Intelligent gas supply system and intelligent gas supply control method for SF6 gas insulation equipment

Abstract

The invention provides an intelligent gas supply system of SF6 gas insulation equipment and an intelligent gas supply control method. The system comprises SF6 gas source equipment, the SF6 gas insulation equipment, a gas pressure sensor, a gas temperature sensor and a gas supply control module, the air supply control module comprises a control unit and an electromagnetic valve; the gas inlet end of the electromagnetic valve is connected with the gas outlet end of the SF6 gas source equipment through a pipeline, and the gas outlet end of the electromagnetic valve is connected with the gas inlet end of the SF6 gas insulation equipment through a pipeline; the control module is used for receiving a gas pressure value collected by the gas pressure sensor and a gas temperature value collected by the gas temperature sensor; dynamically calculating a gas pressure dynamic threshold value for judging SF6 gas leakage; the gas pressure value measured in real time is compared with the current corresponding gas pressure dynamic threshold value, and if the gas pressure value measured in real time is smaller than or equal to the current corresponding gas pressure dynamic threshold value, the electromagnetic valve is controlled to be opened, so that the SF6 gas source equipment supplements gas to the SF6 gas insulation equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic control of operation and maintenance of electrical equipment, and in particular to an intelligent gas replenishment system and an intelligent gas replenishment control method for SF6 gas insulation equipment. Background Art

[0002] SF6 (sulfur hexafluoride) is a colorless, odorless, non-toxic artificial inert gas with excellent insulation properties. In a uniform electric field, the insulation strength of SF6 is 2.5 to 3 times that of air (at 1 atm), and increases nonlinearly with increasing pressure. In addition, SF6 also has excellent arc extinguishing ability. Its molecules can quickly absorb free electrons, and the arc time constant is 2 orders of magnitude lower than that of air. It is widely used in circuit breaker arc extinguishing chambers. Under dry conditions, SF6 does not react with metal materials below 200°C, making it suitable for long-term sealed use.

[0003] Based on the above characteristics, SF6 gas is widely used in 345kV and above substations as the core insulating medium of high / ultra-high voltage equipment (GIS, GIL, circuit breakers).

[0004] Typical SF6 equipment includes a sealed cavity with an aluminum alloy / stainless steel shell, filled with 0.4~0.6MPaSF6 gas, and contains modules such as circuit breakers, disconnectors, and current transformers. However, during operation, SF6 equipment is prone to gas leakage due to poor sealing or material aging, which not only threatens the safe operation of the equipment, but also may pollute the environment.

[0005] In order to prevent SF6 gas leakage, traditional technology uses manual monitoring and gas replenishment solutions. This method is inefficient and cannot accurately grasp the gas conditions inside the equipment in real time, which easily leads to untimely or excessive gas replenishment.

[0006] In some related technologies, attempts are made to use infrared image detection or monitoring of gas pressure thresholds to determine gas leakage in SF6 equipment and to replenish gas. For the latter, since it only relies on a fixed pressure threshold and does not consider the impact of dynamic temperature changes on gas density, it is also easy to cause false replenishment or missed replenishment. For example, fixed threshold alarms cannot distinguish between normal gas expansion (temperature difference between day and night) and real leaks, resulting in false alarms or missed alarms.

[0007] In view of this, the present invention is proposed. Summary of the invention

[0008] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide an intelligent gas replenishment system and an intelligent gas replenishment control method for SF6 gas insulation equipment, so as to solve the technical problems in the related technology that manual gas replenishment easily leads to untimely or excessive gas replenishment, and automatic gas replenishment using a fixed threshold easily leads to incorrect or missed gas replenishment.

[0009] In order to achieve the above technical objectives, the present invention adopts the following technical solutions: According to one aspect of the present invention, there is provided an intelligent gas supply system for SF6 gas-insulated equipment, comprising SF6 gas source equipment, SF6 gas-insulated equipment, a gas pressure sensor for monitoring the gas pressure value of the SF6 gas-insulated equipment, a gas temperature sensor for monitoring the gas temperature value of the SF6 gas-insulated equipment, and a gas supply control module; The gas replenishment control module includes a control unit and a solenoid valve connected to the control unit by signal; the air inlet end of the solenoid valve is connected to the air outlet end of the SF6 gas source equipment through a pipeline, and the air outlet end of the solenoid valve is connected to the air inlet end of the SF6 gas insulation equipment through a pipeline; the control module is also connected to the gas pressure sensor and the gas temperature sensor by signal; The control module is used for: receiving a gas pressure value collected by the gas pressure sensor and a gas temperature value collected by the gas temperature sensor; Dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value; Compare the gas pressure value measured in real time with the current corresponding gas pressure dynamic threshold. If the gas pressure value measured in real time is less than or equal to the current corresponding gas pressure dynamic threshold, control the solenoid valve to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

[0010] Optionally, the dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculating the SF6 gas pressure change rate, and multiplying the SF6 gas pressure change rate by a pre-calibrated pressure change rate gain coefficient to obtain a pressure correction term; Calculating the temperature difference between the gas temperature value received in real time and a preset temperature reference value, and multiplying the temperature difference by a pre-calibrated temperature compensation coefficient to obtain a temperature compensation term; The preset gas pressure reference threshold for determining SF6 gas leakage is corrected at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold.

[0011] Optionally, the gas pressure reference threshold preset for determining SF6 gas leakage is corrected at least according to the pressure correction item and the temperature compensation item to obtain the gas pressure dynamic threshold, specifically comprising: calculating the gas pressure dynamic threshold according to the following formula: P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref), wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, and T_ref is the temperature reference value.

[0012] Optionally, the dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically further includes: Calculate the square of the load current of the SF6 gas-insulated equipment, and then multiply it by a pre-calibrated load current compensation coefficient to obtain a load current compensation term; The step of correcting the preset gas pressure reference threshold for determining SF6 gas leakage at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold specifically includes: According to the pressure correction item, the temperature compensation item, and the load current compensation item, the preset gas pressure reference threshold for determining SF6 gas leakage is corrected to obtain the gas pressure dynamic threshold.

[0013] Optionally, the step of correcting a preset gas pressure reference threshold for determining SF6 gas leakage according to the pressure correction term, the temperature compensation term, and the load current compensation term to obtain the gas pressure dynamic threshold comprises: according to the following formula: P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref)+β*I 2 , the gas pressure dynamic threshold is calculated, wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, T_ref is the temperature reference value, I is the load current, and β is the load current compensation coefficient.

[0014] Optionally, the dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculating the SF6 gas pressure change rate, and multiplying the SF6 gas pressure change rate by a pre-calibrated pressure change rate gain coefficient to obtain a pressure correction term; Determine whether the gas temperature value received in real time is greater than a preset temperature reference value. If so, confirm that the SF6 gas insulation equipment is in a high temperature zone, and calculate the high temperature zone temperature compensation term according to the high temperature zone compensation equation, and then correct the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the high temperature zone temperature compensation term to obtain the gas pressure dynamic threshold; if not, confirm that the SF6 gas insulation equipment is in a low temperature zone, and calculate the low temperature zone temperature compensation term according to the low temperature zone compensation equation, and then correct the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the low temperature zone temperature compensation term to obtain the gas pressure dynamic threshold.

[0015] Optionally, calculating the high temperature zone temperature compensation term according to the high temperature zone compensation equation includes: calculating the high temperature zone temperature compensation term according to the following formula: α(T 高 )=α1(T-T_ref)+α2(T-T_ref) 2 ; The calculation of the low temperature zone temperature compensation term according to the low temperature zone compensation equation includes: calculating the low temperature zone temperature compensation term according to the following formula: α (T 低 )=α3ln(T / T_ref); where, α(T 高 ) is the temperature compensation term in the high temperature zone; α(T 低 ) is the temperature compensation item of the low temperature zone; α1 and α2 are the temperature compensation coefficients of the high temperature zone, and α3 is the temperature compensation coefficient of the low temperature zone; T is the gas temperature value received in real time, and T_ref is the preset temperature reference value.

[0016] Optionally, the control module is further used to: generate simulation data of the operating conditions of the SF6 gas insulation equipment, input the simulation data into the LSTM neural network model for pre-training, the output layer is used to predict the SF6 gas pressure at future time nodes, and perform framework adjustment to obtain the LSTM-based SF6 gas pressure prediction model; The gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor are input into the SF6 gas pressure prediction model to obtain the SF6 gas pressure prediction value at a future time node, and the SF6 gas pressure prediction value at the future time node is compared with the gas pressure dynamic threshold value, and it is determined whether to start the solenoid valve for gas replenishment according to the comparison result.

[0017] Optionally, the gas pressure sensor and the gas temperature sensor both communicate with the control unit via wireless signals, and the control unit is further used to: generate a communication reliability index based on the wireless communication signal packet loss rate, delay and signal strength; if the communication reliability index is less than or equal to a preset first threshold and greater than or equal to a preset second threshold, execute a gas replenishment strategy based on the gas pressure dynamic threshold; if the communication reliability index is less than the second threshold, switch the gas pressure dynamic threshold to a fixed gas pressure threshold, and execute a gas replenishment strategy based on the fixed gas pressure threshold.

[0018] According to another aspect of the present invention, there is also provided an intelligent gas supplement control method, which is implemented using the intelligent gas supplement system of the SF6 gas insulation equipment described above, and the control method comprises: receiving a gas pressure value collected by the gas pressure sensor and a gas temperature value collected by the gas temperature sensor; Dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value; Compare the gas pressure value measured in real time with the current corresponding gas pressure dynamic threshold. If the gas pressure value measured in real time is less than or equal to the current corresponding gas pressure dynamic threshold, control the solenoid valve to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

[0019] The present invention provides an intelligent gas replenishment system and an intelligent gas replenishment control method for SF6 gas-insulated equipment. The system adopts a gas pressure sensor and a temperature sensor to monitor the gas temperature value and the gas pressure value of the SF6 gas-insulated equipment in real time, and judges whether it is necessary to replenish gas for the SF6 gas-insulated equipment in combination with a dynamic threshold. Compared with manual monitoring, the system is more accurate and timely, and the dynamic threshold is obtained after correction based on at least the gas pressure value and the gas temperature value measured in real time. Compared with the fixed threshold, the system is more in line with the SF6 gas working conditions under the actual pressure change state and the temperature change state, and can judge whether the SF6 gas equipment is leaking more truly and accurately, thereby reducing the risk of mis-replenishment and missed replenishment. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A structural schematic diagram of an intelligent gas supply system for SF6 gas-insulated equipment is provided for an embodiment of the present invention; Figure 2 The present invention provides a flow chart of an intelligent gas replenishment control method. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0022] An intelligent gas supply system for SF6 gas insulation equipment provided by the embodiment of the present invention is combined with Figure 1 , including SF6 gas source equipment, SF6 gas insulation equipment, a gas pressure sensor for monitoring the gas pressure value of the SF6 gas insulation equipment, a gas temperature sensor for monitoring the gas temperature value of the SF6 gas insulation equipment; and a gas replenishment control module. As an example, the gas pressure sensor and the gas temperature sensor can be set separately or integrated in one sensor, such as Figure 1 In this embodiment, the gas temperature sensor and the gas pressure sensor are integrated into one SF sensor, which is connected to the pipeline between the SF6 gas insulation equipment and the gas replenishment control module, and is used to monitor the SF6 gas pressure value and gas temperature value in the SF gas insulation equipment in real time.

[0023] As an example, the gas replenishment control module includes a control unit and a solenoid valve connected to the control unit signal; the gas inlet end of the solenoid valve is connected to the gas outlet end of the SF6 gas source device through a pipeline, and the gas outlet end of the solenoid valve is connected to the gas inlet end of the SF6 gas insulation device through a pipeline; the control module is also connected to the gas pressure sensor and the gas temperature sensor signal to receive the gas pressure value signal transmitted from the gas pressure sensor and the gas temperature value signal transmitted from the gas temperature sensor in real time. The control module and the SF6 sensor can be connected by wire or wirelessly.

[0024] The control unit may include a microprocessor and a memory. The microprocessor may be connected to a communication module, such as a Bluetooth module, a 5G module, or a Lora communication module, and connected to the SF6 gas pressure sensor and the gas temperature sensor signal by means of the communication module. Of course, if the distance is short, the control unit may also be connected to the gas pressure sensor and the gas temperature sensor by a signal line. Accordingly, the gas pressure sensor and the gas temperature sensor may also be configured with a communication module of a type corresponding to the control unit for data transmission. Furthermore, the control unit and / or the gas pressure sensor and the gas temperature sensor may also be connected to a proximal smart device (such as a mobile phone, a tablet computer, a laptop computer) or a remote device (such as a cloud server) for communication to transmit the operating data of the SF6 gas-insulated equipment, so as to facilitate monitoring (such as when a leak occurs, the control unit generates an alarm signal and sends it to the proximal smart device or the remote device) or data processing. The control unit may be connected to a power source such as a lithium battery for power supply. The SF6 gas source equipment may include an SF6 gas supply tank, which stores SF6 gas, and the gas pressure of which is higher than that of the SF6 gas-insulated equipment, so that after the solenoid valve is automatically opened, the pressure difference may be used to supply gas to the inside of the SF6 gas-insulated equipment.

[0025] In this embodiment, the control module is used to: The gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor are received.

[0026] Dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage based at least on the gas pressure value and the gas temperature value; The real-time measured gas pressure value is compared with the current corresponding gas pressure dynamic threshold. If the real-time measured gas pressure value is less than or equal to the current corresponding gas pressure dynamic threshold, the solenoid valve is controlled to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

[0027] For example, a computer program may be stored in the memory, and after the computer program is executed by the microprocessor, the above steps are performed (of course, the various steps described in the following embodiments may also be performed, which will not be repeated here).

[0028] In the above embodiment, as an example, the dynamic calculation of the gas pressure dynamic threshold for determining SF6 gas leakage based on at least the gas pressure value and the gas temperature value specifically includes: The pressure change rate of the SF6 gas is calculated, and the pressure correction term is obtained by multiplying the pressure change rate of the SF6 gas by a pre-calibrated pressure change rate gain coefficient.

[0029] Calculate the temperature difference between the gas temperature value received in real time and the preset temperature reference value, and multiply the temperature difference by a pre-calibrated temperature compensation coefficient to obtain a temperature compensation term; A preset gas pressure reference threshold for determining SF6 gas leakage is corrected at least according to the pressure correction term and the temperature compensation term to obtain a gas pressure dynamic threshold.

[0030] In this embodiment, the pressure correction term is obtained by combining the SF6 gas pressure change rate with the pre-calibrated pressure change rate gain coefficient. If the pressure drops sharply due to a sudden leak, this term will dynamically lower the threshold, accelerate the triggering of gas replenishment or alarm, and avoid missed reports. In addition, the temperature compensation term is used to compensate for the thermal expansion / contraction effect to prevent the system from misjudging the pressure as normal when the temperature fluctuates, thereby covering up the real leak. In addition, the combination of the pressure correction term and the temperature compensation term not only covers short-term dynamic disturbances, but also compensates for long-term environmental changes, so that the pressure threshold is adaptive to the actual working conditions of the equipment, significantly improving the robustness of the monitoring system.

[0031] As an example, the above-mentioned method of correcting the preset gas pressure reference threshold for determining SF6 gas leakage at least according to the pressure correction item and the temperature compensation item to obtain the gas pressure dynamic threshold specifically includes: calculating the gas pressure dynamic threshold according to the following formula (hereinafter referred to as formula (1)): P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref), wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, which can be preset, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, and T_ref is the temperature reference value, which can be preset.

[0032] The pressure change rate dP / dt reflects how fast the pressure changes. The absolute value symbol indicates that the system needs to respond to rapid changes regardless of whether the pressure is rising or falling. If the pressure drops sharply due to a sudden leak (dP / dt<0), this item will dynamically reduce the threshold (amplify P_set by 1 + k|dP / dt|), accelerate the triggering of gas replenishment or alarm, and avoid missed alarms. Furthermore, this pressure correction item can also be used to suppress short-term fluctuation interference. For example, if the pressure fluctuates due to noise or short-term disturbances (such as valve action), the gain factor k can adjust the sensitivity to prevent false alarms (for example, setting k is only effective for dP / dt exceeding the threshold).

[0033] For example, if the equipment fails to seal and the pressure drops at dP / dt=-0.1MPa / min, assuming k=2min / MPa, the threshold is corrected to: P_adj = P_set × (1+2*0.1) = 1.2P_set. At this time, if the real-time monitored pressure value is less than 1.2P_set, the control unit directly generates a leakage alarm signal and opens the solenoid valve to automatically replenish air to respond quickly and avoid missed reports due to untimely response.

[0034] According to the ideal gas law, an increase in temperature will cause the gas pressure to naturally increase, and vice versa. The α in the above text is the temperature-pressure coupling coefficient, which is used to quantify the pressure change caused by a unit temperature rise. When the internal temperature of the device rises due to load current or ambient temperature, the actual pressure will be higher than the preset value P_set due to thermal expansion. If not corrected, the system may mistakenly judge that the pressure is normal, covering up the real leak. The threshold is raised to a theoretical value that matches the current temperature through α(T-T_ref), so that the leak judgment is based on the temperature-corrected pressure.

[0035] For example, if the device is at T_ref=20℃ and P_set=0.5MPa, when the temperature rises to T=40℃ and α=0.01MPa / ℃, the threshold is corrected to: P_adj = 0.5+0.01(40-20)=0.7MPa. At this time, if the real-time monitored pressure value is 0.6MPa, although it is greater than P_set, but less than P_adj, the control unit will directly generate a leakage alarm signal and open the solenoid valve to automatically replenish air to avoid missed reports and missed replenishments due to temperature fluctuations.

[0036] In the formula of this embodiment, the pressure change rate gain coefficient k and the temperature compensation coefficient α can be calibrated in advance through experiments. For example, the pressure change rate gain coefficient k simulates different leakage rates (such as slow leakage vs. fast leakage) in a controllable environment, and k is adjusted so that the threshold can accurately distinguish between normal fluctuations and real faults. As a further improvement, k can also be constrained to be less than (P_max-P_set) / (P_set×|dP / dt|_max), where P_max is the maximum gas pressure allowed by SF6 gas insulation equipment, and |dP / dt|_max is the maximum gas pressure change rate allowed by SF6 gas insulation equipment. In this way, the set dynamic gas pressure threshold can be prevented from exceeding the upper limit. The temperature compensation coefficient α measures the pressure at different temperatures, and fits α to minimize the error between P_adj and the measured pressure.

[0037] The inventors found during the research that although the dynamic threshold of gas pressure of SF6 gas-insulated equipment calculated according to the above strategy is more accurate than the fixed threshold, there is still room for improvement. For example, in the formula (1) introduced above, the temperature compensation term is linear, but SF6 gas-insulated equipment often operates in a high temperature environment or a low temperature environment. At this time, when the SF6 gas is at high temperature or low temperature, the pressure-temperature relationship deviates from the ideal gas law, and the linear compensation error is significant.

[0038] In order to solve this problem, when executing the dynamic threshold calculation strategy, the control unit, on the basis of the above embodiment, further includes: judging whether the gas temperature value received in real time is greater than the preset temperature reference value; if so, confirming that the SF6 gas insulation equipment is in the high temperature zone, and calculating the high temperature zone temperature compensation term according to the high temperature zone compensation equation, and then correcting the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the high temperature zone temperature compensation term, to obtain the gas pressure dynamic threshold; if not, confirming that the SF6 gas insulation equipment is in the low temperature zone, and calculating the low temperature zone temperature compensation term according to the low temperature zone compensation equation, and then correcting the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the low temperature zone temperature compensation term, to obtain the gas pressure dynamic threshold. In this way, the temperature is divided into a high temperature zone and a low temperature zone, and different temperature compensation term calculation strategies are called respectively, so that the gas pressure dynamic threshold can be adjusted more accurately, and the accuracy of leakage alarm and gas replenishment can be improved, so that the calculation strategy provided in this embodiment can be applied to the extreme temperature environment where the SF6 equipment works.

[0039] As an example, the high temperature zone temperature compensation term is calculated according to the high temperature zone compensation equation, including: calculating the high temperature zone temperature compensation term according to the following formula: α(T 高 )=α1(T-T_ref)+α2(T-T_ref) 2 ; The calculation of the low temperature zone temperature compensation term according to the low temperature zone compensation equation includes: calculating the low temperature zone temperature compensation term according to the following formula: α (T 低 )=α3ln(T / T_ref); where, α(T 高 ) is the temperature compensation term in the high temperature zone; α(T 低 ) is the temperature compensation term of the low temperature zone; α1 and α2 are the temperature compensation coefficients of the high temperature zone, and α3 is the temperature compensation coefficient of the low temperature zone; T is the gas temperature value received in real time, and T_ref is the preset temperature reference value. Accordingly, formula (1) is further transformed into formula (2), that is, the dynamic threshold of gas pressure is calculated according to formula (2): P_adj = P_set × (1 + k|dP / dt|) + α1(T-T_ref)+α2(T-T_ref) 2 (T>T_ref); P_adj = P_set × (1 + k|dP / dt|) + α3ln(T / T_ref)(T≤T_ref). (The above two sets of equations are formula (2)). For the convenience of calculation, Kelvin temperature can be used when T≤T_ref.

[0040] In the high temperature area, the gas expansion is nonlinearly enhanced, and the pressure increases rapidly with the temperature. In the low temperature area, the gas is close to liquefaction, and the pressure drop rate slows down. Using the above formula (2), in the high temperature area, through the quadratic term α2 (T-T_ref) 2 Capture the nonlinearity of gas expansion. In the low temperature zone, the logarithmic function ln (T / T_ref) is used to fit the attenuation characteristics of pressure changes under the liquefaction trend. In the above calculation strategy, α1, α2, and α3 can also be calibrated experimentally. In the high temperature zone, the equipment can be heated to (T_1, T_2, ..., T_n P_1, P_2, ..., P_n) under constant pressure. Then follow P = P_set + α1 (T-T_ref) + α2 (T-T_ref) 2 Perform quadratic fitting and use the least squares method to solve α1 and α2. In the low temperature zone, cool to (T_1, T_2,..., T_n $ (such as -20°C, -40°C, -60°C), record the pressure. Then perform logarithmic fitting according to P = P_set +α3ln (T / T_ref), and solve α3 by nonlinear regression.

[0041] As an example, assume that after calibration, α1 = 0.02MPa / °C, α2 = 0.0005MPa / °C, α3 = −0.12MPa. T_ref = 20℃, assuming that the SF6 equipment works under extreme conditions of extremely high temperature (T = 80℃) and extremely low temperature (T = -50℃). In the high temperature area, if linear compensation is used, the compensation value is 0.02×(80−20)=1.2MPa. According to nonlinear compensation, the compensation value is 0.02×60+0.0005×60 2 =1.2+1.8=3.0MPa. It can be seen that if the actual pressure reaches 3.0MPa due to nonlinear expansion, insufficient compensation of the linear model will lead to misjudgment of leakage, and the improved nonlinear compensation model can overcome this defect. In the low temperature zone, the linear model compensation value is 0.02×(-70)=−1.4MPa (an unreasonable negative value), and the nonlinear compensation value is -0.12×ln(223.15 / 293.15)=0.033MPa. It can be seen that at extremely low temperatures, the linear model fails at low temperatures, and the nonlinear model is more in line with the SF6 liquefaction trend.

[0042] As a further improvement, in the calculation strategy of the above formula (1) or (2), a load current compensation term can be further introduced, that is: The dynamic calculation of the gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculate the square of the load current of the SF6 gas-insulated equipment, and then multiply it by a pre-calibrated load current compensation coefficient to obtain a load current compensation term; The step of correcting the preset gas pressure reference threshold for determining SF6 gas leakage at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold specifically includes: According to the pressure correction item, the temperature compensation item, and the load current compensation item, the preset gas pressure reference threshold for determining SF6 gas leakage is corrected to obtain the gas pressure dynamic threshold.

[0043] As an example, the preset gas pressure reference threshold for judging SF6 gas leakage is corrected according to the pressure correction term, the temperature compensation term, and the load current compensation term to obtain the gas pressure dynamic threshold, specifically including: according to the following formula P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref)+β*I 2 (Formula (3), the gas pressure dynamic threshold is calculated, wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, T_ref is the temperature reference value, I is the load current, and β is the load current compensation coefficient. (This example introduces the load current compensation term into formula (1)).

[0044] As another example, the preset gas pressure reference threshold for judging SF6 gas leakage is corrected according to the pressure correction item, the temperature compensation item, and the load current compensation item to obtain the gas pressure dynamic threshold, specifically including: when T>T_ref, according to the following formula P_adj = P_set × (1 + k|dP / dt|) + α1(T-T_ref)+α2(T-T_ref) 2 +β*I 2 (Formula (4), calculate the dynamic threshold of the gas pressure, when T≤T_ref, according to the following formula P_adj = P_set × (1 + k|dP / dt|) + α3ln(T / T_ref)+β*I 2 , the gas pressure dynamic threshold is calculated (this example introduces the load current compensation term into formula (2)).

[0045] In the above formulas (1) and (2), the dynamic threshold is corrected only by the temperature compensation term. In addition to the change in external ambient temperature, the temperature change also comes from the change in load current inside the SF6 gas-insulated equipment, resulting in the temperature rise caused by the Joule heat generated. However, the conduction of Joule heat is characterized by locality. It first generates heat around the conductor and then needs to be conducted to the gas temperature sensor through the gas, which causes a significant delay. If only the temperature compensation term is relied upon, when the current suddenly changes (such as a short-circuit shock), the sensor cannot detect the local temperature rise in time, resulting in compensation lag, which may trigger false protection. In order to solve the above problems, the present invention also introduces a load current compensation term to directly compensate for transient thermal effects through real-time current measurement (which can respond as low as milliseconds) to avoid delay problems.

[0046] Similar to the pressure change rate gain coefficient k and the temperature compensation coefficient α, the load current compensation coefficient β can also be pre-calibrated by experiment. For example, when the load current I passes through the equipment, the Joule heat (proportional to the square of I) generated by the conductor resistance will heat the SF6 gas, causing the pressure to rise. The calibration of β needs to be established by fitting experimental data or deriving a thermodynamic model to establish the current square I 2 The linear relationship with the pressure increment △P. The calibration steps include: first, conduct a control variable experiment, fix the ambient temperature T=T_ref, turn off the dynamic pressure change term (k=0), apply different load currents I1, I2, I3...In, record the steady-state pressure values ​​P1, P2, P3...Pn, and calculate the pressure increment △Pi=Pi-P_set (i∈1-n). Use linear regression to fit △P=βI 2 Finally, the model is verified according to the thermodynamic formula △P∝I 2 Rt, verify whether the theoretical value of β is consistent with the actual value. For example, if the experimental measurement shows that I=1kA, △P=0.02MPa, then β=△P / I 2 =0.02MPa / kA 2 .

[0047] In addition, SF6 has certain particularities when it leaks, that is, since SF6 gas is used for insulation of high-voltage equipment, high-voltage arcs are sometimes released, and at this time SF6 will decompose into SO2 and H2S. Therefore, as a further improvement, the system also includes an electrochemical sensor for monitoring the concentration of SO2 and / or H2S to obtain a decomposition product concentration value. The control unit is connected to the electrochemical sensor signal, and the control unit is also used to generate a decomposition product compensation term γ*C based on the received decomposition product concentration value, wherein γ is the decomposition product compensation coefficient, and C is the decomposition product concentration value. Then, based on the decomposition product compensation term, combined with the pressure correction term, temperature compensation term, and load current compensation term mentioned above, the gas pressure dynamic threshold is jointly corrected. For example, it can be calculated according to the following formula: P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref)+β*I 2 +γ*C calculation. This also combines the decomposition product concentration value to correct the dynamic threshold, further improving the accuracy.

[0048] The decomposition product compensation coefficient γ can be set based on empirical values, theoretical derivation, or pre-calibrated using similar experimental methods as above. For example, a controlled environment can be designed, pure SF6 is filled into a closed container, and the initial pressure P0 and temperature T0 are recorded. Then, a known concentration C SO2 SO2 gas, keep the temperature constant, measure the pressure change △P, and then repeat at different C SO2 Working condition, obtain multiple sets of data, and then draw C SO2 The scatter plot of ΔP is finally fitted, and the decomposition product compensation coefficient γ is obtained by linear regression. If the linearity is not significant, a high-order polynomial or piecewise compensation model can also be used. The principle is similar to the other compensation coefficients mentioned above and will not be repeated here.

[0049] As a further improvement, in some embodiments, the control module is further used to: generate simulation data of the operating conditions of the SF6 gas insulation equipment, input the simulation data into the LSTM neural network model for pre-training, the output layer is used to predict the SF6 gas pressure at future time nodes, and adjust the framework to obtain the SF6 gas pressure prediction model based on LSTM; The gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor are input into the SF6 gas pressure prediction model to obtain the SF6 gas pressure prediction value at a future time node, and the SF6 gas pressure prediction value at the future time node is compared with the gas pressure dynamic threshold value, and it is determined whether to start the solenoid valve for gas replenishment according to the comparison result.

[0050] In this embodiment, the present invention can obtain an LSTM neural network model based on the simulated SF6 gas insulation equipment operating data training, which is used to predict the SF6 gas pressure value in the future. When the predicted value is lower than the gas pressure dynamic threshold introduced in the previous embodiment (or other gas replenishment strategies), the solenoid valve can be opened in advance for pre-gas replenishment to prevent the gas pressure from dropping suddenly at a certain time point in the future, resulting in untimely gas replenishment.

[0051] As an example, the simulation data generated for the operating conditions of SF6 gas-insulated equipment can include pressure, temperature, load current, decomposition product concentration, etc. The time window can be configured to be 60 minutes and the sampling interval can be 1 minute. The above data is input into the LSTM neural network model as the input layer, and the feature extraction layer can include the Conv1D layer and the LSTM layer. The Conv1D layer can be configured with 64 filters, a window size of 5, and an activation function ReLU to extract local time series features. The LSTM layer can be configured as a 2-layer stack with 128 units per layer and dropout=0.2 to capture long-term dependencies. Furthermore, the attention mechanism can be called to weight key time steps (such as leakage starting point and load mutation point) to improve prediction sensitivity. The output layer is configured as the pressure prediction value (point prediction) and confidence interval (95% probability range) for the next 30 minutes.

[0052] Next, execute the transfer learning strategy. First, train the network based on simulation data (such as 100,000 groups). The loss function can be set to MAE (mean absolute error) + prediction interval calibration loss. For example, the loss function Loss is: Loss=MAE+λ·Interval_Loss. Interval_Loss=max(0,|P1-P_real|-1.96σ). Among them, MAE is the mean absolute error between the predicted value P1 of the gas pressure and the true value P_real, and Interval_Loss is the interval calibration loss, which forces the prediction confidence interval (95% probability) to cover the actual value to avoid the prediction interval being too narrow or too wide. σ is the prediction standard deviation, calculated by the model output layer, and 1.96 is the Z value coefficient of the 95% confidence interval of the normal distribution. λ is the weight of balancing MAE and Interval_Loss, preventing the interval from being too wide and sacrificing accuracy. The typical value is 0.1-0.5, and the optimal value can be selected on the validation set through cross-validation.

[0053] Then fine-tune the framework. You can use the real data of the device (3 to 5 gas replenishment records) to freeze the Conv1D and LSTM layers, and only train the attention layer and output layer. You can also perform data enhancement, such as adding noise to the real data, time scaling (±10%), and fragment reorganization.

[0054] In this embodiment, for example, SF6 gas leakage is an occasional event, and the number of labeled samples may be small. Direct training of the deep learning model is prone to overfitting. Therefore, through the simulation data pre-training-fine-tuning framework, LSTM is first pre-trained on a large amount of general time series data (such as pressure, temperature, and gas concentration historical data of multiple types of equipment) to learn basic time series features, and then fine-tuned through a small amount of target scene data to improve the generalization ability of the model in the target scene.

[0055] In addition, as a further improvement, the LSTM weight can be reduced from FP32 to INT8, which greatly reduces the model size. The activation functions of the Conv1D and LSTM layers can also be merged to reduce inference latency, making it more suitable for localized embedded deployment of prediction models and improving the local response and execution efficiency of SF6 equipment to gas replenishment strategies.

[0056] As an example, the gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor are input into the SF6 gas pressure prediction model to obtain the SF6 gas pressure prediction value at a future time node, the SF6 gas pressure prediction value at the future time node is compared with the gas pressure dynamic threshold value, and whether to start the solenoid valve for gas replenishment is determined according to the comparison result, specifically including: Follow the following strategy to perform gas replenishment: When P_t is less than P_adj-3σ, and dP / dt<-0.05MPa / min, gas replenishment is performed, otherwise the status quo is maintained. Among them, P_adj and dP / dt have the same meaning as before, P_t is the predicted value of gas pressure (such as after 30 minutes), and 3σ is three times the standard deviation of the prediction interval, which is used to set a safety margin to prevent false triggering of gas replenishment. At the same time, dP / dt is limited to ensure that only significant leaks are responded to.

[0057] The SF6 equipment in the present invention is mostly used in high-voltage substations. Strong electromagnetic pulses lead to a high packet loss rate in wireless communication. In the wireless communication scenario, if leakage judgment and gas replenishment adjustment are performed only based on dynamic thresholds, problems may occur in leakage alarm and gas replenishment due to the degradation of dynamic data collection quality. For this reason, the present invention adopts the following technical solutions: In this embodiment, as a further improvement, the gas pressure sensor and the gas temperature sensor both communicate with the control unit via wireless signals, and the control unit is also used to: generate a communication reliability index based on the wireless communication signal packet loss rate, delay and signal strength; if the communication reliability index is less than or equal to a preset first threshold and greater than or equal to a preset second threshold, the gas replenishment strategy is executed according to the gas pressure dynamic threshold; if the communication reliability index is less than the second threshold, the gas pressure dynamic threshold is switched to a fixed gas pressure threshold, and the gas replenishment strategy is executed according to the fixed gas pressure threshold.

[0058] In this way, when the communication reliability between the control unit and the sensor is high, the gas replenishment strategy is still executed according to the dynamic threshold of gas pressure provided in the above embodiment. When the reliability is low, it is switched to a fixed threshold, which is an alternative solution to a certain extent to ensure the reliability of the system.

[0059] As an example, the communication reliability index Rc may be calculated as follows.

[0060] For example, when the communication module of the SF6 device uses LoRa, 5G, and Bluetooth. For each communication channel (LoRa / Bluetooth / 5G), calculate the normalized health H i (0~1). i =w L *f L (L i )+w D *f D (D i )+w S *f S (S i ).

[0061] Where w is the weight coefficient, which are the packet loss rate weight w L , delay weight w D and signal strength weight w S , must satisfy w L +w D +w S =1. For example, for 5G, the three can be 0.4, 0.4, 0.2 respectively. For Lora, they can be 0.3, 0.2, 0.5. For Bluetooth, they can be 0.1, 0.3, 0.6.

[0062] The three normalization functions f are: f L (L i )=max(0,1-(L i / L max )), where L i is the actual measured packet loss rate, L max is the upper limit of the packet loss rate. As an example, the upper limits of the packet loss rates of all communication modules can be set to 20%.

[0063] f D (D i ) = max (0, 1-(D i -D min / D max -D min )) Among them, D i is the actual measured delay, D max and D minare the upper and lower limits of delay respectively. As an example, the D min Can be 0ms, D max 5G can be set to 50ms, LORA can be set to 1000ms, and Bluetooth can be set to 100ms.

[0064] f s (S i )=1 / (1+e -k(Si-Sth) ), where k=0.1, S th =-100dBm, S i is the actual measured signal strength.

[0065] According to the above formula, if the packet loss rate of the 5G communication channel is 15%, the delay rate is 60ms, and the signal strength is -105dBm, the packet loss rate of the LORA communication channel is 8%, the delay rate is 800ms, and the signal strength is -115dBm, and the packet loss rate of the Bluetooth communication channel is 5%, the delay rate is 50ms, and the signal strength is -75dBm. Substituting into the above formula, we can get: 5G channel communication health H1=0.4×(1-15 / 20)+0.4×(1-60 / 50)+0.2×1 / (1+e -0.1(-105+100) )=0.176.

[0066] LORA channel communication health H2 = 0.3 × (1-8 / 20) + 0.2 × (1-800 / 1000) + 0.5 × 1 / (1 + e -0.1(-115+100) )=0.255.

[0067] Bluetooth channel communication health H3 = 0.1 × (1-5 / 20) + 0.3 × (1-50 / 100) + 0.6 × 1 / (1 + e -0.1(-75+100) )=0.844.

[0068] Communication reliability index Rc=w1*H1+w2*H2+w3*H3, where w1, w2, and w3 are weight values ​​of the quality of the three communication channels, and the sum of the three is 1, for example, 0.3, 0.6, and 0.1 respectively. According to the above calculation, Rc=0.29. Assuming that the first threshold mentioned above is set to 1 and the second threshold is set to 0.5, then Rc is less than the second threshold at this time, and the control unit switches the gas pressure dynamic threshold to a fixed threshold, and uses the fixed threshold as the basis for judging whether there is leakage and replenishing gas.

[0069] The embodiment of the present invention also provides an intelligent gas supplement control method, which is implemented using an intelligent gas supplement system of SF6 gas insulation equipment, combined with Figure 2 , the control method comprises: S1, receiving the gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor; S2, dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage based at least on the gas pressure value and the gas temperature value; S3, compare the gas pressure value measured in real time with the current corresponding gas pressure dynamic threshold value. If the gas pressure value measured in real time is less than or equal to the current corresponding gas pressure dynamic threshold value, control the solenoid valve to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

[0070] It should be understood that the control method of the embodiment of the present invention can refer to the introduction and setting of the corresponding part of the above embodiment, that is, the steps performed by the control unit of the above embodiment can be understood as the control method steps in this implementation, and will not be repeated here.

[0071] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An intelligent gas supply system for SF6 gas-insulated equipment, characterized in that: It includes SF6 gas source equipment, SF6 gas insulation equipment, a gas pressure sensor for monitoring the gas pressure value of the SF6 gas insulation equipment, a gas temperature sensor for monitoring the gas temperature value of the SF6 gas insulation equipment, and a gas supply control module; The gas replenishment control module includes a control unit and a solenoid valve connected to the control unit by signal; the air inlet end of the solenoid valve is connected to the air outlet end of the SF6 gas source equipment through a pipeline, and the air outlet end of the solenoid valve is connected to the air inlet end of the SF6 gas insulation equipment through a pipeline; the control module is also connected to the gas pressure sensor and the gas temperature sensor by signal; The control module is used for: receiving a gas pressure value collected by the gas pressure sensor and a gas temperature value collected by the gas temperature sensor; Dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value; Compare the gas pressure value measured in real time with the current corresponding gas pressure dynamic threshold. If the gas pressure value measured in real time is less than or equal to the current corresponding gas pressure dynamic threshold, control the solenoid valve to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

2. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 1 is characterized in that: The dynamically calculating the gas pressure dynamic threshold value for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculating the SF6 gas pressure change rate, and multiplying the SF6 gas pressure change rate by a pre-calibrated pressure change rate gain coefficient to obtain a pressure correction term; Calculating the temperature difference between the gas temperature value received in real time and a preset temperature reference value, and multiplying the temperature difference by a pre-calibrated temperature compensation coefficient to obtain a temperature compensation term; The preset gas pressure reference threshold for determining SF6 gas leakage is corrected at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold.

3. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 2 is characterized in that: The method of correcting a preset gas pressure reference threshold for determining SF6 gas leakage at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold specifically includes: calculating the gas pressure dynamic threshold according to the following formula: P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref), wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, and T_ref is the temperature reference value.

4. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 2 is characterized in that: The dynamic calculation of the gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculate the square of the load current of the SF6 gas-insulated equipment, and then multiply it by a pre-calibrated load current compensation coefficient to obtain a load current compensation term; The step of correcting the preset gas pressure reference threshold for determining SF6 gas leakage at least according to the pressure correction term and the temperature compensation term to obtain the gas pressure dynamic threshold specifically includes: According to the pressure correction item, the temperature compensation item, and the load current compensation item, the preset gas pressure reference threshold for determining SF6 gas leakage is corrected to obtain the gas pressure dynamic threshold.

5. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 4 is characterized in that: The method of correcting a preset gas pressure reference threshold for judging SF6 gas leakage according to the pressure correction term, the temperature compensation term, and the load current compensation term to obtain the gas pressure dynamic threshold specifically includes: according to the following formula: P_adj = P_set × (1 + k|dP / dt|) + α(T - T_ref)+β*I 2 , the gas pressure dynamic threshold is calculated, wherein P_adj is the gas pressure dynamic threshold; P_set is the gas pressure reference threshold, k is the pressure change rate gain coefficient, α is the temperature compensation coefficient, dP / dt is the SF6 gas pressure change rate, T is the gas temperature value received in real time, T_ref is the temperature reference value, I is the load current, and β is the load current compensation coefficient.

6. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 1, characterized in that: The dynamically calculating the gas pressure dynamic threshold value for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value specifically includes: Calculating the SF6 gas pressure change rate, and multiplying the SF6 gas pressure change rate by a pre-calibrated pressure change rate gain coefficient to obtain a pressure correction term; Determine whether the gas temperature value received in real time is greater than a preset temperature reference value. If so, confirm that the SF6 gas insulation equipment is in a high temperature zone, and calculate the high temperature zone temperature compensation term according to the high temperature zone compensation equation, and then correct the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the high temperature zone temperature compensation term to obtain the gas pressure dynamic threshold; if not, confirm that the SF6 gas insulation equipment is in a low temperature zone, and calculate the low temperature zone temperature compensation term according to the low temperature zone compensation equation, and then correct the preset gas pressure reference threshold for judging SF6 gas leakage at least according to the pressure correction term and the low temperature zone temperature compensation term to obtain the gas pressure dynamic threshold.

7. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 6, characterized in that: The calculation of the high temperature zone temperature compensation term according to the high temperature zone compensation equation includes: calculating the high temperature zone temperature compensation term according to the following formula: α(T 高 )=α1(T-T_ref)+α2(T-T_ref) 2 ; The calculation of the low temperature zone temperature compensation term according to the low temperature zone compensation equation includes: calculating the low temperature zone temperature compensation term according to the following formula: α (T 低 )=α3ln(T / T_ref); where, α(T 高 ) is the temperature compensation term in the high temperature zone; α(T 低 ) is the temperature compensation item of the low temperature zone; α1 and α2 are the temperature compensation coefficients of the high temperature zone, and α3 is the temperature compensation coefficient of the low temperature zone; T is the gas temperature value received in real time, and T_ref is the preset temperature reference value.

8. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 1, characterized in that: The control module is also used to: generate simulation data of the operating conditions of the SF6 gas insulation equipment, input the simulation data into the LSTM neural network model for pre-training, the output layer is used to predict the SF6 gas pressure at future time nodes, and perform framework adjustment to obtain the SF6 gas pressure prediction model based on LSTM; The gas pressure value collected by the gas pressure sensor and the gas temperature value collected by the gas temperature sensor are input into the SF6 gas pressure prediction model to obtain the SF6 gas pressure prediction value at a future time node, and the SF6 gas pressure prediction value at the future time node is compared with the gas pressure dynamic threshold value, and it is determined whether to start the solenoid valve for gas replenishment according to the comparison result.

9. The intelligent gas supply system for SF6 gas-insulated equipment according to claim 1, characterized in that: The gas pressure sensor and the gas temperature sensor both communicate with the control unit via wireless signals. The control unit is also used to: generate a communication reliability index based on the wireless communication signal packet loss rate, delay and signal strength; if the communication reliability index is less than or equal to a preset first threshold and greater than or equal to a preset second threshold, execute a gas replenishment strategy based on the gas pressure dynamic threshold; if the communication reliability index is less than the second threshold, switch the gas pressure dynamic threshold to a fixed gas pressure threshold, and execute a gas replenishment strategy based on the fixed gas pressure threshold.

10. An intelligent gas replenishment control method, characterized in that: The intelligent gas supply system for SF6 gas-insulated equipment according to any one of claims 1 to 9 is used to implement the control method, which includes: receiving a gas pressure value collected by the gas pressure sensor and a gas temperature value collected by the gas temperature sensor; Dynamically calculating a gas pressure dynamic threshold for determining SF6 gas leakage at least based on the gas pressure value and the gas temperature value; Compare the gas pressure value measured in real time with the current corresponding gas pressure dynamic threshold. If the gas pressure value measured in real time is less than or equal to the current corresponding gas pressure dynamic threshold, control the solenoid valve to open so that the SF6 gas source equipment supplies gas to the SF6 gas insulation equipment.

Images