Intensive monitoring and early warning method for grid-connected operation examination of multi-new energy station

Abstract

The application provides a grid-connected operation examination intensive monitoring and early warning method for multiple new energy stations, relates to the field of grid-connected operation examination of multiple new energy stations, and comprises the following steps: collecting operation data with time tags on each station side and uploading the operation data to a centralized control side and a dispatching side, obtaining instructions from the dispatching side on each station side, marking the instructions on each station side to obtain dispatching instructions with time tags; matching the dispatching instructions with time tags with operation data with time tags on the centralized control side to obtain aligned data, calculating the aligned data to obtain real-time centralized control side examination results; correcting the real-time centralized control side examination results based on preset multi-dimensional influence factors and historical deviation data corresponding to examination scenes, performing hierarchical early warning according to the real-time centralized control side correction examination results, and displaying key information corresponding to the hierarchical early warning, the real-time centralized control side correction examination results and the operation data with time tags on each station side.

Description

Technical Field

[0001] This invention relates to the field of grid-connected operation assessment technology for multiple new energy power plants, specifically to an intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants. Background Technology

[0002] Currently, the grid-connected operation assessment of new energy power plants (wind power, photovoltaic, and new energy storage) follows two detailed rules for each region, and the assessment work is mainly completed by the dispatching side. After the end of each day, the dispatching side publishes the final deduction results of the previous day's assessment on a designated official website, without disclosing the detailed process of assessment calculation, data basis, and reasons for deduction. Moreover, this historical assessment data cannot be collected from the power plant end and must be manually logged into the website for querying, which is inefficient and prone to data entry errors.

[0003] The existing new energy power stations lack a professional assessment and management system. Even if a few stations have such systems, they operate on a single-station independent monitoring model. Data from each station's system is isolated, unable to communicate, and redundant construction leads to high costs. The centralized control side struggles to achieve comprehensive control and centralized management of the assessment status of multiple subordinate stations. Even with simple monitoring systems at the station level, basic indicator calculations are only possible based on local station data. Without a unified assessment rule base, the entire assessment process cannot be replicated using the same algorithms as the dispatch side. Furthermore, time misalignment between instructions issued by the dispatch side and data collected by the centralized control side further affects the accuracy of assessment calculations. In addition, historical dispatch assessment data requires manual querying and lacks automated acquisition channels. Stations cannot achieve real-time comparison of assessment results, nor can they efficiently complete post-event comparisons. Maintenance personnel can only learn about deductions manually the following day, unable to proactively identify assessment risks or accurately pinpoint the root causes of deductions. The existing system also lacks a standardized and tiered early warning mechanism for the assessment items of the two detailed rules. Most of the stations make rectifications after the fact, making it difficult to predict and proactively handle assessment risks in advance. This results in a high deduction rate for new energy stations in the assessment, which damages their operational compliance and economic benefits. Summary of the Invention

[0004] This invention addresses the problems existing in the prior art by providing an intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants, enabling centralized collection, unified management, and global monitoring of grid-connected operation assessment data of multiple regions and types of new energy power plants from the centralized control side.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: The time-stamped operation data of each station is collected and uploaded to the central control side and the dispatch side. Each station receives the instructions from the dispatch side and marks the instructions on each station to obtain time-stamped dispatch instructions. On the centralized control side, the time-stamped scheduling instructions are matched with the time-stamped running data to obtain aligned data, and the aligned data is calculated to obtain the real-time centralized control side assessment results. The historical centralized control side assessment results and historical dispatch side assessment results are compared to obtain historical deviation data. Based on the preset multi-dimensional influencing factors and historical deviation data, the real-time centralized control side assessment results are corrected to obtain the real-time centralized control side corrected assessment results. Based on the real-time centralized control side correction assessment results, graded early warnings are issued, and the key information corresponding to the graded early warnings, the real-time centralized control side correction assessment results, and the time-stamped operation data of each station are displayed.

[0006] In some embodiments, the process of collecting time-stamped operational data from each station and uploading it to the central control and dispatching sides includes: Collect operational data from each station; A unified time stamp is applied to the operational data at each station to obtain time-stamped operational data. The time-stamped operational data is uploaded from the station side to the central control side and the dispatch side; On the centralized control side, the time-stamped runtime data is deduplicated, completed, outlier removed, and normalized before being stored in the distributed database.

[0007] In some embodiments, the process of matching time-stamped scheduling instructions with time-stamped runtime data on the centralized control side to obtain aligned data includes: The central control unit receives time-stamped dispatch instructions sent by the station side; In response to time stamp anomalies, the time-stamped scheduling instructions are corrected to obtain scheduling instructions with normal time stamps. The time stamp anomalies include out-of-order time stamp anomalies and missing time stamp anomalies. The scheduling instructions corresponding to out-of-order time stamp anomalies are reordered and put back in place, and the missing time stamp anomalies are filled in by the LSTM timing prediction method. In response to a normal time stamp, the time-stamped scheduling instructions are matched with time-stamped running data on the centralized control side. The distributed database is retrieved based on the time stamp to obtain the corresponding running data. The scheduling instruction data and the running data are then bound together to obtain aligned data. When multiple scheduling instruction data exist under the same time point, the optimal and latest scheduling instruction data is selected based on the principles of optimal data quality and latest time.

[0008] In some embodiments, the process of calculating the alignment data to obtain the real-time centralized control side assessment result includes: On the centralized control side, the alignment data of each station side are calculated separately for each time period according to the preset calculation rules to obtain the cumulative index value for each time period; Obtain the preset assessment rules, perform deductions on the cumulative indicator values ​​over a period of time, and obtain the deduction results; The cumulative indicator values ​​and deduction results for each time period are summarized into the real-time centralized control side assessment results.

[0009] In some embodiments, the process of obtaining historical centralized control side assessment results and historical scheduling side assessment results, comparing them to obtain historical deviation data, and constructing assessment scenarios includes: On the centralized control side, obtain the historical centralized control side assessment results and historical centralized control side input data for a specified date in the past. On the scheduling side, obtain the historical scheduling side assessment results for the corresponding date that have been published. The input data of the historical centralized control side includes time-stamped operation data and alignment data for the corresponding date. Historical deviation data is obtained by comparing the historical centralized control side assessment results and the historical dispatch side assessment results according to the dimensions of station, assessment date, assessment item, indicator name and time scale. The day is divided into several operating periods according to the changes in power grid supply and demand, including periods of high renewable energy generation, peak electricity consumption periods, and other periods; In each time period, the historical deviation value and historical deviation change rate of the historical deviation data are calculated. Based on the historical deviation value and historical deviation change rate and the input data from the historical centralized control side, cluster analysis is performed to obtain the assessment scenario.

[0010] In some embodiments, the process of correcting the real-time centralized control side assessment results based on preset multi-dimensional influencing factors and historical deviation data corresponding to the assessment scenario to obtain the corrected real-time centralized control side assessment results includes: Identify the runtime segment corresponding to the real-time centralized control side assessment results, match the corresponding assessment scenario based on the real-time centralized control side input data, and obtain several matching assessment scenarios. The real-time centralized control side input data includes real-time time-stamped runtime data and alignment data. Obtain multi-dimensional influencing factors, and perform weighted summation on the multi-dimensional influencing factors to obtain the preliminary corrected weights for each matched assessment scenario. The multi-dimensional influencing factors include time period weight, scenario similarity weight, and deviation degree weight. The time period weight is the weight of the running period, the scenario similarity weight is the weight of the similarity between the real-time centralized control side input data and the historical centralized control side input data, and the deviation degree weight is the weight of the historical deviation value and the historical deviation change rate corresponding to the assessment scenario. The deviation correction coefficient for each scenario is calculated based on the historical centralized control side assessment results and historical scheduling side assessment results corresponding to each matching assessment scenario. The deviation correction coefficient for each matched assessment scenario is weighted and calculated with the corresponding initial correction weight to obtain the real-time correction coefficient; The real-time centralized control side assessment results are corrected based on the real-time correction coefficient to obtain the real-time centralized control side corrected assessment results.

[0011] In some embodiments, the process of issuing graded early warnings based on the real-time centralized control side correction assessment results, and displaying the key information corresponding to the graded early warnings, the real-time centralized control side correction assessment results, and the time-stamped operating data of each station includes: The assessment results are corrected based on real-time centralized control and a tiered early warning system is implemented. Specifically, a risk warning is triggered when the deduction result exceeds the first threshold, and a deduction warning is triggered when the deduction result exceeds the second threshold. The visualization terminal displays key information corresponding to the graded early warning, real-time centralized control side correction assessment results, and time-stamped operation data from each station.

[0012] This invention proposes an integrated monitoring and early warning system for grid-connected operation assessment of multiple renewable energy power plants, comprising: The alignment unit is configured to collect time-stamped operation data from each station and upload it to the central control and dispatching sides. Each station receives instructions from the dispatching side and marks the instructions on each station to obtain time-stamped dispatching instructions. The centralized control unit is configured to match time-stamped scheduling instructions with time-stamped running data on the centralized control side to obtain aligned data, and to calculate the aligned data to obtain real-time centralized control side assessment results. The correction unit is configured to obtain historical centralized control side assessment results and historical scheduling side assessment results for comparison to obtain historical deviation data, and correct the real-time centralized control side assessment results based on preset multi-dimensional influence factors and historical deviation data to obtain real-time centralized control side corrected assessment results. The early warning display unit is configured to issue graded early warnings based on the real-time centralized control side's correction assessment results, and to display the key information corresponding to the graded early warnings, the real-time centralized control side's correction assessment results, and the time-stamped operating data from each station.

[0013] This invention proposes a computer device, comprising: At least one processor; and a memory storing a computer program that can run on the processor, wherein the processor executes the program to perform the steps of the intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants.

[0014] This invention proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention proposes an intensive monitoring and early warning method for grid-connected operation assessment of multiple renewable energy power plants. The method includes: collecting time-stamped operation data from each power plant and uploading it to the centralized control and dispatching sides; each power plant receiving instructions from the dispatching side and tagging the instructions at each power plant to obtain time-stamped dispatching instructions; matching the time-stamped dispatching instructions with the time-stamped operation data at the centralized control side to obtain aligned data, and calculating the aligned data to obtain real-time centralized control side assessment results; obtaining historical centralized control side assessment results and historical dispatching side assessment results for comparison to obtain historical deviation data and constructing assessment scenarios; correcting the real-time centralized control side assessment results based on preset multi-dimensional influencing factors and the historical deviation data corresponding to the assessment scenarios to obtain corrected real-time centralized control side assessment results; and providing graded early warnings based on the corrected real-time centralized control side assessment results, and displaying the key information corresponding to the graded early warnings, the corrected real-time centralized control side assessment results, and the time-stamped operation data from each power plant.

[0016] This invention addresses the problems of data silos and high construction costs caused by the lack of professional assessment systems at existing power stations, the independent deployment of systems at a few stations, and the centralized collection, unified management, and global monitoring of grid-connected operation assessment data of multiple regions and types of new energy power stations from the centralized control side, breaking the limitations of single-station monitoring.

[0017] This invention addresses the problems of needing to manually query historical scheduling assessment data, lack of automated acquisition channels, misalignment between scheduling instructions and data collected by the central control side, and lack of multi-dimensional data support for post-event comparison. It achieves basic data alignment by adding time stamps at the station side, corrects misaligned data using intelligent algorithms, designs an automated scheduling assessment data crawling function to efficiently acquire historical assessment data, configures triple data input for dual-end comparison, optimizes the model based on deviations, accurately locates the root cause of deductions, and improves the accuracy of assessment calculations.

[0018] This invention addresses the problems of existing monitoring methods lacking standardized early warning mechanisms and passively handling assessment risks. It constructs a hierarchical early warning system that adapts to both sets of detailed rules for all assessment items, achieves early risk warning based on assessment indicators calculated in real time on the centralized control side, and simultaneously completes integrated monitoring and display of assessment and auxiliary services. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0020] Figure 1The flowchart illustrates the intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants provided by this invention.

[0021] Figure 2 This invention provides a system module diagram for integrated monitoring and early warning of grid-connected operation assessment for multiple new energy power plants.

[0022] Figure 3 A schematic diagram of the structure of an embodiment of the computer device provided by the present invention.

[0023] Figure 4 This is a schematic diagram of an embodiment of the computer-readable storage medium provided by the present invention.

[0024] Figure 5 The system architecture diagram provided by this invention is for the intensive monitoring and early warning of grid-connected operation assessment of multiple new energy power plants.

[0025] Figure 6 This invention provides a multi-terminal data interaction diagram for the intensive monitoring and early warning of grid-connected operation assessment of multiple new energy power plants. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention. It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application.

[0027] It should be noted that all uses of "first" and "second" in the embodiments of the present invention are for the purpose of distinguishing two entities with the same name but different names or different parameters. It is clear that "first" and "second" are only for the convenience of description and should not be construed as limiting the embodiments of the present invention. Subsequent embodiments will not explain this in detail.

[0028] This invention proposes an intensive monitoring and early warning method for grid-connected operation assessment of multiple renewable energy power plants. Please refer to [link / reference]. Figure 1 ,include: The time-stamped operation data of each station is collected and uploaded to the central control side and the dispatch side. Each station receives the instructions from the dispatch side and marks the instructions on each station to obtain time-stamped dispatch instructions. On the centralized control side, the time-stamped scheduling instructions are matched with the time-stamped running data to obtain aligned data, and the aligned data is calculated to obtain the real-time centralized control side assessment results. The historical centralized control side assessment results and historical scheduling side assessment results are compared to obtain historical deviation data and construct assessment scenarios. Based on the preset multi-dimensional influencing factors and the historical deviation data corresponding to the assessment scenarios, the real-time centralized control side assessment results are corrected to obtain the real-time centralized control side corrected assessment results. Based on the real-time centralized control side correction assessment results, graded early warnings are issued, and the key information corresponding to the graded early warnings, the real-time centralized control side correction assessment results, and the time-stamped operation data of each station are displayed.

[0029] This technical solution innovatively adopts a centralized control-side intensive deployment architecture, designing a centralized monitoring and intelligent early warning system for grid-connected operation assessment of multiple new energy power plants. The system consists of seven core functional units: an automated crawling unit for scheduling assessment data, a centralized acquisition unit for multi-source data, an assessment rule base management unit, a centralized control-side assessment calculation unit, a two-end result comparison unit, an assessment deviation correction unit, and a hierarchical early warning and visualization unit. Each unit is directly connected bidirectionally / transmitted unidirectionally for collaborative linkage. Considering the actual situation that scheduling assessment data is historically published and requires website query, the system realizes the fully automated operation of real-time data acquisition and calculation, early risk warning, automatic crawling of historical data, post-event multi-source comparison, deviation correction and modeling, and result feedback optimization.

[0030] In the data acquisition phase, various types of motion machines, gateways, and other data acquisition devices are deployed at each site to collect electrical parameters, equipment operating status parameters, and environmental parameters in real time. Using a timescale as a reference, the scheduling command data and operational data received by each site are time-aligned on the central control side to obtain aligned data. On the central control side, calculations are performed using the aligned data according to preset assessment algorithms and rules to derive the real-time assessment results.

[0031] Next, historical assessment results from the centralized control side and the historical assessment results from the scheduling side are obtained. Differences are identified through data comparison algorithms, and historical deviation data is analyzed. Combined with preset multi-dimensional influencing factors, including time period weight, scene similarity weight, and deviation degree weight, the real-time centralized control side assessment results are corrected to obtain more accurate real-time centralized control side corrected assessment results.

[0032] A tiered early warning mechanism is set up based on the revised assessment results, and key early warning information, revised assessment results, and station-side operational data are displayed on the monitoring interface.

[0033] In some embodiments, please refer to Figure 1 and Figure 6 The process of collecting operational data from each station, setting time stamps to obtain time-stamped operational data, and uploading it to the central control and dispatching sides includes: Collect operational data from each station; A unified time stamp is applied to the operational data at each station to obtain time-stamped operational data. The time-stamped operational data is uploaded from the station side to the central control side and the dispatch side; On the centralized control side, the time-stamped runtime data is deduplicated, completed, outlier removed, and normalized before being stored in the distributed database.

[0034] The system is based on two detailed assessment criteria, embedding the two detailed criteria and calculation logic for each region, such as... Figure 6 As shown, when data is collected, the station automatically adds a timestamp to it to record the specific time the data was generated, ensuring the timeliness, accuracy, and traceability of the data. The station then sends this timestamped operational data to both the dispatching and central control sides, enabling both to promptly grasp the real-time operational status of the station.

[0035] After comprehensively analyzing the received data and the overall power grid operation requirements, the dispatching side issues instructions to the renewable energy power plants to adjust power generation and start / stop equipment. Upon receiving the instructions, the power plants timestamp them to specify the receiving time, and then forward the timestamped instructions to the central control side. This allows the central control side to synchronously understand the dispatching decisions and collaborate with the power plants to ensure the safe and stable operation of the renewable energy power plants, achieving efficient energy production and supply.

[0036] The dispatching side sends real-time command data directly to each station. The station sets a unified timestamp for all command data. There is a delay in the station forwarding the command sent to it by the dispatching side to the central control. After receiving the command, the station immediately sets its local timestamp. Even if there is a delay in forwarding it to the central control, the timestamp of the command is determined. The central control receives the above dispatch command data through the multi-source data centralized acquisition unit. It first completes the data standardization preprocessing and basic time alignment. For dispatch command data that cannot be fully aligned, the time dimension is corrected by intelligent algorithm. Then, it calls the standardized rules in the assessment rule base management unit to complete the calculation of indicators and pre-calculation of deduction results for all dimensions of assessment items such as active power regulation management (wind and solar AGC, energy storage AGC, primary frequency regulation, new energy power prediction), reactive power regulation management (AVC regulation performance), and safety and operation management (black start, unplanned outage / grid disconnection). Based on the pre-calculation results and the manually adjustable yellow / red level two warning threshold, real-time hierarchical warning is realized.

[0037] On the one hand, the station transmits operational data such as measured power, equipment status, and remote signaling and telemetry to the central control side in real time, and the central control side writes it into the distributed database.

[0038] On the other hand, after the dispatching side sends AGC / AVC adjustment commands, voltage curves and other commands to the field station, the field station uses the BeiDou / GPS unified time source to add a unified time stamp accurate to the millisecond level to the command the moment it successfully receives the command, and then encapsulates the command content and time stamp into a message and uploads it to the central control side.

[0039] After receiving a time-stamped instruction message, the centralized acquisition unit of the multi-source data on the control side uses the time stamp as the unique matching benchmark. Based on the time stamp value, it retrieves the measured power, equipment status, and remote signaling and telemetry data of the station within the same time and time interval from the distributed database. It then binds and matches the dispatch instruction data with the retrieved real-time operation data point by point according to the timestamp to complete the basic time alignment. When there are multiple duplicate messages under the same time stamp, the unique valid data is retained according to the principle of the best data quality bit and the latest time, and the rest are automatically filtered.

[0040] Various types of data acquisition equipment are widely deployed at each site to collect electrical parameters. Various sensors are responsible for collecting equipment operating status data as well as environmental data such as temperature, humidity, and wind speed, so as to comprehensively obtain the operating data of each site.

[0041] The dispatching side generates instruction data based on the overall operation plan and actual needs. This instruction data includes key information such as equipment start / stop instructions and power adjustment instructions. The instruction data is then accurately distributed to each site. Upon receiving the instruction data, the site adds a timestamp to it. This timestamp ensures consistency and comparability of instructions received by different sites in terms of time.

[0042] The station transmits time-stamped command data and its own collected operational data to the central control unit via a reliable communication network. Preprocessing is then performed on the received station operational data. This includes deduplication to remove duplicate data generated during data transmission; data completion using interpolation algorithms to fill in missing data based on historical patterns and correlations; outlier removal by identifying and removing data that significantly deviates from the normal range using reasonable threshold ranges or statistical methods; and normalization to unify data of different dimensions into a specific interval, such as [0,1], eliminating the impact of differences in data dimensions on the analysis results, thus obtaining the preprocessed data.

[0043] In some embodiments, please refer to Figure 1 The process of matching time-stamped scheduling instructions with time-stamped running data on the centralized control side to obtain aligned data includes: The central control unit receives time-stamped dispatch instructions sent by the station side; In response to time stamp anomalies, the time-stamped scheduling instructions are corrected to obtain scheduling instructions with normal time stamps. The time stamp anomalies include out-of-order time stamp anomalies and missing time stamp anomalies. The scheduling instructions corresponding to out-of-order time stamp anomalies are reordered and put back in place, and the missing time stamp anomalies are filled in by the LSTM timing prediction method. In response to a normal time stamp, the time-stamped scheduling instructions are matched with time-stamped running data on the centralized control side. The distributed database is retrieved based on the time stamp to obtain the corresponding running data. The scheduling instruction data and the running data are then bound together to obtain aligned data. When multiple scheduling instruction data exist under the same time point, the optimal and latest scheduling instruction data is selected based on the principles of optimal data quality and latest time.

[0044] In the data processing workflow, even after preprocessing, data still exhibits time discrepancies, primarily manifesting as out-of-order data and missing data. Out-of-order data is typically caused by factors such as asynchronous clocks in data acquisition devices, network latency during data transmission, or packet loss, resulting in a discrepancy between the recorded time sequence and the actual order of events. For out-of-order data, it is necessary to utilize the timestamp information inherent in the data itself, employing quicksort or mergesort to rearrange the data according to its chronological order, thus restoring its original temporal logic.

[0045] Missing data arises because of equipment malfunctions, communication interruptions, or accidental errors during data acquisition, resulting in data at certain points in time not being successfully collected or recorded. To fill in these missing data, the LSTM time series prediction method is employed. LSTM, as an advanced recurrent neural network model, captures long-term dependencies and complex patterns in time series data. By fully training the LSTM model with existing complete data, the model learns the inherent patterns and trends of the data. Then, the trained model is used to accurately predict and fill in the missing data points, thereby obtaining corrected preprocessed data.

[0046] After data correction, time alignment is performed. The scheduling instruction data and the corrected preprocessed data are precisely matched and bound according to their time stamps. A time stamp is the temporal coordinate of the data, ensuring that data from different sources correspond accurately in the time dimension. When multiple scheduling instructions appear under the same time stamp, they are filtered according to the principles of optimal data quality and latest time. Optimal data quality ensures the accuracy and reliability of the instructions, while the latest time reflects the most recent scheduling intent and requirements. For data that cannot be matched and aligned due to out-of-order time stamps or time gaps caused by network latency, communication interruptions, etc. Based on the inherent logic of message generation and acquisition time, out-of-order data points are reordered and their timing restored to ensure that the data is arranged in chronological order. For example, if the central control unit receives the 10:00:02 instruction first and then the 10:00:01 instruction, resulting in out-of-order data, the system automatically rearranges the data to 10:00:01→10:00:02, restoring the normal timing sequence.

[0047] An LSTM time series prediction model is used to fill in the missing data points during the interrupted period, and appropriate timestamps are added to the supplemented data to ensure the continuity and integrity of the time period. For example, if there is a 3-second data interruption from 10:00:00 to 10:00:03 with no messages, the system predicts and fills in 3 points based on the normal data before and after, and adds timestamps 00, 01, and 02 in sequence to restore the complete time period.

[0048] In some embodiments, please refer to Figure 1 The process of calculating the alignment data to obtain the real-time centralized control side assessment results includes: On the centralized control side, the alignment data of each station side are calculated separately for each time period according to the preset calculation rules to obtain the cumulative index value for each time period; Obtain the preset assessment rules, perform deductions on the cumulative indicator values ​​over a period of time, and obtain the deduction results; The cumulative indicator values ​​and deduction results for each time period are summarized into the real-time centralized control side assessment results.

[0049] Assessment Rules: The centralized control side uses standardized assessment rules from the assessment rule library that are completely consistent with those of the dispatch side. Typical rules include: availability of active power control systems for wind farms and photovoltaic systems. The availability of active power control systems for wind farms and photovoltaic systems should reach 100%. For every 1% decrease, 1 point is awarded per 10,000 kilowatts. Availability is calculated as the closed-loop time of the power plant's automatic active power control system divided by the time when the output of new energy sources is greater than the total capacity of the prototype machine or prototype inverter. 100%.

[0050] AGC Contribution Power Qualification Rate: The qualification rate for wind power and photovoltaic power contribution should reach 90%, with an additional 0.2 points per 10,000 kilowatts for each 1% decrease. The qualification rate refers to the ratio of the cumulative actual power contribution to the cumulative theoretical power contribution (i.e., the algebraic sum of each contribution) during each AGC adjustment instruction period for new energy units, calculated monthly. Actual power contribution refers to the integral value of the difference between the actual power curve and the initial power curve, where the actual power is in the same direction as the target power in the AGC instruction and the deviation is within 1% of the installed capacity during each AGC adjustment instruction period. Theoretical power contribution refers to the integral value of the difference between the target instruction curve and the initial power curve during each AGC adjustment instruction period.

[0051] AGC response time pass rate: The ratio of the number of times the response time passes to the total number of adjustments should be 99%. For every 1% decrease, 0.25 points per 10,000 kilowatts will be applied. Response time refers to the time from the issuance of the AGC command to a 10% change in power. The response time requirement is less than or equal to 60 seconds.

[0052] According to the calculation methods stipulated in the two detailed rules, the assessment items such as active power regulation, primary frequency regulation, power prediction, and AVC regulation are calculated for each time period to obtain the cumulative index value for each time period.

[0053] Based on the pre-calculated indicator values, combined with the assessment standards, deduction unit price, installed capacity, and time period coefficients stipulated in the two detailed rules, the deduction values, deduction categories, and cumulative deductions that will be generated if the current indicators continue until the end of the assessment period are deduced in real time, and converted into corresponding economic losses. The centralized control unit will perform detailed calculations based on the alignment data transmitted from each power station, according to preset calculation rules that take into account factors such as power system operating characteristics, power station equipment parameters, and assessment targets. Time-period calculations focus on data performance within a specific time interval, reflecting the overall operating status of the power station during that period; specifically, over a six-hour period, the contribution power qualification rate is calculated. , Only the portion that meets the integration requirement (same direction and deviation ≤ 1% of installed capacity) is considered. An AGC command is issued once per minute. The contribution power qualification rate assessment targets the stage where the unit output is greater than 10% of the installed capacity; reaching 90% or higher, 0.2 points are awarded for every 1% decrease. AGC availability rate... The station will send the AGC closed-loop status remote signaling points to the dispatcher to calculate the AGC availability rate, which should reach 99%. For every 1% decrease, 1 cent per 10,000 kilowatts will be used for assessment.

[0054] The centralized control unit will obtain two pre-set detailed rules. These rules clarify the qualification standards for each indicator and the corresponding deduction rules, and extrapolate the deductions based on the cumulative indicator values ​​for the time period calculated earlier. For example, if a wind farm is currently in the peak electricity consumption period of 18:00-22:00, the centralized control unit will calculate the cumulative indicator values ​​for the active power control system of the wind farm at the current time as follows: contribution power qualification rate 92%, AGC availability rate 98%.

[0055] The two detailed rules obtained stipulate that: for the stage where the power contribution qualification rate is greater than 10% of the installed capacity, reaching 90% or above, 0.2 points per 10,000 kilowatts will be assessed for each 1% decrease; the AGC availability rate should reach 100%, and 1 point per 10,000 kilowatts will be assessed for each 1% decrease. The installed capacity of this station is 50,000 kilowatts.

[0056] The centralized control side conducted a simulation, with a contribution power qualification rate of 88%, reaching 90%, a deviation of 2%, resulting in a deduction of 2 points for the contribution power qualification rate; the AGC availability rate was calculated based on a benchmark of 98%, with a target of 99%, a deviation of 1%, and the installed capacity of 50,000 kilowatts, resulting in a deduction of 5 points for the AGC availability rate.

[0057] The current real-time centralized control side assessment results for this station are formed by combining the 88% qualified rate of contributed electricity, the 98% availability rate of AGC, the 2-point deduction for the qualified rate of contributed electricity, and the 5-point deduction for the availability rate of AGC.

[0058] The cumulative indicator values ​​and deduction results for each time period are summarized and integrated to form the real-time centralized control side assessment results.

[0059] In some embodiments, please refer to Figure 1 The process of obtaining historical centralized control side assessment results and historical scheduling side assessment results, comparing them to obtain historical deviation data, and constructing assessment scenarios includes: The system retrieves historical centralized control side assessment results and input data for a specified date from the centralized control side. Simultaneously, it retrieves published historical dispatch side assessment results for the corresponding date from the dispatch side. The input data from the historical centralized control side includes time-stamped operational data and alignment data for the corresponding date. The historical centralized control side assessment results and historical dispatch side assessment results are compared according to the dimensions of station, assessment date, assessment item, indicator name, and time scale to obtain historical deviation data. The day is divided into several operating periods according to the changes in power grid supply and demand, including periods of high renewable energy generation, peak electricity consumption periods, and other periods; In each time period, the historical deviation value and historical deviation change rate of the historical deviation data are calculated. Based on the historical deviation value and historical deviation change rate and the input data from the historical centralized control side, cluster analysis is performed to obtain the assessment scenario.

[0060] On the centralized control side, obtain the historical centralized control side assessment results and the corresponding historical centralized control side input data for a specified date. On the dispatching side, obtain the published historical dispatching side assessment results for the corresponding date. The historical centralized control side assessment results and the historical dispatch side assessment results are compared according to the dimensions of station, assessment date, assessment item, indicator name and time scale. The deviation value between the two is calculated by finding the difference, and the historical deviation data is obtained. The day is divided into three periods based on grid supply and demand changes: peak renewable energy generation, peak electricity consumption, and other periods. For each period, the deviation value and deviation change rate are extracted from historical deviation data as two core feature dimensions. Combined with the historical centralized control side input data for the corresponding period, cluster analysis is performed on the historical data of each period. At the same time, the deviation value and deviation change rate are stored as core features of the assessment scenario. Specifically, the peak renewable energy generation period is from 09:00 to 15:00, which may be extended to 16:00 in summer; the peak electricity consumption period is from 18:00 to 22:00, which may be earlier to 17:00 in winter; adjustments can also be made according to actual supply and demand changes.

[0061] The deviation value and deviation change rate are used only for clustering and feature storage of historical scenarios, and not for real-time scenario matching, to avoid the difficulty of acquiring data in real-time processes. The deviation value is the difference between the historical centralized control side assessment results and the historical dispatch side assessment results within the specified time period; the deviation change rate is the rate of change of the deviation value over time, i.e., the increment of the deviation value per unit time; the input data feature vector includes the corresponding station operation data within the specified time period. Combining the historical centralized control side input data (including time-stamped operation data and aligned data) corresponding to the specified time period, a clustering algorithm is used for classification analysis. Time periods with similar deviation characteristics and input data characteristics are grouped into one category, resulting in several assessment scenarios. Each assessment scenario corresponds to a set of deviation value intervals and deviation change rate intervals, as well as a corresponding historical input data feature template.

[0062] For abnormal deviations with clearly identifiable causes (including only deviations caused by abnormal communication links on the central control side leading to interruption of real-time data acquisition and deviations caused by using fitted data as input data), the deviation causes for the corresponding assessment scenarios are marked separately. For assessment scenarios where the causes of deviations are not clearly identified, only the deviation characteristics and input data characteristics are retained.

[0063] At the end of each day, the dispatching side publishes the previous day's assessment results on a designated website. The centralized control side's automated data crawling unit automatically logs into the published website according to a preset time and designated account, completing the accurate crawling, parsing, and standardization conversion of historical assessment data for each new energy power station. The standardized historical assessment data is then pushed to the dual-end result comparison unit. This unit simultaneously retrieves the pre-calculated historical assessment results from the centralized control side's assessment calculation unit and the historical operation data of the power stations stored in the distributed database, forming a triple data input to complete the precise dimension-by-dimensional comparison of the dual-end results. The assessment deviation correction unit, based on the historical deviation data obtained from the comparison and combined with multi-dimensional influencing factors, completes the differential correction of the assessment results. At the same time, it completes system optimization based on the historical deviation data and historical operation data. The correction results and optimization parameters are fed back to the centralized control side's assessment calculation unit to optimize its real-time pre-calculation strategy, forming a data closed loop.

[0064] Because the scheduling side is a black box, only providing the final assessment results and not detailed data, deviation localization cannot be achieved; only deviation calculation of the final results from both ends can be performed.

[0065] The deviation calculation method uses a combination of the absolute deviation value method and the relative deviation rate method: Absolute deviation value = Historical centralized control side assessment result - Historical dispatch side assessment result; Deviation change rate = |Historical centralized control side assessment result - Historical dispatch side assessment result| ÷ Historical dispatch side assessment result × 100%; When the historical cycle is one day, the centralized control side extracts historical assessment results from its own assessment record system. These results represent the assessment of various operational indicators for each station on the previous day, serving as the centralized control side's preliminary evaluation of the station's operational performance. After completing its assessment of the previous day's dispatch instruction execution and other related matters, the dispatch side publishes historical dispatch side assessment results, which the centralized control side needs to obtain from the dispatch side. Historical station-side operational data is also obtained from the station side, including operating parameters of station equipment and environmental data.

[0066] After obtaining the above data, a detailed comparison is made between the historical centralized control side assessment results and the historical dispatching side assessment results according to multiple dimensions such as station, assessment date, assessment item, indicator name, and time scale. The data that differs between the two is identified as historical deviation data. The above steps are repeated for historical time periods (past 1 day, past 2 days... past n days) to continuously accumulate deviation data for different dates, forming a comprehensive historical deviation database.

[0067] In some embodiments, please refer to Figure 1 The process of correcting the real-time centralized control side assessment results based on preset multi-dimensional influencing factors and historical deviation data corresponding to the assessment scenario, to obtain the corrected assessment results on the real-time centralized control side, includes: Identify the runtime segment corresponding to the real-time centralized control side assessment results, match the corresponding assessment scenario based on the real-time centralized control side input data, and obtain several matching assessment scenarios. The real-time centralized control side input data includes real-time time-stamped runtime data and alignment data. Obtain multi-dimensional influencing factors, and perform weighted summation on the multi-dimensional influencing factors to obtain the preliminary corrected weights for each matched assessment scenario. The multi-dimensional influencing factors include time period weight, scenario similarity weight, and deviation degree weight. The time period weight is the weight of the running period, the scenario similarity weight is the weight of the similarity between the real-time centralized control side input data and the historical centralized control side input data, and the deviation degree weight is the weight of the historical deviation value and the historical deviation change rate corresponding to the assessment scenario. The deviation correction coefficient for each scenario is calculated based on the historical centralized control side assessment results and historical scheduling side assessment results corresponding to each matching assessment scenario. The deviation correction coefficient for each matched assessment scenario is weighted and calculated with the corresponding initial correction weight to obtain the real-time correction coefficient; The real-time centralized control side assessment results are corrected based on the real-time correction coefficient to obtain the real-time centralized control side corrected assessment results.

[0068] During real-time correction, the runtime segment corresponding to the current real-time centralized control side assessment result is identified, and the current real-time centralized control side input data is extracted. This data is then matched with the historical input data feature templates of each assessment scenario to obtain several matching assessment scenarios. After a successful match, the historical deviation value and historical deviation change rate corresponding to each matching assessment scenario are used as the deviation degree weight in the multi-dimensional influencing factors. These weights, along with the time segment weight and scenario similarity weight, are substituted into a weighted sum to calculate the preliminary correction weight for each matching assessment scenario. Finally, the deviation correction coefficient is calculated by combining the historical assessment results of each scenario, resulting in the final corrected assessment result for the real-time centralized control side.

[0069] Identify the current operating period (one of the three: peak renewable energy generation period, peak electricity consumption period, or other periods) corresponding to the real-time centralized control side assessment results, and match multiple assessment scenarios corresponding to the current period based on the input data from the real-time centralized control side. A preset multi-dimensional influencing factor is obtained, which includes time period weight, scenario similarity weight, and deviation degree weight. Among them, the time period weight is preset according to the importance of power grid supply and demand in each time period (peak electricity consumption period > new energy generation period > other time periods); the scenario similarity weight is obtained by calculating the similarity between the real-time centralized control side input data and the historical centralized control side input data of each matching assessment scenario, and the higher the similarity, the greater the weight; the deviation degree weight is set according to the historical deviation value and historical deviation change rate of each matching assessment scenario. Specifically, a weighting ratio is set, and the historical deviation value and historical deviation change rate are weighted and summed to obtain the deviation degree weight.

[0070] For example, the deviation degree weight is set to consist of two parts: deviation value weight and deviation change rate weight. The deviation value weight accounts for δ of the deviation degree weight, and the deviation change rate weight accounts for (1-δ) of the deviation degree weight, where δ is 0.6.

[0071] Scenario A: Historical deviation value is 2.3, and the deviation change rate is 0.38% / h; Scenario B: Historical deviation value is -1.8, and the deviation change rate is -0.30% / h.

[0072] After taking the absolute value of the deviation, multiply it directly by the corresponding proportional coefficient. The initial deviation weight for scenario A is 0.6×|2.3|+0.4×|0.38|=1.38+0.152=1.532; the initial deviation weight for scenario B is 0.6×|-1.8|+0.4×|-0.30|=1.08+0.12=1.2.

[0073] After calculation, the deviation weights of each scenario are normalized. The initial deviation weight of each scenario is divided by the sum of the deviation weights of all matched scenarios to obtain the final deviation weights: the sum of the initial deviation weights of all scenarios = 1.532 + 1.2 = 2.732; the normalized deviation weight of scenario A = 1.532 / 2.732 ≈ 0.56; the normalized deviation weight of scenario B = 1.2 / 2.732 ≈ 0.44.

[0074] That is, the final deviation weight for scenario A is 0.56, and the final deviation weight for scenario B is 0.44.

[0075] Based on the aforementioned multi-dimensional influencing factors, a weighted summation method is used to calculate the preliminary corrected weights for each matched assessment scenario. The calculation formula is as follows: ; Where α, β, and γ are preset weight coefficients, and α+β+γ=1; Based on the historical centralized control side assessment results and historical scheduling side assessment results corresponding to each matching assessment scenario, the deviation correction coefficient for each scenario is calculated, where, ; Among them, the historical centralized control side assessment result is not 0.

[0076] The deviation correction coefficient for each matched assessment scenario is weighted and calculated with the corresponding initial correction weight to obtain the real-time correction coefficient; The real-time centralized control side assessment result is multiplied by the real-time correction coefficient to obtain the real-time centralized control side corrected assessment result; if the historical centralized control side assessment result is 0, the historical deviation value of the corresponding scenario is used to directly correct the real-time centralized control side assessment result.

[0077] Since the scheduling side is a black box, it can only obtain the final scheduling result and the centralized control calculation result. Therefore, a purely result-driven adaptive correction is adopted.

[0078] For example, the centralized control side calculates a real-time assessment score of 12.5 for a wind farm. Based on the real-time input data, the centralized control side matches two assessment scenarios. The preset weighting coefficients are α = 0.3, β = 0.4, and γ = 0.3. The time period weight, scenario similarity weight, and deviation weight for each scenario are as follows: Scenario 1 has a time period weight of 0.6, a scenario similarity weight of 0.5, and a deviation weight of 0.7; Scenario 2 has a time period weight of 0.4, a scenario similarity weight of 0.6, and a deviation weight of 0.4. The initial correction weight for Scenario 1 is 0.3. 0.6 + 0.4 0.5 + 0.3 0.7 = 0.53; the initial adjusted weight for scenario two is 0.3. 0.4 + 0.4 0.6 + 0.3 0.4 = 0.45.

[0079] If the historical assessment results for scenario one are 10 points for the centralized control side and 9 points for the historical scheduling side, then the deviation correction coefficient for scenario one is 0.9. If the historical assessment results for scenario two are 8 points for the centralized control side and 7.2 points for the historical scheduling side, then the deviation correction coefficient for scenario two is 0.9.

[0080] The deviation correction coefficient for each scenario is weighted and calculated with the corresponding initial correction weight. The real-time correction coefficient is 0.53. 0.9 + 0.45 0.9 = 0.882. Multiplying the real-time centralized control side assessment result of 12.5 points by the real-time correction coefficient of 0.882, we get a corrected assessment result of 11.025 points for the real-time centralized control side. By correcting the deviation, the excessively high prediction deduction is adjusted back, making the actual result closer to the final assessment result of the scheduling side.

[0081] In some embodiments, please refer to Figure 1 The process of issuing graded early warnings based on the real-time centralized control side's correction assessment results, and displaying the key information corresponding to the graded early warnings, the real-time centralized control side's correction assessment results, and the time-stamped operational data from each station includes: The assessment results are corrected based on real-time centralized control and a tiered early warning system is implemented. Specifically, a risk warning is triggered when the deduction result exceeds the first threshold, and a deduction warning is triggered when the deduction result exceeds the second threshold. The visualization terminal displays key information corresponding to the graded early warning, real-time centralized control side correction assessment results, and time-stamped operation data from each station.

[0082] The hierarchical early warning and visualization unit enables integrated display of real-time and historical data. It displays the station's assessment indicators, early warning status, and cumulative indicator values ​​and deduction results for the time period when the centralized control side corrects the assessment results in real time. It also displays historical dispatch-side assessment data. At the same time, it supports multi-format export, multi-dimensional statistics, and full-process traceability of all data. Ultimately, it enables the centralized control side to achieve intensive monitoring, accurate early warning, and refined management of the grid-connected operation assessment of multiple new energy power stations.

[0083] When issuing tiered early warnings, corresponding thresholds are set for the deduction results in the real-time centralized control side's correction assessment. A risk warning is issued when the deduction result exceeds a pre-set first threshold, indicating a slight risk to the station's operation during that period. The deduction result includes scores for two dimensions: AGC availability rate and AGC contribution power qualification rate. The first threshold settings differ for each dimension. When the contribution power qualification rate deduction result is 2 points and the corresponding first threshold is 3 points, it indicates normal operation, and no risk warning is issued. When the AGC availability rate deduction result is 5 points and the corresponding first threshold is 4 points, it indicates an anomaly, and a risk warning (yellow warning) is issued.

[0084] The point deduction warning is triggered when the deduction result exceeds the second threshold. Point deductions typically reflect the degree to which the station violates two detailed rules during operation; higher deductions indicate more serious problems. For example, if a station incurs significant point deductions in its assessment due to inaccurate data collection, and the contribution power qualification rate deduction is 6 points while the corresponding second threshold is 10 points, this is considered normal and no point deduction warning is issued. However, if the AGC availability rate deduction is 20 points while the corresponding second threshold is 15 points, this is considered abnormal and a point deduction warning (i.e., a red warning) is issued to remind relevant personnel to pay attention to issues in the data collection process.

[0085] The visualization terminal will display key information corresponding to the graded early warning, including the warning type of risk warning or deduction warning, the name and specific value of the indicator that triggered the warning, the total assessment score, the score of each indicator, the real-time centralized control side correction of the assessment results, and the time-stamped operation data of each station and the corresponding assessment score and whether the warning has been triggered.

[0086] This invention proposes an integrated monitoring and early warning system for grid-connected operation assessment of multiple renewable energy power plants. Please refer to [link / reference]. Figure 2 and Figure 5 ,include: The alignment unit is configured to collect time-stamped operation data from each station and upload it to the central control and dispatching sides. Each station receives instructions from the dispatching side and marks the instructions on each station to obtain time-stamped dispatching instructions. The centralized control unit is configured to match time-stamped scheduling instructions with time-stamped running data on the centralized control side to obtain aligned data, and to calculate the aligned data to obtain real-time centralized control side assessment results. The correction unit is configured to obtain historical centralized control side assessment results and historical scheduling side assessment results for comparison to obtain historical deviation data, and correct the real-time centralized control side assessment results based on preset multi-dimensional influence factors and historical deviation data to obtain real-time centralized control side corrected assessment results. The early warning display unit is configured to issue graded early warnings based on the real-time centralized control side's correction assessment results, and to display the key information corresponding to the graded early warnings, the real-time centralized control side's correction assessment results, and the time-stamped operating data from each station.

[0087] like Figure 5 As shown, in practical applications, this invention is applied to the grid-connected operation system of multiple new energy power stations. The core architecture of the grid-connected operation system of multiple new energy power stations includes: a power station data access terminal (deployed in each new energy power station, with time stamping and coding function), an automated crawling unit for scheduling assessment data, a multi-source data centralized acquisition unit, an assessment rule base management unit, a centralized control side assessment calculation unit, a dual-end result comparison unit, an assessment deviation correction unit, a hierarchical early warning and visualization unit, a distributed database, and a visualization display terminal (including a centralized control screen / computer / mobile terminal).

[0088] The overall hierarchy of component connection relationships is as follows: Data access terminal at the site: Deployed at each wind power / photovoltaic / energy storage site, the terminal receives instruction data issued by the dispatch side in real time, collects local operation data / equipment status data of the site, and uploads the real-time time-stamped data to the multi-source data centralized acquisition unit after uniformly stamping the dispatch instructions.

[0089] Automated Data Crawling Unit for Dispatch Assessment: A dedicated unit on the central control side, independently connected to the official website for publishing dispatch assessment results. The unit automatically crawls, parses, and standardizes historical dispatch assessment data according to preset parameters, and pushes the standardized data to the dual-end result comparison unit. It can retrieve basic information about the stations from the database to achieve accurate data matching.

[0090] Multi-source data centralized acquisition unit: The core of data preprocessing on the centralized control side. The unit performs deduplication, completion, outlier removal, and standardization transformation on the received station data. It completes basic time alignment through time stamping. For data that cannot be fully aligned, it corrects it through intelligent algorithms. The processed standardized data is pushed to the assessment calculation unit on the centralized control side for assessment calculation. At the same time, all data is stored in a distributed database, and historical data can be retrieved as needed to supplement the calculation.

[0091] Assessment Rule Base Management Unit: The unit embeds two detailed rules and auxiliary service rules for each region, supports rule configuration, verification, version management and real-time invocation, provides standardized rules and calculation logic support for the assessment calculation unit on the centralized control side, and can fine-tune rule parameters based on the invocation feedback from the assessment calculation unit.

[0092] Centralized Control Side Assessment Calculation Unit: The core of the centralized control side assessment calculation. Based on standardized real-time data, the unit completes real-time pre-calculation of assessment indicators and deduction simulation, and pushes the real-time pre-calculation results one-way to the hierarchical early warning and visualization unit for real-time early warning; after completing the calculation of historical assessment results, it pushes them to the dual-end result comparison unit for post-event comparison; it receives the optimization parameters / correction results pushed by the assessment deviation correction unit to complete model adjustment and calculation strategy optimization; all calculation results are stored in a distributed database.

[0093] Dual-end result comparison unit: It can retrieve historical data from the distributed database. The unit receives three input data: historical scheduling assessment data pushed by the scheduling assessment data automatic crawling unit, historical control assessment results pushed by the centralized control side assessment calculation unit, and historical station operation data retrieved from the distributed database. It completes dual-end result comparison and deviation calculation by station-assessment date-assessment item-indicator name-time scale, and pushes the historical deviation data to the assessment deviation correction unit.

[0094] Assessment Deviation Correction Unit: The unit completes the differential correction and result verification of assessment results based on historical deviation data and multi-dimensional influencing factors. At the same time, it completes model optimization based on historical deviation data and historical operation data, and feeds back the correction results and optimization parameters to the assessment calculation unit on the centralized control side to realize closed-loop optimization of the model. All correction records and model correction parameters are stored in a distributed database.

[0095] Tiered early warning and visualization unit: The unit completes yellow / red level two real-time early warning based on the real-time pre-calculation results of the centralized control side, and pushes early warning information through multiple channels; at the same time, it integrates real-time assessment data, early warning status, scheduling crawled data, dual-end comparison results, and correction results to complete an integrated visualization display.

[0096] Distributed Database: The sole data storage center for the entire system, enabling unified storage and on-demand retrieval of all collected data, crawled data, assessment calculation results, historical deviation data comparisons, correction records, model modification parameters, and early warning information, providing data support for the entire process operation.

[0097] Visual display terminal: The terminal is operated and displayed by the management personnel on the central control side. It displays the station assessment status and early warning information in real time, and displays the comparison and correction results between the two ends in history. It supports the marking of early warning handling status and the entry of handling records.

[0098] The two detailed rules refer to the "Implementation Rules for Power Grid Connection Operation Management" and the "Implementation Rules for Power Ancillary Service Management" issued by the regional power regulatory departments. These rules serve as the core basis for assessing the grid connection operation and ancillary services of new energy power plants, covering all dimensions of assessment requirements, including active power regulation management, reactive power regulation management, and safety and operation management. They cover all aspects of the operation of new energy power plants, including active power, reactive power voltage, safe operation, and maintenance management, placing high demands on the refined management capabilities of the power plants. To ensure that the central control station can intuitively grasp the assessment risks of indicators such as AGC and power prediction for each new energy power plant, an early warning calculation engine based on the two detailed rules needs to be embedded. This engine should transform the real-time operational data collected by the central control platform into understandable indicators, scores, and risks. When any indicator approaches the assessment threshold or an abnormal event occurs that could lead to point deductions, the system should proactively issue an early warning and inform the user of the potential point deduction impact.

[0099] Centralized monitoring: This refers to abandoning the traditional independent monitoring mode of single stations at the station end and making unified deployments on the centralized control side to realize the centralized collection, unified calculation, global monitoring and large-scale management of grid-connected operation assessment data of multiple regions and types of new energy power stations, breaking down data silos and reducing overall construction and operation and maintenance costs.

[0100] AGC: Automatic Generation Control, refers to the control method by which a new energy power station automatically adjusts the output of its generating units according to instructions issued by the dispatching side, so that the actual power of the station tracks the target power. It is divided into wind and solar AGC and new energy storage AGC, and is the core assessment item of active power regulation management in the two detailed rules.

[0101] AVC: Automatic Voltage Control, refers to the control method by which a new energy power station automatically adjusts its reactive power compensation equipment according to the voltage curve issued by the dispatching side, so as to maintain the power station bus voltage within the specified range. It is the core assessment item of reactive power regulation management in the two detailed rules.

[0102] Primary frequency regulation: refers to the control method in which new energy generating units automatically adjust their active power to suppress grid frequency fluctuations without dispatching instructions when grid frequency fluctuates. Its response performance and disturbance compliance rate are the two core assessment indicators in the detailed rules.

[0103] Yellow / Red Level 2 Warning: This refers to a tiered warning mechanism based on the assessment red lines of the two detailed rules. The yellow warning is a risk warning, triggered when the indicator approaches the assessment red line, prompting maintenance personnel to investigate in a timely manner; the red warning is a deduction warning, triggered when the indicator exceeds the assessment red line / reaches the deduction standard, prompting maintenance personnel to take immediate action. All warning thresholds support flexible manual adjustment.

[0104] Automated crawling of dispatch assessment data: This refers to the process by which the dedicated unit on the centralized control side automatically logs into the official website for publishing dispatch assessment results according to preset parameters (login account / password, crawling time, target URL), and completes the accurate acquisition, parsing, and standardized conversion of historical assessment data for each new energy power station. This replaces manual querying and achieves automated and error-free data acquisition.

[0105] Dual-end multi-source historical post-event comparison: This refers to adapting to the actual situation of historical assessment results published by the dispatching side, using historical assessment data of dispatching, historical assessment results of centralized control, and historical operation data of the station as triple inputs, and completing the matching and deviation calculation of assessment results between the centralized control side and the dispatching side according to a unified dimension. "Dual-end" refers to the centralized control side and the dispatching side.

[0106] Data time stamping: This refers to the operation of adding time stamps to all real-time command data issued by the dispatch side at the station end according to a unified time standard. The purpose is to achieve basic time alignment between the data collected by the dispatch side and the data collected by the central control side, and to solve the problem of data time asynchrony.

[0107] Assessment deviation correction and model optimization: The process involves making differentiated adjustments to the pre-calculated assessment results on the centralized control side, and optimizing the assessment calculation model parameters based on historical deviation data and station historical operation data, so that the error between the corrected results and the dispatching side results is controlled within a preset range. The correction results and optimized parameters are then fed back to the assessment calculation unit on the centralized control side.

[0108] This invention addresses the issues of inadequate assessment systems at stations and isolated data deployments at individual stations. It employs a centralized control-side deployment architecture with direct connections and collaboration among seven units, enabling centralized data collection from multiple stations and resolving the problem of high costs associated with redundant construction. Benefits and effects: The centralized control side provides global oversight of the assessment status of multiple stations, achieving centralized management, reducing overall construction and maintenance costs, and improving the management efficiency of the centralized control layer.

[0109] A new automated data crawling unit for scheduling assessments replaces manual queries, providing triple data input for dual-end comparisons. Combined with time-stamped coding and intelligent correction, it optimizes the model based on deviations and feeds back to the calculation unit, achieving accurate comparisons and model closure. Effects and benefits: It solves the problems of low efficiency and misaligned data times in manual assessment data queries, eliminates human error, improves the accuracy of assessment comparisons and calculations, and accurately pinpoints the root causes of point deductions.

[0110] A two-tiered real-time early warning system (yellow / red) is constructed, adapting to all assessment items and adhering to the detailed assessment rules. Thresholds are manually adjustable, and early warning information includes economic losses. Simultaneously, it achieves integrated display and full data traceability of real-time / historical data and assessment data. Effects and benefits: Proactive identification of assessment risks enables handling, reducing point deductions and losses; comprehensive data support for site operation optimization improves overall economic efficiency.

[0111] Based on the same inventive concept, according to another aspect of the present invention, such as Figure 3 As shown, an embodiment of the present invention also provides a computer device 30, which includes a processor 310 and a memory 320. The memory 320 stores a computer program 321 that can be run on the processor. When the processor 310 executes the program, it performs the steps of the method described above.

[0112] Based on the same inventive concept, according to another aspect of the present invention, such as Figure 4 As shown, embodiments of the present invention also provide a computer-readable storage medium 40, which stores a computer program 410 that, when executed by a processor, performs the methods described above.

[0113] Embodiments of the present invention may also include a corresponding computer device. The computer device includes a memory, at least one processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes any of the methods described above when executing the program.

[0114] The memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions / modules in the embodiments of this application. The processor executes various functional applications and data processing of the device by running the non-volatile software programs, instructions, and modules stored in the memory, thereby implementing the above-described method.

[0115] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the device. Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the local module via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] Finally, it should be noted that those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The storage medium for the program can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. The above computer program embodiments can achieve the same or similar effects as any of the corresponding foregoing method embodiments.

[0117] Those skilled in the art will also understand that the various exemplary logic blocks, modules, circuits, and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the functionality of various illustrative components, blocks, modules, circuits, and steps has been generally described. Whether this functionality is implemented as software or as hardware depends on the specific application and the design constraints imposed on the system as a whole. Those skilled in the art can implement the functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the embodiments disclosed herein.

[0118] The above are exemplary embodiments disclosed in this invention. However, it should be noted that various changes and modifications can be made without departing from the scope of the embodiments of this invention as defined by the claims. The functions, steps, and / or actions of the methods according to the disclosed embodiments described herein do not need to be performed in any particular order. The sequence numbers of the disclosed embodiments of this invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. Furthermore, although the elements disclosed in the embodiments of this invention may be described or claimed individually, they may be understood as multiple unless explicitly limited to a singular number.

[0119] It should be understood that, as used herein, the singular form "one" is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all combinations including one or more of the associated listed items.

[0120] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples. Within the framework of the invention, technical features of the above embodiments or different embodiments can be combined, and many other variations of different aspects of the invention exist, which are not provided in the details for the sake of brevity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A method for intensive monitoring and early warning of grid-connected operation assessment of multiple renewable energy power plants, applied in a centralized control system, characterized in that: include: The time-stamped operation data of each station is collected and uploaded to the central control side and the dispatch side. Each station receives the instructions from the dispatch side and marks the instructions on each station to obtain time-stamped dispatch instructions. On the centralized control side, the time-stamped scheduling instructions are matched with the time-stamped running data to obtain aligned data, and the aligned data is calculated to obtain the real-time centralized control side assessment results. Historical assessment results from the centralized control side and the historical dispatch side are compared to obtain historical deviation data and construct assessment scenarios. Specifically, on the centralized control side, historical assessment results and input data for specified dates are obtained; on the dispatch side, published historical assessment results for corresponding dates are obtained. The input data from the historical centralized control side includes time-stamped operational data and alignment data for the corresponding dates. The historical assessment results from the centralized control side and the historical dispatch side are compared according to the dimensions of station, assessment date, assessment item, indicator name, and time scale to obtain historical deviation data. A day is divided into several operating periods based on grid supply and demand changes, including periods of high renewable energy generation, peak electricity consumption, and other periods. In each period, the historical deviation value and historical deviation change rate are calculated. Cluster analysis is performed based on the historical deviation value, historical deviation change rate, and historical centralized control side input data to obtain assessment scenarios. Based on preset multi-dimensional influencing factors and the historical deviation data corresponding to the assessment scenarios, the real-time centralized control side assessment results are corrected to obtain corrected real-time centralized control side assessment results. In this process, the runtime segment corresponding to the real-time centralized control side assessment results is identified. Based on the real-time centralized control side input data, corresponding assessment scenarios are matched to obtain several matched assessment scenarios. The real-time centralized control side input data includes real-time time-stamped runtime data and alignment data. Multi-dimensional influencing factors are obtained, and a weighted sum is performed on these factors to obtain the preliminary correction weights for each matched assessment scenario. These multi-dimensional influencing factors include time period weight, scenario similarity weight, and deviation degree weight. The time period weight is the weight of the runtime segment; the scenario similarity weight is the weight of the similarity between the real-time centralized control side input data and the historical centralized control side input data; and the deviation degree weight is the weight of the historical deviation value and historical deviation change rate corresponding to the assessment scenario. The deviation correction coefficient for each scenario is calculated based on the corresponding historical centralized control side assessment results and historical scheduling side assessment results in each matched assessment scenario. The deviation correction coefficient for each matched assessment scenario is weighted with the corresponding preliminary correction weight to obtain the real-time correction coefficient. The real-time centralized control side assessment results are then corrected based on the real-time correction coefficient to obtain the corrected real-time centralized control side assessment results. Based on the real-time centralized control side correction assessment results, graded early warnings are issued, and the key information corresponding to the graded early warnings, the real-time centralized control side correction assessment results, and the time-stamped operation data of each station are displayed.

2. The method for intensive monitoring and early warning of grid-connected operation assessment for multiple new energy power plants according to claim 1, characterized in that, The process of collecting time-stamped operational data from each station and uploading it to the central control and dispatching sides includes: Collect operational data from each station; A unified time stamp is applied to the operational data at each station to obtain time-stamped operational data. The time-stamped operational data is uploaded from the station side to the central control side and the dispatch side; On the centralized control side, the time-stamped runtime data is deduplicated, completed, outlier removed, and normalized before being stored in the distributed database.

3. The method for intensive monitoring and early warning of grid-connected operation assessment for multiple new energy power plants according to claim 2, characterized in that, The process of matching time-stamped scheduling instructions with time-stamped running data on the centralized control side to obtain aligned data includes: The central control unit receives time-stamped dispatch instructions sent by the station side; In response to time stamp anomalies, the time-stamped scheduling instructions are corrected to obtain scheduling instructions with normal time stamps. The time stamp anomalies include out-of-order time stamp anomalies and missing time stamp anomalies. The scheduling instructions corresponding to out-of-order time stamp anomalies are reordered and put back in place, and the missing time stamp anomalies are filled in by the LSTM timing prediction method. In response to a normal time stamp, the time-stamped scheduling instructions are matched with time-stamped running data on the centralized control side. The distributed database is retrieved based on the time stamp to obtain the corresponding running data. The scheduling instruction data and the running data are then bound together to obtain aligned data. When multiple scheduling instruction data exist under the same time point, the optimal and latest scheduling instruction data is selected based on the principles of optimal data quality and latest time.

4. The method for intensive monitoring and early warning of grid-connected operation assessment for multiple new energy power plants according to claim 1, characterized in that, The process of calculating the alignment data to obtain the real-time centralized control side assessment results includes: On the centralized control side, the aligned data is calculated for each time period according to the preset calculation rules to obtain the cumulative index value for each time period; Obtain the preset assessment rules, perform deductions on the cumulative indicator values ​​over a period of time, and obtain the deduction results; The cumulative indicator values ​​and deduction results for each time period are summarized into the real-time centralized control side assessment results.

5. The method for intensive monitoring and early warning of grid-connected operation assessment for multiple new energy power plants according to claim 4, characterized in that, The process of issuing graded early warnings based on the real-time centralized control side's correction assessment results, and displaying the key information corresponding to the graded early warnings, the real-time centralized control side's correction assessment results, and the time-stamped operational data from each station includes: The assessment results are corrected based on real-time centralized control and a tiered early warning system is implemented. Specifically, a risk warning is triggered when the deduction result exceeds the first threshold, and a deduction warning is triggered when the deduction result exceeds the second threshold. The visualization terminal displays key information corresponding to the graded early warning, real-time centralized control side correction assessment results, and time-stamped operation data from each station.

6. An integrated monitoring and early warning system for grid-connected operation assessment of multiple new energy power plants, characterized in that: include: The alignment unit is configured to collect time-stamped operation data from each station and upload it to the central control and dispatching sides. Each station receives instructions from the dispatching side and marks the instructions on each station to obtain time-stamped dispatching instructions. The centralized control unit is configured to match time-stamped scheduling instructions with time-stamped running data on the centralized control side to obtain aligned data, and to calculate the aligned data to obtain real-time centralized control side assessment results. The correction unit is configured to compare historical centralized control side assessment results with historical dispatch side assessment results to obtain historical deviation data and construct assessment scenarios. Specifically, it acquires historical centralized control side assessment results and input data for a specified date on the centralized control side, and acquires published historical dispatch side assessment results for the corresponding date on the dispatch side. The input data from the historical centralized control side includes time-stamped operational data and alignment data for the corresponding date. The unit compares the historical centralized control side assessment results with the historical dispatch side assessment results according to the dimensions of station, assessment date, assessment item, indicator name, and time scale to obtain historical deviation data. It divides a day into several operating periods based on grid supply and demand changes, including periods of high renewable energy generation, peak electricity consumption, and other periods. In each period, it calculates the historical deviation value and historical deviation change rate of the historical deviation data, performs cluster analysis based on the historical deviation value, historical deviation change rate, and historical centralized control side input data to obtain assessment scenarios. Finally, it corrects the real-time centralized control side assessment results based on preset multi-dimensional influencing factors and the historical deviation data corresponding to the assessment scenarios, obtaining the real-time centralized control side corrected assessment results. The process involves several steps: identifying the runtime segment corresponding to the real-time centralized control side assessment results; matching the corresponding assessment scenarios based on the real-time centralized control side input data to obtain several matching assessment scenarios; obtaining multi-dimensional influence factors; weighting and summing these factors to obtain preliminary correction weights for each matching assessment scenario; the multi-dimensional influence factors include time period weight, scenario similarity weight, and deviation degree weight; the time period weight is the weight of the runtime segment; the scenario similarity weight is the weight of the similarity between the real-time centralized control side input data and the historical centralized control side input data; and the deviation degree weight is the weight of the historical deviation value and historical deviation change rate corresponding to the assessment scenario; calculating the deviation correction coefficient for each scenario based on the corresponding historical centralized control side assessment results and historical scheduling side assessment results in each matching assessment scenario; weighting the deviation correction coefficient of each matching assessment scenario with the corresponding preliminary correction weight to obtain the real-time correction coefficient; and correcting the real-time centralized control side assessment results based on the real-time correction coefficient to obtain the corrected real-time centralized control side assessment results. The early warning display unit is configured to issue graded early warnings based on the real-time centralized control side's correction assessment results, and to display the key information corresponding to the graded early warnings, the real-time centralized control side's correction assessment results, and the time-stamped operating data from each station.

7. A computer device, comprising: At least one processor; The processor includes a memory storing a computer program that can run on the processor, characterized in that the processor executes the program by performing the steps of the intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it performs the steps of the intensive monitoring and early warning method for grid-connected operation assessment of multiple new energy power plants as described in any one of claims 1 to 5.

Images