Full life cycle management system and method for gas-steam combined cycle units

Abstract

The present invention relates to the technical field of equipment monitoring and control, and discloses a full life cycle management system and method for a gas-steam combined cycle unit. The method first constructs a typical operating condition parameter database and a unit performance prediction model to construct a first performance fingerprint library; simultaneously, historical operating data is acquired to construct a second performance fingerprint library; then, based on the first performance fingerprint library and the second performance fingerprint library, a first performance evaluation report and a second performance evaluation report are generated; abnormal operating conditions are identified based on the two performance evaluation reports, and an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library are generated; finally, abnormality diagnosis is performed on the unit, and a handling plan for abnormal operating conditions of the unit is generated. The present invention can perform comprehensive performance evaluation and abnormal operating condition diagnosis on the gas-steam combined cycle unit, realize intelligent management of the unit throughout its life cycle, and improve the economy and reliability of the unit operation.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring and control, and more particularly to a full life cycle management system and method for a gas-steam combined cycle unit. Background Art

[0002] During the operation of gas-steam combined cycle units, performance evaluation and abnormal condition diagnosis are key to ensuring safe, economical, and efficient operation. Currently, several patents cover intelligent equipment scheduling and full lifecycle management for thermal power plants and wind farms.

[0003] A Chinese patent with authorization publication number CN102541036B discloses an intelligent coal dispatching system for thermal power plants. This system utilizes modules such as online coal handling, intelligent stacking, intelligent coal blending, and coal loading calculation to achieve full lifecycle management and intelligent dispatching of power plant coal, ensuring the safety, economy, and environmental performance of boiler units under blended combustion conditions. However, this patent primarily addresses coal-fired boilers in thermal power plants and does not cover gas-steam combined cycle units. Gas-steam combined cycle units differ significantly from coal-fired boilers in terms of operating characteristics and performance indicators, necessitating the development of specialized performance evaluation and anomaly diagnosis models.

[0004] Chinese patent publication number CN106228262A discloses a system and method for full lifecycle management of wind turbines. This system utilizes data flow management, organically integrating ERP, CRM, PDM, and other systems through data flow to establish a full lifecycle management system for wind turbine operations. This system provides wind farm owners with a comprehensive information management platform and cost-effective O&M planning. However, this patent primarily targets wind turbines, which differ significantly from gas-steam combined cycle units in terms of unit structure, operating conditions, and performance indicators. Therefore, it cannot be directly applied to the performance evaluation and abnormal condition diagnosis of gas-steam combined cycle units.

[0005] In summary, the existing technology lacks an effective method specifically for the full life cycle management of gas-steam combined cycle units. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a full life cycle management system and method for gas-steam combined cycle units, constructs a performance fingerprint library suitable for the operating conditions of gas-steam combined cycle units, accurately evaluates the performance level of the units through typical operating condition analysis and historical data mining; at the same time, establishes an abnormal operating condition diagnosis model to realize online early warning and diagnostic decision-making of abnormal operating conditions of the units, so as to ensure safe, economical and efficient operation of the units.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The full life cycle management approach for gas-steam combined cycle units includes:

[0009] Constructing a typical operating parameter database and a unit performance prediction model, and constructing a first performance fingerprint library based on the typical operating parameter database and the unit performance prediction model; obtaining second historical operating data, and constructing a second performance fingerprint library based on the second historical operating data;

[0010] Generate a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generate a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; identify abnormal operating conditions based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library;

[0011] Provide online warning of abnormal operating conditions for the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, match the abnormal operating condition handling strategy library and generate the unit's abnormal operating condition handling plan.

[0012] Furthermore, the construction of a typical operating condition parameter database includes:

[0013] Obtain the first-level typical operating parameters of the gas-steam combined cycle unit under n1 typical operating conditions; n1 is the total number of typical operating conditions;

[0014] Based on the The first level typical working condition parameters under the typical working condition are calculated to obtain the The second level typical working condition parameters under typical working conditions; 1≤ ≤n1;

[0015] The first The first level typical working condition parameters and the second level typical working condition parameters under the typical working condition The second level typical working condition parameters under the typical working condition constitute the Record the working parameters of typical working conditions;

[0016] The operating condition parameter records of all typical operating conditions constitute a typical operating condition parameter database; the typical operating condition parameter database includes n1 groups of operating condition parameter records.

[0017] Furthermore, the constructing of the unit performance prediction model includes:

[0018] Obtain historical operating data of the gas-steam combined cycle unit within a historical time period T1, marking it as first historical operating data; construct and train a unit performance prediction model based on the first historical operating data;

[0019] The constructing of the first performance fingerprint library includes:

[0020] According to the typical operating parameters and the unit performance prediction model, the predicted performance indicators under n1 typical operating conditions are obtained;

[0021] Based on the typical working condition parameter database Typical working condition parameter records and The predicted performance index under typical working conditions is generated Performance fingerprint under typical working conditions;

[0022] The performance fingerprints under all typical working conditions constitute a first performance fingerprint library.

[0023] Furthermore, the second historical operation data is the historical operation data of the gas-steam combined cycle unit within a historical time period T2, where T2≠T1; the second historical operation data includes n4 groups of historical operating condition parameter records, where n4 is the total number of historical operating conditions;

[0024] The second historical operating data includes a second historical operating condition parameter and a second historical performance index;

[0025] The constructing of the second performance fingerprint library includes: generating a historical performance fingerprint based on the second historical operating condition parameter and the second historical performance index; and forming all the historical performance fingerprints into the second performance fingerprint library.

[0026] Furthermore, generating the first performance evaluation report includes:

[0027] Traverse the performance fingerprint of each typical working condition in the first performance fingerprint library and extract the The prediction performance index of the typical working condition is calculated. The predicted performance index of a typical working condition is consistent with the preset Deviation from the performance evaluation benchmark for a typical working condition;

[0028] For the first The deviation of all the prediction performance indicators of the typical working condition is weighted averaged to obtain the Comprehensive performance deviation score of typical working conditions;

[0029] The comprehensive performance deviation scores of all typical working conditions are compared with the preset deviation threshold, and the typical working conditions that are higher than the deviation threshold are marked as short-board working conditions to generate a first performance evaluation report.

[0030] Furthermore, generating the second performance evaluation report includes:

[0031] Traverse all historical performance fingerprints in the second performance fingerprint library, perform similarity matching with the first performance fingerprint library, and identify the operating conditions of the unit; the operating conditions of the unit are divided into typical operating conditions and atypical operating conditions;

[0032] Based on the identification results of the unit's operating conditions, analyze the unit's cumulative operating time and start-stop frequency under various typical and atypical operating conditions;

[0033] Based on the cumulative operating time and start-stop frequency, the effective utilization rate of the computer group under various typical and atypical operating conditions is calculated; the changing trend of the effective utilization rate of the unit under various atypical operating conditions during the T3 time period is analyzed.

[0034] Furthermore, the identifying the operating condition of the unit includes:

[0035] Calculate the second performance fingerprint library The historical performance fingerprint and the first performance fingerprint library Similarity of performance fingerprints , get the similarity sequence; 1≤ ≤n2,1≤ ≤n3; n2 is the number of historical performance fingerprints in the second performance fingerprint library, n3 is the number of performance fingerprints in the first performance fingerprint library, n3=n1, n2=n4;

[0036] If the similarity sequence is the maximum value, and ≥ , then it is considered that The historical operating condition corresponding to the historical performance fingerprint belongs to Typical working conditions corresponding to each performance fingerprint; is the preset similarity threshold;

[0037] If the similarity sequence is the maximum value, and < , then the The historical operating conditions corresponding to the historical performance fingerprints are marked as atypical operating conditions.

[0038] Furthermore, the identifying of abnormal operating conditions and generating an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library include:

[0039] According to the comprehensive performance deviation score of typical working conditions, the typical working conditions are divided into four abnormal levels: normal, slightly abnormal, moderately abnormal, and severely abnormal. Typical working conditions with slightly abnormal, moderately abnormal, and severely abnormal conditions are marked as abnormal working conditions; normal typical working conditions are marked as stable working conditions.

[0040] Extracting performance indicator characteristics of abnormal working conditions; generating an abnormal working condition fingerprint library based on the working condition parameters, performance indicator characteristics and abnormality level of the abnormal working condition;

[0041] For abnormal operating conditions of different abnormal levels, abnormal operating condition handling strategies are configured according to the first performance evaluation report and the second performance evaluation report, and an abnormal operating condition handling strategy library is generated.

[0042] Furthermore, the online early warning of abnormal operating conditions of the unit includes:

[0043] Collect the real-time operating parameters of the unit and obtain the real-time predicted performance indicators based on the real-time operating parameters and the unit performance prediction model;

[0044] Generate the real-time performance fingerprint of the unit based on real-time operating parameters and real-time predicted performance indicators;

[0045] Calculate the similarity between the real-time performance fingerprint and each abnormal working condition fingerprint in the abnormal working condition fingerprint library to generate a second similarity sequence SI;

[0046] Find the maximum value from the second similarity sequence SI, recorded as SI A , A represents the abnormal condition fingerprint with the highest similarity to the real-time performance fingerprint in the abnormal condition fingerprint library;

[0047] SI A Compared with the preset abnormality judgment threshold θ1, if SI A <θ1, the current real-time operating condition is determined to be a stable operating condition; otherwise, it is determined to be an abnormal operating condition, and the abnormal level of the current real-time operating condition is determined according to the abnormal level of the abnormal operating condition corresponding to the abnormal operating condition fingerprint A.

[0048] A full life cycle management system for a gas-steam combined cycle unit, which is used to implement the above-mentioned full life cycle management method for a gas-steam combined cycle unit, includes:

[0049] Fingerprint library construction module: used to build a typical operating condition parameter database and a unit performance prediction model, and build a first performance fingerprint library based on the typical operating condition parameter database and the unit performance prediction model; obtain second historical operating data, and build a second performance fingerprint library based on the second historical operating data;

[0050] Performance evaluation module: generates a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generates a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library;

[0051] Working condition diagnosis module: identifies abnormal working conditions based on the first performance evaluation report and the second performance evaluation report, and generates an abnormal working condition fingerprint library and an abnormal working condition handling strategy library;

[0052] Online analysis module: used to provide online warning of abnormal operating conditions of the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, it matches the abnormal operating condition handling strategy library and generates the unit's abnormal operating condition handling plan.

[0053] An electronic device includes a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, the full life cycle management method of the gas-steam combined cycle unit is implemented.

[0054] A computer-readable storage medium stores a computer program, which, when executed, implements the above-mentioned full life cycle management method for a gas-steam combined cycle unit.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The gas-steam combined cycle unit full life cycle management system and method provided by the present invention can build a unit performance fingerprint library based on the unit's typical operating parameters and historical operating data, and achieve a comprehensive characterization of the unit's operating conditions and performance. By generating a performance evaluation report, the present invention can accurately evaluate the performance of the unit under different operating conditions, identify the shortcomings and bottlenecks of the unit's operation, and provide a decision-making basis for the unit's optimized operation and maintenance strategy. At the same time, the present invention can also perform online early warning and diagnosis of abnormal operating conditions of the unit. By matching the abnormal operating condition fingerprint library, it can quickly identify the abnormal operating condition level of the unit and automatically generate the corresponding abnormal operating condition handling plan, providing intelligent auxiliary decision support for the safe and efficient operation of the unit. In addition, by analyzing the cumulative operating time and start-stop frequency of the unit under different typical and atypical operating conditions, the present invention can evaluate the adaptability of the unit to different operating conditions and provide guidance for the scheduling optimization of the unit's operating conditions. In summary, the present invention makes full use of the massive operating data of the unit throughout its life cycle, combines data-driven modeling analysis and diagnostic decision-making technology, and realizes the intelligent perception, evaluation, early warning, diagnosis and optimization of the gas-steam combined cycle unit, which can significantly improve the management level and operating efficiency of the unit, reduce the risk of unplanned shutdown, and lay the foundation for the economical operation of the unit throughout its life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 This is a principle flow chart of the full life cycle management method of the gas-steam combined cycle unit in the present invention;

[0059] Figure 2A flow chart of a method for constructing a typical operating condition parameter database in the full life cycle management method of a gas-steam combined cycle unit of the present invention;

[0060] Figure 3 A flow chart of a method for constructing a first performance fingerprint library in the full life cycle management method of a gas-steam combined cycle unit of the present invention;

[0061] Figure 4 A flow chart of a method for constructing a second performance fingerprint library in the full life cycle management method of a gas-steam combined cycle unit of the present invention;

[0062] Figure 5 A flow chart of a method for generating a first performance evaluation report in the full life cycle management method for a gas-steam combined cycle unit of the present invention;

[0063] Figure 6 A flow chart of a method for generating a second performance evaluation report in the full life cycle management method for a gas-steam combined cycle unit of the present invention;

[0064] Figure 7 This is a flow chart of a method for performing online early warning of abnormal operating conditions of a gas-steam combined cycle unit in the full life cycle management method of the present invention;

[0065] Figure 8 This is a functional module diagram of the full life cycle management system of the gas-steam combined cycle unit in the present invention. DETAILED DESCRIPTION

[0066] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0067] Example 1

[0068] See also Figure 1 As shown, this embodiment provides a full life cycle management method for a gas-steam combined cycle unit, including:

[0069] Step S1000: Acquire typical operating parameters of the gas-steam combined cycle unit and construct a typical operating parameter database; construct a unit performance prediction model, and construct a first performance fingerprint library based on the typical operating parameter database and the unit performance prediction model; acquire second historical operating data and construct a second performance fingerprint library based on the second historical operating data;

[0070] Furthermore, step S1000 includes:

[0071] Step S1100, obtaining typical operating parameters of the gas-steam combined cycle unit and constructing a typical operating parameter database, wherein the typical operating parameters include first-level typical operating parameters and second-level typical operating parameters;

[0072] Furthermore, if Figure 2 As shown, step S1100 includes:

[0073] Step S1110, obtaining first-level typical operating condition parameters under n1 typical operating conditions of the gas-steam combined cycle unit; n1 is the total number of typical operating conditions;

[0074] Step S1120, based on The first level typical working condition parameters under the typical working condition are calculated to obtain the The second level typical working condition parameters under typical working conditions; 1≤ ≤n1;

[0075] Step S1130: The first level typical working condition parameters and the second level typical working condition parameters under the typical working condition The second level typical working condition parameters under the typical working condition constitute the Record the working parameters of typical working conditions;

[0076] Step S1140 : Record the operating parameters of all typical operating conditions to form a typical operating parameter database; the typical operating parameter database includes n1 groups of operating parameter records.

[0077] Specifically, the purpose of step S1100 is to build a database that comprehensively describes the performance of the gas-steam combined cycle unit under various typical operating conditions. Each record in the database contains the key parameters of the unit under a specific operating condition. These parameters are divided into two levels:

[0078] First-level typical operating parameters are those that are easily measured and directly obtained, such as unit load (reflecting the unit's power output), ambient temperature (affecting gas turbine inlet conditions), and fuel properties (such as the calorific value and composition of natural gas). Although these parameters are external inputs, they have a significant impact on unit performance. Unit load, referring to the setpoint or actual value of the unit's power output, is one of the key parameters for measuring unit operating status. Under different loads, the unit's combustion, flow, heat transfer, and other operating conditions vary significantly. Ambient temperature, measured in degrees Celsius, refers to the atmospheric temperature at the unit's location. Ambient temperature significantly affects the gas turbine's inlet density and pressure ratio, which in turn affects the operating conditions of the combustor and steam turbine. Fuel properties refer to the type and composition of the fuel used in the unit. Fuel properties, such as calorific value, sulfur content, and ash content, vary significantly, directly affecting combustor operating conditions and exhaust emissions. These first-level operating parameters are typically obtained directly from design manuals, experimental records, or DCS (distributed control system) monitoring data. However, they only reflect the "surface" state of the unit and it is difficult to fully describe the internal thermal system operating conditions.

[0079] Second-level typical operating parameters are those that are difficult to measure directly but are more critical for determining the internal operating status of the unit, such as the combustion chamber outlet temperature, turbine inlet steam temperature, and condenser vacuum. These parameters are estimated based on the first-level parameters using algorithms such as thermodynamic calculations. The combustion chamber outlet temperature refers to the average temperature of the flue gas at the outlet of the gas turbine combustion chamber, measured in Kelvin or °C. This temperature determines the inlet air temperature of the gas turbine and is a key parameter affecting cycle efficiency and turbine life. The turbine inlet steam temperature refers to the temperature of the main steam at the inlet of the high-pressure cylinder of the turbine, measured in °C. The inlet steam temperature directly affects the turbine cycle efficiency and condensation volume. The condenser vacuum refers to the absolute pressure inside the condenser, measured in kPa. The vacuum determines the back pressure of the turbine exhaust, affecting the cycle output and energy utilization.

[0080] The second-level operating parameters are difficult to measure directly and need to be derived from the first-level parameters through algorithms based on thermodynamics and heat transfer knowledge. Taking the combustion chamber outlet temperature as an example, it can be solved according to the combustion chamber energy balance equation:

[0081]

[0082] Where, The heat input to the fuel, is the flue gas mass flow rate, is the constant pressure specific heat of flue gas, and are the compressor outlet temperature and the combustion chamber outlet temperature, is the heat loss of the combustion chamber. It can be obtained from the compressor characteristic curve and , combined with the known fuel lower calorific value, excess air coefficient, etc., the combustion chamber outlet temperature can be solved Other second-level parameters can also be derived in a similar way.

[0083] Step S1130 combines the primary and secondary parameters for each typical operating condition to form a complete operating condition parameter record. Each record comprehensively depicts the unit's thermal system status under that operating condition, from the inside out. Step S1140 aggregates the parameter records for n1 typical operating conditions to construct a typical operating condition parameter database. This database covers a comprehensive description of the unit's operating conditions under different operating scenarios, but does not require that every operating condition actually occur. It is more like an "encyclopedia" of unit performance, used to support subsequent performance prediction, evaluation, and optimization tasks.

[0084] The typical operating condition parameter database covers the complete performance spectrum of the unit, from rated operating conditions to special operating conditions, without missing any possible operating scenarios. This provides a solid operating condition foundation for full lifecycle management. The performance of different typical operating conditions reflects the adaptability of the unit under different operating conditions. They constitute a multi-dimensional, three-dimensional performance evaluation scale. By obtaining the first-level easily measurable parameters and then deriving the second-level deep parameters, a comprehensive description of the internal and external operating conditions of the unit can be achieved. Aggregating records of multiple typical operating conditions into a database can provide an ideal reference and sample support for subsequent tasks such as unit performance prediction, evaluation, and optimization, making performance management more scientific, accurate, and comprehensive.

[0085] Step S1200, constructing a unit performance prediction model;

[0086] Furthermore, step S1200 includes:

[0087] Step S1210: Acquire historical operating data of the gas-steam combined cycle unit within a historical time period T1, marked as first historical operating data; the first historical operating data includes a first historical operating condition parameter and a first historical performance index;

[0088] Step S1220: construct and train a unit performance prediction model based on the first historical operation data.

[0089] Specifically, the unit performance prediction model is a mathematical model based on a machine learning algorithm, which is used to predict the performance indicators of a gas-steam combined cycle unit under given operating parameters. The model learns from a large amount of historical operating data and explores the intrinsic relationship between operating parameters and performance indicators, thereby achieving quantitative performance prediction. The purpose of constructing this model is to be able to accurately estimate the performance of the unit in the absence of a physical mechanism model, and to provide a quantitative basis for performance evaluation, optimization, and management. The input of the model is the operating parameters representing the operating status of the unit, such as load, ambient temperature, fuel characteristics, etc.; the output is the key indicators for measuring the performance of the unit, such as power generation efficiency, fuel consumption rate, etc. By training and optimizing the model parameters, the model can have strong nonlinear fitting and generalization prediction capabilities.

[0090] In step S1210, operating data within a certain historical time range (denoted as T1) is obtained from data sources such as the unit's operating records, monitoring system, and performance test reports as samples for training the performance prediction model. The historical operating data consists of two parts: operating parameters and performance indicators. The first historical operating parameters describe the unit's operating conditions at the time, such as load, ambient temperature, fuel calorific value, etc., which correspond to the input variables of the model; the first historical performance indicators reflect the actual performance of the unit under the corresponding operating conditions, such as power generation efficiency, coal consumption rate, etc., which correspond to the output variables of the model, also called labels or target values. The operating parameters and performance indicators at the same time point are combined into data pairs to form a complete training sample.

[0091] Step S1220 uses the acquired historical operating data to construct and train the unit performance prediction model. The historical data is randomly divided into a training set and a test set. The training set is used to train and optimize the model parameters, and the test set is used to evaluate the performance and generalization ability of the model. A variety of machine learning algorithms can be used, such as support vector machines, random forests, neural networks, etc., and the appropriate algorithm is selected according to the characteristics of the data and the prediction requirements. Taking the neural network as an example, the network structure parameters such as the number of input layer nodes (equal to the number of operating parameters), the number of hidden layers and the number of nodes per layer, and the number of output layer nodes (equal to the number of performance indicators) are determined, and hyperparameters such as the number of training iterations, learning rate, and regularization parameters are set. Input the training set samples and iteratively optimize the model parameters so that the predicted value is as close to the actual value as possible. Then use the test set samples to test the model performance and calculate evaluation indicators such as the mean square error and the coefficient of determination. Optimize the hyperparameters through methods such as cross-validation to improve the robustness and generalization ability of the model.

[0092] After training and optimization, the performance prediction model establishes a nonlinear mapping relationship from operating parameters to performance indicators. When new operating parameters are input, the model can accurately predict the unit's performance indicators under those conditions without requiring tedious physical modeling and calculations. Because the training data covers a long period of actual operating conditions, the model can characterize the unit's performance characteristics under different operating conditions, making the predictions more convincing and reliable.

[0093] Step S1300: constructing a first performance fingerprint library;

[0094] Furthermore, if Figure 3 As shown, step S1300 includes:

[0095] Step S1310, obtaining predicted performance indicators under n1 typical operating conditions based on typical operating condition parameters and the unit performance prediction model;

[0096] Step S1320: Based on the typical working condition parameter database Typical working condition parameter records and The predicted performance index under typical working conditions is generated Performance fingerprint under typical working conditions;

[0097] Step S1330: The performance fingerprints under all typical working conditions are combined into a first performance fingerprint library.

[0098] Specifically, the performance fingerprint library is constructed to systematically and comprehensively describe the unit's performance characteristics under different operating conditions. The first performance fingerprint library, generated based on a database of typical operating parameters and a unit performance prediction model, reflects the unit's multi-condition performance under both designed and ideal conditions. This inherent property of the unit, hence the name "first performance fingerprint," provides an important reference benchmark for quantitatively evaluating the unit's actual operating performance.

[0099] Load the typical operating condition parameter database constructed in step S1100 and traverse the n1 typical operating conditions. The corresponding operating parameters for each typical operating condition are input into the unit performance prediction model trained in step S1220 to obtain the predicted performance indicators for that operating condition. These indicators typically include power generation efficiency, fuel consumption rate, and pollutant emission levels, reflecting the overall performance of the unit under that operating condition. They are the result of a nonlinear mapping from operating parameters to performance indicators. This process is repeated until the predicted performance indicators for all n1 typical operating conditions are obtained.

[0100] It is important to note that before inputting typical operating parameters into the model, necessary data preprocessing, including data cleaning, normalization, and outlier processing, is required. Data cleaning removes noise, outliers, and obviously erroneous records from the raw data to improve data quality. Outlier processing identifies and corrects anomalous data based on specific rules (such as the 3σ criterion) to restore it to normal levels. Data normalization uniformly maps operating parameters of different dimensions to the [0, 1] interval, eliminating dimensionality effects and facilitating model learning. These preprocessing steps can improve the accuracy and robustness of model predictions.

[0101] The performance fingerprint is a combination of operating parameters and performance indicators formed according to a certain logic, just as the fingerprint pattern is composed of fine lines. The system matches and integrates the operating parameter records of a typical operating condition with the corresponding predicted performance indicators, extracts key information, and expresses and stores it using specific data structures such as vectors and matrices to generate a performance fingerprint for that operating condition. The performance fingerprint inherits the basic information of the operating parameters, reflecting the unit's operating condition; it also incorporates information from the performance indicators, reflecting the unit's performance level under that condition. Therefore, the performance fingerprint can be considered a comprehensive performance portrait of the unit under a specific operating condition.

[0102] The performance fingerprints of all typical operating conditions (n1 types in total) generated in step S1320 are aggregated to form a complete performance fingerprint library. The first performance fingerprint library covers all the ideal performance characteristics of the unit from rated operating conditions to special operating conditions, providing important basic data support for full life cycle performance management. The performance fingerprints of different typical operating conditions reflect the ability of the unit to adapt to different operating conditions, reveal the inherent laws of its performance, and help manufacturers optimize the design and improve the unit. The collection of all performance fingerprints constitutes a comprehensive and systematic performance evaluation framework, which facilitates quantitative analysis and comparison of the performance of the unit under different conditions, laying the foundation for subsequent performance evaluation, fault diagnosis, predictive maintenance and other tasks.

[0103] By building the first performance fingerprint library, the following beneficial effects can be achieved:

[0104] 1. Provides an ideal performance reference benchmark. Using the first performance fingerprint library as the "gold standard" for measuring the actual operating performance of the unit, comparative analysis can quantitatively assess the unit's performance level and promptly identify performance anomalies and deterioration trends.

[0105] 2. Revealing the inherent laws of unit performance. Correlation analysis between different fingerprints helps to explore the inherent relationship between operating parameters and performance indicators, and deeply understand the key factors affecting unit performance.

[0106] 3. Supports performance evaluation and optimization. The performance fingerprint library enables multi-dimensional and multi-angle performance comparisons, identifying design and operational deficiencies, providing a practical basis for unit optimization and improvement, and improving unit efficiency and reliability.

[0107] 4. Facilitates the construction of performance prediction models. Using the performance fingerprint library as a training sample for a model that combines mechanism and data-driven approaches can significantly improve the model's fitting and extrapolation capabilities, making performance predictions more accurate and comprehensive.

[0108] In short, the first performance fingerprint database details the performance characteristics of a unit under designed and ideal conditions, representing a concentrated reflection of its inherent endowments. Establishing and fully utilizing this database is crucial for gaining a deeper understanding of unit performance patterns, optimizing design and modification plans, and quantitatively assessing operational performance. It is a crucial cornerstone for achieving refined unit management and intelligent operation and maintenance.

[0109] Step S1400: constructing a second performance fingerprint library;

[0110] Furthermore, if Figure 4 As shown, step S1400 includes:

[0111] Step S1410, obtaining historical operating data of the gas-steam combined cycle unit within a historical time period T2, marked as second historical operating data; the second historical operating data includes a second historical operating condition parameter and a second historical performance index; the second historical operating data includes n4 groups of historical operating condition parameter records, where n4 is the total number of historical operating conditions; T2≠T1;

[0112] Step S1420: generating a historical performance fingerprint based on the second historical operating condition parameter and the second historical performance index;

[0113] Step S1430: All historical performance fingerprints are combined into a second performance fingerprint library.

[0114] Specifically, the second performance fingerprint database is constructed based on actual unit operating data, reflecting the actual performance of the unit over different historical periods and evolving dynamically. Compared to the first performance fingerprint database, the second performance fingerprint database further reflects the impact of factors such as the external environment, equipment status, and personnel operation on unit performance, hence the name "second performance fingerprint." The goal of this fingerprint database is to uncover historical patterns in unit performance through big data analysis, facilitating tasks such as performance evaluation, trend forecasting, and anomaly diagnosis.

[0115] The historical operating data is collected over a different time span, T2, than the first historical operating data used to train the unit performance prediction model. This allows for temporal diversity in the data sources. Operating data from the unit's DCS system, performance monitoring system, and other platforms within this time period serves as the raw material for constructing the historical performance fingerprint library. Similar to the first performance fingerprint library, the second historical operating data consists of operating parameters and performance indicators. The difference is that the second historical operating parameters reflect the operating conditions experienced by the unit during actual operation, including both design and off-design conditions, such as operating data under varying loads and environmental conditions. The second historical performance indicators reflect the actual performance level achieved by the unit under these actual operating conditions, which may deviate from the theoretical performance. Pairing the operating parameters and performance indicators at the same moment in time forms a complete set of historical operating parameter records. If the historical data sampling frequency is 1 hour, 24 sets of records will be generated per day.

[0116] The types of operating parameters and performance indicators in the second historical operating data are similar to those in the first performance fingerprint library, and are also divided into first-level parameters that are directly measured and second-level parameters that are indirectly calculated. However, due to factors such as data missing and noise interference, some first-level parameters may not be directly obtained and need to be estimated using reasonable interpolation and fitting methods. In addition, there may be outliers or inconsistencies in historical data, and a systematic data cleaning process must be adopted, including data screening, noise elimination, and data repair, to ensure data quality and reliability.

[0117] For each set of records in the second historical operating data, its operating parameters and performance indicators are extracted and integrated to generate the corresponding historical performance fingerprint. The content and format of the fingerprint are similar to those of the first performance fingerprint library and are usually expressed using data structures such as vectors, matrices, and time series. However, during the fingerprint generation process, it is important to pay attention to the following issues:

[0118] (1) Working condition clustering: Due to the wide variety of historical working conditions, clustering the performance fingerprints of similar working conditions into one category can avoid the fingerprint library from being too large. Clustering algorithms such as K-means and DBSCAN can be used to automatically divide historical working conditions into several typical categories.

[0119] (2) Missing value processing: Unlike the ideal data integrity in the first performance fingerprint library, historical data may contain missing parameters. For fingerprints with fewer missing values, reasonable interpolation methods can be used to complete them; for fingerprints with serious missing values, they should be removed or marked to avoid introducing errors.

[0120] (3) Statistical feature extraction: In the first performance fingerprint library, only one performance fingerprint is generated for each typical operating condition. However, in historical data, the same operating condition may appear multiple times. Therefore, it is necessary to extract statistical features of these performance fingerprints, such as mean, variance, and peak value, as representative indicators of performance under that operating condition.

[0121] The historical performance fingerprint reflects the performance level of the unit under different operating conditions during actual operation and contains a large amount of valuable information. The historical performance fingerprints generated at all time points are summarized and organized according to time and operating conditions to form a complete second performance fingerprint library. This fingerprint library has the following characteristics:

[0122] (1) Dynamicity: Different from the static characteristics of the first performance fingerprint library, the second performance fingerprint library is continuously updated over time and can reflect the changing trend of unit performance in real time.

[0123] (2) Diversity: The historical performance fingerprint covers various actual operating conditions since the unit was put into operation, reflecting the adaptability and performance of the unit under different conditions, and providing rich samples for comprehensive evaluation of unit performance.

[0124] (3) Correlation: There is often a certain correlation between the performance fingerprints of different periods and different operating conditions, which contains the inherent laws of unit performance evolution. Through correlation analysis and data mining, the influencing factors and deterioration mechanisms of performance can be revealed.

[0125] The benefits of building a second performance fingerprint library are:

[0126] 1. Automatically evaluate performance. Leveraging the historical performance fingerprint library, a performance evaluation model can be developed. By calculating the similarity between measured operating conditions and historical fingerprints, the current performance level can be automatically determined without manual table lookup.

[0127] 2. Revealing performance evolution patterns. Trend analysis of performance fingerprints over different periods helps understand the general patterns of unit performance changes, providing early warning of performance deterioration and optimizing operation and maintenance strategies.

[0128] 3. Diagnose abnormal performance conditions. Cluster historical performance fingerprints into different operating conditions and extract performance benchmarks for each condition. When measured performance deviates significantly from the benchmark, it is identified as abnormal, providing clues for fault diagnosis.

[0129] 4. Accumulate knowledge of optimized control. Conduct sensitivity analysis of performance indicators under different operating conditions to identify key control variables that affect unit performance, which can be used to guide personnel in optimizing control or developing automatic optimization systems.

[0130] 5. Enrich the performance profile of the unit throughout its lifecycle. Over time, the historical performance fingerprint library will become increasingly rich, covering all stages of the unit from commissioning to retirement. Examining these performance fingerprints from a macro perspective provides insight into the full lifecycle characteristics of the unit's performance, providing data support for asset management and retrofit decisions.

[0131] The above steps build a complete second performance fingerprint library by analyzing the unit's historical operating data. Compared to the first performance fingerprint library under ideal operating conditions, this fingerprint library more closely reflects the unit's actual operating conditions and dynamically reflects the unit's performance under different periods and conditions. Through horizontal comparison and vertical analysis of fingerprints, unit performance can be quantitatively evaluated, revealing performance evolution patterns and laying the foundation for subsequent performance management and optimization.

[0132] Step S2000: Generate a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generate a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; identify abnormal operating conditions from typical operating conditions based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library;

[0133] Furthermore, step S2000 includes:

[0134] Step S2100: generating a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark;

[0135] Furthermore, if Figure 5 As shown, step S2100 includes:

[0136] Step S2110: traverse the performance fingerprint of each typical working condition in the first performance fingerprint library and extract the performance fingerprint of the first typical working condition. The prediction performance index of the typical working condition is calculated. The predicted performance index of a typical working condition is consistent with the preset Deviation from the performance evaluation benchmark for a typical working condition;

[0137] Step S2120, The deviation of all the prediction performance indicators of the typical working condition is weighted averaged to obtain the Comprehensive performance deviation score of typical working conditions;

[0138] Step S2130: Compare the comprehensive performance deviation scores of all typical operating conditions with a preset deviation threshold, and mark the typical operating conditions that are higher than the deviation threshold as weak working conditions;

[0139] Step S2140: Generate a first performance evaluation report.

[0140] Specifically, the First Performance Evaluation Report assesses and diagnoses the unit's ideal performance under different typical operating conditions. By comparing the predicted performance in the First Performance Fingerprint Library with industry standards and design indicators, deficiencies in the unit's design and manufacturing can be identified, providing a basis for subsequent optimization and improvement. This report is based on the unit's inherent performance potential, hence the name "First Performance Evaluation." The purpose of generating this evaluation report is to examine the unit's performance characteristics from a macro perspective, identify the "shortcomings" that restrict its performance, and indicate the direction and space for optimization.

[0141] Load the first performance fingerprint library and obtain n1 typical working conditions. For each operating condition, the predicted performance indicators of the unit under this condition are extracted, such as power generation efficiency, fuel consumption rate, NOx emissions, etc. At the same time, a performance evaluation benchmark under this typical operating condition is constructed based on the unit's design documents, industry standards, etc. The evaluation benchmark gives the various performance indicators that the unit should achieve under this operating condition, representing the average or advanced level of similar units. Using appropriate similarity metrics (such as Euclidean distance, KL divergence, etc.), the deviation of each predicted indicator from the benchmark is quantitatively calculated. The greater the deviation, the greater the gap between the unit's performance on this indicator and the ideal level, and there is room for optimization. By summarizing the deviations of all predicted indicators, the overall deviation of the unit performance under this operating condition can be obtained.

[0142] Because different performance indicators have varying impacts on unit performance, appropriate weightings are required. Through expert scoring, hierarchical analysis, and other methods, weight coefficients are determined for each indicator to reflect its importance. Taking power generation efficiency and pollutant emissions as an example, given the emphasis on energy conservation and emission reduction, they can be given greater weighting. For peak-valley regulation power plants, however, indicators such as load change rate and start-up and shutdown times should be given higher weightings. The product of each deviation and weight is summed to obtain the comprehensive performance deviation for that operating condition, which serves as a quantitative score for evaluating overall performance under that condition.

[0143] Sort the comprehensive performance deviation scores of n1 typical operating conditions. Set an appropriate deviation threshold (which can be given by expert experience), and mark the typical operating conditions that are higher than the deviation threshold as short-board conditions. Short-board conditions refer to conditions where the performance of the unit deviates greatly from the ideal level, which restricts the improvement of the overall efficiency of the unit and urgently needs to be optimized. The selection of the threshold needs to balance the sample coverage and optimization effort. The higher the threshold, the fewer the number of short-board conditions and the stronger the targeting, but some conditions with optimization value may be missed. The lower the threshold, the more short-board conditions and the stronger the comprehensiveness, but the optimization investment is large.

[0144] Based on the above analysis, the first performance evaluation report is generated, which mainly includes the following contents:

[0145] (1) Ranking of comprehensive performance deviations under various typical operating conditions and comparison with the deviation threshold; intuitively showing the performance of the unit under different operating conditions.

[0146] (2) A radar chart comparing the predicted performance indicators of the short-board working condition with the benchmark, which details the degree of deviation of each indicator under the working condition and reveals the performance shortcomings.

[0147] (3) Optimization suggestions for weak operating conditions, such as conducting equipment diagnosis, upgrading control logic, etc., indicating the direction and measures for optimization.

[0148] (4) Statistical analysis of the overall performance level of the unit, such as average deviation, percentage of operating conditions that are better than the industry benchmark, etc., to quantitatively evaluate the performance potential of the unit.

[0149] This evaluation report utilizes numerous charts and graphs to transform complex data into intuitive information. This report allows designers and operators to comprehensively review the unit's performance characteristics, prioritize areas where shortcomings are identified, and maximize its efficiency potential. The report also serves as valuable communication material with owners, highlighting the unit's differentiated advantages.

[0150] For example, for a certain type of gas turbine, its first performance fingerprint library contains three typical load conditions: 100%, 75%, and 50%. By comparing with the design indicators, it was found that the overall performance deviation at 75% load was as high as 15%, exceeding the threshold of 10% and being marked as a weak plate condition. Further analysis found that the compressor efficiency and turbine efficiency under this condition were seriously low, 5% and 7% lower than the design values ​​respectively, which was the main shortcoming. It was also noted that its deviation at 100% load was only 5%, which was better than the industry benchmark, reflecting the good peak capacity of the unit.

[0151] Based on the evaluation results, designers focused on analyzing the design parameters of the compressor and turbine components under the 75% operating condition and compared them with similar models. Ultimately, they optimized the blade profile and clearances, bringing the efficiency close to the design value. Operations and maintenance personnel also adjusted the operating mode at this load point accordingly, avoiding frequent operation in this range. This optimization has increased the unit's average annual power generation efficiency by 1.2%, saving 5 million yuan in fuel costs annually.

[0152] This demonstrates that performance evaluation based on design operating conditions is an effective means of systematically diagnosing a unit's inherent deficiencies. By identifying and quantitatively analyzing these deficiencies, precise policies can be implemented to maximize unit efficiency and realize its performance potential. This is a fundamental step in gas turbine performance management.

[0153] The above steps begin with typical operating conditions, using design and industry standards as benchmarks to quantitatively evaluate the unit's performance under different operating conditions, identify shortcomings that restrict its performance, and provide a starting point for optimization. The first level of performance evaluation involves assessing the unit's inherent potential by loading the ideal performance from the first performance fingerprint library.

[0154] Step S2200: generating a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library;

[0155] Furthermore, if Figure 6 As shown, step S2200 includes:

[0156] Step S2210: traverse all historical performance fingerprints in the second performance fingerprint library, perform similarity matching with the first performance fingerprint library, and identify the operating condition of the unit; the operating condition of the unit is divided into typical operating conditions and atypical operating conditions;

[0157] Furthermore, step S2210 includes:

[0158] Step S2211, calculate the second performance fingerprint library The historical performance fingerprint and the first performance fingerprint library Similarity of performance fingerprints , get the similarity sequence; 1≤ ≤n2,1≤ ≤n3; n2 is the number of historical performance fingerprints in the second performance fingerprint library, n3 is the number of performance fingerprints in the first performance fingerprint library, n3=n1, n2=n4;

[0159] The calculation of the second performance fingerprint library The historical performance fingerprint and the first performance fingerprint library Similarity of performance fingerprints include:

[0160]

[0161] in:

[0162] The first item: weighted normalized Euclidean distance part (difference in working condition parameters)

[0163]

[0164] Meaning: Measures the normalized weighted distance between fingerprint parameters and captures the differences in fingerprints in the main feature dimensions.

[0165] Explanation of symbols:

[0166] : The total dimension of the first and second level typical operating parameters.

[0167] :The second performance fingerprint library The first historical performance fingerprint parameter values.

[0168] :The first performance fingerprint library The performance fingerprint parameter values.

[0169] : No. The standard deviation of each parameter is used to normalize each parameter and eliminate the influence of dimension. It measures the degree of dispersion of the parameter. After normalization, it avoids the excessive influence of certain features on the similarity calculation due to their large dimension.

[0170] : No. The weight of a parameter reflects the importance of the parameter to the similarity calculation and is determined by feature selection methods (such as model-based feature importance and expert experience). The parameter weight can highlight the role of key performance indicators in similarity calculation.

[0171] Item 2: Nonlinear performance index difference

[0172]

[0173] Meaning: Measures the nonlinear difference in performance indicators between fingerprints.

[0174] Explanation of symbols:

[0175] : The total number of performance indicators (such as power generation efficiency, fuel consumption rate, etc.).

[0176] :The second performance fingerprint library The first historical performance fingerprint performance indicator value.

[0177] :The first performance fingerprint library The performance fingerprint performance indicator value.

[0178] : Nonlinear adjustment index, used to control the sensitivity of different performance indicators, set by historical data fitting or expert experience. Adjust the impact of the difference between each performance indicator on the similarity. Emphasize the large differences between performance indicators.

[0179] Item 3: Comprehensive similarity formula (combining weights and nonlinear differences)

[0180]

[0181] Meaning: Reflect the distance between the first and second items as a similarity score , the distance is mapped to Similarity of range.

[0182] Explanation of symbols:

[0183] : A balance factor controls the relative importance of the first term (operating parameter differences) and the second term (performance indicator differences) in the similarity calculation. This is determined through parameter tuning, typically optimized on a training set through cross-validation. This balances the contributions of the two feature types to the similarity calculation, preventing one feature type from dominating the calculation.

[0184] : Exponential function with the base e of the natural logarithm as base.

[0185] Range , as the difference in operating parameters increases (i.e. and The gap between them increases), the distance value of the first item increases, resulting in The index decreases, and the similarity decreases. As the performance index difference increases (i.e. and The absolute difference between the two increases), the nonlinear difference of the second term increases, resulting in a decrease in similarity. or nonlinear index Improve, the difference of important features or performance indicators will have a greater impact on the similarity. Balance coefficient The relative importance of control parameters and performance indicators. When increases, the difference in performance indicators has a greater impact on the similarity.

[0186] This formula takes into account the differences between operating parameters and performance indicators and constructs a global similarity metric. and nonlinear adjustment parameters , which can highlight the features and performance indicators that contribute most to similarity. By adjusting the balance coefficient and nonlinear index , which can adapt to the similarity requirements of different working conditions. Normalization eliminates the influence of dimension, allowing different features to be compared on a unified scale. By combining weighted normalized Euclidean distance and nonlinear performance differences, it comprehensively characterizes the complex relationship between working condition parameters and performance indicators. Its complexity and nonlinear characteristics give it a significant advantage in capturing subtle differences between fingerprints and identifying anomalies. It can more accurately identify similar working conditions and potential abnormal conditions, providing a solid foundation for subsequent performance evaluation and diagnosis.

[0187] Step S2212: If is the maximum value, and ≥ , then it is considered that The historical operating condition corresponding to the historical performance fingerprint belongs to Typical working conditions corresponding to each performance fingerprint; is the preset similarity threshold;

[0188] Step S2213: If is the maximum value, and < , then the The historical operating conditions corresponding to the historical performance fingerprints are marked as atypical operating conditions.

[0189] Specifically, step S2210 aims to identify the actual operating conditions of the unit at historical moments through similarity matching. First, historical performance fingerprints are loaded from the second performance fingerprint library. Each historical performance fingerprint corresponds to the actual operating parameters and performance indicators of the unit at a specific historical moment, comprehensively recording the unit's operating trajectory. Next, performance fingerprints from the first performance fingerprint library are loaded as a reference for operating condition identification.

[0190] "Similarity matching" is a common pattern recognition method. For the object to be identified (such as a historical performance fingerprint), the most similar pattern is searched in the known pattern library (such as the typical working condition performance fingerprint library) as the category to which it belongs. The similarity can be measured by the distance between vectors. The smaller the distance, the more similar it is. If the similarity between a historical performance fingerprint and all typical working condition fingerprints is less than , it is identified as a new atypical working condition. This is a preset similarity threshold, typically set at an empirical value such as 0.8. The presence of atypical operating conditions indicates that the unit's actual operating state has deviated from the designed operating conditions, potentially indicating performance issues such as equipment aging or parameter imbalance, which warrant attention.

[0191] For example, it is assumed that the 100th historical performance fingerprint in the second performance fingerprint library is:

[0192] {load: 400MW, exhaustTemp: , steamPressure: 16MPa, efficiency: ;

[0193] Among them, load is the load, exhaustTemp is the exhaust temperature, steamPressure is the main steam pressure, efficiency is the power generation efficiency, is the nitrogen oxide emission concentration.

[0194] Perform similarity matching with the performance fingerprint of the first performance fingerprint library to find the three most similar fingerprints:

[0195] Typical working condition 1 (baseload):

[0196] {load: 450MW, exhaustTemp: , steamPressure: 16.5MPa, efficiency: , similarity: 0.9;

[0197] Typical working condition 2 (75% load):

[0198] load:350MW,exhaustTemp: , steamPressure: 15MPa, efficiency: , similarity: 0.85;

[0199] Typical working condition 4 (hot_day):

[0200] {load: 400MW, exhaustTemp: , steamPressure: 15.5MPa, efficiency: , similarity: 0.8;

[0201] The result of the majority vote is typical operating condition 1, so it is determined that the unit is in full load (baseload) operating condition at that historical moment.

[0202] This operating condition identification method reveals the degree of match between the unit's actual operating status and its designed operating conditions. Frequent occurrence of atypical operating conditions indicates that the unit is struggling to adapt to actual grid demand fluctuations. Excessively short operating times under typical conditions indicate poor unit performance in those conditions, potentially indicating localized equipment adaptability issues. Based on the operating condition identification results, further statistical analysis can be conducted on the unit's utilization and changing trends under each typical and atypical operating condition to identify performance shortcomings, providing guidance for subsequent anomaly diagnosis and performance optimization. A similarity matching algorithm links historical operating data with an ideal operating condition fingerprint, constructing an "actual-ideal" mapping and mathematically characterizing the unit's adaptability to designed operating conditions. Similarity calculations in a multidimensional parameter space provide a more accurate assessment of unit status than single performance indicators. Statistical analysis based on big data can reveal the relative strengths and weaknesses of individual units within a population.

[0203] Step S2210 accurately identifies the operating conditions attributed to historical performance fingerprints and quantifies the gap between actual and designed operating conditions. This lays the foundation for subsequent evaluation of unit performance from the perspective of actual operational capability. Operating condition tags can track the dynamic response of unit performance to changing operating conditions, promptly identifying performance deterioration trends. Furthermore, they can reveal shortcomings in adaptability to special operating conditions such as peak shaving and deep peak shaving, providing a foundation for improving flexible performance.

[0204] Step S2220: Based on the identification result of the unit's operating condition, analyze the cumulative operating time and start-stop frequency of the unit under each typical and atypical operating condition;

[0205] Step S2230: Effective utilization of the computer group under various typical and atypical operating conditions based on the accumulated operating time and start / stop frequency;

[0206] Step S2240, analyzing the changing trend of the effective utilization rate of the unit under various atypical operating conditions during the T3 period;

[0207] Step S2250: horizontally compare the utilization differences of similar units under various historical operating conditions to identify the performance shortcomings of the units;

[0208] Step S2260: Generate a second performance evaluation report.

[0209] Specifically, step S2200 aims to generate a comprehensive performance evaluation report based on the first and second performance fingerprint libraries, from the perspective of the unit's long-term actual operating capability. This report, based on historical data, quantitatively analyzes the unit's utilization, start-stop characteristics, and load adaptability under different operating conditions. It also tracks the evolution of these characteristics, identifies performance shortcomings, and proposes optimization recommendations. This report complements the theoretical performance evaluation based on deviation in the first performance evaluation report, providing a basis for the unit's full lifecycle management.

[0210] Step S2210 identifies operating conditions based on the historical operating data in the second performance fingerprint library. Using algorithms such as similarity matching, each historical performance fingerprint is mapped to the most similar typical operating condition, or identified as an atypical operating condition. This process essentially extracts valuable operating condition information from massive amounts of unlabeled operating data to clarify the unit's actual operating status. Frequent occurrence of atypical operating conditions indicates that the unit's actual operation has significantly deviated from its designed state, potentially posing a risk of performance deterioration.

[0211] On this basis, steps S2220-S2230 count the unit's cumulative operating time, start-stop times and other indicators by operating condition type to quantitatively characterize the unit's adaptability to different operating conditions; construct an indicator reflecting the unit's effective utilization rate (Effective Utilization Rate, EUR), which is the utilization rate of the unit under a certain operating condition, that is, the percentage of the unit's actual operating time under a certain operating condition to the total operating time. Operating conditions with high utilization rates usually represent the unit's main operating mode and areas of expertise. Operating conditions with high utilization rates usually represent the unit's main operating mode and core advantages. Operating conditions with low utilization rates reflect the unit's potential shortcomings, such as insufficient low-load peak-shaving capacity. Pay special attention to the utilization rate of the design operating conditions. If the utilization rate is too low, it means that the actual operation of the unit has seriously deviated from the design state.

[0212] Step S2240 further analyzes the temporal trends of the unit's EUR under atypical operating conditions. A time series curve of the EUR during the T3 period is plotted to visually demonstrate the unit's dynamic adaptation to non-design operating conditions. A monotonically rising curve indicates that the unit's actual operating capability under these conditions is continuously improving, likely due to measures such as equipment modification and operational optimization. A rapidly declining curve warns that the unit's adaptability to these conditions is deteriorating, necessitating prompt diagnosis of the cause. The curve exhibits significant periodicity, typically corresponding to cyclical changes in the unit's external environment, such as seasonal fluctuations in grid demand.

[0213] Step S2250 benchmarks the EUR indicators of different units. Using visualization tools such as radar charts, the target unit's utilization differences with similar units under key operating conditions are displayed in multiple dimensions. This analysis identifies the target unit's weak operating conditions (where utilization is significantly lower than its peers) and its strong operating conditions (where utilization is significantly higher than its peers), and explores the causes of these differences. Performance issues reflected by weak operating conditions are often the focus of subsequent diagnosis and optimization.

[0214] Finally, step S2260 synthesizes the above analysis to form a second performance evaluation report. This report systematically describes the unit's long-term actual operating performance, focusing on topics such as quantitative assessment of the unit's adaptability to actual operating conditions, analysis of historical change trends, and performance benchmarking against similar units.

[0215] The second performance evaluation report mainly includes:

[0216] (1) Statistical indicators such as the cumulative operating time, utilization rate, start-stop frequency, etc. of each historical operating condition, as well as the utilization rate change trend of atypical operating conditions in different time periods (such as monthly, quarterly, and annual). Through intuitive data statistics and trend analysis, managers can accurately grasp the actual long-term operating characteristics of the unit and gain insight into the laws of performance changes.

[0217] (2) Analysis of the causes of long-term utilization trends under atypical operating conditions. This involves in-depth analysis of key characteristic points in the utilization curve, such as inflection points and step increases / decreases. This analysis combines internal and external factors such as equipment status, grid dispatch requirements, and environmental conditions to infer the underlying causes of performance changes. This provides the foundation for transitioning from passive response to proactive management.

[0218] (3) Analyze the utilization differences between the unit and similar units under key operating conditions, identify performance shortcomings, and conduct a preliminary diagnosis of the causes of these shortcomings. Visualize the benchmarking analysis results, using radar charts, comparative bar charts, etc. to intuitively demonstrate the gap between the unit and its peers.

[0219] (4) For performance shortcomings, such as insufficient peak-shaving capacity and frequent starts and stops, macro-level optimization suggestions are proposed to provide guidance for the development of specific diagnostic and treatment measures. Based on the design characteristics of the unit, historical experience, and industry best practices, possible solutions are initially explored to gain the initiative for in-depth diagnosis.

[0220] The production of the second performance evaluation report marks a successful shift in perspective for performance evaluation, achieving a leap from static design deviation analysis to dynamic, realistic capacity characterization. This approach transcends the limitations of traditional thermal engineering calculations, placing the unit within the context of long-term, dynamic production practices for the first time, expanding the breadth and depth of performance evaluation across both time and space.

[0221] The performance shortcomings and historical trends revealed in the report serve as both a summary of past operational practices and a guide to future optimization. The EUR metric, based on big data analysis, is closer to the actual value of a unit than traditional KPIs (such as power generation efficiency and coal consumption rate) and is more readily accepted by owners and managers. The unit's adaptability to operating conditions, as reflected by the dynamic trends in the EUR, is precisely the core competitiveness that power companies are most concerned about in the context of flexible transformation.

[0222] In short, the second performance evaluation report complements the first, providing a comprehensive portrait of the unit's performance at both the theoretical design level and actual operational performance. The combination of the two marks a revolutionary reconstruction of the unit performance evaluation system and serves as a crucial foundation for achieving full lifecycle management. Standing at this new starting point, power companies will be able to more accurately diagnose performance issues, more comprehensively optimize production strategies, and more efficiently adapt to the rapid evolution of grid demands.

[0223] Step S2300: Identify abnormal operating conditions from typical operating conditions based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library;

[0224] Furthermore, step S2300 includes:

[0225] Step S2310: Based on the comprehensive performance deviation score of the typical operating condition, the typical operating condition is divided into four abnormality levels: normal, slightly abnormal, moderately abnormal, and severely abnormal. Typical operating conditions with slightly abnormal, moderately abnormal, and severely abnormal conditions are marked as abnormal operating conditions; and normal typical operating conditions are marked as stable operating conditions.

[0226] Step S2320, extracting the performance indicator features of the abnormal operating condition from the performance indicators of the abnormal operating condition;

[0227] Step S2330: Generate an abnormal operating condition fingerprint library based on the operating condition parameters, performance indicator characteristics and abnormality level of the abnormal operating condition.

[0228] Step S2340 : For abnormal operating conditions of different abnormal levels, configure abnormal operating condition handling strategies based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition handling strategy library.

[0229] Specifically, step S2300 aims to further focus on abnormal operating conditions with significant performance deviations, conducting in-depth diagnosis and identifying the causes of the abnormalities, laying the foundation for subsequent resolution. The core of abnormality diagnosis is to reveal the inherent laws and evolutionary trends of abnormal operating conditions through big data analysis and mechanistic reasoning, thereby forming an interpretable, predictable, and optimizable operating condition profile.

[0230] In step S2310, based on the comprehensive performance deviation scores for each typical operating condition obtained from the first performance evaluation report, a cluster analysis method is used to categorize the operating conditions into normal, slightly abnormal, moderately abnormal, and severely abnormal levels according to the degree of abnormality. Because gas-steam combined cycle units operate under numerous conditions, diagnosing each abnormality individually is costly and difficult. Therefore, it is necessary to prioritize the degree of abnormality so that diagnostic resources can be allocated more effectively.

[0231] Specifically, a common clustering algorithm, such as K-means, can be used to group all operating conditions into a preset number (e.g., four) of abnormality levels by minimizing the sum of squared deviations within each group. The comprehensive deviation score is then used as the clustering feature to optimize the cluster centers, ensuring that operating conditions within the same level have the highest degree of abnormality consistency.

[0232] This anomaly classification method can quickly screen a small number of typical operating conditions with high abnormalities from a large number of operating conditions, narrowing the scope of subsequent anomaly diagnosis and improving diagnostic efficiency. Clustering can reveal similarities between different abnormal operating conditions, providing a reference for summarizing anomaly causes and developing targeted anomaly handling strategies. This hierarchical management facilitates differentiated allocation of diagnostic resources, prioritizing the attention and handling of severe anomalies while also addressing minor anomalies, achieving a cost-effective balance in anomaly management within limited resources.

[0233] For example, a gas-steam combined cycle power plant identified 100 typical operating conditions and calculated their comprehensive performance deviation scores, ranging from 0.5% to 12%. K-means cluster analysis classified 60 conditions with deviations less than 1% as normal, 20 conditions with deviations between 1% and 3% as mildly abnormal, 15 conditions with deviations between 3% and 6% as moderately abnormal, and 5 conditions with deviations greater than 6% as severely abnormal. Moderately abnormal conditions and above are prioritized for attention and resolution. This allows subsequent abnormality diagnosis to focus on these 20 abnormal conditions, significantly improving resource utilization efficiency.

[0234] The performance indicator data for abnormal operating conditions comes from the first performance fingerprint database. Typical performance indicator features reflecting abnormal characteristics, such as low load, high coal consumption, and high pollutant emissions, are extracted from this data. These features help to intuitively describe the performance shortcomings of abnormal conditions and serve as a starting point for abnormal diagnosis and performance optimization. Performance indicator feature extraction can employ data mining techniques such as statistical analysis and pattern recognition to identify common patterns in performance indicators for abnormal conditions. The operating parameters of the abnormal conditions, the extracted performance indicator features, and the abnormality level are matched and integrated, and key information is extracted. These information is expressed and stored using specific data structures (such as vectors and matrices) to generate an abnormal condition fingerprint database. The abnormal condition fingerprint database comprehensively depicts the performance characteristics of various abnormal conditions and serves as an important knowledge base for abnormal diagnosis, prediction, and optimization. The operating parameters of abnormal conditions come from the first performance fingerprint database and the typical operating condition parameter database.

[0235] For different abnormal conditions, we match the preset abnormal handling strategies to form an abnormal condition handling strategy library. We usually adopt the idea of ​​hierarchical management and tailor differentiated handling solutions:

[0236] For severely abnormal operating conditions, it is necessary to formulate systematic optimization projects, carry out equipment technology transformation and management innovation, fundamentally eliminate abnormalities and improve unit performance.

[0237] Moderately abnormal operating conditions need to be listed as daily key focus areas, and online monitoring and trend warnings need to be strengthened to promptly detect and control performance deterioration and prevent abnormalities from escalating.

[0238] Mild abnormal operating conditions can be included in the scope of routine management improvements. By optimizing operating processes, adjusting control parameters, and other means, performance deviations can be reduced and abnormal hidden dangers can be eliminated.

[0239] In addition, it is necessary to establish a long-term mechanism for the graded handling of abnormal working conditions, formulate judgment criteria for abnormal upgrades and releases, clarify the handling procedures, division of responsibilities, and assessment requirements for abnormal working conditions at all levels, form a normalized abnormal management and control system, and continuously optimize abnormal handling capabilities.

[0240] After completing the above abnormality diagnosis steps, gas-steam combined cycle unit managers can clearly understand the unit's current performance shortcomings, the causes of abnormalities, and their severity. This allows them to tailor management resources to the specific symptoms and achieve precise performance improvements. The abnormal condition fingerprint library and treatment strategy library, as a systematic knowledge base, guide subsequent online monitoring, trend prediction, and proactive optimization of unit performance, laying a solid foundation for full lifecycle management.

[0241] Step S3000: Perform online warning of abnormal operating conditions for the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, match the abnormal operating condition handling strategy library to generate the unit's abnormal operating condition handling plan.

[0242] Furthermore, step S3000 includes:

[0243] Step S3100: Perform online warning of abnormal operating conditions on the unit and output abnormality diagnosis results;

[0244] Furthermore, if Figure 7 As shown, step S3100 includes:

[0245] Step S3110: collecting real-time operating parameters of the unit, and obtaining real-time predicted performance indicators based on the real-time operating parameters and the unit performance prediction model;

[0246] Step S3120: Generate a real-time performance fingerprint of the unit based on the real-time operating condition parameters and the real-time predicted performance indicators;

[0247] Step S3130: perform similarity matching between the real-time performance fingerprint and the abnormal working condition fingerprint library, and output the matching result.

[0248] Furthermore, step S3130 includes:

[0249] Step S3131, calculating the similarity between the real-time performance fingerprint and each abnormal operating condition fingerprint in the abnormal operating condition fingerprint library, and generating a second similarity sequence SI;

[0250] Step S3132: Find the maximum value from the second similarity sequence SI, recorded as SI A, A represents the abnormal condition fingerprint with the highest similarity to the real-time performance fingerprint in the abnormal condition fingerprint library;

[0251] Step S3133, SI A Compared with the preset abnormality judgment threshold θ1, if SI A <θ1, the current real-time operating condition is determined to be a stable operating condition; otherwise, it is determined to be an abnormal operating condition, and the abnormal level of the current real-time operating condition is determined according to the abnormal level of the abnormal operating condition corresponding to the abnormal operating condition fingerprint A.

[0252] Specifically, step S3130 determines whether the current real-time operating condition is abnormal and, if so, determines its abnormality level. This step calculates the similarity between the real-time performance fingerprint and samples in the abnormal condition fingerprint library to identify the abnormal condition most similar to the current condition, thereby determining whether it is abnormal and the degree of abnormality.

[0253] In step S3131, a similarity measurement method, such as Euclidean distance or cosine similarity, is used to calculate the similarity between the real-time performance fingerprint and each fingerprint in the abnormal operating condition fingerprint library, thereby obtaining a second similarity sequence SI. The greater the similarity (or the smaller the distance), the more similar the real-time operating condition is to the abnormal operating condition, and the more likely it is to belong to the abnormal type.

[0254] For example, the Euclidean distances between the real-time performance fingerprint Pt and the three fingerprints P1, P2, and P3 in the abnormal working condition fingerprint library are 2, 5, and 3 respectively, and the corresponding cosine similarities are 0.8, 0.5, and 0.7 respectively. Then the second similarity sequence SI = (0.8, 0.5, 0.7).

[0255] Step S2132 selects the abnormal fingerprint with the largest similarity from SI as the best match for the current real-time working condition. Continuing with the above example, the maximum similarity 0.8 is selected from SI = (0.8, 0.5, 0.7), and the corresponding abnormal working condition fingerprint is P1, so A = 1. Step S3133 introduces the abnormality judgment threshold θ1, and by comparing the maximum similarity SI A and θ1, judge whether the current working condition is abnormal. A <θ1, indicating that the current working condition has a low similarity with any known abnormal working condition and can be determined as a stable working condition; otherwise, if SI A ≥θ1, the similarity between the current working condition and the abnormal working condition fingerprint A exceeds the threshold, and it is determined to be an abnormal working condition, and the abnormality level is the same as the abnormal working condition of fingerprint A.

[0256] Setting the anomaly threshold θ1 requires a trade-off between anomaly detection sensitivity and false alarm rate: a larger θ1 results in lower detection sensitivity, a higher probability of missed detection, but a lower false alarm rate; a smaller θ1 results in higher sensitivity, which detects more anomalies but increases the false alarm rate. θ1 can be set empirically or optimized through data analysis.

[0257] For example, the abnormality judgment threshold θ1 is set to 0.75, and the similarity SI between the current real-time performance fingerprint and the most similar abnormal condition (number A=1) is A is 0.8, due to SI A =0.8>θ1=0.75, so the current working condition is judged to be abnormal, and the abnormality level is the same as abnormal working condition 1. A =0.7<θ1=0.75, then although the current operating condition is most similar to abnormal operating condition 1, the similarity is lower than the threshold and it is still determined to be a stable operating condition.

[0258] In summary, step S3130 quickly determines whether the current operating condition is abnormal and what type of abnormality it belongs to by similarity matching the real-time performance fingerprint with the abnormal operating condition fingerprint library, significantly improving the efficiency of real-time diagnosis of abnormal operating conditions. This method fully utilizes the abnormal operating condition fingerprint library knowledge base constructed in step S2300, enabling real-time abnormality diagnosis to be achieved through similarity matching rather than relying on modeling and calculation. This method offers advantages such as simple calculation, timely response, and strong interpretability.

[0259] Step S3100 utilizes the unit's real-time operating parameters and performance prediction model to dynamically generate a real-time fingerprint reflecting the unit's current performance status. This fingerprint is then matched against a database of abnormal operating fingerprints to provide real-time early warning and diagnosis of abnormal unit conditions. This allows for the timely detection of performance anomalies that occur during unit operation, accurately pinpointing their type and severity, and providing a reliable basis for subsequent abnormality resolution.

[0260] Step S3110 is the data preparation stage for generating the real-time performance fingerprint. First, the unit's data acquisition system is used to obtain various real-time operating parameters, including load, ambient temperature, pressure, and flow. These parameters form the basis for evaluating the unit's real-time performance. The load parameter reflects the unit's power output level and is a key factor influencing other parameters. Ambient temperature affects the gas turbine's inlet density and power output. Combustion chamber pressure, main steam pressure, and reheat steam pressure reflect the operating status of the unit's cycle. Fuel flow, feedwater flow, and condensate flow reflect the unit's material balance. Furthermore, exhaust gas temperature and main steam temperature must be collected to determine the actual status of the unit's equipment.

[0261] After obtaining the real-time operating parameters, they are input into the unit performance prediction model constructed in step S1200 to predict real-time performance indicators, such as power generation efficiency, heat rate, power supply coal consumption, etc. These indicators are key parameters for measuring the actual performance of the unit, but they are often impossible to measure directly and must be predicted with the help of models. The unit performance prediction model has been trained and optimized with historical operating data. It can extract deep performance-related features from complex operating parameters and establish a nonlinear mapping relationship between operating conditions and performance. Therefore, the model can more accurately predict the performance of the unit under the current operating conditions based on the input real-time operating parameters. This model-based dynamic performance prediction can make up for the shortcomings of measured performance indicators and provide a more comprehensive and objective performance evaluation perspective.

[0262] Step S3120 extracts features and fuses the acquired real-time operating parameters and predicted performance indicators, ultimately forming a fingerprint that reflects the unit's real-time performance. The key to this step is designing a reasonable fingerprint generation algorithm so that the generated real-time performance fingerprint accurately and comprehensively describes the unit's performance characteristics, facilitating subsequent similarity matching and anomaly detection.

[0263] Real-time performance fingerprints are usually represented in the form of vectors or matrices. Taking vector representation as an example, various operating parameters and performance indicators can be divided into several dimensions, and the value of each dimension corresponds to the real-time value of a parameter or indicator. Different parameters and indicators have different importance in judging the performance of the unit. When constructing the fingerprint vector, necessary feature selection and weight assignment are required. Based on a combination of mechanism analysis and data-driven methods, the parameters with the most significant impact on performance can be screened out, and the importance of different parameters can be measured using methods such as information gain and chi-square test. They are quantified as weight coefficients and assigned to the corresponding dimensions of the fingerprint vector.

[0264] Because the dimensions and numerical ranges of different parameters and indicators vary significantly, real-time data must be normalized before generating fingerprints to eliminate dimensionality effects. Common normalization methods include Min-Max normalization and Z-score normalization. Normalization maps the values ​​of each dimension to the same scale, facilitating the calculation of similarity between different fingerprints.

[0265] The real-time performance fingerprint must inherit the real-time characteristics of operating parameters, ensuring synchronization with unit status updates. It must also take into account the comprehensive characteristics of performance indicators, objectively reflecting the unit's inherent performance. This organic integration of the two enables the performance fingerprint to consistently record and characterize the changing trajectory of unit performance as real-time operating conditions evolve. Therefore, the real-time performance fingerprint can be seen as a bridge that maps and unifies the unit's external manifestations and internal mechanisms, dynamically depicting the unit's consistent performance status.

[0266] Step S3130 uses mathematical methods to measure the proximity between the real-time performance fingerprint and typical abnormal fingerprints in the abnormal condition fingerprint library, determining whether the unit's current operating conditions are abnormal and diagnosing the type and severity of the abnormality. This step is the core of the entire abnormality warning and diagnosis process, and the scientific nature and accuracy of its matching algorithm directly determine the effectiveness of warning and diagnosis.

[0267] The abnormal operating condition fingerprint library is a pre-built dataset containing performance fingerprints for various known abnormal operating conditions. Each abnormal fingerprint corresponds to a specific abnormal operating condition, such as performance anomalies caused by scaling, wear, or component failure. These fingerprints are derived from actual abnormal cases that have occurred in the unit's history, or from theoretical deductions based on mechanistic analysis and empirical experience. Abnormal fingerprints and real-time fingerprints use the same data structure and feature representation, subjecting them to the same criteria for similarity measurement.

[0268] The similarity matching process calculates the "distance" between the real-time fingerprint and each fingerprint in the anomaly fingerprint database, and uses this distance to determine the degree of similarity between the two. The smaller the distance, the higher the similarity, and the closer the current operating condition is to the anomaly. The "distance" here can be measured using a variety of mathematical methods, including Euclidean distance, Manhattan distance, Chebyshev distance, and Mahalanobis distance. Different distance metrics reflect different similarity criteria. For example, Euclidean distance measures absolute numerical differences, while Manhattan distance measures cumulative numerical differences. The appropriate metric should be selected based on actual engineering experience. During the calculation process, fingerprints are typically normalized to eliminate the influence of different feature dimensions.

[0269] The output of the matching results generally includes three aspects: first, the matching degree of the abnormal working condition, that is, the degree of closeness between the real-time fingerprint and the most similar abnormal fingerprint, usually expressed as a value from 0 to 1, the closer to 1, the higher the matching degree; second, the abnormality level, which divides the abnormality into mild, moderate, and severe according to the matching degree, reflecting the severity of the problem; third, the most similar abnormal fingerprint, which provides a reference basis for diagnosing the abnormality.

[0270] Overall, step S3100 makes full use of the multi-source heterogeneous real-time data of the unit and combines it with pre-established mechanism knowledge. Through the fusion of data-driven and knowledge-driven, it dynamically constructs fingerprint features that can objectively reflect the real-time performance status of the unit. On this basis, intelligent algorithms such as similarity matching are used to compare and map the real-time operating conditions of the unit with typical abnormal operating conditions, automatically realizing abnormal warning and diagnosis. Compared with traditional abnormal diagnosis methods that rely on manual experience and threshold rules, this step can greatly improve the intelligence level and real-time performance of diagnosis, and capture abnormal conditions more accurately, quickly and comprehensively, reducing omissions and misjudgments. At the same time, this method fully taps the value of historical data, uses data to drive the optimization of the knowledge base, uses knowledge feedback to guide practice, realizes the solidification and precipitation of experience knowledge and real-time reuse, and safeguards the safe and efficient operation of the unit.

[0271] Step S3200: According to the abnormal diagnosis result, match the abnormal operating condition handling strategy library and generate the abnormal operating condition handling plan of the unit.

[0272] Specifically, step S3200 automatically matches the pre-built abnormal operating condition handling strategy library based on the abnormal diagnosis result output by step S3100, forms a handling plan and control instructions for the current abnormal state of the unit, and sends them to the control system for execution, thereby realizing adaptive control and closed-loop management of the abnormal operating conditions of the unit.

[0273] The abnormal condition handling strategy library is a pre-organized and designed set of strategies corresponding to the abnormal condition fingerprint library. Each abnormal fingerprint corresponds to several handling measures, typically derived by experts based on mechanism analysis, simulation experiments, and operational experience. These measures represent the optimal control strategy for abnormal conditions. These measures include adjusting operating parameters, limiting load change rates, and changing control methods to mitigate the adverse effects of abnormal conditions on the unit and prevent the abnormality from worsening.

[0274] The process of matching the anomaly handling strategy library is similar to the anomaly fingerprint matching principle in step S3130. Based on the anomaly type diagnosis, the control action to be taken in response to the anomaly is further clarified. By calculating the correlation between the anomaly diagnosis results and the various handling measures in the strategy library, the control strategy that best matches the current anomaly state is selected. This "one-to-many" mapping mechanism prioritizes anomalies and ensures precise implementation of handling measures.

[0275] Based on the matching results, an abnormality handling plan is automatically generated, which mainly includes the following aspects: First, the diagnosed abnormality type, matching degree, abnormality level, etc., which comprehensively describe the current abnormal status of the unit; second, the control measures that should be taken, such as adjusting the set values ​​of parameters such as fuel quantity, primary air volume, and water flow, changing the operating mode of the AGC (automatic generation control) system, etc., so that the unit can return to normal as soon as possible; third, the control instructions, which are usually a series of standard data frames, including control objects, operation types, values, time and other elements, which can be directly sent to control systems such as DCS for execution.

[0276] The process of generating a response plan typically follows the principle of prioritizing anomaly levels. This means that the severity of the diagnosed anomaly is prioritized for the corresponding response level. Anomalies are categorized as mild, moderate, and severe, each corresponding to a different level of response intensity and speed. Mild anomalies primarily require local adjustments, such as fine-tuning operating parameters. Moderate anomalies require more drastic adjustments, such as load limiting and removing auxiliary controls. Severe anomalies require a decisive safety shutdown. Furthermore, the coordination of different measures must be assessed to avoid strategic conflicts.

[0277] After the plan is generated, control instructions are promptly issued to the DCS system, which coordinates and controls the actuators of subsystems such as combustion, feedwater, and steam turbines, ultimately achieving adaptive adjustment to abnormal operating conditions. Simultaneously, changes in unit status are closely monitored to evaluate the effectiveness of the response. If necessary, simulation models are used for post-evaluation to support dynamic optimization of the plan.

[0278] The matching and adaptive control of the abnormality handling strategy library achieves a closed-loop system for abnormality management. Diagnostic results trigger early warnings, while rapid feedback provides regulatory instructions. This seamless integration of diagnosis and decision-making creates a real-time closed-loop system that covers the entire process from abnormality generation to resolution. Mild and moderate abnormalities can often be resolved through automated adjustments, eliminating the burden of manual oversight. Even severe abnormalities can be addressed immediately with an emergency response, minimizing losses.

[0279] It's important to note that while this solution achieves a high degree of automation and intelligence, human interaction and decision-making are still required at key points. This is because unit anomalies are often accompanied by numerous uncertainties, especially extreme or complex anomalies, whose diagnosis and resolution require human understanding and judgment. Therefore, the optimal application model for this solution is the deep integration of machine intelligence and expert experience, forming a human-machine collaborative anomaly management paradigm. Drawing on the principle of machine recommendation and human review, the system can provide a set of alternative resolution options, which experts can then make based on the actual situation.

[0280] Overall, step S3000 utilizes a series of key technologies, including real-time fingerprint extraction, similarity matching, and strategy recommendation, to establish an effective abnormal unit operating condition warning and resolution solution. Based on massive amounts of measured and simulated data, driven by mechanistic models and expert experience, this solution integrates multiple intelligent technologies, including data analysis, fault diagnosis, and intelligent decision-making, to streamline the entire process of abnormality monitoring, analysis, warning, diagnosis, and resolution, achieving an end-to-end closed-loop system for abnormal unit management. This not only significantly improves the speed and accuracy of abnormality diagnosis and resolution, reducing accident losses, but also significantly reduces manual oversight and eases the workload of operations and maintenance personnel. Most importantly, while drawing on human experience and wisdom, this solution also forms new knowledge models through inductive deduction, achieving a virtuous cycle of human-computer interaction, knowledge accumulation, and optimization. It is foreseeable that with the continued accumulation of operating data and continuous model iteration, the performance of this solution will continue to improve, making it a key enabling technology for building smart power plants and achieving full unit lifecycle management.

[0281] Example 2

[0282] This embodiment provides a full life cycle management system for a gas-steam combined cycle unit based on embodiment 1, such as Figure 8 Shown, including:

[0283] Fingerprint library construction module: used to build a typical operating condition parameter database and a unit performance prediction model, and build a first performance fingerprint library based on the typical operating condition parameter database and the unit performance prediction model; obtain second historical operating data, and build a second performance fingerprint library based on the second historical operating data;

[0284] Performance evaluation module: generates a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generates a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library;

[0285] Working condition diagnosis module: identifies abnormal working conditions based on the first performance evaluation report and the second performance evaluation report, and generates an abnormal working condition fingerprint library and an abnormal working condition handling strategy library;

[0286] Online analysis module: used to provide online warning of abnormal operating conditions of the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, it matches the abnormal operating condition handling strategy library and generates the unit's abnormal operating condition handling plan.

[0287] In the fingerprint library construction module, the construction of the typical operating condition parameter database includes:

[0288] Step S1110, obtaining first-level typical operating condition parameters under n1 typical operating conditions of the gas-steam combined cycle unit; n1 is the total number of typical operating conditions;

[0289] Step S1120, based on The first level typical working condition parameters under the typical working condition are calculated to obtain the The second level typical working condition parameters under typical working conditions; 1≤ ≤n1;

[0290] Step S1130: The first level typical working condition parameters and the second level typical working condition parameters under the typical working condition The second level typical working condition parameters under the typical working condition constitute the Record the working parameters of typical working conditions;

[0291] Step S1140 : Record the operating parameters of all typical operating conditions to form a typical operating parameter database; the typical operating parameter database includes n1 groups of operating parameter records.

[0292] In the fingerprint library construction module, the construction of the unit performance prediction model includes:

[0293] Step S1210: Acquire historical operating data of the gas-steam combined cycle unit within a historical time period T1, marked as first historical operating data; the first historical operating data includes a first historical operating condition parameter and a first historical performance index;

[0294] Step S1220: construct and train a unit performance prediction model based on the first historical operation data.

[0295] In the fingerprint library construction module, constructing the first performance fingerprint library includes:

[0296] Step S1310, obtaining predicted performance indicators under n1 typical operating conditions based on typical operating condition parameters and the unit performance prediction model;

[0297] Step S1320: Based on the typical working condition parameter database Typical working condition parameter records and The predicted performance index under typical working conditions is generated Performance fingerprint under typical working conditions;

[0298] Step S1330: The performance fingerprints under all typical working conditions are combined into a first performance fingerprint library.

[0299] In the fingerprint library construction module, constructing the second performance fingerprint library includes:

[0300] Step S1410, obtaining historical operating data of the gas-steam combined cycle unit within a historical time period T2, marked as second historical operating data; the second historical operating data includes a second historical operating condition parameter and a second historical performance index; the second historical operating data includes n4 groups of historical operating condition parameter records, where n4 is the total number of historical operating conditions; T2≠T1;

[0301] Step S1420: generating a historical performance fingerprint based on the second historical operating condition parameter and the second historical performance index;

[0302] Step S1430: All historical performance fingerprints are combined into a second performance fingerprint library.

[0303] In the performance evaluation module, generating the first performance evaluation report includes:

[0304] Step S2110: traverse the performance fingerprint of each typical working condition in the first performance fingerprint library and extract the performance fingerprint of the first typical working condition. The prediction performance index of the typical working condition is calculated. The predicted performance index of a typical working condition is consistent with the preset Deviation from the performance evaluation benchmark for a typical working condition;

[0305] Step S2120, The deviation of all the prediction performance indicators of the typical working condition is weighted averaged to obtain the Comprehensive performance deviation score of typical working conditions;

[0306] Step S2130: Compare the comprehensive performance deviation scores of all typical operating conditions with a preset deviation threshold, and mark the typical operating conditions that are higher than the deviation threshold as weak working conditions;

[0307] Step S2140: Generate a first performance evaluation report.

[0308] In the performance evaluation module, generating the second performance evaluation report includes:

[0309] Step S2210: traverse all historical performance fingerprints in the second performance fingerprint library, perform similarity matching with the first performance fingerprint library, and identify the operating condition of the unit; the operating condition of the unit is divided into typical operating conditions and atypical operating conditions;

[0310] Step S2220: Based on the identification result of the unit's operating condition, analyze the cumulative operating time and start-stop frequency of the unit under each typical and atypical operating condition;

[0311] Step S2230: Effective utilization of the computer group under various typical and atypical operating conditions based on the accumulated operating time and start / stop frequency;

[0312] Step S2240, analyzing the changing trend of the effective utilization rate of the unit under various atypical operating conditions during the T3 period;

[0313] Step S2250: horizontally compare the utilization differences of similar units under various historical operating conditions to identify the performance shortcomings of the units;

[0314] Step S2260: Generate a second performance evaluation report.

[0315] The step S2210 includes:

[0316] Step S2211, calculate the second performance fingerprint library The historical performance fingerprint and the first performance fingerprint library Similarity of performance fingerprints , get the similarity sequence; 1≤ ≤n2,1≤ ≤n3; n2 is the number of historical performance fingerprints in the second performance fingerprint library, n3 is the number of performance fingerprints in the first performance fingerprint library, n3=n1, n2=n4;

[0317] Step S2212: If is the maximum value, and ≥ , then it is considered that The historical operating condition corresponding to the historical performance fingerprint belongs to Typical working conditions corresponding to each performance fingerprint; is the preset similarity threshold;

[0318] Step S2213: If is the maximum value, and < , then the The historical operating conditions corresponding to the historical performance fingerprints are marked as atypical operating conditions.

[0319] In the working condition diagnosis module, the identification of abnormal working conditions and the generation of abnormal working condition fingerprint library and abnormal working condition handling strategy library include:

[0320] Step S2310: Based on the comprehensive performance deviation score of the typical operating condition, the typical operating condition is divided into four abnormality levels: normal, slightly abnormal, moderately abnormal, and severely abnormal. Typical operating conditions with slightly abnormal, moderately abnormal, and severely abnormal conditions are marked as abnormal operating conditions; and normal typical operating conditions are marked as stable operating conditions.

[0321] Step S2320, extracting the performance indicator features of the abnormal operating condition from the performance indicators of the abnormal operating condition;

[0322] Step S2330: Generate an abnormal operating condition fingerprint library based on the operating condition parameters, performance indicator characteristics and abnormality level of the abnormal operating condition.

[0323] Step S2340 : For abnormal operating conditions of different abnormal levels, configure abnormal operating condition handling strategies based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition handling strategy library.

[0324] In the online analysis module, the abnormal operating condition online warning of the unit and the output of abnormal diagnosis results include:

[0325] Step S3110: collecting real-time operating parameters of the unit, and obtaining real-time predicted performance indicators based on the real-time operating parameters and the unit performance prediction model;

[0326] Step S3120: Generate a real-time performance fingerprint of the unit based on the real-time operating condition parameters and the real-time predicted performance indicators;

[0327] Step S3130: perform similarity matching between the real-time performance fingerprint and the abnormal working condition fingerprint library, and output the matching result.

[0328] The step S3130 includes:

[0329] Step S3131, calculating the similarity between the real-time performance fingerprint and each abnormal operating condition fingerprint in the abnormal operating condition fingerprint library, and generating a second similarity sequence SI;

[0330] Step S3132: Find the maximum value from the second similarity sequence SI, recorded as SI A , A represents the abnormal condition fingerprint with the highest similarity to the real-time performance fingerprint in the abnormal condition fingerprint library;

[0331] Step S3133, SI A Compared with the preset abnormality judgment threshold θ1, if SI A <θ1, the current real-time operating condition is determined to be a stable operating condition; otherwise, it is determined to be an abnormal operating condition, and the abnormal level of the current real-time operating condition is determined according to the abnormal level of the abnormal operating condition corresponding to the abnormal operating condition fingerprint A.

[0332] Example 3

[0333] This embodiment discloses an electronic device that may include one or more processors and one or more memories. The memories may store computer-readable code that, when executed by the one or more processors, may implement the above-described method for managing the lifecycle of a gas-steam combined cycle unit.

[0334] The method or system according to the embodiments of the present application can also be implemented using the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, read-only memory (ROM), random access memory (RAM), a communication port connected to a network, input / output components, a hard disk, and the like. A storage device in the electronic device, such as a ROM or hard disk, may store the full lifecycle management method for a gas-steam combined cycle unit provided herein. The full lifecycle management method for a gas-steam combined cycle unit may, for example, include: constructing a typical operating parameter database and a unit performance prediction model; constructing a first performance fingerprint library based on the typical operating parameter database and the unit performance prediction model; obtaining second historical operating data and constructing a second performance fingerprint library based on the second historical operating data; generating a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generating a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; identifying abnormal operating conditions based on the first and second performance evaluation reports, generating an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library; performing online abnormal operating condition warnings for the unit and outputting abnormality diagnosis results; and matching the abnormal operating condition handling strategy library with the abnormality diagnosis results to generate a handling plan for abnormal operating conditions for the unit.

[0335] Furthermore, the electronic device may further include a user interface. Of course, the architecture disclosed in the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0336] Example 4

[0337] This embodiment discloses a computer-readable storage medium storing computer-readable instructions. When executed by a processor, the computer-readable instructions can execute the full lifecycle management method for a gas-steam combined cycle unit according to the embodiments of this application. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, and flash memory.

[0338] In addition, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions capable of being executed by a processor to execute instructions corresponding to the method steps provided herein, such as: constructing a typical operating condition parameter database and a unit performance prediction model, and constructing a first performance fingerprint library based on the typical operating condition parameter database and the unit performance prediction model; obtaining second historical operating data and constructing a second performance fingerprint library based on the second historical operating data; generating a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generating a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; identifying abnormal operating conditions based on the first performance evaluation report and the second performance evaluation report, and generating an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library; performing online abnormal operating condition warnings for the unit and outputting abnormality diagnosis results; and matching the abnormal operating condition handling strategy library with the abnormality diagnosis results to generate a unit abnormal operating condition handling plan. When executed by a central processing unit (CPU), this computer program performs the functions defined in the method of the present application.

[0339] The methods, systems, and devices of the present application may be implemented in many ways. For example, the methods, systems, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present application. Therefore, the present application also covers recording media that store programs for executing the methods according to the present application.

[0340] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0341] The above-described specific embodiments further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is merely a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A full life cycle management method for a gas-steam combined cycle unit, characterized in that: The method comprises: Constructing a typical operating parameter database and a unit performance prediction model, and constructing a first performance fingerprint library based on the typical operating parameter database and the unit performance prediction model; obtaining second historical operating data, and constructing a second performance fingerprint library based on the second historical operating data; Generate a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generate a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; identify abnormal operating conditions based on the first performance evaluation report and the second performance evaluation report, and generate an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library; Generating the first performance evaluation report includes: traversing the performance fingerprint of each typical working condition in the first performance fingerprint library, extracting the first The prediction performance index of the typical working condition is calculated. The predicted performance index of a typical working condition is consistent with the preset Deviation of the performance evaluation benchmark for the first typical working condition; The deviation of all the prediction performance indicators of the typical working condition is weighted averaged to obtain the The comprehensive performance deviation scores of typical working conditions are compared with a preset deviation threshold, and the typical working conditions that are higher than the deviation threshold are marked as short board working conditions, and a first performance evaluation report is generated; Generating the second performance evaluation report includes: traversing all historical performance fingerprints in the second performance fingerprint library, performing similarity matching with the first performance fingerprint library, and identifying the operating conditions of the unit; dividing the operating conditions of the unit into typical operating conditions and atypical operating conditions; based on the identification results of the operating conditions of the unit, analyzing the cumulative operating time and start-stop frequency of the unit under each typical operating condition and atypical operating condition; calculating the effective utilization rate of the unit under each typical operating condition and atypical operating condition based on the cumulative operating time and start-stop frequency; and analyzing the changing trend of the effective utilization rate of the unit under each atypical operating condition during the T3 period; Provide online warning of abnormal operating conditions for the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, match the abnormal operating condition handling strategy library and generate the unit's abnormal operating condition handling plan.

2. The full life cycle management method of a gas-steam combined cycle unit according to claim 1, characterized in that: The construction of a typical operating condition parameter database includes: Obtain the first-level typical operating parameters of the gas-steam combined cycle unit under n1 typical operating conditions; n1 is the total number of typical operating conditions; Based on the The first level typical working condition parameters under the typical working condition are calculated to obtain the The second level typical working condition parameters under typical working conditions; 1≤ ≤n1; The first The first level typical working condition parameters and the second level typical working condition parameters under the typical working condition The second level typical working condition parameters under the typical working condition constitute the Record the working parameters of typical working conditions; The operating condition parameter records of all typical operating conditions constitute a typical operating condition parameter database; the typical operating condition parameter database includes n1 groups of operating condition parameter records.

3. The full life cycle management method of a gas-steam combined cycle unit according to claim 2, characterized in that: The constructing of the unit performance prediction model comprises: Obtain historical operating data of the gas-steam combined cycle unit within a historical time period T1, marking it as first historical operating data; construct and train a unit performance prediction model based on the first historical operating data; The constructing of the first performance fingerprint library includes: According to the typical operating parameters and the unit performance prediction model, the predicted performance indicators under n1 typical operating conditions are obtained; Based on the typical working condition parameter database Typical working condition parameter records and The predicted performance index under typical working conditions is generated Performance fingerprint under typical working conditions; The performance fingerprints under all typical working conditions constitute a first performance fingerprint library.

4. The full life cycle management method of a gas-steam combined cycle unit according to claim 3, characterized in that: The second historical operation data is the historical operation data of the gas-steam combined cycle unit within a historical time period T2, where T2≠T1; the second historical operation data includes n4 groups of historical operating condition parameter records, where n4 is the total number of historical operating conditions; The second historical operating data includes a second historical operating condition parameter and a second historical performance index; The constructing of the second performance fingerprint library includes: generating a historical performance fingerprint based on the second historical operating condition parameter and the second historical performance index; All historical performance fingerprints are combined into a second performance fingerprint library.

5. The full life cycle management method of a gas-steam combined cycle unit according to claim 4, characterized in that: The identification unit's operating condition includes: Calculate the second performance fingerprint library The historical performance fingerprint and the first performance fingerprint library Similarity of performance fingerprints , get the similarity sequence; 1≤ ≤n2,1≤ ≤n3; n2 is the number of historical performance fingerprints in the second performance fingerprint library, n3 is the number of performance fingerprints in the first performance fingerprint library, n3=n1, n2=n4; If the similarity sequence is the maximum value, and ≥ , then it is considered that The historical operating condition corresponding to the historical performance fingerprint belongs to Typical working conditions corresponding to each performance fingerprint; is the preset similarity threshold; If the similarity sequence is the maximum value, and < , then the The historical operating conditions corresponding to the historical performance fingerprints are marked as atypical operating conditions.

6. The full life cycle management method of a gas-steam combined cycle unit according to claim 2, characterized in that: The identification of abnormal operating conditions and the generation of an abnormal operating condition fingerprint library and an abnormal operating condition handling strategy library include: According to the comprehensive performance deviation score of typical working conditions, the typical working conditions are divided into four abnormal levels: normal, slightly abnormal, moderately abnormal, and severely abnormal. Typical working conditions with slightly abnormal, moderately abnormal, and severely abnormal conditions are marked as abnormal working conditions; normal typical working conditions are marked as stable working conditions. Extracting performance indicator characteristics of abnormal working conditions; generating an abnormal working condition fingerprint library based on the working condition parameters, performance indicator characteristics and abnormality level of the abnormal working condition; For abnormal operating conditions of different abnormal levels, abnormal operating condition handling strategies are configured according to the first performance evaluation report and the second performance evaluation report, and an abnormal operating condition handling strategy library is generated.

7. The full life cycle management method of a gas-steam combined cycle unit according to claim 6, characterized in that: The online early warning of abnormal operating conditions of the unit includes: Collect the real-time operating parameters of the unit and obtain the real-time predicted performance indicators based on the real-time operating parameters and the unit performance prediction model; Generate the real-time performance fingerprint of the unit based on real-time operating parameters and real-time predicted performance indicators; Calculate the similarity between the real-time performance fingerprint and each abnormal working condition fingerprint in the abnormal working condition fingerprint library to generate a second similarity sequence SI; Find the maximum value from the second similarity sequence SI, recorded as SI A , A represents the abnormal condition fingerprint with the highest similarity to the real-time performance fingerprint in the abnormal condition fingerprint library; SI A Compared with the preset abnormality judgment threshold θ1, if SI A <θ1, the current real-time operating condition is determined to be a stable operating condition; otherwise, it is determined to be an abnormal operating condition, and the abnormal level of the current real-time operating condition is determined according to the abnormal level of the abnormal operating condition corresponding to the abnormal operating condition fingerprint A.

8. A full life cycle management system for a gas-steam combined cycle unit, which is used to implement the full life cycle management method for a gas-steam combined cycle unit according to any one of claims 1 to 7, characterized in that: The system comprises: Fingerprint library construction module: used to build a typical operating condition parameter database and a unit performance prediction model, and build a first performance fingerprint library based on the typical operating condition parameter database and the unit performance prediction model; obtain second historical operating data, and build a second performance fingerprint library based on the second historical operating data; Performance evaluation module: generates a first performance evaluation report based on the first performance fingerprint library and a preset performance evaluation benchmark; generates a second performance evaluation report based on the first performance fingerprint library and the second performance fingerprint library; Working condition diagnosis module: identifies abnormal working conditions based on the first performance evaluation report and the second performance evaluation report, and generates an abnormal working condition fingerprint library and an abnormal working condition handling strategy library; Online analysis module: used to provide online warning of abnormal operating conditions of the unit and output abnormal diagnosis results; based on the abnormal diagnosis results, it matches the abnormal operating condition handling strategy library and generates the unit's abnormal operating condition handling plan.

Images