Intelligent frequency conversion coordinated control system for water circulation heating network of tunnel fluidized bed furnace

Abstract

This invention relates to the field of industrial control technology and discloses an intelligent variable frequency collaborative control system for a tunnel fluidized bed furnace water circulation heating pipe network. The system includes: establishing an axial mapping model between the tunnel furnace body and the heating pipe network; identifying online the hydraulic parameters of the water circulation branches and the bed-side thermal resistance of the fluidized bed; quantifying the water delivery delay, bed-side thermal inertia, and the thermal phase deviation generated when the heat demand arrives, thereby shifting the initial heat demand forward to generate compensating heat demand; using heat residual iterative back-calculation to deduce the target flow rate of each branch; and, under the premise of satisfying the node continuity law and head loss constraints, obtaining the optimal pump frequency with the goal of minimizing the total power consumption of multiple variable frequency pumps, supplemented by actual thermal deviation closed-loop correction to obtain the final pump frequency. This invention effectively improves the matching degree between the pipeline distribution delay and the material arrival time, and reduces the overall operating power consumption of the variable frequency pump group.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, and more specifically, to an intelligent variable frequency collaborative control system for a water circulation heating network of a tunnel fluidized bed furnace. Background Technology

[0002] In continuous production using large-scale tunnel fluidized bed reactors, the furnace is typically divided into multiple independent thermal zones along the material conveying direction, and a heating network with parallel branch lines for supply and return water is used for heat distribution. Currently, the commonly used control strategy is based on the real-time temperature deviation of each zone, using PID loops to independently adjust the operating frequency of the corresponding branch pumps or the opening of regulating valves. However, in actual production, this single-loop control often triggers complex cascading reactions.

[0003] On the one hand, there are multiple and severe time lag effects within the heating system: the transport of fluids in long pipelines takes time, and the penetration of heat through the heating pipe wall and the change of the heat transfer state of the fluidized bed also takes time, while the materials carrying the process requirements are constantly moving forward on the conveyor belt. This means that by the time the variable frequency pump's adjustment action produces a substantial thermal effect, the materials that need that heat have often already left the specific temperature zone, causing frequent localized delayed heating or overheating fluctuations in various temperature zones.

[0004] On the other hand, multiple variable frequency pumps share the same supply and return water header system. Increasing the frequency and flow rate of a single pump inevitably causes a global change in the main pipe pressure, leading to a decrease in available head and passive reduction in flow rate in other parallel branches, resulting in typical fluid flow competition and pressure differential mixing phenomena. To cope with this flow distribution imbalance, operators or the underlying control logic often have to blindly increase the operating frequency of all pumps to maintain pressure and supply. This approach not only masks the actual heat exchange efficiency bottleneck on the fluidized bed side but also keeps the entire pump unit operating at uneconomical high speeds for extended periods, resulting in significant additional energy losses. Summary of the Invention

[0005] This invention provides an intelligent variable frequency collaborative control system for the water circulation heating pipeline network of a tunnel fluidized bed furnace, which solves the technical problems mentioned in the background art.

[0006] This invention provides an intelligent variable frequency collaborative control system for the water circulation heating network of a tunnel fluidized bed furnace. It is applied to a heating system comprising a tunnel furnace body divided into multiple furnace sections along the material movement direction, a fluidized bed, a heating network containing multiple water circulation branches and supply / return water headers, and multiple variable frequency pumps. The system is configured to execute: A mapping model reflecting the axial thermal interaction between the tunnel furnace and the heating pipe network is constructed, and the actual thermal power of each furnace section is analyzed to eliminate the spatial distribution differences of the pipe network. Based on the pipeline network operation data, the hydraulic parameters of the water circulation branch are identified, and a nodal continuity law characterizing flow conservation and pipeline head loss constraints are constructed to isolate fluid interaction coupling interference. Based on the actual thermal power inversion, the bed-side thermal resistance characterizing the gas-solid phase heat transfer characteristics in each of the furnace sections is obtained. The water delivery delay, bed-side thermal inertia, and material heat demand arrival time of each furnace section are quantified and combined to form a thermal phase deviation that characterizes water transport, pipe wall heat transfer, and material displacement time misalignment. Based on the thermal phase deviation, the initial heat demand characterizing the static heating load of the furnace section is shifted forward along the time axis to generate a compensating heat demand to make up for the time misalignment. Construct a heat residual formula and solve iteratively; based on the bed-side thermal resistance, back-calculate the target flow rate of each water circulation branch from the required heat for compensation. Under the premise of satisfying the node continuity law and the head loss constraint, the optimal pump frequency is solved with the goal of minimizing the total power consumption of the variable frequency pump. The actual thermal deviation, which represents the difference between the actual thermal power and the compensation heat required, is extracted and used for closed-loop feedback calculation to correct the unknown external disturbances of the system and obtain the final pump frequency.

[0007] The beneficial effects of this invention include: overcoming the time misalignment problem between heat transfer and material demand in the long-axial furnace pipe network system by coordinating the unified timing alignment of water flow transport lag, heat exchange inertia lag, and material displacement time; upgrading isolated single-point temperature control to global flow feedforward allocation based on future heat demand by online identification of pipe network hydraulic parameters and fluidized bed side thermal resistance, and combining the node continuity law constraint of supply and return water headers, effectively reducing the overall operating energy consumption of the system's variable frequency pump group while suppressing multi-pump parallel flow grabbing and differential pressure oscillation, and improving the dynamic matching accuracy of heat exchange targets in each temperature zone along the process. Attached Figure Description

[0008] Figure 1 This is a flowchart of the intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized furnace of the present invention. Detailed Implementation

[0009] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0010] like Figure 1As shown, the intelligent variable frequency collaborative control system for the water circulation heating network of the tunnel fluidized bed furnace is applied to a heating system that includes a tunnel furnace body divided into multiple furnace sections along the material movement direction, a fluidized bed, a heating network containing multiple water circulation branches and supply and return water headers, and multiple variable frequency pumps. The system is configured to execute: A mapping model reflecting the axial thermal interaction between the tunnel furnace and the heating pipe network is constructed, and the actual thermal power of each furnace section is analyzed to eliminate the spatial distribution differences of the pipe network. Based on the pipeline network operation data, the hydraulic parameters of the water circulation branch are identified, and a nodal continuity law characterizing flow conservation and pipeline head loss constraints are constructed to isolate fluid interaction coupling interference. Based on the actual thermal power inversion, the bed-side thermal resistance characterizing the gas-solid phase heat transfer characteristics in each of the furnace sections is obtained. The water delivery delay, bed-side thermal inertia, and material heat demand arrival time of each furnace section are quantified and combined to form a thermal phase deviation that characterizes water transport, pipe wall heat transfer, and material displacement time misalignment. Based on the thermal phase deviation, the initial heat demand characterizing the static heating load of the furnace section is shifted forward along the time axis to generate a compensating heat demand to make up for the time misalignment. Construct a heat residual formula and solve iteratively; based on the bed-side thermal resistance, back-calculate the target flow rate of each water circulation branch from the required heat for compensation. Under the premise of satisfying the node continuity law and the head loss constraint, the optimal pump frequency is solved with the goal of minimizing the total power consumption of the variable frequency pump. The actual thermal deviation, which represents the difference between the actual thermal power and the compensation heat required, is extracted and used for closed-loop feedback calculation to correct the unknown external disturbances of the system and obtain the final pump frequency.

[0011] The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized furnace provided in this embodiment is applied to a heating system that includes a tunnel furnace body divided into multiple furnace sections along the material movement direction, a fluidized bed, a heating pipeline network containing multiple water circulation branches and supply and return water headers, and multiple variable frequency pumps.

[0012] S201: Determine the corresponding furnace section interval in the tunnel furnace body for each furnace section.

[0013] The furnace segment indexed by i corresponds to the axial spatial range within the tunnel furnace body. The value of i ranges from 1 to n, where n is the total number of furnace segments in the tunnel furnace body. The furnace segment interval is defined by the starting position of that segment. and termination position Limited, forming furnace section intervals The calculation formula is: and The measurement can be directly read from the tunnel furnace design drawings or measured on-site using a laser rangefinder, with the unit being meters. For furnace sections evenly divided along the material movement direction, , Where L is the total length of the tunnel furnace body, in meters.

[0014] S202: Analyze the effective heat exchange ratio of the water circulation branch within the furnace section and extract the branch coverage that reflects the actual thermal radiation coverage area of ​​the pipeline.

[0015] The effective heat transfer of the water circulation branch with index e at axial position x is determined by the space function. Characterization, when position x is within the effective heat exchange coverage area of ​​the heating tube in branch e, ,otherwise The effective heat transfer coverage area of ​​the heating tube refers to the axial region where the outer wall of the heating tube is in direct contact with the fluidized bed, excluding the portion of the heating tube extending out of the furnace wall and the pipe joint area. (Regarding the space function...) In the furnace section Integrate within the interval and then divide by the length of the furnace section to obtain the branch coverage of branch e to furnace section i. The calculation formula is: The value range is [0,1], representing the proportion of the effective heat transfer length of the heating tube in branch e within furnace section i to the total length of the furnace section. For discretely arranged heating tubes, summation can be used instead of integration, i.e. ,in The number of heating tubes in branch e within furnace section i. Let be the effective heat exchange length of the k-th heating tube within furnace section i.

[0016] S203: Synchronously collect the real-time flow rate, inlet water temperature, and outlet water temperature of the water circulation branch.

[0017] An electromagnetic flow meter is installed at a distance of at least 5 times the pipe diameter from the inlet of each water circulation branch to collect the real-time flow of branch e. The unit is cubic meters per second. A Class A platinum resistance temperature sensor is installed at least 10 times the pipe diameter at the inlet of the heating pipe in each water circulation branch to collect the branch's inlet water temperature. A Class A platinum resistance temperature sensor is installed at a distance of at least 10 times the pipe diameter from the outlet of the heating pipe at the outlet of each water circulation branch to collect the outlet water temperature of the branch. Temperature is measured in degrees Celsius. All sensor sampling clocks are synchronized via the NTP network time protocol, with a time synchronization error not exceeding 10% of the sampling period. The sampling period can be set from 1 second to 1 minute depending on the control accuracy requirements; a setting of 10 seconds is recommended.

[0018] The electromagnetic flowmeter has an accuracy class of 0.5 and a measurement range covering 0 to 120% of the branch's maximum design flow. The Class A platinum resistance temperature sensor has a measurement range of 0 to 200 degrees Celsius and a response time of less than 1 second. All sensor signal lines are laid with shielded cables to avoid electromagnetic interference, and the shielding layer is grounded at one end.

[0019] S204: Combine the circulating water density and water specific heat capacity to extract the enthalpy difference of the water path, and correlate the enthalpy difference of the water path, the branch coverage and real-time flow rate to calculate, eliminate the interference of flow segments with no heat exchange effect, and collect the actual heat power that reflects the real net heat energy obtained.

[0020] Circulating water density The value is taken as 998.2 kg / m³ (under standard conditions at 20°C). When the average circulating water temperature deviates from 20°C by more than 10°C, the corresponding density value can be found in the international steam table based on the average water temperature. Specific heat capacity of water. The value is taken as 4182 joules per kilogram of Celsius (under standard conditions at 20°C). When the water temperature varies significantly, the corresponding value can be found in the international steam table. The enthalpy difference of the water circuit is obtained by multiplying the difference between the inlet and outlet water temperatures of each branch by the specific heat capacity of the water and the density of the circulating water. The actual thermal power of furnace section i is obtained by multiplying the enthalpy difference of each branch by the corresponding branch coverage and real-time flow rate, and then summing the results for all branches covering the furnace section. The unit is kilowatt, and the calculation formula is: Where E represents the total number of water circulation branches covering furnace section i. By weighting the branches based on their coverage, interference from flow segments that do not pass through the heat exchange area of ​​the furnace section is eliminated, and the actual thermal power obtained accurately reflects the net thermal energy obtained by furnace section i from the water circulation system.

[0021] S301: Collect the branch pressure difference of the water circulation branch and extract the branch water inertia that characterizes the lag in fluid momentum state.

[0022] High-precision pressure transmitters are installed at the inlet and outlet of each water circulation branch to collect the branch pressure difference. The unit is meters of water column. Branch water inertia. The hysteresis characteristic characterizing changes in the momentum state of a fluid, expressed in seconds squared per square meter. It can be calculated from the pipeline geometric parameters, and the calculation formula is as follows: in The total length of branch e is in meters, which can be read from the pipeline design drawings or measured on site. Let be the cross-sectional area of ​​branch e, in square meters, calculated from the inner diameter of the branch. , denoted as the inner diameter of the branch; g is the gravitational acceleration, with a value of 9.81 m / s².

[0023] The high-precision pressure transmitter has an accuracy class of 0.2 and a measurement range covering 0 to 150% of the maximum design differential pressure of the branch. The pressure transmitter is installed on the straight pipe section at the inlet and outlet of the branch, at least 10 times the pipe diameter away from the nearest bend or valve, to avoid the influence of fluid eddies on the measurement accuracy.

[0024] S302: Based on the characteristics of real-time operation data, the unsteady pressure fluctuation interference caused by the inertia of branch water is removed, and the branch flow resistance, which characterizes the effects of flow along the path and local obstruction, is extracted.

[0025] Select a window containing the current valid running data. The window length is set to 100 sampling points, corresponding to 1000 seconds (when the sampling period is 10 seconds). For each timestamp within the window... Calculate the rate of change of branch flow The central difference method is used for calculation, i.e. ,in The sampling period is defined. For the timestamps at the window boundaries, forward or backward difference methods are used for calculation. After removing the unsteady pressure fluctuations caused by the inertia of the branch water, the steady-state pressure difference component is obtained. Based on the least squares method fitting of the relationship between steady-state pressure difference and the square of flow rate, the branch flow resistance is obtained. The unit is meters per second (cubic meters per second) squared, and the calculation formula is: Branch flow resistance It reflects the combined effect of pipeline friction resistance and local resistance, and is dynamically updated according to factors such as pipeline scaling, valve opening changes, and filter blockage, updating once every 10 control cycles.

[0026] S303: Construct a network association array that maps the network topology, couple it with an array composed of the real-time flow of each water circulation branch, and establish the node continuity law.

[0027] The pipe network correlation matrix B is an N×E matrix, where N is the total number of pipe network nodes and E is the total number of water circulation branches. For each element in the matrix... If node i is the starting point of branch j, then If node i is the endpoint of branch j, then ;otherwise Pipeline nodes refer to the connection points between the supply and return water mains and branches, the connection points between branches, and the inlet and outlet connection points of the variable frequency pump. The pipeline correlation matrix B is then compared with the real-time flow vector. Multiplication results in a zero vector, which is the nodal continuity law: The continuity law of nodes indicates that the inflow rate at each node in the pipeline network is equal to the outflow rate, satisfying the law of conservation of mass.

[0028] S304: Based on the pump head coefficient and pump resistance coefficient of the actual operating curve of the variable frequency pump, extract the variable frequency pump head as it evolves with the current frequency and real-time flow rate of the variable frequency pump.

[0029] The actual operating curve of the variable frequency pump with index j can be obtained by fitting the performance curve provided by the pump manufacturer. First, obtain the rated frequency. The head-flow rate curve at (typically 50 Hz) is fitted to a quadratic polynomial form. ,in The zero-flow head coefficient at rated frequency. This represents the resistance coefficient at the rated frequency. According to the similarity law of centrifugal pumps, the pump head coefficient after frequency conversion is... Pump resistance coefficient Variable frequency pump head With current frequency and real-time traffic The change, measured in meters of water column, is calculated using the following formula: in The unit is Hertz. The unit is cubic meters per second.

[0030] S305: The power of the variable frequency pump is obtained by mapping and converting the pump head, circulating water physical properties and variable frequency pump efficiency.

[0031] Variable frequency pump efficiency The flow rate can be obtained by checking the efficiency curve provided by the pump manufacturer. and frequency The function can be prefitted into a two-dimensional polynomial form. ,in to These are the fitting coefficients. Variable frequency pump power. The unit is kilowatt, and the calculation formula is: This formula converts the pump's hydraulic power into shaft power, reflecting the actual electrical energy consumption of the variable frequency pump.

[0032] S306: Branch flow resistance, variable frequency pump head, and node continuity law together constitute hydraulic parameters.

[0033] Hydraulic parameters are the core set of parameters describing the hydraulic characteristics of a water circulation network, including the flow resistance of each branch. Head of each variable frequency pump and the nodal continuity law characterizing flow conservation These parameters provide the foundation for subsequent flow allocation optimization and pump frequency calculation.

[0034] S401: Extract the equivalent inlet and outlet water temperatures mapped to the furnace section and based on the branch inlet and outlet water temperatures, calculate the logarithmic temperature difference between the equivalent inlet and outlet water temperatures and the fluidized bed temperature, and characterize the driving force of heat exchange temperature inside and outside the furnace.

[0035] Equivalent inlet water temperature of furnace section i and equivalent outlet water temperature It is obtained by weighted averaging of the inlet and outlet water temperatures of each branch covering the furnace section, with the weight being the branch coverage. and branch flow The product of the two is calculated using the following formula: When the sum of the flow rates of all branches covering furnace section i is zero, the equivalent inlet and equivalent outlet water temperatures are taken from the values ​​of the previous control cycle. Fluidized bed temperature Temperature data were collected by armored thermocouples installed 50 mm from the outer wall of the heating tubes within the fluidized bed. Three thermocouple measuring points were evenly distributed in each furnace section, and the average value was taken as the fluidized bed temperature for that section, expressed in degrees Celsius. Logarithmic mean temperature difference. The unit is Celsius, and the calculation formula is: when To avoid singular values ​​in logarithmic operations, the arithmetic mean temperature difference is used instead of the logarithmic mean temperature difference, i.e. .

[0036] The armored thermocouples are of Class A accuracy, with a measurement range of 0 to 1000 degrees Celsius and a response time of less than 2 seconds. The thermocouples are inserted into the fluidized bed to a depth of at least 100 mm and maintain a distance of at least 50 mm from the outer wall of the heating tube to avoid directly measuring the heating tube temperature. Three thermocouple measuring points are evenly distributed along the width of the furnace section, located on the left, middle, and right sides of the furnace body, respectively.

[0037] S402: Combining the logarithmic temperature difference, the heat exchange area of ​​the furnace section, and the actual thermal power, the total heat transfer coefficient reflecting the comprehensive heat exchange intensity inside and outside the tube is calculated.

[0038] furnace section heat exchange area The unit is square meters, calculated from the sum of the outer surface areas of all heating tubes within the furnace section, i.e. ,in This represents the total number of heating tubes within furnace section i. This refers to the outer diameter of the heating element, in meters. The effective heat transfer length of a single heating tube within furnace section i, expressed in meters. Overall heat transfer coefficient. The unit is watts per square meter in degrees Celsius, and the calculation formula is: The overall heat transfer coefficient reflects the comprehensive heat transfer intensity from circulating water to the fluidized bed, including the sum of thermal resistances of three stages: convection heat transfer inside the tube, heat conduction through the tube wall, and gas-solid two-phase heat transfer outside the tube.

[0039] S403: Based on the characteristics of circulating water flow state, Reynolds number and Prandtl number are constructed, and water convection coefficient, which characterizes the forced convection heat transfer capacity of fluid in the pipe, is extracted.

[0040] Average water flow velocity in furnace section i The unit is meters per second, obtained by dividing the total flow rate in the furnace section by the total internal cross-sectional area of ​​the heating tubes. ,in The inner diameter of the heating tube is in meters. Reynolds number. It is a dimensionless parameter characterizing the fluid flow state, and its calculation formula is: in This is the dynamic viscosity of water, measured in Pascals per second (Pa·s). It can be obtained from standard thermodynamic data tables based on the average temperature of the circulating water. Prandtl number. It is a dimensionless parameter characterizing the relative magnitudes of fluid momentum diffusion and heat diffusion, and its calculation formula is: in This is the thermal conductivity of water, measured in watts per meter per degree Celsius. It can be obtained from standard thermodynamic data tables based on the average temperature of the circulating water. For turbulent flow conditions (… The Nuselle number is calculated using the Dittus-Boelter correlation. : Water convection coefficient The unit is watts per square meter in degrees Celsius, and the calculation formula is: The water convection coefficient characterizes the forced convection heat transfer capacity between the circulating water in the pipe and the inner wall of the heating pipe.

[0041] S404: The water convection coefficient is removed from the overall heat transfer coefficient, and the conductive impedance of the heating tube body is eliminated, thus separating the bed-side thermal resistance, which specifically characterizes the heat transfer barrier between the gas and solid phases.

[0042] The conduction impedance of the heating tube body is determined by the wall thickness of the heating tube. and the thermal conductivity of the pipe wall The calculation yields a value measured in degrees Celsius per watt, using the following formula: .in The unit is meters, which can be obtained from the product specifications of the heating element; The unit is watts per meter (°C), determined by the heating element material. Stainless steel heating elements are rated at 16 watts per meter (°C), while carbon steel heating elements are rated at 45 watts per meter (°C). Total thermal resistance. Equal to the thermal resistance of water convection Pipe wall thermal resistance and bed-side thermal resistance The sum of these. Therefore, the bed-side thermal resistance The unit is per square meter per watt (°C), and the calculation formula is: Bed-side thermal resistance specifically characterizes the gas-solid two-phase heat transfer barrier between the outer wall of the heating tube and the fluidized bed, and is a key parameter affecting the heat transfer efficiency of the fluidized bed.

[0043] S405: Convert the bed-side thermal resistance into the bed heat transfer coefficient, and combine the particle density, solid flux, apparent gas velocity and fluidized bed temperature of the fluidized bed to perform gas-solid state correlation, and calibrate the bed heat transfer parameters to eliminate the interference of bed material state fluctuations.

[0044] bed heat transfer coefficient It is the reciprocal of the bed-side thermal resistance, expressed in watts per square meter per degree Celsius. The heat transfer coefficient of the fluidized bed is closely related to the operating state of the fluidized bed. A logarithmic linear model is used to correlate the gas and solid states, and the calculation formula is as follows: in The particle density of the fluidized bed is measured in kilograms per cubic meter. It is obtained by taking samples of the bed material and measuring them using the hydrostatic bottle method. The measurement is performed once a week. Solid flux is expressed in kilograms per square meter per second and is calculated by dividing the material flow rate by the cross-sectional area of ​​the bed. The apparent air velocity is expressed in meters per second and is calculated by dividing the fluidizing airflow by the cross-sectional area of ​​the bed. to The initial bed heat transfer parameters are to be calibrated. These parameters are obtained by least-squares fitting of historical data from steady-state operation, with a minimum of 1000 sets of data.

[0045] S406: Dynamically update bed heat exchange parameters based on real-time operating data to eliminate the impact of bed material condition fluctuations.

[0046] During system operation, the bed heat transfer parameters are dynamically updated using a recursive least squares method with a forgetting factor. The value is set to 0.98 to accommodate the effects of slow changes in bed material particle size, moisture content, and coking degree, while also filtering measurement noise. The updated formula is: in For the regression vector, The covariance matrix is ​​initialized to... , It is a 5th order identity matrix.

[0047] S501: Based on the effective water capacity, branch coverage and real-time flow of each water circulation branch covering the furnace section, extract the water delivery delay that characterizes the physical time required for the water in the pipeline to flow to the heat exchange section.

[0048] The effective water capacity of water circulation branch e relative to furnace section i The unit is cubic meters, which is defined as the volume of water in the pipeline from the water inlet point of branch e (the connection point between the water supply header and branch e) to the inlet of the first heat exchange tube in the furnace section i, excluding the part inside the heat exchange section and the part after the heat exchange section to the return header. The calculation formula is: ,in This refers to the length of the pipeline from the branch point of the water supply main pipe to the inlet of the heat exchange section i of the boiler, expressed in meters. It can be accurately measured from the pipeline network design drawings. Water delivery lag. The unit is seconds, representing the average physical time required for water to flow from the branch inlet to the heat exchange zone i of the furnace section. The calculation formula is: The water delivery delay changes dynamically with the flow rate of the branch; the greater the flow rate, the smaller the water delivery delay.

[0049] S502: Combining the equivalent heat capacity of the tube wall and the bed material with the overall heat transfer coefficient, extract the bed-side thermal inertia that characterizes the thermal response delay time that heat penetrates the tube wall and causes the bed temperature to appear.

[0050] Equivalent heat capacity of tube wall and bed material The unit is joules per degree Celsius, which can be obtained through experimental calibration or calculated based on design parameters. The calculation formula is as follows: ,in The mass of the heating tubes in furnace section i is expressed in kilograms. This refers to the specific heat capacity of the heating element material, expressed in joules per kilogram per degree Celsius. For stainless steel, the value is 500 joules per kilogram per degree Celsius. The mass of the bed material in furnace section i is expressed in kilograms. Specific heat capacity of the bed material, expressed in joules per kilogram of temperature (°C). For quartz sand, the value is taken as 800 joules per kilogram of temperature (°C). Bed side thermal inertia. The unit is seconds, representing the time required for heat to transfer from the inner wall of the heating tube to the fluidized bed and cause a significant change in bed temperature. The calculation formula is: The thermal inertia of the bed side changes dynamically with the change of the overall heat transfer coefficient. The larger the overall heat transfer coefficient, the smaller the thermal inertia of the bed side.

[0051] The experimental calibration procedure for the equivalent heat capacity of the tube wall and bed material is as follows: Under steady-state operation conditions with no material and no fluidizing air, the furnace body is heated with a constant power P. After the bed temperature stabilizes, the bed temperature T1 and time t1 are recorded. Then, heating is suddenly stopped, and the change curve of bed temperature over time is recorded. When the bed temperature drops to T2, time t2 is recorded. According to the law of conservation of energy, the formula for calculating the equivalent heat capacity of the tube wall and bed material is: Repeat the calibration experiment three times and take the average value.

[0052] S503: Based on the material movement direction and speed, the time taken for the physical displacement of the material to reach the spatial position of the furnace section is taken as the heat demand arrival time.

[0053] Furnace section spatial location Defined as the inlet position of furnace section i This refers to the position where the material enters furnace section i and begins to require heating. Material movement speed. The unit is meters per second, representing the speed of material movement at axial position x, which is acquired by a speed encoder of the conveyor belt. The corresponding formula for calculating belt speed (unit: revolutions per minute) is: ,in This refers to the diameter of the conveyor belt rollers, in meters. For a conveyor belt moving at a constant speed, It is a constant. The heat required to reach [the destination] is [a constant]. The unit is seconds, representing the physical time required for material to move from the furnace inlet to the inlet position of furnace section i. The calculation formula is: For materials moving at a constant speed, the heat required to arrive can be simplified to: .

[0054] S504: Integrates water delivery lag and bedside thermal inertia, and performs time axis alignment analysis with the arrival of heat demand to generate a thermal phase deviation quantity that reflects the degree of time misalignment between the heat supply end and the material demand end.

[0055] The total lag time at the heat supply end is the sum of the water delivery lag and the bedside thermal inertia, i.e. Thermal phase deviation The unit is seconds, representing the difference between the total lag time of heat supply and the arrival time of material heat demand. The calculation formula is: when When this occurs, it indicates that the heat supply lags behind the material demand, and heat needs to be supplied in advance; when This indicates that the heat supply is ahead of the material demand, and the heat supply needs to be delayed. The heat phase deviation provides a time reference for subsequent heat demand compensation.

[0056] S601: Extract material flow rate that characterizes the continuous throughput of materials based on their physicochemical properties and loading status.

[0057] Material density The unit is kilograms per cubic meter, determined by the physicochemical properties of the material, and obtained through sampling and measurement; each batch of material is measured once. Effective cross-sectional area of ​​the material. The unit is square meters, representing the cross-sectional area of ​​the material on the conveyor belt. It is obtained by scanning with a laser level gauge installed at least 1 meter from the feed inlet at the furnace inlet. Five measuring points are evenly distributed along the width of the conveyor belt, and the scanning frequency is consistent with the system sampling frequency. Material flow rate. The unit is kilograms per second, representing the mass of material passing through a certain cross-section of the furnace body per unit time. The calculation formula is: The laser level gauge has an accuracy of ±1 mm, a scanning frequency of 10 Hz, and a measurement range covering 0 to 150% of the maximum material layer height on the conveyor belt. The laser level gauge is installed directly above the conveyor belt, at a height of no less than 2 meters above the belt surface, to prevent material splashing from damaging the sensor.

[0058] S602: Combining the material flow rate and the spatial temperature gradient of the material target temperature within the furnace section, quantitatively extract the heat required to meet the specified heating trajectory of the material.

[0059] Material target temperature trajectory The unit is degrees Celsius, set by production process requirements, and obtained in real time from the MES system. It characterizes the temperature that the material should reach at axial position x and time t. Spatial temperature gradient. The unit is degrees Celsius per meter, representing the rate of change of the target temperature of the material along the axial direction. For discrete temperature control points, the central difference method is used for calculation. Heating requires heat. The unit is kilowatt, which represents the amount of heat required to heat the material from the inlet temperature to the outlet temperature of furnace section i. The calculation formula is: in The specific heat capacity of the material is expressed in joules per kilogram per degree Celsius. It is measured by differential scanning calorimetry, and is measured once for each batch of material.

[0060] S603: Based on the heat dissipation characteristics of the furnace structure and the temperature difference between the inside and outside, extract the heat dissipation compensation amount to remove the interference of cold end temperature in the environment.

[0061] Furnace body heat dissipation coefficient The unit is watts per square meter in degrees Celsius, characterizing the heat dissipation rate per unit area of ​​the furnace body, obtained through steady-state calibration experiments. Under steady-state operating conditions with no material and no fluidizing air, the input thermal power and the difference between the bed temperature and the ambient temperature are measured and calculated. Furnace body heat dissipation area The unit is square meters, calculated from the surface area of ​​the furnace wall in furnace section i, including the surface areas of the top, bottom, and sides. Ambient temperature. The unit is degrees Celsius, and the data is collected by a temperature sensor installed at least 5 meters away from the furnace body inside the workshop to avoid the influence of furnace heat radiation. Heat dissipation compensation. The unit is kilowatt, representing the amount of heat required to compensate for the furnace body's heat dissipation to the environment. The calculation formula is: in The target bed temperature for furnace section i, in degrees Celsius, is set according to production process requirements and is obtained in real time from the MES system.

[0062] The dynamic correction method for the furnace body heat dissipation coefficient is as follows: When the ambient temperature deviates from the calibration ambient temperature by more than 5 degrees Celsius, the furnace body heat dissipation coefficient is corrected. The correction formula is as follows: ,in To calibrate the ambient temperature The furnace body heat dissipation coefficient is 0.002, which is a temperature correction coefficient.

[0063] S604: Based on the rate of change of the target bed temperature over time and the equivalent heat capacity of the system, extract the heat storage requirement that characterizes the transformation process of the furnace body's own heat storage state.

[0064] System equivalent heat capacity The unit is joules per degree Celsius, characterizing the total heat storage capacity of the furnace structure and bed material, obtained through dynamic calibration experiments. Under conditions of no material and no fluidizing air, the furnace body is heated at constant power, and the rate of change in bed temperature is measured and calculated. Rate of change of target bed temperature over time The unit is degrees Celsius per second, representing the dynamic rate of change of the target bed temperature, calculated using the central difference method. The unit for thermal storage heat demand is kilowatts, representing the heat required to change the thermal storage state of the furnace body, calculated using the following formula: When the target bed temperature rises, the heat storage requirement is positive, and additional heat needs to be supplied to heat the furnace body and bed material; when the target bed temperature decreases, the heat storage requirement is negative, and the furnace body and bed material will release stored heat.

[0065] The experimental calibration procedure for the system's equivalent heat capacity is as follows: Under conditions of no material and no fluidizing air, the furnace body is heated with a constant power P, and the bed temperature T(t) is recorded every minute. Heating is stopped when the bed temperature rises above 50 degrees Celsius. The bed temperature change rate is calculated based on the bed temperature change curve over time. The formula for calculating the equivalent heat capacity of the system is: The average value of the calculation results under different heating powers is taken as the final calibration value.

[0066] S605: The initial heat requirement for this furnace section is composed of the heat required for polymerization heating, heat dissipation compensation, and heat storage.

[0067] Initial heat requirement The unit is kilowatts, which is the total heat load demand of furnace section i without considering time lag. The calculation formula is: S606: Based on the inherent hysteresis characteristics of water delivery delay and bed-side thermal inertia in the thermal phase deviation, the initial heat demand is forward-shifted and extrapolated in the time domain to generate the compensation heat demand to eliminate the delay in pipeline distribution.

[0068] First, a model for predicting the initial heat demand is established, with the prediction time range T set to 1.5 times the system's maximum lag time. The system's maximum lag time is the time required to predict the initial heat demand across all furnace sections. The maximum value is typically 10 to 30 minutes. Production plan data for the next time period T is obtained from the MES system, including parameters such as material flow rate, target material temperature trajectory, and target bed temperature. For time ranges not covered by the production plan, linear extrapolation is used for prediction, based on linear fitting extrapolation of historical data from the most recent 10 control cycles. Ambient temperature is obtained from hourly weather forecast data provided by the local meteorological department, or based on time series prediction using historical data from the most recent 24 hours. These future parameters are substituted into the initial heat demand calculation formula to obtain the initial heat demand at future times. ,in Simultaneously, a historical compensation heat demand database is established, storing all compensation heat demand values ​​within the past time period T, with the sampling interval consistent with the control cycle. The initial heat demand is shifted forward on the time axis. Time, required to compensate for the heat The unit is kilowatt, and the calculation formula is: The compensation for heat demand takes into account the heat load demand at future moments, so that the heat supply and material demand are synchronized in time.

[0069] S701: Based on the physical relationship between changes in flow rate and changes in heat transfer characteristics, construct a heat residual formula that includes the predicted flow rate, predicted heat transfer rate, and predicted temperature difference.

[0070] Changes in flow rate alter the water velocity within the pipe, thus affecting the water convection coefficient and overall heat transfer coefficient, as well as the inlet and outlet water temperature difference. Therefore, the overall heat transfer coefficient... Logarithmic mean temperature difference All are predicted segment traffic The function. Constructing the heat residual formula. This represents the difference between the predicted thermal power and the compensation heat requirement, calculated using the following formula: in and Based on the predicted segment flow The total heat transfer coefficient and logarithmic mean temperature difference are calculated according to methods S401 to S403.

[0071] S702: The nonlinear iterative algorithm is used to approximate the convergence of the heat residual, eliminating the coupling interference between the flow rate and the change of heat transfer efficiency, and analyzing the target flow rate of the furnace section required to maintain the heat compensation.

[0072] The above nonlinear equations are solved using Newton's iterative method. Given the initial predicted flow rate... The actual flow rate from the previous control cycle is used as the initial value. For the k-th iteration, the heat residual is calculated. and the derivative of residual with respect to flow rate The derivative is calculated using numerical differentiation, i.e. ,in For small flow increments, the value is 1e-4 cubic meters per second. Update the predicted flow: When the iteration satisfies the convergence condition The iteration stops when the predicted flow rate reaches a certain value (e.g., kilowatts) or the maximum number of iterations (50). The converged predicted flow rate is the target flow rate for the furnace section. : If the iteration does not converge, the target flow rate of the previous cycle is multiplied by the compensation coefficient, which is the ratio of the current compensation heat requirement to the compensation heat requirement of the previous cycle.

[0073] S703: Based on the compensation heat demand and branch coverage, evaluate and calculate the total scalar heat demand of each furnace section.

[0074] For each water circulation branch e, calculate the compensated heat demand of all furnace sections multiplied by the sum of the corresponding branch coverage to obtain the scalar heat demand of that branch, in kilowatts. The calculation formula is as follows: , where n is the total number of furnace sections.

[0075] S704: The heat carrying capacity of a unit fluid is extracted by combining the specific heat capacity of water and the predicted temperature difference between inlet and outlet water.

[0076] Predict the outlet water temperature of the branch. The unit is Celsius, calculated using the principle of energy conservation. The characteristic unit of heat carrying capacity of a unit fluid is kilojoules per cubic meter, and the calculation formula is: It represents the amount of heat that a unit volume of circulating water can carry.

[0077] The feedback correction method for predicting the outlet water temperature of the branch is as follows: Calculate the deviation between the predicted outlet water temperature and the actual outlet water temperature from the previous control cycle. Multiply this deviation by a correction factor of 0.1 and add it to the predicted water temperature for the current period. The correction formula is as follows: ,in The original predicted water temperature is obtained based on energy conservation calculations.

[0078] S705: Based on the scalar total heat demand and heat carrying capacity characteristics, physical dimension conversion is performed to obtain the target flow rate that is allocated to each water circulation branch and satisfies the space advance heat supply.

[0079] Dividing the total scalar heat demand of the branch by the heat carrying capacity characteristic of the unit fluid yields the target flow rate of the branch. The unit is cubic meters per second, and the calculation formula is: The target flow rate of the branch line meets the space-advanced heat supply needs of each furnace section, providing a flow rate target for subsequent pump frequency optimization.

[0080] S801: Under the premise that the branch flow is conserved by the node continuity law and the head loss constraint formed by the pipeline correlation array, node pressure array, variable frequency pump head, branch flow resistance and branch water inertia ensures the water pressure balance of the whole network, the calculation is performed with the goal of minimizing the total power of variable frequency pumps, the interference of redundant pressure boosting of single pumps is removed, and the optimal pump frequency is obtained analytically.

[0081] To avoid an unsolvable optimization problem, a soft-constraint optimization problem is constructed, transforming the hard flow constraint into a soft constraint. The objective function is to minimize the total power consumption of the variable frequency pump plus a penalty term for flow deviation. Where M represents the total number of variable frequency pumps. The penalty coefficient is set to 1e7 to ensure that the flow deviation is minimized. and Let these be the positive and negative slack variables of the flow rate, respectively, satisfying... ,and , The constraints include: 1. Nodal continuity law: ; 2. Head loss constraint: Where A is the pipe network correlation matrix, K is the branch flow resistance diagonal matrix, and I is the branch water inertia diagonal matrix. It represents the Hadamardi (or Hadama) stack; 3. Pump frequency constraint: ,in This is the minimum operating frequency of the variable frequency pump, and is set to 10 Hz. The maximum operating frequency is set to 50 Hz. 4. Flow constraints: ; This optimization problem is solved using sequential quadratic programming, employing the IPOPT open-source solver or MATLAB's fmincon function. The solver parameters are set to a maximum of 100 iterations and a convergence threshold of 1e-6. If the solver returns no solution, a penalty coefficient is applied. Reduce the frequency by one order of magnitude and try again; if there is still no solution, use the optimal pump frequency of the previous control cycle as the initial optimal pump frequency of the current cycle.

[0082] Rate of change of flow in head loss constraints The central difference method is used for calculation, i.e. ,in This is the control cycle. For the first calculation step at the beginning of the control cycle, the forward difference method is used for calculation, i.e. .

[0083] S802: Compare the actual thermal power with the actual thermal deviation of the unknown heat dissipation of the characterization system for compensating for the heat required for extraction.

[0084] First, calculate the total lag time at the current moment. Query from the historical compensation heat demand database Caloric value required for time-based compensation If the current value exceeds the database storage range, the earliest historical value in the database will be used. Actual thermal deviation The unit is kilowatt, representing the difference between the actual supplied heat power and the corresponding compensation heat demand. It reflects unmodeled heat dissipation and external disturbances in the system. The calculation formula is: S803: Derive the influence matrix of frequency perturbation of the variable frequency pump on the thermal power output of each furnace section, and quantify the sensitivity of hydraulic and thermal coupling across branches.

[0085] Influence Matrix It is an n×M matrix, where the elements are... This represents the change in thermal power of the i-th furnace section when the frequency of the j-th pump changes by 1 Hz, expressed in kilowatts per Hz. An influence matrix is ​​obtained using a system identification method. During stable system operation, small frequency perturbations are sequentially applied to each variable frequency pump. Hertz, with a duration of twice the system's maximum lag time. During the perturbation, the actual thermal power changes of each furnace section are simultaneously acquired. Calculations yielded To improve identification accuracy, the identification experiment was repeated three times and the average value was taken. The influence matrix was updated every 24 hours, or immediately when there were significant changes in operating conditions such as changes in material type or bed material replacement.

[0086] The operating conditions affecting matrix identification are as follows: Identification experiments must be conducted under steady-state system conditions. During identification, parameters such as material flow rate, material type, fluidizing air flow rate, and ambient temperature must remain stable, with fluctuations not exceeding ±5% of the rated values. Identification experiments should be avoided during shift changes, equipment start-ups and shutdowns, or changes in operating conditions.

[0087] S804: The influence matrix is ​​used to perform feedback decoupling calculation on the actual thermal deviation to obtain the frequency correction amount to offset the interference of unknown pipeline network.

[0088] The actual thermal deviation is used to construct the residual vector. The frequency correction factor is solved using the regularized least squares method. To avoid matrix singularity and improve solution stability, the calculation formula is as follows: in is the regularization coefficient, taking a value of 1e-2, where I is the M-order identity matrix. Frequency correction amount. The unit is Hertz, used to compensate for thermal deviations caused by unknown heat dissipation in the system and interference from the pipeline network.

[0089] S805: The frequency correction is added to the optimal pump frequency to obtain the final pump frequency.

[0090] Final pump frequency The unit is Hertz, which is the operating frequency of each variable frequency pump in the next control cycle. The calculation formula is: in To control the cycle, a value of 10 seconds is used. The final pump frequency is limited to ensure it remains within the operating frequency range of the variable frequency pump. .

[0091] S806: Synchronously correct bed heat transfer parameters based on sensitivity evolution characteristics to eliminate long-term aging drift of fluidized beds.

[0092] Constructing the Jacobian matrix of bed heat transfer parameters , of which elements Based on the relationship between bed heat transfer coefficient and heat power And the logarithmic linear correlation of the bed heat transfer coefficient, we can obtain ,in For regression vectors The k-th element. The bed heat transfer parameter vector is updated using a recursive least squares method with a forgetting factor. The calculation formula is: in This is the regularization coefficient, with a value of 1e-2. By synchronously correcting the bed heat transfer parameters, the drift in heat transfer characteristics caused by long-term factors such as bed material aging and coking is eliminated.

[0093] S807: Sends the final pump frequency to the inverters of each variable frequency pump to perform frequency adjustment.

[0094] The system uses the Modbus TCP protocol via industrial Ethernet to send the final pump frequency command to the inverters of each variable frequency pump, with the communication cycle matching the control cycle. Upon receiving the frequency command, the inverter adjusts the motor speed according to a preset acceleration / deceleration time of 5 seconds to avoid water pressure surges and pipeline vibration. The inverter provides real-time feedback of the actual operating frequency and current to the control system for status monitoring and fault diagnosis.

[0095] The adjustment method for the acceleration and deceleration time of the frequency converter is as follows: When the total pipeline length exceeds 100 meters or the total system water capacity exceeds 10 cubic meters, extend the acceleration and deceleration time to 10 seconds; when the total pipeline length exceeds 200 meters or the total system water capacity exceeds 20 cubic meters, extend the acceleration and deceleration time to 15 seconds. The adjustment of the acceleration and deceleration time should be carried out gradually, with each adjustment not exceeding 5 seconds. After adjustment, observe the pipeline pressure fluctuation to ensure there is no obvious water hammer effect.

[0096] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. An intelligent variable frequency collaborative control system for a tunnel fluidized bed furnace water circulation heating network, applied to a heating system comprising a tunnel furnace body divided into multiple furnace sections along the material movement direction, a fluidized bed, a heating network containing multiple water circulation branches and supply and return water headers, and multiple variable frequency pumps, characterized in that, Configured for execution: A mapping model reflecting the axial thermal interaction between the tunnel furnace and the heating pipe network is constructed, and the actual thermal power of each furnace section is analyzed to eliminate the spatial distribution differences of the pipe network. Based on the pipeline network operation data, the hydraulic parameters of the water circulation branch are identified, and a nodal continuity law characterizing flow conservation and pipeline head loss constraints are constructed to isolate fluid interaction coupling interference. Based on the actual thermal power inversion, the bed-side thermal resistance characterizing the gas-solid phase heat transfer characteristics in each of the furnace sections is obtained. The water delivery delay, bed-side thermal inertia, and material heat demand arrival time of each furnace section are quantified and combined to form a thermal phase deviation that characterizes water transport, pipe wall heat transfer, and material displacement time misalignment. Based on the thermal phase deviation, the initial heat demand characterizing the static heating load of the furnace section is shifted forward along the time axis to generate a compensating heat demand to make up for the time misalignment. Construct a heat residual formula and solve iteratively; based on the bed-side thermal resistance, back-calculate the target flow rate of each water circulation branch from the required heat for compensation. Under the premise of satisfying the node continuity law and the head loss constraint, the optimal pump frequency is solved with the goal of minimizing the total power consumption of the variable frequency pump. The actual thermal deviation, which represents the difference between the actual thermal power and the compensation heat required, is extracted and used for closed-loop feedback calculation to correct the unknown external disturbances of the system and obtain the final pump frequency.

2. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 1, characterized in that, Calculating the actual thermal power of each furnace section includes: Determine the corresponding furnace section interval for each of the aforementioned furnace sections within the tunnel furnace body; The effective heat exchange ratio of the water circulation branch within the furnace section is analyzed, and the branch coverage reflecting the actual thermal radiation coverage area of ​​the pipeline is extracted. The real-time flow rate, inlet water temperature, and outlet water temperature of the water circulation branch are collected simultaneously. The enthalpy difference of the water path is extracted by combining the density of circulating water and the specific heat capacity of water. The enthalpy difference of the water path, the coverage of the branch and the real-time flow are correlated and converted. The interference of flow segments with no heat exchange effect is eliminated, and the actual heat power reflecting the real net heat energy obtained is obtained.

3. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 2, characterized in that, Identifying the hydraulic parameters of each of the water circulation branches and constructing the nodal continuity law includes: The branch pressure difference of the water circulation branch is collected and the branch water inertia, which characterizes the hysteresis of the fluid momentum state, is extracted. Based on the characteristics of real-time operation data, the unsteady pressure fluctuation interference caused by the inertia of the branch water is removed, and the branch flow resistance, which characterizes the effects of flow along the path and local obstruction, is extracted. Construct a network association array that maps the network topology, and couple it with an array composed of the real-time flow of each water circulation branch to establish the node continuity law; Based on the pump head coefficient and pump resistance coefficient of the actual operating curve of the variable frequency pump, the variable frequency pump head that evolves with the current frequency and the real-time flow rate is extracted. The power of the variable frequency pump is obtained by combining the pump head, the physical properties of the circulating water and the pump efficiency through a mapping conversion. The branch flow resistance, the variable frequency pump head, and the node continuity law together constitute the hydraulic parameters.

4. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 3, characterized in that, Based on the actual thermal power, the bed-side thermal resistance of each furnace section is calculated, including: Extract the equivalent inlet and outlet water temperatures mapped to the furnace section and based on the branch inlet water temperature and the branch outlet water temperature, calculate the logarithmic temperature difference between the equivalent inlet and outlet water temperatures and the fluidized bed temperature, and characterize the driving force of heat exchange temperature inside and outside the furnace. By combining the logarithmic temperature difference, the heat exchange area of ​​the furnace section, and the actual thermal power, the total heat transfer coefficient, which reflects the comprehensive heat exchange intensity inside and outside the tube, is calculated. Based on the characteristics of circulating water flow state, Reynolds number and Prandtl number are constructed, and water convection coefficient, which characterizes the forced convection heat transfer capacity of fluid in the pipe, is extracted. The water convection coefficient is removed from the total heat transfer coefficient, and the conductive impedance of the heating tube body is eliminated, thus separating the bed-side thermal resistance, which specifically characterizes the heat transfer barrier between the gas and solid phases. The bed-side thermal resistance is converted into the bed heat transfer coefficient. By combining the particle density, solid flux, apparent gas velocity and fluidized bed temperature of the fluidized bed, the gas-solid state correlation is performed, and the bed heat transfer parameters that eliminate the interference of bed material state fluctuations are calibrated.

5. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 4, characterized in that, Calculating the thermal phase deviation includes: Based on the effective water capacity of each water circulation branch covering the furnace section, the branch coverage and the real-time flow rate, the water delivery hysteresis, which characterizes the physical time required for the water in the pipeline to flow to the heat exchange section, is extracted. By combining the equivalent heat capacity of the tube wall and the bed material and the overall heat transfer coefficient, the bed-side thermal inertia, which characterizes the time delay in the thermal response that causes heat to penetrate the tube wall and cause the bed temperature to appear, is extracted. Based on the material movement direction and material speed, the time taken for the material to reach the spatial position of the furnace section is taken as the time required for heat to arrive. By integrating the water delivery lag and the bedside thermal inertia, and performing time axis alignment analysis with the arrival of heat demand, the thermal phase deviation quantity, which reflects the degree of time misalignment between the heat supply end and the material demand end, is generated.

6. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 5, characterized in that, Generating the compensated heat demand includes: Material flow rate, which characterizes the continuous throughput of materials, is extracted based on the physicochemical properties and loading status of the materials. By combining the material flow rate and the spatial temperature gradient of the target material temperature within the furnace section, the heat required to meet the specified heating trajectory of the material is quantitatively extracted. Based on the heat dissipation characteristics of the furnace structure and the temperature difference between the inside and outside, the heat dissipation compensation amount for the cold end temperature interference of the stripping environment is extracted. Based on the rate of change of the target bed temperature over time and the equivalent heat capacity of the system, the heat storage requirement that characterizes the transformation process of the furnace body's own heat storage state is extracted; The initial heat demand of the furnace section is formed by combining the heat required for heating, the heat dissipation compensation, and the heat storage. Based on the inherent hysteresis characteristics of the water delivery delay and the bed-side thermal inertia in the thermal phase deviation, the initial heat demand is forward-shifted and extrapolated in the time domain to generate the compensation heat demand to eliminate the delay in pipeline distribution.

7. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 6, characterized in that, Reverse calculation of the target traffic includes: Based on the physical relationship between changes in flow rate and changes in heat transfer characteristics, the heat residual formula is constructed, which includes the predicted flow rate, predicted heat transfer rate, and predicted temperature difference. The heat residual is approximated and converged using a nonlinear iterative algorithm to eliminate the coupling interference between the flow rate and the heat transfer efficiency, and to analyze the target flow rate of the furnace section required to maintain the heat compensation. Based on the aforementioned compensation heat demand and the aforementioned branch coverage, the total scalar heat demand of each of the aforementioned furnace sections is evaluated and calculated; The heat carrying capacity of a unit fluid is extracted by combining the specific heat capacity of water and the predicted temperature difference between inlet and outlet water. Based on the scalar total heat demand and the heat carrying capacity characteristics, a physical dimension conversion is performed to obtain the target flow rate that is allocated to each of the water circulation branches and satisfies the spatial advance heat supply.

8. The intelligent variable frequency collaborative control system for the water circulation heating pipeline network of the tunnel fluidized bed furnace according to claim 7, characterized in that, Solving for the optimal pump frequency and the final pump frequency includes: Under the premise that the node continuity law ensures the conservation of branch flow and the head loss constraint formed by the pipeline network association array, node pressure array, variable frequency pump head, branch flow resistance and branch water inertia ensures the water pressure balance of the whole network, the calculation is performed with the goal of minimizing the total power of the variable frequency pumps, the redundancy of single pumps is removed and the optimal pump frequency is obtained analytically. Compare the actual thermal power with the actual thermal deviation of the unknown heat dissipation of the compensation heat extraction characterization system; The influence matrix of the frequency perturbation of the variable frequency pump on the thermal power output of each furnace section is derived, and the sensitivity of hydraulic and thermal coupling across branches is quantified. The influence matrix is ​​used to perform feedback decoupling calculation on the actual thermal deviation to obtain the frequency correction amount to offset the unknown pipeline interference; The frequency correction is superimposed on the optimal pump frequency to obtain the final pump frequency, and the bed heat transfer parameters are simultaneously corrected based on the sensitivity evolution characteristics to eliminate long-term aging drift of the fluidized bed.

Images