Method and equipment for identifying abnormal data of steam turbine operation

Abstract

The present invention provides a method and device for identifying abnormal operation data of a steam turbine. The method collects actual operation data and actual main steam flow of the steam turbine at multiple times within a target time period, compares the collected actual main steam flow with the target main steam flow obtained according to the actual operation data, calculates the abnormality of multiple recognition algorithms to determine the comprehensive abnormality, and then marks the abnormal data in the collected actual main steam flow according to the comprehensive abnormality. This method can prevent erroneous recognition caused by a single recognition algorithm, effectively improve the accuracy of abnormal data recognition, and timely discover the operation abnormality of the steam turbine.

Description

Technical Field

[0001] The present application belongs to the technical field of data identification, and in particular, relates to a method and device for identifying abnormal operation data of a steam turbine. Background Art

[0002] Under the trend of industrial big data, the requirements for process monitoring of steam turbine operation are getting higher and higher, and the engineering data that needs to be processed is also getting larger and larger. Identifying abnormal data during steam turbine operation is a necessary condition to ensure the reliability and effectiveness of steam turbine operation.

[0003] At present, the abnormal data identification methods in engineering include: L'Aida's law judgment method, isolation forest identification method, etc. However, L'Aida's law is too rough for the identification of abnormal data, and many abnormal data points cannot meet the identification criteria, and the identification accuracy is low. When there are a large number of continuous abnormal points, the isolation forest identification method will have more normal points mistakenly identified as abnormalities, and the identification accuracy is low. Summary of the invention

[0004] In view of this, the present invention provides a method and device for identifying abnormal operation data of a steam turbine, aiming to solve the problem of low accuracy in identifying abnormal operation data of a steam turbine.

[0005] A first aspect of an embodiment of the present invention provides a method for identifying abnormal operation data of a steam turbine, comprising:

[0006] Collect actual operation data and actual main steam flow of the steam turbine at multiple times during the target period;

[0007] Inputting the actual operation data into a pre-built regression model to predict the target main steam flow rate of the steam turbine at each time in a target period under normal conditions;

[0008] Determining at least two abnormality levels of each actual main steam flow rate according to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms; wherein each identification algorithm corresponds to one abnormality level;

[0009] According to at least two abnormalities of each actual main steam flow rate, a comprehensive abnormality of each actual main steam flow rate is determined, and the actual main steam flow rate with the comprehensive abnormality greater than or equal to a first preset threshold is marked as abnormal data.

[0010] A second aspect of an embodiment of the present invention provides a device for identifying abnormal operation data of a steam turbine, comprising:

[0011] A collection module is used to collect actual operation data and actual main steam flow of the steam turbine at multiple times within the target period;

[0012] A prediction module, used for inputting the actual operation data into a pre-built regression model to predict the target main steam flow rate of the steam turbine at each moment in a target period under normal conditions;

[0013] A determination module, used to determine at least two abnormality levels of each actual main steam flow rate according to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms; wherein each identification algorithm corresponds to one abnormality level;

[0014] The marking module is used to determine the comprehensive abnormality of each actual main steam flow according to at least two abnormalities of each actual main steam flow, and mark the actual main steam flow with the comprehensive abnormality greater than or equal to a first preset threshold as abnormal data.

[0015] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for identifying abnormal operation data of a turbine as described in the first aspect above are implemented.

[0016] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for identifying abnormal operation data of a steam turbine as described in the first aspect above are implemented.

[0017] The method and device for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention collect actual operation data and actual main steam flow of the steam turbine at multiple times within a target time period, compare the collected actual main steam flow with the target main steam flow obtained according to the actual operation data, calculate the abnormality degrees of multiple recognition algorithms to determine the comprehensive abnormality degree, and then mark the abnormal data in the collected actual main steam flow according to the comprehensive abnormality degree, which can prevent erroneous recognition caused by a single recognition algorithm, effectively improve the accuracy of abnormal data recognition, and timely discover the operation abnormality of the steam turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 labor.

[0019] Figure 1 This is an application scenario diagram of the method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention;

[0020] Figure 2It is a flow chart of the implementation of the method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention;

[0021] Figure 3 is a flowchart for implementing the comprehensive abnormality degree calculation provided by an embodiment of the present invention;

[0022] Figure 4 is a box plot provided by an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram of the structure of a regression model provided by an embodiment of the present invention;

[0024] Figure 6 It is a flow chart of the implementation of the method for identifying abnormal data of steam turbine operation provided by the implementation example of the present invention;

[0025] Figure 7 is a flow chart of a method for identifying abnormal operation data of a steam turbine provided by another embodiment of the present invention;

[0026] Figure 8 It is a structural schematic diagram of a device for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention;

[0027] Fig. 9 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0029] Figure 1 : is an application scenario diagram of the method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention. The method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention can be applied to, but is not limited to, this application environment. Figure 1 As shown, the system includes: a steam turbine 11, a data acquisition device 12, and an electronic device 13.

[0030] The data acquisition device 12 is used to collect the operation data and main steam flow rate in the steam turbine 11 in real time, and send it to the electronic device 13. The electronic device is used to receive the operation data and main steam flow rate collected by each data acquisition device 12, and mark the abnormal data in the main steam flow rate according to the received data. Relevant personnel can view the marked abnormal data through the display screen set on the electronic device 13, or access the electronic device 13 through a terminal device to view the marked abnormal data, which is not limited here.

[0031] The steam turbine 11 may be an impulse steam turbine or a reaction steam turbine, which is not limited here. The data acquisition device 12 may be an electric energy meter, a temperature acquisition device, a pressure acquisition device, a flow acquisition device, etc., which is not limited here. The electronic device 13 may be a server, a terminal, etc., which is not limited here. The server may be implemented as an independent server or a server cluster composed of multiple servers. The terminal may include but is not limited to a desktop computer, a laptop computer, a tablet computer, etc.

[0032] Figure 2 is a flow chart of the method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention. Figure 1 The electronic equipment in the example is used to illustrate. Figure 2 As shown, the method includes:

[0033] S201, collect the actual operation data and actual main steam flow of the steam turbine at multiple times during the target period.

[0034] S202, inputting the actual operation data into a pre-built regression model to predict the target main steam flow rate of the steam turbine at each moment in a normal situation within a target period.

[0035] S203, determining at least two abnormality levels of each actual main steam flow rate according to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms; wherein each identification algorithm corresponds to one abnormality level.

[0036] S204, determining the comprehensive abnormality of each actual main steam flow rate according to at least two abnormality degrees of each actual main steam flow rate, and marking the actual main steam flow rate with a comprehensive abnormality degree greater than or equal to a first preset threshold as abnormal data.

[0037] In this embodiment, the target time period can be one hour, one day, etc., which is not limited here. The actual operation data may include but is not limited to at least one of the following: unit load, main steam pressure, main steam temperature, pressure after regulating stage, temperature after regulating stage, reheat steam pressure, reheat steam temperature. The target main steam flow rate represents the main steam flow rate of the steam turbine under normal conditions when operating with the collected actual operation data. At least two identification algorithms may include but are not limited to at least two of the following: Lleyda's law judgment algorithm, isolation forest recognition algorithm, overshoot judgment algorithm and box plot algorithm.

[0038] In this embodiment, by collecting the actual operating data and actual main steam flow of the steam turbine at multiple times within the target time period, the collected actual main steam flow is compared with the target main steam flow obtained according to the actual operating data, and the abnormality degrees of multiple recognition algorithms are calculated to determine the comprehensive abnormality degree. Then, the abnormal data in the collected actual main steam flow is marked according to the comprehensive abnormality degree. This can prevent erroneous recognition caused by a single recognition algorithm, effectively improve the accuracy of abnormal data recognition, and timely discover the operating abnormality of the steam turbine.

[0039] Figure 3 FIG. 1 is a flowchart of the implementation of the comprehensive abnormality calculation provided by the embodiment of the present invention. Figure 3 As shown, in some embodiments, at least two identification algorithms include: a Lleyda's law determination algorithm, an overshoot determination algorithm, and a box plot algorithm.

[0040] S203 may include:

[0041] S301, determining a first abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and a Lleyda's law judgment algorithm.

[0042] S302, determining a second abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and an overshoot determination algorithm.

[0043] S303, determining a third abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and a box plot algorithm.

[0044] Accordingly, S204 may include:

[0045] S304, adding the first abnormality degree, the second abnormality degree and the third abnormality degree or performing weighted summation to obtain a comprehensive abnormality degree of each actual main steam flow rate.

[0046] In some embodiments, S301 may include:

[0047] Calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate;

[0048] Calculate the standard deviation and each residual error of the series formed by the absolute values ​​of all differences; the residual error corresponds to the actual main steam flow rate one by one;

[0049] The ratio of the absolute value of the residual error of each actual main steam flow rate to the standard deviation is calculated and used as the value of the first abnormality degree of each actual main steam flow rate, or the value of the first abnormality degree of each actual main steam flow rate is found from a preset first abnormality degree table according to the residual error and standard deviation of each actual main steam flow rate.

[0050] In this embodiment, after obtaining the sequence formed by the absolute values ​​of the differences, the arithmetic mean of the sequence is calculated first, and then each value in the sequence is subtracted from the arithmetic mean to obtain the residual error. For example, the standard deviation of the sequence can be obtained based on the obtained residual error and the Bessel formula.

[0051] The Bessel formula is as follows:

[0052]

[0053] Among them, σ is the standard deviation of the series, v i is the ith residual error in the sequence, and n is the number of residual errors.

[0054] In this embodiment, after obtaining the residual error and standard deviation of each actual main steam flow rate, the value of the first abnormality degree of each actual main steam flow rate is determined according to the residual error and standard deviation of each actual main steam flow rate, which can be achieved by one of the following two implementation methods:

[0055] In a first implementation manner, the ratio of the absolute value of the residual error of each actual main steam flow rate to the standard deviation may be calculated and used as the value of the first abnormality degree of each actual main steam flow rate.

[0056] In a second implementation, the value of the first abnormality degree of each actual main steam flow rate is searched from a preset first abnormality degree table according to the residual error and standard deviation of each actual main steam flow rate.

[0057] In this embodiment, the original Læda's law only sets the residual error exceeding 3 times the standard deviation as abnormal data. However, in actual situations, there may be some abnormal data whose residual errors are not particularly large, which will be missed by the law. Therefore, the present invention improves the Læda's law by directly dividing the residual error by the standard deviation to judge abnormal data.

[0058] Exemplarily, the preset first abnormality level is as follows:

[0059] Table 1 First abnormality table

[0060]

[0061]

[0062] Figure 5 It is a box plot provided by an embodiment of the present invention. Figure 5 The box plots shown in FIG. 1 are only for the purpose of illustrating the box plot algorithm and are not intended to be limiting. Figure 5 As shown, in some embodiments, S302 may include:

[0063] Calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate;

[0064] Draw a box plot based on the absolute value of the difference and the box plot algorithm;

[0065] Calculate the second abnormality of each actual main steam flow rate according to the position value of each absolute value of the difference in the box plot, the preset upper / lower boundary value, and the first preset formula, or find the value of the second abnormality of each actual main steam flow rate from a preset second abnormality table according to the position value of the absolute value of the difference in the box plot;

[0066] Among them, the first preset formula is:

[0067]

[0068] Among them, l 2i is the second abnormality of the ith actual main steam flow, Δx i is the position value of the absolute value of the difference corresponding to the ith actual main steam flow in the box plot, x Q is the median of the absolute value of the difference corresponding to each actual main steam flow rate, M is the preset upper / lower boundary value, C i is the preset constant corresponding to the i-th actual main steam flow rate.

[0069] The box plot, also known as the box-and-whisker plot, box-type plot, box-shaped plot or box-and-line plot, is a statistical chart used to display the dispersion of a set of data. It is named because of its box-like shape. It is also often used in various fields, often in quality management, to quickly identify outliers. The biggest advantage of the box plot is that it is not affected by outliers, and can accurately and stably depict the discrete distribution of data, and it is also conducive to data cleaning. The box plot can set a standard for identifying outliers through the program, that is, values ​​greater than or less than the upper and lower limits set by the box plot are identified as outliers.

[0070] Reference Figure 4, upper quartile U: means that only 1 / 4 of all samples have values ​​greater than U, that is, U is at 25% when sorted from large to small. Lower quartile L: means that only 1 / 4 of all samples have values ​​less than L, that is, L is at 75% when sorted from large to small. Median: means the number in the middle of a group of numbers arranged from small to large. If the number of sequences is an even number, the median of the group is the average of the two middle numbers. Upper edge value: means the maximum value within the non-abnormal range. The interquartile range is IQR=UL, then the upper edge value is U+1.5IQR. Lower edge value: means the minimum value within the non-abnormal range. The lower edge value is L-1.5IQR.

[0071] In this embodiment, after drawing the box plot, the second abnormality is determined according to the position value of each absolute value of the difference in the box plot, which can be implemented in one of the following two implementation methods:

[0072] In a first implementation, the second abnormality degree of each actual main steam flow is calculated according to the position value of each absolute value of the difference in the box plot, the preset upper / lower boundary value, and the first preset formula.

[0073] In a second implementation, the second abnormality value of each actual main steam flow rate is searched from a preset second abnormality table according to the position value of the absolute value of the difference in the box plot.

[0074] In this embodiment, the preset constant is used to make the calculated second abnormality degree have the same order of magnitude as the first abnormality degree. In the box plot, the farther the position corresponding to the median is, the larger the preset constant is. For example, if the position value of the absolute value of each difference in the box plot exceeds the preset upper and lower boundary values, then C i = 10. If the absolute value of each difference in the box plot does not exceed the preset upper and lower boundary values, but is greater than U or less than L, then C i =5. If L≤Δx i ≤U, then C i =0.

[0075] Exemplarily, the preset second abnormality table is as follows:

[0076] Table 2 Second abnormality table

[0077]

[0078] In some embodiments, S303 includes:

[0079] Determine the overshoot of each actual main steam flow rate according to a second preset formula;

[0080] The third abnormality degree of each actual main steam flow rate is obtained by multiplying the overshoot amount by a preset parameter, or the value of the third abnormality degree of each actual main steam flow rate is found from a preset third abnormality degree table according to the overshoot amount.

[0081] Among them, the second preset formula is as follows:

[0082]

[0083] Among them, OS is the overshoot, y i is the actual main steam flow rate of the ith 0i is the target main steam flow corresponding to the i-th actual main steam flow.

[0084] In this embodiment, after determining the overshoot of each actual main steam flow rate, determining the second abnormality according to the overshoot of each actual main steam flow rate can be achieved by one of the following two implementation methods:

[0085] In a first implementation manner, the overshoot amount is multiplied by a preset parameter to obtain a third abnormality degree of each actual main steam flow rate.

[0086] In a second implementation, the value of the third abnormality degree of each actual main steam flow rate is searched from a preset third abnormality degree table according to the overshoot amount.

[0087] In this embodiment, the preset parameters are used to make the calculated third abnormality degree have the same order of magnitude as the first abnormality degree. For example, if the decimal order of magnitude is used, and the first abnormality degree is 5, then its order of magnitude is 0, and the third abnormality degree is also 0, that is, the third abnormality degree is greater than or equal to 0 and less than 10.

[0088] Exemplarily, the preset third abnormality table is as follows:

[0089] Table 1 The third abnormality table

[0090] Is it abnormal? The third abnormality OS>20% yes 3 10% ≤ OS ≤ 20% Possibly abnormal 1 OS>10% no 0

[0091] In some embodiments, S304 specifically includes:

[0092] Calculating the abnormality sum of the first abnormality, the second abnormality and the third abnormality, dividing the first abnormality, the second abnormality and the third abnormality by the abnormality sum respectively to obtain the weight of the first abnormality, the weight of the second abnormality and the weight of the third abnormality, and performing weighted summation of the first abnormality, the second abnormality and the third abnormality to obtain the comprehensive abnormality of each actual main steam flow rate;

[0093] Alternatively, the first abnormality degree, the second abnormality degree and the third abnormality degree are added together to obtain a comprehensive abnormality degree of each actual main steam flow rate.

[0094] In this embodiment, after determining the first abnormality, the second abnormality, and the third abnormality, determining the comprehensive abnormality according to the first abnormality, the second abnormality, and the third abnormality can be implemented in one of the following two implementations:

[0095] In a first implementation, the sum of the abnormalities of the first abnormality, the second abnormality, and the third abnormality is calculated, and the first abnormality, the second abnormality, and the third abnormality are respectively divided by the sum of the abnormalities to obtain the weight of the first abnormality, the weight of the second abnormality, and the weight of the third abnormality, and the first abnormality, the second abnormality, and the third abnormality are weighted and summed to obtain the comprehensive abnormality of each actual main steam flow.

[0096] In this implementation, the weight changes with the degree of abnormality, that is, the higher the degree of abnormality, the greater the corresponding weight. If the comprehensive degree of abnormality is obtained by weighted summation, the three recognition algorithms can be combined to identify abnormal data, that is, for a piece of data, if the data is obviously deviated from the normal value in one of the recognition algorithms, but is recognized as a normal value in the other two recognition algorithms, it can still be identified as an abnormal value through the comprehensive degree of abnormality. Therefore, the accuracy of data recognition can be effectively improved.

[0097] In a second implementation, the first abnormality degree, the second abnormality degree and the third abnormality degree are added together to obtain a comprehensive abnormality degree of each actual main steam flow rate.

[0098] In this implementation, the method of adding the three abnormality degrees requires that at least two recognition algorithms indicate that the data is abnormal before the data can be identified as abnormal, so as to overcome the erroneous recognition caused by the defects of the single data recognition method in the prior art. For example, if the first abnormality degree, the second abnormality degree, and the third abnormality degree are all 3, they indicate that the data is abnormal, and the comprehensive abnormality degree must be greater than 3 to indicate that the data is abnormal. That is, if the first abnormality degree, the second abnormality degree, and the third abnormality degree of a data are 3, 0, and 1 respectively, the comprehensive abnormality degree is 4, and the data is identified as abnormal data.

[0099] In some embodiments, after S302, the method further includes:

[0100] According to the first abnormality degree and the preset first abnormality degree table, the second abnormality degree and the preset second abnormality degree table, the third abnormality degree and the preset third abnormality degree table, the first abnormality number, the second abnormality number and the third abnormality number are determined respectively; wherein the first abnormality number represents the number of abnormal data determined by using the Leyda's rule judgment algorithm alone; the second abnormality number represents the number of abnormal data determined by using the box plot algorithm alone; the third abnormality number represents the number of abnormal data determined by using the overshoot judgment algorithm alone;

[0101] According to the first abnormal number, the second abnormal number, the third abnormal number, the number of marked abnormal data, and the correction formula, correction coefficients of the first abnormality degree, the second abnormality degree, and the third abnormality degree are obtained;

[0102] The correction coefficients of the first abnormality, the second abnormality and the third abnormality are respectively multiplied by their corresponding weights and normalized to obtain the correction weights;

[0103] The correction formula is as follows:

[0104]

[0105] Among them, λ1 is the correction coefficient corresponding to the first abnormality, λ2 is the correction coefficient corresponding to the second abnormality, and λ3 is the correction coefficient corresponding to the third abnormality.

[0106] In this embodiment, data collected from different models of steam turbines or different time periods may have different characteristics, and since different data may have different adaptability to several methods, it is necessary to modify the weights accordingly to improve the accuracy of data recognition. The recognition accuracy of each recognition algorithm can be obtained by using the number of abnormal data identified by each algorithm and the number of abnormal data identified in combination, and the accuracy can be used as a correction factor to modify the weights.

[0107] Figure 5 Schematic diagram of the structure of the regression model provided by the embodiment of the present invention. Figure 6 As shown, in some embodiments, before S202, the method further includes:

[0108] Obtain a training sample set;

[0109] A regression model is constructed based on the training sample set and BP neural network algorithm.

[0110] BP neural network is a multi-layer feedforward network trained by the error back propagation algorithm. The learning process consists of two processes: forward propagation of signals and reverse propagation of errors. During forward propagation, each neuron starts from the input layer, receives the input of the previous level, calculates the activation function and inputs it to the next level until the output layer. During reverse propagation, the error between the output value of the output layer and the expected value is returned layer by layer, and the weight matrix is ​​continuously modified. The process of continuous modification of weights is the process of network learning, which continues until the error of the network output is gradually reduced to an acceptable level or the set number of learning times is reached.

[0111] like Figure 5 As shown in Figure 1, the regression model has a three-layer network structure, namely, input layer, hidden layer, and output layer. n ] is the input sample, corresponding to the actual operation data of the steam turbine, [Y1,Y2,…Yn ] is the target main steam flow rate of the corresponding turbine.

[0112] Let the hidden layer weight matrix be W ij , the output layer weight matrix is ​​T jk , then the output of each unit in the hidden layer is calculated as follows:

[0113] O j =f(∑W ij X i -θ j ) (5)

[0114] The output of the output layer is:

[0115] Y j =f(∑T ij O j -θ k ) (6)

[0116] Among them, θ represents the neuron threshold, and f represents the nonlinear function. Then, according to the gradient descent criterion of the BP neural network, the error function e is defined k After that, the weights are updated until the error converges to the expected value and the training ends:

[0117]

[0118] W jk =W jk +ηH j e k (8)

[0119] Among them, i=1,2,…,n; j=1,2,…,l; k=1,2,…,m.

[0120] The above-mentioned method for identifying abnormal operation data of a steam turbine is described below by means of an implementation example, but it is not intended to be limiting. Figure 6 FIG. 1 is a flow chart of a method for identifying abnormal operation data of a steam turbine provided by an embodiment of the present invention. Figure 6 As shown, the specific steps of this implementation example are as follows:

[0121] Step 1: Obtain training samples, build a deep learning BP neural network model, and select model parameters; input the training sample data into the BP neural network, perform fitting regression, and construct a regression model.

[0122] Step 2, input the collected unit load, main steam pressure, main steam temperature, pressure after regulating stage, temperature after regulating stage, reheat steam pressure, and reheat steam temperature at multiple times into the regression model to obtain the target main steam flow at multiple times.

[0123] Step 3: Calculate the absolute value of the difference between the target main steam flow rate and the actual main steam flow rate at each moment Δx i .

[0124] Step 4, calculate Δx i The residual error v i and standard deviation σ, given by |v i | / σCalculate the first abnormality l at each moment 1i .

[0125] Step 5: According to Δx i , draw a box plot, determine the box plot, and calculate the second abnormality l at each moment according to formula (2) 2i Among them, if Δy i If the position value in the box plot exceeds the preset upper and lower boundary values, then C i =10. If Δx i If the position value in the box plot does not exceed the preset upper and lower boundary values, but is greater than U or less than L, then C i =5. If L≤Δx i ≤U, then C i =0.

[0126] Step 6: Calculate the overshoot at each moment according to formula (3), and multiply the overshoot by the preset parameter to obtain the third abnormality degree l at each moment. 3i .

[0127] The execution order of steps 4, 5, and 6 is not limited. For example, the three steps can be executed in parallel or in sequence.

[0128] Step 7: Comprehensive abnormality l i = l 1i ×w1+l 2i ×w2+l 3i ×w3. Among them, w1=l 1i / (l 1i +l 2i +l 3i ), w1=l 2i / (l 1i +l 2i +l 3i ), w1=l 3i / (l 1i +l 2i +l 3i ).

[0129] The actual main steam flow rate whose comprehensive abnormality is greater than or equal to the first preset threshold is marked as abnormal data.

[0130] In this implementation example, the residual error of the data at one time is twice the standard deviation, so its abnormality under the Léida rule is 2. The calculated overshoot is 0.15, so its abnormality under the overshoot method is 1.5. In the box plot, the upper edge value is 100, the lower edge value is 40, the median is 80, and the point corresponding to the data is 120. The point corresponding to the data exceeds the upper limit, so its abnormality under the box plot method is [(120-100) / 80]×10=2.5, so the total abnormality of the data is 2+1.5+2.5=6, the weight of the Leida law is 2÷6=1 / 3, the weight of the overshoot law is 1.5÷6=1 / 4, and the weight under the box plot law is 2.5÷6=5 / 12; the final abnormality is 2×1 / 3+1.5×1 / 4+2.5×5 / 12≈2.08, and the first preset threshold is 2, so the data is marked as abnormal data.

[0131] The above-mentioned method for identifying abnormal operation data of a steam turbine is described below by another implementation example, but it is not intended to be limiting. Figure 7 FIG. 1 is a flow chart of a method for identifying abnormal operation data of a steam turbine provided by another embodiment of the present invention. Figure 7 As shown, the specific steps of this implementation example are as follows:

[0132] Step 1: Obtain training samples, build a deep learning BP neural network model, and select model parameters; input the training sample data into the BP neural network, perform fitting regression, and construct a regression model.

[0133] Step 2, input the collected unit load, main steam pressure, main steam temperature, pressure after regulating stage, temperature after regulating stage, reheat steam pressure, and reheat steam temperature at multiple times into the regression model to obtain the target main steam flow at multiple times.

[0134] Step 3: Calculate the absolute value of the difference between the target main steam flow rate and the actual main steam flow rate at each moment Δx i .

[0135] Step 4, calculate Δx i The residual error v i and standard deviation σ, query the first abnormality degree l at each moment from the preset first abnormality degree table 1i .

[0136] Step 5: According to Δx i , draw a box plot, determine the box plot, query the second abnormality degree l at each moment from the preset second abnormality degree table 2i .

[0137] Step 6: Calculate the overshoot at each moment according to formula (3), and query the third abnormality level l at each moment from the preset second abnormality level table.3i .

[0138] The execution order of steps 4, 5, and 6 is not limited. For example, the three steps can be executed in parallel or in sequence.

[0139] Step 7: Comprehensive abnormality l i = l 1i +l 2i +l 3i The actual main steam flow rate whose comprehensive abnormality is greater than or equal to the first preset threshold is marked as abnormal data.

[0140] In this implementation example, for data at one moment, the first abnormality degree is 3, the second abnormality degree is 0, and the third abnormality degree is 3. If the first preset threshold is 4, 3+0+3>4, and the data is marked as abnormal data.

[0141] In this embodiment, by collecting the actual operating data and actual main steam flow of the steam turbine at multiple times within the target time period, the collected actual main steam flow is compared with the target main steam flow obtained according to the actual operating data, and the abnormality degrees of multiple recognition algorithms are calculated to determine the comprehensive abnormality degree. Then, the abnormal data in the collected actual main steam flow is marked according to the comprehensive abnormality degree. This can prevent erroneous recognition caused by a single recognition algorithm, effectively improve the accuracy of abnormal data recognition, and timely discover the operating abnormality of the steam turbine.

[0142] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0143] Figure 8 It is a structural schematic diagram of a steam turbine operation abnormality data identification device provided by an embodiment of the present invention.

[0144] like Figure 8 As shown, the steam turbine operation abnormality data identification device 8 comprises:

[0145] The acquisition module 810 is used to acquire the actual operation data and the actual main steam flow rate of the steam turbine at multiple times within the target period.

[0146] The prediction module 820 is used to input the actual operation data into the pre-built regression model to predict the target main steam flow rate of the steam turbine at each moment in the target period under normal conditions.

[0147] The determination module 830 is used to determine at least two abnormality levels of each actual main steam flow rate according to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms; wherein each identification algorithm corresponds to one abnormality level.

[0148] The marking module 840 is used to determine the comprehensive abnormality of each actual main steam flow according to at least two abnormalities of each actual main steam flow, and mark the actual main steam flow with a comprehensive abnormality greater than or equal to a first preset threshold as abnormal data.

[0149] Optionally, at least two identification algorithms include: a Lageda's law determination algorithm, an overshoot determination algorithm, and a box plot algorithm.

[0150] The determination module 830 is specifically used to determine the first abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, the target main steam flow rate corresponding to each actual main steam flow rate, and the Lleyda law judgment algorithm;

[0151] Determining the second abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and an overshoot determination algorithm;

[0152] Determining a third abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and a box plot algorithm;

[0153] The marking module 840 is used to add the first abnormality degree, the second abnormality degree and the third abnormality degree or perform weighted summation to obtain the comprehensive abnormality degree of each actual main steam flow rate.

[0154] Optionally, the determination module 830 is specifically used to calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate;

[0155] Calculate the standard deviation and each residual error of the series formed by the absolute values ​​of all differences; the residual error corresponds to the actual main steam flow rate one by one;

[0156] The ratio of the absolute value of the residual error of each actual main steam flow rate to the standard deviation is calculated and used as the value of the first abnormality degree of each actual main steam flow rate, or the value of the first abnormality degree of each actual main steam flow rate is found from a preset first abnormality degree table according to the residual error and standard deviation of each actual main steam flow rate.

[0157] Optionally, the determination module 830 is specifically used to calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate;

[0158] Draw a box plot based on the absolute value of the difference and the box plot algorithm;

[0159] Calculate the second abnormality of each actual main steam flow rate according to the position value of each absolute value of the difference in the box plot, the preset upper / lower boundary value, and the first preset formula, or find the value of the second abnormality of each actual main steam flow rate from a preset second abnormality table according to the position value of each absolute value of the difference in the box plot;

[0160] Among them, the first preset formula is:

[0161]

[0162] Among them, l 2i is the second abnormality of the ith actual main steam flow, Δx i is the position value of the absolute value of the difference corresponding to the ith actual main steam flow in the box plot, x Q is the median of the absolute value of the difference corresponding to each actual main steam flow rate, M is the preset upper / lower boundary value, C i is the preset constant corresponding to the i-th actual main steam flow rate.

[0163] Optionally, the determination module 830 is specifically configured to determine the overshoot of each actual main steam flow rate according to a second preset formula;

[0164] The third abnormality degree of each actual main steam flow rate is obtained by multiplying the overshoot amount by a preset parameter, or the value of the third abnormality degree of each actual main steam flow rate is found from a preset third abnormality degree table according to the overshoot amount.

[0165] Among them, the second preset formula is as follows:

[0166]

[0167] Among them, OS is the overshoot, yi is the i-th actual main steam flow rate, and y0i is the target main steam flow rate corresponding to the i-th actual main steam flow rate.

[0168] Optionally, the marking module 840 is specifically used to calculate the abnormality sum of the first abnormality, the second abnormality and the third abnormality, divide the first abnormality, the second abnormality and the third abnormality by the abnormality sum respectively to obtain the weight of the first abnormality, the weight of the second abnormality and the weight of the third abnormality, and perform weighted summation of the first abnormality, the second abnormality and the third abnormality to obtain the comprehensive abnormality of each actual main steam flow;

[0169] Alternatively, the first abnormality degree, the second abnormality degree and the third abnormality degree are added together to obtain a comprehensive abnormality degree of each actual main steam flow rate.

[0170] Optionally, the steam turbine operation abnormality data identification device 8 further includes: a correction module 850.

[0171] The correction module 850 is used to determine the first abnormality number, the second abnormality number, and the third abnormality number according to the first abnormality degree and the preset first abnormality degree table, the second abnormality degree and the preset second abnormality degree table, and the third abnormality degree and the preset third abnormality degree table, respectively; wherein the first abnormality number represents the number of abnormal data determined by using the Leyda rule judgment algorithm alone; the second abnormality number represents the number of abnormal data determined by using the box plot algorithm alone; and the third abnormality number represents the number of abnormal data determined by using the overshoot judgment algorithm alone;

[0172] According to the first abnormal number, the second abnormal number, the third abnormal number, the number of marked abnormal data, and the correction formula, correction coefficients of the first abnormality degree, the second abnormality degree, and the third abnormality degree are obtained;

[0173] The correction coefficients of the first abnormality, the second abnormality and the third abnormality are respectively multiplied by their corresponding weights and normalized to obtain the correction weights;

[0174] The correction formula is as follows:

[0175]

[0176] Among them, λ1 is the correction coefficient corresponding to the first abnormality, λ2 is the correction coefficient corresponding to the second abnormality, and λ3 is the correction coefficient corresponding to the third abnormality.

[0177] Optionally, the actual operating data may include but is not limited to at least one of the following: unit load, main steam pressure, main steam temperature, pressure after regulating stage, temperature after regulating stage, reheat steam pressure, and reheat steam temperature.

[0178] Optionally, the steam turbine operation abnormality data identification device 8 further includes: a training module 860.

[0179] A training module 860 is used to obtain a training sample set;

[0180] A regression model is constructed based on the training sample set and BP neural network algorithm.

[0181] The device for identifying abnormal operation data of a steam turbine provided in this embodiment can be used to execute the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.

[0182] Fig. 9 Schematic diagram of an electronic device provided by an embodiment of the present invention. Fig. 9As shown, an electronic device 9 provided in one embodiment of the present invention includes: a processor 90, a memory 91, and a computer program 92 stored in the memory 91 and executable on the processor 90. When the processor 90 executes the computer program 92, the steps in the above-mentioned steam turbine operation abnormality data identification method embodiments are implemented, for example Figure 2 Alternatively, when the processor 90 executes the computer program 92, the functions of each module / unit in the above-mentioned device embodiments are implemented, for example Figure 8 Functions of modules 810 to 840 are shown.

[0183] Exemplarily, the computer program 92 may be divided into one or more modules / units, one or more modules / units are stored in the memory 91, and are executed by the processor 90 to implement the present invention. One or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 92 in the electronic device 9.

[0184] The electronic device 9 may be a computing device such as an independent physical server, a server cluster, or a cloud server. Those skilled in the art will appreciate that Fig. 9 The electronic device 9 is merely an example and does not limit the electronic device 9 , and may include more or less components than those shown in the figure, or may combine certain components, or may include different components.

[0185] The processor 90 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0186] The memory 91 may be an internal storage unit of the electronic device 9, such as a hard disk or memory of the electronic device 9. The memory 91 may also be an external storage device of the electronic device 9, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 9. Further, the memory 91 may also include both an internal storage unit of the electronic device 9 and an external storage device. The memory 91 is used to store computer programs and other programs and data required by the terminal. The memory 91 may also be used to temporarily store data that has been output or is to be output.

[0187] An embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method for identifying abnormal operation data of a steam turbine are implemented.

[0188] The computer readable storage medium stores a computer program 92, which includes program instructions. When the program instructions are executed by the processor 90, all or part of the processes in the above-mentioned embodiment method are implemented. The computer program 92 can also be used to instruct the relevant hardware to complete. The computer program 92 can be stored in a computer readable storage medium. When the computer program 92 is executed by the processor 90, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program 92 includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signal and telecommunication signal.

[0189] The computer-readable storage medium may be an internal storage unit of the terminal of any of the foregoing embodiments, such as a hard disk or memory of the terminal. The computer-readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0190] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0191] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0192] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0193] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0194] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0195] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0196] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0197] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0198] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for identifying abnormal operation data of a steam turbine, characterized in that: include: Collect actual operation data and actual main steam flow of the steam turbine at multiple times during the target period; Inputting the actual operation data into a pre-built regression model to predict the target main steam flow rate of the steam turbine at each time in a target period under normal conditions; Determining at least two abnormality levels of each actual main steam flow rate according to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms; wherein each identification algorithm corresponds to one abnormality level; According to at least two abnormalities of each actual main steam flow rate, a comprehensive abnormality of each actual main steam flow rate is determined, and the actual main steam flow rate with the comprehensive abnormality greater than or equal to a first preset threshold is marked as abnormal data.

2. The method for identifying abnormal operation data of a steam turbine according to claim 1, characterized in that: The at least two recognition algorithms include: Ledda's law judgment algorithm, overshoot judgment algorithm and box plot algorithm; According to each actual main steam flow rate, each target main steam flow rate and at least two identification algorithms, at least two abnormalities of each actual main steam flow rate are determined, including: Determine the first abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, the target main steam flow rate corresponding to each actual main steam flow rate, and the Lleyda law judgment algorithm; Determining the second abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and an overshoot determination algorithm; Determining a third abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and a box plot algorithm; According to at least two abnormalities of each actual main steam flow rate, a comprehensive abnormality of each actual main steam flow rate is determined, including: The first abnormality degree, the second abnormality degree and the third abnormality degree are added together or weighted summed to obtain a comprehensive abnormality degree of each actual main steam flow rate.

3. The method for identifying abnormal operation data of a steam turbine according to claim 2, characterized in that: According to each actual main steam flow rate, a target main steam flow rate corresponding to each actual main steam flow rate, and a Ledda's law judgment algorithm, a first abnormality degree of each actual main steam flow rate is determined, including: Calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate; Calculating the standard deviation of the series formed by the absolute values ​​of all the differences and each residual error; the residual error corresponds to the actual main steam flow rate one by one; The ratio of the absolute value of the residual error of each actual main steam flow rate to the standard deviation is calculated and used as the value of the first abnormality degree of each actual main steam flow rate, or the value of the first abnormality degree of each actual main steam flow rate is searched from a preset first abnormality degree table according to the residual error of each actual main steam flow rate and the standard deviation.

4. The method for identifying abnormal operation data of a steam turbine according to claim 2, characterized in that: Determining the second abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, the target main steam flow rate corresponding to each actual main steam flow rate, and the box plot algorithm includes: Calculate the absolute value of the difference between each actual main steam flow rate and its corresponding target main steam flow rate; Draw a box plot according to the absolute value of the difference and the box plot algorithm; Calculate the second abnormality of each actual main steam flow rate according to the position value of each absolute value of the difference in the box diagram, the preset upper / lower boundary value, and the first preset formula, or find the value of the second abnormality of each actual main steam flow rate from a preset second abnormality table according to the position value of each absolute value of the difference in the box diagram; Wherein, the first preset formula is: Among them, l 2i is the second abnormality of the ith actual main steam flow, Δx i is the position value of the absolute value of the difference corresponding to the ith actual main steam flow in the box plot, x Q is the median of the absolute value of the difference corresponding to each actual main steam flow rate, M is the preset upper / lower boundary value, C i is the preset constant corresponding to the i-th actual main steam flow rate.

5. The method for identifying abnormal operation data of a steam turbine according to claim 2, characterized in that: The determining of the third abnormality degree of each actual main steam flow rate according to each actual main steam flow rate, the target main steam flow rate corresponding to each actual main steam flow rate, and the overshoot determination algorithm includes: Determine the overshoot of each actual main steam flow rate according to a second preset formula; The overshoot is multiplied by a preset parameter to obtain a third abnormality degree of each actual main steam flow rate, or, according to the overshoot, a value of the third abnormality degree of each actual main steam flow rate is found from a preset third abnormality degree table; Wherein, the second preset formula is as follows: Among them, OS is the overshoot, y i is the actual main steam flow rate of the ith 0i is the target main steam flow corresponding to the i-th actual main steam flow.

6. The method for identifying abnormal operation data of a steam turbine according to claim 2, characterized in that: The first abnormality degree, the second abnormality degree and the third abnormality degree are added or weighted together to obtain a comprehensive abnormality degree of each actual main steam flow rate, including: Calculating the abnormality sum of the first abnormality, the second abnormality and the third abnormality, dividing the first abnormality, the second abnormality and the third abnormality by the abnormality sum respectively to obtain the weight of the first abnormality, the weight of the second abnormality and the weight of the third abnormality, and performing weighted summation of the first abnormality, the second abnormality and the third abnormality to obtain the comprehensive abnormality of each actual main steam flow rate; Alternatively, the first abnormality degree, the second abnormality degree and the third abnormality degree are added together to obtain a comprehensive abnormality degree of each actual main steam flow rate.

7. The method for identifying abnormal operation data of a steam turbine according to claim 6, characterized in that: After performing weighted summation on the first abnormality degree, the second abnormality degree and the third abnormality degree to obtain the comprehensive abnormality degree of each actual main steam flow rate, the method further includes: According to the first abnormality degree and a preset first abnormality degree table, the second abnormality degree and a preset second abnormality degree table, the third abnormality degree and a preset third abnormality degree table, respectively, a first abnormality number, a second abnormality number, and a third abnormality number are determined; wherein the first abnormality number represents the number of abnormal data determined by using the Leyda's rule judgment algorithm alone; the second abnormality number represents the number of abnormal data determined by using the box plot algorithm alone; the third abnormality number represents the number of abnormal data determined by using the overshoot judgment algorithm alone; According to the first abnormal number, the second abnormal number, the third abnormal number, the number of marked abnormal data, and a correction formula, correction coefficients of the first abnormality degree, the second abnormality degree, and the third abnormality degree are obtained; Multiplying the correction coefficients of the first abnormality degree, the second abnormality degree, and the third abnormality degree by their corresponding weights respectively and normalizing them to obtain correction weights; The correction formula is as follows: Among them, λ1 is the correction coefficient corresponding to the first abnormality, λ2 is the correction coefficient corresponding to the second abnormality, and λ3 is the correction coefficient corresponding to the third abnormality.

8. The method for identifying abnormal operation data of a steam turbine according to any one of claims 1 to 7, characterized in that: The actual operation data includes at least one of the following: unit load, main steam pressure, main steam temperature, pressure after regulating stage, temperature after regulating stage, reheat steam pressure, and reheat steam temperature; Before inputting the actual operation data into the pre-built regression model to obtain the target main steam flow rate of the steam turbine within the target period, the method further includes: Obtain a training sample set; The regression model is constructed according to the training sample set and the BP neural network algorithm.

9. An electronic device comprising a memory, a processor, 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 identifying abnormal operation data of a steam turbine as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for identifying abnormal operation data of a steam turbine as described in any one of claims 1 to 8 are implemented.

Images