Cyclone separator monitoring system and monitoring method
By installing a weighing unit and a data processing system in the cyclone separator, the gas-solid separation parameters can be monitored in real time, solving the problems of accuracy and real-time performance in cyclone separator evaluation methods, and achieving efficient online evaluation and low-cost operation optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DONGFANG BOILER GROUP OF DONGFANG ELECTRIC CORP
- Filing Date
- 2026-02-03
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies lack the ability to directly, accurately, in real-time, and reliably measure the gas-solid separation efficiency of cyclone separators online, resulting in limitations in the accuracy, real-time performance, and industrial applicability of the evaluation methods.
A cyclone separator monitoring system is adopted. By installing weighing units in the return feeder and dust collector, combined with temperature sensors and differential pressure transmitters, the gas-solid separation parameters are acquired in real time. The data is then processed by the data acquisition and processing unit to establish a material quantity change model and realize online evaluation of separation efficiency.
It enables real-time and accurate evaluation of cyclone separator efficiency, reduces interference from measurement actions on the system, improves the timeliness and accuracy of evaluation, and reduces maintenance costs.
Smart Images
Figure CN122016359A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of circulating fluidized bed technology, specifically to a system and method for real-time monitoring and evaluation of cyclone separators. Background Technology
[0002] In industrial plants such as circulating fluidized bed (CFB) boilers and fuel cell (FCC) catalytic cracking, cyclone separators are considered the "heart" of the system, playing a crucial role in separating and returning solid particles from high-speed gas streams. Their gas-solid separation efficiency directly determines the overall system's operating efficiency, economy, and safety. In FCC processes, FCC catalysts are high-cost consumable materials. Even a slight decrease in cyclone separator efficiency (e.g., from 99.9% to 99.5%) can result in several tons or more of fine catalyst particles escaping into the flue daily and ultimately being released into the atmosphere, causing significant material loss and environmental pollution. The core mechanism for stable operation of a CFB boiler is material circulation. The efficiency of the cyclone separator directly affects the quantity and quality of the circulating material. A decrease in separation efficiency directly leads to a significant increase in the carbon content of fly ash, increased mechanical incomplete combustion losses in the boiler, thereby reducing the overall thermal efficiency of the boiler and increasing fuel costs.
[0003] Existing technologies for evaluating the performance of cyclone separators generally employ the following methods: 1. Isokinetic sampling method (offline benchmark method): Although considered a highly accurate measurement method, it relies on manual operation and requires sampling in high-temperature, high-dust environments. This process is time-consuming and labor-intensive, and continuous online operation is not feasible. Significant time delays make it unsuitable for real-time process control. 2. Pressure drop detection method: This method indirectly assesses the operating status by monitoring the pressure difference between the separator's inlet and outlet. However, pressure drop is influenced by multiple factors, including gas flow rate and particle concentration, and its relationship with separation efficiency is not singular. When efficiency declines due to internal component wear, the pressure drop change is often not significant enough to provide effective early warning. 3. Optical measurement method for outlet concentration: This method typically involves installing optical sensors such as turbidimeters on the outlet pipe. It only obtains the outlet particle concentration and cannot directly determine separation efficiency without combining it with inlet concentration data. Furthermore, the sensor mirrors are easily contaminated, requiring high maintenance, and the measurement results are significantly affected by the physical properties of the particles. The above evaluation methods have limitations in terms of accuracy, real-time performance, and industrial applicability.
[0004] In summary, existing technologies lack a method for directly, accurately, in real-time, and reliably measuring the amount of material collected by a cyclone separator online and calculating the cyclone separator efficiency accordingly. Summary of the Invention
[0005] To address the limitations of existing cyclone separator evaluation methods in terms of accuracy, real-time performance, and industrial applicability, this invention proposes a cyclone separator monitoring system and method.
[0006] A cyclone separator monitoring system includes a cyclone separator. The inlet of the cyclone separator is connected to the outlet of a circulating fluidized bed. The outlet of the cyclone separator is connected to a return feeder, and the flue gas outlet is connected to a dust collector. Material enters the cyclone separator from the circulating fluidized bed for gas-solid separation. The gas is transported to the dust collector through the flue gas outlet, and the solids enter the return feeder through the outlet. A first weighing unit is installed at the bottom of the return feeder. The first weighing unit includes a structure that can close or restore the fluidizing air of the return feeder. After the fluidizing air outlet of the return feeder is closed, the weight parameters of the material in the return feeder are collected at a certain frequency. A second weighing unit is installed at the bottom of the dust collector to collect the weight parameters of the solid particles escaping from the dust collector. A monitoring unit is set at the front end of the cyclone separator inlet. The monitoring unit includes a temperature sensor, a differential pressure transmitter, and a flow rate measuring device. A data acquisition and processing unit is also provided, which is connected to the first weighing unit, the second weighing unit, and the monitoring unit to receive and process the collected data in real time.
[0007] Furthermore, the first weighing unit includes a weighing sensor and a quick-closing valve. The weighing sensor is connected to the data acquisition and processing unit, and the quick-closing valve is used to shut off or restore the fluidizing air of the return feeder.
[0008] Furthermore, the second weighing unit employs a device capable of weighing solid particles.
[0009] Furthermore, the cyclone separator is flexibly connected to the return feeder and the dust collector, and the return feeder 2, the dust collector 4 and the rear-end components are flexibly connected, so that the components of the overall structure are not in rigid contact, avoiding errors in the weighing of the first weighing unit and the second weighing unit due to interference from other structures.
[0010] A cyclone separator monitoring method, employing any one of the above-mentioned technical solutions for cyclone separator monitoring systems, includes the following steps: S1: While ensuring the entire circulating fluidized bed boiler is in a stable state, acquire the monitoring unit parameters in real time: real-time temperature T, differential pressure ΔP and gas flow rate v; S2: Set a cycle time t0. Every time t0 elapses, the fluidizing air outlet of the return feeder needs to be closed within a first set time Δt1. During this period, the weight change data of the material in the return feeder is acquired by the first weighing unit at a first frequency and uploaded to the data acquisition and processing unit. The data acquisition and processing unit calculates the weight change ΔW(catch) and calculates the flow rate of solid particles captured by the cyclone separator per unit time as G(catch) = ΔW(catch) / Δt1. S3: After each cycle t0, the second weighing unit acquires the increase in the mass of the escaped solid particles collected by the dust collector during the second set time period Δt2 and uploads it to the data acquisition and processing unit. The data acquisition and processing unit then calculates the flow rate of the escaped solid particles per unit time, G(escape) = ΔW(escape) / Δt2. S4: Calculate the real-time total material flow rate G(total) and separation efficiency η based on the real-time catch material flow rate G(catch) and the real-time escape material flow rate G(escape): G(total) = G(catch) + G(escape), η = (G(catch) / G(escape)) × 100%; S5: Based on the real-time data of T, ΔP, and v measured in step S1, and combined with the real-time data of G(total), a continuous G(total)=f(T,ΔP,v) model is established through data regression analysis. Thus, by monitoring T, ΔP, and v, the corresponding G(total) value and the efficiency η of the cyclone separator can be obtained.
[0011] Furthermore, the above-mentioned cyclone separator monitoring method also includes the following steps: Every third set time interval, steps S1 to S4 are repeated to obtain data for model calibration, avoiding model drift that may occur when the model does not update data over time or when existing data is incorporated.
[0012] Furthermore, in step S2, the short-term weighing operation needs to be repeated N times (N>2, and N is a natural number) within Δt1 to finally obtain the weight change ΔW(catch). Repeated measurements are used to improve the accuracy of the data.
[0013] Furthermore, it also includes the following steps: An alarm unit is also added, with a preset separation efficiency alarm threshold and a fourth preset duration t4. When η is lower than the alarm threshold for a duration of t4, the system issues a warning signal to remind staff to adjust the equipment.
[0014] The beneficial effects of this invention include: This invention proposes a simple and easy-to-implement monitoring system and method, which only requires the addition of parameter acquisition devices for weight, temperature, pressure difference, and flow rate to the original system. Moreover, the method of directly acquiring weight parameters by instantaneous interception weighing can avoid the inherent cumulative error of indirect measurement methods. A material quantity change model is established through data regression analysis, which enables real-time output of separation efficiency through online parameter analysis. It has strong timeliness, and while ensuring data accuracy, it minimizes the adverse effects of measurement actions on normal operation. It solves the technical problems of the limitations of existing cyclone separator evaluation methods in terms of accuracy, real-time performance, and industrial applicability. Attached Figure Description
[0015] Figure 1 This is a schematic diagram of the cyclone separator monitoring system of the present invention.
[0016] Figure 2 This is a logic block diagram of the cyclone separator monitoring method of the present invention.
[0017] The diagram is labeled as follows: 1-Cyclone separator, 2-Return feeder, 201-First weighing unit, 3-Circulating fluidized bed, 301-Monitoring unit, 4-Dust collector, 401-Second weighing unit. Detailed Implementation
[0018] The present invention will now be described with reference to the accompanying drawings. Example 1
[0019] like Figure 1 The cyclone separator monitoring system of the present invention is shown, including a cyclone separator 1. The inlet of the cyclone separator 1 is connected to the outlet of a circulating fluidized bed 3. The outlet of the cyclone separator 1 is connected to a return feeder 2, and the flue gas outlet is connected to a dust collector 4. The dust collector 4 is a bag filter dust collector. A first weighing unit 201 is installed at the bottom of the return feeder 2. The first weighing unit 201 includes a structure that can close or restore the fluidizing air of the return feeder 2. A second weighing unit 301 is installed at the bottom of the dust collector 4. A monitoring unit is set at the inlet front end of the cyclone separator 1. The monitoring unit includes a temperature sensor, a differential pressure transmitter, and a flow rate measuring device. A data acquisition and processing unit is also provided and is signal-connected to the first weighing unit 201, the second weighing unit 301, and the detection unit to receive and process the acquired data in real time. Optionally, the data acquisition and processing unit can be one or more of a programmable logic controller (PLC), a distributed control system (DCS), and an industrial computer.
[0020] The aforementioned cyclone separator monitoring system includes a first weighing unit 201 comprising a weighing sensor and a quick-closing valve. The weighing sensor is connected to the data acquisition and processing unit via a signal connection, and the quick-closing valve is capable of shutting off or restoring the fluidizing air of the return feeder.
[0021] The second weighing unit 301 of the aforementioned cyclone separator monitoring system is a device capable of weighing solid particles.
[0022] In the aforementioned cyclone separator monitoring system, the cyclone separator 1 is flexibly connected to the return feeder 2 and the dust collector.
[0023] Taking a high-temperature cyclone separator of a 4MWth circulating fluidized bed boiler as an example, the cyclone separator monitoring system described in this embodiment, when put into use, firstly installs four high-temperature weighing modules (1t range, 0.05%FS accuracy) on the return feeder 2 and its riser support structure, and installs a pneumatic quick-closing butterfly valve (action time <1s) on the fluidized air header of the return feeder 2. This allows for the acquisition of weight change data over a time period while controlling the impact on the normal operation of the system. Then, an escape dust weighing box is installed on the tail flue dust collector equipment after the outlet of the cyclone separator's central cylinder. A weighing sensor with a range of 0.3t and an accuracy of 0.05%FS is installed at the bottom of the box. Thermocouples and differential pressure transmitters are installed on the furnace equipment at the front end of the cyclone separator, and a gas flow meter is installed on the air inlet pipe of the furnace to obtain the real-time temperature, differential pressure value, and gas flow rate inside the fluidized bed furnace.
[0024] like Figure 2 The diagram illustrates the cyclone separator monitoring method of the present invention. Taking a period of data measurement as an example, and combining it with the aforementioned cyclone separator monitoring system, the method includes the following steps: S1: Acquire monitoring unit parameters: When the circulating fluidized bed boiler is in a stable operating state, continuously collect the real-time temperature T, differential pressure ΔP and gas flow rate v in the furnace or reaction device at the front end of the cyclone separator 1; S2: In this embodiment, the cycle time is set to t0=20min. When the CFB boiler is running stably at 100% load, every 20min, the quick-closing valve is closed at time t1 to temporarily stop the flow of material in the riser and return feeder 2 and start to accumulate. Then, the quick-closing valve is opened at time t2 to restore the normal flow of the system. The first weighing unit 201 acquires the weight change data in the return feeder at the first frequency of 10Hz within the first set time period Δt1=t2-t1=5s and uploads it to the data acquisition and processing unit. The data acquisition and processing unit calculates the weight change ΔW(catch)=95.5kg and obtains the solid particle flow rate captured by the cyclone separator per unit time as G(catch)=ΔW(catch) / Δt=19.1kg / s. S3: After each cycle t0=20min, the second weighing unit obtains the increase in the mass of the escaped solid particles collected by the dust collector during the second set time Δt2=60s, ΔW(escape)=5.8kg, and calculates the flow rate of the escaped solid particles per unit time G(escape)= ΔW(escape) / Δt2=0.097kg / s, and uploads ΔW(escape) and G(escape) to the data acquisition and processing unit; S4: Based on the real-time catch material flow rate G(catch) and the real-time escape material flow rate G(escape), the real-time total material flow rate G(total) = f(T, ΔP,v) = 19.1kg / s + 0.097kg / s ≈ 19.2kg / s and the separation efficiency η = (G(catch) / G(escape)) × 100% ≈ 99.48%.
[0025] S5: Based on the real-time data of T, ΔP, and v measured in step S1, and combined with the real-time data of G(total), a continuous model G(total) = f(T, ΔP, v) is established through data regression analysis. In this embodiment, f is a mathematical model fitted by a multiple regression or machine learning algorithm. For example: G(total)=a*ΔP^α+b*V^β+c*T^γ+d, Where a, b, c, d are fitting coefficients, ΔP is differential pressure, V is flow velocity, T is temperature, and α, β, γ are the power coefficients of the corresponding variables.
[0026] Therefore, by monitoring T, ΔP, and v, the corresponding G(total) value and the efficiency η of the cyclone separator can be obtained. According to calculations, the accuracy of the model's fitting value to G(total) has a coefficient of determination R² > 0.96. Other fitting models that can meet the prediction accuracy of G(total) can also be established to replace the above model. Example 2
[0027] Taking a data measurement process of a 50MWth chemical loop combustion high-temperature cyclone separator as an example, firstly, four high-temperature weighing modules with a range of 10t and an accuracy of 0.05%FS are installed on the return feeder and its riser support structure. A pneumatic quick-closing butterfly valve (action time <1s) is installed on the fluidizing air header of the return feeder. Then, an escape dust weighing box is installed on the tail flue dust collector equipment after the outlet of the central cylinder of the cyclone separator. A weighing sensor with a range of 0.5t and an accuracy of 0.05%FS is installed at the bottom of the box. Finally, a thermocouple and a differential pressure transmitter are installed at the front end of the cyclone separator inlet, and a gas flow meter is installed on the air inlet pipe of the furnace.
[0028] The monitoring method for cyclone separators using the above structure includes the following steps: S1: When the circulating fluidized bed boiler is in a stable state, acquire the monitoring unit parameters: real-time temperature T, differential pressure ΔP and gas flow rate v; S2: Set the cycle time t0, perform 5 short-term weighing operations, with an interval of 5 seconds between each closing and opening of the start-up quick-closing butterfly valve. The weighing system records the material increase ΔW(catch) = 2900kg and uploads it to the data acquisition and processing unit. The data acquisition and processing unit then calculates G(catch) = ΔW(catch) / 5s = 580kg / s. S3: After each cycle t0, the second weighing unit obtains the weight of the escaped material in the dust collector equipment hopper downstream of the tail flue of the cyclone separator central cylinder outlet within 60s as ΔW(escape)=66kg and uploads it to the data acquisition and processing unit. The data acquisition and processing unit then calculates G(escape)= ΔW(escape) / 60s=1.1kg / s. S4: Based on the real-time catch material flow rate G(catch) and the real-time escape material flow rate G(escape), the real-time total material flow rate G(total) = f(T, ΔP, v) = G(catch) + G(escape) = 581.1 kg / s and the separation efficiency η = (G(catch) / G(escape)) × 100% ≈ 99.81% S5: Based on the real-time data of T, ΔP and v measured in step S1, and combined with the real-time data of G(total), a continuous G(total)=f(T, ΔP,v) model is established through data regression analysis. Thus, by monitoring T, ΔP and v, the corresponding G(total) value and the efficiency η of the cyclone separator can be obtained. S6: Set the third setting duration to 24h. Every 24h, repeat steps S1 to S4 to obtain data for model calibration. S7: An additional alarm unit is provided, with a preset separation efficiency alarm threshold η=99.6% and a fourth preset duration t4=120s. When η is lower than the alarm threshold and the duration reaches 120s, the system issues a warning to remind the operators to check and adjust the equipment.
[0029] Compared with existing technologies, this invention fits a model G(total)=f(T, ΔP,v) using real-time monitoring parameters and total material flow rate at certain operating points. This allows the model to learn the mapping relationship between parameters. In subsequent continuous operation, the system no longer needs to perform frequent weighing measurements. It only needs to measure temperature, pressure difference, and flow rate and input them into the trained model to predict the working efficiency data of the cyclone separator. This is simple and easy to implement. Furthermore, the model is remeasured every 24 hours to correct it and avoid possible model drift. This application uses high-precision direct weighing measurements to calibrate an implementation calculation model based on easily measurable process parameters, resulting in high accuracy and reliability, and low maintenance costs. In addition, the amount of solid circulating material obtained from the cyclone separator can provide key data support for the optimized operation of the entire device (such as bed temperature control and combustion efficiency optimization).
Claims
1. A cyclone separator monitoring system, comprising a cyclone separator (1), wherein the inlet of the cyclone separator (1) is connected to the outlet of a circulating fluidized bed (3), the outlet of the cyclone separator (1) is connected to a return feeder (2), and the flue gas outlet is connected to a dust collector (4), characterized in that: The bottom of the return feeder (2) is equipped with a first weighing unit (201), which includes a structure that can shut off or restore the fluidizing air of the return feeder (2). The bottom of the dust collector (4) is equipped with a second weighing unit (301). A monitoring unit is provided at the front end of the inlet of the cyclone separator (1). The monitoring unit includes a temperature sensor, a differential pressure transmitter and a flow rate measuring device. A data acquisition and processing unit is also provided, which is connected to the first weighing unit (201), the second weighing unit (301) and the monitoring unit to receive and process the acquired data in real time.
2. The cyclone separator monitoring system according to claim 1, characterized in that: The first weighing unit (201) includes a weighing sensor and a quick-closing valve. The weighing sensor is connected to the data acquisition and processing unit. The quick-closing valve is used to close or restore the fluidizing air of the return feeder (2).
3. The cyclone separator monitoring system according to claim 1, characterized in that: The second weighing unit (301) employs a device capable of weighing solid particles.
4. The cyclone separator monitoring system according to claim 1, characterized in that: The cyclone separator (1) is flexibly connected to the return feeder (2) and the dust collector (4), and the return feeder (2), the dust collector (4) and the rear-end components are flexibly connected.
5. A cyclone separator monitoring method, employing any one of the cyclone separator monitoring systems described in claims 1 to 4, characterized in that: Includes the following steps: S1: While ensuring the entire circulating fluidized bed boiler is in a stable operating state, acquire the monitoring unit parameters in real time: real-time temperature T, differential pressure ΔP and gas flow rate v; S2: Set a cycle time t0. Every time t0 elapses, the fluidizing air outlet of the return feeder (2) needs to be closed within a first set time Δt1. During this period, the first weighing unit (201) is used to acquire the weight change data of the material in the return feeder (2) at a first frequency and upload it to the data acquisition and processing unit. The data acquisition and processing unit calculates the weight change ΔW(catch) and obtains the flow rate of solid particles captured by the cyclone separator per unit time as G(catch) = ΔW(catch) / Δt1. S3: After each cycle t0, the second weighing unit (301) acquires the increase in mass of the escaped solid particles collected by the dust collector (4) during the second set time Δt2 and uploads it to the data acquisition and processing unit. The data acquisition and processing unit then calculates the flow rate of the escaped solid particles G(escape) = ΔW(escape) / Δt2. S4: Calculate the real-time material flow rate G(total) and separation efficiency η based on the real-time material flow rate G(catch) and the real-time material flow rate G(escape) per unit time: G(total) = G(catch) + G(escape), η = (G(catch) / G(escape)) × 100%; S5: Based on the real-time data of T, ΔP and v measured in step S1, and combined with the real-time data of G(total), a continuous G(total)=f(T,ΔP,v) model is established through data regression analysis. Thus, by monitoring T, ΔP and v, the corresponding G(total) value and the efficiency η of the cyclone separator (1) can be obtained.
6. The cyclone separator monitoring method according to claim 5, characterized in that: It also includes the following steps: Every third set time interval, steps S1 to S4 are repeated to obtain data for model calibration.
7. The cyclone separator monitoring method according to claim 5, characterized in that: In step S2, the weighing operation needs to be repeated N times (N>2, and N is a natural number) within Δt1 to finally obtain the weight change ΔW(catch).
8. The cyclone separator monitoring method according to claim 5, characterized in that: It also includes the following steps: An alarm unit is also added, with a preset separation efficiency alarm threshold and a fourth preset duration t4. When η is lower than the alarm threshold for a duration of t4, the alarm unit issues a warning signal.