Control method of gas chromatography electronic gas circuit

Abstract

The invention discloses a control method of a gas chromatography electronic gas circuit, and particularly relates to the technical field of electronic gas circuit control. Dynamic adjustment of integral terms of the PID controller is realized by constructing a gas path elastic parameter model containing the maximum instantaneous flow rate, the maximum pressure difference response rate and the gas source recovery delay time; in the control process, feedback values and control signals are collected in real time, a system operation scene is judged, a control structure is automatically switched to be in a non-PID mode in an unstable state, and an integral path is frozen; after the system is recovered to be stable, smoothly recovering a PID structure, and adaptively optimizing gas path elastic parameters based on an adjusting effect; the method can be widely applied to a high-precision gas chromatograph, and safer, more efficient and more intelligent gas path control is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic gas circuit control, and in particular to a method for controlling a gas chromatograph electronic gas circuit. Background Art

[0002] Electronic gas flow control in gas chromatography (GC) refers to the precise electronic regulation and management of the flow of carrier gases (such as helium and nitrogen) within a GC. It utilizes an electronic pressure controller (EPC) or electronic flow controller (EFC) to monitor and automatically adjust gas pressure or flow in real time, ensuring stable and repeatable gas flow conditions during analysis, thereby improving analytical accuracy, sensitivity, and system automation. This control method is widely used in modern GCs and is a key component in achieving efficient and high-precision analysis.

[0003] The existing technology has the following shortcomings: When using the PID algorithm in gas chromatography electronic gas flow control, if a sudden change in gas source pressure (such as low cylinder pressure at startup) causes the system to fail to reach the setpoint for an extended period, the PID controller's integral term will accumulate errors, locking the control signal at maximum output and driving the proportional solenoid valve fully open. Once the gas source returns to normal, a large influx of gas causes a dramatic pressure overshoot, potentially damaging the chromatographic column, inlet, or detector. This loss of control stems from the controller's lack of integral limiting or anti-saturation mechanisms, resulting in a lack of adaptive response to sudden anomalies. This is a hidden but potentially devastating problem in PID control. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for controlling a gas chromatograph electronic gas circuit to solve the deficiencies in the background technology.

[0005] In order to achieve the above object, the present invention provides the following technical solution: a method for controlling a gas chromatograph electronic gas circuit, comprising: Constructing a gas path elasticity parameter model, wherein the gas path elasticity parameters include the maximum instantaneous flow rate of the gas channel, the maximum pressure difference response rate, and the gas source recovery delay time; The gas path elasticity parameter is embedded in a PID controller structure to construct an integral regulation function, wherein the integral regulation function adjusts the accumulation rate of the integral term in real time according to the gas path elasticity parameter; Collect current gas state data and calculate the deviation between the actual feedback value and the target set value as the error input; Outputting a control signal according to the error input and the integral regulation function to drive a proportional solenoid valve to regulate gas flow or pressure; classifying the current control scenario based on the feedback value, control signal, and historical status data; When it is determined that the control scenario is an unstable scenario, the controller structure is automatically adjusted to a non-PID structure, and the update of the integral term is frozen or suppressed; After detecting that the system state has returned to stability, the PID control structure is restored, and the air path elasticity parameters are adaptively optimized and updated based on historical adjustment effects.

[0006] Preferably, constructing the gas path elastic parameter model includes: During the initial operation of the system or after each gas source change, a preload test is performed on each gas channel based on the set step pressure excitation signal; The maximum instantaneous flow rate is obtained by sampling the output signal of the flow sensor into a discrete time series, calculating the maximum difference between adjacent time points, and dividing the difference by the sampling interval. The obtained value is the maximum instantaneous flow rate estimate; The maximum pressure difference response rate is calculated using an exponential fitting method. The collected pressure change curve is fitted into an exponential growth model, and its derivative value at the maximum gradient is calculated by derivation. The derivative value corresponds to the maximum pressure difference response rate. The gas source recovery delay time is: the time when the first non-zero effective flow signal appears after the gas source is restored, and subtracted from the timestamp when the interruption occurs, and the resulting time difference is the recovery delay time.

[0007] Preferably, constructing the integral adjustment function includes: Set up a baseline structure for the integral regulation function, where the cumulative rate of the integral term is an adjustable parameter rather than a fixed proportional constant; The maximum instantaneous flow parameter of the target channel is extracted from the gas path elastic parameter model and used to set the initial gain upper limit of the integral regulation function. The integral gain upper limit is derived from the ratio between the maximum instantaneous flow and the control period. When the control system detects a trend of continuous error accumulation, it adjusts the integral term accumulation step size based on the maximum pressure difference response rate, automatically reducing the accumulation rate when the system's response capability is weak. A combined judgment of the error change direction and rate of change is performed before integral accumulation. If repeated or oscillating error directions are detected, and the maximum pressure difference response rate of the corresponding channel is less than a preset threshold, the integral adjustment function is immediately triggered to reduce the integral step size. The gas source recovery delay time is used to set the minimum delay threshold for integral accumulation, that is, the integral accumulation is frozen before the delay time is reached.

[0008] Preferably, calculating the deviation between the actual feedback value and the target set value as the error input includes: The actual state value of the gas in the current channel is collected at fixed time intervals, and the sensor output is denoised using a sliding time window. The denoised current state value is compared with the target set value of the corresponding channel to obtain a real-time deviation value, which is used as the error input of the controller.

[0009] Preferably, driving the proportional solenoid valve to adjust the gas flow or pressure includes: The error input is combined with the integral regulation function output to form a set of multi-dimensional control instructions, including the basic signal value for adjusting the proportional valve voltage, current or duty cycle; Determine the current operating state of the channel. If the control signal change amplitude exceeds the preset threshold within a continuous cycle, the signal change amplitude is compressed; When the gas source pressure fluctuates violently, the pressure differential response rate parameter is called to impose a dynamic upper limit constraint on the control signal. The dynamic upper limit constraint is determined based on the maximum pressure differential response rate of the channel. If the response rate is lower than the system's preset critical value, the control signal change rate is automatically limited to a range that matches the physical response capability of the channel.

[0010] Preferably, classifying the current control scenario includes: Collect the feedback value and control signal output value of the current cycle and construct a set of historical state sequences together with the data of the previous cycles; Calculate the response consistency index between the feedback value and the control signal to reflect whether the control action causes the expected pressure or flow change. The response consistency index compares the direction of change of the control signal and the direction of change of the feedback value to see if they are consistent. If the directions are consistent but the amplitude ratio is continuously low, it indicates a low response rate state. If the directions are opposite, it is marked as a reverse feedback anomaly. Based on the consistency index and the feedback trend curve, the current gas circuit operation state is predicted. If the control signal is continuously applied but the feedback value does not change significantly, the state is marked as a response hysteresis scenario. If the feedback value changes suddenly without any obvious change in the control signal, and the change exceeds the maximum pressure difference response capability of the current channel, the state is marked as an external disturbance scenario.

[0011] Preferably, automatically adjusting the controller structure to a non-PID structure includes: Based on the response relationship between the feedback value and the control signal, identify the current control scenario as any unstable scenario such as response hysteresis, reverse feedback or external disturbance; When the system is in this scenario state for multiple consecutive control cycles, and the error direction fluctuates frequently or the control output amplitude is abnormally amplified, the control structure switching logic is triggered; Temporarily disconnect the integral path in the original PID structure, retain only the proportional term and the differential term to participate in the control decision, and construct a simplified PD control structure; After the control structure is switched, the integral term accumulator state is frozen. The integral term freeze adopts a two-layer freezing mechanism. The first layer is periodic freezing, which prohibits integral updates within a fixed time period. The second layer is dynamic freezing, which automatically determines whether the freezing period should be extended or released in advance based on the system feedback recovery speed and pressure difference response capability.

[0012] Preferably, restoring the PID control structure after the detection system state returns to stability includes: Continuously monitor the rate and direction of change of the feedback value during the freezing period to determine whether it is stable and close to the target value, and calculate whether the range of change is continuously within the set stability threshold; The time period response trend analysis method is used. That is, if the rate of change of the feedback value gradually decreases and the error direction does not reverse within multiple consecutive control cycles, it is judged that the control has entered a stable state and the PID structure recovery process is allowed to begin. After recovery, the PID control structure reactivates the joint regulation mode of the integral term, proportional term, and differential term, and limits the integral output amplitude in the first two cycles; The feedback response difference before and after the control structure is restored is recorded to generate a regulation performance score for elastic parameter optimization, which is used to dynamically adjust the maximum pressure difference response rate and recovery delay time in the gas path elastic parameters.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. This invention dynamically adjusts the integral term in the PID controller by introducing a gas path elasticity parameter model, effectively avoiding problems such as integral saturation and control overshoot caused by gas source disturbances, feedback hysteresis, or error accumulation. During the control process, the system determines the current operating scenario based on real-time feedback values and control signal behavior, and automatically switches the control structure to a non-PID mode when an unstable state is detected, enhancing the stability and safety of the regulation process.

[0014] 2. This invention also possesses self-learning capabilities, enabling smooth restart of the PID control structure after system status recovery. It also adaptively optimizes and updates the gas path elasticity parameters based on historical regulation results, enabling the control system to continuously evolve over time, improving regulation accuracy and response efficiency. This method is suitable for multi-channel, complex gas path control scenarios, and is particularly well-suited for gas chromatography systems requiring high sensitivity and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0018] For examples, see Figure 1 As shown, the method for controlling the gas chromatography electronic gas circuit of this embodiment includes: Constructing a gas path elasticity parameter model, wherein the gas path elasticity parameters include the maximum instantaneous flow rate of the gas channel, the maximum pressure difference response rate, and the gas source recovery delay time; The gas path elasticity parameter is embedded in a PID controller structure to construct an integral regulation function, wherein the integral regulation function adjusts the accumulation rate of the integral term in real time according to the gas path elasticity parameter; Collect current gas state data and calculate the deviation between the actual feedback value and the target set value as the error input; Outputting a control signal according to the error input and the integral regulation function to drive a proportional solenoid valve to regulate gas flow or pressure; classifying the current control scenario based on the feedback value, control signal, and historical status data; When it is determined that the control scenario is an unstable scenario, the controller structure is automatically adjusted to a non-PID structure, and the update of the integral term is frozen or suppressed; After detecting that the system state has returned to stability, the PID control structure is restored, and the air path elasticity parameters are adaptively optimized and updated based on historical adjustment effects.

[0019] When the gas chromatograph system is first operated, or each time the gas cylinder is replaced or the gas source channel is switched, performance tests must be performed on all key gas channels (such as the inlet, pre-column gas line, split gas line, detector makeup gas line, etc.).

[0020] The test steps are as follows: The control system sends a preset step pressure command to the target channel (e.g., instantaneous jump from 0 kPa to 200 kPa); At this time, the solenoid valve is kept open, and the pressure and flow rate in the channel are collected over time. The sampling frequency is not less than 100 Hz, and the recording time span is 5 to 10 seconds to ensure that the gas filling process and the steady-state delay period are covered; The sampled data is sent to the subsequent calculation module to extract the dynamic characteristic indicators of the channel.

[0021] In order to describe the maximum flow fluctuation capacity that a gas channel can withstand per unit time, the present invention provides a maximum instantaneous flow calculation method based on differential sequence extraction, as follows: Processing the output signal of the flow sensor during the test into a set of discrete time series; Perform differential processing on adjacent time points to obtain a set of instantaneous flow changes; Extract the difference with the largest absolute value from the difference sequence and divide it by the corresponding sampling interval to obtain the maximum instantaneous flow estimation value; If the flow sensor has analog output, A / D conversion and zero drift calibration must be performed first to ensure calculation accuracy.

[0022] Through this indicator, the control system can determine the maximum gas inrush tolerance of each channel, and then set the dynamic upper limit of the pressure control response range and integration rate in the controller.

[0023] In view of the response speed of the air path under step pressure input, the present invention introduces an exponential fitting model to model the pressure change curve to extract its maximum pressure difference response rate.

[0024] The specific steps are as follows: Select the data segment where the channel pressure changes with time, recorded as P(t); fit the data segment to a first-order exponential function: ;in, is the initial pressure, A is the pressure rise amplitude, and k is the response rate constant. Taking the derivative of the above fitting function, we get the pressure difference slope expression: ; Calculate the maximum value of the derivative function, that is, obtain the maximum derivative at t=0: Maximum pressure difference response rate = ; This rate value is used to reflect the response limit of the channel to the instantaneous pressure regulation instruction, and can provide the upper limit boundary of the pressure difference derivative for the subsequent integral term dynamic function.

[0025] In order to accurately determine the system's ability to recover after a sudden gas source interruption, the present invention defines the gas source recovery delay time as: the delay time required for the system to detect a valid airflow signal for the first time after the gas source is restored.

[0026] The specific measurement method is as follows: A gas cut-off trigger point is set in the control system. That is, when the pressure is continuously lower than a certain safety lower limit (such as 20 kPa) for more than a set time (such as 1 second), it is determined that the gas source is cut off, and the cut-off timestamp T0 is recorded; Continuously monitor the flow sensor output and record the timestamp T1 corresponding to the first non-zero effective gas flow signal (e.g., ≥ 0.2 mL / min); Calculate the gas source recovery delay time as ; To improve robustness, an additional layer of pre-column pressure change rate judgment can be added: if the flow signal recovers but the pressure rises slowly, T1 will be delayed until the pre-column pressure reaches the set stable value (such as 90% of the target value).

[0027] Through this delay parameter, the system can determine whether the current gas source is in a fast response state, thereby dynamically suppressing or releasing the integral term accumulation rate.

[0028] In summary, through the four-step modeling process described in this embodiment, the system can form a set of real-time updated gas path elastic parameter models. This model not only dynamically reflects the responsiveness of different channels under specific environmental and gas source conditions, but also provides multi-dimensional input constraints for the controller. This fundamentally improves the intelligence and safety of electronic gas path control. In particular, it effectively avoids hidden dangers such as integral saturation and pressure overshoot in scenarios with sudden gas source fluctuations, thereby improving the overall stability and service life of the gas chromatography system.

[0029] In the PID controller architecture, the integral term uses a constant gain constant to continuously accumulate the error. This structure is common in traditional industrial control, but it lacks the ability to identify system hysteresis and response differences, and is prone to excessive integral response when input conditions change suddenly or the gas source is unstable.

[0030] Therefore, this paper proposes a "dynamic integral control structure." Its core concept is to set the amplitude or frequency of the integral term accumulation as a variable parameter, which is controlled in real time by the gas path elasticity parameters of the current system state. By dynamically linking the integral behavior with the gas path responsiveness, the controller can continuously assess and limit the integral risk during operation.

[0031] In actual operation, the maximum instantaneous flow rate of each gas channel reflects its carrying capacity for inrushing gas per unit time. The present invention uses this parameter to set the dynamic upper limit of the integral term accumulation amplitude.

[0032] Specifically, at the beginning of each regulation cycle, the controller calls the channel's corresponding maximum instantaneous flow parameter and sets the maximum allowable integral term variation for that cycle based on a preset conversion relationship. If the actual error exceeds a threshold and the channel's gas-carrying capacity is weak, the upper limit is automatically tightened, suppressing the integral accumulation. This prevents system overshoot caused by the integral response exceeding the channel's physical capacity.

[0033] This mechanism ensures that the integral term does not amplify infinitely above the channel flow limit, so that the controller output is always controlled within the safe regulation range.

[0034] Different channels have significantly different response rates to pressure changes. This is especially true at low pressure starting points or when the air supply is unstable. Some channels may exhibit significant hysteresis or slow response during pressure regulation. If the integral term continues to accumulate error at a fixed rate during this process, the system output can easily exceed the feedback response capability, resulting in a typical "overshoot-pullback-overshoot" oscillation pattern.

[0035] This invention dynamically adjusts the frequency or step size of the integral term updates by analyzing the maximum differential pressure response rate parameter. For example, if the controller detects a weak differential pressure response in the current channel, it will automatically switch to a slow integral cadence and may activate an internal freeze mechanism to pause integral updates. Normal integral updates will only resume when the feedback signal is detected to be within an acceptable range of variation.

[0036] This structure allows the controller to prioritize proportional and differential adjustments when the system response lags severely, delaying or suppressing the integral action, thereby avoiding the accumulation of continuous integral errors caused by misjudgment.

[0037] After a sudden loss or replacement of the gas source, the system undergoes an uncontrolled recovery process. During this period, the pressure and flow signals may experience drastic jumps. If the controller fails to correctly recognize this state and continues to accumulate the integral term, it will misjudge the state after the gas source is restored and amplify the integral output, ultimately causing a severe pressure overshoot.

[0038] To address this situation, the present invention proposes incorporating a gas supply restoration delay time as one of the integral activation thresholds. After the system detects the restoration of gas supply, the controller will not activate the integral path until at least this delay time has elapsed, during which time it will continuously evaluate whether the feedback value has reached a preliminary steady state. Accumulation of the integral term resumes only after the delay time has expired and the feedback trend has entered a stable range.

[0039] In addition, if it is detected that the gas source recovery process is still accompanied by high-frequency fluctuations or repeated interruptions, the system can further extend the integral freeze period and temporarily switch to non-integral adjustment mode to ensure the output stability and safety of the control system.

[0040] To achieve long-term adaptability to complex operating scenarios, the present invention also incorporates an integral behavior feedback recording module. During operation, the system monitors the integral term, error trends, and system output status in real time within each cycle, storing this information in a loop buffer. This information can be used to subsequently optimize the parameter matching strategy of the integral control function, enabling the controller to automatically develop an integral control model adapted to specific gas source and channel conditions after extended operation.

[0041] For example, if a channel operates in a slow response state for a long time, the system can gradually lower the upper limit or frequency of its integral term to construct a more conservative integral response curve, thereby making the overall system more robust and dynamic adaptable.

[0042] In the gas chromatography electronic gas circuit control method of the present invention, to achieve accurate, efficient, and safe control input to the controller, it is necessary to obtain the current actual operating state of the gas channel and compare it with the target set value to form a feedback error input. This error serves as the core calculation basis of the PID control algorithm and has a decisive influence on the overall system output behavior.

[0043] However, in actual operation, sensor data is often affected by a variety of disturbance factors, such as: gas source fluctuations, ambient temperature changes, electromagnetic interference, sensor zero drift, pipeline micro-vibration, etc. If the difference between the unprocessed real-time sampling value and the set target value is directly calculated, it is easy to cause over-adjustment, misadjustment and even system oscillation. Therefore, the present invention proposes a set of highly robust error input calculation methods. The core idea is: through multi-layer data processing and control strategy correction, the error calculation results can reflect the system status as realistically as possible, ensuring that the PID controller output is stable and responsive.

[0044] During system operation, the controller deploys high-precision pressure and flow sensors at key points in the gas path. These sensors sample data at regular intervals, typically ranging from tens to hundreds of milliseconds, to ensure continuous monitoring of gas conditions.

[0045] The signal collected by each sensor is converted to a digital quantity, timestamped, and then transmitted to the data processing module. The collected data includes, but is not limited to, pre-column pressure, split flow, makeup flow, and detector gas flow rate, which are used to evaluate the operating status of each channel.

[0046] To enhance data anti-interference capabilities and extract more reliable state values, the present invention introduces a time-weighted median filtering algorithm. Specifically, the system calls several previous historical sampling values during each calculation to form a time sliding window.

[0047] Within this time window, all sampled values are first sorted, and the middle value is selected as the representative value for the current period. To enhance responsiveness to trends, the algorithm also assigns higher weights to values closer to the current moment. In other words, sampling points closer to the current moment have a greater impact on the filtered results. This approach mitigates the sensitivity of traditional mean filtering to outliers while retaining the ability to identify fluctuating trends in gas conditions.

[0048] In addition, for instantaneous spikes in the data (such as outliers caused by valve vibration or electromagnetic interference), the system automatically eliminates or smoothes them by setting abnormality identification rules (for example, the mutation rate exceeds the normal change rate by several times), preventing such outliers from being misused in error calculations and causing the controller to respond incorrectly.

[0049] After obtaining the filtered current state value, the system compares it with the preset target value to form a deviation value. Unlike traditional algorithms that directly use the difference as the error, this invention further introduces an "error direction trend judgment" mechanism.

[0050] Specifically: In addition to calculating the deviation of this cycle in each cycle, the system also records the direction of error change in recent cycles and analyzes whether there is a convergence, divergence or oscillation trend between the current state and the target value.

[0051] If the current error direction is determined to be changing repeatedly (i.e., the error value frequently jumps between positive and negative), and is accompanied by poor feedback data stability, the system will identify this as an "error uncertainty phase" and will automatically reduce the participation weight of the current error to prevent controller misadjustment due to short-term fluctuations.

[0052] On the other hand, when the system detects that the error direction is completely opposite to the target change direction and the error value continues to expand, the system will enable the "trend correction mechanism" and use the historical state average as the weight compensation benchmark to avoid extreme deviation of the error term.

[0053] In a complex multi-channel operating environment, the response capabilities, inertia, and flow rate variation characteristics of each gas channel vary significantly. To further enhance the adaptability of error input, the present invention introduces gas path elasticity parameters to dynamically weight the error calculation.

[0054] The specific method is: the system reads the maximum pressure difference response rate parameter and the gas source recovery delay time parameter corresponding to the current channel, and judges the current response capability status of the channel accordingly.

[0055] If the maximum differential pressure response rate of the channel is low, or if the gas supply has just been interrupted and has not yet stabilized, the system will automatically reduce the weight of the current error value in the PID input and use the average deviation of the historical stable period as the primary input. This approach can prevent excessive transient errors from triggering strong regulation when the system is experiencing response hysteresis or gas source fluctuations, causing pressure overshoot or sudden flow changes.

[0056] In addition, the system is set with a dynamic threshold. When the gas source recovery time exceeds a preset safety window, or the sensor status value shows high stability for several consecutive cycles, the error input weight will automatically return to normal, ensuring that the system gradually enters the high-response control stage.

[0057] Taking into account the possibility of serious sensor drift, data loss or abnormal mutation in actual applications, the present invention provides a two-level protection mechanism of hardware and software in the error input module.

[0058] When the error value suddenly changes beyond the safety limit range set by the controller, the system will trigger the error blocking logic, suspend the integral and proportional control actions, and switch to the mode of maintaining only the minimum flow or pressure regulation to ensure that the entire gas system is in a safe operating state.

[0059] At the same time, the system will record the abnormal status and push it to the log module for subsequent self-learning mechanism to optimize the error response model.

[0060] In each control cycle, the system first obtains the real-time error value calculated by the pre-processing module and the output of the integral control function constructed with the gas path elastic parameters. Together, they are used to generate the controller's control signal for the current cycle.

[0061] Unlike traditional linear superposition methods, this invention employs a fusion mapping strategy, combining the error term and the integral adjustment value in a weighted manner based on the current channel operating state. If the system is experiencing error fluctuations, the proportional component is weighted more heavily; as the system approaches steady state, the integral component is weighted more heavily to promote error convergence. This strategy dynamically allocates the proportion of the two components based on the channel response phase, enhancing the control's adaptability to specific scenarios.

[0062] The fusion results form a set of control instruction parameters, including control direction (open or close the valve), target adjustment amplitude, recommended signal change rate, etc., which are used by the subsequent drive circuit to generate the actual drive signal.

[0063] To prevent the proportional solenoid valve from frequently opening and closing in a short period of time due to repeated changes in the error direction, the present invention incorporates a control signal smoothing module into the signal output channel. This module determines the trend of the control signal's continuous periodic changes and initiates a smoothing strategy if it detects that the control signal's direction is continuously reversing or the amplitude of the change is excessive.

[0064] The smoothing strategy includes the following specific actions: Limit the maximum change range of the signal. Even if the error between the previous cycle and the current cycle changes significantly, the control signal is only allowed to increase or decrease slowly within a limited range to prevent the valve from jumping suddenly. Delay the execution of unnecessary reversal instructions. For those fluctuations where the signal reverse amplitude is lower than the system noise threshold, this module will temporarily suspend the output and re-evaluate after the signal trend stabilizes; A buffer step table is established. Based on historical valve response behavior records, the system can calculate the actual reaction time of various channels to control signals of different amplitudes, thereby automatically releasing the pressure in stages before a sudden change in the signal, improving system stability.

[0065] Through the above processing, the control signal output process becomes smoother and the rhythm perception is stronger, which greatly reduces valve fatigue and air pressure fluctuations caused by small errors frequently triggering large actions.

[0066] Considering that different gas channels have different response upper limits at the hardware level (for example, the regulation capabilities of small-diameter tail blowing ports and large-flow diversion ports are significantly different), if the control signal exceeds the physical response capability of the channel, it may cause abnormalities such as overpressure, overcurrent or valve blockage.

[0067] Therefore, the present invention further proposes a "control signal dynamic limiting mechanism" for applying safety boundary constraints to the fused output control signal.

[0068] Before calculating the control signal, the system first reads the channel's maximum pressure differential response rate parameter (defined in the elastic parameter model above). This parameter estimates the maximum acceptable pressure or flow change per unit time for the channel. If the control signal's suggested change exceeds this estimate, the output amplitude is automatically reduced to keep the actual signal within the valve's controllable range.

[0069] At the same time, if the channel is in a state of delayed gas source recovery in the previous cycle, the system will also superimpose a "change buffer period" mechanism to further tighten the fluctuation amplitude of the control signal to avoid issuing drastic adjustment instructions because the gas source has not yet returned to stability.

[0070] This approach ensures that the controller output does not "get ahead of the gas path response" and establishes a dynamic coordination mechanism between signals and physical behavior.

[0071] After completing the above processing, the final control signal is sent to the proportional solenoid valve driver module. The driver module matches the signal type (such as voltage, PWM duty cycle, current, etc.) and converts it into a control format suitable for the specific proportional valve.

[0072] To further improve system response quality, the driver module also features a signal rate-of-change recording function. This function records the signal change value within each cycle and compares it with the signal of the previous cycle, forming a continuous signal change curve. This curve is not only used for post-diagnosis of valve operating conditions, but also allows the system's self-learning module to analyze and adjust the control strategy.

[0073] For example, if a channel frequently experiences drastic signal changes but small feedback fluctuations during long-term operation, the system will identify that the channel may have control redundancy and appropriately reduce its control signal sensitivity in subsequent cycles to extend the life of the valve body and improve overall control smoothness.

[0074] In the inlet split control of a gas chromatography system, after applying this control signal generation and output strategy, compared with the traditional PID control method, the system error convergence time was shortened by about 30%, the valve operation frequency was reduced by nearly 50%, and the gas pressure amplitude was controlled within ±0.5 kPa under the scenarios of carrier gas fluctuation and sudden change of pre-column pressure, significantly improving the control accuracy and hardware stability.

[0075] During each control cycle, the system collects the following three core variables from key data sources: Sensor feedback values of the current cycle (including pressure, flow or related gas status indicators); The control signal value of the current cycle (such as proportional solenoid valve voltage, current or PWM duty cycle); Control the signal change trend (i.e. the direction and magnitude of change between this cycle and the previous cycle).

[0076] The system will form a sliding time window with the data collected in the current cycle and the data in the previous cycles. The window length is usually set to 5 to 10 control cycles to ensure that the analysis has sufficient time resolution and trend characteristics.

[0077] This historical data is cached in a structured format for subsequent consistency analysis and pattern recognition operations.

[0078] Within this time window, the system matches the control signal's changing behavior with the feedback value's response behavior and calculates a response consistency index. This index reflects whether the control signal's effect causes the feedback value to change as expected.

[0079] The system executes the following judgment logic in each cycle: If the control signal and the feedback value change in the same direction, and the feedback value change amplitude is roughly proportional to the control signal amplitude, it is judged as "normal response"; If the control signal is continuously adjusted, but the feedback value hardly changes or changes very slowly, it is marked as a "low response rate scenario"; If the direction of the control signal is opposite to the direction of change of the feedback value (for example, increasing the opening but the pressure decreases), it is marked as a "reverse feedback scenario", indicating that there is a physical anomaly or control failure.

[0080] This directional analysis method abandons the traditional crude method of judging whether it is abnormal based on the absolute value of the error, and pays more attention to the consistency of the system's dynamic behavior and the rationality of the trend, and has stronger engineering practicality.

[0081] Based on the response consistency index and historical feedback trends, the system divides control scenarios into the following categories and records their frequency and duration in the cache: Normal control scenario: The control signal and feedback value are highly consistent, and the feedback is adjusted in time with the control output; Hysteresis response scenarios: The control signal direction is clear, but the feedback changes with significant lag, small fluctuations, or no response; External disturbance scenario: The feedback value changes significantly without any change in the control signal. This is usually caused by unstable air source, electromagnetic interference, or hardware jitter. Reverse feedback scenario: The feedback value changes in the opposite direction to the control signal, which may be caused by valve sticking, return air shock, or sensor false alarm; Repetitive structural disturbance scenario: If the same abnormal feature sequence appears multiple times in the historical status data, such as periodic abnormal feedback values at regular intervals, it indicates the presence of hardware structural interference or algorithm jitter.

[0082] After each scene type is identified, the system will assign a scene label and record its entry time, duration period and context status (such as gas source pressure, temperature, operation stage, etc.) to provide a reference for subsequent processing.

[0083] When the system identifies that it is currently in an "abnormal control scenario", the controller will dynamically adjust its regulation strategy. The specific response mechanism includes: If the response is to a hysteresis scenario, suspend the integral term update and reduce the proportional adjustment speed to avoid overshoot due to misjudgment; If it is an external disturbance scenario, the controller enters protection mode, maintaining only the minimum airflow output and waiting for the feedback value to stabilize before resuming normal regulation; If it is a reverse feedback scenario, the controller will trigger the diagnostic module to further determine whether a soft restart or alarm processing is required; If it is a repetitive structural disturbance scenario, the system will record the scenario trigger cycle and automatically include the channel in the "abnormal learning list", entering the low-frequency response mode to reduce hardware fatigue caused by repeated control.

[0084] This scenario-driven strategy switching mechanism enables PID control to no longer execute a fixed set of adjustment formulas, but instead "selects behavior" based on the scenario classification results, greatly improving the control system's adaptability in complex and uncertain environments.

[0085] In order to improve the long-term stability and adaptability of the system, the present invention also designs a self-learning mechanism based on the control scene recognition results.

[0086] The system stores the characteristic parameters and control response results of all abnormal scenarios in a local learning model, forming an abnormal pattern library. In the next round of control tasks, the system compares the current state with historical patterns. If the current characteristics are found to be highly similar to a historical abnormal scenario, it will predict and switch to the corresponding response strategy in advance.

[0087] For example, if the system detects that the feedback value enters a "slow decrease followed by a sharp increase" pattern, and this pattern was previously marked as an "unstable carrier gas after injection" scenario, the slow pressure start function can be enabled in advance or the integral term response time can be delayed to prevent system misadjustment.

[0088] This mechanism makes the controller more stable as it runs longer, and it has intelligent judgment capabilities similar to "expert operator experience".

[0089] The system first evaluates the operating status in real time. Typical characteristics of unstable scenarios include: Hysteresis response scenarios: The control signal continues to change, but the feedback value changes slightly or has a significant delay; Reverse feedback scenario: The feedback value changes in the opposite direction to the control signal, such as the pressure drops after the valve is opened; External disturbance scenarios: The feedback value changes suddenly without any change in the control signal, exceeding the normal fluctuation range; Integral saturation scenario: After the integral term accumulates to the upper or lower limit, it continues to maintain without regression, and the control output becomes invalid.

[0090] When the system identifies any of the above states by analyzing historical feedback trends, control signal change patterns and error behaviors, and confirms that it persists for more than a preset judgment period (such as 3 to 5 sampling periods), it is determined to be an "unstable scenario" and triggers the subsequent structural switching logic.

[0091] After identifying the unstable state, the system immediately executes the controller structure conversion process to convert the standard PID controller into a non-integral structure. The conversion method is as follows: Disconnect the integral path: temporarily cancel the error accumulation link and prohibit the error from being added to the control signal; Preserve proportional and differential terms: The system responds to control based only on the current error value and its changing trend, avoiding incorrect adjustment of the integral term due to feedback hysteresis; Update flag: Write the "integral frozen state" flag into the controller status register to ensure that subsequent logic judgments are executed in the frozen state.

[0092] This non-PID structure is a simplified control form with faster response and stronger suppression ability. It is especially suitable for dealing with short-term disturbances or responding to unknown abnormal conditions.

[0093] After the control structure is switched, the system performs a "freeze" process on the current integration state, that is: Lock the integral accumulator: no longer update the historical error accumulation value; Prevent negative integral: If the error direction is reversed, the accumulated integral value is not released immediately to prevent overcompensation; During the freezing period, the control output does not contain the integral component, ensuring that the regulation output only reflects the current actual system requirements.

[0094] In order to ensure that the freezing behavior can prevent system misadjustment without causing long-term deviation backlog, the present invention proposes a double integral unlocking mechanism: Fixed cycle unlocking: When the integral freeze state lasts for more than the set period (such as 10 control cycles), the system determines whether the current state has been restored. If the feedback value enters the stable window, it will try to restore the PID structure. Dynamic unlocking: If during the freeze period the rate of change of the system feedback value is synchronized with the control signal trend again and meets the response capability of the channel (such as within the maximum pressure difference response rate), the system can release the integral freeze in advance.

[0095] This mechanism ensures that the system has the ability to "soft reset" after handling sudden abnormalities, avoiding long-term degradation of control accuracy.

[0096] In some cases, the integral term is not accumulated by mistake in unstable scenarios, but rather leads to "integral saturation" due to ineffective regulation for a long time, for example: If the air source of a certain control channel fails, the target pressure can never be reached; The control signal has reached the maximum output but the feedback value has not increased significantly.

[0097] To prevent such phenomena from worsening, the present invention has a built-in integral saturation detection module. When the controller detects that the integral term is continuously at the upper and lower limit boundaries and the control signal tends to a fixed value, it automatically switches to a non-PID control structure and enters a "saturation frozen state".

[0098] The system maintains low-response control in this state, waiting for the fault to be eliminated or feedback to be restored before reactivating the integral path.

[0099] In a gas chromatography system with multi-channel gas path control, the structure switching and integral freezing mechanism deployed by the embodiment of the present invention exhibits significant advantages in the following scenarios: During the carrier gas bottle switching process, the system enters a feedback oscillation state due to short-term unstable gas supply. Traditional PID adjustment repeatedly operates, resulting in excessive pressure fluctuations. However, the system of the present invention automatically switches to non-integral control, shortening the pressure recovery time by 40% and reducing the fluctuation amplitude by more than 50%. When the hydrogen channel of a detector responds slowly due to aging of the pressure regulating valve, the system identifies a "low response scenario" and freezes the integral term to avoid misadjustment, allowing the system to continue to operate stably and avoiding detection interruption.

[0100] When the controller is in non-PID structure operation (such as PD structure or proportional mode), the system will continuously collect feedback values, error trends and control signal change data to determine whether the current operating state has entered a stable range.

[0101] The system sets a time sliding window, the window length of which can be set to 5 to 10 control cycles. During this period, if the following conditions are met, it is determined that "the system has returned to stability": Feedback value fluctuation is small: the maximum fluctuation range of feedback values such as pressure or flow within this window is less than the stability tolerance defined by the system; Error change rate convergence: The error value decreases cycle by cycle without positive or negative reversal, indicating that the error is approaching the target; No sudden changes in the control signal: The control output maintains the same direction and changes slowly, without any regulation jump or output boundary saturation.

[0102] When the above three judgment conditions are all met, the controller considers that the current gas path state has the conditions to restore complete control capability.

[0103] After the state is determined to be stable, the system will trigger the control structure recovery process, the specific operations are as follows: Unfreeze flag: The system first clears the points freeze status flag and unlocks the points module; Reactivate the integral path: Allow the integral term to re-participate in the control calculation, but limit its output to a minimum change range for the first two cycles to avoid sudden rebound caused by error backlog; PID structure is fully restored: the controller structure is restored to the full-function PID form including proportional, integral and differential parts; Switching smooth transition mechanism: In the initial recovery stage, the built-in signal transition module is used to fuse and smooth the output signals of the non-PID structure and the recovered PID structure to ensure that there are no sudden changes or pulse jumps in the system adjustment process.

[0104] Through the above process, the controller can achieve an intelligent and smooth transition from safety mode to normal operation mode, improve regulation consistency, and avoid system oscillation or secondary interference caused by control structure switching.

[0105] The present invention further proposes that after the control structure is restored, the system response behavior before and after the recovery is used to retrospectively evaluate the gas path elastic parameters previously used for limiting regulation (including the maximum pressure difference response rate, maximum instantaneous flow rate and gas source recovery delay time), and these parameter values are automatically optimized and updated accordingly.

[0106] The optimization process is as follows: Extract key data before and after the control structure switching: including feedback response data within a certain period before and after the structure switching point, control signal change rate and error convergence time; Generate regulation performance scoring indicators: Comprehensively evaluate the following three indicators: recovery time (i.e., the time it takes from structural recovery to the feedback value entering the stable zone), control signal smoothness (i.e., whether the signal changes smoothly without sudden changes), and feedback convergence trend (i.e., whether the feedback value is stable and close to the target value); Establish a scoring and parameter mapping mechanism: If the recovery speed is fast, the control signal is stable, and the feedback trend is good, it means that the current parameters accurately describe the actual channel behavior; otherwise, it means that the response capability is overestimated or underestimated; Optimization parameter updates: Based on the scoring results, the values of the corresponding fields in the gas path elasticity parameters are appropriately adjusted. For example, if the channel is found to be still responsive after control is restored, the maximum differential pressure response rate parameter is reduced. If the recovery speed is significantly faster than predicted, the response capability value is increased and the delay time parameter is shortened.

[0107] This optimization process establishes a closed-loop self-correction mechanism for the system. As the controller runs longer, the gas path response parameters in the model evolve, becoming more aligned with the physical characteristics of the actual channel, thereby improving future control accuracy and regulation efficiency.

[0108] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0109] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0110] It should be understood that the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone, where A and B may be singular or plural. In addition, the character " / " herein generally indicates that the objects associated with each other are in an "or" relationship, but it may also indicate an "and / or" relationship, which can be understood by referring to the context. A person of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0111] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for controlling a gas chromatograph electronic gas circuit, characterized in that: include: Constructing a gas path elasticity parameter model, wherein the gas path elasticity parameters include the maximum instantaneous flow rate of the gas channel, the maximum pressure difference response rate, and the gas source recovery delay time; The gas path elasticity parameter is embedded in a PID controller structure to construct an integral regulation function, wherein the integral regulation function adjusts the accumulation rate of the integral term in real time according to the gas path elasticity parameter; Collect current gas state data and calculate the deviation between the actual feedback value and the target set value as the error input; Outputting a control signal according to the error input and the integral regulation function to drive a proportional solenoid valve to regulate gas flow or pressure; classifying the current control scenario based on the feedback value, control signal, and historical status data; When it is determined that the control scenario is an unstable scenario, the controller structure is automatically adjusted to a non-PID structure, and the update of the integral term is frozen or suppressed; After detecting that the system state has returned to stability, the PID control structure is restored, and the air path elasticity parameters are adaptively optimized and updated based on historical adjustment effects.

2. The method for controlling a gas chromatography electronic circuit according to claim 1, characterized in that: Constructing the gas path elastic parameter model includes: During the initial operation of the system or after each gas source change, a preload test is performed on each gas channel based on the set step pressure excitation signal; The maximum instantaneous flow rate is obtained by sampling the output signal of the flow sensor into a discrete time series, calculating the maximum difference between adjacent time points, and dividing the difference by the sampling interval. The obtained value is the maximum instantaneous flow rate estimate; The maximum pressure difference response rate is calculated using an exponential fitting method. The collected pressure change curve is fitted into an exponential growth model, and its derivative value at the maximum gradient is calculated by derivation. The derivative value corresponds to the maximum pressure difference response rate. The gas source recovery delay time is: the time when the first non-zero effective flow signal appears after the gas source is restored, and subtracted from the timestamp when the interruption occurs, and the resulting time difference is the recovery delay time.

3. The method for controlling a gas chromatography electronic circuit according to claim 1, characterized in that: Constructing the integral adjustment function includes: Set up a baseline structure for the integral regulation function, where the cumulative rate of the integral term is an adjustable parameter rather than a fixed proportional constant; The maximum instantaneous flow parameter of the target channel is extracted from the gas path elastic parameter model and used to set the initial gain upper limit of the integral regulation function. The integral gain upper limit is derived from the ratio between the maximum instantaneous flow and the control period. When the control system detects a trend of continuous error accumulation, it adjusts the integral term accumulation step size based on the maximum pressure difference response rate, automatically reducing the accumulation rate when the system's response capability is weak. A combined judgment of the error change direction and rate of change is performed before integral accumulation. If repeated or oscillating error directions are detected, and the maximum pressure difference response rate of the corresponding channel is less than a preset threshold, the integral adjustment function is immediately triggered to reduce the integral step size. The gas source recovery delay time is used to set the minimum delay threshold for integral accumulation, that is, the integral accumulation is frozen before the delay time is reached.

4. The method for controlling a gas chromatography electronic circuit according to claim 1, wherein: Calculating the deviation between the actual feedback value and the target set value as the error input includes: The actual state value of the gas in the current channel is collected at fixed time intervals, and the sensor output is denoised using a sliding time window. The denoised current state value is compared with the target set value of the corresponding channel to obtain a real-time deviation value, which is used as the error input of the controller.

5. The method for controlling a gas chromatography electronic circuit according to claim 1, wherein: Driving the proportional solenoid valve to regulate gas flow or pressure includes: The error input is combined with the integral regulation function output to form a set of multi-dimensional control instructions, including the basic signal value for adjusting the proportional valve voltage, current or duty cycle; Determine the current operating state of the channel. If the control signal change amplitude exceeds the preset threshold within a continuous cycle, the signal change amplitude is compressed; When the gas source pressure fluctuates violently, the pressure differential response rate parameter is called to impose a dynamic upper limit constraint on the control signal. The dynamic upper limit constraint is determined based on the maximum pressure differential response rate of the channel. If the response rate is lower than the system's preset critical value, the control signal change rate is automatically limited to a range that matches the physical response capability of the channel.

6. The method for controlling a gas chromatography electronic circuit according to claim 1, characterized in that: Classification of current control scenarios includes: Collect the feedback value and control signal output value of the current cycle and construct a set of historical state sequences together with the data of the previous cycles; Calculate the response consistency index between the feedback value and the control signal to reflect whether the control action causes the expected pressure or flow change. The response consistency index compares the direction of change of the control signal and the direction of change of the feedback value to see if they are consistent. If the directions are consistent but the amplitude ratio is continuously low, it indicates a low response rate state. If the directions are opposite, it is marked as a reverse feedback anomaly. Based on the consistency index and the feedback trend curve, the current gas circuit operation state is predicted. If the control signal is continuously applied but the feedback value does not change significantly, the state is marked as a response hysteresis scenario. If the feedback value changes suddenly without any obvious change in the control signal, and the change exceeds the maximum pressure difference response capability of the current channel, the state is marked as an external disturbance scenario.

7. The method for controlling a gas chromatography electronic circuit according to claim 1, characterized in that: Automatically tuning the controller structure to a non-PID structure includes: Based on the response relationship between the feedback value and the control signal, identify the current control scenario as any unstable scenario such as response hysteresis, reverse feedback or external disturbance; When the system is in this scenario state for multiple consecutive control cycles, and the error direction fluctuates frequently or the control output amplitude is abnormally amplified, the control structure switching logic is triggered; Temporarily disconnect the integral path in the original PID structure, retain only the proportional term and the differential term to participate in the control decision, and construct a simplified PD control structure; After the control structure is switched, the integral term accumulator state is frozen. The integral term freeze adopts a two-layer freezing mechanism. The first layer is periodic freezing, which prohibits integral updates within a fixed time period. The second layer is dynamic freezing, which automatically determines whether the freezing period should be extended or released in advance based on the system feedback recovery speed and pressure difference response capability.

8. The method for controlling a gas chromatography electronic circuit according to claim 1, characterized in that: Restoring the PID control structure after the detection system state returns to stability includes: Continuously monitor the rate and direction of change of the feedback value during the freezing period to determine whether it is stable and close to the target value, and calculate whether the range of change is continuously within the set stability threshold; The time period response trend analysis method is used. That is, if the rate of change of the feedback value gradually decreases and the error direction does not reverse within multiple consecutive control cycles, it is judged that the control has entered a stable state and the PID structure recovery process is allowed to begin. After recovery, the PID control structure reactivates the joint regulation mode of the integral term, proportional term, and differential term, and limits the integral output amplitude in the first two cycles; The feedback response difference before and after the control structure is restored is recorded to generate a regulation performance score for elastic parameter optimization, which is used to dynamically adjust the maximum pressure difference response rate and recovery delay time in the gas path elastic parameters.

Images