Working condition optimizing method and system for operation of thermal power generating unit

Abstract

The invention relates to the technical field of thermal power generating unit operation optimization, and discloses a thermal power generating unit operation condition optimization method and system, and the method comprises the steps: obtaining first historical operation data and second historical operation data of a unit, carrying out the abnormal processing of the data, and obtaining first processing data and second processing data; performing steady-state screening on the first processing data to obtain first steady-state data; performing working condition division on the first steady-state data by using a fuzzy C-means clustering algorithm to obtain first working condition categories, screening out a first optimal working condition category from the first working condition categories, and extracting corresponding operation parameters to form a first optimal working condition table; on-line working condition judgment and optimization are carried out on the second processing data, and the optimal working condition table is updated; the system comprises a data acquisition module, a data processing module, a steady-state screening module, a working condition division module, an optimal working condition screening module and an online working condition optimization module. The operation efficiency and economical efficiency of the thermal power generating unit can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power unit operation optimization, and more specifically, to a method and system for optimizing the operating conditions of a thermal power unit. Background Art

[0002] As the main power generation equipment in my country's power system, the operating efficiency and economy of thermal power units directly affect the safety, stability and economic operation of the power system. How to optimize the operating conditions of thermal power units and improve the efficiency of units according to the actual operating status of the units is an urgent problem that thermal power plants need to solve.

[0003] At present, there are some related technologies for optimizing the operation of thermal power units. For example, the Chinese patent with publication number CN109685264A discloses a method, device and computer equipment for optimizing the operation of thermal power units. The method establishes multiple subsystem optimization models based on the operation history data and the coupling degree and importance of the production process, and uses the optimization algorithm to iteratively optimize the model to quickly and effectively obtain the optimal process parameters for the operation of the thermal power unit. Another example is the Chinese patent with authorization announcement number CN108037748B, which discloses a method, device and equipment for optimizing the operation of thermal power units under all working conditions. The method screens historical data by data mining to obtain economic conditions and economic rules that meet the preset conditions, and supplements the missing content in the benchmark condition library through unit tests, and finally adds the optimized economic rules to the benchmark condition library to guide the subsequent unit operation.

[0004] Although the above patents provide some methods for optimizing the operation of thermal power units, they lack online real-time optimization capabilities and cannot perform dynamic optimization based on the real-time operating conditions of the units. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for optimizing the operating conditions of a thermal power unit.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for optimizing the operating conditions of a thermal power unit comprises:

[0008] Acquire first historical operation data and second historical operation data of the unit, perform abnormal processing on the acquired first historical operation data and second historical operation data, and obtain first processed data and second processed data;

[0009] Perform steady-state screening on the first processed data to obtain first steady-state data; use a fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions to obtain a first operating condition category, screen out a first optimal operating condition category from the first operating condition category, and extract corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table;

[0010] Perform online operating condition judgment and optimization on the second processed data, and update the first optimal operating condition table.

[0011] Furthermore, the first historical operation data is the historical operation data of the unit in the last 12 months obtained from the real-time database, with a sampling interval of 1 minute; the second historical operation data is the operation data within the previous 24 hours obtained from the real-time database, with a sampling interval of 1 second.

[0012] Furthermore, the performing exception processing on the acquired first historical operation data and the second historical operation data includes:

[0013] Perform missing value detection on the first historical operation data and the second historical operation data item by item. If the data is missing, it is marked as missing data. If the data is not missing, it is marked as non-missing data. For missing data, the average value of the 5 adjacent data before and after the adjacent time is used to replace it. If there are 2 or more consecutive missing data in the adjacent data, it is marked as unavailable data.

[0014] Perform physical threshold detection on each non-missing data item by item, and set the physical threshold range. If it exceeds the physical threshold range, it is marked as the first abnormal data;

[0015] Perform statistical threshold detection on each data item that does not exceed the physical threshold range, with 3 times the standard deviation as the statistical threshold. If the deviation from the data mean value exceeds 3 times the standard deviation, it is marked as the second abnormal data;

[0016] Perform noise detection on each data item without abnormality detection, calculate the peak-to-peak value of the data, and if the peak-to-peak value exceeds the set peak-to-peak threshold and the short-term variance of the data is higher than 3 times the long-term variance, it is marked as the third abnormal data;

[0017] Unusable data, first abnormal data, second abnormal data and third abnormal data are eliminated to obtain processed first processed data and second processed data.

[0018] Further, the step of performing steady-state screening on the first processed data to obtain the first steady-state data includes:

[0019] Extracting time series data of steady-state parameters from the first processed data, dividing the time series data of each steady-state parameter into a sliding time window of length L, and sliding the window continuously with a step length of 1 sampling point;

[0020] The steady-state index SI is calculated for the data in each time window of each steady-state parameter;

[0021] Calculate SI for multiple steady-state parameters in each time window and take the maximum SI of each SI max As a comprehensive homeostatic index for this time window;

[0022] The average of the timestamps of all sampling points in the sliding time window is used as the comprehensive steady-state index SI max At the corresponding time t, the time series steady-state exponential curve SI is obtained. max (t);

[0023] Set the first steady-state threshold SI th , find out SI max All the values ​​in (t) less than SI th Time interval T s , and T s If the interval is greater than the interval threshold T, the interval is determined to be a steady-state interval; and the first processed data in the steady-state interval is extracted as the first steady-state data.

[0024] Furthermore, the first steady-state data is divided into the first operating condition category by the operating condition classification, which includes:

[0025] Performing data standardization processing on the first steady-state data to obtain standardized steady-state data;

[0026] Initialize the parameters of the fuzzy C-means clustering algorithm, including the number of clusters C, the membership matrix U, the cluster center V, and the maximum number of iterations MAX ITER , objective function termination threshold ε;

[0027] Based on the standardized steady-state data, the membership matrix U and cluster center V are iteratively optimized;

[0028] According to the final membership matrix U, each data point is divided into the category with the largest membership, forming C working condition clusters, namely the first working condition category;

[0029] The iterative optimization of the membership matrix U and the cluster center V includes:

[0030] Based on the current membership matrix U, update the cluster centers V of each category k , k is the index of the number of cluster categories, V k represents the cluster center of the kth class;

[0031] Based on the current cluster center V, update the membership U of the data point to each category kj , U kj Indicates the membership of the jth sample to the kth class;

[0032] Calculate the objective function J and determine whether J is less than ε or the number of iterations reaches MAX ITER , if yes, generate the final membership matrix U; if not, continue iterating.

[0033] Further, the filtering out the first optimal operating condition category from the first operating condition category includes:

[0034] For each operating condition category, calculate its various performance indicators;

[0035] Perform weighted summation on each performance index to obtain the comprehensive performance evaluation value of each working condition category;

[0036] The operating condition category with the best comprehensive performance evaluation value is selected as the first optimal operating condition category.

[0037] Furthermore, the online working condition judgment and optimization for the second processed data includes:

[0038] Perform steady-state judgment on the second processed data. If it is steady-state, divide the second processed data into working conditions to obtain a second working condition category; perform benchmark analysis on the second working condition category and the first optimal working condition table. If the second working condition category is better than the first optimal working condition category, update the first optimal working condition table with the operating parameters corresponding to the second working condition category; if the second working condition category is not better than the first optimal working condition category, maintain the first optimal working condition table unchanged, wait for 24 hours, store the latest 24-hour data as the new second historical operating data in the real-time database, and perform the next round of online working condition judgment and optimization;

[0039] If it is a non-steady state, the first optimal operating condition table is maintained unchanged. After waiting for 24 hours, the latest 24-hour data is stored in the real-time database as the new second historical operation data for the next round of online operating condition judgment and optimization.

[0040] A system for optimizing the operating conditions of a thermal power unit, which is used to implement the above-mentioned method for optimizing the operating conditions of a thermal power unit, and the system comprises:

[0041] Data acquisition module: used to obtain the first historical operation data and the second historical operation data of the unit;

[0042] Data processing module: used for performing exception processing on the acquired first historical operation data and second historical operation data to obtain first processed data and second processed data;

[0043] Steady-state screening module: used for performing steady-state screening on the first processed data to obtain first steady-state data;

[0044] Working condition classification module: used for performing working condition classification on the first steady-state data by using fuzzy C-means clustering algorithm to obtain a first working condition category;

[0045] Optimal operating condition screening module: used for screening out a first optimal operating condition category from the first operating condition category, and extracting corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table;

[0046] Online operating condition optimization module: used to perform online operating condition judgment and optimization on the second processed data, and update the first optimal operating condition table.

[0047] An electronic device comprises a memory, a central processing unit and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method for optimizing the operating conditions of a thermal power unit are implemented.

[0048] A computer-readable storage medium stores a computer program, which, when executed, implements the steps in the above-mentioned method for optimizing the operating conditions of a thermal power unit.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention can timely discover and eliminate abnormal data by acquiring and processing the historical operating data of the unit, thereby ensuring the accuracy and reliability of the data; the fuzzy C-means clustering algorithm is used to divide the steady-state data into operating conditions, so as to accurately identify the operating status of the thermal power unit under different operating conditions, screen out the optimal operating condition, and thus improve the operating efficiency of the unit.

[0051] The present invention can effectively reduce data noise and errors by handling the exceptions of the first and second historical operating data, avoid unnecessary maintenance and overhaul due to data problems, and thus reduce operating costs; through online operating condition judgment and optimization, the optimal operating condition table can be updated in real time to ensure that the thermal power unit is always in the best operating state, reducing energy consumption and equipment wear.

[0052] The present invention adopts advanced data processing methods, including missing value detection, physical threshold detection, statistical threshold detection and noise detection, which can comprehensively and efficiently process data and improve the accuracy and efficiency of data processing; using the sliding time window and steady-state index calculation method, it can accurately screen out steady-state data to ensure the accuracy of subsequent working condition division.

[0053] The present invention can adaptively adjust the cluster center and the membership matrix under different working conditions through the fuzzy C-means clustering algorithm to ensure the accuracy and robustness of the working condition division; the online working condition optimization module can judge and optimize new data in real time, has strong adaptive ability, and can cope with various changes in the operation of thermal power units.

[0054] The present invention can accurately evaluate the comprehensive performance of each operating condition by weighted summing up the performance indicators of each operating condition category, ensuring that the selected optimal operating condition has the best operating performance; the method can effectively apply the performance evaluation results to actual operation, guide the optimal scheduling and operation management of thermal power units, and improve the overall operation effect.

[0055] The system in the present invention includes multiple modules such as data acquisition, data processing, steady-state screening, operating condition division, optimal operating condition screening and online operating condition optimization. The modules cooperate with each other to form a complete operating condition optimization system. The modular design of the system makes it have good scalability and flexibility, and can be adjusted and optimized according to actual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0057] Figure 1 It is a principle flow chart of a method for optimizing the operating conditions of a thermal power unit in the present invention;

[0058] Figure 2 A flow chart showing the principle of performing abnormal data processing on first historical operating data to obtain first processed data in a method for optimizing the operating conditions of a thermal power unit according to the present invention;

[0059] Figure 3 A flow chart of a method for dividing the working condition of the first steady-state data in a working condition optimization method for a thermal power unit operation according to the present invention;

[0060] Figure 4 A flow chart of a method for iteratively optimizing a membership matrix U and a cluster center V in a method for optimizing an operating condition of a thermal power unit according to the present invention;

[0061] Figure 5 A flow chart of a method for forming a first optimal operating condition table from a first operating condition category in an operating condition optimization method for a thermal power unit of the present invention;

[0062] Figure 6 It is a functional module diagram of a system for optimizing the operating conditions of a thermal power unit in the present invention;

[0063] Figure 7 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application;

[0064] Figure 8A schematic diagram of the structure of a computer-readable storage medium provided for one embodiment of the present application. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0066] Example 1

[0067] See also Figure 1 As shown, this embodiment provides a method for optimizing the operating conditions of a thermal power unit, including:

[0068] Step S1000, obtaining first historical operation data and second historical operation data of the unit, performing abnormal processing on the obtained first historical operation data and second historical operation data to obtain first processed data and second processed data;

[0069] Furthermore, step S1000 includes:

[0070] Step S1100, obtaining the historical operation data of the unit in the last 12 months from the real-time database as the first historical operation data, with a sampling interval of 1 minute; obtaining the operation data within the previous 24 hours from the real-time database as the second historical operation data, with a sampling interval of 1 second;

[0071] Specifically, the historical operation data of the unit in the last 12 months is obtained from the real-time database through the data interface API. The data items include main steam pressure, main steam temperature, reheat steam temperature, generator power, furnace negative pressure, induced draft fan inlet static pressure, economizer outlet flue gas temperature, exhaust gas temperature, flame brightness, unit heat consumption, unit coal consumption, unit power generation coal consumption, boiler efficiency, auxiliary machine power, low calorific value of coal entering the furnace, sulfur content of coal entering the furnace, volatile matter of coal entering the furnace, nitrogen oxide emission concentration, sulfur dioxide emission concentration, etc. The sampling interval is 1 minute and stored as the first historical operation data; the operation data of the unit in the last 24 hours is obtained from the real-time database through the same data interface API. The data items are the same as the first historical operation data, the sampling interval is 1 second, and it is stored as the second historical operation data. The use of 12 months of historical data is to cover various operating conditions under different loads, different coal types, and different environmental conditions of the unit, so that the operating condition division is more comprehensive; and the use of operating data within 24 hours is to obtain the current operating condition information of the unit and provide a basis for operating condition optimization.

[0072] A real-time database is a database system specially designed for real-time data processing and analysis, which can capture, store, process and analyze data while it is being generated. The establishment of a real-time database usually involves selecting a suitable software platform, configuring hardware resources, defining data models and establishing data interfaces. Specifically, a real-time database automatically collects data from various data sources, such as sensors and control systems, through a data interface API, to achieve low-latency processing and high-throughput storage of data. This database system can provide real-time or near-real-time data access and analysis to ensure the timeliness of data. In the operating condition optimization method of thermal power units, the real-time database can be used to obtain the historical operating data of the unit in the last 12 months and the operating data in the previous 24 hours, which are stored and processed at sampling intervals of 1 minute and 1 second respectively. These data include key parameters such as main steam pressure, temperature, and generator power, covering the full operating range of the unit, and can reflect the operating conditions under different loads, different coal types, and different environmental conditions. Using a real-time database, the automatic collection and fusion of historical operating data and real-time operating data of the unit can be realized, providing a solid data foundation for operating condition analysis and optimization, and ensuring the comprehensiveness and timeliness of the data. Therefore, the real-time database plays a key role in this application scenario, providing a unified data access interface, shielding the heterogeneity of the underlying system, and improving the efficiency of data collection and processing.

[0073] The data interface API is used to realize the automatic collection of historical operation data and real-time operation data of the unit. The API accesses the real-time database of each control system of the power plant in a unified way, shielding the heterogeneity of the underlying system and providing convenience for data fusion. For historical data, a time span of 12 months is selected to cover the entire operating range of the unit, including operating data under different loads, different coal types, and different environmental conditions, thereby improving the comprehensiveness of the operating condition division. For real-time data, a 24-hour time window is selected, which can obtain sufficient data samples on the one hand, and also take into account the calculation efficiency on the other hand, avoiding the timeliness affected by excessive data volume. The sampling frequency of the two types of data is also differentiated according to actual needs: historical data uses minute-level sampling to meet the accuracy requirements of statistical analysis; real-time data uses second-level sampling to achieve dynamic tracking of unit operating conditions. In addition, the selection of data items covers key parameters reflecting the unit operating conditions, including thermal parameters, electrical parameters, fuel parameters, environmental protection parameters, etc. Multi-parameter fusion can improve the differentiation of operating condition characteristics. Step S1100 efficiently collects multi-source heterogeneous data of the unit's full operating conditions and sets appropriate time and space granularity, laying a solid data foundation for subsequent operating condition analysis.

[0074] Step S1200, performing abnormal data processing on the first historical operation data and the second historical operation data respectively to obtain first processed data and second processed data;

[0075] Furthermore, if Figure 2 As shown, step S1200 includes:

[0076] Step S1210, performing missing value detection on the first historical operation data and the second historical operation data item by item, if the data is missing, it is marked as missing data, if the data is not missing, it is marked as non-missing data; for the missing data, the average value of the 5 adjacent data before and after the adjacent time is used to replace it; if there are 2 or more consecutive missing data in the adjacent data, it is marked as unavailable data;

[0077] Step S1220, performing physical threshold detection on each non-missing data item, setting the physical threshold range of the measuring point, and marking it as the first abnormal data if it exceeds the physical threshold range;

[0078] Step S1230, performing statistical threshold detection on the data that does not exceed the physical threshold range in step S1220 item by item, taking 3 times the standard deviation as the statistical threshold, and marking as second abnormal data if the deviation from the data average value exceeds 3 times the standard deviation;

[0079] Step S1240, performing noise detection on the data that are not abnormal in step S1230 one by one, calculating the peak-to-peak value of the data, and if the peak-to-peak value exceeds the set peak-to-peak threshold, and the short-term variance of the data is higher than 3 times the long-term variance, it is marked as the third abnormal data;

[0080] Step S1250 , removing the unusable data, the first abnormal data, the second abnormal data and the third abnormal data detected in steps S1210 to S1240 to obtain processed first processed data and second processed data.

[0081] Specifically, step S1200 uses a multiple anomaly detection method to achieve layer-by-layer identification and elimination of missing values, over-limit values, outliers, and noise values. First, missing value detection is performed to ensure the integrity of the data. The sporadic missing values ​​are replaced by the mean of the data at adjacent moments, which can repair the data to a certain extent. The continuous large amount of missing data is directly marked as unavailable to prevent the introduction of errors.

[0082] For example, suppose there is a set of time series data where some data points are missing:

[0083]

[0084] In this example, the data points of t3, t4, and t5 are missing, and they are three consecutive missing points; the data point of t8 is missing, but its adjacent data points (t7 and t9) exist, so it can be replaced by the average of the five data before and after the adjacent moments; for t3, t4, and t5, since their adjacent data points (t2 and t6) cannot provide enough valid data to calculate the average (because the adjacent t3, t4, and t5 themselves are continuously missing), these data points are marked as unavailable data.

[0085] Then, physical threshold detection is performed. The hard upper and lower limits of the parameters are set according to the physical characteristics of the unit. Data that exceeds the physical threshold range is obviously unreasonable and needs to be eliminated. The setting of the physical threshold range needs to be based on the design parameters of the unit, such as the rated steam parameters of the boiler and the rated power of the turbine. Then, the 3σ criterion is used for statistical threshold detection to identify outliers that deviate too much from the mean value. The 3σ criterion assumes that the data follows a normal distribution, and the probability of data falling outside 3 times the standard deviation is very small, which can be regarded as abnormal. Finally, noise detection is performed to judge the violent volatility of the data by the peak-to-peak value and short-term variance. Although the amplitude of the noise data may not be large, it changes very quickly and needs to be identified from the perspective of signal stability. Through step-by-step screening, clean processed data is finally obtained, where the first processed data represents the historical data of each operating condition of the unit, and the second processed data represents the real-time data of the current operating condition. Anomaly detection avoids the impact of dirty data on modeling analysis and is a key link in data preprocessing.

[0086] Step S1200 adopts a multi-level, multi-indicator abnormal data detection method, which can comprehensively identify various abnormal situations in the operating condition data. Missing value detection repairs the integrity of the data and reduces information loss; the physical threshold eliminates erroneous data that obviously violates the characteristics of the unit; the statistical threshold identifies accidental outliers that deviate from the normal level; and noise detection removes high-frequency interference and equipment failure data. By comprehensively utilizing mechanism knowledge and statistical means, a high-quality operating condition data set is finally obtained, thus providing reliable data support for operating condition optimization. At the same time, the anomaly detection method is applied to the first and second types of data, which not only ensures the data quality of offline operating condition analysis, but also takes into account the real-time requirements of online optimization. After data cleaning in step S1200, the success rate and convergence speed of operating condition optimization will be significantly improved.

[0087] Step S2000, performing steady-state screening on the first processed data to obtain first steady-state data; using the fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions to obtain a first operating condition category, screening out a first optimal operating condition category from the first operating condition category, and extracting the corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table;

[0088] Furthermore, step S2000 includes:

[0089] Step S2100, performing steady-state screening on the first processed data to obtain first steady-state data;

[0090] Furthermore, step S2100 includes:

[0091] Step S2110, extracting time series data of steady-state parameters from the first processed data, dividing the time series data of each steady-state parameter into sliding time windows of length L, and sliding the windows continuously with a step length of 1 sampling point, where L is 2 to 60 minutes;

[0092] Step S2120, calculating the steady-state index SI for the data in each time window of each steady-state parameter;

[0093] Step S2130, calculate SI for each of the multiple steady-state parameters in each time window, and take the maximum SI of each SI max As a comprehensive homeostatic index for this time window;

[0094] Step S2140: average the timestamps of all sampling points in the sliding time window as the comprehensive steady-state index SI max At the corresponding time t, the time series steady-state exponential curve SI is obtained. max (t);

[0095] Step S2150, setting the first steady-state threshold SI th , find out SI max All the values ​​in (t) less than SI th Time interval T s , and T s If the interval is greater than the interval threshold T, the interval is determined to be a steady-state interval; and the first processed data in the steady-state interval is extracted as the first steady-state data.

[0096] Specifically, step S2100 adopts the method of sliding time window and comprehensive steady-state index to realize the automatic identification of multi-parameter steady-state conditions of the unit. Traditional steady-state judgment often adopts a single parameter super-threshold method, which is difficult to characterize complex conditions. This method integrates multiple key parameters reflecting the state of the unit to comprehensively examine the stability of the unit. First, the time period reflecting the stable operation state of the unit is identified from the massive first processed data, and the time series of each process parameter in these steady-state time periods are extracted to form the time series data of each steady-state parameter; the steady-state parameters include main steam pressure, main steam temperature, reheat steam temperature, generator power, furnace negative pressure, induced draft fan inlet static pressure, economizer outlet flue gas temperature, exhaust gas temperature, flame brightness, etc.

[0097] A time window is constructed for each steady-state parameter. The selection of the window length needs to balance the timeliness and reliability of the steady-state criterion. The value of L is determined according to the time scale of the unit operating condition change, and can be selected as 2 to 60 minutes. If it is too short, it is susceptible to transient interference, and if it is too long, it may miss the judgment. Then, the dimensionless steady-state index SI is constructed using the mean and standard deviation of the data in the window. The smaller the SI value, the more stable the parameter is within the window. The definition of SI refers to the capability index in statistical process control, but the standard deviation is used instead of the upper and lower limits of the specification to make it suitable for steady-state judgment. The comprehensive steady-state index SI is extracted by the maximum value of the multi-parameter SI. max , it can be judged whether all parameters meet the stable state at the same time, avoiding the limitation of single parameter judgment. SI is obtained by sliding window technology max The continuous curve of the unit is then segmented by the threshold to obtain the interval where the steady-state duration exceeds the given threshold. The historical data in the interval is the steady-state operating condition data, which can be used for subsequent operating condition clustering. Step S2100 makes full use of the dynamic characteristics and correlation of the operating condition data, and describes the steady-state characteristics of the unit through quantitative indicators. It is more objective and comprehensive than the traditional method, and provides a high-quality sample set for online operating condition optimization.

[0098] The method for calculating the steady-state index SI comprises:

[0099]

[0100] in:

[0101] x i : The value of the i-th sample data point, which is extracted from the time series data of the steady-state parameters;

[0102] μ: The average value of all sample data points in the time window, which represents the average level of the data in the time window and reflects the central trend of the data;

[0103] σ: The standard deviation of all sample data points in the time window, which indicates the degree of dispersion or volatility of the data and reflects the stability of the data;

[0104] w i : The weight of the i-th sample data point. The weight can be assigned according to the importance of the sample data point to increase the flexibility of the formula. It can usually be set to 1. If some data points are more important than other data points, they can be assigned higher weights.

[0105] κ: adjustment coefficient, used to adjust the sensitivity of the steady-state index to volatility to make the formula more flexible in different applications;

[0106] ν: The rate of change of data points within the time window, calculated as It reflects the dynamics and volatility of data and helps identify situations where there are short-term and dramatic changes;

[0107] N: The number of sample data points in the time window.

[0108] When the data point x i When a drastic change occurs, |x i -μ| will also increase accordingly, causing SI to increase, indicating that the steady state is reduced; the change in the average value μ will affect |x i -μ|, thus affecting the value of SI. If μ increases and x i If the standard deviation σ is unchanged, the overall SI may increase; an increase in the standard deviation σ means an increase in the volatility of the data, resulting in a decrease in the SI, reflecting a decrease in the stability of the data; increasing the weights w of certain data points i It will increase the influence of these points on SI, making the formula more flexible; increasing the adjustment coefficient κ will increase the sensitivity of the formula to the rate of change ν, making the steady-state index more able to reflect the drastic changes in the data.

[0109] Weight w i The numerical range and selection of the two parameters and the adjustment coefficient κ have an important impact on the calculation of the steady-state index. For different application scenarios and data characteristics, the preferred numerical range is as follows:

[0110] Weight w i The preferred value range is 0≤w i ≤2;

[0111] When i =0, it means ignoring the data point;

[0112] When i =1, it means that the importance of the data point is the standard value;

[0113] When i When >1, it means increasing the importance of the data point.

[0114] The importance of data points is usually obtained based on historical data analysis. For key data points (for example, key parameters reflecting unit operating conditions), their weights can be appropriately increased. Excessively high weights may cause certain data points to have an excessive impact on the steady-state index, so they need to be set with caution.

[0115] The preferred value range of the adjustment coefficient κ is 0≤κ≤1;

[0116] When κ = 0, it means that the effect of the rate of change on the steady-state index is ignored;

[0117] When κ=1, it means maximizing the impact of the rate of change on the steady-state index.

[0118] The adjustment coefficient is determined based on the actual needs of the unit operation and historical data analysis. A larger κ indicates greater sensitivity to data fluctuations and is suitable for scenarios with high steady-state requirements; a smaller κ is suitable for scenarios with high tolerance to fluctuations.

[0119] This formula comprehensively considers the discrete degree (standard deviation) of the data, so that the steady-state index can fully reflect the stability of the data; by adjusting the coefficient κ and the weight w i , the sensitivity of the formula and the importance of data points can be adjusted according to actual needs; by calculating the mean and standard deviation of the data, the formula can effectively smooth out accidental fluctuations in the data and provide a more stable steady-state index.

[0120] Exemplarily, it is assumed that there are the following first processed data, which include three steady-state parameters: main steam pressure, main steam temperature, and generator power; the data sampling interval is 1 minute, and the data length is 10 time points.

[0121] The original data is:

[0122]

[0123] Set the window length L = 3 minutes, slide in steps of 1 sampling point, and the window is as follows:

[0124] Window 1: T1-T3

[0125] Window 2: T2-T4

[0126] Window 3: T3-T5

[0127] Window 4: T4-T6

[0128] Window 5: T5-T7

[0129] Window 6: T6-T8

[0130] Window 7: T7-T9

[0131] Window 8: T8-T10

[0132] according to Calculate the SI of each steady-state parameter in each time window, w i 1, κ 0.5;

[0133] Window 1 (T1-T3):

[0134] Main steam pressure data points: [9.5, 10.0, 9.8];

[0135] Mean value (μ): 9.7667;

[0136] Standard deviation (σ): 0.2517;

[0137] Change rate (ν): 0.35;

[0138] Steady-state index of main steam pressure:

[0139]

[0140] Main steam temperature data points: [540,542,541];

[0141] Mean value (μ): 541;

[0142] Standard deviation (σ): 1;

[0143] Change rate (ν): 1.5;

[0144] Steady-state index of main steam temperature:

[0145]

[0146] Data points: [100,102,101];

[0147] Mean value (μ): 101;

[0148] Standard deviation (σ): 1;

[0149] Change rate (ν): 1.5;

[0150] Steady-state index of generator power:

[0151]

[0152] …(compute SI for each window);

[0153] Calculate SI for multiple steady-state parameters in each time window and take the maximum SI of each SI max As a comprehensive homeostatic index for this time window;

[0154] Window 1 (T1-T3): SI max =max(1.807,1.143,1.143)=1.807;

[0155] …(calculated similarly for each window);

[0156] The integrated steady-state index SI of each window max Draw a time series steady-state exponential curve according to the time point to obtain SI max (t).

[0157] Set the first steady-state threshold SI th , for example, set it to 2.0, find SI max All the values ​​in (t) less than SI thIf the time interval is greater than the interval threshold T, the interval is determined to be a steady-state interval. Finally, the first processed data is extracted from the steady-state interval to obtain the first steady-state data.

[0158] Step S2200, using a fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions to obtain a first operating condition category;

[0159] Furthermore, if Figure 3 As shown, step S2200 includes:

[0160] Step S2210, performing data standardization processing on the first steady-state data to obtain standardized steady-state data;

[0161] Step S2220, initializing the parameters of the fuzzy C-means clustering algorithm, including the number of clusters C, the membership matrix U, the cluster center V, the maximum number of iterations MAX ITER , objective function termination threshold ε;

[0162] Step S2230, iteratively optimizing the membership matrix U and the cluster center V based on the standardized steady-state data;

[0163] Furthermore, if Figure 4 As shown, step S2230 includes:

[0164] Step S2231: based on the current membership matrix U, update the cluster centers V of each category k , k is the index of the number of cluster categories, V k represents the cluster center of the kth class;

[0165] Step S2232: based on the current cluster center V, update the membership degree U of the data point to each category kj , U kj Indicates the membership of the jth sample to the kth class;

[0166] Step S2233, calculate the objective function J, and determine whether J is less than ε or the number of iterations reaches MAX ITER If yes, generate the final membership matrix U and execute step S2240; if no, return to step S2231 and continue iterating.

[0167] Step S2240: According to the final membership matrix U, each data point is divided into the category with the largest membership, forming C operating condition clusters, namely the first operating condition category.

[0168] Specifically, step S2200 uses the fuzzy C-means (FCM) clustering algorithm to divide the steady-state operating condition data. The FCM algorithm is an unsupervised learning method that adaptively discovers the cluster structure of data by optimizing the objective function. Unlike traditional hard-partition clustering methods (such as K-means), FCM is based on fuzzy set theory and allows data points to belong to multiple categories. The membership reflects the degree of similarity between the data points and each cluster center. This soft partitioning method can better describe the transition characteristics between complex operating conditions, making the operating condition division more reasonable and natural.

[0169] Before applying the FCM algorithm, the raw data needs to be standardized. Unit parameters have different physical dimensions and numerical ranges. Directly using the raw data may cause some parameters to dominate the distance calculation, affecting the clustering effect. Through data standardization, each parameter can be mapped to the same numerical interval (such as [0,1]), eliminating dimensional differences and making different parameters contribute equally to clustering. Commonly used standardization methods include maximum-minimum standardization, zero-mean standardization, decimal calibration standardization, etc.

[0170] The FCM algorithm requires some preset parameters, including the expected number of operating condition clusters C, the membership matrix U, the cluster center V, and the iterative control parameter MAX. ITER and ε, etc. Among them, C can be determined based on the actual operating experience of the unit and historical data analysis, and 310 categories are usually selected; U and V can be initialized randomly or with other clustering results (such as K-means); parameter m is the fuzziness index, which controls the degree of "softening" in the clustering process; MAX ITER and ε are used to control the convergence of the algorithm and prevent too many invalid iterations. The selection of these parameters has a certain impact on the clustering results and requires repeated trials and tuning.

[0171] The core of FCM is to alternately optimize the membership matrix U and the cluster center V. Each iteration is divided into two steps: first, fix the membership matrix U and update the cluster center based on the current category division; then, fix the cluster center V and update the membership matrix based on the current center position. These two steps are performed alternately, so that the cluster center continues to move to the data dense area. At the same time, the membership of the data point is updated according to the distance to the cluster center, and finally converges to the local optimal solution. The objective function J measures the sum of the weighted distances from the data point to the center of its category. The smaller J is, the better the clustering effect. When the change in the objective function is less than the threshold ε, or the number of iterations exceeds MAX ITER When , the algorithm terminates.

[0172] After the algorithm converges, for each data point, the category with the largest membership is selected as the operating condition to which it belongs. In this way, the original steady-state operating condition data set is divided into C non-overlapping subsets, each of which represents a typical unit operating condition. This division makes full use of the inherent structural information of the data and can adaptively extract the common features of different operating conditions. Compared with manual division, FCM clustering is more objective and comprehensive, and can discover potential operating condition patterns in the data.

[0173] The mathematical formula of the objective function J includes:

[0174]

[0175] in:

[0176] n: number of samples for standardized steady-state data;

[0177] C: number of clustered categories;

[0178] U kj : The membership of the jth sample to the kth class is initialized to a random value and obtained through iterative optimization;

[0179] m: fuzziness index of membership matrix U, usually in the range of [1.5, 2.5];

[0180] Y j : jth sample data;

[0181] V k : The cluster center of the kth class is initialized to a random value and obtained through iterative optimization;

[0182] ∥·∥: L2 norm (Euclidean distance) of the vector, used to measure the distance between the sample and the cluster center;

[0183] λ: distance weighting coefficient, which controls the influence of distance on membership. The value range is usually [0.1, 10] and is adjusted according to the clustering effect;

[0184] δ: The scale parameter of the Gaussian kernel function, which controls the rate of distance attenuation. It is usually in the range of [0.1, 1] and is selected according to the characteristics of the data distribution;

[0185] η: Category balance penalty coefficient, which controls the degree of penalty for category imbalance. The value range is usually [0.01, 0.1] and is adjusted according to the category distribution;

[0186] n k : The number of samples in the kth class, obtained by summing the membership matrix U;

[0187] p: The power exponent of the class balance penalty, which controls the nonlinearity of the penalty. The value range is usually [0.5, 2], which is selected according to the severity of class imbalance.

[0188] The objective function consists of two parts: the first is the objective function of traditional fuzzy C-means clustering, which is used to minimize the sum of the weighted distances between the sample and the cluster center with the highest membership; the second is the newly introduced category balance penalty term, which is used to penalize the imbalance of categories and promote the balance of the number of samples in each category.

[0189] In the first item, the Gaussian kernel function is introduced to weight the distance, so that the samples with closer distance have a greater impact on the membership, and the influence of the samples with farther distance is weakened, which improves the robustness of clustering and the tolerance to noise. As the distance between the sample and the cluster center increases, the value of the Gaussian kernel function will decay rapidly, making the sample far away from the cluster center contribute less to the objective function.

[0190] In the second term, the class balance penalty term increases as the class imbalance increases. When the number of samples in a certain class is much smaller than that in other classes, the value of the penalty term will increase, forcing the optimization process to adjust in the direction of increasing the number of samples in that class to achieve class balance. The power exponent p in the penalty term controls the nonlinearity of the penalty. The larger p is, the greater the penalty for class imbalance.

[0191] By minimizing this objective function, we can achieve sample clustering while taking into account the balance of categories, and obtain more reasonable and meaningful clustering results. During the iterative optimization process, the membership matrix U and the cluster center V will be continuously updated until the objective function value converges or the maximum number of iterations is reached. In general, this objective function introduces distance weighting and category balance penalty on the basis of traditional fuzzy C-means clustering, which improves the robustness and category balance of clustering, and helps to obtain more accurate and reliable results in the clustering analysis of working condition data sets.

[0192] Exemplarily, an example is given to illustrate the working condition division process of the fuzzy C-means clustering algorithm.

[0193] Assume that there is the following first steady-state data, which includes 10 data samples, each of which has 3 characteristic variables: main steam pressure, main steam temperature, and generator power.

[0194] Sample No. Main steam pressure(MPa) Main steam temperature℃ Generator power (MW) 1 17.1 535 300 2 16.8 530 290 3 17.5 540 305 4 17.3 538 308 5 16.5 525 285 6 17.0 532 295 7 16.9 529 293 8 17.4 542 310 9 16.7 526 288 10 17.2 537 303

[0195] The data after standardization of the original data are as follows:

[0196]

[0197]

[0198] Initialize clustering parameters, set the number of clusters C = 3, and the maximum number of iterations MAX ITER =100, objective function threshold ε=0.01; membership matrix U and cluster center V are initialized randomly.

[0199] The initial membership matrix U is as follows:

[0200] Sample No. Category 1 Category 2 Category 3 1 0.3 0.5 0.2 2 0.7 0.2 0.1 3 0.2 0.4 0.4 4 0.1 0.3 0.6 5 0.8 0.1 0.1 6 0.5 0.2 0.3 7 0.6 0.3 0.1 8 0.2 0.2 0.6 9 0.7 0.2 0.1 10 0.4 0.1 0.5

[0201] The initial cluster center V is:

[0202]

[0203]

[0204] Continuously iterate the cluster center V and membership degree U until the termination condition is reached: the objective function changes less than ε or the number of iterations reaches MAX ITER .

[0205] Assume m = 2, λ = 1, δ 2 =0.5, η=0.1, p=1, after 20 iterations,

[0206] The objective function J drops from the initial 8.37 to 0.58, which is less than ε. At this time, the optimized membership matrix U and cluster center V are as follows:

[0207] Membership matrix U:

[0208] Sample No. Category 1 Category 2 Category 3 1 0.05 0.93 0.02 2 0.08 0.89 0.03 3 0.02 0.95 0.03 4 0.01 0.97 0.02 5 0.92 0.05 0.03 6 0.14 0.83 0.03 7 0.21 0.76 0.03 8 0.01 0.02 0.97 9 0.85 0.09 0.06 10 0.02 0.95 0.03

[0209] Cluster center V:

[0210] category Main steam pressure Main steam temperature Generator power Category 1 -1.00 -1.80 -1.28 Category 2 0.39 0.34 0.36 Category 3 1.50 2.00 2.00

[0211] Working conditions division:

[0212] According to the optimized membership matrix U, the category with the largest membership is taken as the working condition of the sample. It can be seen that samples 1-4, 6-7, and 10 are divided into category 2; samples 5 and 9 are divided into category 1; and sample 8 is divided into category 3. These three categories are the working condition division results of this group of steady-state working condition data, that is, the first working condition category. Category 1 represents low-pressure, low-temperature, and low-load working conditions, category 2 represents medium-pressure, medium-temperature, and medium-load working conditions, and category 3 represents high-pressure, high-temperature, and high-load working conditions. Fuzzy C-means clustering objectively characterizes the characteristics of different steady-state working conditions by optimizing the membership of samples and the clustering center of categories, which can be used as an important basis for subsequent optimization control. The objective function J comprehensively considers factors such as the distance between the sample and the clustering center, the fuzzy degree of membership, and the balance of categories, making the clustering results more reasonable. It should be noted that the setting of clustering parameters needs to be repeatedly tested and adjusted according to the specific data distribution and physical background to obtain the best working condition division effect.

[0213] This example shows how to use the fuzzy C-means clustering algorithm to divide multiple steady-state operating condition samples into several categories, each of which represents a typical operating condition. Operating condition division is the basis for optimizing the whole operating condition of the unit. Only by accurately characterizing the characteristics of different operating conditions can we design an optimized control strategy on this basis. Step S2200 provides an effective operating condition division method, which lays the foundation for subsequent operating condition optimization control.

[0214] Step S2300, selecting a first optimal operating condition category from the first operating condition category, and extracting operating parameters corresponding to the first optimal operating condition category to form a first optimal operating condition table;

[0215] Furthermore, if Figure 5 As shown, step S2300 includes:

[0216] Step S2310, for each operating condition category, calculating various performance indicators;

[0217] Step S2320, performing weighted summation on each performance index to obtain a comprehensive performance evaluation value for each operating condition category;

[0218] Step S2330, selecting the operating condition category with the best comprehensive performance evaluation value as the first best operating condition category;

[0219] Step S2340, extracting typical operating parameters under the first optimal operating condition category to form a first optimal operating condition table.

[0220] Specifically, step S2300 selects the optimal operating condition from multiple operating condition categories, and a scientific evaluation index system needs to be established. Since different performance indicators may have a mutually restrictive relationship, a single indicator cannot fully reflect the pros and cons of the operating condition, and a comprehensive evaluation is required. The various performance indicators of each operating condition category include load level, heat consumption, coal consumption, NOx emissions, pressure, temperature, etc.; by calculating the statistical averages of load level, heat consumption, coal consumption, NOx emissions, pressure, temperature, etc. under each operating condition category, the performance differences of different operating conditions can be preliminarily compared. On this basis, combined with the actual management objectives and needs of the unit, different weights are assigned to each indicator to construct a weighted comprehensive evaluation model.

[0221] The selection of weights needs to take into account various factors. From an economic perspective, the weights of heat consumption and coal consumption should be higher to reduce the cost of power generation; from a safety perspective, the weights of pressure and temperature should be appropriate to ensure that the unit does not exceed the allowable value; from an environmental protection perspective, the weight of NOx emissions should be prominent to meet the environmental protection requirements of ultra-low emissions. In the absence of a clear optimization target, the same weight coefficient can be given to each indicator. The rationality of the weight needs to be verified through historical data analysis and expert judgment to balance the needs of unit efficiency, safety and environmental protection.

[0222] After weighted summation, there may be significant differences in the comprehensive performance evaluation values ​​of different operating condition categories. The operating condition with the highest evaluation value usually takes into account the various performance aspects of the unit, represents the optimal operating state, and can be used as the target of operating condition optimization. Generally speaking, the optimal operating condition corresponds to the high-efficiency, low-cost, stable and safe, clean and environmentally friendly operating state of the unit. A series of typical operating parameters are extracted from the optimal operating conditions, including load, main steam pressure / temperature, reheat steam temperature, furnace negative pressure, oxygen content, primary and secondary air volume, coal consumption, pulverizer combination, feed water temperature, etc., to form an optimal operating condition table, which provides an operational reference range for real-time operating condition optimization.

[0223] For example, it is assumed that FCM clustering divides the unit operating conditions into three categories: high load condition (A1), low load condition (A2), and peak load condition (A3). The performance indicators of each condition are as follows:

[0224] Working condition A1: average load 95% rated, power supply coal consumption 280g / kWh, NOx emission 30mg / m3;

[0225] Working condition A2: average load 60% rated, power supply coal consumption 320g / kWh, NOx emission 80mg / m3;

[0226] Working condition A3: average load 80% rated, power supply coal consumption 300g / kWh, NOx emission 100mg / m3.

[0227] If the coal consumption and NOx emission indicators are considered, with weights of 0.6 and 0.4 respectively, the comprehensive performance evaluation values ​​of the three working conditions are:

[0228] Working condition A1: 280×0.6+30×0.4=180;

[0229] Working condition A2: 320×0.6+80×0.4=224;

[0230] Working condition A3: 300×0.6+100×0.4=220.

[0231] It can be seen that the comprehensive evaluation value of operating condition A1 is the lowest, which is the optimal operating condition. The key parameter values ​​such as load, main steam temperature, and coal mill combination are extracted from it to form the optimal operating condition operating parameter table:

[0232] Load: 950MW±50MW;

[0233] Main steam pressure: 24MPa±1MPa;

[0234] Main steam temperature: 560℃±10℃;

[0235] …

[0236] The optimal operating condition table specifies the typical parameter setting range under the optimal operating state of the unit, which can guide the operator to optimize the operating conditions. When the real-time data deviates from the setting range of the optimal operating condition table, the relevant parameters can be adjusted in time to make the unit operating condition approach the optimal direction, thus achieving the purpose of operating condition optimization control.

[0237] Step S3000, performing online working condition judgment and optimization on the second processed data;

[0238] Furthermore, step S3000 includes:

[0239] Step S3100, performing steady-state judgment on the second processed data, if it is steady-state, executing step S3200; if it is non-steady-state, maintaining the first optimal operating condition table unchanged, and executing step S3300;

[0240] Specifically, step S3100 uses the same time series steady-state determination method as step S2000 to achieve stability determination of the real-time operating condition of the unit. The timeliness of online steady-state determination is higher, and it is necessary to reflect the dynamic changes of the unit operating condition in a timely manner to provide a basis for optimized control. Therefore, step S3100 selects the latest 24-hour second processed data for sliding window calculation to obtain the second steady-state index curve SI reflecting the current unit status max1 (t). By setting the second steady-state threshold SI th1The duration threshold T1 can be used to determine whether the unit operating condition is in a stable operating state at the current moment. If the current operating condition is steady, it means that the unit is running smoothly and the operating condition optimization control can be performed; if it is unsteady, it means that the unit is in a state of operating condition conversion or fluctuation, and it is not suitable for optimization. At this time, the original first optimal operating condition table is maintained unchanged, and optimization is performed after the operating condition stabilizes. This real-time steady-state judgment strategy can avoid drastic fluctuations in the unit operating condition caused by optimization control and ensure safe and stable operation of the unit.

[0241] Step S3200, dividing the second processed data into operating conditions to obtain a second operating condition category; performing a benchmark analysis on the second operating condition category and the first optimal operating condition table; if the second operating condition category is better than the first optimal operating condition category, then updating the first optimal operating condition table with the operating parameters corresponding to the second operating condition category to obtain the second optimal operating condition table; if the second operating condition category is not better than the first optimal operating condition category, then maintaining the first optimal operating condition table unchanged, and then executing step S3300;

[0242] Specifically, step S3200 updates the optimal operating condition knowledge base online in real time, reflecting the operating condition self-learning function. After the current unit operating condition is determined to be steady state, the characteristics of its process parameters in the parameter space can be extracted, and the same fuzzy C-means clustering algorithm as step S2200 can be used for classification to obtain the operating condition category to which the steady state operating condition belongs. By comparing each operating condition category in the first optimal operating condition table one by one, the optimal operating condition most similar to the current unit operating condition is found. If the current operating condition is better than the existing optimal operating condition, the optimal operating condition table can be updated with its corresponding operating parameters to obtain the second optimal operating condition table. The "better than" criterion here can be determined according to the unit performance indicators, such as heat consumption, coal consumption, power generation efficiency, pollutant emissions, etc. If there are multiple criteria, they can also be weighted to obtain a comprehensive performance index as a measure of the quality of the operating condition. Through this real-time benchmarking learning, the optimal operating condition of the unit can be continuously discovered and solidified, and a mapping relationship between configuration parameters and performance indicators can be formed in the knowledge base, providing decision support for real-time optimization control. This process, on the one hand, makes full use of the unit's historical optimal operating condition knowledge, and on the other hand, can dynamically adapt to the changing trend of the operating condition, continuously iterate the operating condition optimization model in inheritance and innovation, and improve the adaptability of the unit. If the current operating condition is equivalent to or worse than the existing optimal operating condition, there is no need to update the optimal operating condition table, and the first optimal operating condition table is maintained unchanged as the target operating condition set for optimization control.

[0243] Step S3300, after waiting for 24 hours, the latest 24-hour data is stored in the real-time database as the new second historical operation data, and the process returns to step S3100 to perform the next round of online working condition judgment and optimization.

[0244] Specifically, step S3300 realizes the rolling update and closed-loop control of online working condition judgment. After each decision cycle, a new cycle of unit operation data will be added to the real-time database. The latest 24-hour data is used as the new second historical operation data to replace the data of the previous cycle to participate in the next round of steady-state judgment and working condition benchmarking. At the same time, the second historical operation data of the previous cycle will enter the historical database and become a sample for the long-term working condition analysis of the unit. By continuously iterating steps S3100 to S3300, dynamic closed-loop control of working condition optimization can be achieved, that is, according to the online working condition judgment results, the optimal working condition knowledge base is automatically updated, the real-time optimization control strategy is generated, the unit working condition is adjusted, and the optimization effect is evaluated before entering the next optimization cycle. This "identification-decision-execution-evaluation" cycle structure can continuously tap the optimization potential of unit operation, conduct online verification and correction of each working condition optimization strategy, continuously accumulate new working condition knowledge, build a working condition big data analysis platform, and provide experience support for the full life cycle management of the unit.

[0245] For example, assuming that a supercritical thermal power unit has been running stably for 60 minutes, online steady-state judgment and operating condition optimization are triggered. The unit operation data for the 60 minutes are as follows:

[0246]

[0247] The 60-minute second steady-state index SI max1 Less than the set second steady-state threshold SI th1 , and determine that the current unit is in steady-state condition. Cluster analysis is performed on the parameters of the steady-state condition, and it is found that it belongs to condition B1. The first optimal condition table is queried as follows:

[0248]

[0249]

[0250] Comparing the current unit operating condition with the first optimal operating condition table, it is found that it is closest to operating condition B1, and the unit heat consumption is lower than operating condition B1. Therefore, the current unit operating condition is used to update operating condition B1 to obtain the second optimal operating condition table:

[0251]

[0252] When the next 24-hour cycle arrives, the latest 24-hour data will replace the above 60-minute data, and a new round of steady-state judgment and working condition optimization will begin, and the working condition knowledge base will be continuously updated.

[0253] In summary, the method of the present invention realizes the combination of online optimization and offline analysis of unit operating conditions through the interaction of real-time database and historical database. The operating condition knowledge can be learned online, accumulated offline, applied online, and updated offline, forming a data-driven closed-loop optimization control system. Compared with traditional offline optimization and fixed-cycle online optimization, this method can significantly improve the timeliness of optimization, shorten the knowledge iteration cycle, and enhance the dynamic adaptive ability of unit operating conditions. The present invention provides a new idea for the intelligent optimization operation of thermal power units, which is of great significance for energy saving, emission reduction, and efficiency improvement.

[0254] Example 2

[0255] This embodiment provides a system for optimizing the operating conditions of a thermal power unit based on the first embodiment. Figure 6 As shown, including:

[0256] Data acquisition module: used to obtain the first historical operation data and the second historical operation data of the unit;

[0257] Data processing module: used for performing exception processing on the acquired first historical operation data and second historical operation data to obtain first processed data and second processed data;

[0258] Steady-state screening module: used for performing steady-state screening on the first processed data to obtain first steady-state data;

[0259] Working condition classification module: used for performing working condition classification on the first steady-state data by using fuzzy C-means clustering algorithm to obtain a first working condition category;

[0260] Optimal operating condition screening module: used for screening out a first optimal operating condition category from the first operating condition category, and extracting corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table;

[0261] Online operating condition optimization module: used to perform online operating condition judgment and optimization on the second processed data, and update the first optimal operating condition table.

[0262] Example 3

[0263] This embodiment discloses an electronic device 500, such as Figure 7 As shown, the electronic device 500 may include one or more processors and one or more memories. The memories store computer-readable codes, which, when executed by one or more processors, may execute the above-mentioned method for optimizing the operating conditions of a thermal power unit.

[0264] The method or system according to the embodiment of the present application can also be used by Figure 7 The electronic device architecture shown in FIG. Figure 7As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store a method for optimizing the operating conditions of a thermal power unit provided in the present application. A method for optimizing the operating conditions of a thermal power unit may, for example, include: obtaining first historical operating data and second historical operating data of the unit, performing abnormal processing on the first historical operating data and the second historical operating data obtained to obtain first processed data and second processed data; performing steady-state screening on the first processed data to obtain first steady-state data; using a fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions to obtain a first operating condition category, screening out a first optimal operating condition category from the first operating condition category, and extracting the corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table; performing online operating condition judgment and optimization on the second processed data, and updating the first optimal operating condition table. Furthermore, the electronic device 500 may also include a user interface 508. Figure 7 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 7 One or more components of an electronic device are shown.

[0265] Example 4

[0266] This embodiment discloses a computer-readable storage medium 600, such as Figure 8 As shown, a computer-readable storage medium 600 stores computer-readable instructions. When the computer-readable instructions are executed by the processor, a method for optimizing the operating conditions of a thermal power unit according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0267] In addition, according to the implementation mode of the present application, the process described in the above reference flow chart can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, the non-temporary machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: obtaining the first historical operation data and the second historical operation data of the unit, performing abnormal processing on the acquired first historical operation data and the second historical operation data, and obtaining the first processed data and the second processed data; performing steady-state screening on the first processed data to obtain the first steady-state data; using the fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions, obtaining the first operating condition category, screening the first optimal operating condition category from the first operating condition category, and extracting the corresponding operating parameters under the first optimal operating condition category to form the first optimal operating condition table; performing online operating condition judgment and optimization on the second processed data, and updating the first optimal operating condition table. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0268] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, 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 method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0269] In addition, the parts of the above-mentioned 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.

[0270] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for optimizing the operating conditions of a thermal power unit, characterized in that: The method comprises: Acquire first historical operation data and second historical operation data of the unit, perform abnormal processing on the acquired first historical operation data and second historical operation data, and obtain first processed data and second processed data; Perform steady-state screening on the first processed data to obtain first steady-state data; use a fuzzy C-means clustering algorithm to divide the first steady-state data into operating conditions to obtain a first operating condition category, screen out a first optimal operating condition category from the first operating condition category, and extract corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table; Perform online operating condition judgment and optimization on the second processed data, and update the first optimal operating condition table.

2. A method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The first historical operation data is the historical operation data of the unit in the last 12 months obtained from the real-time database, with a sampling interval of 1 minute; the second historical operation data is the operation data within the previous 24 hours obtained from the real-time database, with a sampling interval of 1 second.

3. The method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The performing abnormal processing on the acquired first historical operation data and second historical operation data comprises: Perform missing value detection on the first historical operation data and the second historical operation data item by item. If the data is missing, it is marked as missing data. If the data is not missing, it is marked as non-missing data. For missing data, the average value of the 5 adjacent data before and after the adjacent time is used to replace it. If there are 2 or more consecutive missing data in the adjacent data, it is marked as unavailable data. Perform physical threshold detection on each non-missing data item, set the physical threshold range, and if it exceeds the physical threshold range, mark it as the first abnormal data; Perform statistical threshold detection on each data item that does not exceed the physical threshold range, with 3 times the standard deviation as the statistical threshold. If the deviation from the data mean value exceeds 3 times the standard deviation, it is marked as the second abnormal data; Perform noise detection on each data item without abnormality detection, calculate the peak-to-peak value of the data, and if the peak-to-peak value exceeds the set peak-to-peak threshold and the short-term variance of the data is higher than 3 times the long-term variance, it is marked as the third abnormal data; Unusable data, first abnormal data, second abnormal data and third abnormal data are eliminated to obtain processed first processed data and second processed data.

4. A method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The performing steady-state screening on the first processed data to obtain the first steady-state data comprises: Extracting time series data of steady-state parameters from the first processed data, dividing the time series data of each steady-state parameter into a sliding time window of length L, and sliding the window continuously with a step length of 1 sampling point; The steady-state index SI is calculated for the data in each time window of each steady-state parameter; Calculate SI for multiple steady-state parameters in each time window and take the maximum SI of each SI max As a comprehensive homeostatic index for this time window; The average of the timestamps of all sampling points in the sliding time window is used as the comprehensive steady-state index SI max At the corresponding time t, the time series steady-state exponential curve SI is obtained. max (t); Set the first steady-state threshold SI th , find out SI max All the values ​​in (t) less than SI th Time interval T s , and T s If the interval is greater than the interval threshold T, the interval is determined to be a steady-state interval; and the first processed data in the steady-state interval is extracted as the first steady-state data.

5. The method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The first steady-state data is divided into the first working condition category by the working condition classification, which includes: Performing data standardization processing on the first steady-state data to obtain standardized steady-state data; Initialize the parameters of the fuzzy C-means clustering algorithm, including the number of clusters C, the membership matrix U, the cluster center V, and the maximum number of iterations MAX ITER , objective function termination threshold ε; Based on the standardized steady-state data, the membership matrix U and cluster center V are iteratively optimized; According to the final membership matrix U, each data point is divided into the category with the largest membership, forming C working condition clusters, namely the first working condition category; The iterative optimization of the membership matrix U and the cluster center V includes: Based on the current membership matrix U, update the cluster centers V of each category k , k is the index of the number of cluster categories, V k represents the cluster center of the kth class; Based on the current cluster center V, update the membership U of the data point to each category kj , U kj Indicates the membership of the jth sample to the kth class; Calculate the objective function J and determine whether J is less than ε or the number of iterations reaches MAX ITER , if yes, generate the final membership matrix U; if not, continue iterating.

6. A method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The step of selecting the first optimal operating condition category from the first operating condition category includes: For each operating condition category, calculate its various performance indicators; Perform weighted summation on each performance index to obtain the comprehensive performance evaluation value of each working condition category; The operating condition category with the best comprehensive performance evaluation value is selected as the first optimal operating condition category.

7. A method for optimizing the operating conditions of a thermal power unit according to claim 1, characterized in that: The online working condition judgment and optimization for the second processed data includes: Perform steady-state judgment on the second processed data. If it is steady-state, divide the second processed data into working conditions to obtain a second working condition category; perform benchmark analysis on the second working condition category and the first optimal working condition table. If the second working condition category is better than the first optimal working condition category, update the first optimal working condition table with the operating parameters corresponding to the second working condition category; if the second working condition category is not better than the first optimal working condition category, maintain the first optimal working condition table unchanged, wait for 24 hours, store the latest 24-hour data as the new second historical operating data in the real-time database, and perform the next round of online working condition judgment and optimization; If it is a non-steady state, the first optimal operating condition table is maintained unchanged. After waiting for 24 hours, the latest 24-hour data is stored in the real-time database as the new second historical operation data for the next round of online operating condition judgment and optimization.

8. A system for optimizing the operating conditions of a thermal power unit, which is used to implement a method for optimizing the operating conditions of a thermal power unit according to any one of claims 1 to 7, characterized in that: The system comprises: Data acquisition module: used to obtain the first historical operation data and the second historical operation data of the unit; Data processing module: used for performing exception processing on the acquired first historical operation data and second historical operation data to obtain first processed data and second processed data; Steady-state screening module: used for performing steady-state screening on the first processed data to obtain first steady-state data; Working condition classification module: used for performing working condition classification on the first steady-state data by using fuzzy C-means clustering algorithm to obtain a first working condition category; Optimal operating condition screening module: used for screening out a first optimal operating condition category from the first operating condition category, and extracting corresponding operating parameters under the first optimal operating condition category to form a first optimal operating condition table; Online operating condition optimization module: used to perform online operating condition judgment and optimization on the second processed data, and update the first optimal operating condition table.

9. An electronic device comprising a memory, a central processing unit, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for optimizing the operating conditions of a thermal power unit according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the steps in the method for optimizing the operating conditions of a thermal power unit according to any one of claims 1 to 7 are implemented.

Images