A temperature equalization control method for an explosion-proof electric appliance

Abstract

The application relates to the technical field of temperature control, and discloses a temperature control method for an explosion-proof electric appliance, which has the technical scheme as follows: based on a thermal steady state condition, an initial swarm intelligence model reference baseline is obtained through initial orthogonal micro-perturbation identification; through an ant colony pheromone evaporation and deposition mechanism and a ladder type reverse heat pulse, an updated gain matrix and a new layout reference baseline are obtained in combination with updated orthogonal micro-perturbation identification; a suction cup power reference value is obtained through double-mode cooperative control including a bat echo location frequency decoupling mode and a pseudo-inverse decoupling mode. The adaptive modeling and decoupling control framework based on swarm intelligence realizes rapid and robust temperature regulation and control of a three-dimensional space temperature field under a movable layout through the ant colony pheromone evaporation and deposition mechanism and the swarm intelligence mechanism of bat echo location, and the adaptive capacity of the system to layout changes, the working condition control stability and the operation safety of a sealed environment are improved.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, and more specifically to a temperature control method for explosion-proof electrical appliances. Background Technology

[0002] With the increasing demand for enclosed spaces in industrial production and warehousing, modular wall-mounted infrared radiant heating systems have gained widespread application due to their advantages of flexible layout, high thermal efficiency, and precise local heating. These systems achieve fine-tuning of the three-dimensional temperature field through multiple independently controllable heating unit matrices, significantly improving work environment comfort and energy efficiency. Current mainstream temperature control methods are mostly based on heat transfer gain matrix modeling and inverse calculation to solve power distribution. While they can achieve basic temperature control in fixed layout scenarios, in applications with frequent layout changes, the residual heat field formed by the walls as large heat capacities can interfere with the accuracy of the system model identification, leading to decreased control performance. Furthermore, when heating units are arranged similarly or symmetrically, the model is prone to ill-conditioned characteristics, causing power oscillations and control instability. In addition, existing methods often fail to deeply integrate operational safety constraints into the entire control process, making it difficult to simultaneously meet the requirements of temperature control and safe operation in enclosed environments. Therefore, existing technologies have shortcomings. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide a method for temperature uniformity control of explosion-proof electrical appliances. This method constructs a system heat transfer model through initial orthogonal perturbation identification and updated orthogonal perturbation identification. Based on the ant colony pheromone evaporation and deposition mechanism combined with a stepped anti-phase heat pulse, it achieves adaptive model updates after layout changes. According to the bat echolocation swarm intelligence mechanism, it achieves precise power allocation through dual-mode collaborative control. This enables rapid and robust temperature uniformity control of the three-dimensional spatial temperature field under arbitrary layouts. It solves the problems of wall thermal memory interference model identification accuracy and decoupling failure caused by ill-conditioned gain matrix in movable layout scenarios, improving the system's adaptability to layout changes, the stability of the control process, and the operational safety in confined environments.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] In a first aspect, the present invention provides a method for temperature uniformity control of explosion-proof electrical appliances, comprising:

[0006] Based on the thermal steady-state condition, the initial reference baseline of the swarm intelligence model is obtained through initial orthogonal perturbation identification;

[0007] Based on the initial swarm intelligence model reference baseline and layout change trigger signal, the updated gain matrix and new layout reference baseline are obtained by combining the ant colony pheromone evaporation and deposition mechanism, the step-by-step anti-phase thermal pulse, and the updated orthogonal perturbation identification.

[0008] Based on the updated gain matrix and the new layout reference baseline, and according to the bat echolocation swarm intelligence mechanism, the suction cup power reference value is obtained through dual-mode collaborative control including bat echolocation frequency decoupling mode and pseudo-inverse decoupling mode.

[0009] Furthermore, based on the thermal steady-state condition, the initial swarm intelligence model reference baseline is obtained through initial orthogonal perturbation identification. This initial orthogonal perturbation identification includes Walsh orthogonal perturbation sequence synchronization excitation and orthogonal decoupling, comprising:

[0010] Based on thermal steady-state conditions, a perturbation temperature response dataset is obtained by synchronous excitation using a Walsh orthogonal perturbation sequence.

[0011] Based on the perturbation temperature response dataset, the initial pure gain matrix is ​​obtained through orthogonal decoupling calculation.

[0012] The initial swarm intelligence model reference baseline is obtained through non-volatile storage based on the initial pure gain matrix.

[0013] Furthermore, the process of obtaining the updated gain matrix and new layout reference baseline based on the initial swarm intelligence model reference baseline and layout change trigger signal, through ant colony pheromone evaporation and deposition mechanisms, step-wise anti-phase thermal pulses, and combined with updated orthogonal perturbation identification, includes:

[0014] Based on the layout change trigger signal, the real-time power of the suction cup and the temperature distribution of the wall surface are obtained and a thermal status profile is constructed. The layout change trigger signal is determined according to the online status of the suction cup communication node or user interface instructions.

[0015] Based on the initial swarm intelligence model reference baseline and thermal state archive, the old thermal memory elimination state is obtained through the ant colony pheromone evaporation and deposition mechanism, the step-by-step reverse phase thermal pulse, and the simultaneous execution of gas concentration safety monitoring.

[0016] Based on the old thermal memory elimination state, a new layout reference temperature vector is obtained by power reset and establishing a new layout surrounding quasi-steady state.

[0017] Based on the new layout reference temperature vector, the updated gain matrix and the new layout reference baseline are obtained by updating the orthogonal perturbation identification.

[0018] Furthermore, the process of obtaining the old thermal memory elimination state based on the initial swarm intelligence model reference baseline and thermal state profile, through ant colony pheromone evaporation and deposition mechanisms, step-by-step anti-phase thermal pulses, and simultaneous gas concentration safety monitoring, includes:

[0019] Based on the initial swarm intelligence model reference baseline and thermal state profile, an evaporation power execution sequence is obtained by applying a stepped anti-phase thermal pulse and calculating the amplitude of the anti-phase thermal pulse.

[0020] Based on the evaporation power execution sequence, the safety monitoring status of the evaporation process is obtained by performing gas concentration safety monitoring.

[0021] Based on the evaporation power execution sequence and the evaporation process safety monitoring status, the old thermal memory elimination status is obtained by comparing the deviation threshold.

[0022] Further, the step of obtaining the updated gain matrix and the new layout reference baseline based on the new layout reference temperature vector through updated orthogonal perturbation identification includes:

[0023] Based on the new layout reference temperature vector, the new layout perturbation temperature response dataset is obtained by synchronous excitation using the same Walsh orthogonal perturbation sequence as the initial orthogonal perturbation identification.

[0024] Based on the new layout perturbation temperature response dataset, the full pure gain matrix of the new layout is obtained through orthogonal decoupling calculation.

[0025] Based on the full pure gain matrix of the new layout, the updated gain matrix and the new layout reference baseline are obtained by replacing the matrix and storing it.

[0026] Furthermore, based on the updated gain matrix and the new layout reference baseline, and according to the bat echolocation swarm intelligence mechanism, the suction cup power reference value is obtained through dual-mode cooperative control including a bat echolocation frequency decoupling mode and a pseudo-inverse decoupling mode, including:

[0027] Based on the updated gain matrix and the new layout reference baseline, the target power reference value of the suction cup is obtained by quadratic programming optimization.

[0028] Based on the updated gain matrix and the target power reference value of the suction cup, the control mode selection signal and the smooth initial power reference value are obtained through power smooth transition scheduling.

[0029] Based on the new layout reference baseline, the jitter frequency range is determined, and the smooth initial power reference value is used as the bias reference. The independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying component of the temperature measurement point are obtained by sinusoidal jitter superposition and orthogonal demodulation of the sliding window.

[0030] Based on the control mode selection signal, the suction cup power reference value is obtained through dual-mode collaborative control.

[0031] Further, the step of obtaining the control mode selection signal and the smooth initial power reference value through power smoothing transition scheduling based on the updated gain matrix and the suction cup target power reference value includes:

[0032] Based on the updated gain matrix, the current condition number is obtained by calculating the ratio of the maximum singular value to the minimum singular value of the matrix;

[0033] A control mode selection signal is obtained by comparing the current condition number with a preset state determination threshold.

[0034] Based on the control mode selection signal and the target power reference value of the suction cup, a smooth initial power reference value is obtained by gradually switching within the transition window using a power smooth transition algorithm in a linear interpolation manner.

[0035] Further, the step of determining the jitter frequency range based on the new layout reference baseline, using the smoothed initial power reference value as a bias reference, and obtaining the independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components of the temperature measurement point through sinusoidal jitter superposition and orthogonal demodulation of the sliding window, includes:

[0036] Based on the new layout reference baseline, the jitter frequency is determined by calculating the thermal cutoff frequency and taking a safety factor, and then the jitter frequency of the suction cup is allocated by a geometric series to obtain a frequency allocation scheme.

[0037] Based on the frequency allocation scheme and the smoothed initial power reference value, the real-time output power of the suction cup containing the characteristic frequency identifier is obtained by superimposing a sinusoidal dithering signal.

[0038] Based on the real-time output power of the suction cup, the independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components of the temperature measurement point are obtained by orthogonal demodulation through a sliding window.

[0039] Furthermore, the suction cup power reference value is obtained through dual-mode collaborative control based on the control mode selection signal. The control mode selection signal includes a first mode selection signal and a second mode selection signal, including:

[0040] When the control mode selection signal is the first mode selection signal, the bat echo location frequency decoupling mode is selected, and the suction cup power reference value is obtained by frequency decoupling deviation distribution and incremental proportional-integral-derivative control.

[0041] When the control mode selection signal is the second mode selection signal, the pseudo-inverse decoupling mode is selected, and the suction cup power reference value is obtained by superimposing discrete proportional-integral-derivative control on the pseudo-inverse decoupling feedforward.

[0042] Furthermore, the step of obtaining the suction cup power reference value through frequency decoupling deviation allocation and incremental proportional-integral-derivative control includes:

[0043] Based on the gradually varying component of the temperature measurement point and the global temperature setpoint, the global temperature deviation is obtained by calculating the difference.

[0044] Based on the global temperature deviation and the independent heat transfer contribution amplitude matrix of the suction cup, the comprehensive temperature deviation of the suction cup is obtained by weighting the contribution ratio.

[0045] Based on the comprehensive temperature deviation of the suction cup, the reference value of the suction cup power in the bat echolocation frequency decoupling mode is calculated by incremental proportional-integral-derivative control.

[0046] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0047] This invention establishes an initial swarm intelligence model reference baseline based on initial orthogonal perturbation identification under thermal steady-state conditions. It then uses an ant colony pheromone evaporation and deposition mechanism, combined with a stepped antiphase heat pulse to update the orthogonal perturbation identification, to achieve adaptive updates of the heat transfer model. Based on the bat echolocation swarm intelligence mechanism, it realizes dual-mode collaborative control of bat echolocation frequency decoupling mode and pseudo-inverse decoupling mode. This achieves refined and robust uniform temperature control of the three-dimensional spatial thermal field under arbitrary layouts. It solves the problems of wall thermal memory interference model identification accuracy and decoupling failure and power oscillation caused by ill-conditioned gain matrix in portable heating systems, improving the system's rapid response to layout changes, full-condition control stability, and operational safety in enclosed special environments. Attached Figure Description

[0048] Figure 1 This is a flowchart of the temperature equalization control method for the explosion-proof electrical appliance of the present invention;

[0049] Figure 2 This is a structural diagram of a wall-mounted suction cup bracket;

[0050] Figure 3 A flowchart illustrating the steps involved in identifying the initial orthogonal perturbation to obtain the initial reference baseline for the swarm intelligence model;

[0051] Figure 4 The flowchart shows the steps to obtain the updated gain matrix and new layout reference baseline based on the layout change trigger signal through the ant colony pheromone evaporation and deposition mechanism.

[0052] Figure 5 A flowchart illustrating the steps for obtaining a suction cup power reference value using a dual-mode cooperative control system based on a bat echolocation swarm intelligence mechanism.

[0053] Reference numerals: 1. Main body of bracket; 2. Long hole for wall mounting; 3. Mounting position for wall temperature sensor; 4. Suction cup mounting hole; 5. Grounding terminal; 6. Communication interface position. Detailed Implementation

[0054] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof.

[0055] Identical parts are indicated by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, while the terms "bottom surface," "top surface," "inner," and "outer" refer to directions toward or away from the geometric center of a specific part, respectively.

[0056] The term "and / or" in the following text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0057] Example 1:

[0058] like Figure 1 As shown, this embodiment provides a method for temperature uniformity control of explosion-proof electrical appliances, including:

[0059] Based on the thermal steady-state condition, the initial reference baseline of the swarm intelligence model is obtained through initial orthogonal perturbation identification;

[0060] Based on the initial swarm intelligence model reference baseline and layout change trigger signal, the updated gain matrix and new layout reference baseline are obtained by combining the ant colony pheromone evaporation and deposition mechanism, the step-by-step anti-phase thermal pulse, and the updated orthogonal perturbation identification.

[0061] Based on the updated gain matrix and the new layout reference baseline, and according to the bat echolocation swarm intelligence mechanism, the suction cup power reference value is obtained through dual-mode collaborative control including bat echolocation frequency decoupling mode and pseudo-inverse decoupling mode.

[0062] In this embodiment, the explosion-proof electrical appliance is an explosion-proof electric heater. This heater is designed for enclosed environments and uses a modular wall-mounted infrared radiation electromagnetic induction suction cup matrix as its core heating unit. Each suction cup is an independent, controllable heating module of the explosion-proof heater. Each suction cup integrates an isolated drive module, a surface temperature sensor, and an ambient gas concentration monitoring unit. It is suspended from the wall by a bracket, and the suction cups are connected to a multi-channel intelligent controller via an industrial communication bus. The suction cups are suspended from the wall by a wall-mounted suction cup bracket, the specific structure of which is as follows: Figure 2As shown, the main body of the bracket (1) is made of galvanized steel plate bent and formed, and the surface is sprayed with an antistatic coating to meet the requirements of explosion-proof environment. The wall mounting holes (2) are evenly distributed along the length of the bracket, which facilitates the fine adjustment of the position during the installation process to ensure the flatness of the installation. The wall temperature sensor mounting position (3) is coaxially arranged with the wall mounting holes (2) and is used to embed the temperature sensor to monitor the temperature change of the contact point between the bracket and the wall in real time, and to provide data support for the determination of thermal steady state conditions and the elimination of thermal memory. The suction cup mounting holes (4) are evenly arranged along the length of the bracket and are used to fix the modular heating suction cup. The number and arrangement of the suction cups can be flexibly selected according to the actual heating needs. The grounding terminal (5) is set at the end of the bracket to achieve reliable explosion-proof safety grounding. The communication interface position (6) is reserved on the side of the bracket to install the explosion-proof gland to connect to the industrial communication bus and realize the data transmission and command interaction between each suction cup and the multi-channel intelligent controller. The multi-channel intelligent controller assigns a unique communication address to each suction cup and establishes a mapping table between physical location and address. It can flexibly adjust the number, installation position and arrangement of suction cups according to the actual size, layout and temperature requirements of the heating space. Through independent and precise control of the output power of each suction cup, it can achieve fine and uniform temperature control of the three-dimensional spatial temperature field.

[0063] Thermal steady-state conditions refer to a state where the wall temperature essentially remains unchanged. This is determined by real-time monitoring data from temperature sensors embedded in the contact points between the bracket and the wall. Thermal steady-state is achieved when the wall temperature change rate falls below a preset temperature change rate threshold within a continuous judgment period. The preset temperature change rate threshold ranges from 0.05℃ / min to 0.2℃ / min, typically set at 0.1℃ / min. The continuous judgment period ranges from 2 to 5 minutes, typically set at 3 minutes. Initial orthogonal perturbation identification is a multi-input multi-output system identification method based on orthogonal binary sequences. It is used to simultaneously acquire the independent heat transfer contribution of each suction cup to each spatial temperature measurement point without interrupting the heating process. Specifically, under thermal steady-state conditions, with a DC bias applied to the enclosing power reference value, an orthogonal binary perturbation sequence generated by the Walsh function as the excitation signal, and the perturbation amplitude set to a preset proportion of the enclosing power, all suction cups are synchronously driven and the spatial temperature response is collected. Then, the independent temperature of each suction cup is separated through orthogonal decoupling calculation. The contribution of the encirclement power reference value is a preset proportion of the rated power of the suction cup, which ranges from 30% to 40%, and is usually 35%. The proportion of the perturbation amplitude to the encirclement power ranges from 5% to 8%, and is usually 6%. The perturbation step size ranges from 30 to 60 seconds. The initial swarm intelligence model reference baseline is a set of reference parameters characterizing the heat transfer relationship between the suction cup and the spatial temperature measurement point and the thermal characteristics of the wall in the initial state of the system. It includes the initial pure gain matrix composed of the static heat transfer influence of each suction cup on each temperature measurement point obtained by orthogonal decoupling calculation, the wall reference temperature distribution at the end of the encirclement preheating, the spatial distance matrix of each suction cup, and the wall dominant thermal time constant. The wall dominant thermal time constant is identified by the heating curve. The identification method is to take the time required for the wall temperature to rise from the start of the encirclement preheating to the final steady state value of 63.2%. The wall dominant thermal time constant is stored in non-volatile memory as the basis for subsequent bat echolocation jitter frequency allocation.

[0064] This embodiment obtains the initial swarm intelligence model reference baseline through initial orthogonal perturbation identification based on thermal steady-state conditions, achieving accurate calibration of the independent heat transfer contribution of each suction cup under multi-suction cup coupled heat transfer conditions. By combining the ant colony pheromone evaporation and deposition mechanism with a stepped anti-phase heat pulse and updated orthogonal perturbation identification, it achieves rapid elimination of wall thermal memory and accurate reconstruction of the heat transfer model after layout changes. By applying the bat echolocation frequency adaptive mechanism to thermal field decoupling control, it achieves natural separation of the heat transfer contribution of each suction cup at the physical layer when the gain matrix is ​​ill-conditioned, solving the problem of errors in traditional pseudo-inverse decoupling. The problem of differential amplification and power oscillation is addressed. Smooth switching and coordinated control of dual modes are achieved through gain matrix condition number monitoring, forming a complete identification and control closed loop. Through dual-mode coordinated control, including a bat echolocation frequency decoupling mode and a pseudo-inverse decoupling mode, and by dynamically scheduling the decoupling strategy based on the gain matrix condition number, adaptive isothermal control is achieved, enabling stable decoupling without matrix inversion under ill-conditioned gain matrix conditions and efficient decoupling using pseudo-inversion under benign gain matrix conditions. Ultimately, under safety constraints, the uniformity of the three-dimensional temperature field and the system's adaptability to layout changes are improved.

[0065] Furthermore, this embodiment provides a step for obtaining an initial swarm intelligence model reference baseline based on thermal steady-state conditions through initial orthogonal perturbation identification. The initial orthogonal perturbation identification includes Walsh orthogonal perturbation sequence synchronization excitation and orthogonal decoupling, comprising:

[0066] Based on thermal steady-state conditions, a perturbation temperature response dataset is obtained by synchronous excitation using a Walsh orthogonal perturbation sequence.

[0067] Based on the perturbation temperature response dataset, the initial pure gain matrix is ​​obtained through orthogonal decoupling calculation.

[0068] The initial swarm intelligence model reference baseline is obtained through non-volatile storage based on the initial pure gain matrix.

[0069] Among them, the Walsh orthogonal perturbation sequence synchronous excitation is a method for synchronously exciting a multi-input multi-output system based on the orthogonality of the Walsh function. By assigning mutually orthogonal binary power perturbation sequences to each independent heating unit, it enables all heating units to be excited simultaneously without interference. This method is used to synchronously acquire the independent temperature response characteristics of each heating unit to each temperature measurement point in space without interrupting the normal heating process of the system. The perturbation temperature response dataset is a set of temperature values ​​of all space temperature measurement points synchronously collected at a fixed time step during the Walsh orthogonal perturbation sequence synchronous excitation process, containing information on the space temperature change caused by the power perturbation of each heating unit. Orthogonal decoupling is a method for separating the independent contributions of each channel of a multi-input multi-output system by utilizing the cross-correlation characteristics of orthogonal functions. Based on the orthogonality of the Walsh function, it uses cross-correlation... The calculation extracts the static heat transfer influence of a single heating unit on a single temperature measuring point from the mixed temperature response data, which is used to eliminate coupling interference between multiple heating units and obtain a pure system heat transfer model. The initial pure gain matrix is ​​a two-dimensional matrix characterizing the static heat transfer influence of each heating unit on each spatial temperature measuring point under the initial layout of the system. The larger the element value, the stronger the temperature regulation capability of the corresponding heating unit on the corresponding temperature measuring point. Non-volatile storage is a storage method that can retain data without loss after power failure. In this embodiment, it is implemented by the flash memory chip built into the controller to store the system reference parameters, ensuring that the system can run normally without recalibration after restart. The initial swarm intelligence model reference baseline is the set of thermal characteristic reference parameters under the initial layout of the system, which serves as a reference for model updates and real-time temperature equalization control after subsequent layout changes.

[0070] Specifically, such as Figure 3 As shown, firstly, based on the thermal steady-state condition, a perturbation temperature response dataset is obtained through synchronous excitation using a Walsh orthogonal perturbation sequence. Once the system reaches the wall's thermal steady state, the controller adjusts the parameters based on the number of suction cups. generate Group length is The Walsh orthogonal binary perturbation sequence, where the sequence length is... satisfy To ensure orthogonality, the sequence length can take values ​​ranging from 1 to 2. to The value is usually taken as The perturbation amplitude is taken from the enclosing power reference value. The preset ratio, the preset ratio value range is to The value is usually taken as That is, the maximum fluctuation of the perturbation power is perturbation step size The range of values ​​is Instant Seconds, typically taken as a value Seconds ensure that the temperature response reaches a quasi-steady state within each perturbation step; all suction cups surround the power reference value. With DC bias, they synchronously execute their respective Walsh quadrature perturbation sequences, i.e., the... The suction cup in the first The output power of each perturbation step is ,in For the first The suction cup in the first The perturbation power of a step size, This represents the perturbation amplitude; at the end of each perturbation step, the controller synchronously acquires all... Temperature values ​​at various spatial temperature measurement points ,in , The dimension is obtained as Perturbation temperature response dataset .

[0071] Then, based on the perturbation temperature response dataset, the initial pure gain matrix is ​​obtained through orthogonal decoupling calculation. This orthogonal decoupling calculation is derived based on the orthogonality of the Walsh function, meaning that the cross-correlation between any two distinct Walsh sequences is zero, and the autocorrelation is equal to the sequence length. The initial pure gain matrix... elements pass Calculated; where Temperature measurement point at the end of preheating. The steady-state temperature, i.e., the temperature measurement point under thermal steady-state conditions. The reference temperature; For the first The suction cup in the first Perturbation power of a perturbation step size; molecule as temperature measurement point Temperature fluctuations and suction cups The cross-correlation of the perturbation sequences characterizes the suction cup. The effect of power change on the temperature measurement point The total contribution of temperature change; the denominator is the suction cup. The autocorrelation of the perturbation sequence is used for normalization; since the wall is in a thermal steady state and the background temperature fluctuation is zero, the calculated... Suction cups only Temperature measurement point The independent radiation and convective heat transfer contributions, without any coupling interference terms; for example, when , , When, the dimension can be calculated as The initial pure gain matrix, where each element represents the steady-state temperature change at the corresponding temperature measurement point caused by the unit power change of the corresponding suction cup, with units of ℃ / W.

[0072] Finally, based on the initial pure gain matrix, the initial swarm intelligence model reference baseline is obtained through non-volatile storage. The controller then calculates the initial pure gain matrix. The baseline temperature distribution of the wall surface at the end of the preheating process. Spatial distance matrix of each suction cup and the dominant thermal time constant of the wall The data is packaged into an initial swarm intelligence model reference baseline and written into the controller's built-in non-volatile flash memory chip; among which, the wall reference temperature distribution... This is the set of temperature values ​​collected by all temperature sensors embedded in the contact points between the bracket and the wall under thermal steady-state conditions; the spatial distance matrix of each suction cup. elements For the first The suction cup and the first The Euclidean distance between the suction cups is calculated by the controller based on the input installation coordinates of each suction cup; the wall's dominant thermal time constant. The temperature rise curve was used to identify the wall surface temperature from the start of preheating to its final steady-state value. The required time, typically within a certain range, is... minutes to The time varies depending on the wall material and thickness, ranging from minutes. After storage, the controller generates a storage check code to ensure data integrity. This baseline data can be read at any time during subsequent system operation for model updates and control calculations. In this embodiment, parameters such as sequence length, perturbation amplitude ratio, and perturbation step size are merely examples. Those skilled in the art can adjust these parameters according to actual environmental conditions, the number of suction cups, and the number of temperature measurement points. This embodiment does not impose any limitations on these adjustments.

[0073] This embodiment achieves simultaneous excitation of multiple suction cups without interference by employing Walsh orthogonal perturbation sequence synchronous excitation, significantly shortening the system identification time. Through orthogonal decoupling calculation based on the orthogonality of Walsh functions, coupled heat transfer interference between multiple suction cups is eliminated, resulting in a pure initial gain matrix. By storing the initial pure gain matrix along with wall thermal characteristic parameters and suction cup spatial parameters as the initial swarm intelligence model reference baseline, a unified benchmark is provided for subsequent incremental model updates and real-time temperature control after layout changes. Non-volatile storage ensures the reliability and persistence of the benchmark data, avoiding the tedious process of repeated calibration after system restarts, and improving the system's initialization efficiency and operational stability.

[0074] Furthermore, this embodiment provides a step-by-step approach to obtain an updated gain matrix and a new layout reference baseline based on an initial swarm intelligence model reference baseline and a layout change trigger signal, through ant colony pheromone evaporation and deposition mechanisms, a stepped anti-phase thermal pulse, and updated orthogonal perturbation identification. The steps include:

[0075] Based on the layout change trigger signal, the real-time power of the suction cup and the temperature distribution of the wall surface are obtained and a thermal status profile is constructed. The layout change trigger signal is determined according to the online status of the suction cup communication node or user interface instructions.

[0076] Based on the initial swarm intelligence model reference baseline and thermal state archive, the old thermal memory elimination state is obtained through ant colony pheromone evaporation and deposition mechanisms, step-by-step anti-phase thermal pulses, and simultaneous gas concentration safety monitoring; specifically including:

[0077] Based on the initial swarm intelligence model reference baseline and thermal state archive, the evaporation power execution sequence is obtained by applying a stepped anti-phase thermal pulse and calculating the amplitude of the anti-phase thermal pulse; based on the evaporation power execution sequence, the safety monitoring status of the evaporation process is obtained by performing gas concentration safety monitoring; based on the evaporation power execution sequence and the safety monitoring status of the evaporation process, the old thermal memory elimination status is obtained by comparing the deviation threshold.

[0078] Based on the old thermal memory elimination state, the new layout reference temperature vector is obtained by power reset and establishing a new layout surrounding quasi-steady state.

[0079] Based on the new layout reference temperature vector, the updated gain matrix and the new layout reference baseline are obtained through updated orthogonal perturbation identification; specifically including:

[0080] Based on the new layout reference temperature vector, the new layout perturbation temperature response dataset is obtained by synchronous excitation using the same Walsh orthogonal perturbation sequence as the initial orthogonal perturbation identification. Based on the new layout perturbation temperature response dataset, the new layout full pure gain matrix is ​​obtained by orthogonal decoupling calculation. Based on the new layout full pure gain matrix, the updated gain matrix and the new layout reference baseline are obtained by replacing the matrix and storing it.

[0081] Among them, the layout change trigger signal is a control signal indicating that the physical layout of the system suction cups has changed, and is determined according to the online status of the suction cup communication node or user interface instructions; the real-time suction cup power is the actual output power value of each suction cup at the moment the layout change is triggered, which is acquired in real time through the industrial communication bus between the controller and each suction cup; the wall temperature distribution is the set of temperature values ​​collected by all temperature sensors embedded in the contact point between the bracket and the wall at the moment the layout change is triggered, which is obtained through synchronous sampling by the embedded temperature sensors; the thermal status archive is a data set recording the thermal operating status of the system at the moment of the layout change, which is obtained by freezing and storing the current suction cup real-time power vector and the wall temperature distribution data by the controller.

[0082] The ant colony pheromone evaporation and deposition mechanism is a model adaptive update mechanism designed by drawing on the dynamic pheromone update principle in the ant colony algorithm. It maps the elimination of the old layout's thermal memory to the active evaporation process of ant colony pheromones, and the establishment of the new layout's thermal characteristics to the redeposition process of ant colony pheromones. This mechanism enables rapid and interference-free adaptive updates of the system's heat transfer model after layout changes. The stepped anti-phase heat pulse is a safe thermal excitation method for actively eliminating residual thermal memory in the walls. By applying a stepped power pulse with the opposite direction to the old power bias, it quickly cancels the residual temperature gradient in the walls, eliminating interference from the old layout's thermal memory on the new model's identification without causing power abrupt changes. The gas concentration safety monitoring is a full-process embedded safety monitoring mechanism. By synchronously collecting data from the environmental gas concentration monitoring units built into each suction cup and executing graded response logic, it ensures the smooth operation of the heat pulse execution process. Safety is paramount; the old thermal memory elimination state is the system state in which the residual temperature distribution of the wall has been basically restored to the initial reference state; the anti-phase thermal pulse amplitude is the power adjustment amount used to offset the old power bias, which is calculated by the enclosing power reference value in the initial swarm intelligence model reference baseline and the real-time power of the suction cups in the old layout thermal state file; the evaporation power execution sequence is the step-like power change sequence of each suction cup in the thermal memory elimination stage, which is generated by the anti-phase thermal pulse amplitude and the step-like power change rules; the evaporation process safety monitoring state is the system operation state in which the gas concentration is within a safe range during the thermal memory elimination stage, which is determined by the gas concentration safety monitoring and graded response logic; the deviation threshold comparison is the process of calculating the deviation between the current wall temperature distribution and the wall reference temperature distribution in the initial swarm intelligence model reference baseline and comparing it with a preset threshold, used to determine whether the old thermal memory has been basically eliminated.

[0083] Power reset is the operation of synchronously restoring the output power of all suction cups to the bounding power reference value, which is used to establish a unified thermal reference under the new layout; the new layout bounding quasi-steady state is the system thermal state under the new layout where the temperature change rate of all spatial temperature measurement points is lower than a preset threshold, which is determined by continuously monitoring the temperature change rate of the spatial temperature measurement points; the new layout reference temperature vector is the set of temperature values ​​of all spatial temperature measurement points under the new layout bounding quasi-steady state, which is obtained by synchronously collecting spatial temperature measurement point data at the quasi-steady state moment.

[0084] The updated orthogonal perturbation identification process is performed under the new layout's enclosed quasi-steady state after the layout change. It uses the same Walsh orthogonal perturbation sequence as the initial orthogonal perturbation identification, calculating the incremental temperature response with the new layout's reference temperature vector as a reference. This is used to obtain the pure gain matrix that truly reflects the heat transfer characteristics of the new layout. The new layout perturbation temperature response dataset is the set of incremental temperature values ​​collected at each temperature measurement point during the updated orthogonal perturbation identification process. It is obtained by synchronously collecting the spatial temperature response under perturbation excitation and subtracting the new layout's reference temperature vector. The new layout's full pure gain matrix characterizes the static heat transfer influence of each suction cup on each spatial temperature measurement point under the new layout. The two-dimensional matrix is ​​obtained by performing orthogonal decoupling calculations on the new layout perturbation temperature response dataset; the replacement matrix is ​​stored as the operation of replacing the current effective gain matrix in the controller with the full pure gain matrix of the new layout, and storing the new layout reference parameters in non-volatile memory, which is used to complete the update and persistence of the system thermal model; the updated gain matrix is ​​the full pure gain matrix of the new layout, which serves as the core model parameter for the uniform temperature control under the new layout; the new layout reference baseline is the set of thermal characteristic reference parameters under the new layout, including the new layout reference temperature vector, the current wall temperature distribution, and the updated gain matrix, which serves as the reference for real-time uniform temperature control under the new layout.

[0085] Specifically, such as Figure 4 As shown, firstly, based on the layout change trigger signal, the real-time power of the suction cup and the wall temperature distribution are obtained and a thermal state profile is constructed. The controller then polls at a fixed interval. Continuously poll the online status of each suction cup node on the industrial communication bus, with a polling cycle of [period missing]. The value ranges from 100ms to 500ms, with a typical value of 200ms; simultaneously, it monitors layout change confirmation commands input from the user interface in real time, and detects when the number of suction cup communication nodes decreases from the initial number. Become a new quantity ( When the controller receives a layout change confirmation command from the user, it immediately generates a layout change trigger signal, and at the moment the signal is generated... Freeze the real-time power vector of all current suction cups. and the points where the bracket contacts the wall Wall temperature distribution collected by a temperature sensor The above data, along with the trigger time timestamp, is packaged into an old layout hot state file and stored in the controller's temporary storage.

[0086] Then, based on the initial swarm intelligence model reference baseline and thermal state archive, the old thermal memory elimination state is obtained through ant colony pheromone evaporation and deposition mechanisms, step-by-step anti-phase thermal pulses, and simultaneous gas concentration safety monitoring. This is first based on the enclosing power reference value in the initial swarm intelligence model reference baseline. With the suction cup real-time power vector in the thermal state file Calculate the amplitude of the inverse thermal pulse. In the pheromone evaporation and deposition mechanism of ant colonies, the pheromone concentration is mapped to the residual temperature deviation at each temperature measurement point on the wall; a higher pheromone concentration indicates a stronger residual thermal memory at that location. The evaporation power of each suction cup is through The calculation yielded, where The evaporation intensity coefficient is used to adjust the intensity of the reverse-phase thermal pulse, with a value ranging from 0.3 to 0.5, typically 0.4. The pheromone deposition trigger condition is: after the new layout surrounds the quasi-steady state, the deposition process is triggered when the temperature change rate at any temperature measurement point is detected to be below 0.05℃ / min for three consecutive control cycles; the deposition rate... according to Calculation, where The deposition rate coefficient ranges from 0.01 to 0.03. The current average wall temperature is 0°C. The coupling logic between evaporation and deposition is as follows: after the evaporation stage is completed, if the maximum residual temperature deviation is less than 0.5°C, the system will automatically switch to the deposition stage. During the deposition stage, the system will continue to operate at the power reference value while monitoring the temperature change rate of each temperature measuring point. The deposition is considered complete when the temperature change rate of all temperature measuring points is lower than the preset threshold.

[0087] The evaporation power calculation formula is derived based on the principle of ant colony pheromone evaporation. By applying a power adjustment amount opposite to the direction of the old power bias, the residual temperature gradient of the wall is actively offset. Then, a stepped evaporation power execution sequence is generated, with all power changes using a stepped increase / decrease method, and the power change amount at each step... Not exceeding the rated power of the suction cup 5%, duration of each step The value range is from 5 to 15 seconds, with a typical value of 10 seconds to ensure a smooth and abrupt power change. During the evaporation power operation, the controller simultaneously performs gas concentration safety monitoring, collecting data from the ambient gas concentration monitoring units built into all suction cups in each control cycle. ( ), and the preset warning threshold and alarm threshold Comparison, among which the warning threshold The value is set at 20% to 30% of the lower explosive limit of a specific gas in the corresponding environment, typically 25%, for the alarm threshold. The value is taken as 40% to 60% of the lower explosive limit of a specific gas in the corresponding environment, and is usually taken as 50%; if the concentration of any gas exceeds the warning threshold, the warning threshold will be triggered. Immediately terminate the evaporation process and reduce all suction cup power to a low-power maintenance level. Low power consumption maintenance power The value is taken as 5% to 15% of the suction cup's rated power, typically 10%; if any gas concentration is detected to exceed the alarm threshold... Immediately execute the system shutdown procedure and trigger an audible and visual alarm; continuously monitor the wall surface temperature distribution. Calculate the current wall temperature and the wall reference temperature distribution in the initial swarm intelligence model reference baseline. Maximum deviation , when the maximum deviation When the preset temperature deviation threshold is reached and the state lasts for at least 1 minute, it is determined that the old thermal memory has been basically eliminated, and the old thermal memory elimination state is obtained. The temperature deviation threshold is determined comprehensively based on the wall thermal steady-state identification accuracy requirements, the temperature sensor measurement accuracy, and the maximum allowable error of subsequent gain matrix identification. The value ranges from 0.3℃ to 0.8℃, and is usually 0.5℃.

[0088] Then, based on the old thermal memory elimination state, a new layout reference temperature vector is obtained by power reset and establishing a new layout surrounding quasi-steady state. The controller then synchronously switches the output power of all suction cups back to the surrounding power reference value. And it continues to operate at that power; real-time monitoring of all Temperature change rate at each spatial temperature measurement point ( The temperature change rate is calculated by dividing the temperature difference between two adjacent control cycles by the control cycle. Calculations show that when the temperature change rate at all temperature measurement points is lower than the preset temperature change rate threshold within the continuous judgment period, the judgment system enters the enclosed quasi-steady state under the new layout. At this time, the controller synchronously collects the temperature values ​​of all spatial temperature measurement points to obtain the new layout reference temperature vector. The new layout reference temperature vector represents the quasi-steady-state temperature distribution of the new layout under the bounding power reference value and when the wall surface is under the initial reference temperature background.

[0089] Finally, based on the new layout reference temperature vector, the updated gain matrix and the new layout reference baseline are obtained through updated orthogonal perturbation identification. First, the power reference value is enclosed. With DC bias applied, the controller again synchronously applies the same Walsh orthogonal perturbation sequence as the initial orthogonal perturbation identification to each suction cup, with the sequence length remaining the same. At the end of each perturbation step, the controller synchronously collects the temperature values ​​of all space temperature measurement points. ( , ), and calculate the incremental temperature response at each temperature measurement point. The dimension is obtained as New layout perturbation temperature response dataset Next, based on the orthogonality of the Walsh function, orthogonal decoupling calculations are performed on the new layout perturbation temperature response dataset to obtain the full pure gain matrix of the new layout. Its elements pass Calculated; where For the first The suction cup in the first The perturbation power for each perturbation step size is exactly the same as the perturbation sequence used in the initial orthogonal perturbation identification; since the old thermal memory has been completely eliminated, the calculated... Includes only the new layout lower suction cup Temperature measurement point The independent radiation and convective heat transfer contributions are calculated, without any historical interference terms; finally, a replacement matrix and storage operation are performed, and a fully pure gain matrix with a new layout is generated. Replace the currently active gain matrix in the controller and the new layout reference temperature vector Current wall surface temperature distribution and updating the gain matrix The data is packaged into a new layout reference baseline and written to the controller's built-in non-volatile flash memory chip, while a storage checksum is generated to ensure data integrity; for example, when the number of new suction cups... Number of temperature measurement points Sequence length When, the dimension can be calculated as The updated gain matrix, where each element physically represents the steady-state temperature change at the corresponding temperature measurement point caused by a unit power change in the corresponding suction cup, is expressed in °C / W. In this embodiment, parameters such as the polling period, evaporation intensity coefficient, step size of the stepped power change, and temperature deviation threshold are merely examples. Those skilled in the art can adjust these parameters according to actual environmental conditions, wall thermal characteristics, and safety requirements; this embodiment does not impose any limitations on these adjustments.

[0090] This embodiment applies the ant colony pheromone evaporation and deposition mechanism to model updates after layout changes, achieving proactive and rapid elimination of old thermal memories and accurate reconstruction of the new model. It eliminates the need to passively wait for the walls to cool naturally, significantly shortening the model reconstruction time after layout changes. By employing a stepped anti-phase thermal pulse technology, it avoids the safety risks caused by power surges while ensuring the efficiency and effectiveness of thermal memory elimination. Simultaneous execution of full-process gas concentration safety monitoring and graded response ensures operational safety during the model update process. The use of an updated orthogonal perturbation identification method consistent with the initial identification ensures the consistency and comparability of the old and new models, improving the system's layout adaptability and control stability.

[0091] Furthermore, this embodiment provides a step for obtaining a suction cup power reference value based on an updated gain matrix and a new layout reference baseline, according to the bat echolocation swarm intelligence mechanism, through dual-mode cooperative control including a bat echolocation frequency decoupling mode and a pseudo-inverse decoupling mode, including:

[0092] Based on the updated gain matrix and the new layout reference baseline, the target power reference value of the suction cup is obtained by quadratic programming optimization.

[0093] Based on the updated gain matrix and the target power reference value of the suction cup, the control mode selection signal and the smooth initial power reference value are obtained through power smooth transition scheduling; specifically including:

[0094] Based on the updated gain matrix, the current condition number is obtained by calculating the ratio of the maximum singular value to the minimum singular value of the matrix; the control mode selection signal is obtained by comparing the current condition number with the preset state judgment threshold; based on the control mode selection signal and the suction cup target power reference value, the power smooth transition algorithm is used to gradually switch in the transition window in a linear interpolation manner to obtain a smooth initial power reference value.

[0095] The jitter frequency range is determined based on the new layout reference baseline, with a smoothed initial power reference value as the bias reference. The independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components at the temperature measurement point are obtained through sinusoidal jitter superposition and orthogonal demodulation via a sliding window; specifically including:

[0096] Based on the new layout reference baseline, the jitter frequency is determined by calculating the thermal cutoff frequency and taking a safety factor. Then, the jitter frequency of the suction cup is allocated by a geometric series to obtain a frequency allocation scheme. Based on the frequency allocation scheme and the smoothed initial power reference value, the real-time output power of the suction cup containing the characteristic frequency identifier is obtained by superimposing a sinusoidal jitter signal. Based on the real-time output power of the suction cup, the independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components of the temperature measurement point are obtained by orthogonal demodulation with a sliding window.

[0097] Based on the control mode selection signal, a suction cup power reference value is obtained through dual-mode collaborative control; the control mode selection signal includes a first mode selection signal and a second mode selection signal; specifically, it includes:

[0098] When the control mode selection signal is the first mode selection signal, the bat echo localization frequency decoupling mode is selected. The suction cup power reference value is obtained through frequency decoupling deviation distribution and incremental proportional-integral-derivative control; specifically including:

[0099] Based on the gradually varying component of the temperature measurement point and the global temperature setpoint, the global temperature deviation is obtained by calculating the difference; based on the global temperature deviation and the independent heat transfer contribution amplitude matrix of the suction cup, the comprehensive temperature deviation of the suction cup is obtained by weighting the contribution ratio; based on the comprehensive temperature deviation of the suction cup, the suction cup power reference value in the bat echolocation frequency decoupling mode is calculated by incremental proportional-integral-derivative control.

[0100] When the control mode selection signal is the second mode selection signal, the pseudo-inverse decoupling mode is selected. The suction cup power reference value is obtained by superimposing the pseudo-inverse decoupling feedforward with discrete proportional-integral-derivative control.

[0101] Among them, quadratic programming optimization is a mathematical optimization method for finding the minimum value of a constrained quadratic function. In this embodiment, it is used to solve the suction cup power allocation scheme that makes the spatial temperature distribution most uniform under power and rate of change constraints. The suction cup target power reference value is the ideal output power value of each suction cup that makes the spatial temperature distribution most uniform when the mode switching transition process is not considered. It is obtained by solving the quadratic programming optimization. Power smooth transition scheduling is a scheduling mechanism to achieve shock-free switching between different control modes, used to avoid power jumps during mode switching. The control mode selection signal is a control signal that indicates which decoupling control mode the system should currently adopt. It is determined based on the comparison result of the condition number of the updated gain matrix and the preset state judgment threshold. The smooth initial power reference value is the initial power value after smoothing during the mode switching process. The algorithm calculates the singular values, which are numerical values ​​in linear algebra that characterize the properties of a matrix and describe the degree to which the matrix stretches or compresses a vector. These values ​​are obtained by performing singular value decomposition on the gain matrix. The current condition number is the ratio of the maximum to the minimum singular value of the gain matrix, used to measure the ill-conditioning of the matrix. The larger the ratio, the more ill-conditioning the matrix. The preset state judgment threshold is a critical value used to determine whether the gain matrix is ​​ill-conditioning. It is preset according to the system's requirements for control stability, with a value range of 50 to 200, and usually 100. The power smooth transition algorithm is an algorithm that achieves smooth power changes through linear interpolation within a transition window, used to eliminate power abrupt changes during mode switching. The transition window is the time interval for smooth power transition during mode switching, with a value range of 5 to 15 control cycles, and usually 10 control cycles.

[0102] The jitter frequency range is the frequency interval of the sinusoidal jitter signal assigned to each suction cup, determined based on the wall's thermal cutoff frequency and engineering feasibility; the bias reference is the DC power reference value superimposed on the sinusoidal jitter signal, i.e., the smoothed initial power reference value; sinusoidal jitter superposition is the operation of superimposing small sinusoidal power fluctuations on the DC power reference value, used to assign a unique frequency identifier to each suction cup; the thermal cutoff frequency is the cutoff frequency of the wall's thermal response system, characterizing the wall's response capability to thermal excitation at different frequencies, calculated through the wall's dominant thermal time constant; the safety factor is a coefficient used to ensure that the jitter frequency is much higher than the wall's thermal cutoff frequency, ranging from 3 to 10, typically taken as 5; the suction cup jitter frequency is the frequency of the unique sinusoidal jitter signal assigned to each suction cup, within the jitter frequency range. The frequency allocation scheme is a set of specific values ​​for the jitter frequencies of each suction cup, obtained through a geometric progression. The characteristic frequency identifier is a unique jitter frequency for each suction cup, used to distinguish the heat transfer contribution of different suction cups in the frequency domain. The real-time output power of the suction cups is the actual output power of each suction cup after superimposing a sinusoidal jitter signal. Sliding window orthogonal demodulation is a signal processing method that extracts the amplitude of specific frequency components from a mixed temperature signal, used to separate the independent heat transfer contribution of each suction cup. The independent heat transfer contribution amplitude matrix of the suction cups is a two-dimensional matrix characterizing the magnitude of the independent heat transfer contribution of each suction cup to each temperature measurement point, obtained through sliding window orthogonal demodulation. The slowly varying component at the temperature measurement point is the component in the temperature signal with a frequency lower than the lowest jitter frequency, reflecting the overall trend of temperature change in the space.

[0103] The dual-mode cooperative control is an adaptive control mechanism that dynamically switches between bat echolocation frequency decoupling mode and pseudo-inverse decoupling mode based on the gain matrix condition number, used to achieve stable and efficient temperature equalization control under different operating conditions; the suction cup power reference value is the target power value of each suction cup finally output by the controller, which is sent to the suction cup drive module for execution after being limited; the first mode selection signal is a control signal indicating that the system adopts the bat echolocation frequency decoupling mode, generated when the gain matrix condition number is greater than the preset state judgment threshold; the bat echolocation frequency decoupling mode is a frequency domain decoupling control mode designed based on the bat echolocation principle, which achieves natural separation of the heat transfer contribution of each suction cup in the frequency domain by assigning a unique characteristic frequency to each suction cup, used to achieve stable temperature equalization control when the gain matrix is ​​ill-conditioned; the frequency decoupling deviation allocation is the process of distributing the global temperature deviation to each suction cup according to the heat transfer contribution ratio of each suction cup, used to generate the independent control deviation of each suction cup; incremental proportional-integral-derivative control is a calculation-only control... The proportional-integral-derivative (PID) control algorithm for the incremental quantity is used to avoid integral saturation and abrupt changes in the control quantity. The global temperature setpoint is the user-desired average spatial temperature value, determined through input via the controller's human-machine interface. The global temperature deviation is the difference between the global temperature setpoint and the gradually varying components at each temperature measurement point. The suction cup comprehensive temperature deviation is the comprehensive control deviation allocated to each suction cup, obtained through frequency decoupling deviation allocation. The second mode selection signal is a control signal indicating that the system adopts the pseudo-inverse decoupling mode, generated when the gain matrix condition number is less than or equal to the preset state judgment threshold. The pseudo-inverse decoupling mode is a conventional mode for decoupling control of a multi-input multi-output system using the Moore-Penrose pseudo-inverse of the gain matrix, used to achieve fast and efficient temperature equalization control when the gain matrix is ​​in a good state. The pseudo-inverse decoupling feedforward superimposed discrete PID control is a control strategy that superimposes the feedforward decoupling component and the feedback adjustment component, where the feedforward component is used to quickly eliminate spatial temperature deviation, and the feedback component is used to stabilize the suction cup surface temperature.

[0104] Specifically, such as Figure 5 As shown, firstly, based on the updated gain matrix and the new layout reference baseline, the target power reference value of the suction cup is obtained through quadratic programming optimization. The controller then sets the control cycle. Seconds, each control cycle collects all current data. Temperature vector of each spatial temperature measurement point Calculate the maximum temperature difference in the space. If the maximum temperature difference in the space Greater than the preset temperature difference threshold Among them, the preset temperature difference threshold The value of is in the range of 3℃ to 8℃, and is usually taken as 5℃. Therefore, a constrained quadratic programming problem is constructed:

[0105]

[0106]

[0107]

[0108] in, Set the global temperature value. It is a vector of all 1s. The power adjustment amount for each suction cup. and These are the minimum and maximum permissible operating power of the suction cup, respectively. The maximum allowable power change rate between adjacent periods is defined as 5% to 15% of rated power per second, typically set to 10% of rated power per second. The power adjustment amount for each suction cup is obtained by solving this quadratic programming problem using the effective set method. Update the suction cup target power reference value. If the maximum temperature difference in the space Less than or equal to If so, the target power reference value of the suction cup in the previous cycle remains unchanged.

[0109] Then, based on the updated gain matrix and the target power reference value of the suction cup, the control mode selection signal and the smooth initial power reference value are obtained through power smooth transition scheduling. The controller then performs this process every [period]. Calculate the current gain matrix in each control cycle. The condition number, where The value ranges from 5 to 20, and is usually 10; for the current gain matrix Singular value decomposition yields ,in It is a singular value matrix, and ; Calculate the current condition number ;Will Compared with the preset state determination threshold Compare, if the current condition number Greater than the preset state determination threshold Generate the first mode selection signal; if the current condition number Less than or equal to the preset state determination threshold When the control mode selection signal changes, a second mode selection signal is generated; when the control mode selection signal changes, a power smooth transition algorithm is executed, and the transition window duration is set. One control cycle, of which The value ranges from 5 to 15, with 10 being the most common value; within the transition window, linear interpolation is used to calculate the value from the current power. Gradually transition to the initial calculated values ​​of the target mode , No. The smoothed initial power reference value for each transition cycle is After the transition window ends, the calculated value of the target mode is directly used as the initial power reference value for smoothing.

[0110] Then, based on the new layout reference baseline, the jitter frequency range is determined, with the smoothed initial power reference value as the bias reference. The amplitude matrix of the independent heat transfer contribution of the suction cup and the slowly varying components of the temperature measurement point are obtained by sinusoidal jitter superposition and orthogonal demodulation of the sliding window. First, the dominant thermal time constant of the wall is read from the new layout reference baseline. Calculate the thermal cutoff frequency of the wall Take a safety factor Determine the minimum jitter frequency Maximum jitter frequency ;exist Distribute within the range according to a geometric sequence The vibration frequency of the suction cup, of which the first... The vibration frequency of each suction cup is Ensure that the difference between any two frequencies is greater than the reciprocal of the resolution of the locked analysis window. The duration of locking the analysis window ; to smooth the initial power reference value Using the bias reference, a sinusoidal dithering signal of the corresponding frequency is superimposed on each suction cup, where the th... The real-time output power of each suction cup is ;in The jitter amplitude is taken as 1% to 2% of the suction cup's rated power, typically 1.5%; for the temperature signal at each spatial temperature measurement point. To process this, first use a cutoff frequency of... First-order low-pass filter for extracting slowly varying components The filter transfer function is ,in Then, subtract the slowly varying component from the original temperature signal to obtain the fluctuation component. ; with length as The sliding window performs orthogonal demodulation on the wave components and extracts the first wave. The amplitude corresponding to each frequency The dimension is obtained suction cup independent heat transfer contribution amplitude matrix Simultaneously output the slowly varying components at each temperature measurement point. .

[0111] Finally, based on the control mode selection signal, the suction cup power reference value is obtained through dual-mode collaborative control. When the control mode selection signal is the first mode selection signal, the bat echo localization frequency decoupling mode is adopted. In the bat echo localization frequency decoupling mode, the physical correspondence of the bat echo is the characteristic frequency sinusoidal jitter signal superimposed in the output power of each suction cup and the temperature response component of the same frequency caused at the temperature measurement point; the target recognition corresponds to separating each frequency component from the mixed temperature signal through quadrature demodulation of a sliding window, thereby identifying the independent heat transfer contribution of each suction cup. The correspondence between frequency encoding and suction cup spatial position is as follows: the suction cup number corresponds one-to-one with the jitter frequency. The lower the frequency, the closer the spatial position is to the center area of ​​the wall, and the higher the frequency, the closer it is to the edge area. The frequency allocation scheme is stored in the controller as a frequency-position mapping table. The mathematical principle of the bat echo localization frequency decoupling mode in solving the ill-conditioned problem of the gain matrix is: when the gain matrix is ​​ill-conditioned, its condition number is much greater than 1, and the amplification factor of the measurement noise by the traditional pseudo-inverse decoupling tends to infinity as the degree of ill-conditioning intensifies. The frequency decoupling mode assigns a unique characteristic frequency to each suction cup, naturally orthogonally separating them in the frequency domain. The heat transfer contribution extraction of each channel depends only on the amplitude demodulation of the corresponding frequency. The amplitude estimation error is independent of the matrix singular values, and the error amplification factor remains constant. This bypasses the matrix inversion operation at the physical layer.

[0112] Calculate the global temperature deviation vector The global temperature deviation is distributed according to the heat transfer contribution ratio of each suction cup. The overall temperature deviation of the suction cups is ,in The average temperature weight of the spatial temperature measurement points and satisfying Equal weights are usually used. , To prevent tiny positive numbers with a denominator of zero, the value is taken as... ;by Incremental proportional-integral-derivative (PID) control is performed on the input to calculate the power correction. in , , These are the proportional, integral, and derivative coefficients, which are tuned according to the system response characteristics and are typically set to values ​​of [value missing]. , , Update suction cup power reference value When the control mode selection signal is the second mode selection signal, a pseudo-inverse decoupling mode is adopted, and the Moore-Penrose pseudo-inverse of the gain matrix is ​​calculated. ; Calculate the feedforward decoupling power components Simultaneously, the surface temperature of each suction cup is collected. Calculate surface temperature deviation ,in Set the suction cup surface temperature; execute discrete proportional-integral-derivative control to obtain the feedback regulation power component. The suction cup power reference value is obtained by superimposing the feedforward and feedback components. Finally, the power reference values ​​for each suction cup were limited to... Within the specified range, data is transmitted via an industrial communication bus to each suction cup isolated drive module for execution. In this embodiment, parameters such as control cycle, temperature difference threshold, condition number threshold, transition window duration, and jitter amplitude are merely examples. Those skilled in the art can adjust these parameters according to actual environmental conditions, system response requirements, and safety standards; this embodiment does not impose any limitations on these adjustments.

[0113] This embodiment, drawing on the frequency adaptive principle of bat echolocation, designs a naturally decoupled control mode in the frequency domain. Bypassing matrix inversion operations at the physical layer, it solves the error amplification and power oscillation problems of pseudo-inverse decoupling when the gain matrix is ​​ill-conditioned. Automatic and smooth switching between the bat echolocation frequency decoupling mode and the pseudo-inverse decoupling mode is achieved through gain matrix condition number monitoring. The speed of pseudo-inverse decoupling is maintained when the gain matrix is ​​well-conditioned, while control stability is ensured when the gain matrix is ​​ill-conditioned. Incremental proportional-integral-derivative control and full-process power smoothing constraints are used to avoid safety risks caused by power mutations. By organically combining swarm intelligence mechanisms with traditional control methods, an adaptive dual-mode collaborative control system is formed, improving the uniform temperature control effect and robustness of the system under arbitrary layouts.

[0114] In summary, this invention proposes a temperature control method for explosion-proof electrical appliances that integrates the pheromone evaporation and deposition mechanism of ant colonies with the frequency adaptive mechanism of bat echolocation. By constructing a complete closed-loop system of initial swarm intelligence model calibration, adaptive incremental update for layout changes, and dual-mode collaborative decoupling control, it solves the common industry problem of wall thermal memory interference model identification and decoupling failure caused by ill-conditioned gain matrix in movable layouts. This invention achieves synchronous, interference-free, and accurate calibration of independent heat transfer contributions under strong coupling heat transfer conditions with multiple suction cups through Walsh orthogonal perturbation identification technology based on thermal steady state, shortening the initial modeling time of the system. By mapping the pheromone dynamic update principle of the ant colony algorithm to the process of thermal memory elimination and new model reconstruction, it actively cancels the residual temperature gradient of the wall using stepped anti-phase heat pulses, eliminating the need to passively wait for the wall to cool naturally, thus shortening the model reconstruction time after layout changes. By drawing on the frequency coding and target recognition mechanism of bat echolocation, it designs a control mode with natural frequency domain decoupling, bypassing the physical layer... This invention eliminates the risk of error amplification and power oscillation caused by pseudo-inverse decoupling when the gain matrix is ​​ill-conditioned. Through real-time monitoring of the gain matrix condition number and multi-cycle smooth power transition scheduling, it achieves adaptive and seamless switching between bat frequency decoupling mode and pseudo-inverse decoupling mode. It maintains the fast response characteristics of traditional methods when the gain matrix is ​​well-conditioned and ensures long-term robust stability of the system when ill-conditioned. Simultaneously, this invention embeds safety constraints such as power amplitude limiting, stepped power changes, and graded response to ambient gas concentrations into the entire control process to avoid safety hazards caused by power surges and electrical anomalies. This invention improves the uniformity of the three-dimensional temperature field and the system's adaptability to arbitrary layout changes, achieving precise temperature uniformity control of a modular, portable heating system in a closed environment.

[0115] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0116] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0117] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0118] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for temperature uniformity control of explosion-proof electrical appliances, characterized in that, include: Based on the thermal steady-state condition, the initial reference baseline of the swarm intelligence model is obtained through initial orthogonal perturbation identification; Based on the initial swarm intelligence model reference baseline and layout change trigger signal, the updated gain matrix and new layout reference baseline are obtained by combining the ant colony pheromone evaporation and deposition mechanism, the step-by-step anti-phase thermal pulse, and the updated orthogonal perturbation identification. Based on the updated gain matrix and the new layout reference baseline, and according to the bat echolocation swarm intelligence mechanism, the suction cup power reference value is obtained through dual-mode collaborative control including bat echolocation frequency decoupling mode and pseudo-inverse decoupling mode.

2. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 1, characterized in that, The initial reference baseline for the swarm intelligence model is obtained based on thermal steady-state conditions through initial orthogonal perturbation identification. This initial orthogonal perturbation identification includes Walsh orthogonal perturbation sequence synchronization excitation and orthogonal decoupling, comprising: Based on thermal steady-state conditions, a perturbation temperature response dataset is obtained by synchronous excitation using a Walsh orthogonal perturbation sequence. Based on the perturbation temperature response dataset, the initial pure gain matrix is ​​obtained through orthogonal decoupling calculation. The initial swarm intelligence model reference baseline is obtained through non-volatile storage based on the initial pure gain matrix.

3. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 1, characterized in that, The method, based on the initial swarm intelligence model reference baseline and layout change trigger signal, utilizes ant colony pheromone evaporation and deposition mechanisms, step-wise anti-phase thermal pulses, and updated orthogonal perturbation identification to obtain the updated gain matrix and new layout reference baseline, including: Based on the layout change trigger signal, the real-time power of the suction cup and the temperature distribution of the wall surface are obtained and a thermal status profile is constructed. The layout change trigger signal is determined according to the online status of the suction cup communication node or user interface instructions. Based on the initial swarm intelligence model reference baseline and thermal state archive, the old thermal memory elimination state is obtained through the ant colony pheromone evaporation and deposition mechanism, the step-by-step reverse phase thermal pulse, and the simultaneous execution of gas concentration safety monitoring. Based on the old thermal memory elimination state, a new layout reference temperature vector is obtained by power reset and establishing a new layout surrounding quasi-steady state. Based on the new layout reference temperature vector, the updated gain matrix and the new layout reference baseline are obtained by updating the orthogonal perturbation identification.

4. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 3, characterized in that, The process, based on the initial swarm intelligence model reference baseline and thermal state profile, utilizes ant colony pheromone evaporation and deposition mechanisms, a stepped anti-phase thermal pulse, and simultaneous gas concentration safety monitoring to obtain the old thermal memory elimination state, including: Based on the initial swarm intelligence model reference baseline and thermal state profile, an evaporation power execution sequence is obtained by applying a stepped anti-phase thermal pulse and calculating the amplitude of the anti-phase thermal pulse. Based on the evaporation power execution sequence, the safety monitoring status of the evaporation process is obtained by performing gas concentration safety monitoring. Based on the evaporation power execution sequence and the evaporation process safety monitoring status, the old thermal memory elimination status is obtained by comparing the deviation threshold.

5. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 4, characterized in that, The step of obtaining the updated gain matrix and the new layout reference baseline based on the new layout reference temperature vector through updated orthogonal perturbation identification includes: Based on the new layout reference temperature vector, the new layout perturbation temperature response dataset is obtained by synchronous excitation using the same Walsh orthogonal perturbation sequence as the initial orthogonal perturbation identification. Based on the new layout perturbation temperature response dataset, the full pure gain matrix of the new layout is obtained through orthogonal decoupling calculation. Based on the full pure gain matrix of the new layout, the updated gain matrix and the new layout reference baseline are obtained by replacing the matrix and storing it.

6. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 1, characterized in that, Based on the updated gain matrix and the new layout reference baseline, and according to the bat echolocation swarm intelligence mechanism, the suction cup power reference value is obtained through dual-mode cooperative control including a bat echolocation frequency decoupling mode and a pseudo-inverse decoupling mode, including: Based on the updated gain matrix and the new layout reference baseline, the target power reference value of the suction cup is obtained by quadratic programming optimization. Based on the updated gain matrix and the target power reference value of the suction cup, the control mode selection signal and the smooth initial power reference value are obtained through power smooth transition scheduling. Based on the new layout reference baseline, the jitter frequency range is determined, and the smooth initial power reference value is used as the bias reference. The independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying component of the temperature measurement point are obtained by sinusoidal jitter superposition and orthogonal demodulation of the sliding window. Based on the control mode selection signal, the suction cup power reference value is obtained through dual-mode collaborative control.

7. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 6, characterized in that, The step of obtaining the control mode selection signal and the smooth initial power reference value based on the updated gain matrix and the suction cup target power reference value through power smooth transition scheduling includes: Based on the updated gain matrix, the current condition number is obtained by calculating the ratio of the maximum singular value to the minimum singular value of the matrix; A control mode selection signal is obtained by comparing the current condition number with a preset state determination threshold. Based on the control mode selection signal and the target power reference value of the suction cup, a smooth initial power reference value is obtained by gradually switching within the transition window using a power smooth transition algorithm in a linear interpolation manner.

8. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 6, characterized in that, The process of determining the jitter frequency range based on the new layout reference baseline, using the smoothed initial power reference value as a bias reference, and obtaining the independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components of the temperature measurement point through sinusoidal jitter superposition and orthogonal demodulation of the sliding window includes: Based on the new layout reference baseline, the jitter frequency is determined by calculating the thermal cutoff frequency and taking a safety factor, and then the jitter frequency of the suction cup is allocated by a geometric series to obtain a frequency allocation scheme. Based on the frequency allocation scheme and the smoothed initial power reference value, the real-time output power of the suction cup containing the characteristic frequency identifier is obtained by superimposing a sinusoidal dithering signal. Based on the real-time output power of the suction cup, the independent heat transfer contribution amplitude matrix of the suction cup and the slowly varying components of the temperature measurement point are obtained by orthogonal demodulation through a sliding window.

9. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 6, characterized in that, The suction cup power reference value is obtained through dual-mode collaborative control based on the control mode selection signal. The control mode selection signal includes a first mode selection signal and a second mode selection signal, including: When the control mode selection signal is the first mode selection signal, the bat echo location frequency decoupling mode is selected, and the suction cup power reference value is obtained by frequency decoupling deviation distribution and incremental proportional-integral-derivative control. When the control mode selection signal is the second mode selection signal, the pseudo-inverse decoupling mode is selected, and the suction cup power reference value is obtained by superimposing discrete proportional-integral-derivative control on the pseudo-inverse decoupling feedforward.

10. The method for temperature uniformity control of an explosion-proof electrical appliance according to claim 9, characterized in that, The process of obtaining the suction cup power reference value through frequency decoupling deviation allocation and incremental proportional-integral-derivative control includes: Based on the gradually varying component of the temperature measurement point and the global temperature setpoint, the global temperature deviation is obtained by calculating the difference. Based on the global temperature deviation and the independent heat transfer contribution amplitude matrix of the suction cup, the comprehensive temperature deviation of the suction cup is obtained by weighting the contribution ratio. Based on the comprehensive temperature deviation of the suction cup, the reference value of the suction cup power in the bat echolocation frequency decoupling mode is calculated by incremental proportional-integral-derivative control.

Images