point, a correction operation of dragging and adjusting the specified point of the to-be-corrected data curve is adopted.

[0042] Specifically, in step C1, the correction processing module calculates the overall deviation rate of the to-be-corrected data and the historical same-period measured power data. The historical same period refers to the historical data of the same date and the same time period corresponding to the to-be-corrected data. For example, when the to-be-corrected data is a predicted power curve of a certain day, the historical same-period measured power data is the power data actually run on that day. The overall deviation rate is calculated by statistically calculating the average deviation percentage of the predicted power at all time points in the to-be-corrected data and the historical same-period measured power. When the historical same-period measured power data does not exist, the correction processing module uses the average value of the measured power in the same time period in the recent period as a reference. In step C2, when the overall deviation rate exceeds a set threshold, the correction processing module determines that the to-be-corrected data has overall deviation, and uses a correction operation of applying a uniform correction parameter to the to-be-corrected data in the selected time period. The set threshold is configured according to the characteristics of the site and the requirements of the dispatching assessment, and is usually set to 10% to 20%. The correction processing module displays the comparison chart of the to-be-corrected data curve and the historical same-period measured power curve to the user through the interactive interface, and prompts the overall deviation rate value. The user selects the target time period to be corrected through the interactive interface and inputs the uniform correction parameter, which includes a multiplication coefficient or an addition coefficient. The correction processing module calculates the to-be-corrected data at each time point in the target time period with the uniform correction parameter to generate the corrected to-be-corrected data. The calculation method is that when the uniform correction parameter is a multiplication coefficient, the to-be-corrected data is multiplied by the coefficient, and when the uniform correction parameter is an addition coefficient, the to-be-corrected data is added to the coefficient. The correction processing module simultaneously displays the to-be-corrected data curve before correction and the to-be-corrected data curve after correction on the interactive interface, so that the user can visually compare the correction effect. When the user issues a confirmation instruction through the interactive interface, the correction processing module completes the correction and takes the corrected to-be-corrected data as the final predicted power data. In step C3, when the overall deviation rate does not exceed the set threshold but the to-be-corrected data curve has local mutation points, the correction processing module determines that the to-be-corrected data does not have overall deviation but has local abnormalities, and uses a correction operation of dragging and adjusting the specified points of the to-be-corrected data curve. The local mutation point refers to a point in the to-be-corrected data curve where the power change between adjacent time points exceeds the climbing ability of the site. The correction processing module identifies the local mutation point by calculating the power difference between adjacent time points and comparing it with the rated climbing rate of the site. The correction processing module displays the to-be-corrected data curve on the interactive interface and marks the detected local mutation point position with a special marker. The user drags the key point of the to-be-corrected data curve on the interactive interface through the mouse or touch method. The correction processing module captures the user's drag operation in real time and records the coordinates of the target point after dragging.The correction processing module performs interpolation calculation on the to-be-corrected data curve according to the target point coordinates, and re-generates an adjusted to-be-corrected data curve by using a cubic spline interpolation algorithm under the premise of maintaining the smoothness of the curve. In the interpolation calculation process, the values of the points not dragged by the user remain unchanged, and only the dragged points and their adjacent regions are adjusted. The correction processing module updates and displays the adjusted to-be-corrected data curve in real time on the interactive interface, and the user can repeatedly drag and adjust until satisfied. When the user completes all the dragging operations, the correction processing module takes the adjusted to-be-corrected data as the final predicted power data.

[0043] Further, applying a uniform correction parameter to the to-be-corrected data of the selected time period includes receiving a target time period selected by the user through the interactive interface; obtaining a uniform correction parameter input by the user for the target time period; operating the to-be-corrected data at each time in the target time period with the uniform correction parameter to generate corrected to-be-corrected data; displaying the to-be-corrected data curve before correction and the to-be-corrected data curve after correction on the interactive interface simultaneously, and completing the correction after receiving the confirmation instruction of the user.

[0044] Further, the dragging adjustment of the specified point of the to-be-corrected data curve includes displaying the to-be-corrected data curve on the interactive interface; capturing the operation of the user dragging the key point on the to-be-corrected data curve, and recording the target point coordinates after the dragging; performing interpolation calculation on the to-be-corrected data curve according to the target point coordinates to generate an adjusted to-be-corrected data curve; updating and displaying the adjusted to-be-corrected data curve in real time on the interactive interface.

[0045] In step 4: the corrected predicted power data is transmitted to the gateway machine of each station through the reverse isolation device, including the following steps D1-D3: D1: the centralized prediction center generates a to-be-transmitted data packet, which contains the corrected predicted power data, the target station identifier, the data generation timestamp and the check code; D2: the to-be-transmitted data packet is sent to the reverse isolation device, and the reverse isolation device performs safety inspection on the to-be-transmitted data packet, including data format verification, target station legality verification and data content compliance verification; D3: when the safety inspection passes, the reverse isolation device transmits the to-be-transmitted data packet to the gateway machine of the corresponding station, and the gateway machine receives the to-be-transmitted data packet, analyzes it, extracts the corrected predicted power data and reports it to the dispatching system.

[0046] In the step 4, the reverse isolation device can realize data delivery by adopting a double-buffer mechanism, the centralized prediction center writes the corrected predicted power data into a first buffer area of the reverse isolation device, the reverse isolation device transfers the data from the first buffer area to a second buffer area after passing the preset safety check, and the gateway machine reads the corrected predicted power data from the second buffer area.

[0047] In the step 4, the reverse isolation device can realize data delivery by adopting a double-buffer mechanism, the centralized prediction center writes the corrected predicted power data into a first buffer area of the reverse isolation device, the reverse isolation device transfers the data from the first buffer area to a second buffer area after passing the preset safety check, and the gateway machine reads the corrected predicted power data from the second buffer area.

[0048] In the step 4, the reverse isolation device can realize data delivery by adopting a double-buffer mechanism, the centralized prediction center writes the corrected predicted power data into a first buffer area of the reverse isolation device, the reverse isolation device transfers the data from the first buffer area to a second buffer area after passing the preset safety check, and the gateway machine reads the corrected predicted power data from the second buffer area.

[0049] Embodiment 3 is an embodiment of the present application, which provides a centralized power prediction management system for new energy power stations, comprising: a plurality of prediction sub-stations, each prediction sub-station comprising a plurality of prediction systems and a gateway machine, the plurality of prediction systems being configured to generate predicted power data, and the gateway machine being configured to collect the predicted power data of each prediction system and encapsulate the predicted power data into a predicted data packet; a forward isolation device arranged between each prediction sub-station and the centralized prediction center, and configured to upload the predicted data packet to the centralized prediction center through a unidirectional transmission channel; the centralized prediction center comprising a data receiving module configured to receive the predicted data packet uploaded through the forward isolation device; a precision evaluation module configured to calculate an adaptation score of each prediction system for each station based on the historical prediction precision of each prediction system; a prediction tuning module configured to select corresponding predicted power data from the plurality of prediction systems of each station as to-be-corrected data according to the adaptation score; a correction processing module configured to respond to a correction operation of a user on the to-be-corrected data, the correction operation comprising applying a uniform correction parameter to the to-be-corrected data of a selected time period, or performing drag adjustment on a specified point of the to-be-corrected data curve; a data delivery module configured to generate corrected predicted power data; The reverse isolation device is arranged between the centralized prediction center and each prediction substation, and is used for issuing the corrected predicted power data to a gateway machine of a corresponding prediction substation.

[0050] The embodiment further provides an electronic device suitable for the new energy power station centralized power prediction management method, including a memory and a processor; the memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions to realize the new energy power station centralized power prediction management method.

[0051] The embodiment further provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to realize the new energy power station centralized power prediction management method.

[0052] The storage medium provided by the embodiment and the new energy power station centralized power prediction management method provided by the above embodiment belong to the same inventive concept, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0053] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by software and necessary general hardware, and of course can be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0054] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A centralized power prediction and management method for new energy power plants, characterized in that: include, Receive prediction data packets uploaded by multiple power stations through a forward isolation device, wherein the prediction data packets contain prediction power data generated by multiple prediction systems of each power station; Based on the historical prediction accuracy of each prediction system, the adaptability score of each prediction system for each site is calculated, and the corresponding predicted power data is selected from multiple prediction systems of each site as the data to be corrected according to the adaptability score. Responding to user correction operations on the data to be corrected, the correction operations include applying uniform correction parameters to the data to be corrected within a selected time period, or dragging and adjusting specified points on the curve of the data to be corrected. The corrected predicted power data is sent to the gateway of each site through the reverse isolation device.

2. The centralized power prediction and management method for new energy power plants as described in claim 1, characterized in that: The process of receiving prediction data packets uploaded by multiple sites through a forward isolation device includes collecting prediction data packets from the gateway of each site. The prediction data packets are transferred to the receiving buffer of the centralized prediction center via a one-way optical shutter mechanism of the forward isolation device. Perform integrity checks on the predicted data packets in the receive buffer, and write the predicted data packets to the historical database once the check passes.

3. The centralized power prediction and management method for new energy power plants as described in claim 2, characterized in that: The calculation of the adaptability score of each prediction system for each site includes extracting the predicted power data and the corresponding measured power data of each prediction system within a set time window from the historical database. Calculate the deviation statistics between the predicted power data and the measured power data of each prediction system; Based on the aforementioned deviation statistics, each prediction system generates an adaptation score for each site.

4. The centralized power prediction and management method for new energy power plants as described in claim 3, characterized in that: The step of selecting the corresponding predicted power data as the data to be corrected from multiple prediction systems of each site based on the fit score includes comparing the fit score values ​​of each prediction system for each site. The predicted power data of the prediction system with the highest fit score is determined as the data to be corrected for this site; Establish a mapping table between each site and its corresponding prediction system, and record the identifier of the prediction system currently selected by each site.

5. The centralized power prediction and management method for new energy power plants as described in claim 4, characterized in that: The step of applying uniform correction parameters to the data to be corrected within a selected time period includes receiving the target time period selected by the user through an interactive interface. Obtain the uniform correction parameters input by the user for the target time period; The data to be corrected at each time point within the target time period is calculated with the unified correction parameter to generate the corrected data to be corrected. The interactive interface simultaneously displays the data curve to be corrected before and after correction, and completes the correction after receiving the user's confirmation command.

6. The centralized power prediction and management method for new energy power plants as described in claim 5, characterized in that: The step of dragging and adjusting the specified points of the data curve to be corrected includes displaying the data curve to be corrected on the interactive interface. Capture the user's dragging operation on the data curve to be corrected, and record the coordinates of the target point after dragging; Based on the coordinates of the target point, interpolation calculations are performed on the data curve to be corrected to generate the adjusted data curve to be corrected. The adjusted data curve to be corrected is updated and displayed in real time on the interactive interface.

7. The centralized power prediction and management method for new energy power plants as described in claim 6, characterized in that: The method of performing the correction operation is determined based on the deviation characteristics of the data to be corrected, including calculating the overall deviation rate between the data to be corrected and the historical measured power data of the same period; When the overall deviation rate exceeds the set threshold, a correction operation is performed by applying uniform correction parameters to the data to be corrected within the selected time period. When the overall deviation rate does not exceed the set threshold but there are local abrupt changes in the data curve to be corrected, a correction operation is performed by dragging and adjusting the specified points of the data curve to be corrected.

8. A centralized power prediction and management system for new energy power plants, employing the centralized power prediction and management method for new energy power plants as described in any one of claims 1 to 7, characterized in that, include: Multiple prediction substations, each prediction substation including multiple prediction systems and a gateway machine, the multiple prediction systems are used to generate prediction power data, and the gateway machine is used to collect the prediction power data of each prediction system and encapsulate it into prediction data packets; A forward isolation device is installed between each prediction substation and the central prediction center to upload the prediction data packet to the central prediction center through a one-way transmission channel. The centralized prediction center includes: a data receiving module for receiving prediction data packets uploaded through a forward isolation device; The accuracy evaluation module is used to calculate the suitability score of each prediction system for each site based on the historical prediction accuracy of each prediction system. The prediction optimization module is used to select the corresponding predicted power data from multiple prediction systems of each site as the data to be corrected based on the adaptability score. The correction processing module is used to respond to the user's correction operation on the data to be corrected. The correction operation includes applying a uniform correction parameter to the data to be corrected in a selected time period, or dragging and adjusting a specified point of the curve of the data to be corrected. The data delivery module is used to generate corrected predicted power data; A reverse isolation device is installed between the centralized prediction center and each prediction substation, and is used to send the corrected prediction power data to the gateway of the corresponding prediction substation.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the centralized power prediction and management method for new energy power plants according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the centralized power prediction and management method for new energy power plants according to any one of claims 1 to 7.

Images