A method for real-time online prediction of fan operation status based on data mining
The method uses data mining and neural networks to predict fan operating status changes, ensuring safe and efficient fan operation by preventing stalling and optimizing load capacity in power plant exhaust gas systems.
Patent Information
- Application Number
- JP2023540848
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2022-10-28
- Filing Date
- 2023-04-24
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2043-04-24
AI Technical Summary
Existing power plant fan monitoring systems fail to accurately predict fan operating status changes in exhaust gas systems, leading to conservative adjustments that reduce system resistance and limit load capacity, especially after equipment modifications.
A real-time online prediction method using data mining and artificial neural networks to construct predictive models for fan operating status, incorporating past parameter data to estimate fan performance and safety margins, and adjust fan operation to prevent stalling.
Accurately predicts fan operating status changes, enabling safe and efficient fan operation by preventing stalling and optimizing load capacity.
Smart Images

Figure 2024542332000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to axial fans (including static vane adjustable axial fans and dynamic vane adjustable axial fans, etc.) used in exhaust gas systems of coal-fired power plants, and in particular to a method for predicting the operating status of the fans online in real time based on data mining. [Background technology]
[0002] At present, with the deep implementation of policies such as deep adjustment of flexibility, the fan of the power plant faces the requirement of frequent adjustment of the width load, but after the implementation of the ultra-low emission modification of the unit, the requirements for environmental protection indicators are increasingly higher, and the environmental protection equipment in the exhaust gas system is also increasing. In actual operation, with the improvement of environmental protection requirements, the amount of ammonia scattering in the exhaust gas system increases, and the abnormal blockage phenomenon of the equipment formed by the presence of ammonium sulfate in each equipment in the exhaust gas system becomes more and more serious, and the operating parameters of the exhaust gas system deviate from their design parameters, and the stall of the induced draft fan in the high load operating mode and the reduction of the load capacity of the unit frequently occur. For the stall phenomenon of the induced draft fan at high load, the power plant operators usually use the method of limiting the fan opening and current to prevent the stall of the fan, but because the blockage situation of the exhaust gas system and the operating situation of the induced draft fan cannot be accurately predicted, it is too conservative to use the method of limiting the fan opening and current to prevent the stall of the fan, and even after the maintenance of the exhaust gas system to remove the blockage, the resistance of the exhaust gas system is greatly reduced. In addition, the load capacity of the unit is limited by the fan's anti-stall operation strategy. The conventional power plant fan monitoring system can monitor the state parameters such as the fan's inlet and outlet pressure, flow rate, current and inlet temperature in real time, and evaluate the real-time performance state of the fan in real time. Therefore, based on data mining of the past state parameters of the power plant fan, an online prediction method for the operation state of the power plant fan is established, and the operation status of the fan after the blockage of the exhaust gas system is accurately evaluated, providing a basis for the formulation of the fan's anti-stall operation strategy and the operation adjustment of the fan. For the change of the state parameters of the exhaust gas system after adding new equipment to the exhaust gas system, it is necessary to accurately predict the output state of the fan in order to provide a basis for the formulation of technical modification solutions related to the exhaust gas system. Summary of the Invention [Problem to be solved by the invention]
[0003] The present invention proposes a method for real-time online prediction of the fan operating status based on data mining, in order to accurately predict the operating status of the fan after the operating status of the exhaust gas system changes.
[0004] The present invention is realized by the following technical solutions:
[0005] Based on data mining, the method for real-time online prediction of fan operation status is 1) obtaining historical operating status parameters of the unit and the fan based on the fan online monitoring system and the DCS system; 2) using data analysis techniques to process the historical operating parameters of the fan, extract data under multiple normal and stable operating modes, and build a sample collection; 3) training the extracted sample collection based on an artificial neural network to obtain a prediction model of the fluctuation relationship of the fan inlet flow rate with the unit load or the main steam flow rate, and a prediction model of the system resistance from the furnace room to the induced draft fan inlet stage and the system resistance from the induced draft fan outlet to the chimney outlet stage with the fluctuation relationship of the exhaust gas volume; 4) obtaining a fan status operating parameter based on the two prediction models by providing a fluctuation prediction value of the status parameter of the exhaust gas system; 5) calculating a theoretical stall safety factor, a pressure margin factor, and a flow margin factor for the fan based on the fan performance curve and the predicted values of the fan operating parameters; 6) Setting thresholds for each stall margin coefficient based on past stall operation mode analysis and a large amount of stall test statistical analysis, and evaluating whether the operating status of the fan after the fluctuation of the exhaust gas system status parameters is safe by comparing the deviation between the theoretical stall safety coefficient, the pressure margin coefficient and the flow margin coefficient and the thresholds.
[0006] A further improvement of the present invention is that in step 1), the past operating status parameters of the unit and the fan are: Unit load L, boiler evaporation D b , fan inlet temperature T in , the fan inlet volume flow rate Q v , Fan inlet total pressure P t、in , total pressure at the fan outlet P t、out , fan opening degree β, fan inlet static pressure P e、in and fan outlet static pressure P e、out The time period t is 10 to 30 days, and the time interval Δt is 1 to 5 minutes.
[0007] JPEG2024542332000002.jpg68159
[0008] JPEG2024542332000003.jpg21159
[0009] A further improvement of the present invention is to use an artificial neural network to train based on the sample points in steps (a) and (b), respectively, to calculate the fan inlet volume flow rate Q v and the flow rate of main steam D b The variation relationship model Q v =f(D b ), fan inlet static pressure P e、in and the fan inlet volume flow rate Q v The change relationship model P e、in =f(Q v ), fan outlet static pressure P e、out and the fan inlet volume flow rate Q v The change relationship model P e、out =f(Q v ), fan inlet total pressure P t、in and the fan inlet volume flow rate Q v The change relationship model P t、in =f(Q v ) and the total pressure at the outlet of the fan P t、out and the fan inlet volume flow rate Q v The change relationship model P t、out =f(Q v ) to obtain
[0010] A further improvement of the present invention is to predict the state fluctuation of the exhaust gas system in steps and 3) and obtain a fluctuation value corresponding to the front and rear resistance of the induced draft fan in the driving mode, i.e., (D' b , ΔP in ), (D' b , ΔP out ) to obtain
[0011] A further improvement of the present invention is that in steps and 4), the status operating parameters of the fan are: Based on the basic design parameters of the boiler and the actual operation mode of the unit, the boiler evaporation range [D b、min ,D b、BMCR ] is determined, m typical operation modes are selected in the section, and a set of boiler evaporation amounts {D b、i} is obtained, and a set of situation parameter points in each driving mode, i.e., {(D b , T in , Q v , P e、in , P e、out , P t、in , P t、out , ...) i} where i=1, 2, 3, ..m and m>=3.
[0012] JPEG2024542332000004.jpg59165
[0013] According to the fluctuation of the fan inlet smoke temperature and static pressure, the fan inlet volume flow rate is corrected, and according to the predicted values of the fan status parameters in each typical operation mode, the fan total pressure Pt and specific pressure energy Y are calculated and obtained, and the fan operation prediction parameters are {(Q' v , P' t , Y') i}.
[0014] A further improvement of the present invention is to provide, in steps 5), the fan operation estimation point parameters {(Q' v , P' t , Y')i} (i=1, 2, 3, ... m), the fan performance curve is displayed, and the opening β corresponding to the fan operation point of each operation mode is calculated. i Obtain the intersection point of the isotropic line and the theoretical stall line {(Q v、s , P t、s ) i} and obtain it.
[0015] JPEG2024542332000005.jpg66156
[0016] JPEG2024542332000006.jpg39159
[0017] If the operating point parameters of each fan are predicted to not satisfy the above conditions, it is necessary to explain that the fan cannot meet the full load range condition even if the resistance of the exhaust gas system is increased, and to carry out fan stall prevention control. The increase value of the resistance of the exhaust gas system ΔP in and ΔP out The actual fan inlet / outlet static pressure difference ΔP is then lowered until the operating condition parameters that satisfy the conditions are obtained. e and by monitoring the fan opening β0, a fan anti-stall control strategy is formulated, i.e. ΔP e <(P e , out +ΔP out ) BMCR -(P e、in -ΔP in ) BMCR、 β0<0.8β max Effect
[0018] The present invention has at least the following beneficial technical effects: The present invention provides a method for real-time online prediction of the operation status of a fan based on data mining. By mining and utilizing the past monitoring data of the fan, a prediction model for each status parameter can be built, and the main operation status parameters of the fan in a typical operation mode can be obtained. In addition, by predicting the fluctuation value of the resistance of the exhaust gas system, the operation status parameters of the fan after the resistance of the exhaust gas system fluctuates can be accurately predicted. According to the design performance curve of the fan, the stall safety system, the stall pressure margin coefficient and the stall flow margin coefficient corresponding to the operation status parameters of the fan in multiple operation modes can be calculated. According to the stall margin coefficient and the opening degree of the fan, the operation safety of the fan can be evaluated, and whether the fan can operate safely and stably can be determined, providing a basis for the operation adjustment of the fan, improving the rationality of the stall prevention control strategy of the fan, and improving the operation safety and economy of the fan. [Brief description of the drawings]
[0019] [Figure 1] FIG. 1 is a schematic diagram illustrating the principle of the present invention. [Diagram 2] FIG. 2 is a schematic diagram showing the distribution of operating points and corresponding stall points of the fan of the present invention on the fan performance curve. [Diagram 3] FIG. 2 is a schematic diagram showing the distribution of actual operating points and estimated operating points of a fan on its performance curve in a typical operating mode. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0020] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are illustrated in the accompanying drawings, it should be understood that the present disclosure should not be limited by the embodiments described herein, but can be realized in various forms. On the contrary, these embodiments are provided to enable a more complete understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other if they do not conflict. Hereinafter, the present invention will be described in detail in combination with the embodiments with reference to the accompanying drawings.
[0021] The present invention provides a method for real-time online prediction of fan operation status based on data mining, which includes the steps of: 1) acquiring historical operation status parameters of the unit and the fan based on the fan online monitoring system and the DCS system; 2) processing the historical operation status parameters of the fan using data analysis technology to extract data in multiple normal and stable operation modes and constructing a sample set; and 3) training the extracted sample set based on an artificial neural network to obtain a prediction model of the fluctuation relationship of the fan inlet flow rate with the unit load or the main steam flow rate, and a prediction model of the system resistance from the furnace room to the induced draft fan inlet stage and the system resistance from the induced draft fan outlet to the chimney outlet stage with the fluctuation relationship of the flue gas volume. 4) given the predicted fluctuation values of the exhaust gas system state parameters, and obtaining the state operating parameters of the fan based on the two prediction models, 5) calculating the theoretical stall safety factor, pressure margin factor and flow margin factor of the fan based on the performance curve of the fan and the predicted values of the operating parameters of the fan, and 6) setting thresholds for each stall margin factor based on the past stall operating mode analysis and a large amount of stall test statistical analysis, and evaluating whether the operating conditions of the fan after the fluctuations of the exhaust gas system state parameters are safe by comparing the deviations between the theoretical stall safety factor, pressure margin factor and flow margin factor and the thresholds. A specific implementation method of the present invention is as follows:
[0022] 1. From the power plant fan online monitoring system and DCS system, the unit load L and boiler evaporation D b , fan inlet temperature T in , the fan inlet volume flow rate Q v , Fan inlet total pressure P t、in , total pressure at the fan outlet P t、out , fan opening degree β, fan inlet static pressure P e、in and fan outlet static pressure P e、out A data set of major monitoring parameters of recent boilers and fans, including parameters such as above, is extracted, the time period t is 10 to 30 days, and the time interval △t = 1 to 5 minutes. The data set should cover most of the normal operating load range.
[0023] 2. Organize past data, submit key monitoring parameters under multiple stable load modes, and establish a sample collection.
[0024] JPEG2024542332000007.jpg34159
[0025] JPEG2024542332000008.jpg21159
[0026] 3. Based on the sample points, the artificial neural network is trained to obtain the fan inlet volume flow rate Q v and the flow rate of main steam D b The variation relationship model Q v =f(D b ), fan inlet static pressure P e、in and the fan inlet volume flow rate Q v The change relationship model P e、in =f(Q v ), fan outlet static pressure P e、out and the fan inlet volume flow rate Q v The change relationship model P e、out =f(Q v ), fan inlet total pressure P t、in and the fan inlet volume flow rate Q v The change relationship model P t、in =f(Q v) and the total pressure at the outlet of the fan P t、out and the fan inlet volume flow rate Q v The change relationship model P t、out =f(Q v ) to get the
[0027] 4. Combine the actual operating conditions of the unit and the proposed modification solutions to predict the fluctuation of the exhaust gas system conditions, and calculate the fluctuation value of the front and rear resistance of the induced draft fan in the operating mode, i.e. (D' b , ΔP in ), (D' b , ΔP out ) to get the
[0028] JPEG2024542332000009.jpg155161
[0029] JPEG2024542332000010.jpg108159
[0030] JPEG2024542332000011.jpg46159
[0031] If the operating point parameters of each fan are predicted to not satisfy the above conditions, it is necessary to explain that the fan cannot meet the full load range condition even if the resistance of the exhaust gas system is increased, and to carry out fan stall prevention control. The increase value of the resistance of the exhaust gas system ΔP in and ΔP out The actual fan inlet / outlet static pressure difference ΔP is then lowered until the operating condition parameters that satisfy the conditions are obtained. e and by monitoring the fan opening β0, a fan anti-stall control strategy is formulated, i.e. ΔP e <(P e , out +ΔP out ) BMCR -(P e、in -ΔP in ) BMCR、 β0<0.8β max Meet the following.
[0032] 5. The method has the characteristics of high efficiency, high reliability and strong robustness, and can be applied to the anti-stall monitoring and prediction of the fan operating state of the induced draft fan of large coal-fired power units.
[0033] Working Example The induced draft fan of a 300MW unit in China is a dynamic vane adjustment type axial fan, and each environmental protection equipment in the exhaust gas system is clogged to different degrees due to the presence of ammonium hydrogen sulfate, which causes stalling in high load operation mode. Therefore, in order to ensure the safe and stable operation of the unit, the power plant adopts an adjustment method to limit the opening and current to limit the load capacity of the unit. In addition, after the unit is repaired and the congestion is cleared, there is no reliable basis for evaluation, so the output of the unit is limited by operating the unit with the established fan adjustment method. The method of the present invention is realized using an object-oriented programming language, and this function module is incorporated into the online monitoring and fault alarm system of the induced draft fan, so that the operating state of the fan can be estimated in real time according to the resistance fluctuation situation of the exhaust gas system, the safety of operation can be evaluated, and early warning can be given. Based on this method, a more reliable fan adjustment method is established, and a reliable basis for fan adjustment can be provided by comprehensively monitoring the resistance changes of the equipment that is mainly prone to congestion in the exhaust gas system, the static pressure difference between the inlet and outlet of the fan, the opening and current of the fan, etc., thereby improving the output of the fan and the load capacity of the unit. The algorithm of the present invention is efficient and reliable. Under typical operating modes, the distribution of actual fan operating points and predicted operating points on the performance curve is shown in Figure 3, and the condition assessment calculation results for each operating point are shown in Table 1.
[0034] JPEG2024542332000012.jpg243165JPEG2024542332000013.jpg246165JPEG2024542332000014.jpg21165
[0035] Although the present invention has been described in detail above using general descriptions and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made based on the present invention. Therefore, any modifications or improvements made without departing from the spirit of the present invention are within the scope of protection claimed by the present invention. [Explanation of symbols]
[0036] L is the power generation load of the unit, in kW D b The amount of steam generated by the boiler is expressed in t / h. T in Fan inlet smoke temperature, unit is ℃ Q v Fan inlet volumetric flow rate, unit is m 3 / s P t、in Fan inlet total pressure, unit is Pa P t、out Fan outlet total pressure, unit is Pa P e、in Fan inlet static pressure, unit is Pa P e、out Fan outlet total pressure, unit is Pa P t Fan total pressure, unit is Pa Y Fan specific pressure energy, unit is kJ / kg β DCS fan opening feedback value, unit is degrees l Data point order (l=1, 2, ..., n) j time interval (j=1, 2, ..., n) I Number of load modes (i=1, 2, ..., m) S standard variance k s Stall Safety Factor Kp Stall pressure margin coefficient kq Stall flow margin coefficient β min The minimum fan opening value, in degrees. β max The maximum fan opening value, in degrees. β BMCRThe opening value corresponding to the operating point of the fan in BMCR operation mode, in degrees.
Claims
1. A method for predicting a fan operation status online in real time based on data mining, comprising: 1) obtaining historical operating status parameters of the unit and the fan based on the fan online monitoring system and the DCS system; 2) using data analysis techniques to process the historical operating parameters of the fan, extract data under multiple normal and stable operating modes, and build a sample set; 3) training the extracted sample collection based on an artificial neural network to obtain a prediction model of the fluctuation relationship of the fan inlet flow rate with the unit load or the main steam flow rate, and a prediction model of the system resistance from the furnace room to the induced draft fan inlet stage and the system resistance from the induced draft fan outlet to the chimney outlet stage with the fluctuation relationship of the exhaust gas volume; 4) obtaining a fan status operating parameter based on two prediction models given a fluctuation prediction value of the exhaust gas system status parameter; 5) calculating a theoretical stall safety factor, a pressure margin factor, and a flow margin factor for the fan based on the fan performance curve and predicted values of the fan operating parameters; 6) A method for online, real-time prediction of the operating status of a fan based on data mining, comprising: a step of setting a threshold value for each stall margin coefficient based on past stall operating mode analysis and a large amount of stall test statistical analysis; and evaluating whether the operating status of the fan is safe after the fluctuation of the status parameters of the exhaust gas system is safe by comparing the deviation between the theoretical stall safety coefficient, the pressure margin coefficient and the flow margin coefficient and the threshold value.
2. In step 1), the past operating status parameters of the unit and the fan are: Unit load L, boiler evaporation D b , fan inlet temperature T in , fan inlet volume flow rate Q v , Fan inlet total pressure P t、in , total pressure at the fan outlet P t、out, fan opening degree β, fan inlet static pressure P e、in and the fan outlet static pressure P e、out The method for real-time online prediction of the operating status of a fan based on data mining according to claim 1, characterized in that: the method includes:
3.
4. In step 3), an artificial neural network is trained based on the sample points to obtain the fan inlet volume flow rate Q v and the flow rate of main steam D b The change relationship model Q v = f (D b ), fan inlet static pressure P e、in and the fan inlet volume flow rate Q v The fluctuation relationship model P e、in = f (Q v ), fan outlet static pressure P e、out and the fan inlet volume flow rate Q v The fluctuation relationship model P e、out = f (Q v ), total pressure at the fan inlet P t、in and the fan inlet volume flow rate Q v The fluctuation relationship model P t、in = f (Q v ) and the fan outlet total pressure P t、out and the fan inlet volume flow rate Q v The fluctuation relationship model P t、out = f (Q v 4. The method for predicting the operation status of a fan in real time online based on data mining according to claim 3, further comprising: acquiring the operation status of the fan based on the data mining.
5. In step 3), a fluctuation in the state of the exhaust gas system is predicted, and a fluctuation value corresponding to the front and rear resistance of the induced draft fan in the driving mode, i.e., (D' b , ΔP in ), (D' b , ΔP out 5. The method for predicting the operation status of a fan in real time online based on data mining according to claim 4, further comprising: acquiring the fan operation status information based on the fan operation status information. Claim 6 Claim 7 Claim 8
Citation Information
Patent Citations
Method for calibrating actual stall line of axial flow fan of power station in operation state
CN114776619A
