A GIS mixed gas supplementing self-adaptive control method, system, device and medium

Abstract

The application discloses a GIS mixed gas supplementing self-adaptive control method, system, equipment and medium, relates to the GIS supplementing control technical field, and realizes the deep perception and early warning of the gas state through the multi-source data fusion trend prediction, and changes the passive supplementing to the active prevention. The self-adaptive control method adopts the fuzzy PID control algorithm, finely adjusts the controller parameters according to the deviation, the deviation change rate and the equipment load current, utilizes the predicted gas leakage rate as a feedforward signal to compensate the leakage amount, finally makes the pressure stable at the dynamic pressure set point, and the gas proportion is always kept constant. The whole process is smooth and impact-free, can adapt to different leakage rates, environmental conditions and equipment load changes, and has high robustness. In addition, the self-adaptive control method also adopts the matching strategy based on the mass flow control, ensures that the pressure and proportion of the supplemented gas reach the optimal set values.

Description

Technical Field

[0001] This application relates to the field of GIS gas replenishment control technology, specifically to an adaptive control method, system, device, and medium for GIS mixed gas replenishment. Background Technology

[0002] Gas-insulated switchgear (GIS), which uses a mixture of gases such as SF6 / N2 as the insulating and arc-extinguishing medium, is increasingly used in power systems. During the manufacturing, installation, commissioning, and operation of GIS, defects such as pinholes in the casing, poor-quality seals, improper on-site sealing surface treatment, and aging of sealing materials after long-term operation can all lead to leaks in the GIS gas chamber, resulting in reduced chamber pressure and abnormal gas pressure. If not addressed promptly, the continued pressure drop will cause a sharp decline in the equipment's arc-extinguishing capability, easily leading to breakdown accidents and power outages. Therefore, it is necessary to use a gas replenishment device to replenish the gas chamber under energized conditions and to implement leak prevention measures. However, existing gas replenishment technologies have the following limitations:

[0003] (1) Lag: It is impossible to provide early warning and intervention in the early stage when gas pressure is slowly leaking or slightly imbalanced, which poses a great safety hazard;

[0004] (2) Insufficient accuracy: Some automatic gas replenishment devices adopt a simple control logic of "starting at low pressure threshold and stopping at high pressure threshold". Manual gas replenishment or such switch control is difficult to accurately control the total amount of gas replenished and the mixing ratio of each component, which may cause the gas ratio after replenishment to deviate from the optimal design value (such as a decrease in the proportion of SF6), directly affecting the insulation and arc extinguishing performance of the equipment.

[0005] (3) Most existing solutions rely only on a single absolute pressure signal and do not comprehensively consider the complex influence of operating state variables such as temperature and load current on gas density and pressure, and cannot achieve optimized control decisions based on the actual operating conditions of the equipment (such as high load heating).

[0006] (4) Existing automatic control logic (such as fixed-parameter PID or switching control) cannot adaptively adjust the control strategy according to the dynamic changes in leakage rate, the diurnal and seasonal fluctuations in ambient temperature, and the changes in equipment load current. When faced with changing operating conditions, problems such as control oscillation, excessive gas supply, or sluggish response are likely to occur. Summary of the Invention

[0007] To overcome the shortcomings of existing gas replenishment control technologies, this application proposes an adaptive control method, system, device, and medium for GIS mixed gas replenishment. This method enables early warning, precise proportioning, and smooth gas replenishment under GIS energized operation by using multi-source data fusion sensing, leakage trend prediction, and adaptive composite control algorithms.

[0008] This application is achieved through the following technical solution:

[0009] An adaptive control method for GIS mixed gas replenishment includes:

[0010] The system simultaneously acquires and preprocesses multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes gas pressure, gas temperature, mixed gas ratio, and equipment load current.

[0011] Based on the gas law, the gas pressure is compensated by the gas temperature to obtain the normalized pressure at the standard temperature.

[0012] Based on historical normalized pressure time series data, the future pressure change trend is calculated through a time prediction model, and the gas leakage rate is estimated.

[0013] A tiered early warning system is implemented based on the predicted normalized pressure and gas leakage rate: when the predicted normalized pressure and gas leakage rate are both within the normal range or the deviation from the normal range is less than the threshold, the subsequent steps continue; when the predicted normalized pressure or gas leakage rate deviates from its normal range by more than the threshold, the system first enters a safety restriction mode, and the subsequent steps continue only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal.

[0014] The pressure control setpoint is dynamically generated based on the target pressure and the gas leakage rate.

[0015] The deviation between the predicted normalized pressure and the pressure control setpoint, the rate of change of the deviation, and the equipment load current are used as inputs. The parameters of the controller are adjusted online adaptively through a fuzzy inference mechanism, and the first control quantity is calculated.

[0016] Based on the gas leakage rate and combined with the feedforward control gain, the feedforward compensation control quantity is calculated.

[0017] The first control quantity and the feedforward compensation control quantity are superimposed and limited to generate a total control quantity, which is then converted into a corresponding control signal and sent to the execution unit to achieve precise and continuous adjustment of the supplementary gas flow rate and the ratio of the mixed gas.

[0018] In some embodiments, the calculation of gas pressure compensation using the gas temperature includes:

[0019] Convert the gas temperature into thermodynamic temperature;

[0020] Based on the aforementioned thermodynamic temperature, the compressibility factor under the current state and the compressibility factor at the standard temperature are calculated using a method based on a simplified form of the modified Benedict-Wade-Rubin equation of state.

[0021] Calculate the normalized pressure based on the compressibility factor under the current conditions and the compressibility factor at standard temperature:

[0022] P_norm = P_m * (293.15 / T_k) * (Z2 / Z1);

[0023] Where P_norm is the normalized pressure, P_m is the gas pressure, T_k is the thermodynamic temperature, and Z1 and Z2 are the compressibility factors under the current state and the compressibility factors under the standard temperature, respectively.

[0024] In some embodiments, calculating the compressibility factor under the current state and the compressibility factor at the standard temperature includes:

[0025] Calculate the compression factor Z1 in the current state: Z1 = 1 + (a1 + a2 / T_k) * P_m + (a3 + a4 / T_k) * P_m^2;

[0026] Calculate the initial estimated normalized pressure P_est: P_est = P_m * 293.15 / T_k;

[0027] Calculate the compressibility factor Z2 at standard temperature: Z2 = 1 + (a1 + a2 / 293.15) * P_est + (a3 + a4 / 293.15) * P_est^2;

[0028] Among them, a1, a2, a3 and a4 are coefficients related to the proportion of the mixed gas.

[0029] In some implementations, the calculation of future pressure change trends and estimation of gas leakage rates using a time prediction model includes:

[0030] Acquire historical normalized pressure time series data and set a sliding analysis window of a preset length;

[0031] A second-order autoregressive model is used to describe the dynamic change process of pressure, and the model parameters are estimated online in real time using the recursive least squares method with a forgetting factor.

[0032] In each control cycle, the normalized pressure prediction sequence for a future period is iteratively calculated using the latest estimated parameters and the current sliding analysis window data.

[0033] Extract the normalized pressure values ​​of the last segment of the normalized pressure prediction sequence, and perform linear least squares fitting based on the extracted normalized pressure values ​​and their corresponding time indices to obtain the slope of the fitted line.

[0034] The gas leakage rate V_leak is calculated based on the slope of the fitted line: V_leak = - (k_slope / Δt) * 3600; where k_slope is the slope of the fitted line and Δt is the sampling period.

[0035] The new normalized pressure data is used to update the sliding analysis window to prepare for model parameter estimation in the next cycle.

[0036] In some implementations, the online adaptive adjustment of the controller parameters via fuzzy inference mechanism and the calculation of the first control quantity include:

[0037] The deviation and rate of change of the normalized pressure obtained from the current input prediction from the pressure control setpoint and the equipment load rate are fuzzified, and the corresponding fuzzy rules are matched from the pre-built fuzzy rule library according to the fuzzy results to obtain the corresponding output fuzzy set; wherein the output fuzzy set is the controller parameter fuzzy set.

[0038] The output fuzzy set is synthesized by MAX, and the fuzziness is defuzzified by the centroid method to obtain the precise controller parameter adjustment amount;

[0039] Update the controller parameters according to the adjustment amount of the controller parameters;

[0040] Based on the updated controller parameters, the first control quantity is calculated using an incremental PID algorithm.

[0041] In some implementations, converting the total control quantity into a corresponding control signal includes:

[0042] The total control quantity is converted into the total make-up gas flow rate under standard conditions by looking up a table.

[0043] Based on the proportion of the mixed gas, calculate the flow rate setpoint for each component and input the flow rate setpoint for each component into the corresponding mass flow controller.

[0044] In some embodiments, converting the total control quantity into a corresponding control signal further includes:

[0045] When an online gas proportioning analyzer is configured to monitor the volume concentration of sulfur hexafluoride in the mixed gas in real time, an outer loop correction algorithm is executed: the actual concentration of sulfur hexafluoride measured by the online gas proportioning analyzer is read, and the proportioning error is calculated; the flow correction amount is calculated through the PI controller; and the flow correction amount is used to correct the flow setpoint of each component.

[0046] And / or, when the total replenishment flow rate decreases according to control requirements, the flow rate setpoints of each component are updated proportionally, while the update process meets the constraint that the change in the flow rate setpoint per second does not exceed 20% of the full scale.

[0047] Secondly, this application proposes a GIS mixed gas replenishment adaptive control system, comprising:

[0048] The preprocessing unit is configured to: synchronously acquire and preprocess multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes gas pressure, gas temperature, mixed gas ratio, and equipment load current;

[0049] The compensation unit is configured to: perform compensation calculations on the gas pressure based on the gas state equation, using the gas temperature to obtain the normalized pressure at the standard temperature;

[0050] The trend prediction unit is configured to: calculate the future pressure change trend and estimate the gas leakage rate based on historical normalized pressure time series data using a time prediction model;

[0051] The graded early warning unit is configured to: perform decomposed early warning based on the normalized pressure and gas leakage rate predicted by the trend prediction unit; when the predicted normalized pressure and gas leakage rate are both within the normal range or the deviation from the normal range is less than the threshold, the adaptive decision unit is activated; when the predicted normalized pressure or gas leakage rate deviates from its normal range by a value greater than the threshold, the unit first enters the safety restriction mode, and the adaptive decision unit is activated only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal.

[0052] Furthermore, the adaptive decision unit is configured to: dynamically generate a pressure control setpoint based on the target pressure and the gas leakage rate; use the deviation and rate of change of the predicted normalized pressure from the pressure control setpoint, as well as the equipment load current, as inputs to adaptively adjust the controller parameters online through a fuzzy inference mechanism and calculate a first control quantity; calculate a feedforward compensation control quantity based on the gas leakage rate and combined with the feedforward control gain; superimpose and limit the first control quantity and the feedforward compensation control quantity to generate a total control quantity, and convert the total control quantity into a corresponding control signal and send it to the execution unit to achieve precise and continuous adjustment of the supplementary gas flow rate and the mixed gas ratio.

[0053] Thirdly, this application proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the above-described embodiments of the GIS mixed gas replenishment adaptive control method.

[0054] Fourthly, this application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described embodiments of the GIS mixed gas replenishment adaptive control method.

[0055] This application proposes an adaptive control method for GIS mixed gas replenishment. By fusing multi-source data and predicting trends, it achieves deep perception and early warning of gas status, transforming passive gas replenishment into proactive prevention. This adaptive control method employs a fuzzy PID control algorithm, which finely adjusts controller parameters based on deviation, deviation change rate, and equipment load current. It also uses the predicted gas leakage rate as a feedforward signal to compensate for leakage, ultimately stabilizing the pressure at the dynamic pressure setpoint while maintaining a constant gas ratio. The entire process is smooth and shock-free, capable of adapting to different leakage rates, environmental conditions, and equipment load changes, exhibiting high robustness. Furthermore, this adaptive control method also employs a proportioning strategy based on mass flow control to ensure that the pressure and ratio of the replenished gas reach the optimal setpoint.

[0056] Accordingly, the GIS mixed gas replenishment adaptive control system, electronic equipment, and computer-readable storage medium proposed in this application have the same technical effects as described above. Attached Figure Description

[0057] The accompanying drawings, which are included to provide a further understanding of the embodiments of this application and form part of this application, do not constitute a limitation on the embodiments of this application. In the drawings:

[0058] Figure 1 This is a flowchart of the adaptive control method proposed in the embodiments of this application;

[0059] Figure 2 This is a block diagram illustrating the principle of the adaptive control system proposed in an embodiment of this application.

[0060] Figure 3 This is a schematic diagram of the electronic device proposed in the embodiments of this application;

[0061] Figure 4 This is a schematic diagram of a computer-readable storage medium proposed in an embodiment of this application;

[0062] Figure reference numerals and corresponding component names:

[0063] 200 - Adaptive control system; 210 - Air replenishment control module; 211 - Preprocessing unit; 212 - Compensation unit; 213 - Trend prediction unit; 214 - Hierarchical early warning unit; 215 - Adaptive decision-making unit; 220 - Sensor module; 230 - Air replenishment execution module; 240 - Remote monitoring module; 300 - Electronic equipment; 310 - Memory; 320 - Processor; 311 - Computer program A; 400 - Computer-readable storage medium; 411 - Computer program B. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0065] like Figure 1 As shown in the embodiment of this application, an adaptive control method for GIS mixed gas replenishment is proposed. The adaptive control method includes the following steps:

[0066] Step 1: Simultaneously acquire and preprocess multi-source sensor data from the GIS gas chamber. This multi-source sensor data includes at least gas pressure, gas temperature, mixed gas ratio, and equipment load current. It should be noted that the preprocessing in Step 1 involves standard filtering, calibration, and outlier removal processes, which will not be elaborated upon here.

[0067] Step 2: Based on the gas law, the gas pressure is compensated by the gas temperature to obtain the normalized pressure at the standard temperature.

[0068] Furthermore, the specific implementation process of step 2 includes:

[0069] Obtain the current gas pressure P_m, gas temperature T_g, and gas mixture ratio; the gas mixture ratio includes at least two ratio values. For SF6 / N2 gas mixture, the gas mixture ratio includes the proportion of SF6 (sulfur hexafluoride) x_sf6 and the proportion of N2 (nitrogen) x_n2, and the sum of the two is 1.

[0070] Convert the gas temperature to a thermodynamic temperature: T_k = T_g + 273.15; where T_g is the gas temperature, T_k is the thermodynamic temperature, and 273.15 is the thermodynamic temperature corresponding to 0℃.

[0071] A compressibility factor calculation method based on a modified Benedict-Webb-Rubin (MBWR) simplified equation of state is adopted to balance computational accuracy and embedded system resource consumption: The compressibility factor Z1 under the current state is calculated as follows: Z1 = 1 + (a1 + a2 / T_k) * P_m + (a3 + a4 / T_k) * P_m^2; where a1, a2, a3, and a4 are coefficients related to the proportion of the mixed gas; for the SF6 / N2 mixture, a1 = -0.0012, a2 ​​= 15.6, a3 = 3.4e-6, and a4 = -0.021 can be taken. The initially estimated normalized pressure P_est is calculated as follows: P_est = P_m * 293.15 / T_k; The compressibility factor Z2 at standard temperature is calculated as follows: Z2 = 1 + (a1 + a2 / 293.15) * P_est + (a3 + a4 / T_k) * P_m^2; a4 / 293.15) * P_est^2;

[0072] Calculate the final normalized pressure P_norm (normalized pressure at standard temperature 20℃): P_norm = P_m *(293.15 / T_k) * (Z2 / Z1), where 293.15 is the thermodynamic temperature corresponding to standard temperature 20℃;

[0073] The normalized pressure output is made available for use by subsequent modules and stored in the historical data queue for trend prediction.

[0074] Step 3: Based on historical normalized pressure time series data, calculate the future pressure change trend and estimate the gas leakage rate using a time prediction model.

[0075] Furthermore, the specific implementation process of step 3 includes:

[0076] Data preparation: Continuously acquire temperature-compensated normalized pressure (i.e., normalized pressure at standard temperature). The sampling period can be set to 5 seconds. In this embodiment, a first-in-first-out data queue P_window with a length of N=360 (covering 30 minutes) can be used as a sliding analysis window.

[0077] Online Modeling and Parameter Estimation: A second-order autoregressive model is used to describe the dynamic pressure change process: P_norm(t) = φ1·P_norm(t-1) + φ2·P_norm(t-2) + ε; where φ1 and φ2 represent the autoregressive coefficients, which are parameters that need to be estimated online; ε represents white noise, representing fluctuations that the model cannot explain; P_norm(t), P_norm(t-1), and P_norm(t-2) represent the normalized pressure at the current time t, historical time t-1, and historical time t-2, respectively. The model parameters (i.e., the autoregressive coefficients) are estimated online in real time using recursive least squares with a forgetting factor. This eliminates the need for pre-training and allows the model to adapt to the characteristics of different GIS gas chambers and changing leakage conditions. The forgetting factor is set to 0.995, enabling the model to continuously track slowly changing leakage trends.

[0078] Online real-time assessment:

[0079] S1: Initialization. The parameter estimation vector is set during system startup. covariance matrix Forgetting factor ,in, Represents the transpose matrix; Represents the identity matrix.

[0080] S2: Data Preparation. In each sampling period, obtain the current normalization pressure. And read the values ​​from the previous two moments from the history queue. and Constructing regression vectors .

[0081] S3: Error calculation. Based on the parameter estimation vector from the previous time step. Calculate the predicted value Thus, the prediction error is obtained. .

[0082] S4: Gain Calculation. Calculate the gain vector. :

[0083] ;

[0084] in, Let represent the covariance matrix of the previous time step.

[0085] S5: Parameter update. Update autoregressive coefficients:

[0086] ;

[0087] in, This represents the parameter estimation vector at the current moment.

[0088] S6: Covariance Update. Update the covariance matrix to prepare for the next time step:

[0089] ;

[0090] in, This represents the covariance matrix at the current time.

[0091] S7: Output and Iteration. The two estimated autoregressive coefficients are output to the multi-step pressure prediction section to calculate future pressure trends and leakage rates. Wait for the next sampling point and repeat S2-S7.

[0092] Through the above online recursive estimation, the model parameters can track the slow changes in the leakage characteristics of GIS gas chambers in real time without offline training, and achieve the ability to adapt to different gas chambers and different operating conditions.

[0093] Multi-step stress prediction: In each control cycle, using the latest estimated parameters and current window data, the normalized stress prediction sequence P_pred[1] to P_pred[M] for a future time step M (e.g., M=720 time steps, covering 1 hour) is iteratively calculated;

[0094] Leakage rate calculation: Extract the normalized pressure value P_tail = [P_pred[M-L+1], ..., P_pred[M]] from the last time step L (e.g., L=12 time steps) in the prediction sequence;

[0095] Linear least squares fitting is performed based on the normalized pressure values ​​of the last time step extracted and their corresponding time indices to obtain the slope k_slope of the fitted line.

[0096] The gas leakage rate V_leak is calculated based on the slope of the fitted straight line: V_leak = - (k_slope / Δt) * 3600; where Δt is the sampling period; and the negative sign indicates a pressure drop. The gas leakage rate characterizes the instantaneous rate of pressure decrease in the near future, predicted based on current and historical conditions.

[0097] The P_window queue is updated with the latest P_norm value to prepare for the recursive parameter estimation in the next cycle.

[0098] Step 4: Based on the predicted normalized pressure and gas leakage rate, a graded early warning system is implemented: When both the normalized pressure and gas leakage rate are within the normal range or slightly deviate from the normal range (i.e., the deviation is less than the threshold), full-function adaptive control is performed (i.e., steps 5-8); when the normalized pressure or gas leakage rate deviates abnormally (i.e., the deviation is greater than the threshold), a safety restriction mode is automatically entered. Full-function adaptive control is then implemented after maintenance personnel eliminate safety hazards and confirm that normalcy has been restored (i.e., steps 5-8). Specifically, when an abnormal deviation is detected, the safety restriction mode is automatically entered. The specific implementation process includes: reducing the controller output limit to a safe range; limiting the flow rate change rate per second; pausing adaptive parameter updates; recording the abnormal state and issuing an alarm. The confirmation of normalcy is achieved through at least one of the following methods (a)-(b): (a) the normalized pressure and leakage rate are detected to return to the normal range for multiple consecutive cycles; (b) maintenance personnel confirm no leakage or fault through on-site or remote inspection; (c) the safety restriction is manually lifted on the monitoring interface or maintenance terminal, restoring full-function adaptive control.

[0099] Step 5: Dynamically generate the pressure control setpoint based on the target pressure and gas leakage rate;

[0100] Furthermore, in step 5 of this paper, the process of dynamically generating the pressure control setpoint is as follows:

[0101] S_dynamic = P_target + K * V_leak; where S_dynamic is the pressure control setpoint, P_target is the target pressure at standard temperature, V_leak is the estimated gas leakage rate, and K is a coefficient related to the look-ahead time of the gas replenishment response.

[0102] Step 6: Using the predicted deviation between the normalized pressure and the pressure control setpoint, the rate of change of the deviation, and the equipment load current as inputs, the parameters of the controller are adjusted online adaptively through a fuzzy inference mechanism, and the first control quantity is calculated.

[0103] Furthermore, in step 6, the equipment load current is introduced as an environmental factor. When the equipment load current exceeds a preset threshold, the fuzzy rule base output tends to make the controller's parameter adjustments more conservative, thereby reducing the impact of control actions on high-load operating equipment. The specific implementation process includes:

[0104] Controller input definition:

[0105] The predicted deviation e between the normalized pressure and the pressure control setpoint ranges from -0.1 to 0.1 (which can be adjusted according to the actual GIS pressure range). The fuzzy subset is: {Negative Large (NB), Negative Medium (NM), Negative Small (NS), Zero (ZO), Positive Small (PS), Positive Medium (PM), Positive Large (PB)}, a total of 7.

[0106] The predicted deviation rate ec between the normalized pressure and the pressure control setpoint ranges from -0.01 to 0.01, and the fuzzy subset is: {Negative Large (NB), Negative Medium (NM), Negative Small (NS), Zero (ZO), Positive Small (PS), Positive Medium (PM), Positive Large (PB)}, a total of 7.

[0107] For the equipment load current, this embodiment of the application adopts the normalized load current I_norm = (I_load - I_min) / (I_max - I_min); where I_min and I_max are the minimum and maximum currents during normal operation of the equipment; I_load is the real-time collected equipment load current; the fuzzy subset is: {Low, Medium, High}, a total of 3;

[0108] Each input is fuzzified using a predefined triangular membership function to obtain the membership degree of each fuzzy subset.

[0109] Controller parameter adjustment definition: In this embodiment, a PID controller is used. Therefore, the controller parameter adjustment includes proportional adjustment ΔKp, integral adjustment ΔKi, and derivative adjustment ΔKd, all within the range of [-0.5, 0.5], representing the adjustment ratio relative to the initial value. The fuzzy subset is: {Negative Large (NB), Negative Medium (NM), Negative Small (NS), Zero (ZO), Positive Small (PS), Positive Medium (PM), Positive Large (PB)}, a total of 7. The actual parameter update formula is expressed as:

[0110] Kp_new = Kp_initial × (1 + ΔKp);

[0111] Ki_new = Ki_initial × (1 + ΔKi);

[0112] Kd_new = Kd_initial × (1 + ΔKd);

[0113] Where Kp_new, Ki_new, and Kd_new are the proportional, integral, and derivative parameters of the updated controller, respectively, and Kp_initial, Ki_initial, and Kd_initial are the proportional, integral, and derivative parameters of the original controller, respectively. A fuzzy rule base is constructed based on the fuzzy subsets of the controller input and the fuzzy subsets of the controller parameters.

[0114] Fuzzy inference mechanism: Based on the input, it matches the corresponding fuzzy rules from the fuzzy rule base to obtain the corresponding output fuzzy set. The system has a pre-built fuzzy rule base containing 7*7*3 rules, for example:

[0115] If e is PB AND ec is ZO AND I_norm is Low, then ΔKp is PB, ΔKi is NB, and ΔKd is PS. This means that if the input deviation e is PB, the deviation change rate ec is ZO, and the normalized load current I_norm is Low, then ΔKp takes PB, ΔKi takes NB, and ΔKd takes PS.

[0116] Defuzzing to obtain parameter adjustment amounts: Maximize the output fuzzy sets of all activation rules, and defuzzify using the centroid method to obtain accurate ΔKp, ΔKi, and ΔKd.

[0117] It is understandable that each fuzzy rule generates an output fuzzy set after inference. For example, if ΔKp is positive, this fuzzy set is described by a membership function. When multiple rules are activated simultaneously, multiple output fuzzy sets are generated. In this embodiment, MAX synthesis is used to merge the fuzzy sets output by all activated rules to obtain a comprehensive fuzzy output set. MAX synthesis uses the maximum value method: that is, for each point in the output universe (i.e., each possible ΔKp value), the maximum value of the membership degrees of all rules at that point is taken as the membership degree of the synthesized fuzzy set at that point. This is expressed as:

[0118] μ combined (x)=max(μ R1 (x),μ R2 (x),…,μ Rn (x)); where μ Ri (x) is the membership degree of the fuzzy set output by the i-th rule (i=1,2,⋯,n) at x, and n represents the number of activation rules.

[0119] The centroid method for fuzzy resolution aims to calculate specific parameter adjustments from the synthesized fuzzy output set, serving as the actual input to the controller. Its principle involves calculating the abscissa of the centroid (centroid) of the region enclosed by the membership function curve of the synthesized fuzzy output set and the horizontal axis (sampling coordinates). On the continuous universe of discourse, this is expressed as:

[0120]

[0121] In practical discretization calculations, the universe of discourse is usually discretized into several points and approximated by a weighted average:

[0122]

[0123] in, For discrete sampling points, The membership degree of the comprehensive fuzzy output set at this point.

[0124] Based on the parameter adjustment amount obtained from defuzzification, the controller parameters are updated, and the new parameters are calculated based on the initial parameters. The calculation formula adopts the parameter update formula mentioned above.

[0125] The first control quantity u_pid is calculated using an incremental PID algorithm:

[0126] Δu=Kp_new×(e(k)-e(k-1)) +Ki_new×0.1×e(k) + (Kd_new / 0.1)×[e(k)-2e(k-1)+e(k-2)];

[0127] u_pid(k) = u_pid(k-1) + Δu;

[0128] Where Δu is the increment; e(k) is the deviation at the current time k, that is, the difference between the predicted normalized pressure and the pressure control setpoint; e(k-1) is the deviation at the previous time k-1; e(k-2) is the deviation between the two previous times k-2; and u_pid(k-1) is the first control variable at the previous time k-1.

[0129] Step 7: Based on the gas leakage rate and combined with the feedforward control gain, calculate the feedforward compensation control quantity.

[0130] Furthermore, in step 7, the feedforward compensation control quantity u_ff = K_ff * V_leak, where K_ff represents the feedforward gain.

[0131] Step 8: The first control quantity and the feedforward compensation control quantity are superimposed and limited to generate a total control quantity. The total control quantity is then converted into an analog quantity or PWM control signal to drive the proportional regulating valve in the gas injection unit, so as to achieve precise and continuous adjustment of the gas injection flow rate and the ratio of the mixed gas.

[0132] Furthermore, in step 8, the specific implementation process for the precise and continuous adjustment of the supplementary gas flow rate and the ratio of the mixed gas includes:

[0133] Basic flow control: The first control quantity and the feedforward compensation control quantity are superimposed and limited to generate a total control quantity. This total control quantity is then converted into the total make-up gas flow rate under standard conditions via a lookup table. Based on the target gas mixture ratio, the setpoints for the flow rates of each component are calculated and input to the corresponding mass flow controllers. For example, for SF6 and N2, the actuator contains two independent gas path controllers: a mass flow controller MFC1 for the SF6 gas path and a mass flow controller MFC2 for the N2 gas path. Based on the target SF6 and N2 ratio, the setpoints for the flow rates of each component are calculated. MFC1 and MFC2 receive their respective setpoints and run their built-in closed-loop control algorithms to ensure that the actual flow rate quickly and accurately reaches the setpoint.

[0134] Optionally, an ultrasonic gas proportioning analyzer is installed at the mixer outlet to monitor the volume concentration of SF6 in the mixed gas in real time. Step 8 also includes online proportioning feedback correction: when an online gas proportioning analyzer is configured, an outer-loop correction algorithm can be executed with a cycle of 1 second: the actual SF6 concentration measured by the analyzer is read; the proportioning error is calculated; and the flow correction amount is calculated via the PI controller. The setpoints of the two MFCs are corrected to maintain a constant total flow rate. The corrected setpoints are then sent to MFC1 and MFC2.

[0135] Optionally, step 8 also includes dynamic process ratio maintenance: when the total replenishment flow rate decreases according to control requirements, the set value is updated proportionally, while meeting the constraint that the change in the flow rate set value per second does not exceed 20% of the full scale, ensuring a smooth transition.

[0136] It is understood that the adaptive control method proposed in this application uses a mixture of SF6 and N2 gas as an example for illustrative purposes, but it does not limit the mixed gas in the GIS. It can be adaptively applied to different proportions or types of insulating gases.

[0137] Based on the same technical concept described above, this application also proposes a GIS mixed gas replenishment adaptive control system, such as... Figure 2 As shown, the adaptive control system 200 includes: a gas replenishment control module 210, which includes:

[0138] The preprocessing unit 211 is configured to: synchronously acquire and preprocess multi-source sensor data of the GIS gas chamber; the multi-source sensor data includes at least gas pressure, gas temperature, mixed gas ratio and equipment load current.

[0139] The compensation unit 212 is configured to perform gas pressure compensation calculations based on the gas state equation, using gas temperature to obtain the normalized pressure at the standard temperature. The specific compensation process is as described in step 2 above, and will not be repeated here.

[0140] The trend prediction unit 213 is configured to calculate the future pressure change trend and estimate the gas leakage rate based on historical normalized pressure time-series data using a time prediction model. The specific pressure change trend prediction and gas leakage rate estimation process is as described in step 3 above, and will not be repeated here.

[0141] The graded early warning unit 214 is configured to provide graded early warnings based on the normalized pressure and gas leakage rate predicted by the trend prediction unit 213: when both the normalized pressure and gas leakage rate are within the normal range or slightly deviate from the normal range (i.e., the deviation is less than the threshold), the adaptive decision unit 215 is activated; when the normalized pressure or gas leakage rate deviates abnormally (i.e., the deviation is greater than the threshold), a safety restriction mode is first entered, and the adaptive decision unit 215 is activated only after the maintenance personnel have eliminated the safety hazards and confirmed that normalcy has been restored. The specific safety restriction mode and the method for confirming normalcy are as described in step 4 above, and will not be repeated here.

[0142] Furthermore, the adaptive decision unit 215 is configured to: dynamically generate a pressure control setpoint based on the target pressure and gas leakage rate; use the deviation and rate of change of the predicted normalized pressure from the pressure control setpoint, as well as the equipment load current, as inputs, and adaptively adjust the controller parameters online through a fuzzy inference mechanism to calculate the first control quantity; calculate the feedforward compensation control quantity based on the gas leakage rate and combined with the feedforward control gain; superimpose and limit the first control quantity and the feedforward compensation control quantity to generate a total control quantity, and convert the total control quantity into an analog quantity or PWM control signal to drive the proportional regulating valve in the gas replenishment execution unit, so as to achieve precise and continuous adjustment of the gas replenishment flow rate and the ratio of the mixed gas. The specific adaptive control process is as described in steps 5 to 8 above, and will not be repeated here.

[0143] Furthermore, the adaptive control system 200 proposed in this application embodiment also includes:

[0144] Sensor module 220 is used to collect physical state data of the GIS gas chamber in real time, i.e., multi-source sensor data. Specifically, sensor module 220 includes, but is not limited to: a gas pressure sensor for real-time measurement of absolute gas pressure inside the GIS gas chamber; a gas temperature sensor for real-time measurement of gas temperature inside the GIS gas chamber; a gas proportioning sensor (such as ultrasonic or thermal conductivity type) for real-time measurement of the volume ratio or density ratio of the mixed gas inside the GIS gas chamber; and an equipment load current sensor that obtains the GIS bus current through a CT (current transformer) to reflect the equipment operating load.

[0145] Furthermore, the adaptive control system 200 proposed in this application embodiment also includes:

[0146] The gas replenishment execution module 230 is used to receive control signals from the gas replenishment control module 210 and precisely execute gas replenishment and mixing operations. Further, the gas replenishment execution module 230 includes: at least two independent high-pressure gas source storage tanks, respectively storing SF6 gas and buffer gas (e.g., N2); a mass flow controller or high-precision proportional control valve corresponding to each gas source; and a gas mixing chamber for receiving and proportionally mixing gases from each gas path; wherein the mass flow controller or proportional control valve is controlled by the control signal output by the gas replenishment control module 210.

[0147] Furthermore, the adaptive control system 200 proposed in this application embodiment also includes:

[0148] The remote monitoring module 240 receives all raw data, status assessment results, control command logs, alarm information, etc. uploaded by the gas supply control module 210 through a communication module (such as 4G / 5G, fiber optic Ethernet) for data visualization, historical analysis and remote parameter management.

[0149] Based on the same technical concept described above, this application also proposes an electronic device, such as... Figure 3 As shown, the electronic device 300 includes: a memory 310, a processor 320, and a computer program A311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program A311, it performs the following steps:

[0150] Simultaneously acquire and preprocess multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes at least gas pressure, gas temperature, mixed gas ratio, and equipment load current;

[0151] Based on the gas law, the gas pressure is compensated by the gas temperature to obtain the normalized pressure at the standard temperature.

[0152] Based on historical normalized pressure time series data, the future pressure change trend is calculated through a time prediction model, and the gas leakage rate is estimated.

[0153] A tiered early warning system is implemented based on the predicted normalized pressure and gas leakage rate: when both the normalized pressure and gas leakage rate are within the normal range or slightly deviate from the normal range (i.e., the deviation is less than the threshold), the subsequent steps continue; when the normalized pressure or gas leakage rate deviates abnormally (i.e., the deviation is greater than the threshold), the system automatically enters a safety restriction mode, and the subsequent steps continue only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal.

[0154] The pressure control setpoint is dynamically generated based on the target pressure and the gas leakage rate.

[0155] The predicted normalized pressure is used as input to the deviation and rate of change of the pressure control setpoint, as well as the equipment load current. The parameters of the controller are adjusted online adaptively through a fuzzy inference mechanism, and the first control quantity is calculated.

[0156] Based on the gas leakage rate and combined with the feedforward control gain, the feedforward compensation control quantity is calculated.

[0157] The first control quantity is superimposed and limited with the feedforward compensation control quantity to generate the total control quantity. The total control quantity is then converted into an analog quantity or PWM control signal to drive the proportional regulating valve in the gas injection unit, so as to achieve precise and continuous adjustment of the gas injection flow rate and the ratio of the mixed gas.

[0158] Optionally, when the processor 320 executes the computer program A311, it can implement any of the embodiments in the corresponding examples of the above-described adaptive control method.

[0159] It should be noted that the electronic device proposed in this application embodiment is a device used to implement the above-mentioned adaptive control method. Therefore, based on the above-mentioned adaptive control method proposed in this application embodiment, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this application embodiment. Therefore, how the electronic device specifically implements the above-mentioned adaptive control method will not be described in detail here. Any electronic device used by those skilled in the art to implement the above-mentioned adaptive control method falls within the scope of protection of this application.

[0160] Based on the same technical concept described above, embodiments of this application also propose a computer-readable storage medium, such as... Figure 4 As shown, the computer-readable storage medium 400 stores a computer program B411, which, when executed by a processor, performs the following steps:

[0161] Simultaneously acquire and preprocess multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes at least gas pressure, gas temperature, mixed gas ratio, and equipment load current;

[0162] Based on the gas law, the gas pressure is compensated by the gas temperature to obtain the normalized pressure at the standard temperature.

[0163] Based on historical normalized pressure time series data, the future pressure change trend is calculated through a time prediction model, and the gas leakage rate is estimated.

[0164] A tiered early warning system is implemented based on the predicted normalized pressure and gas leakage rate: when both the normalized pressure and gas leakage rate are within the normal range or slightly deviate from the normal range (i.e., the deviation is less than the threshold), the subsequent steps continue; when the normalized pressure or gas leakage rate deviates abnormally (i.e., the deviation is greater than the threshold), the system automatically enters a safety restriction mode, and the subsequent steps continue only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal.

[0165] The pressure control setpoint is dynamically generated based on the target pressure and the gas leakage rate.

[0166] The predicted normalized pressure is used as input to the deviation and rate of change of the pressure control setpoint, as well as the equipment load current. The parameters of the controller are adjusted online adaptively through a fuzzy inference mechanism, and the first control quantity is calculated.

[0167] Based on the gas leakage rate and combined with the feedforward control gain, the feedforward compensation control quantity is calculated.

[0168] The first control quantity is superimposed and limited with the feedforward compensation control quantity to generate the total control quantity. The total control quantity is then converted into an analog quantity or PWM control signal to drive the proportional regulating valve in the gas injection unit, so as to achieve precise and continuous adjustment of the gas injection flow rate and the ratio of the mixed gas.

[0169] Optionally, when the computer program B411 is executed by the processor, it can implement any of the embodiments corresponding to the above-described adaptive control method.

[0170] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0171] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0172] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0173] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0174] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0175] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An adaptive control method for GIS mixed gas replenishment, characterized in that, include: The system simultaneously acquires and preprocesses multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes gas pressure, gas temperature, mixed gas ratio, and equipment load current. Based on the gas law, the gas pressure is compensated by the gas temperature to obtain the normalized pressure at the standard temperature. Based on historical normalized pressure time series data, the future pressure change trend is calculated through a time prediction model, and the gas leakage rate is estimated. A tiered early warning system is implemented based on the predicted normalized pressure and gas leakage rate: when the predicted normalized pressure and gas leakage rate are both within the normal range or the deviation from the normal range is less than the threshold, the subsequent steps continue; when the predicted normalized pressure or gas leakage rate deviates from its normal range by more than the threshold, the system first enters a safety restriction mode, and the subsequent steps continue only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal. The pressure control setpoint is dynamically generated based on the target pressure and the gas leakage rate. The deviation between the predicted normalized pressure and the pressure control setpoint, the rate of change of the deviation, and the equipment load current are used as inputs. The parameters of the controller are adjusted online adaptively through a fuzzy inference mechanism, and the first control quantity is calculated. Based on the gas leakage rate and combined with the feedforward control gain, the feedforward compensation control quantity is calculated. The first control quantity and the feedforward compensation control quantity are superimposed and limited to generate a total control quantity, which is then converted into a corresponding control signal and sent to the execution unit to achieve precise and continuous adjustment of the supplementary gas flow rate and the ratio of the mixed gas.

2. The adaptive control method for GIS mixed gas replenishment according to claim 1, characterized in that, The calculation of gas pressure compensation using the gas temperature includes: Convert the gas temperature into thermodynamic temperature; Based on the aforementioned thermodynamic temperature, the compressibility factor under the current state and the compressibility factor at the standard temperature are calculated using a method based on a simplified form of the modified Benedict-Wade-Rubin equation of state. Calculate the normalized pressure based on the compressibility factor under the current conditions and the compressibility factor at standard temperature: P_norm = P_m * (293.15 / T_k) * (Z2 / Z1); Where P_norm is the normalized pressure, P_m is the gas pressure, T_k is the thermodynamic temperature, and Z1 and Z2 are the compressibility factors under the current state and the compressibility factors under the standard temperature, respectively.

3. The adaptive control method for GIS mixed gas replenishment according to claim 2, characterized in that, The calculation of the compressibility factor under the current state and the compressibility factor at the standard temperature includes: Calculate the compression factor Z1 in the current state: Z1 = 1 + (a1 + a2 / T_k) * P_m + (a3 + a4 / T_k) *P_m^2; Calculate the initial estimated normalized pressure P_est: P_est = P_m * 293.15 / T_k; Calculate the compressibility factor Z2 at standard temperature: Z2 = 1 + (a1 + a2 / 293.15) * P_est + (a3 + a4 / 293.15) * P_est^2; Among them, a1, a2, a3 and a4 are coefficients related to the proportion of the mixed gas.

4. The adaptive control method for GIS mixed gas replenishment according to claim 1, characterized in that, The method of calculating future pressure change trends and estimating gas leakage rates using a time-based prediction model includes: Acquire historical normalized pressure time series data and set a sliding analysis window of a preset length; A second-order autoregressive model is used to describe the dynamic change process of pressure, and the model parameters are estimated online in real time using the recursive least squares method with a forgetting factor. In each control cycle, the normalized pressure prediction sequence for a future period is iteratively calculated using the latest estimated parameters and the current sliding analysis window data. Extract the normalized pressure values ​​of the last segment of the normalized pressure prediction sequence, and perform linear least squares fitting based on the extracted normalized pressure values ​​and their corresponding time indices to obtain the slope of the fitted line. The gas leakage rate V_leak is calculated based on the slope of the fitted line: V_leak = - (k_slope / Δt) * 3600; where k_slope is the slope of the fitted line and Δt is the sampling period. The new normalized pressure data is used to update the sliding analysis window to prepare for model parameter estimation in the next cycle.

5. The adaptive control method for GIS mixed gas replenishment according to claim 1, characterized in that, The method of online adaptive adjustment of controller parameters through fuzzy inference mechanism and calculation of the first control quantity includes: The deviation and rate of change of the normalized pressure obtained from the current input prediction from the pressure control setpoint and the equipment load rate are fuzzified, and the corresponding fuzzy rules are matched from the pre-built fuzzy rule library according to the fuzzy results to obtain the corresponding output fuzzy set; wherein the output fuzzy set is the controller parameter fuzzy set. The output fuzzy set is synthesized by MAX, and the fuzziness is defuzzified by the centroid method to obtain the precise controller parameter adjustment amount; Update the controller parameters according to the adjustment amount of the controller parameters; Based on the updated controller parameters, the first control quantity is calculated using an incremental PID algorithm.

6. The adaptive control method for GIS mixed gas replenishment according to any one of claims 1-5, characterized in that, The process of converting the total control quantity into a corresponding control signal includes: The total control quantity is converted into the total make-up gas flow rate under standard conditions by looking up a table. Based on the proportion of the mixed gas, calculate the flow rate setpoint for each component and input the flow rate setpoint for each component into the corresponding mass flow controller.

7. The adaptive control method for GIS mixed gas replenishment according to claim 6, characterized in that, The method of converting the total control quantity into a corresponding control signal further includes: When an online gas proportioning analyzer is configured to monitor the volume concentration of sulfur hexafluoride in the mixed gas in real time, an outer loop correction algorithm is executed: the actual concentration of sulfur hexafluoride measured by the online gas proportioning analyzer is read, and the proportioning error is calculated; the flow correction amount is calculated through the PI controller; and the flow correction amount is used to correct the flow setpoint of each component. And / or, when the total replenishment flow rate decreases according to control requirements, the flow rate setpoints of each component are updated proportionally, while the update process meets the constraint that the change in the flow rate setpoint per second does not exceed 20% of the full scale.

8. A GIS mixed gas replenishment adaptive control system, characterized in that, include: The preprocessing unit is configured to: synchronously acquire and preprocess multi-source sensor data from the GIS gas chamber; the multi-source sensor data includes gas pressure, gas temperature, mixed gas ratio, and equipment load current; The compensation unit is configured to: perform compensation calculations on the gas pressure based on the gas state equation, using the gas temperature to obtain the normalized pressure at the standard temperature; The trend prediction unit is configured to: calculate the future pressure change trend and estimate the gas leakage rate based on historical normalized pressure time series data using a time prediction model; The graded early warning unit is configured to: perform decomposed early warning based on the normalized pressure and gas leakage rate predicted by the trend prediction unit; when the predicted normalized pressure and gas leakage rate are both within the normal range or the deviation from the normal range is less than the threshold, the adaptive decision unit is activated; when the predicted normalized pressure or gas leakage rate deviates from its normal range by a value greater than the threshold, the unit first enters the safety restriction mode, and the adaptive decision unit is activated only after the maintenance personnel have eliminated the safety hazards and confirmed that the system has returned to normal. And, the adaptive decision unit is configured to: dynamically generate a pressure control setpoint based on the target pressure and the gas leakage rate; take the deviation and rate of change of the predicted normalized pressure from the pressure control setpoint, and the equipment load current as inputs, and adaptively adjust the parameters of the controller online through a fuzzy inference mechanism, and calculate the first control quantity; Based on the gas leakage rate and combined with the feedforward control gain, the feedforward compensation control quantity is calculated; the first control quantity and the feedforward compensation control quantity are superimposed and limited to generate a total control quantity, and the total control quantity is converted into a corresponding control signal and sent to the execution unit to achieve precise and continuous adjustment of the supplementary gas flow rate and the ratio of the mixed gas.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the adaptive control method for GIS mixed gas replenishment as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the adaptive control method for GIS mixed gas replenishment as described in any one of claims 1-7.

Images