A new energy power generation loss electric quantity accurate accounting method and system

Abstract

This invention relates to the field of energy management technology, specifically to a method and system for accurately calculating the power loss of new energy power generation. The method includes: simultaneously collecting data from generator sets at new energy power plants in multiple dimensions to obtain generator set operation datasets; based on the operation datasets, conducting command limit gap assessments, maintenance failure loss measurement, and fault source tracing to obtain explicit power curtailment, maintenance, and fault loss; selecting historical windows with no power curtailment and no faults to analyze the power group correlation of generator sets, completing wake effect correction during power curtailment periods, and constructing a wake correction reference benchmark; based on the wake correction reference benchmark, measuring the power curtailment gains and losses of restricted generator sets, and calculating the implicit power curtailment loss; during the period of implicit power curtailment, calculating the incremental loss of power regulation of restricted generator sets, and updating the wake correction reference benchmark; calculating compensation for subsequent implicit power curtailment based on the updated benchmark, merging multiple loss data, and statistically analyzing the total generalized power curtailment loss of new energy power plants.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, and in particular to a method and system for accurately calculating the power loss of new energy power generation. Background Technology

[0002] The large-scale and clustered development and application of new energy power plants has become the core development direction of the low-carbon transformation of the power system. As the core power generation equipment of new energy power plants, the long-term stable operation of wind turbine generators directly determines the power generation efficiency of the power plant and the quality of power grid supply and demand balance regulation. At present, new energy power plants have widely deployed various types of sensing and data acquisition equipment and online monitoring systems, which can continuously collect multi-dimensional operating data such as active power, speed, temperature and humidity, electrical parameters and wind conditions of the power plant, providing basic data support for the quantification of power generation losses and the management and control of unit operation status.

[0003] Existing power generation loss accounting models for renewable energy power plants are limited in scope, only capable of statistically analyzing power losses in explicit scenarios such as direct-controlled power curtailment, shutdown for maintenance, and equipment failure. They fail to address implicit influencing factors such as wind-induced interference and the interconnectedness of unit cluster operations. For instance, in a scenario where upstream and downstream units are arranged within the same wind farm, airflow obstruction caused by the output control behavior of upstream units can continuously reduce the power generation capacity of downstream units. Such implicit output losses are often overlooked in statistics, resulting in a one-sided and distorted measurement of overall power generation losses at the power plant. Furthermore, traditional wake effect analysis uses fixed static parameter modeling methods without dynamically adapting to and correcting real-time power fluctuations, frequent adjustment behaviors, and equipment fatigue degradation characteristics. For example, high-frequency power adjustments under long-term power curtailment conditions can lead to mechanical performance degradation, thereby altering the actual wind energy capture capacity of the units. Static benchmark models cannot adapt to the dynamic changes in unit operating conditions, resulting in large deviations in output benchmark judgments and insufficient accuracy in loss quantification, making it difficult to meet the practical application needs of refined power management and loss tracing for renewable energy power plants. Summary of the Invention

[0004] This invention provides a method and system for accurately calculating the power loss of new energy power generation, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for accurately calculating the power loss of new energy power generation, comprising: S1, performing multi-dimensional synchronous data collection on the generator sets in the new energy power station to obtain the generator set operation dataset; S2. Simultaneously perform instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset to obtain the explicit curtailment power, maintenance loss power and fault loss power of the generator set, respectively. S3. In the historical window of no power curtailment and no faults in the running dataset, the power behavior of the generator set is analyzed by group correlation. During the period when the restricted generator set enters the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for the new energy power station. S4. Using the wake correction reference benchmark as the basis for judgment, measure the power curtailment loss of the centrally constrained generator units in the operating data to obtain the implicit power curtailment loss of the new energy power station. S5. During the period when the restricted generator unit is affected by implicit curtailment, perform high-frequency incremental loss conversion on the power regulation record of the restricted generator unit, and update the wake correction reference benchmark based on the conversion result. S6. Based on the updated wake correction reference benchmark, the implicit curtailment loss of electricity in subsequent periods is compensated and calculated. The calculation results are then combined with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the new energy power station.

[0006] In a preferred embodiment, the process of obtaining the generator set's operating dataset is as follows: The initial data stream of generator units in the new energy power station is time-series aligned and checked to obtain the synchronization data sequence of the generator units. Based on the synchronous data sequence, the active power signal of the generator set and the wind speed and direction signal of the wind measuring tower at the station are decoupled and mapped by information entropy to obtain the equivalent wind condition sequence of the generator set. Based on the equivalent wind condition sequence of the generator set, the operating status of the generator set is identified by a multi-threshold hierarchical method to obtain the operating status label of the generator set. By performing multi-source time-series fusion of synchronous data sequences, equivalent wind condition sequences of generating units, and operating condition label sets, an operational dataset for new energy power plants is obtained.

[0007] In a preferred embodiment, the process of obtaining the explicit power curtailment loss, maintenance loss, and fault loss of the generator set is as follows: By identifying deviation periods in the scheduling instruction records and actual power generation files of the generating units in the operation dataset, the power curtailment period markers of the new energy power plants are obtained; Based on the instruction deviation time period mark, the power generation gap of the generator set is differentially aggregated to obtain the explicit power curtailment loss of the generator set. Based on the maintenance records and unit status logs in the operational dataset, the unavailable periods of the generator set are extracted to obtain the maintenance downtime calibration of the generator set. Based on the maintenance and shutdown period calibration, the benchmark capacity of the generator set is extracted within the historical window of no power curtailment and no faults to obtain the normal output benchmark of the generator set. Based on the normal output benchmark, the power loss of the generator units during the outage period is recursively compensated to obtain the maintenance power loss of the new energy power station. Based on the fault alarm codes and state change events in the operational dataset, fault attribution and source analysis are performed on the abnormal operation events of the generator set to obtain the fault responsibility time period of the generator set. Based on the division of fault responsibility periods, the power loss of the generator set during the fault period is accumulated by fault deficit integral to obtain the power loss of the generator set due to fault.

[0008] In a preferred embodiment, the process of obtaining the wake correction reference for the renewable energy power station is as follows: In the historical window of no power curtailment and no faults in the running dataset, the power fluctuations between generator sets are analyzed in a coordinated manner to obtain the directional dependence strength coefficient of the generator sets. Based on the directed dependency strength coefficient, a topological association mapping is performed on the generator set to obtain the body dependency topological network of the generator set; Using the free-flow state power of generator units in the unit group dependency topology network as a reference, the benchmark output of generator units is calibrated to obtain the dynamic behavior reference benchmark of new energy power stations. During the period when the restricted generator units enter the power curtailment state, the assessment results are corrected for wake effect to obtain the wake correction reference benchmark for new energy power stations.

[0009] In a preferred embodiment, the process of obtaining the wake correction reference for the renewable energy power station is as follows: Based on the power curtailment command records in the operational dataset, the constrained generator units in the operational dataset are identified, and the downstream impact links of the constrained generator units are traced level by level according to the machine body dependency topology network to obtain the wake-affected generator units of the constrained generator units. During the period when the restricted generator unit enters the power curtailment state, the virtual free flow wake field of the restricted generator unit is inverted and reconstructed to obtain the virtual free flow wake distribution of the restricted generator unit. Based on the virtual free-flow wake distribution, the equivalent wind speeds of the wind turbine surfaces of the downstream generator units affected by the wake are inverted to obtain the equivalent wind speed sequence of the wind turbine surfaces of the downstream generator units. Based on the equivalent wind speed sequence of the wind turbine surface, the expected power output of downstream generator units in the dynamic behavior reference benchmark is corrected for the wake effect, thus obtaining the wake correction reference benchmark for new energy power plants.

[0010] In a preferred embodiment, the process of obtaining the implicit power curtailment loss of renewable energy power plants is as follows: Using the wake correction reference standard as the basis for judgment, the restricted period of the restricted generator unit is identified to obtain the implicit curtailment period of the restricted generator unit. Based on the implicit power curtailment period, the expected power output of the restricted generator units in the wake correction reference benchmark is compared with the actual power output point by point to obtain the implicit power deficit sequence of the restricted generator units. Based on the implicit power deficit sequence, the deficit time-domain integral of the constrained generator set is performed to obtain the implicit power curtailment loss of the constrained generator set. The implicit power curtailment loss is aggregated across the entire region to obtain the implicit power curtailment loss of new energy power plants.

[0011] In a preferred embodiment, the process of updating the wake correction reference based on the calculation result is as follows: During the period when constrained generator units are affected by implicit power curtailment, amplitude features are extracted from the power regulation records of constrained generator units to obtain the regulation amplitude sequence of constrained generator units. The fatigue equivalent aggregation of the regulation amplitude sequence is used to obtain the cumulative fatigue damage increment of the constrained generator set. Based on the cumulative fatigue damage increment, the performance degradation degree of the restricted generator set is evaluated, and the active power output attenuation coefficient of the restricted generator set is obtained. Based on the active power output attenuation coefficient, the expected power output of the restricted generator unit in the wake correction reference is corrected by the reference capacity attenuation to obtain the updated wake correction reference.

[0012] In a preferred embodiment, the process of compensating for the implicit power curtailment loss in subsequent time periods is as follows: Based on the updated wake correction reference, the compensation period for the subsequent time period of the restricted generator unit is determined, and the compensation calculation period for the restricted generator unit is obtained. Based on the compensation calculation period, the expected power output in the updated wake correction reference benchmark is extracted to obtain the compensation expected power output sequence of the constrained generator set. By comparing the expected power output of the compensation-expected sequence with the actual power output of the restricted generator units, the uncompensated power deficit sequence of the restricted generator units is obtained. The compensation deficit sequence is integrated to obtain the compensation calculation power of the restricted generator units.

[0013] In a preferred embodiment, the process for obtaining the total generalized curtailment loss of renewable energy power plants is as follows: The compensated electricity volume, explicit curtailment loss volume, maintenance loss volume, and fault loss volume are uniformly calibrated for multi-source losses to obtain the set of calibrated loss components for new energy power plants. The total generalized curtailment loss of new energy power plants is obtained by summarizing and merging the calibrated loss component sets.

[0014] To address the aforementioned issues, this invention also provides a system for accurately calculating power loss during new energy power generation. The system includes: an operation data acquisition module, which performs multi-dimensional synchronous acquisition of generator sets in new energy power plants to obtain the operation dataset of the generator sets. The three loss accounting modules simultaneously perform instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset, and obtain the explicit curtailment power loss, maintenance loss power and fault loss power of the generator set, respectively. The wake correction benchmark construction module performs group correlation analysis on the power behavior of generator sets in the historical window of no power curtailment and no faults in the running dataset. During the period when the restricted generator sets enter the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for new energy power stations. The implicit curtailment loss calculation module uses the wake correction reference benchmark as the basis for judgment to measure the curtailment loss of the centrally constrained generator units in the operating data, and obtains the implicit curtailment loss of the new energy power station. The wake reference update module performs high-frequency incremental loss calculation on the power regulation records of the restricted generator units during the period when the restricted generator units are affected by implicit curtailment, and updates the wake correction reference reference based on the calculation results. The generalized curtailment loss calculation module compensates for the implicit curtailment loss in subsequent periods based on the updated wake correction reference benchmark. It then merges the calculation results with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the renewable energy power station.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Time-series alignment verification and anomaly screening are performed on multi-dimensional raw monitoring data of generator units. Information entropy decoupling mapping is used to remove interference components from wind conditions and the unit's own operation. Multi-threshold hierarchical identification is used to complete accurate classification and labeling of operating conditions. Multiple types of time-series data are integrated to form a unified standardized operation dataset. Then, through precise time period division, differential rolling accumulation and integral calculation mode, the power loss corresponding to explicit curtailment, maintenance shutdown, and equipment failure is calculated independently. This breaks through the limitations of narrow statistical dimensions in traditional accounting models, unifies the time benchmark and accounting caliber of all data, and eliminates the problems of vague classification, coarse calculation and missing data of explicit loss under single statistical methods. It can perform fine-grained decomposition and quantitative calculation of various explicit power generation losses, comprehensively covering the conventional loss statistics scenarios of power plants, and providing accurate data basis for the standardized management of losses in the power energy field.

[0016] 2. By screening for interference-free and stable operation windows, analyzing the power coordination fluctuation patterns among units, quantifying the directional dependency coefficients, and building a unit topology network, the influence transmission links between upstream and downstream units are accurately identified. For units operating under power curtailment, the virtual wake field is reconstructed through inversion. Wind speed attenuation correction is completed by combining spatial distance and relative orientation. The theoretical expected output value of downstream units is corrected step by step, and a dynamic wake correction reference benchmark is constructed. This method abandons the outdated approach of static modeling of wake effects with fixed parameters, effectively quantifying hidden interference factors such as wind farm unit cluster layout, airflow obstruction, and wake coupling linkage. It accurately identifies the hidden output loss of downstream units caused by power curtailment control of upstream units, solves the long-standing industry pain point of missed statistics and difficulty in tracing the source of hidden power curtailment loss, improves the full-dimensional metering system of power generation loss in the station, and enhances the comprehensiveness of power loss assessment under complex wind conditions and cluster operation scenarios.

[0017] 3. Collect full power regulation records of the unit under implicit curtailment conditions, extract regulation amplitude characteristics, and combine them with mechanical fatigue parameters to complete damage equivalent aggregation. Match the corresponding output attenuation coefficient according to the fatigue damage level, dynamically iterate and optimize the wake correction reference benchmark, and combine the actual operating state after equipment performance degradation to carry out output deviation discrimination and loss compensation calculation for subsequent operating periods. This makes up for the shortcomings of traditional fixed output benchmarks that cannot adapt to the dynamic operating changes of the unit. It fully combines the actual operating condition changes such as high-frequency power regulation, component fatigue aging, and gradual performance degradation, continuously corrects the theoretical output judgment standard, reduces the calculation deviation caused by static models, and coordinates short-term implicit curtailment losses and long-term power generation losses caused by equipment degradation. It realizes the integrated calculation of all types of losses of generalized curtailment. The measurement results are highly consistent with the actual operating status of new energy power stations and meet the actual needs of power stations for refined loss control and efficient operation. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for accurately calculating power loss in new energy power generation, provided in an embodiment of the present invention. Figure 2 A functional module diagram of a new energy power generation loss accurate calculation system provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This application provides a method for accurately calculating the power loss of new energy power generation. The executing entity of the method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server and a terminal. In other words, the method for accurately calculating the power loss of new energy power generation can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating a method for accurately calculating power loss in new energy power generation according to an embodiment of the present invention. In this embodiment, the method for accurately calculating power loss in new energy power generation includes: S1. Perform multi-dimensional synchronous data collection on the generator sets in the new energy power station to obtain the generator set operation dataset; In this embodiment of the invention, the process of obtaining the generator set's operating dataset is as follows: The initial data stream of generator units in the new energy power station is time-series aligned and checked to obtain the synchronization data sequence of the generator units. The initial data stream of the generator sets at new energy power plants consists of all the original operational monitoring data collected in real time during generator set operation, including active power data, speed data, temperature data, voltage data, current data, wind speed data, and wind direction data collected by the wind measurement tower at the power plant. All initial data streams are time-matched according to a unified time scale, adjusting data collected by different acquisition devices at different time points to the same time reference, and eliminating abnormal data with mismatched time scales. Valid data with completely consistent time dimensions are retained, ultimately forming a synchronous data sequence of the generator sets. The synchronous data sequence is a set of generator set operation data in which all data are arranged in an orderly manner on the same time axis after time sequence alignment and verification, and there is no time deviation.

[0022] Based on the synchronous data sequence, the active power signal of the generator set and the wind speed and direction signal of the wind measuring tower at the station are decoupled and mapped by information entropy to obtain the equivalent wind condition sequence of the generator set. The active power signal in the synchronous data sequence is the real-time active power output data of the generator set. The wind speed and direction signals from the wind measurement tower at the power station are the real-time wind speed and direction data collected in the synchronous data sequence. First, the discrete state distribution of the active power signal in the synchronous data sequence is extracted, and then the discrete state distribution of the wind speed and direction signals is extracted. Through the correlation decomposition of the state distribution, the part of the active power signal affected by wind speed and direction is separated from the part affected by the unit's own operating characteristics. Then, the separated wind speed and direction affected parts are combined with the unit's operating status to form the wind condition correlation data of the corresponding generator set. Finally, the equivalent wind condition sequence of the generator set is obtained. The equivalent wind condition sequence of the generator set is an ordered data set that can reflect the actual degree of wind condition influence on the generator set.

[0023] Based on the equivalent wind condition sequence of the generator set, the operating status of the generator set is identified by a multi-threshold hierarchical method to obtain the operating status label of the generator set. Based on the equivalent wind condition sequence of the generator set, the first-level operating status threshold is divided into two categories: normal operation and abnormal operation. Then, the second-level threshold is divided for the normal operation status, which is further subdivided into full-power operation, partial-output operation, and low-output operation. The second-level threshold is also divided for the abnormal operation status, which is further subdivided into power curtailment operation, fault shutdown, and maintenance shutdown. The operating status is accurately distinguished step by step, and finally the operating condition status label of the generator set is obtained. The operating condition status label is the identification data that directly reflects the operating status of the generator set during each operating period.

[0024] The synchronous data sequence, the unit equivalent wind condition sequence, and the operating condition label set are fused together using multi-source time series to obtain the operation dataset of the new energy power station. Based on a unified timeline, the basic operating data of the synchronous data sequence, the wind condition correlation data of the equivalent wind condition sequence of the generator unit, and the status identification data of the operating status label set are spliced ​​together one by one according to the same time node to form a complete data set containing multi-dimensional information and with completely unified time dimension. Finally, the operating dataset of the new energy power station is obtained. The operating dataset is a complete set of operating data of the generator unit of the new energy power station covering the basic operating data of the generator unit, the wind condition impact data, and the operating status data.

[0025] S2. Simultaneously perform instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset to obtain the explicit power curtailment loss, maintenance loss and fault loss of the generator set, respectively. In this embodiment of the invention, the process of obtaining the explicit power curtailment loss, maintenance loss, and fault loss of the generator set is as follows: By identifying deviation periods in the scheduling instruction records and actual power generation files of the generating units in the operation dataset, the power curtailment period markers of the new energy power plants are obtained; The operational dataset contains dispatch instruction records and actual power generation files of the generator units at new energy power plants. The dispatch instruction records are the time-series data of power control instructions issued by the grid dispatching terminal to the generator units, while the actual power generation files are the continuously collected time-series data of the actual output power of the generator units. The dispatch instruction records and the actual power generation files are aligned on a unified time axis. After alignment, the power instruction values ​​and actual power generation values ​​are compared hourly. When the power instruction value is consistently greater than the actual power generation value, and the duration of this continuous state reaches the minimum length standard for the conventional power curtailment judgment of the power plant, the continuous period is determined to be a power curtailment period. All determined power curtailment periods are integrated in chronological order to form the power curtailment period markers for new energy power plants. The power curtailment period markers are a complete set of time intervals that indicate that the generator units are in a power curtailment operation state.

[0026] Based on the instruction deviation time period mark, the power generation gap of the generator set is differentially aggregated to obtain the explicit power curtailment loss of the generator set. The calculation scope is defined by the time interval corresponding to the power curtailment period. Within this range, the expected and actual power output of the generator units are extracted hourly. The expected power output is directly taken from the power command value in the dispatch instruction record, and the actual power output is directly taken from the actual output power value in the unit's actual power output file. The calculation is performed point-by-point using a fixed time step, which is the minimum time interval for data acquisition at the power station. The difference between the expected and actual power output within each time step is accumulated. The formula for calculating the explicit power curtailment loss is: E1=Σ(P0-P In the formula E1×Δt, E1 represents the explicit power curtailment loss, Σ represents the summation of the calculation results for all time steps within the calculation period, P0 represents the generating power of the generator unit, and P Δt represents the actual power generated by the generator set, Δt represents the fixed time step used in the calculation, and the explicit power curtailment loss is the direct power curtailment loss caused by the generator set being affected by the dispatch command limit.

[0027] Based on the maintenance records and unit status logs in the operational dataset, the unavailable periods of the generator set are extracted to obtain the maintenance downtime calibration of the generator set. The operational dataset contains generator set maintenance records and generator set status logs. Maintenance records are data recording the time and operating status of planned maintenance and shutdowns of the generator set. Generator set status logs are continuous monitoring data recording the real-time operating status of the generator set. Time intervals marked as shutdowns and maintenance are extracted from the maintenance records, and time intervals in which the generator set is in an inoperable state are extracted from the generator set status logs. The two types of time intervals are matched for overlap, and the time periods that completely overlap and are continuous are retained. These time periods are the unavailable periods of the generator set. All unavailable periods are organized according to the time axis to form the maintenance shutdown period label of the generator set. The maintenance shutdown period label is a set of time intervals that mark the generator set in an inoperable state due to maintenance work.

[0028] Based on the maintenance and shutdown period calibration, the benchmark capacity of the generator set is extracted within the historical window of no power curtailment and no faults to obtain the normal output benchmark of the generator set. The no-waste and no-fault historical window is the historical data range of continuous normal operation of generator sets after removing power curtailment and fault periods from the operational data. The duration of this window is no less than the normal operating cycle of the generator set. With the maintenance shutdown period as a reference, all normal operation power generation data of the corresponding generator set are extracted from the no-waste and no-fault historical window, and the output power value of the generator set is collected at each time. All the collected output power values ​​are calculated as the arithmetic mean, which is the normal output benchmark of the generator set. The normal output benchmark is the standard output power value of the generator set under the condition of no-waste and no fault.

[0029] Based on the normal output benchmark, the power loss of the generator units during the outage period is recursively compensated to obtain the maintenance power loss of the new energy power station. Using the time interval specified during maintenance outages as the calculation scope, the normal output benchmark and the actual output power of the generator units during the outage period are obtained moment by moment. The calculation is performed point by point using a fixed time step. The difference between the normal output benchmark and the actual output power within each time step is accumulated. The formula for calculating maintenance loss is E2=Σ(P -P )×Δt, where E2 represents the maintenance loss electricity, Σ represents the summation of the calculation results for all time steps within the accounting period, and P P represents the normal output reference of the generator set. Δt represents the actual power output of the generator set, and Δt represents the fixed time step used in the calculation. The maintenance loss power of the new energy power station is obtained by recursively filling in the loss power. The maintenance loss power is the value of the power loss caused by the generator set being shut down for maintenance.

[0030] Based on the fault alarm codes and state change events in the operational dataset, fault attribution and source analysis are performed on the abnormal operation events of the generator set to obtain the fault responsibility time period of the generator set. The operational dataset contains fault alarm codes and state change events for the generator sets. Fault alarm codes are fault type codes automatically generated when a fault is triggered in the generator set. State change events are monitoring records of generator set operating parameters suddenly deviating from the normal range. The fault alarm codes and state change events are precisely matched by timestamps. The fault type corresponding to each fault alarm code is analyzed one by one to determine the start and end times of the fault. Combined with the abnormal parameter periods of the state change events, the responsibility periods that the fault directly affects the power generation output are defined. All responsibility periods are integrated in chronological order to form the fault responsibility period division of the generator set. The fault responsibility period division is a set of time intervals that mark the abnormal power generation caused by the fault of the generator set.

[0031] Based on the fault responsibility period division, the fault deficit of the generator set is accumulated by fault deficit integral of the power loss of the generator set during the fault period to obtain the fault loss of the generator set. The calculation scope is defined by the time interval of the fault responsibility period. The normal output benchmark and the actual output power of the generator set during the fault period are extracted moment by moment. The calculation is performed point by point using a fixed time step. The difference between the normal output benchmark and the actual output power within each time step is accumulated. The formula for calculating the fault-related power loss is E3=Σ(P -P )×Δt, where E3 represents the power loss due to fault, Σ represents the summation of the calculation results for all time steps within the calculation period, and P P represents the normal output reference of the generator set. Δt represents the actual power output of the generator set, and Δt represents the fixed time step used in the calculation. The fault loss power of the generator set is calculated by accumulating the fault deficit integral. The fault loss power is the power loss value caused by the generator set to stop or reduce power due to faults.

[0032] S3. In the historical window of no power curtailment and no faults in the running dataset, the power behavior of the generator set is analyzed by group correlation. During the period when the restricted generator set enters the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for the new energy power station. In this embodiment of the invention, the process of obtaining the wake correction reference standard for new energy power stations is as follows: In the historical window of no power curtailment and no faults in the running dataset, the power fluctuations between generator sets are analyzed in a coordinated manner to obtain the directional dependence strength coefficient of the generator sets. The zero-waste and zero-fault historical window is obtained by filtering from the operational dataset, which contains full-time power generation operation data of all generator units in the renewable energy power station. During the filtering process, all periods marked with power curtailment status and periods marked with fault status are first removed, and normal power generation data intervals whose duration meets the requirements for continuous and stable operation of the power station are retained. Power fluctuation is the change in the output power of a generator unit per unit time. Coordinated fluctuation analysis first aligns the output power of all generator units in the zero-waste and zero-fault historical window with a unified time axis, and then extracts the power fluctuation values ​​of any two units at each time step to calculate the power fluctuation of the upstream unit. The temporal fit between the numerical values ​​and the power fluctuation values ​​of downstream units is converted into a quantified value according to a fixed ratio. This quantified value is the directed dependence strength coefficient of the generator units. The directed dependence strength coefficient is used to uniquely characterize the degree of correlation between the upstream and downstream power influence between any two generator units in the field. The calculation formula for the cooperative fluctuation analysis is: R=Σ(W1×W2) / N, where R represents the directed dependence strength coefficient, Σ represents the time-by-time cumulative calculation, W1 represents the power fluctuation value of the upstream unit, W2 represents the power fluctuation value of the downstream unit, and N represents the total number of time moments involved in the calculation.

[0033] Based on the directed dependency strength coefficient, a topological association mapping is performed on the generator set to obtain the body dependency topological network of the generator set; Topology mapping treats each generator set as an independent network node, and uses the directed dependency strength coefficient between two generator sets as the connection strength value between nodes. According to the actual physical arrangement of generator sets in the power plant, all network nodes are connected sequentially according to the upstream and downstream influence relationship. Connection paths with connection strength values ​​higher than the power plant's set association threshold are retained, while connection paths with connection strength values ​​lower than the power plant's set association threshold are removed. The final complete network structure is the generator set's body dependency topology network. The body dependency topology network fully presents the power influence transmission path and association strength between all generator sets in the new energy power plant.

[0034] Using the free-flow state power of generator units in the unit group dependency topology network as a reference, the benchmark output of generator units is calibrated to obtain the dynamic behavior reference benchmark of new energy power stations. Free-flow power is the standard power output value of a generator unit under ideal operating conditions, without wake obstruction from other units, scheduling power restrictions, or equipment performance degradation, corresponding to real-time wind conditions. This value is extracted from the period when the unit operates independently within a historical window with no power curtailment and no faults. The benchmark output calibration matches and binds the free-flow power corresponding to each unit in the dependent topology network to the unit number one by one according to the time sequence, forming a standard output data set covering all units and all time periods. This data set is the dynamic behavior reference benchmark of the new energy power station, which serves as the output judgment standard for the power station's generator units when there is no external interference.

[0035] Based on the power curtailment command records in the operational dataset, the constrained generator units in the operational dataset are identified, and the downstream impact links of the constrained generator units are traced level by level according to the machine body dependency topology network to obtain the wake-affected generator units of the constrained generator units. The power curtailment instruction record is a complete data content that centrally records the time, unit number, and power curtailment value of the power curtailment instruction issued by the dispatch terminal. The unit number in the power curtailment instruction record is read line by line and compared with the unit numbers of all generator units in the station. The generator unit with the same number is the restricted generator unit. The downstream impact link is a continuous connection path in the machine-dependent topology network that starts from the restricted generator unit node and extends to all downstream related nodes. The path is traversed level by level, starting from the restricted generator unit node and visiting each downstream connection node in the machine-dependent topology network in turn, recording the unit numbers of all visited units. These units together constitute the wake-affected units of the restricted generator unit. The power output of the wake-affected units will be directly affected by the wake effect of the upstream restricted generator unit.

[0036] During the period when the restricted generator unit enters the power curtailment state, the virtual free flow wake field of the restricted generator unit is inverted and reconstructed to obtain the virtual free flow wake distribution of the restricted generator unit. The virtual free-flow wake field is a stable wake wind field formed downstream of a restricted generator unit during power curtailment periods, assuming no power curtailment command is received and the unit maintains free-flow operation. The inversion reconstruction first extracts real-time wind data, generator hub height, and impeller diameter during the power curtailment period. Then, combined with free-flow power values, it reverse-engineers the wind speed attenuation values ​​at different distances and directions downstream of the unit during the power curtailment period. The attenuation coefficient is then exponentially calculated based on the distance from the restricted generator unit, and the free-flow wind speed is multiplied by this exponentiation result to obtain the wake wind speed at the corresponding location. All wind speed values ​​are integrated according to spatial location to form spatial distribution data. This spatial distribution data is the virtual free-flow wake distribution of the restricted generator unit. The virtual free-flow wake distribution completely records the spatial wind speed variation law of the ideal wake during the power curtailment period of the restricted generator unit.

[0037] Based on the virtual free-flow wake distribution, the equivalent wind speeds of the wind turbine surfaces of the downstream generator units affected by the wake are inverted to obtain the equivalent wind speed sequence of the wind turbine surfaces of the downstream generator units. The wake propagation inversion first extracts the spatial wind speed attenuation law in the virtual free-flow wake distribution. The spatial wind speed attenuation law is that the wind speed in the wake region gradually decreases by a fixed proportion as the distance from the upstream restricted unit increases, and different relative orientations correspond to different wind speed correction proportions. Then, the actual distance and relative orientation between the wake-affected unit and the upstream restricted unit are obtained. The basic wind speed value of the upstream restricted unit under the virtual free-flow state is extracted at each time step. At each time step, the corresponding distance attenuation proportion is matched according to the actual distance and the distance attenuation of wind speed is calculated. At each time step, the corresponding orientation correction proportion is matched according to the relative orientation and the orientation correction of wind speed is calculated. After distance attenuation and orientation correction, the equivalent wind speed value at the center position of the wind turbine sweep surface of each downstream wake-affected unit is obtained. The equivalent wind speed values ​​of each downstream unit at each time step are arranged continuously in chronological order to form a continuous time series data set. This time series data set is the equivalent wind speed sequence of the wind turbine surface of the downstream generator unit. The equivalent wind speed sequence of the wind turbine surface directly reflects the actual impact of the wake effect on the wind energy captured by the downstream unit.

[0038] Based on the equivalent wind speed sequence of the wind turbine surface, the expected power output of the downstream generator unit in the dynamic behavior reference benchmark is corrected for the wake effect, and the wake correction reference benchmark of the new energy power station is obtained. The expected power output is the standard output value of the corresponding unit at the corresponding time in the dynamic behavior reference. Wake effect correction first reads the equivalent wind speed sequence values ​​of the downstream unit's rotor surface time-by-time, then compares these values ​​with the standard wind speed values ​​corresponding to the dynamic behavior reference. The formula for calculating the corrected expected power output is P. =P0×(V / V In the formula, P P0 represents the expected power output after correction, and V represents the expected power output before correction. V represents the equivalent wind speed on the wind turbine surface. The dynamic behavior reference standard wind speed is used to determine the expected power output. When the equivalent wind speed at the wind turbine surface is higher than the standard wind speed, the expected power output is increased proportionally. When the equivalent wind speed at the wind turbine surface is lower than the standard wind speed, the expected power output is decreased proportionally. The corrected expected power output values ​​of all units are integrated by time and unit number to form unified reference data. This reference data is the wake correction reference benchmark for new energy power plants. The wake correction reference benchmark is the accurate power output calculation benchmark for generator units after incorporating the actual impact of wake.

[0039] S4. Using the wake correction reference benchmark as the basis for judgment, measure the power curtailment loss of the centrally constrained generator units in the operating data to obtain the implicit power curtailment loss of the new energy power station. In this embodiment of the invention, the process of obtaining the implicit power curtailment loss of new energy power plants is as follows: Using the wake correction reference standard as the basis for judgment, the restricted period of the restricted generator unit is identified to obtain the implicit curtailment period of the restricted generator unit. The wake correction reference benchmark is the standard output calculation benchmark for generator sets obtained after incorporating wake effect correction; restricted generator sets are those that have received power curtailment orders and entered power curtailment operation; restricted period identification is based on the actual power generation time-series data of restricted generator sets and the wake correction reference benchmark as the basic data; the actual power generation time-series data of restricted generator sets are matched and aligned one-to-one with the expected power generation time-series data of the corresponding units in the wake correction reference benchmark, using a unified timestamp as the benchmark; the aligned actual power generation value is compared with the expected power generation value moment by moment in chronological order. The fixed judgment duration is the minimum continuous effective duration preset by the power station based on the data acquisition frequency, used to filter instantaneous fluctuation interference; when the actual generated power value is continuously less than the expected generated power value, and the duration of this continuous state reaches the fixed judgment duration, the continuous period is marked as an effective restricted period; all effective restricted periods are integrated in chronological order to form a set of continuous time intervals without interruption or repetition, which is the implicit curtailment period of the restricted generator; the implicit curtailment period is the complete time interval in which the restricted generator is affected by implicit curtailment, resulting in insufficient output.

[0040] Based on the implicit power curtailment period, the expected power output of the restricted generator units in the wake correction reference benchmark is compared with the actual power output point by point to obtain the implicit power deficit sequence of the restricted generator units. The implicit curtailment period is the complete time interval during which the power output of the constrained generator units is insufficient due to implicit curtailment. The fixed time step is the minimum time interval consistent with the data acquisition frequency of the power station. Within the implicit curtailment period, each acquisition time is traversed sequentially from the start time to the end time according to the fixed time step. For each traversed time, the expected power output value of the constrained generator unit in the wake correction reference is extracted, as well as the actual operating power output value at that time. The power deficit value at a single moment is obtained by subtracting the actual power output value from the expected power output value. If the expected power output value is less than or equal to the actual power output value, the power deficit value at that moment is recorded as zero. The power deficit values ​​at all moments are arranged continuously in chronological order to form a complete time series data set, which is the implicit power deficit sequence of the constrained generator units. The implicit power deficit sequence is the time series data characterizing the implicit curtailment power gap of the constrained generator units at each moment.

[0041] Based on the implicit power deficit sequence, the deficit time-domain integral of the constrained generator set is performed to obtain the implicit power curtailment loss of the constrained generator set. The implicit power deficit sequence is time-series data characterizing the implicit power curtailment gap of constrained generator units at each time step. The time-domain integral of the deficit is based on the implicit power deficit sequence and a fixed time step. The power deficit value corresponding to each time step is extracted sequentially from the implicit power deficit sequence in ascending order of timestamp. Each extracted power deficit value is multiplied by the fixed time step to obtain the implicit power curtailment loss value at that time step. The implicit power curtailment loss values ​​at all times are summed sequentially in chronological order. During the summation process, only the currently calculated sum is added to the value at the next time step, without performing other calculations. The summation stops when all values ​​in the implicit power deficit sequence have been calculated. The final sum is the implicit power curtailment loss value of a single constrained generator unit. This value represents the total power loss of a single constrained generator unit during the implicit power curtailment period due to the actual output being lower than the expected power output.

[0042] The implicit power curtailment loss is aggregated across the entire region to obtain the implicit power curtailment loss of new energy power plants. The implicit curtailment loss of a renewable energy power station is the sum of the implicit curtailment losses of all restricted generator units in the station. The overall aggregation is based on the individual implicit curtailment losses of each restricted generator unit. All restricted generator units in the renewable energy power station are traversed, and the individual implicit curtailment loss value of each unit is extracted sequentially. All extracted values ​​are then summed sequentially, without distinguishing between unit numbers or time periods; the values ​​are simply added directly. If there are no restricted generator units in the station or the implicit curtailment losses of all restricted units are zero, the final summation result is zero. The total sum is the implicit curtailment loss of the renewable energy power station, representing the total power loss caused by implicit curtailment for the entire station.

[0043] S5. During the period when the restricted generator unit is affected by implicit curtailment, perform high-frequency incremental loss conversion on the power regulation record of the restricted generator unit, and update the wake correction reference benchmark based on the conversion result. In this embodiment of the invention, the process of updating the wake correction reference based on the calculation result is as follows: During the period when constrained generator units are affected by implicit power curtailment, amplitude features are extracted from the power regulation records of constrained generator units to obtain the regulation amplitude sequence of constrained generator units. The implicit curtailment impact period is the time interval during which the output of the constrained generator unit is continuously lower than the standard value and frequently adjusts its power due to implicit curtailment. The power adjustment record is the full power adjustment data collected and stored by the control system during the operation of the constrained generator unit. This data includes the precise time of occurrence of each power adjustment action, the stable output power value before the adjustment action is initiated, the stable output power value after the adjustment action is completed, and the duration of the adjustment action. During the implicit curtailment impact period, all power adjustment action data are traversed frame by frame according to the fixed time interval of high-frequency acquisition at the station. The stable output power value before and after the adjustment corresponding to each adjustment action is extracted one by one. The change value of the single power adjustment is obtained by subtracting the stable output power value before the adjustment from the stable output power value after the adjustment. The absolute value of the change value is taken to obtain the amplitude value of the single power adjustment. All the calculated single power adjustment amplitude values ​​are arranged continuously in the order of the occurrence of the adjustment action. The continuous time series data set formed by the arrangement is the adjustment amplitude sequence of the constrained generator unit. The adjustment amplitude sequence completely records the amplitude change of each power adjustment of the constrained generator unit during the implicit curtailment impact period.

[0044] The fatigue equivalent aggregation of the regulation amplitude sequence is used to obtain the cumulative fatigue damage increment of the constrained generator set. The regulation amplitude sequence is a continuous time-series data set of successive power regulation amplitudes of the constrained generator set. The fatigue damage value per unit amplitude is a fixed value pre-calibrated based on the generator set's model parameters, core component design life, and mechanical fatigue characteristics. This value represents the degree of fatigue damage caused to the generator set's core components by the unit power regulation amplitude. The fatigue equivalent aggregation involves iterating through each single power regulation amplitude value in the regulation amplitude sequence, multiplying each single power regulation amplitude value by the pre-calibrated unit amplitude fatigue damage value, and obtaining the single fatigue damage value generated by that power regulation action. All single fatigue damage values ​​are then accumulated sequentially in chronological order, retaining the precise results of all calculated values ​​during the accumulation process. The total value obtained after accumulation is the cumulative fatigue damage increment of the constrained generator set. The cumulative fatigue damage increment is the sum of all mechanical fatigue damage caused by frequent power regulation during the implicit curtailment period of the constrained generator set.

[0045] Based on the cumulative fatigue damage increment, the performance degradation degree of the restricted generator set is evaluated, and the active power output attenuation coefficient of the restricted generator set is obtained. The cumulative fatigue damage increment is the sum of all mechanical fatigue damage of the constrained generator unit during the period of implicit power curtailment. The fatigue damage threshold range is a fixed numerical range pre-divided according to the unit's mechanical fatigue safety standards and performance degradation law. This range is divided into three levels: low fatigue damage range, medium fatigue damage range, and high fatigue damage range. Each level corresponds to a fixed output attenuation ratio. The performance degradation degree assessment is to compare the total value of the cumulative fatigue damage increment with the three fatigue damage threshold ranges one by one to determine the unique threshold range to which the cumulative fatigue damage increment belongs, and extract the fixed output attenuation ratio value corresponding to the threshold range. This value is the active power output attenuation coefficient of the constrained generator unit. The active power output attenuation coefficient is used to quantitatively characterize the specific degree of decline in the actual active power output capacity of the constrained generator unit due to mechanical fatigue damage.

[0046] Based on the active power output attenuation coefficient, the expected power output of the restricted generator unit in the wake correction reference is corrected by the reference capacity attenuation to obtain the updated wake correction reference. The active power output attenuation coefficient is a quantitative value characterizing the degree of decline in the output capacity of a restricted generator unit. The wake correction reference benchmark is an output calculation benchmark that includes the expected output power of the restricted generator unit throughout the entire time period. The benchmark capacity attenuation correction is to extract the expected output power of the restricted generator unit corresponding to the wake correction reference benchmark at each time moment, multiply the expected output power value by the active power output attenuation coefficient, and obtain the actual expected output power value after considering the fatigue damage of the unit. The actual expected output power values ​​corrected at all times are re-integrated with the unit number in chronological order. The newly integrated output calculation benchmark is the updated wake correction reference benchmark. The updated wake correction reference benchmark incorporates the output attenuation characteristics caused by the fatigue damage of the unit and can fit the actual operating capacity of the unit.

[0047] S6. Based on the updated wake correction reference benchmark, the implicit curtailment loss of electricity in subsequent periods is compensated and calculated. The calculation results are then combined with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the new energy power station. In this embodiment of the invention, the process of obtaining the total generalized curtailment loss of new energy power plants is as follows: Based on the updated wake correction reference, the compensation period for the subsequent time period of the restricted generator unit is determined, and the compensation calculation period for the restricted generator unit is obtained. The updated wake correction reference benchmark is an output calculation benchmark that closely matches the actual operating capacity of the unit after incorporating the output attenuation characteristics of fatigue damage in the constrained generator set. The subsequent period is the continuous operating period after the end of the implicit curtailment period of the constrained generator set. The compensation period determination is to extract the actual power output value of the constrained generator set in the subsequent period and the expected power output value at the corresponding time in the updated wake correction reference benchmark. The two types of values ​​are matched and aligned one by one with a unified timestamp as the benchmark. The aligned actual power output value and the expected power output value are compared in chronological order. The fixed judgment duration is the minimum continuous effective duration preset by the station based on the data acquisition frequency. It is used to filter instantaneous power fluctuation interference. When the actual power output value is continuously less than the expected power output value and the duration of this continuous state reaches the fixed judgment duration, the continuous period is marked as an effective compensation period. All effective compensation periods are integrated in chronological order to form a set of time intervals without repetition or interruption. This set of time intervals is the compensation calculation period of the constrained generator set.

[0048] Based on the compensation calculation period, the expected power output in the updated wake correction reference benchmark is extracted to obtain the compensation expected power output sequence of the constrained generator set. The compensation calculation period is the effective time interval during which implicit curtailment loss compensation calculations need to be carried out for constrained generator units. Expected value extraction is performed within the compensation calculation period by sequentially traversing each acquisition moment from the start to the end moment according to a fixed time step consistent with the data acquisition frequency of the power station. The expected power output corresponding to the constrained generator unit in the updated wake correction reference is extracted moment by moment. All extracted expected power output values ​​are arranged continuously in chronological order, and the resulting continuous time series data set is the expected compensation output sequence for the constrained generator unit.

[0049] By comparing the expected power output of the compensation-expected sequence with the actual power output of the restricted generator units, the uncompensated power deficit sequence of the restricted generator units is obtained. The expected compensation power sequence is a time-series data set of the standard expected power output of the restricted generator set within the compensation calculation period. The deviation compensation comparison is to compare the expected compensation power output value with the actual operating power output value of the restricted generator set at fixed time steps within the compensation calculation period. The power deficit value to be compensated at a single moment is obtained by subtracting the actual power output value from the expected compensation power output value. If the expected compensation power output value is less than or equal to the actual power output value, the power deficit value to be compensated at that moment is recorded as zero. The time-series data set formed by arranging all the power deficit values ​​to be compensated at a single moment in chronological order is the power deficit sequence to be compensated for of the restricted generator set.

[0050] The compensation deficit sequence is integrated to obtain the compensation calculation power of the restricted generator units; The uncompensated deficit sequence is a time-series data set of the uncompensated power deficit of the restricted generator unit at each moment during the compensation calculation period. The compensation deficit integral is to multiply the uncompensated power deficit value at each moment in the uncompensated deficit sequence by a fixed time step to obtain the corresponding single-moment compensation calculation power value. All single-moment compensation calculation power values ​​are accumulated and summed in chronological order. During the accumulation process, only the currently calculated sum is added to the value at the next moment, and no other additional calculations are performed. The accumulation stops after all values ​​in the uncompensated deficit sequence have participated in the calculation. The final accumulated sum value is the compensation calculation power of a single restricted generator unit.

[0051] The compensated electricity volume, explicit curtailment loss volume, maintenance loss volume, and fault loss volume are uniformly calibrated for multi-source losses to obtain the set of calibrated loss components for new energy power plants. Compensated power loss is the implicit power curtailment loss generated by a single constrained generator unit during the subsequent compensation period. Explicit power curtailment loss is the direct power curtailment loss caused by the generator unit due to dispatch command limits. Maintenance loss is the power loss caused by the generator unit's maintenance shutdown. Fault loss is the power loss caused by the generator unit's fault operation. Unified calibration of multi-source losses unifies the four types of data—compensated power loss, explicit power curtailment loss, maintenance loss, and fault loss—with the same accounting caliber and unit of measurement. It involves checking and eliminating any overlapping time periods and statistical deviations in the four types of data, and then accurately matching and integrating the four types of data according to the generator unit number and time period. The standardized loss data set formed after integration is the calibration loss component set of the new energy power station.

[0052] The total generalized curtailment loss of new energy power plants is obtained by summarizing and merging the calibrated loss component sets. The calibration loss component set is a complete data set composed of four types of standardized power loss from renewable energy power plants. The component summation is achieved by traversing all standardized power loss values ​​in the calibration loss component set and directly adding all values ​​without discrimination. During the summation process, the accurate calculation results of all values ​​are fully preserved without any rounding. The final total value obtained after the summation is the generalized curtailment loss of renewable energy power plants, which is the sum of all types of power generation losses at renewable energy power plants.

[0053] like Figure 2 The diagram shown is a functional block diagram of a system for accurately calculating the power loss of new energy power generation according to an embodiment of the present invention.

[0054] The new energy power generation loss accurate calculation system 100 described in this invention can be installed in an electronic device. According to the functions implemented, the new energy power generation loss accurate calculation system 100 may include an operation data acquisition module 101, a three-type loss calculation module 102, a wake correction benchmark construction module 103, an implicit curtailment loss calculation module 104, a wake benchmark update module 105, and a generalized curtailment loss calculation module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0055] In this embodiment, the functions of each module / unit are as follows: The data acquisition module 101 is used to perform multi-dimensional synchronous data acquisition on the generator sets in the new energy power station to obtain the generator set operation dataset. The three-category loss accounting module 102 simultaneously performs instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset, and obtains the explicit curtailment power loss, maintenance loss power and fault loss power of the generator set, respectively. The wake correction benchmark construction module 103 performs group correlation analysis on the power behavior of generator sets in the historical window of no power curtailment and no faults in the running dataset. During the period when the restricted generator sets enter the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for new energy power stations. The implicit curtailment loss calculation module 104 uses the wake correction reference benchmark as the judgment basis to measure the curtailment loss of the centrally constrained generator units in the operating data, and obtains the implicit curtailment loss of the new energy power station. The wake reference update module 105 performs high-frequency incremental loss calculation on the power regulation records of the restricted generator set during the period when the restricted generator set is affected by implicit curtailment, and updates the wake correction reference benchmark based on the calculation results. The generalized curtailment loss calculation module 106 calculates the implicit curtailment loss in subsequent periods based on the updated wake correction reference benchmark, and merges the calculation results with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the renewable energy power station.

[0056] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of modules is merely a logical functional division, and there may be other division methods in actual implementation.

[0057] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0058] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0059] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0060] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for accurately calculating the power loss in new energy power generation, characterized in that, The method includes: S1. Perform multi-dimensional synchronous data collection on the generator sets in the new energy power station to obtain the generator set operation dataset; S2. Simultaneously perform instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset to obtain the explicit power curtailment loss, maintenance loss and fault loss of the generator set, respectively. S3. In the historical window of no power curtailment and no faults in the running dataset, the power behavior of the generator set is analyzed by group correlation. During the period when the restricted generator set enters the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for the new energy power station. S4. Using the wake correction reference benchmark as the basis for judgment, measure the power curtailment loss of the centrally constrained generator units in the operating data to obtain the implicit power curtailment loss of the new energy power station. S5. During the period when the restricted generator unit is affected by implicit curtailment, perform high-frequency incremental loss conversion on the power regulation record of the restricted generator unit, and update the wake correction reference benchmark based on the conversion result. S6. Based on the updated wake correction reference benchmark, the implicit curtailment loss of electricity in subsequent periods is compensated and calculated. The calculation results are then combined with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the new energy power station.

2. The method for accurately calculating the power loss of new energy power generation as described in claim 1, characterized in that, The process of obtaining the generator set's operating data set is as follows: The initial data stream of generator units in the new energy power station is time-series aligned and checked to obtain the synchronization data sequence of the generator units. Based on the synchronous data sequence, the active power signal of the generator set and the wind speed and direction signal of the wind measuring tower at the station are decoupled and mapped by information entropy to obtain the equivalent wind condition sequence of the generator set. Based on the equivalent wind condition sequence of the generator set, the operating status of the generator set is identified by a multi-threshold hierarchical method to obtain the operating status label of the generator set. By performing multi-source time-series fusion of synchronous data sequences, equivalent wind condition sequences of generating units, and operating condition label sets, an operational dataset for new energy power plants is obtained.

3. The method for accurately calculating the power loss of new energy power generation as described in claim 1, characterized in that, The process of obtaining the explicit power curtailment loss, maintenance loss, and fault loss of the generator set is as follows: By identifying deviation periods in the scheduling instruction records and actual power generation files of the generating units in the operation dataset, the power curtailment period markers of the new energy power plants are obtained; Based on the instruction deviation time period mark, the power generation gap of the generator set is differentially aggregated to obtain the explicit power curtailment loss of the generator set. Based on the maintenance records and unit status logs in the operational dataset, the unavailable periods of the generator set are extracted to obtain the maintenance downtime calibration of the generator set. Based on the maintenance and shutdown period calibration, the benchmark capacity of the generator set is extracted within the historical window of no power curtailment and no faults to obtain the normal output benchmark of the generator set. Based on the normal output benchmark, the power loss of the generator units during the outage period is recursively compensated to obtain the maintenance power loss of the new energy power station. Based on the fault alarm codes and state change events in the operational dataset, fault attribution and source analysis are performed on the abnormal operation events of the generator set to obtain the fault responsibility time period of the generator set. Based on the division of fault responsibility periods, the power loss of the generator set during the fault period is accumulated by fault deficit integral to obtain the power loss of the generator set due to fault.

4. The method for accurately calculating the power loss of new energy power generation as described in claim 1, characterized in that, The process of obtaining the wake correction reference standard for new energy power plants is as follows: In the historical window of no power curtailment and no faults in the running dataset, the power fluctuations between generator sets are analyzed in a coordinated manner to obtain the directional dependence strength coefficient of the generator sets. Based on the directed dependency strength coefficient, a topological association mapping is performed on the generator set to obtain the body dependency topological network of the generator set; Using the free-flow state power of generator units in the unit group dependency topology network as a reference, the benchmark output of generator units is calibrated to obtain the dynamic behavior reference benchmark of new energy power stations. During the period when the restricted generator units enter the power curtailment state, the assessment results are corrected for wake effect to obtain the wake correction reference benchmark for new energy power stations.

5. The method for accurately calculating the power loss of new energy power generation as described in claim 4, characterized in that, The process of obtaining the wake correction reference standard for new energy power plants is as follows: Based on the power curtailment command records in the operational dataset, the constrained generator units in the operational dataset are identified, and the downstream impact links of the constrained generator units are traced level by level according to the machine body dependency topology network to obtain the wake-affected generator units of the constrained generator units. During the period when the restricted generator unit enters the power curtailment state, the virtual free flow wake field of the restricted generator unit is inverted and reconstructed to obtain the virtual free flow wake distribution of the restricted generator unit. Based on the virtual free-flow wake distribution, the equivalent wind speeds of the wind turbine surfaces of the downstream generator units affected by the wake are inverted to obtain the equivalent wind speed sequence of the wind turbine surfaces of the downstream generator units. Based on the equivalent wind speed sequence of the wind turbine surface, the expected power output of downstream generator units in the dynamic behavior reference benchmark is corrected for the wake effect, thus obtaining the wake correction reference benchmark for new energy power plants.

6. The method for accurately calculating the power loss of new energy power generation as described in claim 5, characterized in that, The process of obtaining the implicit power curtailment loss of renewable energy power plants is as follows: Using the wake correction reference standard as the basis for judgment, the restricted period of the restricted generator unit is identified to obtain the implicit curtailment period of the restricted generator unit. Based on the implicit power curtailment period, the expected power output of the restricted generator units in the wake correction reference benchmark is compared with the actual power output point by point to obtain the implicit power deficit sequence of the restricted generator units. Based on the implicit power deficit sequence, the deficit time-domain integral of the constrained generator set is performed to obtain the implicit power curtailment loss of the constrained generator set. The implicit power curtailment loss is aggregated across the entire region to obtain the implicit power curtailment loss of new energy power plants.

7. The method for accurately calculating the power loss of new energy power generation as described in claim 1, characterized in that, The process of updating the wake correction reference based on the conversion results is as follows: During the period when constrained generator units are affected by implicit power curtailment, amplitude features are extracted from the power regulation records of constrained generator units to obtain the regulation amplitude sequence of constrained generator units. The fatigue equivalent aggregation of the regulation amplitude sequence is used to obtain the cumulative fatigue damage increment of the constrained generator set. Based on the cumulative fatigue damage increment, the performance degradation degree of the restricted generator set is evaluated, and the active power output attenuation coefficient of the restricted generator set is obtained. Based on the active power output attenuation coefficient, the expected power output of the restricted generator unit in the wake correction reference is corrected by the reference capacity attenuation to obtain the updated wake correction reference.

8. The method for accurately calculating the power loss of new energy power generation as described in claim 1, characterized in that, The process for compensating for implicit power curtailment losses in subsequent periods is as follows: Based on the updated wake correction reference, the compensation period for the subsequent time period of the restricted generator unit is determined, and the compensation calculation period for the restricted generator unit is obtained. Based on the compensation calculation period, the expected power output in the updated wake correction reference benchmark is extracted to obtain the compensation expected power output sequence of the constrained generator set. By comparing the expected power output of the compensation-expected sequence with the actual power output of the restricted generator units, the uncompensated power deficit sequence of the restricted generator units is obtained. The compensation deficit sequence is integrated to obtain the compensation calculation power of the restricted generator units.

9. The method for accurately calculating the power loss of new energy power generation as described in claim 8, characterized in that, The process of obtaining the total generalized curtailment loss of renewable energy power plants is as follows: The compensated electricity volume, explicit curtailment loss volume, maintenance loss volume, and fault loss volume are uniformly calibrated for multi-source losses to obtain the set of calibrated loss components for new energy power plants. The total generalized curtailment loss of new energy power plants is obtained by summarizing and merging the calibrated loss component sets.

10. A system for accurately calculating power loss in new energy power generation, characterized in that, The system is used to implement the method for accurately calculating the power loss of new energy power generation according to any one of claims 1-9, the system comprising: The data acquisition module is used to collect data from the generator sets in the new energy power station in multiple dimensions and synchronously to obtain the generator set operation dataset. The three loss accounting modules simultaneously perform instruction limit gap assessment, maintenance failure loss measurement and fault source tracing loss measurement on the running dataset, and obtain the explicit curtailment power loss, maintenance loss power and fault loss power of the generator set, respectively. The wake correction benchmark construction module performs group correlation analysis on the power behavior of generator sets in the historical window of no power curtailment and no faults in the running dataset. During the period when the restricted generator sets enter the power curtailment state, the wake effect is corrected on the analysis results to obtain the wake correction reference benchmark for new energy power stations. The implicit curtailment loss calculation module uses the wake correction reference benchmark as the basis for judgment to measure the curtailment loss of the centrally constrained generator units in the operating data, and obtains the implicit curtailment loss of the new energy power station. The wake reference update module performs high-frequency incremental loss calculation on the power regulation records of the restricted generator units during the period when the restricted generator units are affected by implicit curtailment, and updates the wake correction reference reference based on the calculation results. The generalized curtailment loss calculation module compensates for the implicit curtailment loss in subsequent periods based on the updated wake correction reference benchmark. It then merges the calculation results with the explicit curtailment loss, maintenance loss, and fault loss to obtain the total generalized curtailment loss of the renewable energy power station.

Images