A multi-working-condition PID dynamic compensation method for a temperature control module

Abstract

The application discloses a temperature control module multi-working condition PID dynamic compensation method, and relates to the technical field of dynamic compensation, and comprises the following steps: based on the working condition switching event mark of a temperature control module, determining a first dialing action and a second dialing action; based on the first dialing action, generating a first direction interpretation result, and based on the second dialing action, generating a second direction interpretation result; based on the first direction interpretation result and the second direction interpretation result, performing state determination on the hot on-off path response of the temperature control module, and obtaining a structure state label; performing action combination generation on an action library corresponding to the structure state label, and obtaining an action combination graph; based on the action combination graph, constructing a candidate action combination path, and determining the path cost of the candidate action combination path; and the application improves the adaptability of control behavior under multiple working conditions.

Description

Technical Field

[0001] This invention relates to the field of dynamic compensation technology, and in particular to a multi-condition PID dynamic compensation method for a temperature control module. Background Technology

[0002] In microcircuit temperature control and protection applications, temperature control modules typically need to perform continuous adjustment and rapid switching across multiple operating states. These include processes such as chip startup preheating, steady-state temperature control, local overheat suppression, thermal protection recovery, thermal on / off path switching, and temperature recovery after sudden load changes. In these operating states, the controlled objects not only include conventional heating units but may also include microchannel thermal switches used to change the thermal on / off path. For temperature control modules employing thermomagnetic liquid metal microchannel thermal switches, the combination of liquid metal and magnetic particles can generate thermal on / off switching actions according to temperature changes, thus controlling the heat... The switching between heat conduction and heat insulation paths enables thermal management under different operating conditions. Gallium-based liquid metal surfaces have an oxide skin. During the reciprocating motion within the microchannel, the oxide skin and its derivatives may remain locally on the quartz or glass walls, gradually forming a tongue-shaped anchoring structure with one end attached to the wall and the other end able to adhere to or detach with the movement of the droplet. This causes changes in the contact state of the heat transfer interface, the distribution of thermal resistance, and the conductivity of the heat conduction path. As a result, the thermal response of the temperature control module under multiple operating conditions exhibits a more complex variation pattern than that of a conventional single heat capacity object.

[0003] In existing technologies, temperature control modules mostly employ conventional PID control to perform closed-loop adjustment of the error between the target temperature and the current temperature. This is usually based on the premise that the control output and the temperature response maintain a relatively stable correspondence. In the above-mentioned thermal switching scenario of thermomagnetic liquid metal microchannels, when a tongue-shaped anchoring structure is formed on the inner wall of the microchannel, the interface contact relationship of the thermal on / off path will change with the direction of the control output and the movement state of the droplets. This leads to inconsistent effects of the control output on temperature changes under the same temperature error conditions, which in turn causes the integral term in the PID control algorithm to accumulate deviations and the derivative term to produce response distortions. This results in unstable control results under the same operating conditions and inconsistent compensation effects when switching between different operating conditions. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies, such as unstable control results under similar operating conditions and inconsistent compensation effects when switching between different operating conditions, and to propose a multi-condition PID dynamic compensation method for temperature control modules.

[0005] To address the problems existing in the prior art, the present invention adopts the following technical solution: A multi-condition PID dynamic compensation method for a temperature control module includes: Based on the operating condition switching event flag of the temperature control module, the first toggle action and the second toggle action are determined; a first direction interpretation result is generated based on the first toggle action, and a second direction interpretation result is generated based on the second toggle action; Based on the interpretation results of the first direction and the interpretation results of the second direction, the state of the thermal on / off path response of the temperature control module is determined, and the structural state label is obtained. Action combination is generated from the action library corresponding to the structural state label to obtain the action combination graph; candidate action combination paths are constructed based on the action combination graph, and the path cost of the candidate action combination paths is determined. The pheromone parameters are updated based on the path cost to obtain the optimal action combination node, and the PID control algorithm is dynamically compensated based on the optimal action combination node.

[0006] Preferably, determining the first actuation action and the second actuation action includes: Obtain the operating condition switching event flag of the temperature control module; Based on the operating condition switching event flags and the PID control algorithm, obtain the control output of the temperature control module; The toggle increment is generated based on the control output; Based on the control output and the toggle increment, the first toggle action and the second toggle action are determined.

[0007] Preferably, generating a first directional interpretation result based on the first toggle action includes: During the first toggle action, the first temperature change is determined based on the first temperature sample value collected by the temperature sensor and the first reference temperature sample value at the start of the first toggle action. Determine the direction of the first temperature change based on the first temperature change amount; Based on the first toggle output and control output corresponding to the first toggle action, determine the change in the first control output; The direction of change of the first control output is determined based on the amount of change in the first control output. Based on the first temperature change direction and the first control output change direction, a first direction interpretation result is generated.

[0008] Preferably, generating a second directional interpretation result based on the second toggle action includes: During the second toggle action, determine the second temperature change. Based on the second toggle output and control output corresponding to the second toggle action, determine the change in the second control output; The second direction interpretation result is generated based on the second temperature change and the second control output change.

[0009] Preferably, the structural state label is obtained, including: The state in which the thermal on / off path response of the temperature control module is consistent is defined as the symmetrical path state. The state in which the thermal on / off response of the temperature control module is inconsistent is defined as the shell tongue direction bias state. When the interpretation results in the first direction are the same as those in the second direction, the symmetric path state is used as the structural state label. When the interpretation results in the first direction differ from those in the second direction, the shell tongue orientation bias state is used as the structural state label.

[0010] Preferably, the obtained action combination diagram includes: When the structural state label is in the shell tongue direction bias state: Based on the structural state label, the gating rules for integral writing gating actions in the PID control algorithm are determined; Based on structural state labels, the gating rules for derivative protection actions in the PID control algorithm are determined. Based on the control output and the toggle increment, the toggle sequence of the symmetrical reset toggle action is determined; The integral is written into the gating rules of the gating action, the gating rules of the differential protection action, and the toggle sequence of the symmetrical reset toggle action, and then associated and configured to obtain the action library; The action library is used to generate action combination graphs.

[0011] When the structure status label is in symmetrical path state, the integral write gating action, differential protection action, and symmetrical reset toggle action are not enabled.

[0012] Preferably, constructing candidate action combination paths based on the action combination graph includes: The ant colony algorithm is used to initialize the action combination graph to obtain pheromone parameters. Based on pheromone parameters, candidate action combination paths are constructed.

[0013] Preferably, determining the path cost of the candidate action combination path includes: Obtain the target temperature from the temperature control module; Obtain the current temperature of the temperature control module; The temperature error is obtained by performing a difference calculation between the target temperature and the current temperature. Obtain the change in control output; Based on the temperature error and the change in control output, the path cost of candidate action combination paths is determined.

[0014] Preferably, the optimal action combination node is obtained, including: Based on the path cost, the pheromone parameters are updated to obtain the updated pheromone parameters; When the iteration termination condition is met, the optimal action combination node is determined based on the updated pheromone parameters.

[0015] Preferably, dynamic compensation control is applied to the PID control algorithm, including: When the optimal action combination node indicates that the integral write gating action is enabled, the integral write process of the integral term in the PID control algorithm is dynamically compensated and controlled according to the gating rules of the integral write gating action. When the optimal action combination node indicates that the differential protection action is enabled, dynamic compensation control is performed on the differential contribution process of the differential term in the PID control algorithm according to the gating rules of the differential protection action. When the optimal action combination node indicates that the symmetrical reset toggle action is enabled, dynamic compensation control is performed on the control output process of the temperature control module.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. In this invention, by introducing a first toggle action and a second toggle action and generating directional interpretation results, the difference in response of the heat on / off path under opposite adjustment directions is realized. This enables accurate determination of whether the current structural state belongs to the symmetrical path state or the shell tongue direction offset state. The temperature control process no longer relies on a single error feedback, but is judged in combination with the actual heat transfer path state. This effectively solves the problem of unstable relationship between control output and temperature response, and improves the adaptability and discrimination accuracy of control behavior under multiple operating conditions.

[0017] 2. In this invention, by constructing an action combination graph and iteratively updating the pheromone parameters using path cost, adaptive filtering of multiple control strategy combinations is achieved. This enables the integral writing gating action, differential protection action, and symmetrical reset toggle action to be dynamically selected based on the actual path effect, thereby avoiding the problem of error accumulation or response distortion caused by fixed control strategies under complex interface conditions, and improving the matching and stability of control strategies under different operating conditions.

[0018] 3. In this invention, by dynamically compensating the integral term, derivative term, and control output process under the guidance of the optimal action combination node, the formation process of the control quantity can be modified separately to address the directional dependence caused by changes in interface contact, thereby restoring the correspondence between the control output and the temperature response, reducing the accumulation of control deviation and suppressing abnormal fluctuations, and improving the consistency and reliability of the temperature regulation process under multiple operating conditions. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating a multi-condition PID dynamic compensation method for a temperature control module according to an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] This embodiment provides a multi-condition PID dynamic compensation method for a temperature control module. (See also...) Figure 1 Specifically, this includes: S1, using the operating condition switching event flag of the temperature control module to determine the first toggle action and the second toggle action; In embodiments of the present invention, determining the first toggle action and the second toggle action includes: Obtain the operating condition switching event flag of the temperature control module; Operating condition switching event markers refer to the status identification information formed when the temperature control module changes from one operating state to another, which is used to characterize that the working environment, load state, or thermal on / off state of the controlled thermal system has changed.

[0022] Specifically, the system continuously collects target temperature data, temperature sensor output data, actuator current output data, and thermal on / off execution unit conduction status data during the current operation of the temperature control module. The target temperature data and temperature sensor output data are synchronized and aligned to form a current temperature deviation sequence. The actuator current output data is arranged according to the control cycle sequence to form an output change sequence. The thermal on / off execution unit conduction status data is recorded in chronological order to form an on / off status sequence. The current temperature deviation sequence, output change sequence, and on / off status sequence are then compared within the same control cycle. When at least one of the following changes occurs within any control cycle: target temperature change, actuator output adjustment direction change, or thermal on / off execution unit conduction status change, the corresponding operating state of that control cycle is recorded as a new operating condition. The change in this new operating condition relative to the operating condition of the previous control cycle is generated as a condition switching event marker.

[0023] Based on the operating condition switching event flags and the PID control algorithm, obtain the control output of the temperature control module; PID control algorithm refers to a control operation method that adjusts the input of the actuator by proportional, integral and derivative based on the deviation between the target temperature and the current temperature. It corresponds to the control mechanism inside the temperature control module used to adjust the heat input or the degree of heat blockage. The control output refers to the actual control quantity given to the actuator by the temperature control module after completing the control operation. This control quantity corresponds to the heating power, drive current, duty cycle, switch conduction degree or other physical quantities that can change the state of the heat path.

[0024] The system reads the target temperature data and current temperature data corresponding to the control cycle marked by the operating condition switching event. It calculates the difference between the target temperature data and the current temperature data to obtain the current temperature error. It then calculates the proportional adjustment amount within the current control cycle based on the current temperature error. Finally, it accumulates the temperature errors of the current control cycle and all previous control cycles in chronological order to obtain the integral adjustment amount. The system calculates the difference between the temperature error of the current control cycle and the temperature error of the previous control cycle to obtain the derivative adjustment amount. Finally, it sums the proportional, integral, and derivative adjustment amounts to obtain the PID control algorithm output value. This PID control algorithm output value is then verified against the operating condition corresponding to the operating condition marked by the operating condition switching event. If both conditions belong to the same control cycle, the PID control algorithm output value is determined as the control output of the temperature control module and sent to the corresponding actuator.

[0025] The toggle increment is generated based on the control output; The incremental adjustment refers to a small change added to the current control output. It corresponds to a short-term increase or decrease in the original thermal drive level and is used to cause a change in the response of the thermal on / off path.

[0026] Read the control output value corresponding to the current control cycle, extract the absolute value of the control output value as the reference output amplitude, calculate the change amplitude of the control output value in multiple consecutive control cycles before the current control cycle and form an output change sequence, calculate the absolute value of the control output difference between each adjacent control cycle in the output change sequence and obtain the average change amplitude, compare the reference output amplitude with the average change amplitude, select the smaller value as the toggle generation base value, and then match the toggle generation base value with the positive and negative directions of the control output to form a positive toggle increment consistent with the current control output direction and a reverse toggle increment opposite to the current control output direction. The positive toggle increment and the reverse toggle increment are equal in magnitude and opposite in direction. The positive toggle increment and the reverse toggle increment are output as the toggle increment corresponding to the current control cycle.

[0027] Based on the control output and the toggle increment, the first toggle action and the second toggle action are determined.

[0028] The first toggle action refers to the first control adjustment process formed by applying a toggle increment to the control output in one direction, which corresponds to a directional disturbance to the thermal path; the second toggle action refers to the second control adjustment process formed by applying a toggle increment to the control output in the opposite direction to the first toggle action, which corresponds to a reverse disturbance to the thermal path, so that the thermal response difference of the temperature control module in two opposite directions can be identified and utilized.

[0029] The numerical value and sign direction of the control output are obtained. The forward toggle increment is used as the increment value consistent with the control output direction, and the reverse toggle increment is used as the increment value opposite to the control output direction. A numerical addition operation is performed on the control output and the forward toggle increment to obtain the first toggle output. The addition operation is an algebraic summation of two values ​​with the same dimension while preserving the direction attribute of the result. A numerical addition operation is performed on the control output and the reverse toggle increment to obtain the second toggle output. The reverse toggle increment participates in the operation in the form of a signed numerical value to achieve a superposition effect with opposite directions. The first toggle output is mapped to the execution control quantity of the first toggle action, and the second toggle output is mapped to the execution control quantity of the second toggle action. The first toggle action and the second toggle action are written into the execution sequence according to the control cycle order. At the same time, the first toggle output, the second toggle output, and the corresponding control output and toggle increment are stored in correspondence for subsequent temperature change calculation and direction interpretation result generation.

[0030] S2. Generate a first direction interpretation result based on the first toggle action, and generate a second direction interpretation result based on the second toggle action; In an embodiment of the present invention, generating a first directional interpretation result based on a first tossing action includes: During the first toggle action, the first temperature change is determined based on the first temperature sample value collected by the temperature sensor and the first reference temperature sample value at the start of the first toggle action. The first toggle action period refers to the time period during which the actuator controls and adjusts according to the first toggle output. During this time period, the state of the thermal on / off path changes under control. The first temperature sampling value refers to the temperature data collected in real time by the temperature sensor during this time period, reflecting the result of heat transfer in the path. The first reference temperature sampling value refers to the temperature sensor data collected at the start of the first toggle action, which is used as a reference starting point for temperature change. The first temperature change amount refers to the difference between the first temperature sampling value and the first reference temperature sampling value, which is used to characterize the temperature response amplitude of the thermal path under the toggle action.

[0031] At the start of the first toggle action, the temperature data currently output by the temperature sensor is read and recorded as the first reference temperature sampling value. During the continuous execution of the first toggle action, the temperature data output by the temperature sensor is read according to the sampling time corresponding to the control cycle and recorded as the first temperature sampling value. The first temperature sampling value and the first reference temperature sampling value are numerically subtracted to obtain the first temperature change.

[0032] Determine the direction of the first temperature change based on the first temperature change amount; The first temperature change direction refers to the positive or negative trend of the first temperature change, which is used to characterize the trend of increasing or decreasing heat in the pathway.

[0033] Read the numerical result of the first temperature change, and determine whether the first temperature change is greater than zero, equal to zero, or less than zero. If the first temperature change is greater than zero, determine the direction of the first temperature change as the heating direction; if the first temperature change is less than zero, determine the direction of the first temperature change as the cooling direction; and if the first temperature change is equal to zero, determine the direction of the first temperature change as the no-change direction.

[0034] Based on the first toggle output and control output corresponding to the first toggle action, determine the change in the first control output; The first toggle output refers to the actuator control quantity formed by superimposing the toggle increment on the control output, which is used to drive the thermal on / off path to undergo directional changes; the control output refers to the control quantity output to the actuator by the temperature control module after control calculation, which corresponds to the actual driving force applied to the thermal on / off path.

[0035] Read the first toggle output value corresponding to the first toggle action, read the control output value corresponding to the control cycle of the first toggle action, perform a numerical subtraction operation between the first toggle output value and the control output value, and determine the result of the operation of subtracting the control output value from the first toggle output value as the first control output change.

[0036] The direction of change of the first control output is determined based on the amount of change in the first control output. The direction of change of the first control output refers to the positive or negative trend of the change in the first control output, which is used to characterize the direction of increase or decrease of the control quantity.

[0037] Read the numerical result of the change in the first control output, determine whether the change in the first control output is greater than zero, equal to zero, or less than zero. If the change in the first control output is greater than zero, determine the direction of change of the first control output as the increasing direction. If the change in the first control output is less than zero, determine the direction of change of the first control output as the decreasing direction. If the change in the first control output is equal to zero, determine the direction of change of the first control output as the direction of no change. Store the corresponding change direction of the first control output with the change in the first control output.

[0038] Based on the first temperature change direction and the first control output change direction, a first direction interpretation result is generated.

[0039] The first direction interpretation result refers to the result obtained based on the correspondence between the first temperature change direction and the first control output change direction, which is used to reflect the consistency or inconsistency between the control input and the temperature response.

[0040] The system reads the first temperature change direction and the first control output change direction corresponding to the same first toggle action. It then compares the directional correspondence between the first temperature change direction and the first control output change direction. If the first temperature change direction is a heating direction and the first control output change direction is an increasing direction, they are determined to be in the same direction. If the first temperature change direction is a cooling direction and the first control output change direction is a decreasing direction, they are determined to be in the same direction. If the first temperature change direction is a heating direction and the first control output change direction is a decreasing direction, they are determined to be in opposite directions. If the first temperature change direction is a cooling direction and the first control output change direction is an increasing direction, they are determined to be in opposite directions. If the first temperature change direction is not changing or the first control output change direction is not changing, they are determined to be in zero direction. The same-direction correspondence, opposite-direction correspondence, or zero-direction correspondence is then written into the result recording unit corresponding to the first toggle action, and the written correspondence result is determined as the first direction interpretation result.

[0041] In an embodiment of the present invention, generating a second direction interpretation result based on a second tossing action includes: During the second toggle action, determine the second temperature change. The second temperature change refers to the difference between the temperature data collected by the temperature sensor during the second toggle action and the temperature data corresponding to the start of the second toggle action, which is used to characterize the magnitude of heat change under the reverse toggle action.

[0042] At the start of the second toggle action, the temperature data currently output by the temperature sensor is read and recorded as the second reference temperature sampling value. During the continuous execution of the second toggle action, the temperature data output by the temperature sensor is continuously read according to the sampling time corresponding to the control cycle and recorded as the second temperature sampling value. The second temperature sampling value is matched one-to-one with the second reference temperature sampling value. The second temperature sampling value at each moment is subtracted from the second reference temperature sampling value to obtain the temperature difference value at the corresponding moment. Then, the temperature difference value corresponding to the end of the second toggle action is extracted from all the temperature difference values ​​as the second temperature change.

[0043] Based on the second toggle output and control output corresponding to the second toggle action, determine the change in the second control output; The second toggle output refers to the actuator control quantity formed by superimposing the reverse toggle increment on the control output, which is used to drive the thermal on / off path to produce adjustment in the opposite direction to the first toggle action; the second control output change refers to the difference between the second toggle output and the control output, which is used to characterize the degree of change of the actuator control quantity during the reverse toggle process.

[0044] Read the second toggle output value corresponding to the second toggle action, read the control output value corresponding to the control cycle of the second toggle action, perform a numerical subtraction operation on the same dimension between the second toggle output value and the control output value, and determine the result of subtracting the control output value from the second toggle output value as the second control output change.

[0045] The second direction interpretation result is generated based on the second temperature change and the second control output change.

[0046] The second direction interpretation result refers to the result obtained based on the correspondence between the second temperature change and the second control output change, which is used to reflect the matching between the control input and the temperature response under reverse toggle conditions.

[0047] Read the second temperature change and the second control output change corresponding to the same second toggle action, perform sign judgment on the numerical result of the second temperature change, determine the second temperature change to correspond to a temperature rise when the second temperature change is greater than zero, determine the second temperature change to correspond to a temperature drop when the second temperature change is less than zero, and determine the second temperature change to correspond to no temperature change when the second temperature change is equal to zero. The sign of the numerical result of the change in the second control output is determined. When the change in the second control output is greater than zero, the control output is determined to increase. When the change in the second control output is less than zero, the control output is determined to decrease. When the change in the second control output is equal to zero, the control output is determined to remain unchanged. The corresponding states of the second temperature change and the second control output change are compared. When the temperature increase and the control output increase occur simultaneously, it is determined to be a same-direction correspondence. When the temperature decrease and the control output decrease occur simultaneously, it is determined to be a same-direction correspondence. When the temperature increase and the control output decrease occur simultaneously, it is determined to be a opposite-direction correspondence. When the temperature decrease and the control output increase occur simultaneously, it is determined to be a opposite-direction correspondence. When there is no temperature change or no change in control output, it is determined to be a zero-direction correspondence. The same-direction correspondence, opposite-direction correspondence, or zero-direction correspondence is then written into the result recording unit corresponding to the second toggle action, and the correspondence result after writing is determined as the second direction interpretation result.

[0048] S3. Based on the interpretation results of the first direction and the interpretation results of the second direction, the state of the thermal on / off path response of the temperature control module is determined to obtain the structural state label. In an embodiment of the present invention, obtaining a structural state label includes: The state in which the thermal on / off path response of the temperature control module is consistent is defined as the symmetrical path state. The state in which the thermal on / off response of the temperature control module is inconsistent is defined as the shell tongue direction bias state. The thermal on / off path response refers to the temperature change behavior exhibited by the temperature control module in order to open or close the heat transfer path under the control output; the symmetrical path state refers to the state in which the response of the thermal on / off path to temperature change remains consistent under both forward and reverse toggling actions.

[0049] The shell-tongue directional bias state refers to the state in which the heat transfer interface of the internal heat on / off path of the temperature control module has an asymmetrical contact relationship due to the locally attached shell structure and the tongue structure formed by the extension of the free end when subjected to control output. This results in differences in the heat transfer impedance and conductivity in different directions. In this state, heat exhibits directional dependence when passing through the path. That is, when it is transferred in one direction, the interface contact is tight and the thermal resistance is low, while when it is transferred in the opposite direction, the interface contact is limited and the thermal resistance is high. This leads to different temperature response results depending on the direction of the control input.

[0050] The state in which the response of the thermal on / off path is consistent is defined as the symmetrical path state. This is because, in this state, the heat transfer in the path is not affected by direction, and the relationship between the control output and the temperature change remains consistent under both positive and negative conditions. This indicates that the thermal on / off interface is stable and the heat transfer path is continuous, so this state can be regarded as the thermal path being in a balanced conduction state. The state in which the response of the thermal on / off path is inconsistent is defined as the shell tongue direction bias state. This is because, in this state, the heat transfer process is affected by the local attachment structure at the interface, which leads to differences in the contact conditions and heat transfer capacity of the path in different directions. This causes the correspondence between the control output and the temperature change to change with the direction, thus reflecting that the thermal on / off path has a direction-dependent non-equilibrium conduction state.

[0051] When the interpretation results in the first direction are the same as those in the second direction, the symmetric path state is used as the structural state label. When the interpretation results in the first direction are the same as those in the second direction, the symmetrical path state is used as the structural state label because the correspondence between temperature change and control output remains consistent under the control action in two opposite directions. This indicates that the thermal on / off path has the same heat transfer behavior under different action directions, the interface contact state is stable and the thermal resistance distribution is uniform, and the relationship between control input and temperature response does not change with the direction. Therefore, this state can be classified as the symmetrical path state to characterize that the thermal on / off path is in a balanced conduction state.

[0052] When the interpretation results in the first direction differ from those in the second direction, the shell tongue orientation bias state is used as the structural state label.

[0053] The structural status label refers to the identification result obtained by classifying the current thermal on / off path status based on the relationship between the first direction interpretation result and the second direction interpretation result. It is used to reflect whether the temperature control module is currently in a symmetrical path state or a shell tongue direction offset state.

[0054] When the interpretation results in the first direction differ from those in the second direction, the shell tongue directional bias state is used as the structural state label because the correspondence between temperature change and control output differs under control action in opposite directions. This indicates that there are inconsistent contact conditions and heat transfer capabilities in different directions of the heat on / off path. The local attached structure has a directional dependence on heat transfer, causing the relationship between control input and temperature response to shift. This reflects that the heat on / off path is in a non-equilibrium conduction state and has directional bias.

[0055] S4. Generate action combinations from the action library corresponding to the structural state labels to obtain the action combination diagram. In an embodiment of the present invention, an action combination diagram is obtained, including: When the structural state label is in the shell tongue direction bias state: Based on the structural state label, the gating rules for integral writing gating actions in the PID control algorithm are determined; Integral write gating refers to the adjustment method that restricts or allows the accumulation process of the integral term in the PID control algorithm, which corresponds to whether the accumulation of the control quantity over time participates in the formation of the control output.

[0056] Read the structural state label corresponding to the current control cycle. When the structural state label represents the shell tongue direction offset state, extract the temperature error data of the current control cycle, the temperature error data of the previous consecutive control cycles, and the cumulative result of the integral term of the current control cycle. Determine whether the direction of temperature error change in the current control cycle is consistent with the direction of thermal on / off path response. If they are consistent, the temperature error of the current control cycle is allowed to participate in the integral accumulation. If they are inconsistent, the temperature error of the current control cycle is prohibited from being written into the integral term, and the cumulative result of the integral term of the previous control cycle is kept as the output of the integral term of the current control cycle. When the structural state label represents the symmetrical path state, the temperature error of the current control cycle is allowed to be continuously written into the integral term in the order of the control cycle, thereby obtaining the gating rule of the integral write gating action corresponding to the current structural state.

[0057] Based on structural state labels, the gating rules for derivative protection actions in the PID control algorithm are determined. The differential protection action refers to the adjustment method that suppresses or releases the effect of the rate of change of the differential term in the PID control algorithm, which corresponds to the degree of influence of the rate of change of the control quantity on the actuator drive.

[0058] The structural state label corresponding to the current control cycle is read, and the temperature error data of the current control cycle and the temperature error data of the previous control cycle are extracted. The temperature error change is obtained by subtracting the temperature error data of the previous control cycle from the temperature error data of the current control cycle. The change direction of the differential term in the current control cycle is then determined based on the temperature error change. When the structural state label represents the shell tongue direction offset state, it is determined whether the change direction of the differential term is consistent with the response direction of the thermal on / off path. If they are consistent, the differential term is kept to participate in the current control output calculation. If they are inconsistent, the contribution of the differential term to the control output in the current control cycle is canceled. When the structural state label represents the symmetrical path state, the differential term is kept to participate in the control output calculation according to the temperature error change of the current control cycle, thereby obtaining the gating rules of the differential protection action corresponding to the current structural state.

[0059] Based on the control output and the toggle increment, the toggle sequence of the symmetrical reset toggle action is determined; Symmetrical reset toggle action refers to the toggle adjustment process in which equal amplitude and opposite direction are applied in both positive and negative directions based on the control output. It corresponds to the symmetrical disturbance adjustment of the hot on / off path. Toggle sequence refers to the toggle control quantity change process arranged in time sequence. It corresponds to the adjustment trajectory of the actuator in multiple consecutive control cycles.

[0060] Read the control output value corresponding to the current control cycle, read the toggle increment value corresponding to the control output, perform numerical addition on the control output value and the toggle increment value under the same dimension to obtain the forward toggle output, perform numerical addition on the control output value and the reverse value of the toggle increment under the same dimension to obtain the reverse toggle output, arrange the control output value, forward toggle output, control output value, reverse toggle output and control output value in chronological order to form a toggle sequence of reference hold, forward toggle, reference recovery, reverse toggle and reference recovery.

[0061] The integral is written into the gating rules of the gating action, the gating rules of the differential protection action, and the toggle sequence of the symmetrical reset toggle action, and then associated and configured to obtain the action library; The system reads the integral enable or integral disable states corresponding to the gating rules of the integral writing gating action, reads the differential participation or differential cancellation states corresponding to the gating rules of the differential protection action, and reads the output values ​​and execution order of each sequence corresponding to the toggle sequence of the symmetrical reset toggle action. The integral enable or integral disable states, differential participation or differential cancellation states, and the output values ​​and execution order of each sequence are mapped one-to-one according to the same control cycle and written into the same action recording unit. This ensures that each action recording unit simultaneously contains the integral term processing method, the differential term processing method, and the toggle execution method, thereby forming an action library that can be directly used for the generation of subsequent action combinations.

[0062] The action library is used to generate action combination graphs.

[0063] An action library refers to a set of control strategies formed by combining different control and adjustment methods. It corresponds to a set of control actions that can be used to adjust the state of the hot-on-off circuit. An action combination diagram refers to a structural representation formed by organizing the various control strategies in the action library according to possible combination relationships. It corresponds to the switching relationship and combination path between different control strategies and is used to describe the selection space of control and adjustment methods.

[0064] Read all action recording units in the action library, extract the integral term processing method, differential term processing method, and toggle execution method contained in each action recording unit, and arrange and combine action recording units with different integral term processing methods, different differential term processing methods, and different toggle execution methods to form multiple action combination results. Assign a unique combination identifier to each action combination result and determine each action combination result as an action combination node. Then, compare any two action combination nodes one by one to determine whether there is a relationship between them where only one processing method or one execution method changes. If such a relationship exists, establish a switching connection relationship between the two action combination nodes; otherwise, do not establish a switching connection relationship. Then, summarize and record all action combination nodes and their corresponding switching connection relationships, using the action combination nodes as node content in the graph and the switching connection relationships as connection content in the graph. Generate an action combination graph that can represent the switching path between each action combination according to the correspondence between nodes and connection relationships.

[0065] When the structure status label is in symmetrical path state, the integral write gating action, differential protection action, and symmetrical reset toggle action are not enabled.

[0066] Read the structural state label corresponding to the current control cycle to confirm that the current thermal on / off path response is in a consistent state. Read the temperature error data of the current control cycle, the temperature error data of the previous control cycle, and the integral result accumulated from the previous control cycle. Directly write the temperature error data of the current control cycle into the integral accumulation process to form the integral term of the current control cycle. Perform a difference operation on the temperature error data of the current control cycle and the temperature error data of the previous control cycle to form the derivative term of the current control cycle. This allows the integral term and the derivative term to participate in the calculation of the current control output without gating restrictions. At the same time, read the control output data corresponding to the current control cycle and send it directly to the actuator in the original control cycle order without inserting forward and reverse toggle outputs. This ensures that when the structural state label is in the symmetrical path state, the integral term of the PID control algorithm is continuously accumulated, the derivative term participates continuously, and the control output is continuously executed.

[0067] It should be noted that when the structural status label is in symmetrical path state, the integral write gating action, the derivative protection action, and the symmetrical reset toggle action are not enabled. This is because at this time, the thermal on / off path exhibits a consistent heat transfer response under both forward and reverse action. The interaction between the control output and temperature change is stable and does not shift with the change of direction. The continuous accumulation of the integral term can accurately reflect the long-term temperature deviation and drive the control quantity to converge smoothly. The response of the derivative term to the rate of temperature change can truly reflect the current dynamic change trend. At the same time, the control output can maintain the balanced state of heat transfer by executing the original continuous adjustment method. Therefore, there is no need to introduce additional gating restrictions or disturbance adjustments, thereby avoiding unnecessary interference to the original stable control process.

[0068] S5. Construct candidate action combination paths based on the action combination graph and determine the path cost of the candidate action combination paths. In an embodiment of the present invention, constructing candidate action combination paths based on an action combination graph includes: The ant colony algorithm is used to initialize the action combination graph to obtain pheromone parameters. Ant colony algorithm refers to a search method that simulates the continuous accumulation and updating of information by individuals in a group during path selection to form the optimal selection result. It corresponds to the decision-making process of iteratively filtering among multiple control strategies. Initialization process refers to the process of assigning initial weights to each connection path in the relational structure. It corresponds to the state in which each path has the same selection basis before filtering. Pheromones parameter refers to the quantity used to represent the degree of preference of each path. It corresponds to the preference intensity formed by different control strategies in the historical selection process.

[0069] Read all action combination nodes and all connections between nodes in the action combination graph. Number each node and its corresponding connection according to the recording order in the graph. Create an independent information record unit for each connection. Write the start node number, end node number, and current reachable state of the corresponding connection into each information record unit. Arrange all information record units sequentially to form an information record table. Then, assign the same initial information value to each information record unit in the information record table. Summarize all information record units after writing the initial information value to form an action combination. Figure 1 A set of corresponding pheromone parameters.

[0070] Based on pheromone parameters, candidate action combination paths are constructed.

[0071] A candidate action combination path refers to a continuous connection path from the starting node to the target node in the action combination diagram. It corresponds to a set of control and adjustment methods executed in sequence, which are used to implement a specific adjustment scheme for the temperature control process.

[0072] Read all information recording units in the pheromone parameter set, determine the starting action combination node in the action combination graph, read all information recording units corresponding to all subsequent nodes directly connected to the starting action combination node, compare the amount of information in each information recording unit, select subsequent nodes in the order of node number if the amount of information is the same, select the subsequent node with larger information amount as the next action combination node of the current path if the amount of information is different, take the selected next action combination node as the new current node and continue to read all information recording units corresponding to all subsequent nodes directly connected to the node, and continue to select nodes in the same way until the end node with no subsequent connection relationship is reached or the node marked as the termination position in the action combination graph is reached. Combine and record all action combination nodes and connection relationships passed from the starting action combination node to the termination node in chronological order, thereby forming a candidate action combination path.

[0073] In embodiments of the present invention, determining the path cost of a candidate action combination path includes: Obtain the target temperature from the temperature control module; Obtain the current temperature of the temperature control module; The temperature error is obtained by performing a difference calculation between the target temperature and the current temperature. The target temperature refers to the temperature value that the temperature control module expects to achieve under the current operating conditions. It corresponds to the set thermal state of the controlled object. The current temperature refers to the actual temperature value collected by the temperature sensor within the current control cycle. It corresponds to the real-time thermal state of the controlled object. The temperature error refers to the difference between the target temperature and the current temperature. It is used to characterize the degree of deviation between the actual thermal state and the set thermal state.

[0074] Obtaining the target temperature of the temperature control module is to clarify the temperature reference required for the current control process. Obtaining the current temperature of the temperature control module is to determine the actual thermal state of the controlled object at the current moment. After both the target temperature and the current temperature are read, the temperature error is obtained by performing difference processing on the two. The target temperature is used as the set reference in the difference calculation, and the current temperature is used as the actual measured value in the difference calculation. The resulting temperature error is used to characterize the degree of deviation of the actual temperature from the set temperature. This forms a continuous processing logic from target determination and state acquisition to deviation quantification, and provides a direct basis for subsequent control output adjustment.

[0075] Obtain the change in control output; The change in control output refers to the difference between control output values ​​within adjacent control cycles, which is used to characterize the magnitude and trend of change of the control quantity over time.

[0076] Read the control output value corresponding to the current control cycle, read the control output value corresponding to the previous control cycle, perform a numerical subtraction operation on the control output value of the current control cycle and the control output value of the previous control cycle under the same dimension, determine the change in control output by subtracting the control output value of the previous control cycle from the control output value of the current control cycle, and associate and record the change in control output with the corresponding current control cycle, previous control cycle, and control output value.

[0077] Based on the temperature error and the change in control output, the path cost of candidate action combination paths is determined.

[0078] The path cost of the candidate action combination path refers to the evaluation result obtained by combining temperature error and control output change. It is used to reflect the effectiveness and control stability of a certain control and regulation scheme in achieving the temperature approximation target.

[0079] Read the temperature error value corresponding to the current control cycle, and read the control output change value corresponding to the same control cycle. Convert the temperature error value and the control output change value into two independent evaluation values ​​in the same evaluation record. Establish a path record unit for each candidate action combination path. Write the temperature error value and control output change value in the control process corresponding to each candidate action combination path into the corresponding path record unit. Perform numerical accumulation on the temperature error value and control output change value in the path record unit, and determine the accumulation result as the path cost of the candidate action combination path.

[0080] S6. Update the pheromone parameters according to the path cost to obtain the optimal action combination node, and perform dynamic compensation control on the PID control algorithm based on the optimal action combination node.

[0081] In an embodiment of the present invention, obtaining the optimal action combination node includes: Based on the path cost, the pheromone parameters are updated to obtain the updated pheromone parameters; The updated pheromone parameter refers to the result of adjusting the original pheromone parameter after considering the path cost, which corresponds to the degree of path selection preference redistributed according to the path quality.

[0082] Read the path cost value corresponding to each of the candidate action combination paths in the current iteration round, read the current pheromone parameter corresponding to each of the connection relationships in the action combination graph, and perform a uniform decay process on the current pheromone parameter of each connection relationship. The decay process is to multiply the current pheromone parameter of each connection relationship by the same decay coefficient to obtain the decayed pheromone parameter corresponding to each connection relationship. Next, the path cost of all candidate action combination paths is converted into its reciprocal value. Specifically, for each candidate action combination path, the value is divided by the path cost of that candidate action combination path to obtain the reciprocal value of that candidate action combination path. Then, the reciprocal values ​​of all candidate action combination paths are summed to obtain the total reciprocal value. Finally, the reciprocal value of each candidate action combination path is divided by the total reciprocal value to obtain the incremental allocation ratio of each candidate action combination path. Since the smaller the path cost value, the larger the corresponding reciprocal value, and therefore the larger the corresponding incremental allocation ratio, and since the larger the path cost value, the smaller the corresponding reciprocal value, and therefore the smaller the corresponding incremental allocation ratio, the incremental allocation ratio of all candidate action combination paths is then multiplied by the total incremental base value of the same round to obtain the pheromone increment corresponding to each candidate action combination path, where the total incremental base value is the average of the path cost values ​​of all candidate action combination paths. Next, all connections traversed by each candidate action combination path are identified, and the pheromone increment corresponding to the candidate action combination path is superimposed one by one onto the decayed pheromone parameter of each connection traversed by the candidate action combination path. If the same connection is traversed by multiple candidate action combination paths, the pheromone increments corresponding to the multiple candidate action combination paths are accumulated at the connection. Finally, the decayed pheromone parameter of each connection is merged with the superimposed pheromone increment to obtain the updated pheromone parameter corresponding to the connection. The updated pheromone parameters corresponding to all connections are then written into their respective information recording units.

[0083] When the iteration termination condition is met, the optimal action combination node is determined based on the updated pheromone parameters.

[0084] The iteration termination condition refers to the criterion used to determine whether to end the current multi-round path search process. It corresponds to the state where the path selection process reaches a stable state or reaches the specified search range. The optimal action combination node refers to the node with the largest pheromone parameter and the best path cost performance among all candidate action combination paths. It corresponds to the control strategy combination that can achieve a good adjustment effect in the current temperature control process.

[0085] After each round of pheromone parameter update, the updated pheromone parameters corresponding to all connections in the action combination graph are read. The connections are then categorized and organized according to the starting and ending action combination nodes to which they belong, forming a set of connection information for each action combination node. The updated pheromone parameters in the set of connection information for each action combination node are then summed to obtain the total pheromone amount for each action combination node. At the same time, the candidate action combination path records corresponding to each action combination node are read, and the number of times each action combination node appears in all candidate action combination paths and the sum of the associated path costs are calculated. When the iteration termination condition is met, the total amount of node pheromone for each action combination node is sorted by size. If the total amount of node pheromone is the same, the cumulative result of the path cost associated with the corresponding action combination node is compared. The action combination node with the smaller cumulative path cost is given priority to be the candidate optimal node. If the total amount of node pheromone is different, the action combination node with the largest total amount of node pheromone is determined as the candidate optimal node. Then, all the connection relationships and candidate action combination paths corresponding to the candidate optimal node are read. After confirming that the candidate optimal node corresponds to the maximum aggregation region of the updated pheromone parameters, the candidate optimal node is determined as the optimal action combination node. The optimal action combination node is associated and recorded with the corresponding total amount of node pheromone, candidate action combination path, and path cost.

[0086] In embodiments of the present invention, dynamic compensation control of the PID control algorithm includes: When the optimal action combination node indicates that the integral write gating action is enabled, the integral write process of the integral term in the PID control algorithm is dynamically compensated and controlled according to the gating rules of the integral write gating action. Read the optimal action combination node corresponding to the current control cycle, confirm that the integral write gating action in the optimal action combination node is enabled, read the gating rule of the integral write gating action corresponding to the current control cycle, read the temperature error data of the current control cycle, read the integral term value that has been accumulated at the end of the previous control cycle, read the structural state label corresponding to the current control cycle, when the structural state label represents the shell tongue direction offset state, determine the correspondence between the temperature error data of the current control cycle and the current thermal on / off path response state, and if the gating rule allows the temperature error to be written into the integral term of the current control cycle, add the temperature error data of the current control cycle to the integral term value that has been accumulated at the end of the previous control cycle according to the control cycle order to obtain the updated integral term value of the current control cycle; When the gating rules prohibit the writing of the temperature error of the current control cycle into the integral term, the value of the integral term accumulated at the end of the previous control cycle remains unchanged, and the unupdated integral term value is used as the integral term output of the current control cycle. When the structure state label represents the symmetrical path state, the temperature error data of the current control cycle is directly written into the integral accumulation process to obtain the updated integral term value of the current control cycle. Then, the integral term value obtained in the current control cycle replaces the original integral term of the PID control algorithm and participates in the calculation of the current control output, thereby completing the dynamic compensation control of the integral term writing process in the PID control algorithm.

[0087] When the optimal action combination node indicates that the differential protection action is enabled, dynamic compensation control is performed on the differential contribution process of the differential term in the PID control algorithm according to the gating rules of the differential protection action. Read the optimal action combination node corresponding to the current control cycle, confirm that the differential protection action in the optimal action combination node is enabled, read the gating rule of the differential protection action corresponding to the current control cycle, read the temperature error data of the current control cycle, read the temperature error data of the previous control cycle, subtract the temperature error data of the previous control cycle from the temperature error data of the current control cycle to obtain the temperature error change of the current control cycle, calculate the differential contribution value corresponding to the original differential term of the PID control algorithm based on the temperature error change of the current control cycle, and read the structural state label corresponding to the current control cycle. When the structural state label indicates a shell tongue direction bias state, the temperature error change of the current control cycle is correlated with the current thermal on / off path response state. If the gating rules of the differential protection action allow the differential contribution of the current control cycle to participate in the control output calculation, the differential contribution value calculated in the current control cycle remains unchanged, and this differential contribution value is used as the differential term output of the current control cycle. If the gating rules of the differential protection action prohibit the differential contribution of the current control cycle from participating in the control output calculation, the differential contribution value of the current control cycle is set to zero, and this zeroed differential contribution value is used as the differential term output of the current control cycle. When the structural state label indicates a symmetrical path state, the differential contribution value calculated based on the temperature error change of the current control cycle is directly used as the differential term output of the current control cycle. Then, the differential term output obtained in the current control cycle replaces the original differential term of the PID control algorithm in the current control output calculation, thereby completing the dynamic compensation control of the differential contribution process of the differential term in the PID control algorithm.

[0088] When the optimal action combination node indicates that the symmetrical reset toggle action is enabled, dynamic compensation control is performed on the control output process of the temperature control module.

[0089] Read the optimal action combination node corresponding to the current control cycle, confirm that the symmetrical reset toggle action in the optimal action combination node is enabled, read the control output value corresponding to the current control cycle, read the toggle increment value corresponding to the control output, generate a symmetrical reset toggle sequence based on the control output value and the toggle increment value, use the control output value as the reference output, perform numerical addition operation on the control output value and the toggle increment value to obtain the forward toggle output, and perform numerical addition operation on the control output value and the reverse value of the toggle increment to obtain the reverse toggle output; The reference output, forward toggle output, reference output, reverse toggle output, and reference output are then arranged in chronological order to form a complete toggle sequence. Each output value is then sent to the actuator in sequence, causing the actuator to sequentially execute reference hold, forward toggle, reference recovery, reverse toggle, and reference recovery within the current control cycle. This provides symmetrical reciprocating adjustment of the control output process. Simultaneously, temperature data collected by the temperature sensor is read during the execution of each sequence output value. The control output value, execution time, and temperature data corresponding to each sequence output value are linked and recorded. The final reference output after execution is used as the control output at the end of the current control cycle, thus completing the dynamic compensation control of the temperature control module's control output process.

[0090] It should be noted that after introducing a thermomagnetic liquid metal microchannel thermal switch into the temperature control module, the oxide skin on the surface of the liquid metal may form a one-sided residue and a tongue-shaped anchoring structure on the inner wall of the quartz or glass microchannel during repeated reciprocating motions. This alters the interfacial contact relationship of the thermal on / off path, causing the correspondence between the control output, the thermal on / off state, and the temperature response to no longer remain stable and consistent. Under the same temperature error conditions, the effect of the control output on temperature changes will shift due to different interfacial contact states, exhibiting a non-equilibrium response with path dependence and direction dependence. This disrupts the mechanism of the integral and derivative terms in the traditional PID control algorithm, which is based on the assumption of continuous and consistent response, leading to the accumulation of control deviations or response distortion. Therefore, dynamic compensation control is required for the PID control algorithm to adapt to changes in the thermal on / off path under different structural states, enabling the control output to re-establish a matching relationship with the actual temperature response, thereby ensuring the stability and consistency of the temperature regulation process.

[0091] It should be noted that the tongue-shaped anchoring structure refers to the localized adhesion and gradual extension of the surface oxide layer or oxygen hydroxylation component of liquid metal during its reciprocating movement along the inner wall of a microchannel, forming a sheet-like or strip-like solid structure with one end fixed and the other end freely extending. This structure is similar to a tongue-shaped overhang, with one end maintaining stable contact with the channel wall and forming an anchor point, while the other end can bend, adhere, or detach under the action of flow or force. This creates a local interface structure between the droplet and the channel wall that can be repeatedly contacted and reconstructed, and has a continuous impact on the droplet's movement path and the interface heat transfer contact state.

[0092] When the optimal action combination node indicates the activation of integral write gating, derivative protection, and symmetrical reset toggle actions respectively, dynamic compensation control is performed on each part of the PID control algorithm according to the corresponding gating rules or toggle sequence. This is because, under the influence of asymmetric interface structure, the integral term, derivative term, and control output itself no longer maintain a stable and consistent mechanism of action on temperature changes. The integral term is prone to amplifying deviations due to continuous error accumulation, and the derivative term is prone to excessive amplification due to abnormal local response. When the control output directly acts on the actuator, it may cause asymmetric response of the thermal on / off path. By gating the integral write process, selectively suppressing the derivative contribution process, and compensating for symmetrical disturbances in the control output process, sub-term corrections can be made for different sources of influence, so that the formation process of the control quantity and the actual temperature response can be re-established to match, thereby improving the stability and consistency of temperature control under complex interface conditions.

[0093] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multi-condition PID dynamic compensation method for a temperature control module, characterized in that, Includes the following steps: Based on the operating condition switching event flag of the temperature control module, the first toggle action and the second toggle action are determined; a first direction interpretation result is generated based on the first toggle action, and a second direction interpretation result is generated based on the second toggle action; Based on the interpretation results of the first direction and the interpretation results of the second direction, the state of the thermal on / off path response of the temperature control module is determined, and the structural state label is obtained. Action combination is generated from the action library corresponding to the structural state label to obtain the action combination graph; candidate action combination paths are constructed based on the action combination graph, and the path cost of the candidate action combination paths is determined. The pheromone parameters are updated based on the path cost to obtain the optimal action combination node, and the PID control algorithm is dynamically compensated based on the optimal action combination node.

2. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, Determine the first and second toggle actions, including: Obtain the operating condition switching event flag of the temperature control module; Based on the operating condition switching event flags and the PID control algorithm, obtain the control output of the temperature control module; The toggle increment is generated based on the control output; Based on the control output and the toggle increment, the first toggle action and the second toggle action are determined.

3. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, The first direction interpretation result is generated based on the first tug action, including: During the first toggle action, the first temperature change is determined based on the first temperature sample value collected by the temperature sensor and the first reference temperature sample value at the start of the first toggle action. Determine the direction of the first temperature change based on the first temperature change amount; Based on the first toggle output and control output corresponding to the first toggle action, determine the change in the first control output; The direction of change of the first control output is determined based on the amount of change in the first control output. Based on the first temperature change direction and the first control output change direction, a first direction interpretation result is generated.

4. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, The second directional interpretation result is generated based on the second flicking action, including: During the second toggle action, determine the second temperature change. Based on the second toggle output and control output corresponding to the second toggle action, determine the change in the second control output; The second direction interpretation result is generated based on the second temperature change and the second control output change.

5. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, Obtain the structure state labels, including: The state in which the thermal on / off path response of the temperature control module is consistent is defined as the symmetrical path state. The state in which the thermal on / off response of the temperature control module is inconsistent is defined as the shell tongue direction bias state. When the interpretation results in the first direction are the same as those in the second direction, the symmetric path state is used as the structural state label. When the interpretation results in the first direction differ from those in the second direction, the shell tongue orientation bias state is used as the structural state label.

6. The multi-condition PID dynamic compensation method for a temperature control module according to claim 2, characterized in that, The resulting action combination diagram includes: When the structural state label is in the shell tongue direction bias state: Based on the structural state label, the gating rules for integral writing gating actions in the PID control algorithm are determined; Based on structural state labels, the gating rules for derivative protection actions in the PID control algorithm are determined. Based on the control output and the toggle increment, the toggle sequence of the symmetrical reset toggle action is determined; The integral is written into the gating rules of the gating action, the gating rules of the differential protection action, and the toggle sequence of the symmetrical reset toggle action, and then associated and configured to obtain the action library; The motion library is used to generate motion combination graphs. When the structure status label is in symmetrical path state, the integral write gating action, differential protection action, and symmetrical reset toggle action are not enabled.

7. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, Constructing candidate action combination paths based on the action combination graph includes: The ant colony algorithm is used to initialize the action combination graph to obtain pheromone parameters. Based on pheromone parameters, candidate action combination paths are constructed.

8. The multi-condition PID dynamic compensation method for a temperature control module according to claim 1, characterized in that, Determine the path cost of candidate action combination paths, including: Obtain the target temperature from the temperature control module; Obtain the current temperature of the temperature control module; The temperature error is obtained by performing a difference calculation between the target temperature and the current temperature. Obtain the change in control output; Based on the temperature error and the change in control output, the path cost of candidate action combination paths is determined.

9. The multi-condition PID dynamic compensation method for a temperature control module according to claim 7, characterized in that, The nodes that yield the optimal action combination include: Based on the path cost, the pheromone parameters are updated to obtain the updated pheromone parameters; When the iteration termination condition is met, the optimal action combination node is determined based on the updated pheromone parameters.

10. A multi-condition PID dynamic compensation method for a temperature control module according to claim 6, characterized in that, Dynamic compensation control of the PID control algorithm includes: When the optimal action combination node indicates that the integral write gating action is enabled, the integral write process of the integral term in the PID control algorithm is dynamically compensated and controlled according to the gating rules of the integral write gating action. When the optimal action combination node indicates that the differential protection action is enabled, dynamic compensation control is performed on the differential contribution process of the differential term in the PID control algorithm according to the gating rules of the differential protection action. When the optimal action combination node indicates that the symmetrical reset toggle action is enabled, dynamic compensation control is performed on the control output process of the temperature control module.

Images