Intelligent temperature field prediction and furnace fine temperature control system

Abstract

The application provides a kind of intelligent temperature field prediction and furnace fine temperature control system, it is related to industrial furnace heat treatment control technical field, system includes load support heat flux acquisition component, agent core temperature with furnace probe component, in-furnace environment and execution state acquisition component, edge data processing unit, load endothermic fingerprint generation unit, thermal inertia compensation temperature control unit, process completion degree determination unit and safety protection and data traceability unit.System obtains the real heat absorption state of load by supporting path heat flow, heating power balance, empty furnace background deduction and furnace agent block core temperature correction;According to load net heat absorption power, cumulative net heat absorption, corrected agent core temperature, temperature hysteresis, heat absorption decay ratio, generate load heat absorption fingerprint;Then use load heat absorption fingerprint to correct temperature rising, soaking and holding control output, and heat absorption decay and core temperature stability are used to determine process completion degree.

Description

Technical Field

[0001] This invention relates to the field of industrial furnace heat treatment control technology, specifically to an intelligent temperature field prediction and refined furnace temperature control system. Background Technology

[0002] Industrial furnace temperature control systems typically use furnace atmosphere temperature or thermocouple temperature near the furnace wall as feedback, and achieve formula curve tracking through zoned proportional-integral-derivative (PID) control. This type of control can bring the furnace measuring point temperature close to the set value, but the furnace measuring point temperature cannot directly represent the workpiece core temperature, the heat absorption state of the material tray, or the cumulative heat absorption under the furnace load. For heat treatment processes involving thick-walled workpieces, multi-layer material trays, stacked loading, fixture obstruction, and significant batch quality variations, even when the furnace atmosphere temperature has reached the formula set value, there may still be a temperature lag in the workpiece core. For lightly loaded furnaces, a fixed formula holding time leads to excessive holding and energy waste; for heavily loaded furnaces, a fixed formula holding time results in underheating of the core and decreased process consistency.

[0003] Existing furnace control schemes improve temperature uniformity by increasing in-furnace temperature measurement points, establishing furnace temperature field models, predicting furnace temperature distribution, and optimizing control. These schemes primarily control the furnace space temperature, making it difficult to directly reflect the actual heat absorption process of the workpiece or pallet load. Purely data-driven prediction schemes rely on extensive historical furnace data; their applicability decreases when the loading method, pallet structure, workpiece material, furnace lining condition, and air duct condition change. If the patent text only describes data acquisition, temperature prediction, and optimized control without specifying sensor installation methods, data sources, calibration processes, data processing formulas, control constraints, safety logic, and complete application examples, those skilled in the art will find it difficult to implement directly, posing a risk of insufficient disclosure during the examination stage.

[0004] Therefore, an adaptive temperature control system for the furnace body is needed that shifts the control basis from the temperature measured at the furnace chamber to the heat absorption state of the load. This system requires clearly defined hardware structure, installation layout, data acquisition, calibration methods, data processing, fingerprint generation, control output, completion determination, and anomaly protection, enabling those skilled in the art to directly configure, install, debug, and operate it according to the application documents. Summary of the Invention

[0005] Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an intelligent temperature field prediction and furnace body fine temperature control system, which solves the following problems:

[0007] First, the problem is that the furnace atmosphere temperature reaches the set value, but the core of the workpiece does not meet the process requirements.

[0008] Second, the problem of mismatch in fixed formula caused by different furnace loading, different material tray arrangements and different workpiece thermal inertia;

[0009] Third, the existing furnace temperature control system lacks a closed-loop system that accurately reflects the actual heat absorption state of the load.

[0010] Fourth, the existing solutions lack sufficient description of the data acquisition, calibration, processing, and control processes;

[0011] Fifth, the problem lies in relying on a fixed heat treatment time rather than the actual degree of heat treatment completion to determine the completion of the process.

[0012] Technical solution

[0013] To achieve the above objectives, the present invention provides the following technical solution: an intelligent temperature field prediction and refined furnace temperature control system, comprising: a load-supported heat flux acquisition component, a proxy core temperature probe component, a furnace environment and execution status acquisition component, an edge data processing unit, a load heat absorption fingerprint generation unit, a thermal inertia compensation temperature control unit, a process completion determination unit, and a safety protection and data traceability unit; the load-supported heat flux acquisition component includes four to eight supporting heat flow paths, each supporting heat flow path including a high-temperature resistant support, a heat flow metering layer, a hot end temperature measuring point, a cold end temperature measuring point, a heat insulation and flow limiting layer, and a contact layer. The system includes a status detection component and shielded leads. The load-bearing heat flux acquisition component is configured to collect heat flow data transmitted by the load of the material tray through the support path. The proxy core temperature probe component includes a proxy block, a core temperature sensor, a surface temperature sensor, and a proxy block identification mark. The proxy block is placed at the most unfavorable heating position of the load of the material tray, and the proxy core temperature probe component is configured to collect the temperature data of the proxy block, which represents the lag in heating inside the workpiece. The furnace environment and execution status acquisition component is configured to collect furnace atmosphere temperature, furnace ambient temperature, circulating air velocity, furnace pressure, furnace door status, and actual input power of each heating zone. The edge data processing unit is configured to perform unified time base alignment, outlier detection, filtering, validity marking, unit conversion, support path heat flux conversion, empty furnace background subtraction, power balance calculation, load net heat absorption power fusion, and cumulative net heat absorption integral on the collected data. The load heat absorption fingerprint generation unit is configured to generate a load heat absorption fingerprint based on the load net heat absorption power, cumulative net heat absorption, corrected proxy core temperature, temperature hysteresis, heat absorption power change slope, heat absorption attenuation ratio, load heat capacity estimate, and support heat flux reliability. The thermal inertia compensation temperature control unit is configured to perform the following based on the load... The heating zone control output of the heat absorption fingerprint correction heating section, heat soaking section and heat preservation section is configured; the process completion determination unit is configured to determine the process completion status based on the slope of heat absorption power change, heat absorption attenuation ratio, temperature hysteresis, cumulative net heat absorption and continuous confirmation time; the safety protection and data traceability unit is configured to trigger protection actions when there are abnormalities in the heat flow channel, abnormalities in the proxy core temperature probe, abnormal opening of the furnace door, abnormal heating power feedback, temperature exceeding the limit, unreachable control output and abnormal data recording, and save the original furnace data, alignment data, channel quality mark, load heat absorption fingerprint, control command, alarm event and completion determination result.

[0014] Preferably, the heat flow metering layer is a high-temperature resistant heat-conducting block with a calibrated thermal conductivity. The hot end temperature measuring points and the cold end temperature measuring points are arranged at intervals along the main heat flow direction of the support path. The edge data processing unit calculates the heat flow density and heat flow power of a single support heat flow path according to the hot end temperature, cold end temperature, calibrated thermal conductivity of the heat-conducting block, measuring point spacing and effective conversion area.

[0015] Preferably, the contact status detection device includes a load switch installed below the high-temperature resistant support and a tray identification mark installed on the tray. The edge data processing unit generates a valid contact mark for the support path based on the load switch status, the tray identification mark, the temperature difference change rate of the support path, and the heat flow mutation of the support path. Only the support heat flow paths with valid contact marks and valid channel quality marks are included in the calculation of the net heat absorption power of the load.

[0016] Preferably, the furnace-mounted proxy block is a cylindrical proxy block made of the same material as the target workpiece. The diameter of the cylindrical proxy block is 0.8 to 1.2 times the thickness of the maximum thermal resistance path of the target workpiece. The core temperature sensor is located at the geometric center of the cylindrical proxy block, and the surface temperature sensor is located 2 mm to 5 mm inward from the outer surface of the cylindrical proxy block. The proxy block identification mark stores the proxy block number, material number, size number, and proxy core temperature correction coefficient number.

[0017] Preferably, the edge data processing unit performs time alignment on the data of different acquisition channels according to a unified sampling period, which is 0.5s to 2s; when a certain acquisition channel has two original sample values ​​before and after the unified sampling time, the edge data processing unit uses linear interpolation to obtain the alignment value of the acquisition channel at the unified sampling time.

[0018] Preferably, the edge data processing unit performs a confidence-weighted fusion of the heat absorption power of the load converted from the support side and the heat absorption power of the load converted from the power balance side to obtain the net heat absorption power of the load; the heat absorption power of the load converted from the support side is obtained by summing the heat flow power of the effective support heat flow path; the heat absorption power of the load converted from the power balance side is calculated by the actual input power of the heating zone, the efficiency of the heating zone, the background power consumption of the empty furnace, and the change in the heat storage of the furnace body; the fusion weight of the confidence-weighted fusion is calculated by the uncertainty of the power estimation on the support side and the uncertainty of the power estimation on the power balance side.

[0019] Preferably, the load heat absorption fingerprint generated by the load heat absorption fingerprint generation unit is an eight-dimensional vector, and the components of the eight-dimensional vector are, in order, the net heat absorption power of the load, the cumulative net heat absorption, the corrected proxy core temperature, the temperature hysteresis, the slope of the change in heat absorption power, the heat absorption attenuation ratio, the estimated value of the load heat capacity, and the reliability of the support heat flow.

[0020] Preferably, the thermal inertia compensation temperature control unit superimposes a load cumulative heat absorption compensation term, a temperature lag compensation term, and a heat absorption power change slope suppression term on the basic control output to obtain the original compensation output of the heating zone, and then sends the original compensation output of the heating zone to the heating zone power regulator after applying amplitude and rate of change limits.

[0021] Preferably, the process completion determination unit outputs the completion status only when the following conditions are met simultaneously and continuously for a continuous confirmation time: the absolute value of the slope of the net heat absorption power change is not greater than the slope threshold, the heat absorption attenuation ratio is not greater than the attenuation ratio threshold, the temperature hysteresis is not greater than the hysteresis threshold, the cumulative net heat absorption is not less than the target cumulative heat absorption, the support heat flow reliability is not less than the reliability threshold, and no abnormalities occur in the heat flow channel, the proxy core temperature probe, the furnace door, the heating power feedback, or the temperature exceeding the limit.

[0022] Preferably, a temperature control method corresponding to an intelligent temperature field prediction and refined furnace temperature control system includes the following steps:

[0023] Sp1, four to eight supporting heat flow paths are installed at the material tray support position in the furnace body, and a furnace agent block is placed at the position where the material tray load is most unfavorable for heating. An isolation acquisition module, an energy metering module, an edge controller and a safety relay are installed in the furnace body electrical control cabinet.

[0024] Sp2, establish a channel configuration table, which records the channel number, data type, installation location, unit, range, sampling period, communication address, filter coefficient and alarm threshold;

[0025] Sp3 performs empty furnace background calibration, support heat flow path calibration, agent block core temperature correction calibration, heating zone power feedback calibration, and safety interlock test, and writes the calibration parameters into the edge data processing unit;

[0026] Sp4 collects the following data in each uniform sampling period: hot end temperature, cold end temperature, core temperature of the proxy block, surface temperature of the proxy block, furnace atmosphere temperature, external ambient temperature, circulating wind speed, furnace pressure, furnace door status, actual input power of the heating zone, and control output of the heating zone.

[0027] SP5 performs time alignment, outlier detection, filtering, validity marking, and unit conversion on the collected data, while saving both the original data and the processed data.

[0028] Sp6 calculates the equivalent load heat absorption power on the support side based on the support heat flow path data, calculates the equivalent load heat absorption power on the power balance side based on the actual input power of the heating zone, the background power consumption of the empty furnace and the change in the furnace body heat storage, and performs a confidence-weighted fusion of the two to obtain the net load heat absorption power.

[0029] Sp7, the net heat absorption power of the load is integrated over time to obtain the cumulative net heat absorption. The corrected agent core temperature is calculated based on the agent block temperature data and the agent core temperature correction coefficient. The load heat absorption fingerprint is generated based on the net heat absorption power of the load, the cumulative net heat absorption, the corrected agent core temperature, the temperature hysteresis, the slope of the heat absorption power change, the heat absorption attenuation ratio, the estimated value of the load heat capacity and the reliability of the support heat flow.

[0030] Sp8 calculates the heating zone compensation control output based on the load heat absorption fingerprint, and issues control commands within the limits of amplitude, rate of change, furnace door status, hardware temperature and communication health.

[0031] Sp9 determines the process completion rate based on the load heat absorption fingerprint and continuous confirmation time, and writes the original data, processing data, fingerprint data, control output, alarm records and completion determination results of the furnace batch into the traceability database.

[0032] Beneficial effects

[0033] This invention provides an intelligent temperature field prediction and precise furnace temperature control system. It has the following beneficial effects:

[0034] This invention directly collects and calculates the load heat absorption state, reducing the judgment deviation between furnace temperature and workpiece core temperature; it improves adaptability under different furnace loading and material tray layouts by integrating support path heat flow, power balance, and proxy block core temperature; it avoids mistaking furnace body heat dissipation and heat storage as load heat absorption by subtracting empty furnace background; it identifies the true thermal state during heating, homogenization, and heat preservation processes by identifying load heat absorption fingerprints; it reduces over-insulation under light load and under-insulation under heavy load by completing judgments under complete channel quality marking and furnace traceability data; and it improves engineering maintenance capabilities and review feasibility by using complete channel quality marking and furnace traceability data. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0036] Figure 2 This is a schematic diagram of the load-supported heat flux acquisition component structure of the present invention;

[0037] Figure 3 This is a schematic diagram of the core temperature probe assembly of the present invention.

[0038] Figure 4 This is a schematic diagram of the data processing and control flow of the present invention;

[0039] Figure 5 This is a screenshot of the system's main page for this invention;

[0040] Figure 6 This is a screenshot of the system functions of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation Example 1:

[0043] like Figures 1 to 6 As shown, an intelligent temperature field prediction and refined furnace temperature control system is presented. The system collects heat flow data transmitted by the load through the material tray support position via four to eight support heat flow paths. It also collects representative core heating lag data through a furnace-mounted agent block. An empty furnace background and power balance model is established using heating zone power, furnace atmosphere temperature, ambient temperature, circulating wind speed, furnace pressure, and furnace door status. An edge data processing unit performs unified time alignment, channel validity determination, filtering, support-side heat absorption power calculation, power balance-side heat absorption power calculation, and confidence-weighted fusion on the multi-source data to obtain the load's net heat absorption power and cumulative net heat absorption. A load heat absorption fingerprint generation unit generates an eight-dimensional load heat absorption fingerprint. A thermal inertia compensation temperature control unit corrects the control output of each heating zone based on the load heat absorption fingerprint. A process completion determination unit outputs the completion status based on the heat absorption slope, heat absorption attenuation ratio, temperature lag, cumulative net heat absorption, and continuous confirmation time.

[0044] The system is divided into four layers according to engineering implementation: field sensing layer, data acquisition and isolation layer, edge computing layer, control execution layer, completion determination layer, and safety traceability layer. The field sensing layer includes the support heat flow path, the core temperature channel of the furnace-mounted agent block, the furnace atmosphere temperature channel, the external ambient temperature channel, the circulating air velocity channel, the furnace pressure channel, the furnace door status channel, and the heating power channel. The data acquisition and isolation layer includes thermocouple acquisition modules, analog acquisition modules, digital acquisition modules, energy metering modules, isolation communication modules, and a unified clock module. The edge computing layer includes data alignment, filtering, channel quality evaluation, support-side heat flow conversion, empty furnace background subtraction, power balance calculation, data fusion, cumulative heat absorption integration, load heat absorption fingerprint generation, and control compensation calculation. The control execution layer includes heating zone power regulators, circulating fans, exhaust valves, furnace door interlocks, and safety relays. The completion determination layer includes core temperature stability determination, heat absorption attenuation determination, cumulative heat absorption determination, continuous confirmation timing, and completion status latching. The safety traceability layer includes hardware over-temperature protection, furnace door interlock, sensor failure degradation, communication anomaly protection, alarm recording, furnace data storage, and report export.

[0045] The load-supported heat flux acquisition components are arranged along the tray support path. Each supported heat flux path consists of a high-temperature resistant support, a heat flux metering layer, hot-end temperature measuring points, cold-end temperature measuring points, a thermal insulation and flow-limiting layer, a contact status detection element, and shielded leads. The heat flux metering layer uses high-temperature resistant alloy or ceramic thermally conductive blocks, whose calibrated thermal conductivity is obtained through calibration. The hot-end temperature measuring points are located near the tray contact surface, and the cold-end temperature measuring points are located near the furnace body support base, both arranged along the dominant heat flux direction. The thermal insulation and flow-limiting layer is located outside the heat flux metering layer to suppress bypass heat flux. For rectangular trays, four supported heat flux paths are arranged at the four corners; for trays longer than 1200 mm, two additional supported heat flux paths are added in the middle; for trays with a load capacity greater than 500 kg, eight supported heat flux paths are arranged. The contact status detection element uses a combination of a high-temperature resistant mechanical load switch and a tray identification mark to ensure that only valid contact paths are included in the calculation.

[0046] The in-furnace core temperature probe assembly is placed at the most unfavorable temperature rise location under the load of the material tray. The most unfavorable temperature rise location is determined according to process verification and includes the center of the material tray, the middle area of ​​the stacked layers, the vicinity of the center of thick-walled workpieces, and areas obstructed by air ducts. The in-furnace probe block is a cylindrical block made of the same material as the target workpiece. The diameter of the probe block is 0.8 to 1.2 times the thickness of the target workpiece's maximum thermal resistance path, and the height of the probe block is 0.8 to 1.5 times the diameter. The core temperature sensor is embedded in the geometric center of the probe block, and the surface temperature sensor is positioned 2mm to 5mm inward from the outer surface of the probe block. The probe block is identified using a high-temperature resistant QR code nameplate or a passive RF marker, recording the probe block number, material number, size number, and probe core temperature correction factor number.

[0047] The furnace environment and execution status acquisition components include furnace atmosphere temperature thermocouples, furnace external ambient temperature sensors, circulating air velocity sensors, furnace pressure transmitters, furnace door status switches, an energy metering module, and zoned power feedback interfaces. The furnace atmosphere temperature thermocouples are installed in a position that does not contact the material tray and represents the furnace circulating atmosphere. The circulating air velocity sensor is installed in the return air section of the circulating air duct. The furnace pressure transmitter is connected to the furnace pressure tap. The energy metering module measures voltage, current, active power, and cumulative energy for each heating zone. All analog channels enter the isolated acquisition module, all digital channels enter the opto-isolated digital module, and all power data enters the edge controller via Modbus TCP or Modbus RTU.

[0048] The system uses a uniform sampling period run, The sampling time is taken from 0.5s to 2s; the example uses 1s. One furnace record is written for each sampling period. The furnace record includes furnace number, process number, material tray number, proxy block number, and unified sampling sequence number. Unified sampling time All original acquired values, all aligned values, all filtered values, all channel validity markers, and the calculated heat absorption power of the support-side load. Calculated load heat absorption power on the power balance side Net heat absorption power of load Cumulative net heat absorption Correct agent core temperature Temperature hysteresis Load heat absorption fingerprint Control output Alarm status and completion judgment quantity .

[0049] A channel configuration table is created for each acquisition channel. The channel configuration table includes the channel number. Channel name, data type, unit, lower limit of range, upper limit of range, installation location, sampling interface, original sampling period, filter coefficient, and abnormal threshold. Missing data handling strategies, alarm levels, and traceability retention periods are defined. The edge controller stores data from the most recent 7200 sampling periods in a circular cache and stores all furnace data in a furnace database.

[0050] To avoid multiple meanings for the same character, the following global symbols are used throughout the text. Unless otherwise specified in this table, subsequent formulas must not change the meaning of the same symbol.

[0051] Global symbol table:

[0052] symbol Unique meaning Unit or value Unified sampling sequence number positive integer No. A unified sampling time s Unified sampling period s Acquisition Channel Number positive integer Original sampling sequence number of a certain acquisition channel positive integer Heating zone number positive integer Support heat flow path number positive integer Total number of acquisition channels indivual Total number of heating zones indivual Total number of support heat flow paths indivual No. Each channel in the original sampling sequence number raw value Determined by the channel No. Each channel at a unified sampling time alignment value Determined by the channel No. Each channel at a unified sampling time Filter value Determined by the channel No. The mean of each channel within the sliding window Determined by the channel No. Standard deviation of each channel within the sliding window Determined by the channel No. Anomaly detection multiplier threshold for each channel Dimensionless No. The first channel in the Validity markers for uniform sampling times 0 or 1 No. The first-order filter coefficients of each channel 0 to 1 No. Furnace atmosphere temperature at a uniform sampling time ℃ No. The ambient temperature outside the furnace at a uniform sampling time ℃ No. Hot end temperature of the support heat flow path ℃ No. cold end temperature of the support heat flow path ℃ Core temperature of the furnace-mounted agent block ℃ Surface temperature of the furnace-mounted agent block ℃ Correct agent core temperature ℃ Temperature hysteresis between furnace atmosphere temperature and corrected agent core temperature ℃ No. Actual input power of each heating zone kW No. Thermal efficiency coefficient of each heating zone 0 to 1 No. Final control output of each heating zone % or kW No. Basic control output for each heating zone % or kW No. Heat flux density along the support heat flow path W / square meter No. The thermal conductivity of the heat flow metering layer supported by the heat flow path is calibrated. W / (m·K) No. Spacing between hot and cold end measuring points along the support heat flow path m No. Effective converted area of ​​the support heat flow path square meters No. Effective contact markings for support heat flow path 0 or 1 Equivalent load heat absorption power on the support side kW Background power consumption of empty furnace kW Power balance side converted load heat absorption power kW Net heat absorption power of the fused load kW Cumulative net heat absorption kJ Slope of net heat absorption power change under load kW / s Load heat absorption attenuation ratio Dimensionless Estimated load heat capacity kJ / ℃ Support heat flow reliability 0 to 1 Load-absorbing fingerprint vector Eight-dimensional vector Uncertainty of support side power estimation kW Uncertainty of power estimation on the power balance side kW Support side fusion weight 0 to 1 Circulation speed m / s Furnace pressure Pa Furnace door status marker 0 or 1 Furnace body heat storage coefficient kJ / ℃ constants of the empty furnace background model kW Temperature difference coefficient in empty furnace background model kW / ℃ Wind speed coefficient in empty furnace background model kW / (m / s) Furnace pressure coefficient in empty furnace background model kW / Pa Furnace door coefficient in empty furnace background model kW No. Power term coefficients of the empty furnace background model for each heating zone Dimensionless Calculation interval for slope of heat absorption power change Sampling points Calculation interval for heat attenuation ratio Sampling points The completion condition is to continuously confirm the number of sampling points. Sampling points Agent core temperature correction constant term ℃ Agent core temperature correction factor Dimensionless Agent's watch core temperature difference correction coefficient Dimensionless Atmosphere and agent block temperature difference correction coefficient Dimensionless Heat absorption power correction factor ℃ / kW No. The target cumulative heat absorption at a unified sampling time kJ No. Target temperature hysteresis at a uniform sampling time ℃ No. Cumulative heat absorption compensation coefficient for each heating zone Control quantity / kJ No. Temperature lag compensation coefficient for each heating zone Control quantity / ℃ No. Heat absorption slope suppression coefficient of each heating zone Controlled quantity (kW / s) No. Lower limit of each heating zone control output % or kW No. Each heating zone controls the output limit. % or kW No. Upper limit of the rate of change for each heating zone. Control quantity / s Complete the determination of slope threshold kW / s Complete the determination of attenuation ratio threshold Dimensionless Complete the determination of temperature hysteresis threshold ℃ Complete the minimum cumulative heat absorption determination kJ Complete the determination of the reliability of the minimum support heat flow. 0 to 1 No. Completion determination quantity at a unified sampling time 0 or 1

[0053] Each acquisition module generates raw data at its own original sampling period. The edge controller establishes a unified sampling time. And map all channel data to a unified sampling time. For the first... Each acquisition channel, if the original sampling time meets... Then the alignment value is obtained according to formula (1).

[0054]

[0055] In formula (1), For the first Each channel at a unified sampling time The alignment value; For the first Each channel in the original sampling sequence number The original value; For the first Each channel in the original sampling sequence number The original value; for The corresponding original sampling time; for The corresponding original sampling time; For the first A unified sampling time; Assign a channel number to the data acquisition channel; The original sampling sequence number; To standardize the sampling sequence number.

[0056] The alignment value is filtered by first-order filtering to obtain the filtered value, and the calculation formula is formula (2).

[0057]

[0058] In formula (2), For the first The first channel in the The filtered value at a uniform sampling time; For the first The filter coefficients for each channel; For the first The first channel in the Alignment values ​​at a uniform sampling time; For the first The first channel in the The filtered value at a uniform sampling time.

[0059] The channel validity marker is determined according to formula (3).

[0060]

[0061] In formula (3), For the first The first channel in the A validity marker for each uniform sampling time; Alignment value; For the first The average value of each channel sliding window; For the first Standard deviation of each channel sliding window; For the first The threshold for anomaly detection multiple for each channel; numerical value Indicates validity; numerical value Indicates invalid.

[0062] In the data processing flow, raw values, aligned values, filtered values, and validity markers are all written to the furnace database. Control calculations only use... Channel data. When the critical channel When 10 consecutive sampling cycles are reached, the system triggers an alarm and enters degradation logic.

[0063] No. The heat flux density of the support heat flow path is calculated according to formula (4).

[0064]

[0065] In formula (4), For the first The support heat flow path is in the first Heat flux density at a uniform sampling time; For the first The calibrated thermal conductivity of the heat flow metering layer supporting the heat flow path; For the first The hot end temperature of the support heat flow path; For the first Cold end temperature of the support heat flow path; For the first The distance between the hot end measuring point and the cold end measuring point of the support heat flow path.

[0066] The heat absorption power of the converted load on the support side is calculated according to formula (5).

[0067]

[0068] In formula (5), Calculate the heat absorption power of the load on the support side; The conversion factor from watts to kilowatts; This represents the total number of supporting heat flow paths; For the first Effective contact markings for the support heat flow path; For the first Effective marking of the hot end temperature channel of the support heat flow path; For the first Effective marking of cold end temperature channel in support heat flow path; For the first Heat flux density along the support heat flow path; For the first Effective converted area of ​​the heat flow path of the support.

[0069] The background power consumption of the empty furnace was obtained by fitting empty furnace calibration data. The empty furnace calibration covered the heating stage, the isothermal stage, and different fan speeds, and collected data... , , , , and The background model of the empty furnace is calculated according to formula (6).

[0070]

[0071] In formula (6), Power consumption is consumed in the background of an empty furnace; For the constant term of the empty furnace background model; The coefficient of the temperature difference term in the empty furnace background model; This refers to the temperature of the atmosphere inside the furnace. The ambient temperature outside the furnace; The wind speed coefficient is the background coefficient of the empty furnace model. This refers to the circulating wind speed; The furnace pressure coefficient is the background model coefficient for an empty furnace. For furnace pressure; For the furnace door term coefficients in the empty furnace background model; For furnace door status marking; This represents the total number of heating zones. For the first Power term coefficients of the empty furnace background model for each heating zone; For the first The actual input power of each heating zone.

[0072] The background parameter vector of the empty furnace is calibrated according to formula (7).

[0073]

[0074] In formula (7), This is the background parameter vector for an empty furnace; The characteristic matrix for calibrating an empty furnace; for The transpose of the matrix; Calibrate the regularization coefficients for the background parameters of the empty furnace; Calibrate the unit matrix for background parameters of the empty furnace; Calibrate the target power vector for the empty furnace; superscript Indicates matrix transpose; superscript This represents finding the inverse of a matrix.

[0075] The heat absorption power of the converted load on the power balance side is calculated according to formula (8).

[0076]

[0077] In formula (8), Calculate the load heat absorption power on the power balance side; This represents the total number of heating zones. For the first Thermal efficiency coefficient of each heating zone; For the first Actual input power of each heating zone; Power consumption is consumed in the background of an empty furnace; The furnace body heat storage coefficient; For the first The furnace atmosphere temperature at a uniform sampling time; For the first The furnace atmosphere temperature at a uniform sampling time; To standardize the sampling period.

[0078] The fusion weight on the support side is calculated according to formula (9).

[0079]

[0080] In formula (9), For the support side fusion weight; Uncertainty in estimating the power on the support side; This formula assigns a higher weight to the side with lower uncertainty in the power balance estimate.

[0081] The net heat absorption power of the load is calculated according to formula (10).

[0082]

[0083] In formula (10), This represents the net heat absorption power of the load. For the support side fusion weight; Calculate the heat absorption power of the load on the support side; The heat absorption power of the load is converted to the power balance side.

[0084] The cumulative net heat absorption is calculated according to formula (11).

[0085]

[0086] In formula (11), For the first The cumulative net heat absorption at a uniform sampling time; For the first The cumulative net heat absorption at a uniform sampling time; This represents the net heat absorption power of the load. To standardize the sampling period. Because The unit is kW. The unit is seconds (s), and the unit of the calculation result is kJ.

[0087] The agent core temperature correction coefficient is obtained through calibration in a verification furnace with the actual workpiece core temperature. The calibration formula is formula (12).

[0088]

[0089] In formula (12), For the agent core temperature correction coefficient vector, including , , , and ; For the core temperature correction feature matrix of the proxy block; for The transpose of the matrix; To determine the regularization coefficient for core temperature correction; The core temperature correction calibration unit matrix; This is the temperature calibration vector for the core of the actual workpiece.

[0090] The corrected agent core temperature is calculated according to formula (13).

[0091]

[0092] In formula (13), To correct the agent's core temperature; For the agent core temperature correction constant term; For agent block core temperature correction coefficient; To determine the core temperature of the furnace-mounted agent block; For the agent's watch core temperature difference correction coefficient; The surface temperature of the furnace-fed agent block; This is a correction factor for the temperature difference between the atmosphere and the agent block. This refers to the temperature of the atmosphere inside the furnace. This is the heat absorption power correction factor; This represents the net heat absorption power of the load.

[0093] The temperature lag is calculated according to formula (14).

[0094]

[0095] In formula (14), This is the temperature hysteresis; This refers to the temperature of the atmosphere inside the furnace. To correct the agent core temperature.

[0096] The slope of the heat absorption power change is calculated according to formula (15).

[0097]

[0098] In formula (15), The slope of the change in net heat absorption power of the load; For the first Net heat absorption power of the load at a uniform sampling time; For the first Net heat absorption power of the load at a uniform sampling time; The interval is calculated for the slope; To standardize the sampling period.

[0099] The heat absorption attenuation ratio is calculated according to formula (16).

[0100]

[0101] In formula (16), The load heat absorption attenuation ratio; For the first Net heat absorption power of the load at a uniform sampling time; For the first Net heat absorption power of the load at a uniform sampling time; The interval for calculating the heat attenuation ratio is as follows; To prevent the power lower limit constant from having a denominator of zero; This indicates taking the larger value within the parentheses.

[0102] The estimated load heat capacity is calculated according to formula (17).

[0103]

[0104] In formula (17), This is an estimate of the load's heat capacity; This is the cumulative net heat absorption; To correct the agent's core temperature; The corrected agent core temperature at the start of the furnace cycle; To prevent the lower limit constant of the denominator from being zero; This indicates taking the larger value within the parentheses.

[0105] The reliability of the support heat flow is calculated according to formula (18).

[0106]

[0107] In formula (18), To ensure the reliability of the heat flow; This represents the total number of supporting heat flow paths; For the first Effective contact markings for the support heat flow path; For the first Effective marking of the hot end temperature channel of the support heat flow path; For the first Effectiveness marking of cold end temperature channel of the support heat flow path.

[0108] The load heat absorption fingerprint vector is generated according to formula (19).

[0109]

[0110] In formula (19), This is the load heat absorption fingerprint vector; This represents the net heat absorption power of the load. This is the cumulative net heat absorption; To correct the agent's core temperature; This is the temperature hysteresis; The slope of the change in net heat absorption power of the load; The load heat absorption attenuation ratio; This is an estimate of the load's heat capacity; To support the reliability of heat flow; superscript This represents the transpose of a vector.

[0111] Basic control output The original furnace body formula controller provides the output. The thermal inertia compensation temperature control unit calculates the original compensation output according to formula (20).

[0112]

[0113] In formula (20), For the first The original compensation output of each heating zone; For the first Basic control outputs for each heating zone; For the first Cumulative heat absorption compensation coefficient for each heating zone; Accumulate heat for the target; This is the cumulative net heat absorption; For the first Temperature lag compensation coefficient for each heating zone; This is the temperature hysteresis; The target temperature hysteresis; For the first The heat absorption slope suppression coefficient of each heating zone; The slope of the change in net heat absorption power of the load.

[0114] The final control output is subject to amplitude and rate of change limits according to formula (21).

[0115]

[0116] In formula (21), For the first The final control output for each heating zone; For the first Each heating zone controls the upper limit of output; For the first Each heating zone control output lower limit; For the first The final control output of each heating zone at a unified sampling time; For the first The upper limit of the rate of change for each heating zone; To standardize the sampling period; For the first Original compensation output for each heating zone; This indicates taking the smaller value within the parentheses; This indicates taking the larger value within the parentheses.

[0117] During control execution, if the furnace door status flag... When the furnace door is open, the system freezes the heating compensation items and limits them. The system will not exceed the safety maintenance output; if any critical temperature exceeds the hardware threshold, the safety relay will cut off the heating power; if the communication interruption exceeds 3 sampling cycles, the system will maintain the previous safety output and trigger an audible and visual alarm.

[0118] The determination of process completion is based on formula (22).

[0119]

[0120] In formula (22), For the first The completion determination quantity for a unified sampling time; This is an indicator function; it takes the value 1 if all conditions within the parentheses are true, and 0 if any condition is false. For the first The slope of the change in net heat absorption power of the load at a uniform sampling time; To complete the determination of the slope threshold; For the first The heat absorption attenuation ratio at a uniform sampling time; To determine the attenuation ratio threshold; For the first Temperature lag at a uniform sampling time; To complete the determination of the temperature hysteresis threshold; For the first The cumulative net heat absorption at a uniform sampling time; To determine the minimum cumulative heat absorption; For the first Reliability of support heat flow at a uniform sampling time; To determine the reliability of the minimum support heat flow; To ensure the completion of the conditions, the number of sampling points must be continuously confirmed; This refers to the sampling sequence number within the continuous confirmation window.

[0121] when When the insulation is complete, the system outputs the insulation completion status and records the completion time. If any abnormality occurs in the heat flow channel, the agent core temperature probe, the furnace door is opened abnormally, the heating power feedback is abnormal, or the temperature exceeds the limit within the continuous confirmation window, the system will... Set to 0 and restart the continuous confirmation timer.

[0122] The empty furnace background calibration is performed according to the following steps: First, in an empty furnace state, raise the furnace body from room temperature to the maximum process temperature, and record the heating process. , , , , and Second, maintain the temperature at the commonly used process temperature platform for 30 to 60 minutes and collect data during the stable period; third, change the frequency of the circulating fan to create three wind speed conditions; fourth, construct the empty furnace calibration feature matrix. and target power vector Calculated using formula (7) Fifth, The data is written to the edge data processing unit and verified in the next empty furnace check that the average error is no greater than 5%.

[0123] The support heat flow path calibration is performed according to the following steps: First, place a standard heat source block above the support heat flow path; second, run the system at three stable heat input levels, each lasting 20 minutes; third, record the results. , and standard heat input power; fourth, calculate and correct. and The product; fifth, repeat the calibration three times, and write the result into the parameter table when the repeatability deviation is no greater than 8%.

[0124] The calibration of the surrogate block core temperature correction is performed according to the following steps: First, temporarily install a verification temperature sensor in the core of the target workpiece; second, place the surrogate block and the target workpiece together in the most unfavorable heating position; third, run the complete heat treatment formula and collect the actual workpiece core temperature, surrogate block core temperature, surrogate block surface temperature, atmosphere temperature, and net heat absorption power of the load; fourth, construct... and Calculated using formula (12) Fifth, after removing the temporary verification temperature sensor, the subsequent production furnaces will use the furnace-mounted agent block to estimate and correct the agent core temperature.

[0125] The system employs a three-tiered security mechanism: hardware protection, software protection, and data protection. Hardware protection consists of an independent temperature controller, safety relays, and a heating main circuit contactor; heating power is directly cut off when the temperature exceeds a hardware threshold. Software protection is implemented by an edge controller; when… The key channel has reached 10 consecutive sampling cycles. When the furnace door status is abnormal, communication is interrupted, or the heating power feedback deviation exceeds 10%, the system enters degraded control. Degraded control retains basic recipe control, disables load thermal inertia compensation, prohibits premature judgment, and sets the upper limit of the control change rate. Reduced to 50% of normal value. Data protection requires that the raw data, processed data, formula calculation results, control outputs and alarm events of each furnace batch be written to both the local database and the host computer database simultaneously. If the write fails for three consecutive sampling cycles, a data recording anomaly alarm will be triggered. Specific Implementation Example 2:

[0127] This embodiment applies the invention to a four-zone circulating wind-powered heating heat treatment furnace. The effective furnace chamber dimensions are 1800mm × 1200mm × 900mm, the rated temperature is 650℃, and the number of heating zones is... Each zone has a rated power of 18kW, and the circulating fan has a rated power of 5.5kW. The material tray size is 1200mm×800mm, and the weight of gear steel parts loaded in a single furnace ranges from 280kg to 520kg. The process formula is to heat to 540℃, homogenize for 30min, and hold for 120min.

[0128] The hardware installation is as follows: Six supporting heat flow paths are installed at the four corners and the middle of the two long sides of the tray, i.e. The thickness of the heat flow metering layer supporting each heat flow path. m, effective converted area The area is square meters, and both the hot and cold end temperatures are achieved using K-type armored thermocouples. The furnace-mounted thermocouple block is a cylindrical block made of the same material as the gear steel components, with a diameter of 60mm and a height of 60mm. The core temperature sensor is located at the geometric center, and the surface temperature sensor is 3mm from the surface. The edge controller sampling period... s, slope calculation interval Calculation interval for heat absorption attenuation ratio The condition for completion is to continuously confirm the number of sampling points. .

[0129] The calibration process is as follows: The empty furnace calibration is performed at three platforms: 300℃, 450℃, and 540℃, with each platform maintained for 45 minutes, and three fan speeds set at each platform. The result is obtained through formula (7). kW kW / ℃ kW / (m / s), kW / Pa, kW, four heating zones The values ​​are 0.035, 0.032, 0.036, and 0.034, respectively. The agent obtained the core temperature correction calibration. , , , , .

[0130] A certain production batch in the 1st At each sampling time, data was collected. , , m / s, Pa, The actual input power of the four zones is 8.0kW, 7.5kW, 7.8kW, and 7.6kW. The heat flux densities of the six supporting heat flow paths are 1680W / m², 1715W / m², 1650W / m², 1690W / m², 1740W / m², and 1665W / m², respectively. All six supporting heat flow paths are in effective contact and the channels are effective. According to formula (5), kW. According to formulas (6) and (8), we obtain... kW. The system obtains the current channel noise. kW and kW, obtained from formula (9) From formula (10), we get kW.

[0131] At the same time, the core temperature of the agent block Agent block surface temperature According to formula (13), we get According to formula (14), we get The system calculates... kW / s , kJ, The heat absorption fingerprint of the load is .because Temperature hysteresis threshold higher than the completion judgment threshold The system does not output a completion status, and increases the control output of the two heating zones in the middle by 2.0% according to formula (20).

[0132] When the furnace reaches the [number]th [time] At each sampling time, the system calculates... kW / s , , kJ, If the above conditions are met continuously for 600 seconds according to formula (22), the system output will be... The system records the completion status of the heat preservation process and exports the furnace report. Compared with the fixed heat preservation time of 120 minutes, the heat preservation time of this furnace is shortened by 13 minutes; in the heavy-load furnace with a loading mass of 520 kg, the system automatically extends the heat preservation time by 8 minutes to avoid underheating of the core.

[0133] An example of handling an anomaly is as follows: In another heat cycle, the hot end temperature of the fourth support heat flow path jumps by 80°C within 5 seconds, exceeding the anomaly threshold for that channel. The edge controller will use this channel Set it to 0, and remove the 4th support heat flow path from the calculation of formula (5). Since there are still five support heat flow paths valid, the system continues to operate and reduces If the number of effective support heat flow paths is less than two, the system enters degraded control, disables load thermal inertia compensation, retains basic formula control, and prohibits output. .

[0134] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0135] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent temperature field prediction and refined furnace temperature control system, characterized in that, include: The system comprises a load-supported heat flux acquisition component, a proxy core temperature probe component, a furnace environment and execution status acquisition component, an edge data processing unit, a load heat absorption fingerprint generation unit, a thermal inertia compensation temperature control unit, a process completion determination unit, and a safety protection and data traceability unit. The load-supported heat flux acquisition component includes four to eight supporting heat flow paths. Each supporting heat flow path includes a high-temperature resistant support, a heat flow metering layer, a hot-end temperature measuring point, a cold-end temperature measuring point, a heat insulation and flow-limiting layer, a contact status detection component, and a shielded lead wire. The load-supported heat flux acquisition component is configured to acquire heat flow data transmitted by the load of the material tray through the supporting paths. The proxy core temperature probe component includes a furnace-mounted proxy block, a core temperature sensor, a surface temperature sensor, and a proxy block identification mark. The furnace-mounted proxy block is placed at the most unfavorable heating position of the load of the material tray. The proxy core temperature probe component is configured to acquire temperature data of the proxy block representing the lag in the internal heating of the workpiece. The furnace environment and execution status acquisition component is configured to acquire... The data processing unit is configured to perform unified time reference alignment, outlier detection, filtering, validity marking, unit conversion, support path heat flow conversion, empty furnace background subtraction, power balance calculation, load net heat absorption power fusion, and cumulative net heat absorption integral on the collected data. The load heat absorption fingerprint generation unit is configured to generate a load heat absorption fingerprint based on the load net heat absorption power, cumulative net heat absorption, corrected proxy core temperature, temperature hysteresis, heat absorption power change slope, heat absorption attenuation ratio, estimated load heat capacity, and support heat flow reliability. The thermal inertia compensation temperature control unit is configured to correct the heating zone control outputs of the heating rise section, homogenization section, and heat preservation section based on the load heat absorption fingerprint. The process completion determination unit is configured to determine the process completion status based on the heat absorption power change slope, heat absorption attenuation ratio, temperature hysteresis, cumulative net heat absorption, and continuous confirmation time. The safety protection and data traceability unit is configured to trigger protection actions when there are abnormalities in the heat flow channel, abnormalities in the agent core temperature probe, abnormal opening of the furnace door, abnormal heating power feedback, temperature exceeding the limit, unreachable control output, and abnormal data recording, and to save the original furnace data, alignment data, channel quality mark, load heat absorption fingerprint, control commands, alarm events, and completion determination results.

2. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The heat flow metering layer is a high-temperature resistant heat-conducting block with a calibrated thermal conductivity. The hot end temperature measuring points and the cold end temperature measuring points are arranged at intervals along the main heat flow direction of the support path. The edge data processing unit calculates the heat flow density and heat flow power of a single support heat flow path according to the hot end temperature, cold end temperature, calibrated thermal conductivity of the heat-conducting block, measuring point spacing and effective conversion area.

3. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The contact status detection device includes a load switch installed under the high-temperature resistant support and a tray identification mark installed on the tray. The edge data processing unit generates a valid contact mark for the support path based on the load switch status, the tray identification mark, the temperature difference change rate of the support path, and the heat flow mutation of the support path. Only the support heat flow paths with valid contact marks and valid channel quality marks are included in the calculation of the net heat absorption power of the load.

4. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The accompanying furnace agent block is a cylindrical agent block made of the same material as the target workpiece. The diameter of the cylindrical agent block is 0.8 to 1.2 times the thickness of the maximum thermal resistance path of the target workpiece. The core temperature sensor is located at the geometric center of the cylindrical agent block, and the surface temperature sensor is located 2 mm to 5 mm inward from the outer surface of the cylindrical agent block. The agent block identification mark stores the agent block number, material number, size number, and agent core temperature correction coefficient number.

5. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The edge data processing unit performs time alignment on the data from different acquisition channels according to a unified sampling period, which is 0.5s to 2s. When an acquisition channel has two original sampled values ​​before and after the unified sampling time, the edge data processing unit uses linear interpolation to obtain the alignment value of the acquisition channel at the unified sampling time.

6. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The edge data processing unit performs a confidence-weighted fusion of the heat absorption power of the load converted from the support side and the heat absorption power of the load converted from the power balance side to obtain the net heat absorption power of the load. The heat absorption power of the load converted from the support side is obtained by summing the heat flow power of the effective support heat flow path. The heat absorption power of the load converted from the power balance side is calculated by the actual input power of the heating zone, the efficiency of the heating zone, the background power consumption of the empty furnace, and the change in the heat storage of the furnace body. The fusion weight of the confidence-weighted fusion is calculated by the uncertainty of the power estimation on the support side and the uncertainty of the power estimation on the power balance side.

7. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The load heat absorption fingerprint generated by the load heat absorption fingerprint generation unit is an eight-dimensional vector. The components of the eight-dimensional vector are, in order, the net heat absorption power of the load, the cumulative net heat absorption, the corrected proxy core temperature, the temperature hysteresis, the slope of the heat absorption power change, the heat absorption attenuation ratio, the estimated value of the load heat capacity, and the reliability of the support heat flow.

8. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The thermal inertia compensation temperature control unit superimposes a load cumulative heat absorption compensation term, a temperature lag compensation term, and a heat absorption power change slope suppression term onto the basic control output to obtain the original compensation output of the heating zone. After applying amplitude and rate of change limits to the original compensation output of the heating zone, it sends it to the heating zone power regulator.

9. The intelligent temperature field prediction and refined furnace temperature control system according to claim 1, characterized in that, The process completion determination unit outputs the completion status only when the following conditions are met simultaneously and the continuous confirmation time is reached: the absolute value of the slope of the change in net heat absorption power of the load is not greater than the slope threshold, the heat absorption attenuation ratio is not greater than the attenuation ratio threshold, the temperature hysteresis is not greater than the hysteresis threshold, the cumulative net heat absorption is not less than the target cumulative heat absorption, the support heat flow reliability is not less than the reliability threshold, and no abnormalities occur in the heat flow channel, the proxy core temperature probe, the furnace door, the heating power feedback, or the temperature exceeding the limit.

10. A temperature control method corresponding to an intelligent temperature field prediction and refined furnace temperature control system according to any one of claims 1 to 9, characterized in that, Includes the following steps: Sp1, four to eight supporting heat flow paths are installed at the material tray support position in the furnace body, and a furnace agent block is placed at the position where the material tray load is most unfavorable for heating. An isolation acquisition module, an energy metering module, an edge controller and a safety relay are installed in the furnace body electrical control cabinet. Sp2, establish a channel configuration table, which records the channel number, data type, installation location, unit, range, sampling period, communication address, filter coefficient and alarm threshold; Sp3 performs empty furnace background calibration, support heat flow path calibration, agent block core temperature correction calibration, heating zone power feedback calibration, and safety interlock test, and writes the calibration parameters into the edge data processing unit; Sp4 collects the following data in each uniform sampling period: hot end temperature, cold end temperature, core temperature of the proxy block, surface temperature of the proxy block, furnace atmosphere temperature, external ambient temperature, circulating wind speed, furnace pressure, furnace door status, actual input power of the heating zone, and control output of the heating zone. SP5 performs time alignment, outlier detection, filtering, validity marking, and unit conversion on the collected data, while saving both the original data and the processed data. Sp6 calculates the equivalent load heat absorption power on the support side based on the support heat flow path data, calculates the equivalent load heat absorption power on the power balance side based on the actual input power of the heating zone, the background power consumption of the empty furnace and the change in the furnace body heat storage, and performs a confidence-weighted fusion of the two to obtain the net load heat absorption power. Sp7, the net heat absorption power of the load is integrated over time to obtain the cumulative net heat absorption. The corrected agent core temperature is calculated based on the agent block temperature data and the agent core temperature correction coefficient. The load heat absorption fingerprint is generated based on the net heat absorption power of the load, the cumulative net heat absorption, the corrected agent core temperature, the temperature hysteresis, the slope of the heat absorption power change, the heat absorption attenuation ratio, the estimated value of the load heat capacity and the reliability of the support heat flow. Sp8 calculates the heating zone compensation control output based on the load heat absorption fingerprint, and issues control commands within the limits of amplitude, rate of change, furnace door status, hardware temperature and communication health. Sp9 determines the process completion rate based on the load heat absorption fingerprint and continuous confirmation time, and writes the original data, processing data, fingerprint data, control output, alarm records and completion determination results of the furnace batch into the traceability database.

Images