Variable frequency step speed regulation control method for a moving blade regulating axial induced draft fan

Abstract

The application discloses a variable-frequency and grading speed regulation control method of a moving blade regulation axial flow induced draft fan, and belongs to the technical field of energy-saving control of boiler auxiliary machines. The method comprises the following steps: in the initial operation stage, a full-working-condition self-scanning program based on multi-physical field coupling is executed, and an end-to-end system power consumption and safety characteristic correlation model is constructed; a resonance forbidden area is marked by identifying the coupling boundary of vibration energy and aerodynamic parameters, and the optimal combination is solved by combining the measured aerodynamic stall boundary and the regulation nonlinear area, superimposing in a multidimensional space and projecting to form a composite safety boundary; the safety screening is executed with the boundary as a hard constraint; and the optimal combination with the lowest total input power of the system considering the chain efficiency of the frequency converter and the motor in the full working condition is selected and an instruction is issued. The application also monitors the energy accumulation trend through online vibration spectrum analysis, dynamically updates the composite safety boundary, and realizes adaptive optimization control in the whole life cycle.

Description

Technical Field

[0001] This invention relates to the field of energy-saving control technology for boiler auxiliary equipment, and in particular to a variable frequency speed control method for a variable blade axial flow induced draft fan. Background Technology

[0002] Under deep peak-shaving conditions, the boiler induced draft fans of thermal power generating units operate at low loads for extended periods, leading to increasingly prominent issues of low fan efficiency and high plant power consumption. Taking a 350MW supercritical steam turbine generator unit as an example, statistics show that the total plant power consumption of the unit's forced draft fans, induced draft fans, and primary air fans is high, especially the induced draft fans, which account for the largest proportion. In 2023, the unit's annual load rate was only 50.3%, with the induced draft fans operating at low loads or deep peak-shaving conditions for most of the time, resulting in fan efficiency below 60%, indicating significant potential for efficiency improvement and energy saving. Adopting variable frequency speed control retrofitting is a feasible path to improve the operating efficiency of induced draft fans under low-load conditions. Its energy-saving principle lies in fixing the fan blades at a high-efficiency opening and using variable speed operation to ensure the induced draft fan always operates within its high-efficiency range.

[0003] However, when an adjustable-blade axial-flow induced draft fan operates using variable frequency speed control, the natural frequency of the shaft system may fall within a certain order or harmonic range of the inverter's output frequency, causing structural resonance and potentially leading to blade breakage in severe cases. How to ensure the safety of the induced draft fan itself (avoiding resonance) while simultaneously optimizing operating efficiency across all operating conditions is a pressing technical problem that needs to be solved in this field.

[0004] Currently, control methods for avoiding resonance zones in variable frequency wind turbines have been disclosed. For example, Chinese patent document CN102116305A discloses a control system and method for preventing resonance fracture of variable frequency wind turbine blades. After resonance is detected, a resonance alarm triggers an automatic frequency regulator and a damper regulator to make the wind turbine frequency jump out of the resonance zone. However, this method is a passive resonance avoidance method, only executing avoidance actions after resonance has occurred and been detected. It cannot eliminate resonance risks in advance during the operation decision-making stage, and this method does not involve the coordinated efficiency optimization of variable frequency speed regulation and blade opening.

[0005] In recent years, improved schemes in this field have also emerged that incorporate the resonance prohibition range into the frequency converter and blade joint control strategy. For example, Chinese patent document CN121738929A discloses a method, device, equipment, and medium for the joint control of frequency converter and blades of an induced draft fan. This method obtains the performance and structural parameters of the induced draft fan, simulates the fan based on these parameters, and generates an initial frequency converter-blade joint operation strategy. It then determines the resonance speed prohibition range of the induced draft fan, generates a final frequency converter-blade joint operation strategy based on the initial strategy and the resonance speed prohibition range, generates a speed-blade opening mapping table based on this strategy and the target induced draft fan's preset frequency converter-blade coupling operation zone, and performs frequency converter and blade joint control of the induced draft fan based on this mapping table and preset multi-modal adjustment rules. This scheme clearly addresses the technical problem of "optimizing operating efficiency across the entire operating range while ensuring the safety of the induced draft fan equipment (avoiding frequency converter harmonic resonance)," and achieves graded control by setting a resonance speed prohibition range and generating a deterministic speed-blade opening mapping table.

[0006] In addition, the journal article "Application Example of Efficiency Improvement Technology for Low-Load Peak Shaving Operation of Exhaust Fans" (Fan Technology, 2023, S1) takes a 300MW thermal power unit with adjustable blades in an axial flow induced draft fan as the research object. It adopts a variable frequency stepped speed regulation mode for the induced draft fan, and coordinates the fan speed and blade opening under different load conditions to achieve the effect of improving the efficiency of low-load peak shaving operation of the induced draft fan. The application effect shows that this method can avoid the resonance range.

[0007] However, the aforementioned existing technologies still have the following problems in practical engineering applications:

[0008] First, the construction of safety boundaries relies on theoretical simulation models rather than field measurement data, and cannot characterize multi-physics coupling effects. Taking the existing technology represented by CN121738929A as an example, its resonant speed prohibition range is generated by simulation based on acquired performance and structural parameters. However, through long-term field practice and data mining, it has been found that the vibration exceeding limits of induced draft fans is not a simple single-variable problem of speed, but is caused by the cross-coupling of multiple physical field parameters such as speed, blade opening, and airflow pressure pulsation within a specific range. The static boundary method in existing technologies, which relies on a single factory simulation curve or a single vibration test, cannot characterize this multi-dimensional coupling effect. Therefore, in actual engineering, existing technologies either sacrifice energy-saving space due to overly conservative boundaries or create safety hazards because the boundaries fail to capture coupling risks.

[0009] Meanwhile, after long-term service, the actual vibration and aerodynamic characteristics of induced draft fans will deviate significantly due to factors such as flue gas wear, ash accumulation, and changes in pipeline resistance characteristics, potentially leading to a mismatch between theoretical safety boundaries and actual conditions. Furthermore, for induced draft fans in actual engineering projects, due to differences in on-site installation conditions, pipeline characteristics, and combustion media, there is a significant deviation between the actual vibration and aerodynamic characteristics of each fan and its factory performance curve. Therefore, relying solely on the safety boundaries of the factory performance curve cannot accurately reflect the true situation under actual service conditions.

[0010] Second, the optimization objective is one-sided, failing to consider overall energy efficiency. Existing technologies typically aim for maximum efficiency solely in the wind turbine itself when optimizing efficiency. However, in actual variable frequency drive (VFD) operation, motor and inverter efficiencies vary significantly with load rate. Optimizing only the efficiency of a single wind turbine may not achieve the minimum total system input power, meaning it may not achieve the true lowest equivalent plant power consumption. In other words, maximum wind turbine efficiency does not equate to the lowest total system input power from the high-voltage inverter input to the wind turbine output. Existing technologies do not comprehensively incorporate the changes in high-voltage inverter and motor efficiency curves under different load rates into the optimization calculation, resulting in a one-sided optimization objective.

[0011] Third, the control strategy is rigid and lacks adaptability throughout the entire lifecycle. Existing technologies, such as CN121738929A, ultimately generate a deterministic speed-blade opening mapping table. This table, once generated, is fixed in the controller and cannot be automatically corrected for characteristic drift or new resonance risks that occur during long-term operation. Over the service life of an induced draft fan, which can last for several years or even longer, the natural frequency of the shaft system may shift due to wear, loosening, crack initiation, and other factors. The originally set safety boundaries need to be dynamically updated; otherwise, safety hazards exist. This one-time setting and non-updating technical solution lacks the ability to perceive and adaptively adjust to changes in the equipment's state throughout its entire lifecycle.

[0012] In summary, the three types of defects in existing technologies are interconnected systemic problems: static safety boundaries relying on simulation models cannot reflect the multi-physics coupling effects of real-world conditions, resulting in inaccuracies in safety constraints; focusing solely on the efficiency of the fan itself while neglecting the overall energy efficiency of the transmission chain leads to a one-sided optimization objective; and one-time fixed control strategies cannot adapt to long-term equipment state drift, resulting in a lack of timeliness in the control strategies. These three defects collectively prevent existing technologies from continuously achieving optimal overall system energy consumption under safety constraints throughout the entire lifecycle of the induced draft fan.

[0013] Based on the shortcomings of the existing technologies, there is an urgent need to provide a variable frequency speed control method for axial flow induced draft fans with adjustable blades. This method can construct multi-dimensional safety boundaries based on field measured data, optimize for minimizing the total system input power, and possess online self-learning capabilities throughout the entire lifecycle. This method ensures the safe operation of the induced draft fan under all operating conditions while achieving continuous optimization of the system's overall energy consumption. Summary of the Invention

[0014] To overcome the shortcomings in the prior art, the present invention discloses a variable frequency speed control method for a moving blade adjustable axial flow induced draft fan.

[0015] To achieve the above objectives, the present invention adopts the following technical solution:

[0016] A variable frequency speed control method for an axial flow induced draft fan with adjustable blades includes the following steps:

[0017] Step 1: In the initial stage of commissioning of the induced draft fan, execute the full-condition autonomous network scanning program based on multi-physics coupling to construct an end-to-end system power consumption and safety characteristic correlation model of the induced draft fan in the real service environment, and obtain the operating data of the induced draft fan at multiple discrete speed levels and multiple discrete blade openings. The operating data includes at least the real-time vibration intensity and real-time pressure pulsation amplitude as coupling variables, as well as the total system input power used for system efficiency modeling.

[0018] Step 2: Based on the coupled variable data obtained in Step 1, the coupling boundary between vibration energy and aerodynamic parameters is identified and determined in the speed-opening coordinate system to mark the resonance exclusion zone; based on the operating data obtained in Step 1, the measured aerodynamic stall boundary is determined; and the area where the blade opening is lower than a preset threshold is marked as the adjustment nonlinear zone; the resonance exclusion zone, the measured aerodynamic stall boundary, and the adjustment nonlinear zone are superimposed in multidimensional space and projected onto the speed-opening operating plane to form a composite safety boundary;

[0019] Step 3: Receive the target air volume and target air pressure from the distributed control system of the unit in real time, and generate a candidate matrix containing multiple speed range-blade opening combinations; perform safety screening using the composite safety boundary established in Step 2 as a hard constraint condition, eliminate speed range-blade opening combinations that fall into the composite safety boundary in the candidate matrix, and solve for the achievable combinations that can stably reach the target air volume and target air pressure under the end-to-end system power consumption and safety characteristic correlation model from the remaining speed range-blade opening combinations;

[0020] Step four: For each reachable combination selected in step three, the end-to-end system power consumption and safety characteristic correlation model is invoked to calculate the total system input power, taking into account the chain efficiency of the frequency converter and motor under all operating conditions. Through a global extreme value search with safety constraints, the reachable combination with the lowest total system input power is selected as the optimal speed gear-blade opening combination, and its corresponding target speed gear command is sent to the high-voltage frequency converter, and its corresponding target blade opening command is sent to the wind turbine blade actuator.

[0021] Furthermore, in step one, the autonomous scanning program based on multi-physics coupling under all operating conditions specifically involves: controlling the high-voltage frequency converter to gradually increase from the lower limit of the set frequency to the upper limit of the set frequency with a preset frequency step size; controlling the fan blade actuator to scan from the lower limit of the set opening to the upper limit of the set opening with a preset opening step size at each frequency and opening point; and after stable operation for a preset time under the combined operating conditions of each frequency and opening point, synchronously collecting and recording the vibration intensity, pressure pulsation amplitude, fan total pressure, fan flow rate, and total system input power at that operating point.

[0022] Furthermore, in step two, the construction of the composite safety boundary includes: marking the rotational speed regions corresponding to all operating points in the multi-physics field coupling-based full-condition autonomous sweeping program where the vibration intensity or pressure pulsation amplitude exceeds their respective preset safety thresholds as resonance exclusion zones;

[0023] Connect the lowest pressure points on the pressure-flow characteristic curve before entering the stall condition at each blade opening scanned by the multi-physics field coupling-based full-condition autonomous scanning program, and mark the area outside this line near the low-pressure side as the measured aerodynamic stall boundary.

[0024] The region corresponding to all operating points where the blade opening is below 25% is marked as the adjustment nonlinearity region;

[0025] The resonance exclusion zone, the measured aerodynamic stall boundary, and the adjustment nonlinear region are superimposed in a multidimensional space and projected onto the speed-opening operating plane. The area occupied by their combination is the composite safety boundary. The operating region outside the composite safety boundary is the safe solution space that allows for optimization.

[0026] Furthermore, in step three, the solution for a stable achievable combination of the target airflow and target air pressure specifically involves: for each of the remaining speed range-blade opening combinations that do not fall within the composite safety boundary, based on the operating data obtained by the multi-physics field coupling-based full-condition autonomous scanning program, determining whether, under the corresponding speed range of the speed range-blade opening combination, the pressure-flow characteristic curve can pass the target operating point determined by the target airflow and target air pressure within the allowable operating range by adjusting the blade opening; if so, the combination is determined to be an achievable combination.

[0027] Furthermore, in step four, the calculation of the total system input power, which incorporates the chain efficiency of the inverter and motor under all operating conditions, specifically involves: reading the measured value of the total system input power under the corresponding operating condition in the autonomous scanning program based on multi-physics coupling for the reachable combination; or, calculating the calculated value of the total system input power based on the speed range and blade opening of the reachable combination, as well as the built-in inverter efficiency curve and motor efficiency curve.

[0028] Furthermore, the variable frequency speed control method for the axial flow induced draft fan with adjustable blades also includes online self-learning and dynamic correction steps:

[0029] Step 5: During unit operation, when the preset triggering conditions are met, perform online vibration spectrum analysis on the induced draft fan to obtain the energy accumulation trend of the vibration characteristic frequency at the current operating speed;

[0030] Step 6: When the energy of a certain vibration characteristic frequency is detected to be monotonically increasing and approaching the preset safety threshold, the rotational speed region corresponding to the vibration characteristic frequency is automatically marked as a new resonance forbidden zone and updated to the currently effective composite safety boundary.

[0031] Step 7: When a new control cycle arrives, steps 3 and 4 are automatically invoked. Based on the updated composite safety boundary, the optimal speed gear-blade opening combination is re-solved through the global extreme value search under safety constraints. The command corresponding to the new optimal speed gear-blade opening combination is then sent to the high-voltage frequency converter and the wind turbine blade actuator.

[0032] Furthermore, in step five, the preset triggering conditions include the unit's cumulative operating time reaching a preset interval, or the unit's load remaining stable in a preset load range for a duration exceeding a set time.

[0033] Furthermore, in step six, marking the rotational speed region corresponding to a certain vibration characteristic frequency as a new resonance forbidden zone specifically includes: when the latest vibration amplitude of a certain vibration characteristic frequency reaches more than 80% of the preset safety threshold, and the moving average value of the vibration amplitude of the vibration characteristic frequency shows a continuous upward trend over multiple consecutive sampling periods, even if the vibration characteristic frequency has not yet exceeded the preset safety threshold, the rotational speed region corresponding to the vibration characteristic frequency is also marked as a new resonance forbidden zone.

[0034] Furthermore, in step one, the total system input power obtained refers to the sum of the input power of the motor driving the induced draft fan and the power loss of the frequency converter itself.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. This invention abandons the approach of relying on factory performance curves or simulation models to construct safety boundaries. In step one, a multi-physics field coupling-based autonomous network scanning program is executed during the initial commissioning of the induced draft fan to obtain real-time vibration intensity, real-time pressure pulsation amplitude, and total system input power at multiple discrete speed levels and multiple discrete blade openings. Based on this, the composite safety boundary established in step two consists of three parts: a resonance exclusion zone, a measured aerodynamic stall boundary, and a regulation nonlinearity zone, which are superimposed in a multi-dimensional space and projected onto the speed-opening operation plane. Among them, the resonance exclusion zone is determined by identifying the coupling boundary between vibration energy and aerodynamic parameters, rather than a simple single-variable threshold judgment, and can capture the vibration over-limit risk caused by the cross-coupling of multiple physical field parameters such as speed, blade opening, and airflow pressure pulsation; the measured aerodynamic stall boundary is determined based on the real pressure-flow characteristics obtained by the multi-physics field coupling-based autonomous network scanning program, accurately reflecting the stall boundary of the induced draft fan under specific pipeline characteristics; and the regulation nonlinearity zone excludes the region of poor regulation linearity at small blade openings. This modeling method based on measured data can accurately reflect the actual vibration and aerodynamic characteristics of the induced draft fan under specific installation environments, pipeline characteristics, and coal combustion conditions, effectively avoiding the mismatch between theoretical safety boundaries and actual service characteristics in existing technologies. In particular, by incorporating the measured aerodynamic stall boundary and the regulation nonlinearity region into the safety constraints, compared to existing technologies that rely solely on a single resonance prohibition interval, it provides a more comprehensive multi-dimensional safety guarantee. Under the unit's low-load, deep peak-shaving conditions, it ensures that the fan blades always operate within a safe region with good regulation linearity, contributing to stable boiler combustion.

[0037] 2. This invention overcomes the one-sidedness of existing technologies that only aim for the highest efficiency of the wind turbine itself. In the operational data acquired in step one, the total system input power is explicitly included in the data collection scope. This total system input power is the sum of the input power of the motor driving the induced draft fan and the power loss of the high-voltage frequency converter itself. In step four, with the lowest total system input power as the optimization objective, the optimal combination is selected from all safe and achievable speed range-blade opening combinations through a global extreme value search under safety constraints. Because the efficiency of both the motor and the high-voltage frequency converter changes under different load rates, the highest efficiency of a single wind turbine is not equivalent to the lowest total system input power. This invention comprehensively incorporates the changes in the efficiency curves of the high-voltage frequency converter and the motor under different load rates into the optimization calculation. It calls the end-to-end system power consumption and safety characteristic correlation model to calculate the total system input power, including the chain efficiency of the frequency converter and motor under all operating conditions. This ensures that the finally determined speed range-blade opening combination is the true point of lowest overall system energy consumption, rather than just the point of highest wind turbine efficiency.

[0038] 3. Compared with existing technologies that generate a fixed speed-opening mapping table and whose control strategy is not updated after being set once, this invention introduces an online self-learning and dynamic correction closed-loop mechanism in steps five to seven. In step five, when the preset triggering conditions are met, online vibration spectrum analysis is performed on the induced draft fan to obtain the energy accumulation trend of the vibration characteristic frequency at the current operating speed. In step six, when the energy of a certain vibration characteristic frequency is detected to be monotonically increasing and approaching a preset safety threshold, even if it has not yet exceeded the limit, the speed region corresponding to that vibration characteristic frequency is proactively marked as a new resonance forbidden zone, and the composite safety boundary is updated. Specifically, when the latest vibration amplitude of a certain vibration characteristic frequency reaches more than 80% of the preset safety threshold and its moving average value over multiple consecutive sampling periods shows a continuous upward trend, it is determined that the energy of that characteristic frequency is monotonically increasing and approaching the preset safety threshold, and the potential resonance risk region is included in the composite safety boundary in advance. In step seven, in the new control cycle, the optimal combination is re-solved through a global extreme value search under safety constraints based on the updated composite safety boundary. This closed-loop feedback mechanism enables the control system to sense the long-term characteristic drift and potential fatigue risks of the equipment, and can proactively adjust the control strategy before the risk increases significantly, thus achieving true full life cycle adaptive optimization control. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the primary electrical wiring of the high-voltage frequency converter in an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram showing the position of the typical operating point of the induced draft fan on the variable speed curve of the moving blade opening +4° line;

[0041] Figure 3 This is a schematic diagram showing the position of the typical operating point of the induced draft fan on the variable speed curve of the moving blade opening line of -4°.

[0042] Figure 4 This is a schematic diagram showing the position of the typical operating point of the induced draft fan on the variable speed curve of the moving blade opening line -8°. Detailed Implementation

[0043] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Please refer to the instruction manual appendix. Figures 1 to 4 The present invention provides the following technical solutions:

[0045] Example 1: This example uses the HU27050 dual-stage adjustable axial flow induced draft fan configured in Unit 1 of Company A as the implementation object. The fan is designed to run at 745 r / min, and the matching motor is model YKK900-6 with a rated power of 5600 kW and a rated voltage of 6 kV. An 8500 kVA high-voltage frequency converter is added to the front end of the motor, using a one-to-one automatic bypass wiring configuration. The frequency converter is a harmonic-free, series-connected multi-level type with a built-in phase-shifting transformer, and its efficiency is no less than 97%. The bypass circuit consists of three high-voltage vacuum circuit breakers QF1 to QF3 and two high-voltage disconnect switches QS1 and QS2, ensuring that the motor can be switched to power frequency bypass operation during frequency converter maintenance.

[0046] The system hardware configuration in this embodiment is as follows:

[0047] Acceleration vibration sensors with a frequency response range of not less than 0.5Hz to 500Hz are installed on the front and rear bearing seats of the induced draft fan to collect vibration intensity.

[0048] A dynamic pressure sensor is installed on the outlet pipe of the induced draft fan to collect the pressure pulsation amplitude.

[0049] A power measurement device is installed on the 6kV power input side of the high-voltage frequency converter to collect the total input power of the system. The total input power of the system is the sum of the input power of the motor driving the induced draft fan and the power loss of the frequency converter itself.

[0050] A speed sensor is installed on the coupling between the electric motor and the induced draft fan to accurately measure the actual speed of the fan.

[0051] The signals from the aforementioned sensors are connected to a newly installed optimization control station via a 4-20mA analog signal channel or an RS485 digital communication channel. The optimization control station is connected to the unit's distributed control system (DCS), the high-voltage frequency converter main controller, and the vibration data acquisition system via hardwiring and communication protocols.

[0052] The following is combined Figure 1 The electrical wiring diagram shown illustrates the variable frequency speed control method for the axial flow induced draft fan with adjustable blades in this embodiment. The method includes the following steps:

[0053] Step 1: Execute the autonomous network scanning program based on multi-physics coupling under all operating conditions to build an end-to-end system power consumption and security characteristic correlation model.

[0054] During the initial startup phase after the first commissioning or planned shutdown of Unit 1's induced draft fan, the optimization control station automatically triggers a full-condition autonomous network scan program based on multi-physics coupling. The purpose of this program is not simply to collect data, but to treat the induced draft fan as a multi-input multi-output system. Through a comprehensive scan of control variables (speed range and blade opening), it simultaneously acquires the mechanical response (vibration intensity), aerodynamic response (pressure pulsation amplitude), total fan pressure, and fan flow rate, as well as the electrical response (total system input power). This allows for the establishment of an end-to-end system power consumption and safety characteristic correlation model that accurately maps the triple input-output relationships of control variables, energy efficiency variables, and safety variables within the actual pipeline network. This model serves as the data-driven kernel for subsequent safety optimization.

[0055] Specifically, the high-voltage frequency converter starts from the frequency corresponding to the lower speed limit (set to 25Hz in this embodiment) and gradually increases the frequency step by step to the frequency corresponding to the upper speed limit (set to 50Hz in this embodiment) at a preset frequency step size of 1Hz. After running stably for 30 seconds at each frequency point, it enters the opening degree scanning subroutine at that frequency. In the opening degree scanning subroutine, the wind turbine blade actuator controls the blade opening degree to scan from the set lower opening degree limit of 10% to the set upper opening degree limit of 95% at a preset opening degree step size of 5%. After running stably for 10 seconds under the combined operating conditions of each frequency point and opening degree point, the optimization control station synchronously collects and records the following operating data for that operating condition point through the aforementioned sensors:

[0056] The actual speed of the fan (directly measured by the speed sensor and converted into the corresponding speed range);

[0057] Leaf opening;

[0058] Fan inlet volumetric flow rate (m³ / s);

[0059] Fan total pressure (Pa);

[0060] Fan inlet temperature (°C);

[0061] Vibration intensity of front and rear bearing housings (mm / s);

[0062] Fan outlet pressure pulsation amplitude (kPa);

[0063] The total system input power (kW) is the measured value of the power measurement device on the 6kV power input side of the frequency converter, which is the sum of the input power of the motor driving the induced draft fan and the power loss of the high-voltage frequency converter itself.

[0064] Using the measured speed range as the first dimension and the blade opening as the second dimension, the above-mentioned operating data of all operating points are stored in a two-dimensional data matrix, which forms the data basis for the end-to-end system power consumption and safety characteristics correlation model of the induced draft fan in the real service environment.

[0065] Step 2: Construct a composite security boundary based on the coupled variable data.

[0066] The purpose of this step is to determine the operating region that must not be entered during subsequent online operation, i.e., the composite safety boundary, based on the coupling variable data obtained in step one. The composite safety boundary is composed of three parts: the resonance exclusion zone, the measured aerodynamic stall boundary, and the regulation nonlinearity region, which are superimposed in multidimensional space and projected onto the speed-opening operating plane.

[0067] S2.1 Marked Resonance Forbidden Zone

[0068] In the two-dimensional data matrix obtained in step one, all operating points are traversed. Since the vibration exceeding limits of the induced draft fan is not a simple single-variable problem of rotational speed, but rather caused by the cross-coupling of multiple physical field parameters such as rotational speed, blade opening, and airflow pressure pulsation within a specific range, this step does not rely solely on the threshold of a single variable. Instead, it determines the boundary of the resonance exclusion zone by identifying the coupling relationship between vibration energy and aerodynamic parameters.

[0069] Specifically, when the vibration intensity at a certain operating point exceeds the preset safety threshold (4.6 mm / s in this embodiment) or the pressure pulsation amplitude exceeds the preset safety threshold (2.0 kPa in this embodiment), the speed gear corresponding to that operating point and the speed range within ±1 Hz adjacent to it are marked as resonance forbidden zones in the speed-opening coordinate system.

[0070] It should be noted that in the multi-physics coupling-based full-condition autonomous scanning program, one speed gear corresponds to multiple different blade opening operating points. If any operating point at any blade opening under that speed gear has a vibration intensity or pressure pulsation amplitude exceeding its respective preset safety threshold, then that entire speed gear and its adjacent speed region are marked as resonance forbidden zones.

[0071] S2.2 Determine the measured aerodynamic stall boundary

[0072] For each blade opening (e.g., 10%, 15%, 20%, ..., 95%) scanned by the multi-physics field coupling-based full-condition autonomous scanning program in step one, the total fan pressure and fan flow rate recorded at all speed ranges under that opening are used as data sources, and the pressure-flow characteristic curve under that blade opening is plotted in the pressure-flow coordinate system.

[0073] On each pressure-flow characteristic curve, by identifying the inflection point where the pressure begins to drop sharply as the flow rate decreases, the stable operating point with the lowest air pressure before entering the stall condition at that blade opening is determined. Connecting these lowest pressure points determined at all blade openings sequentially forms a continuous curve, which is the measured aerodynamic stall boundary. The region of this curve closer to the low-pressure side (i.e., the region far from the normal stable operating region) is marked as the stall zone where operation is prohibited.

[0074] S2.3 Marker Adjustment Nonlinear Region

[0075] The region occupied by all operating points with blade opening less than 25% of the preset threshold in the speed-opening coordinate system in the multi-physics field coupling-based full-condition autonomous scanning program is marked as the adjustment nonlinear region.

[0076] S2.4 constitutes a composite safety boundary.

[0077] The resonance exclusion zone, the measured aerodynamic stall boundary, and the regulation nonlinearity region are superimposed in a multidimensional space and projected onto the speed-aperture operating plane. The region occupied by their combined set is the composite safety boundary. The remaining operating region outside the composite safety boundary is the safe solution space that allows for subsequent optimization.

[0078] Step 3: Receive the target operating conditions and filter the achievable combinations.

[0079] During normal operation of the unit, the distributed control system of the unit sends the target air volume and target air pressure to the optimization control station in real time.

[0080] The optimization control station first generates a candidate matrix. The candidate matrix consists of multiple combinations of speed ranges and blade opening degrees. The speed ranges are determined by the discrete frequencies output by the inverter (26 ranges in this embodiment, corresponding to 25Hz to 50Hz, with a step size of 1Hz), and the blade opening degree is determined by the adjustment range of the wind turbine actuator (18 opening values ​​in this embodiment, corresponding to 10% to 95%, with a step size of 5%). Theoretically, the candidate matrix contains 26 × 18 = 468 combinations of speed ranges and blade opening degrees.

[0081] Next, safety screening is performed using the composite safety boundary established in step two as a hard constraint condition. The first round of screening is carried out on the candidate matrix: all speed gear-blade opening combinations that fall within the range of the composite safety boundary in the candidate matrix are eliminated.

[0082] Then, for the remaining speed range-blade opening combinations that do not fall within the composite safety boundary, the reachability is solved one by one under the end-to-end system power consumption and safety characteristic correlation model.

[0083] The specific method for determining reachability is as follows: Based on the operational data obtained by the autonomous scanning program under all operating conditions based on multi-physics coupling in step one, determine whether the pressure-flow characteristic curve can accurately pass through the target operating point determined by the target air volume and target air pressure within the allowable operating range at the speed range corresponding to the speed range and blade opening combination. If it can pass, the combination is determined to be a reachable combination and retained; otherwise, it is discarded.

[0084] Step 4: Select the optimal combination and issue the command.

[0085] For each reachable combination selected in step three, the end-to-end system power consumption and safety characteristic correlation model is invoked to calculate the total system input power, taking into account the chain efficiency of the frequency converter and motor under all operating conditions.

[0086] In this embodiment, if a reachable combination has the same or similar operating point in the database of the autonomous scanning program based on multi-physics coupling, the measured value of the total system input power recorded in step one is directly read as the power value of that reachable combination. For reachable combinations that do not have a corresponding operating point in the scanning database, the calculated value of the total system input power is obtained by calling the high-voltage inverter efficiency curve and motor efficiency curve built into the optimization control station, based on the speed gear and blade opening of the reachable combination, and combining the fan air power data obtained in step one.

[0087] By performing a global extreme value search under safety constraints, the total system input power of all reachable combinations is compared, and the reachable combination with the lowest total system input power is selected as the optimal speed gear-blade opening combination in the current control cycle.

[0088] The optimization control station sends the target speed command corresponding to the optimal speed gear-blade opening combination to the high-voltage frequency converter main controller via a communication protocol, so that the main controller adjusts the output frequency to the frequency value corresponding to the target speed gear. At the same time, the target blade opening command corresponding to the optimal speed gear-blade opening combination is sent to the unit's distributed control system via hard wiring or a communication protocol. The distributed control system then controls the wind turbine blade actuator to adjust the blade opening to the target value.

[0089] This completes one control cycle. When the unit load changes or the distributed control system issues new target air volume and target air pressure, steps three and four are repeated.

[0090] Example 2: Based on Example 1, the method in this example also includes online self-learning and dynamic correction functions during unit operation to address the characteristic drift and new resonance risks that may occur after the induced draft fan has been in service for a long time.

[0091] Step 5: Perform online vibration spectrum analysis when the triggering conditions are met.

[0092] During unit operation, the optimization control station continuously determines whether preset trigger conditions are met. This embodiment sets two trigger conditions, and online spectrum analysis is triggered when either one is met: Condition 1 is that the unit's cumulative operating time reaches a preset interval of 720 hours; Condition 2 is that the unit load operates stably at 50% of the rated load range (i.e., around 175MW) for more than 2 hours.

[0093] Once the triggering conditions are met, the optimized control station, at the current operating speed, collects vibration signals at a sampling frequency of no less than 1 kHz using acceleration vibration sensors installed on the front and rear bearing housings of the fan. The vibration spectrum is then obtained by performing a Fast Fourier Transform on the signal. The vibration amplitude at the rotational frequency and its harmonics is extracted, and the energy accumulation trend of each vibration characteristic frequency within one monitoring period is calculated. The energy accumulation trend refers to the direction of change of the moving average value of the vibration amplitude at a certain characteristic frequency over multiple consecutive sampling periods.

[0094] Step 6: Dynamically update the composite security boundary.

[0095] When the optimization control station detects that the energy of a certain vibration characteristic frequency (in this embodiment, for example, a harmonic of the motor's rotational frequency) shows a monotonically increasing trend, and the latest vibration amplitude of this characteristic frequency is approaching the preset safety threshold set in step two (for example, reaching more than 80% of the preset safety threshold), even if the characteristic frequency has not yet exceeded the safety threshold, the optimization control station automatically marks the rotational speed region corresponding to this vibration characteristic frequency as a new resonance exclusion zone and updates this new resonance exclusion zone to the composite safety boundary based on the current control cycle. The updated composite safety boundary is larger than the previous one, including newly discovered potential resonance risk areas.

[0096] Step 7: Re-optimize based on the updated composite security boundary.

[0097] When a new control cycle arrives, the optimization control station automatically invokes the control logic of steps three and four. Since the composite safety boundary has been updated in step six, step three, when screening candidate matrices, will use the updated composite safety boundary as a hard constraint, automatically excluding speed range-blade opening combinations that fall within the newly added resonance exclusion zone. Finally, the optimization control station, through a global extremum search under safety constraints, solves for the new optimal speed range-blade opening combination that minimizes the total system input power under the updated safety constraints, and sends the corresponding commands to the high-voltage frequency converter and the wind turbine actuator.

[0098] Through the closed-loop feedback mechanism in steps five to seven above, the control method of this embodiment can continuously sense the subtle changes in the equipment status throughout the entire service life of the induced draft fan, and proactively adjust the control strategy before the risk increases significantly, thus achieving true full life cycle adaptive optimization control.

[0099] It is worth noting that, using the control method provided in this embodiment, after the HU27050 induced draft fan of Company A's Unit 1 was retrofitted with variable frequency speed control, the total system input power of the induced draft fan was significantly reduced under three typical load conditions: 350MW, 263MW, and 175MW, compared to the original fixed-speed operation. According to the unit's annual operating data for 2023, the weighted average reduction rate of plant power consumption for the induced draft fan was approximately 0.249%, resulting in an annual power saving of approximately 3,219,978.5 kWh per unit, achieving excellent energy-saving results.

[0100] Meanwhile, during the long-term operational assessment within two years of unit commissioning, no safety accidents occurred due to variable frequency speed regulation, such as fan blade resonance, abnormal bearing vibration, or operational stall. Furthermore, no deterioration in furnace negative pressure regulation quality occurred due to the operating point falling into the nonlinearity region of regulation, verifying the multi-dimensional safety assurance effect of this control method. After one year of operation, the online vibration spectrum analysis function successfully predicted a potential resonance frequency drift trend. The optimized control station automatically updated the composite safety boundary, incorporating the new resonance exclusion zone into the optimization constraints, thus preventing the operating point from gradually approaching this risk area. This verified the online self-learning and dynamic correction capabilities of this control method.

[0101] Furthermore, using the control method provided in this embodiment, performance tests were conducted on Unit 1 of Company A under three typical load conditions: 350MW, 263MW, and 175MW, and compared with the original fixed-speed operation mode at the power frequency. The total system input power was defined as the sum of the input power of the motor driving the induced draft fan and the power loss of the frequency converter itself. The comparison results are shown in the table below:

[0102]

[0103] As shown in the table above, the lower the unit load rate, the more significant the energy-saving effect is achieved while ensuring safety. According to the unit's annual operating data in 2023, the weighted average reduction rate of electricity consumption by the induced draft fan plant was approximately 0.249%, the annual electricity saving of a single unit was approximately 3,219,978.5 kWh, and the annual electricity cost saving of two units was approximately 2.704 million yuan.

[0104] The overall system efficiency is defined as the ratio of the fan's air power to the total system input power, i.e., overall system efficiency = (fan air power / total system input power) × 100%. The percentage point improvement in overall system efficiency is the difference between the overall system efficiency after frequency conversion and the overall system efficiency during power frequency operation.

[0105] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention; therefore, the above embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and thus all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention.

Claims

1. A variable frequency speed control method for an axial flow induced draft fan with adjustable blades, characterized in that, Includes the following steps: Step 1: In the initial stage of commissioning of the induced draft fan, execute the full-condition autonomous network scanning program based on multi-physics coupling to construct an end-to-end system power consumption and safety characteristic correlation model of the induced draft fan in the real service environment, and obtain the operating data of the induced draft fan at multiple discrete speed levels and multiple discrete blade openings. The operating data includes at least the real-time vibration intensity and real-time pressure pulsation amplitude as coupling variables, as well as the total system input power used for system efficiency modeling. Step 2: Based on the coupled variable data obtained in Step 1, the coupling boundary between vibration energy and aerodynamic parameters is identified and determined in the speed-opening coordinate system to mark the resonance exclusion zone; based on the operating data obtained in Step 1, the measured aerodynamic stall boundary is determined; and the area where the blade opening is lower than a preset threshold is marked as the adjustment nonlinear zone; the resonance exclusion zone, the measured aerodynamic stall boundary, and the adjustment nonlinear zone are superimposed in multidimensional space and projected onto the speed-opening operating plane to form a composite safety boundary; Step 3: Receive the target air volume and target air pressure from the distributed control system of the unit in real time, and generate a candidate matrix containing multiple speed range-blade opening combinations; perform safety screening using the composite safety boundary established in Step 2 as a hard constraint condition, eliminate speed range-blade opening combinations that fall into the composite safety boundary in the candidate matrix, and solve for the achievable combinations that can stably reach the target air volume and target air pressure under the end-to-end system power consumption and safety characteristic correlation model from the remaining speed range-blade opening combinations; Step four: For each reachable combination selected in step three, the end-to-end system power consumption and safety characteristic correlation model is invoked to calculate the total system input power, taking into account the chain efficiency of the frequency converter and motor under all operating conditions. Through a global extreme value search with safety constraints, the reachable combination with the lowest total system input power is selected as the optimal speed gear-blade opening combination, and its corresponding target speed gear command is sent to the high-voltage frequency converter, and its corresponding target blade opening command is sent to the wind turbine blade actuator.

2. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 1, characterized in that, In step one, the autonomous network scanning program based on multi-physics coupling under all working conditions specifically includes: The high-voltage frequency converter is controlled to gradually increase from the lower limit of the set frequency to the upper limit of the set frequency in a preset frequency step size; at each frequency point, the fan blade actuator is controlled to scan from the lower limit of the set opening to the upper limit of the set opening in a preset opening step size. After a preset time of stable operation under the combined operating conditions of each frequency point and opening point, the vibration intensity, pressure pulsation amplitude, fan total pressure, fan flow rate, and total system input power at that operating point are collected and recorded synchronously.

3. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 2, characterized in that, In step two, the construction of the composite security boundary includes: In the multi-physics field coupling-based full-condition autonomous scanning program, the rotational speed regions corresponding to all operating points where the vibration intensity or pressure pulsation amplitude exceeds their respective preset safety thresholds are marked as resonance forbidden zones. Connect the lowest pressure points on the pressure-flow characteristic curve before entering the stall condition at each blade opening scanned by the multi-physics field coupling-based full-condition autonomous scanning program, and mark the area outside this line near the low-pressure side as the measured aerodynamic stall boundary. The region corresponding to all operating points where the blade opening is below 25% is marked as the adjustment nonlinearity region; The resonance exclusion zone, the measured aerodynamic stall boundary, and the adjustment nonlinear region are superimposed in a multidimensional space and projected onto the speed-opening operating plane. The area occupied by their combination is the composite safety boundary. The operating region outside the composite safety boundary is the safe solution space that allows for optimization.

4. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 1, characterized in that, In step three, the solution that stably achieves the attainable combination of the target air volume and the target air pressure specifically refers to: For each of the remaining speed range-blade opening combinations that do not fall within the composite safety boundary, based on the operating data obtained by the full-condition autonomous scanning program based on multi-physics coupling, it is determined whether the pressure-flow characteristic curve can pass the target operating point determined by the target air volume and target air pressure within the allowable operating range under the speed range corresponding to the speed range of the speed range-blade opening combination; if so, the combination is determined to be a reachable combination.

5. The variable frequency speed control method for an axial flow induced draft fan with adjustable blades according to claim 1, characterized in that, In step four, the calculation of the total system input power, which includes the chain efficiency of the frequency converter and motor under all operating conditions, specifically involves reading the measured value of the total system input power under the corresponding operating condition in the autonomous scanning program based on multi-physics coupling for the reachable combination.

6. The variable frequency speed control method for an axial flow induced draft fan with adjustable blades according to claim 1, characterized in that, In step four, the calculation of the total system input power, which incorporates the chain efficiency of the frequency converter and motor under all operating conditions, is specifically calculated based on the speed range and blade opening corresponding to the achievable combination, as well as the built-in frequency converter efficiency curve and motor efficiency curve, to obtain the calculated value of the total system input power.

7. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 3, characterized in that, It also includes the following steps: Step 5: During unit operation, when the preset triggering conditions are met, perform online vibration spectrum analysis on the induced draft fan to obtain the energy accumulation trend of the vibration characteristic frequency at the current operating speed; Step 6: When the energy of a certain vibration characteristic frequency is detected to be monotonically increasing and approaching the preset safety threshold, the rotational speed region corresponding to the vibration characteristic frequency is automatically marked as a new resonance forbidden zone and updated to the currently effective composite safety boundary. Step 7: When a new control cycle arrives, steps 3 and 4 are automatically invoked. Based on the updated composite safety boundary, the optimal speed gear-blade opening combination is re-solved through the global extreme value search under safety constraints. The command corresponding to the new optimal speed gear-blade opening combination is then sent to the high-voltage frequency converter and the wind turbine blade actuator.

8. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 7, characterized in that, In step five, the preset triggering conditions include the unit's cumulative operating time reaching a preset interval, or the unit's load remaining stable in a preset load range for a duration exceeding a set time.

9. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to claim 7, characterized in that, In step six, the rotational speed region corresponding to a certain vibration characteristic frequency is marked as a new resonance forbidden zone, specifically including: When the latest vibration amplitude of a certain vibration characteristic frequency reaches more than 80% of the preset safety threshold, and the moving average value of the vibration amplitude of the vibration characteristic frequency shows a continuous upward trend over multiple consecutive sampling periods, the rotational speed region corresponding to the vibration characteristic frequency will be marked as a new resonance forbidden zone, even if the vibration characteristic frequency has not exceeded the preset safety threshold.

10. The variable frequency speed control method for the axial flow induced draft fan with adjustable blades according to any one of claims 1-9, characterized in that, In step one, the total system input power obtained refers to the sum of the input power of the motor driving the induced draft fan and the power loss of the frequency converter itself.

Images