Temperature control method, system and equipment based on zirconia oxygen analyzer and storage medium

Abstract

The invention relates to the technical field of industrial process control, in particular to a temperature control method, system and equipment based on a zirconia oxygen analyzer and a storage medium. A temperature signal and an oxygen potential signal of a zirconium oxide sensor are obtained, a step response test is performed on a newly-installed heating wire, a heating time constant, heat capacity and a cooling time constant are calculated, a heating wire thermodynamic parameter fingerprint database is constructed, and automatic identification and parameter matching of heating wires of different models are achieved. And the self-tuning time is shortened from more than 30 minutes to less than 3 minutes. An adaptive model prediction control algorithm is established based on a three-order state space model, complex constraint optimization control is realized on an embedded platform, the temperature control precision is improved from + / -2 DEG C to + / -0.3 DEG C, the temperature rise rate is strictly controlled within the range of 2-8 DEG C / min, and zirconia ceramic cracking is effectively prevented. And a recursive least square method with a forgetting factor is adopted to identify the parameter change of the heating wire in real time, so that dynamic adaptive control and predictive maintenance are realized.

Description

Technical Field

[0001] This invention relates to the field of industrial process control technology, and in particular to a temperature control method, system, device and storage medium based on a zirconia oxygen analyzer. Background Technology

[0002] Zirconia oxygen analyzers, as important industrial process control instruments, are widely used in combustion control, environmental monitoring, and chemical production. This equipment detects the oxygen content in a gas by measuring the oxygen concentration potential generated by zirconia ceramics at high temperatures. Its core operating principle requires the zirconia sensor to operate stably in a high-temperature environment of approximately 750℃. Precise control of the sensor temperature directly affects the measurement accuracy and the equipment's lifespan.

[0003] In current industrial production environments, numerous brands and models of zirconia oxygen analyzers are widely used, including products from well-known manufacturers such as AMETEK, ABB, and Ireland's NTRON. However, existing technologies have significant shortcomings in temperature control. Traditional PID control methods require manual parameter retuning when changing heating wires of different models, taking more than 30 minutes; they have weak anti-interference capabilities, with ambient temperature changes of ±10℃ causing temperature fluctuations exceeding 5℃; and overshoot can reach 8-15℃, easily jeopardizing the structural integrity of zirconia ceramics. Furthermore, issues such as parameter drift due to heating wire aging, ceramic cracking caused by rapid temperature changes, and the lack of hardware-level protection in IGBT drives further exacerbate the control difficulties.

[0004] The oxygen potential signal output by the zirconia sensor is only a microvolt-level signal of 0-50mV, which is highly susceptible to 10-100mV power frequency interference in industrial environments. Traditional signal processing techniques are unable to effectively suppress various noise interferences, resulting in a significant decrease in measurement accuracy. In addition, existing safety protection mechanisms mostly rely on a single software level, which cannot meet the IEC 61508 SIL-2 safety level requirements and poses a risk of overheating and burnout. Summary of the Invention

[0005] In view of the problems existing in the prior art, the present invention is proposed.

[0006] Therefore, the problem to be solved by this invention is how to address the technical issues of insufficient temperature control accuracy, long time consumption for heating wire replacement and parameter adjustment, poor anti-interference ability, imperfect safety protection mechanism, and difficulty in processing microvolt-level oxygen potential signals in zirconia oxygen analyzers, and to provide a high-precision, adaptive, and multi-safety-protection intelligent temperature control method.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] In a first aspect, embodiments of the present invention provide a temperature control method based on a zirconia oxygen analyzer, which includes acquiring the temperature signal and oxygen potential signal of a zirconia sensor, performing a step response test on a newly installed heating wire, calculating the heating time constant, heat capacity and cooling time constant, and constructing a fingerprint database of heating wire thermodynamic parameters.

[0009] An adaptive model predictive control algorithm is established based on a third-order state-space model to control the temperature of the zirconia sensor, achieving precise control of 750℃±0.3℃.

[0010] The recursive least squares method with a forgetting factor is used to identify changes in the resistance and heat capacity parameters of the heating wire in real time.

[0011] A multi-layered coupling safety protection mechanism is implemented to prevent zirconia ceramics from being damaged by overheating.

[0012] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, the construction of the heating wire thermodynamic parameter fingerprint database includes:

[0013] Apply a standard step power test signal and record the dynamic temperature-time-current-voltage response curve from below 650℃ to steady state;

[0014] The heating time constant was calculated by the time difference between when the temperature reaches 63% and 10% of its steady-state value.

[0015] Calculate the heat capacity parameters based on the applied power, the time to reach steady state, and the temperature rise.

[0016] The heating time constant, heat capacity, cooling time constant, and resistance temperature coefficient are stored as fingerprint data for the heating wire.

[0017] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, the adaptive model predictive control algorithm includes:

[0018] A third-order state-space model is established with the state variables being sensor temperature, heater temperature, and temperature change rate;

[0019] Set constraints to limit the heating rate to 2-8℃ / min and the PWM duty cycle to 0-1;

[0020] By solving the quadratic programming problem through rolling optimization, temperature setpoint tracking and control changes can be smoothed.

[0021] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, the recursive least squares parameter identification includes:

[0022] The forgetting factor is set to 0.98, and the parameter update is triggered when the predicted temperature error is greater than 2.0℃ and lasts for 5 seconds.

[0023] The parameters to be identified in real time are equivalent resistance, heat capacity and heating time constant;

[0024] When the resistance drift exceeds 10%, an alarm for heating wire aging is triggered; when the heat capacity change exceeds 15%, an alarm for insulation material deterioration is triggered.

[0025] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, the multi-layer coupling safety protection mechanism includes:

[0026] The control layer protection enforces power change rate, temperature change rate, and temperature upper and lower limit constraints within the algorithm.

[0027] The monitoring layer protection uses an independent safety thread to force the power to drop to 30% within 200ms when the temperature rise rate exceeds 10℃ / min;

[0028] Hardware layer protection physically disconnects the main power circuit through a 780°C bimetallic switch and blocks the drive signal through a hardware voltage comparator.

[0029] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, it further includes:

[0030] The 0-50mV oxygen potential signal output by zirconium oxide is protected by a front-end bipolar TVS and shielded by an electromagnetic isolation chamber.

[0031] A differential amplifier is used to achieve a signal amplification with a common-mode rejection ratio of 120dB;

[0032] Interference is eliminated by differential calculation after synchronous sampling of the oxygen potential channel and the environmental noise channel;

[0033] Signal integrity is improved by using isolated DC-DC power supplies and three-level ground reconfiguration technology.

[0034] As a preferred embodiment of the temperature control method based on a zirconia oxygen analyzer described in this invention, it further includes signal processing:

[0035] The SMAJ5.0A bipolar TVS array is used to protect against electrostatic surges, and a MuMetal alloy shield is used to attenuate electromagnetic interference by 80dB.

[0036] Microvolt-level signal capture was achieved using an INA188 differential amplifier and an ADS1256 synchronous sampling ADC.

[0037] A dedicated PT1000 resistor is set to simulate the zirconium oxide impedance as a noise channel, and a differential operation is performed between the oxygen potential sample at time t0 and the noise sample at time t0+1μs.

[0038] The LT3045 ultra-low noise LDO is used to generate ±5V sensor power, achieving three-level grounding isolation between signal ground, analog ground, and chassis ground.

[0039] Secondly, embodiments of the present invention provide a temperature control system based on a zirconia oxygen analyzer, which includes a signal acquisition module for acquiring the temperature signal and oxygen potential signal of the zirconia sensor, performing a step response test on the newly installed heating wire, calculating the heating time constant, heat capacity and cooling time constant, and constructing a fingerprint library of heating wire thermodynamic parameters.

[0040] The control execution module is used to establish an adaptive model predictive control algorithm based on a third-order state-space model and to perform temperature control on the zirconia sensor, achieving precise control of 750℃±0.3℃.

[0041] The parameter identification module is used to identify changes in the resistance and heat capacity parameters of the heating wire in real time using a recursive least squares method with a forgetting factor.

[0042] The safety protection module is used to implement a multi-layered coupled safety protection mechanism to prevent zirconia ceramics from being damaged by overheating.

[0043] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, they implement the steps of the temperature control method based on a zirconia oxygen analyzer as described in the first aspect of the present invention.

[0044] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of the temperature control method based on a zirconia oxygen analyzer as described in the first aspect of the present invention.

[0045] The beneficial effects of this invention are as follows: By constructing a fingerprint database of heating wire thermodynamic parameters, this invention achieves automatic identification and parameter matching of different models of heating wires, completely solving the technical problem of traditional methods requiring manual parameter readjustment when replacing heating wires, which takes more than 30 minutes. The self-tuning time is shortened to within 3 minutes, significantly improving equipment maintenance efficiency and reducing operational skill requirements, achieving a plug-and-play effect for heating wires. By establishing an adaptive model predictive control algorithm based on a third-order state-space model, the limitations of traditional PID control in handling multi-variable constraints are overcome. Complex constraint optimization control is implemented on an embedded platform, strictly limiting the heating rate within a safe range of 2-8℃ / min. Simultaneously, the temperature control accuracy is significantly improved from the traditional ±2℃ to ±0.3℃, effectively preventing thermal stress cracking of zirconia ceramics due to rapid temperature changes and overshoot, achieving unexpectedly high-precision temperature control. By employing a recursive least squares method with a forgetting factor for online parameter identification, the control system can adaptively track the resistance drift and heat capacity changes of the heating wire during long-term operation. This transforms traditional static control into dynamic adaptive control, maintaining high-precision control performance throughout its lifespan and innovatively converting parameter drift information into predictive maintenance data, achieving a shift from passive maintenance to proactive prevention. Through a multi-layered coupled safety protection mechanism involving the control, monitoring, and hardware layers, a comprehensive, in-depth protection system is constructed, encompassing software algorithm constraints and physical device disconnection. With a response time down to the millisecond level, this system completely solves the safety hazards of traditional single-software protection failing to cope with extreme faults, meeting the stringent requirements of industrial safety level IEC 61508SIL-2 and achieving 100% protection against overheating burn-out accidents. Through four-level anti-interference signal processing technology, the system successfully solves the technical challenge of 0-50mV microvolt-level oxygen potential signals being overwhelmed by 10-100mV strong interference in industrial environments, improving signal processing accuracy by two orders of magnitude and achieving an oxygen content measurement resolution of 0.01%, providing precise oxygen control capabilities for ultra-low emission monitoring. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart of a temperature control method based on a zirconia oxygen analyzer;

[0048] Figure 2 A diagram of a computer device for a temperature control method based on a zirconia oxygen analyzer;

[0049] Figure 3 Another flowchart for the temperature control method based on a zirconia oxygen analyzer. Detailed Implementation

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0053] Example 1

[0054] Reference Figures 1-2 This is the first embodiment of the present invention, which provides a temperature control method based on a zirconia oxygen analyzer, including:

[0055] S100: Acquires the temperature signal and oxygen potential signal from the zirconia sensor, performs a step response test on the newly installed heating wire, calculates the heating time constant, heat capacity and cooling time constant, and constructs a fingerprint library of heating wire thermodynamic parameters;

[0056] S200: An adaptive model predictive control algorithm is established based on a third-order state-space model to control the temperature of the zirconia sensor, achieving precise control of 750℃±0.3℃.

[0057] S300: Employs a recursive least squares method with a forgetting factor to identify changes in heating wire resistance and heat capacity parameters in real time;

[0058] S400: Implements a multi-layered coupling safety protection mechanism to prevent zirconia ceramics from being damaged by overheating.

[0059] The core problems faced by zirconia oxygen analyzers in practical industrial applications mainly manifest in four aspects. First, in the signal acquisition and fingerprint database construction stage of S100, the 0-50mV microvolt oxygen potential signal output by the zirconia sensor is easily submerged by the 10-100mV power frequency interference in the industrial environment. Traditional signal processing methods cannot effectively suppress ground loop noise, power supply ripple, and electromagnetic interference, resulting in signal distortion exceeding 100%, which seriously affects the subsequent temperature control accuracy. At the same time, the resistance difference of different models of heating wires can reach ±30%, and their thermodynamic characteristics vary, making it impossible for traditional methods to quickly identify and adapt them. Second, in the model predictive control stage of S200, zirconia ceramics have strict requirements for the rate of temperature rise; rapid temperature changes exceeding 5℃ / min will cause ceramic cracking. However, traditional PID control has an overshoot of 8-15℃, which cannot meet the unique temperature constraint requirements of zirconia. Secondly, during the parameter identification phase of S300, the heating wire exhibits a resistance drift of 0.5% per 1000 hours during long-term operation, leading to control parameter mismatch. Traditional methods cannot track this slow change in real time. Finally, during the safety protection phase of S400, existing protection mechanisms largely rely on a single software layer. When software crashes or hardware malfunctions, they cannot provide reliable protection, posing a risk of overheating and burnout.

[0060] Claim 1 of this invention constitutes a complete intelligent temperature control technology solution for a zirconia oxygen analyzer through four core steps. By acquiring temperature and oxygen potential signals and constructing a fingerprint database of heating wire thermodynamic parameters, automatic identification and parameter matching of different heating wire models are achieved, completely solving the problem of manual parameter adjustment of more than 30 minutes required by traditional methods, and shortening the self-tuning time to within 3 minutes. The adaptive model predictive control algorithm based on a third-order state-space model breaks through the limitations of traditional PID control. Through online optimization and constraint processing, the temperature control accuracy is improved from ±2℃ to ±0.3℃, while strictly controlling the heating rate within the safe range of 2-8℃ / min, effectively preventing cracking of zirconia ceramics. The recursive least squares method with a forgetting factor is used to achieve real-time tracking of heating wire aging parameters, enabling the control system to adaptively compensate for parameter drift during long-term operation and maintain high-precision control throughout its life. The multi-layered coupled safety protection mechanism ensures equipment safety even under extreme fault conditions through triple protection of software constraints, hardware monitoring, and physical disconnection, with a response time of less than 50ms, meeting industrial safety level requirements.

[0061] Key English terms involved in the technical solution include: MPC (Model Predictive Control), an advanced control algorithm based on predictive models, which achieves precise control by predicting the future behavior of the system and optimizing the control sequence online; RLS (Recursive Least Squares), an online parameter identification algorithm that can update model parameters in real time to adapt to system changes; TVS (Transient Voltage Suppressor), a protective device used to protect circuits from voltage surges and electrostatic discharge damage; CMRR (Common Mode Rejection Ratio), which measures the ability of a differential amplifier to suppress common-mode interference signals; ADC (Analog-to-Digital Converter), which converts continuous analog signals into discrete digital signals; LDO (Low Dropout Regulator), which provides high-precision, low-noise, stable power output; PWM (Pulse Width Modulation), which controls power output by adjusting the pulse width duty cycle; and IGBT (Insulated Gate Bipolar Transistor). Transistor (insulated gate bipolar transistor) is a key switching device in power electronics; SPI (Serial Peripheral Interface) is a high-speed, full-duplex synchronous serial communication protocol.

[0062] Example 2

[0063] Reference Figures 1-3 This is the second embodiment of the present invention.

[0064] In this embodiment of the application, the step S100 of constructing the thermodynamic parameter fingerprint library of the heating wire includes the following steps A1-A4:

[0065] A1: Constructing the thermodynamic parameter fingerprint library for heating wires includes:

[0066] Apply a standard step power test signal and record the dynamic temperature-time-current-voltage response curve from below 650℃ to steady state;

[0067] The heating time constant was calculated by the time difference between when the temperature reaches 63% and 10% of its steady-state value.

[0068] Calculate the heat capacity parameters based on the applied power, the time to reach steady state, and the temperature rise.

[0069] The heating time constant, heat capacity, cooling time constant, and resistance temperature coefficient are stored as fingerprint data for the heating wire.

[0070] Step response feature recognition test

[0071] Upon initial activation or replacement of an unidentified heating wire, the system automatically performs a standard step power test. The test begins in the target low-temperature range (e.g., <650℃), applying 50% of the initial power as a standard step signal. The dynamic temperature-time-current-voltage response curve is recorded with high precision from the start of the target low-temperature range to near steady state (e.g., 650℃). The entire testing process strictly controls the heating rate within the range of 2-8℃ / min to ensure no thermal shock damage to the zirconia ceramic.

[0072] The data acquisition system uses K-type thermocouples for temperature measurement, covering a range of -200℃ to 1372℃, suitable for the high-temperature environment of zirconium-tipped thermocouples. For the weak voltage signal output by the thermocouples, the system uses a dedicated thermocouple signal conditioning chip, MAX6675, to convert the thermocouple voltage signal into a digital temperature value, which communicates with the microcontroller via an SPI interface. The sampling frequency is set to an appropriate value to capture the system's dynamic characteristics while avoiding data redundancy caused by oversampling.

[0073] During testing, the system continuously monitors environmental parameters to ensure a stable testing environment. When external interference or abnormal conditions are detected, the test program automatically pauses and resumes once conditions are restored, ensuring the accuracy and repeatability of the test data.

[0074] A2: Intelligent identification of physical parameters

[0075] Using dynamic response data, the system accurately calculates key thermodynamic parameters. Heating time constant (T) rise By analyzing the time difference between the initial temperature reaching 63% and 10% of the steady-state value, combined with the formula:

[0076]

[0077] Obtain the system's inertial characteristics. This parameter directly reflects the thermal response speed of the heating wire-sensor system.

[0078] Heat capacity (C) is determined by the applied constant power P, the time Δt required to reach steady state, and the temperature rise ΔT, through...

[0079]

[0080] Calculate thermal storage capacity. Cooling time constant (T) cool) The cooling characteristics of the system are determined by analyzing the time it takes for the system temperature to drop from steady state to 37% after a power outage.

[0081] The temperature coefficient of resistance α is obtained through multi-point temperature-resistance measurements, establishing a precise relationship between resistance and temperature. Maximum power P max and the maximum safe temperature rise rate dT max These parameters are determined through safety boundary tests, providing fundamental data for subsequent safety protection mechanisms. The system also measures and records the impedance characteristics of the heating wire at different temperatures, forming a complete electro-thermal characteristic database.

[0082] A3: Fingerprint Data Storage and Identification

[0083] The system will use the above parameters (including the temperature coefficient of resistance α, maximum power P) max Maximum safe temperature rise rate dT max The data is structured and stored in a specific format (similar to JSON), forming a unique "fingerprint" for the heating wire. The storage format includes complete information such as parameter name, value, unit, measurement accuracy, and measurement conditions, ensuring data accuracy and traceability.

[0084] Each fingerprint dataset contains a unique identifier and checksum to prevent data confusion or tampering. The system supports offline storage of thousands of fingerprint datasets, employing high-speed Flash memory to ensure fast data access and reliable long-term storage. The data structure design considers backward compatibility, allowing newer software versions to recognize and use older fingerprint data.

[0085] A4: Intelligent Calibration and Identification System

[0086] The system automatically generates a QR code containing a unique ID based on the heating wire's fingerprint data, which is then affixed to the heating wire itself. The QR code uses a high-fault-tolerance encoding format, ensuring reliable identification even in harsh industrial environments such as high temperature, humidity, and pollution. The QR code encodes the heating wire's model number, batch number, key parameter summary, and data verification information.

[0087] When replacing the heating wire, operators can instantly load its optimal control parameters simply by scanning a code, achieving "plug and play." The scanning process includes QR code recognition, data decoding, parameter verification, and system configuration, all completed within 3 minutes. The system also supports a backup method of manually entering the heating wire ID, ensuring normal recognition even if the QR code is damaged.

[0088] It should be noted that this heating wire fingerprint database technology completely eliminates the manual parameter adjustment step after replacing the heating wire, shortening the self-tuning time from >30 minutes in traditional PID to <3 minutes. It provides a heating wire life warning function based on resistance drift (ΔR / R0>5%) and is compatible with different models of heating wires with resistance ranges of 0.8 to 1.2Ω.

[0089] In this embodiment of the application, the adaptive model predictive control algorithm in step S200 includes the following steps B1-B4:

[0090] B1: Adaptive model predictive control algorithms include:

[0091] A third-order state-space model is established with the state variables being sensor temperature, heater temperature, and temperature change rate;

[0092] Set constraints to limit the heating rate to 2-8℃ / min and the PWM duty cycle to 0-1;

[0093] By solving the quadratic programming problem through rolling optimization, temperature setpoint tracking and control changes can be smoothed.

[0094] Embedded Optimized Constrained Model Predictive Control Core Design

[0095] The system implements the real-time operation of complex MPC algorithms on embedded processors with limited computing power, such as the STM32H743 (480MHz), strictly meeting the safety temperature constraints (750±0.3℃) and heating rate limits (2-8℃ / min) of zirconia ceramics, and solving the problems of large overshoot (8-15℃) and poor disturbance rejection (±10℃ disturbance fluctuation >5℃) of traditional PID.

[0096] The state variables of the third-order state-space model are designed as follows:

[0097]

[0098] Output y = T 传感器温度 The control input u = PWM duty cycle (equivalent power), limited by 0 ≤ u ≤ 1. The dynamic equation is x(k+1) = A × x(k) + B × u(k).

[0099] The A and B matrices are initialized by the heating wire fingerprint parameters and continuously updated through online identification.

[0100] The optimization objective of model predictive control is designed as follows:

[0101] minΣ[(T 传感器温度 (k)-750) 2 ×Q+Δu(k) 2 ×R]

[0102] Q (temperature weight = 10.0) prioritizes ensuring accurate tracking of the setpoint (750℃), while R (control change weight = 0.1) suppresses power surges and protects the actuator and heating wire.

[0103] B2: Design of Key Constraints

[0104] The system is designed with strict constraints to ensure the safety of the zirconia ceramic. |Δu|≤0.05 / s constrains the power change rate to ≤5% / second to prevent thermal shock. Strictly limited to the safe heating / cooling window of zirconia ceramics; 700≤T 传感器温度 Temperature hard boundary protection is set for ≤780℃.

[0105] These constraints are treated as hard constraints during the optimization process, and any control sequence that violates these constraints will be automatically rejected. The constraint implementation employs a strategy combining the active set method and the interior point method to ensure optimal control performance while satisfying the constraints. As the system approaches the constraint boundaries, the algorithm automatically adjusts the weight coefficients to prioritize safety.

[0106] B3: Embedded High-Efficiency Solving Techniques

[0107] To achieve real-time MPC on an embedded platform, the system employs several key optimization techniques. Hessian matrix pre-computation stores the fixed portion of the core quadratic programming problem offline in Flash memory, reducing online computational power consumption by 85%. Condense technology compresses the long prediction time domain of Np = 60 steps into a low-dimensional optimization problem related only to the control time domain of Nc = 5 steps, reducing computational complexity from O(Np...) 3 ) decreased to O(Nc 2 ).

[0108] The warm-start initialization utilizes the continuity and similarity of the optimization solution from the previous control cycle as the initial value for this optimization, reducing the number of iterations by 40% and ensuring a single solution time of <15ms. Combining these optimization techniques, the system achieves real-time MPC control with a 100ms control cycle on a 480MHz STM32H743.

[0109] The prediction time domain is set to 60 steps (6 seconds), and the control time domain is set to 5 steps (0.5 seconds), achieving an optimal balance between control performance and computational load. The optimization algorithm employs a modified interior-point method, specifically optimized for the memory and computational limitations of embedded platforms.

[0110] B4: Control Performance and Innovative Value

[0111] The system achieves an ultra-high steady-state accuracy of ±0.3℃ (compared to ±2.5℃ for traditional PID), with the temperature rise rate strictly controlled within a safe range of 2-8℃ / min, reducing ceramic cracking rate by 90%. Energy consumption is reduced by 18%, and it exhibits strong anti-interference capabilities (±10℃ disturbance fluctuation <0.5℃). These performance indicators are achieved thanks to the predictive and constraint handling capabilities of the MPC algorithm.

[0112] In one optional implementation, the system can also automatically switch control parameters according to different operating modes (startup, normal operation, maintenance, etc.). Startup mode uses more conservative control parameters to ensure safe temperature rise; normal operation mode pursues optimal control performance; and maintenance mode prioritizes system stability. This multi-mode control strategy further enhances the system's adaptability and reliability.

[0113] It should be noted that this embedded MPC control technology breaks through the bottleneck of model predictive control in real-time applications of resource-constrained embedded systems. While ensuring the core functions of the algorithm, it strictly meets the physical constraints of the heating rate unique to zirconia ceramics, achieving real-time requirements of 15ms computation time.

[0114] In this embodiment of the application, the recursive least squares parameter identification in step S300 includes the following steps C1-C3:

[0115] C1: Parameter identification for recursive least squares includes:

[0116] The forgetting factor is set to 0.98, and the parameter update is triggered when the predicted temperature error is greater than 2.0℃ and lasts for 5 seconds.

[0117] The parameters to be identified in real time are equivalent resistance, heat capacity and heating time constant;

[0118] When the resistance drift exceeds 10%, an alarm for heating wire aging is triggered; when the heat capacity change exceeds 15%, an alarm for insulation material deterioration is triggered.

[0119] The system addresses the issue of model parameters (such as resistance R, heat capacity C, and time constant T) being affected by aging of the heating wire during long-term operation. rise To address the drift problem and ensure the accuracy of the MPC model, a recursive least squares method with a forgetting factor is employed. The parameters to be identified are θ = [R (equivalent resistance), C (heat capacity), T]. rise (Heating time constant) T Data vector φ(k) = [current squared (I 2 ), power change rate (Δu) / temperature change (ΔT), temperature difference between heater and environment.

[0120] A forgetting factor λ = 0.98 is designed to give higher weight to new measurement data, enabling the algorithm to track parameter changes over time (such as aging). Adaptive gain update is based on real-time prediction error (y). mes -y predicted The covariance matrix P is dynamically adjusted. Robustness measures include adding a "blowout protection" mechanism for the covariance matrix (such as numerical stability checks and lower bound constraints).

[0121] The algorithm implementation includes the parameter update formula:

[0122] θ(k)=θ(k-1)+K(k)[y(k)-φ T (k)θ(k-1)]

[0123] Gain vector:

[0124] K(k)=P(k-1)φ(k) / [λ+φ T (k)P(k-1)φ(k)]

[0125] Covariance matrix:

[0126] P(k)=[P(k-1)-K(k)φ T [(k)P(k-1)] / λ

[0127] This recursive form avoids matrix inversion operations and is suitable for real-time applications.

[0128] C2: Smart Trigger Strategy

[0129] The system activates RLS updates only when the |predicted temperature error| > 2.0℃ and lasts for 5 seconds, avoiding unnecessary computational overhead in steady state. This triggering strategy can respond promptly to actual parameter changes while avoiding erroneous updates caused by measurement noise. The triggering judgment also considers the system's operating state, appropriately relaxing the triggering conditions during startup or when there are significant changes in setpoints.

[0130] To address the nonlinear and time-varying characteristics of the system, the algorithm also integrates multi-model adaptive techniques. The system maintains multiple parallel parameter estimators, each corresponding to different operating conditions, and adaptively selects the most suitable parameter combination through weighted selection. This design improves the algorithm's adaptability under complex operating conditions.

[0131] C3: Integrating fault diagnosis and predictive maintenance

[0132] The system transforms parameter drift into valuable maintenance information. (Resistance drift) This triggers a heating wire aging alarm; the threshold, determined based on extensive field data, provides early warning before significant performance degradation. (Heat capacity change (|C)) est -C nom A threshold of 15% triggers an early warning for the deterioration of insulation materials, reflecting the health status of the insulation layer.

[0133] The fault diagnosis function also includes sensor fault detection. By analyzing the convergence and consistency of parameter estimates, the system can identify sensor offset, drift, or intermittent faults. When a sensor abnormality is detected, the system switches to fault-tolerant control mode, using the remaining healthy sensors to maintain control functionality.

[0134] Predictive maintenance predicts future maintenance needs based on parameter change trends. The system records historical parameter changes and uses trend analysis to predict the remaining lifespan of critical components such as heating wires and sensors. Maintenance recommendations are provided in report form, including the expected maintenance time window, required spare parts, and maintenance procedures.

[0135] In one alternative implementation, the system can also upload parameter change data to a cloud-based analysis platform via a network, leveraging big data analytics to further improve the accuracy of fault prediction. The cloud platform can integrate operational data from multiple devices, identify common problems and best practices, and provide more precise maintenance guidance for on-site equipment.

[0136] It should be noted that this online parameter identification technology enables the control system to adaptively resist long-term aging effects, maintain high-precision control throughout its life, and transform the control performance degradation caused by parameter drift into a basis for predictive maintenance.

[0137] In this embodiment of the application, the multi-layered coupling security protection mechanism in step S400 includes the following steps D1-D4:

[0138] D1: Multi-layered coupling security protection mechanisms include:

[0139] The control layer protection enforces power change rate, temperature change rate, and temperature upper and lower limit constraints within the algorithm.

[0140] The monitoring layer protection uses an independent safety thread to force the power to drop to 30% within 200ms when the temperature rise rate exceeds 10℃ / min;

[0141] Hardware layer protection physically disconnects the main power circuit through a 780°C bimetallic switch and blocks the drive signal through a hardware voltage comparator.

[0142] Control layer MPC core protection

[0143] The algorithm enforces fundamental physical constraints such as the rate of power change (|Δu|≤0.05 / s), the rate of temperature change (2≤dT / dt≤8℃ / min), and upper and lower limits (700≤T≤780℃). These constraints are treated as hard constraints during optimization, and any control sequence that violates them will be automatically rejected. In the near-critical region (e.g., T>740℃), the control weight R is automatically increased (to suppress power fluctuations) and the temperature weight Q is decreased (to moderately relax accuracy), prioritizing safety.

[0144] The control layer protection is implemented using a combination of barrier functions and penalty functions. When the system state approaches the constraint boundary, the objective function increases sharply, forcing the optimizer to select control actions away from the danger zone. Simultaneously, the system implements hierarchical constraint processing, setting safety-related constraints as the highest priority and performance-related constraints as the second highest priority.

[0145] When multiple safety indicators are detected approaching critical values ​​simultaneously, the control algorithm automatically switches to conservative mode. In conservative mode, the system employs lower control gain and a longer response time. Although control performance is reduced, absolute safety is ensured.

[0146] Independent security thread in the monitoring layer

[0147] A high-priority monitoring daemon thread, running independently of the main control loop (10Hz) within the MCU, continuously monitors the temperature rise rate and current temperature. Tiered protection actions include: forcing power down to 30% within 200ms when the temperature rise rate is >10℃ / min; switching to a simplified standby PID algorithm within 500ms when the sensor temperature is >760℃; and triggering final protection within <50ms when the temperature is >780℃ (or any reading fails).

[0148] The monitoring thread employs a state machine design, featuring three main states: normal monitoring, early warning, and emergency. In the normal monitoring state, the thread checks key parameters at a frequency of 50Hz; upon entering the early warning state, the checking frequency increases to 100Hz, and abnormal events begin to be recorded; in the emergency state, the thread focuses on executing protective actions to ensure rapid response.

[0149] The backup PID algorithm is specially designed and tested, employing extremely conservative control parameters. Although its control precision is lower, it can maintain basic temperature control functions and prevent the system from completely losing control when the main control algorithm fails. The parameters of the backup algorithm have undergone worst-case analysis to ensure that it will not lead to system instability under any conditions.

[0150] Hardware layer protection cannot be bypassed

[0151] The mechanical temperature control switch uses independent bimetallic contact structure (e.g., normally closed at 780℃), physically cutting off the main power circuit when the temperature exceeds the limit. A pure hardware voltage comparator directly compares the thermocouple signal between high and low limits (760 / 780℃), immediately gating and blocking the IGBT drive signal if the limit is exceeded. An independent hardware watchdog monitors the MCU's operating heartbeat; if the main program freezes for ≥1.6 seconds, the watchdog strongly resets the MCU and locks the power output to a safe state.

[0152] The hardware protection circuit adopts a fail-safe design principle, ensuring that the system enters a safe state rather than a dangerous state upon the failure of any protective device. The bimetallic switch is a high-reliability industrial-grade product, verified through temperature cycling and mechanical life testing. The voltage comparator is implemented using a high-speed operational amplifier, with a response time of less than 1 microsecond.

[0153] The watchdog circuit not only monitors the presence of the heartbeat signal, but also its frequency and waveform characteristics. If an abnormality is detected in the heartbeat signal (such as excessively high or low frequency, or waveform distortion), the watchdog will immediately initiate a protection program. This window-type watchdog provides stronger fault detection capabilities than traditional watchdogs.

[0154] D2: also includes:

[0155] The 0-50mV oxygen potential signal output by zirconium oxide is protected by a front-end bipolar TVS and shielded by an electromagnetic isolation chamber.

[0156] A differential amplifier is used to achieve a signal amplification with a common-mode rejection ratio of 120dB;

[0157] Interference is eliminated by differential calculation after synchronous sampling of the oxygen potential channel and the environmental noise channel;

[0158] Signal integrity is improved by using isolated DC-DC power supplies and three-level ground reconfiguration technology.

[0159] D3: Also includes signal processing:

[0160] The SMAJ5.0A bipolar TVS array is used to protect against electrostatic surges, and a MuMetal alloy shield is used to attenuate electromagnetic interference by 80dB.

[0161] Microvolt-level signal capture was achieved using an INA188 differential amplifier and an ADS1256 synchronous sampling ADC.

[0162] A dedicated PT1000 resistor is set to simulate the zirconium oxide impedance as a noise channel, and a differential operation is performed between the oxygen potential sample at time t0 and the noise sample at time t0+1μs.

[0163] The LT3045 ultra-low noise LDO is used to generate ±5V sensor power, achieving three-level grounding isolation between signal ground, analog ground, and chassis ground.

[0164] To address the interference challenges posed by the microvolt-level electromotive force signal (typically only 0-50mV) output by zirconium oxide in industrial environments, the system employs a four-level collaborative solution. Front-end signal isolation and protection utilize a bipolar TVS array (SMAJ5.0A) to protect against electrostatic discharge and surges. RF suppression components include a common-mode choke (TDK ACM2012) to suppress interference >10MHz, and a feedthrough capacitor (100nF / 1kV) to filter out wire coupling noise.

[0165] The high common-mode rejection ratio (CMRR) signal chain employs an INA188 instrumentation amplifier (CMRR = 120dB@50Hz) and a programmable gain stage LTC6912, a dual T-type notch filter (f0 = 50 / 60Hz, Q = 30), and a 24-bit Σ-Δ ADC ADS1256. Dynamic baseline tracking acquires the open-circuit voltage every 10 seconds (at the moment of zirconium oxide power failure) and automatically subtracts background noise. An adaptive gain control algorithm automatically switches the gain (1x / 10x / 100x) based on the signal-to-noise ratio.

[0166] Synchronous sampling and digital filtering employ differential calculation of oxygen potential channel sampling (t0) - noise channel sampling (t0 + 1μs) = real signal, using a dedicated PT1000 resistor to simulate zirconium oxide impedance (no current flows, only environmental interference is captured). Power supply and grounding reconstruction utilize an ADUM6000 isolated DC-DC converter, with an LT3045 ultra-low noise LDO generating ±5V sensor power, output voltage noise of 0.8μV RMS, and power supply rejection ratio of 100dB@100Hz.

[0167] In one alternative implementation, the system can also implement adaptive filtering technology, automatically adjusting filter parameters according to the current interference environment. When strong interference at a specific frequency is detected, the system dynamically generates a corresponding notch filter to precisely suppress that frequency component. This adaptive capability enables the system to cope with complex and ever-changing industrial electromagnetic environments.

[0168] It should be noted that the above-mentioned four-level anti-interference technology, through millivolt-level signal protection, hundred-dB-level noise suppression, and intelligent dynamic compensation three-level hard core technology, completely solves the problem of effective extraction of zirconium oxide signals in harsh industrial environments. The measured oxygen content resolution reaches 0.01% (compared to only 0.1% by traditional technology), which helps to achieve precise oxygen control in ultra-low emission monitoring.

[0169] In summary, this invention completely solves the key technical challenges in temperature control of zirconia oxygen analyzers by constructing a complete heating wire fingerprint recognition system, implementing an embedded constraint-based MPC algorithm, establishing an online parameter identification mechanism, and designing a multi-layered coupled safety protection system. The system improves temperature control accuracy from ±2℃ to ±0.3℃, achieves self-tuning within 3 minutes after heating wire replacement, and establishes a comprehensive safety protection system from software to hardware, providing reliable technical support for the industrial application of zirconia oxygen analyzers.

[0170] Example 3

[0171] The above is a schematic scheme of a temperature control method based on a zirconia oxygen analyzer. It should be noted that the technical solution of this temperature control system based on a zirconia oxygen analyzer and the technical solution of the temperature control method based on a zirconia oxygen analyzer described above belong to the same concept. Details not described in detail in this embodiment regarding the temperature control system based on a zirconia oxygen analyzer can be found in the description of the temperature control method based on a zirconia oxygen analyzer described above.

[0172] This embodiment also provides a temperature control system based on a zirconium oxide oxygen analyzer, including:

[0173] The signal acquisition module is used to acquire the temperature signal and oxygen potential signal of the zirconia sensor, perform a step response test on the newly installed heating wire, calculate the heating time constant, heat capacity and cooling time constant, and construct a fingerprint library of heating wire thermodynamic parameters.

[0174] The control execution module is used to establish an adaptive model predictive control algorithm based on a third-order state-space model and to control the temperature of the zirconia sensor, achieving precise control of 750℃±0.3℃.

[0175] The parameter identification module is used to identify changes in the resistance and heat capacity parameters of the heating wire in real time using a recursive least squares method with a forgetting factor.

[0176] The safety protection module is used to implement a multi-layered coupled safety protection mechanism to prevent zirconia ceramics from being damaged by overheating.

[0177] This embodiment also provides an electronic device suitable for temperature control based on a zirconia oxygen analyzer, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the temperature control method based on a zirconia oxygen analyzer as proposed in the above embodiment.

[0178] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the temperature control method based on a zirconia oxygen analyzer as proposed in the above embodiments.

[0179] The storage medium proposed in this embodiment and the temperature control method based on the zirconia oxygen analyzer proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0180] Based on the above description of the implementation methods, those skilled in the art will clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0181] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A temperature control method based on a zirconia oxygen analyzer, characterized in that: This includes acquiring the temperature and oxygen potential signals from the zirconia sensor, performing a step response test on the newly installed heating wire, calculating the heating time constant, heat capacity, and cooling time constant, and constructing a fingerprint database of the heating wire's thermodynamic parameters. An adaptive model predictive control algorithm is established based on a third-order state-space model to control the temperature of the zirconia sensor, achieving precise control of 750℃±0.3℃. The recursive least squares method with a forgetting factor is used to identify changes in the resistance and heat capacity parameters of the heating wire in real time. A multi-layered coupling safety protection mechanism is implemented to prevent zirconia ceramics from being damaged by overheating.

2. The temperature control method based on a zirconia oxygen analyzer as described in claim 1, characterized in that: The construction of the heating wire thermodynamic parameter fingerprint library includes: Apply a standard step power test signal and record the dynamic temperature-time-current-voltage response curve from below 650℃ to steady state; The heating time constant was calculated by the time difference between when the temperature reaches 63% and 10% of its steady-state value. Calculate the heat capacity parameters based on the applied power, the time to reach steady state, and the temperature rise. The heating time constant, heat capacity, cooling time constant, and resistance temperature coefficient are stored as fingerprint data for the heating wire.

3. The temperature control method based on a zirconia oxygen analyzer as described in claim 2, characterized in that: The adaptive model predictive control algorithm includes: A third-order state-space model is established with the state variables being sensor temperature, heater temperature, and temperature change rate; Set constraints to limit the heating rate to 2-8℃ / min and the PWM duty cycle to 0-1; By solving the quadratic programming problem through rolling optimization, temperature setpoint tracking and control changes can be smoothed.

4. The temperature control method based on a zirconia oxygen analyzer as described in claim 3, characterized in that: The recursive least squares parameter identification includes: The forgetting factor is set to 0.98, and the parameter update is triggered when the predicted temperature error is greater than 2.0℃ and lasts for 5 seconds. The parameters to be identified in real time are equivalent resistance, heat capacity and heating time constant; When the resistance drift exceeds 10%, an alarm for heating wire aging is triggered; when the heat capacity change exceeds 15%, an alarm for insulation material deterioration is triggered.

5. The temperature control method based on a zirconia oxygen analyzer as described in claim 4, characterized in that: The multi-layered coupling security protection mechanism includes: The control layer protection enforces power change rate, temperature change rate, and upper and lower temperature limits within the algorithm. The monitoring layer protection uses an independent safety thread to force the power to drop to 30% within 200ms when the temperature rise rate exceeds 10℃ / min; Hardware layer protection physically disconnects the main power circuit through a 780°C bimetallic switch and blocks the drive signal through a hardware voltage comparator.

6. The temperature control method based on a zirconia oxygen analyzer as described in claim 5, characterized in that: Also includes: The 0-50mV oxygen potential signal output by zirconium oxide is protected by a front-end bipolar TVS and shielded by an electromagnetic isolation chamber. A differential amplifier is used to achieve a signal amplification with a common-mode rejection ratio of 120dB; Interference is eliminated by differential calculation after synchronous sampling of the oxygen potential channel and the environmental noise channel; Signal integrity is improved by using isolated DC-DC power supplies and three-level ground reconfiguration technology.

7. The temperature control method based on a zirconia oxygen analyzer as described in claim 6, characterized in that: It also includes signal processing: The SMAJ5.0A bipolar TVS array is used to protect against electrostatic surges, and a MuMetal alloy shielding cover is used to attenuate electromagnetic interference by 80dB. Microvolt-level signal capture was achieved using an INA188 differential amplifier and an ADS1256 synchronous sampling ADC. A dedicated PT1000 resistor is set to simulate the zirconium oxide impedance as a noise channel, and a differential operation is performed between the oxygen potential sample at time t0 and the noise sample at time t0+1μs. The LT3045 ultra-low noise LDO is used to generate ±5V sensor power, achieving three-level grounding isolation between signal ground, analog ground, and chassis ground.

8. A temperature control system based on a zirconia oxygen analyzer, based on the temperature control method based on a zirconia oxygen analyzer according to any one of claims 1 to 7, characterized in that: It also includes a signal acquisition module, which is used to acquire the temperature signal and oxygen potential signal of the zirconia sensor, perform a step response test on the newly installed heating wire, calculate the heating time constant, heat capacity and cooling time constant, and construct a fingerprint library of heating wire thermodynamic parameters; The control execution module is used to establish an adaptive model predictive control algorithm based on a third-order state-space model and to perform temperature control on the zirconia sensor, achieving precise control of 750℃±0.3℃. The parameter identification module is used to identify changes in the resistance and heat capacity parameters of the heating wire in real time using a recursive least squares method with a forgetting factor. The safety protection module is used to implement a multi-layered coupled safety protection mechanism to prevent zirconia ceramics from being damaged by overheating.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the temperature control method based on a zirconia oxygen analyzer as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the temperature control method based on a zirconia oxygen analyzer as described in any one of claims 1 to 7.

Images