Coal-fired power plant parallel operation axial flow type induced draft fan stall fault early warning method

Abstract

According to the coal-fired power plant parallel operation axial flow type induced draft fan stall fault early warning method, data screening judgment under the steady-state working condition is carried out through a sliding window based on information entropy, and a normal current curve band corresponding to unit loads is given. The method comprises the following steps: carrying out trend inspection on key data of the induced draft fan based on a nonparametric statistical inspection method, giving out significant trend judgment under confidence coefficient, carrying out real-time calculation on the collected data of the induced draft fan based on historical fault data and a fault early warning model, and comprehensively considering a relation between the opening degree of a movable blade and smoke flow and a current state. And giving out fan stall early warning information by considering the running state of the other induced draft fan running in parallel. The system has a self-learning function, an early warning model can be optimized along with accumulation of equipment operation data, and the early warning system can be integrated with a monitoring system of a power station to realize automatic early warning. And finally, the accuracy and the timeliness of fault early warning are improved, and the method has a relatively high popularization prospect.

Description

Technical Field

[0001] The invention relates to the technical field of fault warning of large axial flow induced draft fans, and in particular to a stall fault warning method for parallel-operated axial flow induced draft fans in coal-fired power plants. Background Art

[0002] Axial induced draft fans are important large auxiliary machines in the wind and smoke system of traditional coal-fired thermal power plants. They are used to exhaust the flue gas in the boiler and maintain the stable negative pressure environment required for the combustion of pulverized coal in the furnace. Their operating status directly affects the stability and safety of the entire thermal production system. However, since the induced draft fan is in a complex working environment for a long time, its operating conditions are poor, and the temperature of the air duct medium is above 100°C. As the equipment operation cycle lengthens, the duct dust accumulation, blade wear, oil system failure and other reasons lead to changes in the air duct resistance characteristics, fan characteristic curve deviation, and sudden closure of the blades during operation. During the operation of the induced draft fan, if its working point deviates from the normal working condition, a stall failure is prone to occur. This stall failure will not only cause air flow fluctuations in the fan, but may also cause equipment vibration, deterioration of thermal effects, and decreased efficiency. In severe cases, it may even cause damage to the induced draft fan blades or cause other system failures, which will have adverse effects on the economic and safe operation of the power plant. At present, the monitoring and early warning of the stall failure of the induced draft fan mainly rely on manual experience and traditional monitoring methods, and there are problems such as untimely early warning and low accuracy.

[0003] Traditional monitoring methods often only focus on one or several parameters of the induced draft fan, such as vibration amplitude, bearing temperature, etc. However, the stall failure of the induced draft fan is a complex process, and the change of a single parameter may not accurately reflect the occurrence of the fault. In addition, traditional monitoring methods can usually only detect abnormalities when the fault has occurred or is about to occur, and lack the ability to warn in advance. This makes power station staff often passive when facing faults and it is difficult to take effective preventive measures. In addition, traditional monitoring methods are easily affected by external interference, such as ambient temperature, electromagnetic interference, etc., resulting in low accuracy and reliability of monitoring results. In summary, traditional induced draft fan stall fault monitoring systems mostly rely on offline diagnosis, which has lags and is difficult to detect and warn of stall faults in time, which may lead to serious production accidents and economic losses. Therefore, it is of great practical significance to develop an effective fault warning method for large axial flow induced draft fans.

[0004] The comparison with the existing technology is as follows:

[0005] Technical comparison with patent CN113653607B "An intelligent early warning diagnosis method for power station fan stall based on system efficiency model"

[0006] The present invention uses a sliding window based on information entropy and non-parametric statistical tests to perform real-time calculation and trend judgment on the induced draft fan data to achieve stall fault warning; while CN113653607B focuses on early warning of stall status by calculating the total volume flow and real-time efficiency of the two fans. There are essential differences between the two in terms of warning mechanism and data processing method.

[0007] The present invention comprehensively considers the relationship between the opening of the moving blades and the flue gas flow rate, and combines the status of the parallel-operating induced draft fans to provide comprehensive stall warning information; while CN113653607B focuses on accurately locating the fan when stall occurs, providing intervention time to avoid accidents. The two have different focuses and methods for fault handling.

[0008] The early warning system of the present invention has a self-learning function and can optimize the early warning model, while CN113653607B focuses on real-time monitoring and early warning, and does not mention self-learning or model optimization. There is a significant difference between the two in the degree of system intelligence.

[0009] Technical comparison with patent CN115596696A "A real-time online prediction method of wind turbine operating status based on data mining"

[0010] The present invention uses information entropy and non-parametric statistical methods to calculate the induced draft fan data in real time and determine the stall state; while CN115596696A builds multiple prediction models based on data mining and focuses on the extraction and analysis of historical operating state parameters. There are essential differences between the two in terms of data processing methods and model construction.

[0011] The present invention focuses on the comprehensive consideration of the opening of the moving blades, the flue gas flow rate and the state of the parallel fans, while CN115596696A evaluates the state through the theoretical stall safety factor, the pressure margin factor and the flow margin factor. The two are different in evaluation indicators and emphasis.

[0012] The present invention aims to improve the accuracy and timeliness of fault warning and has self-learning ability, while CN115596696A mainly improves the operating safety and economy of the fan and emphasizes the rationality of the control strategy. There are essential differences between the two in terms of ultimate goals and functional realization. Summary of the invention

[0013] To solve the above technical problems, the present invention proposes a stall fault warning method for parallel-operated axial flow induced draft fans in coal-fired power plants, aiming to overcome the shortcomings of existing methods and improve the operating safety and reliability of coal-fired power plants.

[0014] To achieve the above object, the technical solution adopted by the present invention is:

[0015] A stall fault warning method for parallel-operated axial-flow induced draft fans in coal-fired power plants is provided. The data under steady-state conditions are screened and judged by a sliding window based on information entropy, and the normal current curve band corresponding to the unit load is given. The trend test of the key data of the induced draft fan is performed based on the non-parametric statistical test method, and a significant trend judgment under the confidence level is given. The collected induced draft fan data is calculated in real time based on the historical fault data and the fault warning model. The fan stall warning information is given by comprehensively considering the relationship between the opening of the moving blades and the flue gas flow rate and the current state, and considering the operating state of another induced draft fan running in parallel. The specific steps include the following:

[0016] S1: Based on the historical operation data of axial flow induced draft fans, sensor noise and outliers are removed through data cleaning, and the data under steady state is screened out through the sliding window steady state judgment method based on information entropy. Then, the curve band of induced draft fan current and unit load is fitted by statistical methods, and the calculation function relationship between the fan current and the unit fitting curve of the upper and lower limits of the induced draft fan current curve band corresponding to the unit load under a certain confidence level is given;

[0017] S2: Based on the DCS or SIS system of the thermal power plant, real-time collection of the operating sequence data of the induced draft fan, including but not limited to the opening degree of the induced draft fan, flue gas flow, motor current, fan speed, unit load, pressure, temperature, and vibration amplitude;

[0018] S3: Based on historical fault data and normal operation data, the collected data is subjected to feature extraction using signal processing technology to extract key feature quantities that can characterize the stall state. Feature parameters include parameter change trends, fluctuation amplitudes, and spectrum characteristics.

[0019] S4: Based on the extracted characteristic parameters, a non-parametric statistical test method is used to establish an induced draft fan stall fault warning model. For the axial flow induced draft fans running in parallel, the trend test of the time series data of the induced draft fan opening, flue gas flow, motor current, and unit load of the two induced draft fans is performed in real time based on the non-parametric statistical test method. The trend test detects the trend of the time series data curve, including a significant upward trend, a significant upward trend, and no significant trend;

[0020] S5: input the real-time collected operating parameters into the early warning model to make a fault judgment. If the model judges that the induced draft fan may have a stall fault, a warning signal is issued;

[0021] S6: When the early warning model determines that the induced draft fan is stalled, the early warning signal will be promptly notified to the power station staff so that corresponding preventive measures can be taken. At the same time, the early warning information will be uploaded to the power station's monitoring system to integrate the early warning system with the power station's monitoring system for remote monitoring and decision-making by operation and maintenance personnel.

[0022] As a further improvement of the present invention, in step S1, the upper and lower limits of the wind turbine current and the unit fitting curve are determined by using a confidence interval method based on polynomial fitting according to historical operation data;

[0023] Polynomial fitting is an extension of linear regression, and the model form is y = β0 + β1x + β2x 2 +L+β m x m +∈, the principle of calculating the confidence interval is similar to that of linear regression, and the calculation steps are as follows:

[0024] S101: Estimating Polynomial Coefficients by Least Squares Method Get the fitting polynomial

[0025] S102: Calculate the covariance matrix Its elements Represents the estimated coefficient and The covariance of .

[0026] S103: For a given x0, the variance of the predicted value is estimated to be

[0027] S104: Calculate the confidence interval as:

[0028] Lower limit

[0029] Upper limit: where t α / 2,n-(m+1) is the two-sided quantile of the t-distribution with n-(m+1) degrees of freedom.

[0030] As a further improvement of the present invention, in step S2, historical operation data of the induced draft fan is obtained based on the DCS or SIS system of the thermal power plant, and a data sequence required for steady state determination is obtained from the thermal process;

[0031] S201: First, data preprocessing is performed to remove sensor noise and outliers through data cleaning to ensure data accuracy. For missing values, mean filling, interpolation and other methods are used for processing. For outliers, including values ​​that obviously deviate from the normal range, they are identified and corrected by setting thresholds. Data points that exceed the normal operating range by ±20% are considered outliers and are remeasured or corrected according to actual conditions.

[0032] S202: Since different parameters in the thermal process may have different dimensions and numerical ranges, in order to facilitate the subsequent information entropy calculation, the data needs to be normalized using the minimum-maximum normalization formula: Where x is the original data, x min and x maxare the minimum and maximum values ​​in the original data, respectively, new is the normalized data;

[0033] S203: according to the characteristics of the thermal process and the time scale of the data, the time series data is divided into a plurality of sliding windows, the time width of the sliding window is greater than the sampling period of the DCS or SIS system, and further, the time width of the sliding window is at least 6 times the sampling period of the DCS or SIS system;

[0034] S204: For each data window, calculate its information entropy. Assume that the data in the data window is x1, x2, L, x n ,, probability distribution p(x i ) is determined by counting the frequency of occurrence of each data value. The calculation formula for information entropy is:

[0035] S205: When the information entropy of the unit load and current are both less than the set threshold, the boiler is considered to be in a relatively stable operating state. The information entropy calculated for each data window is compared with the determined steady-state threshold. If the information entropy is less than or equal to the threshold, the data in the data window is considered to be in a steady state; if the information entropy is greater than the threshold, the data is considered to be in a non-steady state, and then the data under steady-state conditions is obtained.

[0036] As a further improvement of the present invention, step S2 is a method for obtaining real-time data from a DCS or SIS system of a thermal power plant based on a network communication protocol, and the network communication protocol includes but is not limited to OPC UA, OPC UA, Modbus TCP, WebAPI, and WebSocket.

[0037] As a further improvement of the present invention, the step S4 uses a non-parametric statistical test method based on a statistical method to determine the trend of the time series data curve, which has no requirements on the distribution of the time series data and is suitable for processing time series data with non-normal distribution or outliers.

[0038] Null hypothesis H0: There is no monotonic trend;

[0039] Alternative hypothesis H1: There is a monotonic trend;

[0040] The initial hypothesis is that H0 is true, and before rejecting H0 and accepting H1, the data must be beyond reasonable doubt—reach a certain level of confidence.

[0041] The steps are as follows:

[0042] S401: Data preparation: There is a set of time series data X = {x1, x2, ..., x n}, where x irepresents the i-th observation in the time series, and n is the total number of observations;

[0043] S402: Calculate the S statistic and define each pair of data (x i ,x j ), where j>i, the S statistic is obtained by symbolically calculating all the differences;

[0044]

[0045] in,

[0046]

[0047] The meaning of is the sum of the signs of the differences between all pairs of observations. If there is an upward trend in the data, the value of S should be positive; conversely, if there is a downward trend, the value of S should be negative.

[0048] S403: Calculate the variance of S. Under the assumption that there is no trend, calculate the variance Var(S) of the statistic S. When there are repeated data points in the time series, the calculation formula of the variance is as follows:

[0049]

[0050] in:

[0051] n is the total number of data points;

[0052] m is the number of repeating value groups;

[0053] t k is the number of data points in each replicate group;

[0054] When there are no repeated data points in the time series, or the repeated data points are removed, the formula for the variance is simplified to:

[0055]

[0056] S404: Calculate the standardized test statistic Z, and obtain the Z value by standardizing the S statistic. The calculation formula of Z is as follows:

[0057]

[0058] In the formula, the Z value is used to judge the significance of the trend;

[0059] S405: Determine the significance level, select the significance level α, which is 0.01 or 0.05 or 0.1, and then find the corresponding critical Z according to the normal distribution table α2 ;

[0060] If |Z|>Z α2, then reject the null hypothesis and believe that there is a significant trend in the data;

[0061] If |Z|≤Z α2 , then the null hypothesis cannot be rejected, and it is believed that there is no significant trend in the data;

[0062] S406: Determine the trend direction;

[0063] When Z>0, it indicates that there is a significant upward trend in the time series;

[0064] When Z<0, it indicates that there is a significant downward trend in the time series;

[0065] When Z = 0, it means there is no significant trend;

[0066] S407: Calculate the gradient slope by the Theil-Sen estimator;

[0067] In order to estimate the strength of the trend, the Theil-Sen slope estimator is used to calculate the trend slope of the time series, and its calculation formula is:

[0068]

[0069] This estimator gives the median slope between each pair of data points in the time series, which is used to indicate the size of the trend.

[0070] As a further improvement of the present invention, in step S5, for two fans running in parallel, respectively called fan A and fan B, firstly, the changing trends of the opening degree of induced draft fan A and the smoke volume of the fan are calculated based on the non-parametric statistical test method. If the opening degree of induced draft fan A shows a significant upward trend, and the smoke volume flowing through induced draft fan A shows a significant downward trend, and the operating current of induced draft fan A exceeds the limit, then the changing trend of the current of induced draft fan B is calculated based on the non-parametric statistical test method. If the current of induced draft fan B shows a significant upward trend at this time, it is considered that fan A is stalled.

[0071] As a further improvement of the present invention, the types of warning information in step S6 include but are not limited to:

[0072] (1) Sound and light alarm signal: The alarm emits sound and flashes light to alert the operator to the fault condition;

[0073] (2) Digital display alarm: Displays fault code, fault type and severity on the control panel or monitoring system to facilitate quick problem location;

[0074] (3) Remote alarm notification: Notify relevant personnel via SMS, email or APP to ensure rapid response when a fault occurs;

[0075] (4) Visual monitoring: The monitoring system provides real-time video monitoring to observe the operating status of the induced draft fan and detect abnormal conditions in a timely manner;

[0076] (5) Data recording and analysis report: After a fault occurs, the system generates data recording and analysis reports to help technicians diagnose the fault and conduct subsequent processing;

[0077] (6) Status indicator: An indicator light is set on the device to indicate normal, warning or fault status through different colors or flashing patterns.

[0078] As a further improvement of the present invention, when the operating status of the fan is adjusted or overhauled, the fault warning model is automatically updated, and the calculation function relationship of the induced draft fan current curve with upper and lower limit curves corresponding to the unit load under a certain confidence level is obtained again, and the monitoring thresholds of the induced draft fan opening, flue gas flow relationship, and current state trend curve are obtained again to ensure the accuracy of the warning.

[0079] As a further improvement of the present invention, in step S6, the warning information is uploaded to the monitoring system of the power station, so that the warning system can be integrated with the monitoring system of the power station, and the data uploading method includes but is not limited to HttpAPI interface, WebSocket interface, and IEC104 interface.

[0080] Compared with the prior art, the present invention has the following beneficial effects:

[0081] The present invention provides a stall fault warning method for large axial flow induced draft fans in parallel operation for coal-fired power station boilers, which can effectively solve the problem of difficulty in early warning of stall faults of large axial flow induced draft fans in the prior art.

[0082] The method can timely detect the signs of stall failure and give early warning through real-time monitoring and intelligent analysis. By comprehensively considering the relationship between the opening of the moving blades and the flue gas flow rate and the current state, and considering the operating state of another induced draft fan running in parallel, the operating state of the induced draft fan can be more comprehensively understood, the signs of stall can be discovered in time, and the accuracy and reliability of the stall judgment of the axial flow induced draft fan can be improved. The system has a self-learning function, and can continuously optimize the early warning model as the equipment operation data accumulates, and adapt to different working conditions. The early warning system can be integrated with the monitoring system of the power station to realize automatic early warning, reduce manual intervention, and improve the operation efficiency and safety of the power station. The method and system of the present invention are easy to implement and can be applied to existing axial flow induced draft fan equipment without large-scale transformation. The accuracy and timeliness of fault early warning are improved. The method has high real-time and accuracy, can provide a strong guarantee for the safe and stable operation of coal-fired power plants, and has high practical value and promotion and application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a schematic flow chart of the induced draft fan stall fault warning method in the present invention;

[0084] Figure 2 It is a schematic diagram of steady-state data screening based on a sliding window in the present invention;

[0085] Figure 3 It is a schematic diagram of the execution of the stall fault warning algorithm established by the non-parametric statistical test method in the present invention. DETAILED DESCRIPTION

[0086] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0087] The present invention provides a stall fault warning method for large axial flow induced draft fans in parallel operation for coal-fired power station boilers, such as Figure 1 As shown, the following steps are included:

[0088] S1: Based on the historical operation data of the axial flow induced draft fan, sensor noise and outliers are removed by data cleaning, and the data under steady state is screened out by the sliding window steady-state judgment method based on information entropy. Then, the curve band of the induced draft fan current and the unit load is fitted by a statistical method, and the calculation function relationship of the upper and lower limit curves of the induced draft fan current curve band corresponding to the unit load under a certain confidence level is given.

[0089] S2: Based on the DCS or SIS system of the thermal power plant, the operation sequence data of the induced draft fan is collected in real time, including but not limited to the opening degree of the induced draft fan, flue gas flow, motor current, fan speed, unit load, pressure, temperature, and vibration amplitude.

[0090] S3: Based on historical fault data and normal operating data, signal processing technology is used to extract features from the collected data to extract key feature quantities that can characterize the stall state. These feature parameters may include parameter change trends, fluctuation amplitudes, spectral characteristics, etc.

[0091] S4: Based on the extracted characteristic parameters, a non-parametric statistical test method is used to establish an induced draft fan stall fault warning model. For the axial flow induced draft fans running in parallel, the trend test of the time series data of the induced draft fan opening, flue gas flow, motor current, and unit load of the two induced draft fans is performed in real time based on the non-parametric statistical test method. The trend test includes a significant upward trend, a significant upward trend, and no significant trend.

[0092] S5: Input the real-time collected operating parameters into the early warning model to make a fault judgment. If the model judges that the induced draft fan may have a stall fault, a warning signal is issued.

[0093] S6: When the early warning model determines that the induced draft fan is stalled, the early warning signal will be promptly notified to the power station staff so that they can take corresponding preventive measures. At the same time, the early warning information can also be uploaded to the power station's monitoring system, so that the early warning system can be integrated with the power station's monitoring system for remote monitoring and decision-making by operation and maintenance personnel.

[0094] According to the aerodynamic model of the fan, the motor current of the axial flow fan is proportional to the cube of the flue gas flow rate flowing through the induced draft fan, the volume flow rate of the fan is proportional to the fuel amount, the fuel amount is proportional to the load, so the fan current is proportional to the cube of the load.

[0095] In step S1, the confidence interval method based on polynomial fitting is used to determine the upper and lower limits of the wind turbine current and the unit fitting curve according to the historical operation data. Polynomial fitting is an extension of linear regression, and the model form is y=β0+β1x+β2x 2 +L+β m x m +∈. The principle of calculating the confidence interval is similar to that of linear regression, and the calculation steps are as follows:

[0096] S101: Estimating Polynomial Coefficients by Least Squares Method Get the fitting polynomial

[0097] S102: Calculate the covariance matrix Its elements Represents the estimated coefficient and The covariance of .

[0098] S103: For a given x0, the variance of the predicted value is estimated to be

[0099] S104: Calculate the confidence interval as:

[0100] Lower limit

[0101] Upper limit: where t α / 2,n-(m+1) is the two-sided quantile of the t-distribution with n-(m+1) degrees of freedom.

[0102] Furthermore, in step S2, historical operation data of the induced draft fan is obtained based on the DCS or SIS system of the thermal power plant, and a data sequence required for steady-state determination is obtained from the thermal process.

[0103] S201: First, data preprocessing is performed to remove sensor noise and outliers through data cleaning to ensure data accuracy. For missing values, mean filling, interpolation and other methods can be used for processing. For outliers (such as values ​​that are significantly deviated from the normal range), they can be identified and corrected by setting thresholds. For example, data points that exceed the normal operating range by ±20% are considered outliers and can be remeasured or corrected according to actual conditions.

[0104] S202: Since different parameters in the thermal process may have different dimensions and value ranges, in order to facilitate the subsequent information entropy calculation, the data needs to be normalized. Use minimum-maximum normalization, the formula is Where x is the original data, x min and x max are the minimum and maximum values ​​in the original data, respectively, new is the normalized data.

[0105] S203: According to the characteristics of the thermal process and the time scale of the data, the time series data is divided into a number of sliding windows, and the time width of the sliding window is greater than the sampling period of the DCS or SIS system. Furthermore, the time width of the sliding window is at least 6 times the sampling time of the DCS or SIS system.

[0106] S204: For each data window, calculate its information entropy. Assume that the data in the data window are x1, x2, L, x n ,, probability distribution p(x i ) can be determined by counting the frequency of occurrence of each data value. The formula for calculating information entropy is:

[0107] S205: When the information entropy of the unit load and current are both less than the set threshold, the boiler is considered to be in a relatively stable operating state. The information entropy calculated for each data window is compared with the determined steady-state threshold. If the information entropy is less than or equal to the threshold, the data in the data window is considered to be in a steady state; if the information entropy is greater than the threshold, the data is considered to be in a non-steady state. Then the data under steady-state conditions is obtained.

[0108] like Figure 2 As shown in the figure, it is a schematic diagram of a sliding window. Select a suitable window size (N) to slide on the data sequence and extract subsequences. The choice of window size should be adjusted according to the data characteristics and analysis requirements. Apply a sliding window on the data sequence to gradually extract subsequences of fixed size. First, set the starting position and place the window at the beginning of the data sequence. Then calculate the statistical features (such as mean, variance, etc.) within the current window. Finally, move the window forward by one data point and repeat the calculation until the entire data sequence is traversed.

[0109] Furthermore, step S2 is a method for obtaining real-time data from a DCS or SIS system of a thermal power plant based on a network communication protocol, wherein the network communication protocol includes but is not limited to OPC UA, OPC UA, Modbus TCP, Web API, and WebSocket.

[0110] In step S4, a non-parametric statistical test method based on statistical methods is used to determine the trend of the time series data curve. This method has no requirements on the distribution of time series data and is suitable for processing time series data with non-normal distribution or outliers.

[0111] Null hypothesis H0: There is no monotonic trend;

[0112] Alternative hypothesis H1: There is a monotonic trend;

[0113] The initial hypothesis is that H0 is true, and before rejecting H0 and accepting H1, the data must be beyond reasonable doubt—reach a certain level of confidence.

[0114] The steps are as follows:

[0115] S401: Data preparation: There is a set of time series data X = {x1, x2, ..., x n}, where x i represents the i-th observation in the time series and n is the total number of observations.

[0116] S402: Calculate the S statistic and define each pair of data (x i ,x j The S statistic is obtained by performing a sign calculation on all the differences.

[0117]

[0118] in,

[0119]

[0120] The meaning of is the sign and difference of the values ​​between all pairs of observations. If there is an upward trend in the data, the value of S should be positive; conversely, if there is a downward trend, the value of S should be negative.

[0121] S403: Calculate the variance of S. Under the assumption that there is no trend, calculate the variance Var(S) of the statistic S. When there are repeated data points in the time series, the calculation formula of the variance is as follows:

[0122]

[0123] in:

[0124] n is the total number of data points.

[0125] m is the number of repeating value groups.

[0126] t k is the number of data points in each replicate group.

[0127] When there are no repeated data points in the time series, or the repeated data points are removed, the formula for the variance is simplified to:

[0128]

[0129] S404: Calculate the standardized test statistic Z, and obtain the Z value by standardizing the S statistic. The calculation formula of Z is as follows:

[0130]

[0131] In the formula, the Z value is used to judge the significance of the trend.

[0132] S405: Determine the significance level, select the significance level α (usually 0.01, 0.05 or 0.1), and then find the corresponding critical Z according to the normal distribution table α2 .

[0133] If |Z|>Z α2 , then the null hypothesis is rejected and it is believed that there is a significant trend in the data.

[0134] If |Z|≤Z α2 , then the null hypothesis cannot be rejected and it is believed that there is no significant trend in the data.

[0135] S406: Determine the trend direction

[0136] When Z>0, it indicates that there is a significant upward trend in the time series.

[0137] When Z<0, it indicates that there is a significant downward trend in the time series.

[0138] When Z = 0, it means there is no significant trend.

[0139] S407: Calculate the gradient slope (Theil-Sen estimator)

[0140] In order to estimate the strength of the trend, the Theil-Sen slope estimator is used to calculate the trend slope of the time series. The calculation formula is:

[0141]

[0142] This estimator gives the median slope between each pair of data points in the time series, which can be used to indicate the size of the trend.

[0143] For example, for the fan opening time series data, taking α=0.1 as an example, Query the standard normal distribution table, Z0.95=1.645, so when Z≥1.645, the 90% significance test is passed, the H0 hypothesis is not established, Z>0, the sequence has an upward trend, that is, when Z≥1.645, there is a 90% probability that the fan opening shows a significant upward trend; if Z<0, the sequence has a downward trend, that is, when Z≤-1.645, there is a 90% probability that the fan opening shows a significant downward trend. For another example, for the fan current time series data, taking α=0.05 as an example, Query the standard normal distribution table Z0.975 = 1.96, so when Z ≥ 1.96, the 95% significance test is passed, and the H0 hypothesis is not established. Z> 0, the sequence has an upward trend, that is, when Z ≥ 1.96, there is a 95% probability that the wind turbine current presents a significant upward trend; if Z < 0, the sequence has a downward trend, that is, when Z ≤ -1.96, there is a 95% probability that the wind turbine current presents a significant upward trend.

[0144] like Figure 3 Shown is a schematic diagram of the execution of the stall fault warning algorithm established by the non-parametric statistical test method of the present invention. For an axial flow fan with adjustable blades, under normal circumstances, an increase in the blade opening will increase the fan flow rate, and there is a strong linear relationship between the fan's air volume and the blade opening. According to the above fitting results, it can be considered that under the normal operation of the fan, its blade opening and the fan flow rate change linearly. However, when the fan is about to stall, the relationship between its air volume and blade opening will deviate from the relationship during normal operation. Before the fan stalls, its operating current will gradually increase, and a warning of fan stall is issued based on the comprehensive criterion of the air volume-blade opening relationship and the current. In addition, when put into "automatic" mode, the current and volume ratio of another fan running in parallel with the stalled fan can be greatly increased;

[0145] The output increases, which can easily cause motor overload. When the fan stalls, the air volume and air pressure drop significantly, causing drastic changes in furnace combustion, which can easily cause fire accidents; when the other fan running in parallel is put into "automatic", the output increases, which can easily cause motor overload; the vibration of the stalled fan increases significantly, which may cause serious damage to the fan equipment and air duct vibration; if the handling process is not correct, it is easy to cause the fan to "surge" and damage the equipment. Based on the above characteristics

[0146] In step S5, for two fans running in parallel, respectively called fan A and fan B, firstly, the change trend of the opening degree of induced draft fan A and the fan smoke volume is calculated based on the non-parametric statistical test method. If the opening degree of induced draft fan A shows a significant upward trend, and the smoke volume flowing through induced draft fan A shows a significant downward trend, and the operating current of induced draft fan A exceeds the limit, then the change trend of the current of induced draft fan B is calculated based on the non-parametric statistical test method. If the current of induced draft fan B shows a significant upward trend at this time, it is considered that fan A is stalled.

[0147] Furthermore, the types of warning information in step S6 include but are not limited to:

[0148] (1) Audible and visual alarm signals: The alarm emits sounds and flashes lights to alert the operator to fault conditions.

[0149] (2) Digital display alarm: The fault code, fault type and severity are displayed on the control panel or monitoring system to facilitate quick problem location.

[0150] (3) Remote alarm notification: Notify relevant personnel via SMS, email or APP to ensure a quick response when a fault occurs.

[0151] (4) Visual monitoring: The monitoring system provides real-time video monitoring to observe the operating status of the induced draft fan and detect abnormal conditions in a timely manner.

[0152] (5) Data recording and analysis report: After a fault occurs, the system generates data recording and analysis reports to help technicians diagnose the fault and conduct subsequent processing.

[0153] (6) Status indicator: An indicator light is set on the device to indicate normal, warning or fault status through different colors or flashing patterns.

[0154] Furthermore, when the fan operating status is adjusted or overhauled, the fault warning model is automatically updated to obtain the calculation function relationship of the induced draft fan current curve with upper and lower limit curves corresponding to the unit load under a certain confidence level, and to obtain the monitoring threshold of the induced draft fan opening, flue gas flow relationship, and current state trend curve to ensure the accuracy of the warning.

[0155] Furthermore, in step S6, the warning information is uploaded to the monitoring system of the power station, so that the warning system can be integrated with the monitoring system of the power station. The data uploading method includes but is not limited to HttpAPI interface, WebSocket interface, and IEC104 interface.

[0156] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A method for early warning of stall failure of parallel-operated axial flow induced draft fans in coal-fired power plants, characterized in that: The steps include: S1: Based on the historical operation data of the axial flow induced draft fan, the sensor noise and outliers are removed by data cleaning, and the data under steady state is screened out by the sliding window steady state judgment method based on information entropy. Then, the curve band of the induced draft fan current and the unit load is fitted by statistical methods, and the calculation function relationship of the upper and lower limit curves of the induced draft fan current curve band corresponding to the unit load under a certain confidence level is given; S2: Based on the DCS or SIS system of the thermal power plant, real-time collection of the operating sequence data of the induced draft fan, including but not limited to the opening degree of the induced draft fan, flue gas flow, motor current, fan speed, unit load, pressure, temperature, and vibration amplitude; S3: Based on historical fault data and normal operation data, the collected data is subjected to feature extraction using signal processing technology to extract key feature quantities that can characterize the stall state. Feature parameters include parameter change trends, fluctuation amplitudes, and spectrum characteristics. S4: Based on the extracted characteristic parameters, a non-parametric statistical test method is used to establish an induced draft fan stall fault warning model. For the axial flow induced draft fans running in parallel, the trend test of the time series data of the induced draft fan opening, flue gas flow, motor current, and unit load of the two induced draft fans is performed in real time based on the non-parametric statistical test method. The trend test detects the trend of the time series data curve, including a significant upward trend, a significant upward trend, and no significant trend; S5: inputting the real-time collected operating parameters into the early warning model to make a fault judgment. If the model judges that the induced draft fan may have a stall fault, a warning signal is issued; S6: When the early warning model determines that the induced draft fan is stalled, the early warning signal will be promptly notified to the power station staff so that corresponding preventive measures can be taken. At the same time, the early warning information will be uploaded to the power station's monitoring system to integrate the early warning system with the power station's monitoring system for remote monitoring and decision-making by operation and maintenance personnel.

2. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 1, characterized in that: The step S1 is to determine the upper and lower limits of the wind turbine current and the unit fitting curve using a confidence interval method based on polynomial fitting according to historical operation data; Polynomial fitting is an extension of linear regression, and the model form is y = β0 + β1x + β2x 2 +L+β m x m +∈, the principle of calculating the confidence interval is similar to that of linear regression, and the calculation steps are as follows: S101: Estimating Polynomial Coefficients by Least Squares Method Get the fitting polynomial S102: Calculate the covariance matrix Its elements Represents the estimated coefficient and The covariance of S103: For a given x0, the variance of the predicted value is estimated to be S104: Calculate the confidence interval as: Lower limit Upper limit: where t α / 2,n-(m+1) is the two-sided quantile of the t-distribution with n-(m+1) degrees of freedom.

3. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 1, characterized in that: The step S2 is to obtain historical operation data of the induced draft fan based on the DCS or SIS system of the thermal power plant, and obtain the data sequence required to judge the steady state from the thermal process; S201: First, data preprocessing is performed to remove sensor noise and outliers through data cleaning to ensure data accuracy. For missing values, mean filling, interpolation and other methods are used for processing. For outliers, including values ​​that obviously deviate from the normal range, they are identified and corrected by setting thresholds. Data points that exceed the normal operating range by ±20% are considered outliers and are remeasured or corrected according to actual conditions. S202: Since different parameters in the thermal process may have different dimensions and numerical ranges, in order to facilitate the subsequent information entropy calculation, the data needs to be normalized using the minimum-maximum normalization formula: Where x is the original data, x min and x max are the minimum and maximum values ​​in the original data, respectively, new is the normalized data; S203: according to the characteristics of the thermal process and the time scale of the data, the time series data is divided into a plurality of sliding windows, the time width of the sliding window is greater than the sampling period of the DCS or SIS system, and further, the time width of the sliding window is at least 6 times the sampling period of the DCS or SIS system; S204: For each data window, calculate its information entropy. Assume that the data in the data window is x1, x2, L, x n ,, probability distribution p(x i ) is determined by counting the frequency of occurrence of each data value. The calculation formula for information entropy is: S205: When the information entropy of the unit load and current are both less than the set threshold, the boiler is considered to be in a relatively stable operating state. The information entropy calculated for each data window is compared with the determined steady-state threshold. If the information entropy is less than or equal to the threshold, the data in the data window is considered to be in a steady state; if the information entropy is greater than the threshold, the data is considered to be in a non-steady state, and then the data under steady-state conditions is obtained.

4. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 3, characterized in that: The step S2 is a method for acquiring real-time data from a DCS or SIS system of a thermal power plant based on a network communication protocol, wherein the network communication protocol includes but is not limited to OPC UA, OPC UA, Modbus TCP, Web API, and WebSocket.

5. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 1, characterized in that: The step S4 uses a non-parametric statistical test method based on statistical methods to determine the trend of the time series data curve. There is no requirement for the distribution of the time series data, and it is suitable for processing time series data with non-normal distribution or outliers. Null hypothesis H0: There is no monotonic trend; Alternative hypothesis H1: There is a monotonic trend; The initial hypothesis is that H0 is true. Before rejecting H0 and accepting H1, the data must be beyond reasonable doubt—reach a certain level of confidence. The steps are as follows: S401: Data preparation: There is a set of time series data X = {x1, x2, ..., x n }, where x i represents the i-th observation in the time series, and n is the total number of observations; S402: Calculate the S statistic and define each pair of data (x i ,x j ), where j>i, the S statistic is obtained by symbolically calculating all the differences; in, The meaning of is the sum of the signs of the differences between all pairs of observations. If there is an upward trend in the data, the value of S should be positive; conversely, if there is a downward trend, the value of S should be negative. S403: Calculate the variance of S. Under the assumption that there is no trend, calculate the variance Var(S) of the statistic S. When there are repeated data points in the time series, the calculation formula of the variance is as follows: in: n is the total number of data points; m is the number of repeating value groups; t k is the number of data points in each replicate group; When there are no repeated data points in the time series, or the repeated data points are removed, the formula for the variance is simplified to: S404: Calculate the standardized test statistic Z, and obtain the Z value by standardizing the S statistic. The calculation formula of Z is as follows: In the formula, the Z value is used to judge the significance of the trend; S405: Determine the significance level, select the significance level α, which is 0.01 or 0.05 or 0.1, and then find the corresponding critical Z according to the normal distribution table α2 ; If |Z|>Z α / 2 , then reject the null hypothesis and believe that there is a significant trend in the data; If |Z|≤Z α / 2 , then the null hypothesis cannot be rejected, and it is believed that there is no significant trend in the data; S406: Determine the trend direction; When Z>0, it indicates that there is a significant upward trend in the time series; When Z<0, it indicates that there is a significant downward trend in the time series; When Z = 0, it means there is no significant trend; S407: Calculate the gradient slope by the Theil-Sen estimator; In order to estimate the strength of the trend, the Theil-Sen slope estimator is used to calculate the trend slope of the time series, and its calculation formula is: This estimator gives the median slope between each pair of data points in the time series, which is used to indicate the size of the trend.

6. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 1, characterized in that: In step S5, for two fans running in parallel, respectively called fan A and fan B, firstly, the changing trends of the opening degree of induced draft fan A and the smoke volume of the fan are calculated based on the non-parametric statistical test method. If the opening degree of induced draft fan A shows a significant upward trend, and the smoke volume flowing through induced draft fan A shows a significant downward trend, and the operating current of induced draft fan A exceeds the limit, then the changing trend of the current of induced draft fan B is calculated based on the non-parametric statistical test method. If the current of induced draft fan B shows a significant upward trend at this time, it is considered that fan A is stalled.

7. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 1, characterized in that: The types of warning information in step S6 include but are not limited to: (1) Sound and light alarm signal: The alarm emits sound and flashes light to alert the operator to the fault condition; (2) Digital display alarm: Displays fault code, fault type and severity on the control panel or monitoring system to facilitate quick problem location; (3) Remote alarm notification: Notify relevant personnel via SMS, email or APP to ensure rapid response when a fault occurs; (4) Visual monitoring: The monitoring system provides real-time video monitoring to observe the operating status of the induced draft fan and detect abnormal conditions in a timely manner; (5) Data recording and analysis report: After a fault occurs, the system generates data recording and analysis reports to help technicians diagnose the fault and conduct subsequent processing; (6) Status indicator: An indicator light is set on the device to indicate normal, warning or fault status through different colors or flashing patterns.

8. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 7, characterized in that: When the fan operating status is adjusted or overhauled, the fault warning model is automatically updated to obtain the calculation function relationship of the induced draft fan current curve with upper and lower limit curves corresponding to the unit load under a certain confidence level, and to obtain the monitoring threshold of the induced draft fan opening, flue gas flow relationship, and current status trend curve to ensure the accuracy of the warning.

9. The method for early warning of stall failure of parallel-operated axial flow induced draft fans in a coal-fired power plant according to claim 7, characterized in that: In step S6, the warning information is uploaded to the monitoring system of the power station, so that the warning system can be integrated with the monitoring system of the power station. The data uploading method includes but is not limited to HttpAPI interface, WebSocket interface, and IEC104 interface.

Images