Gas monitoring systems and respective methods

Abstract

The present invention relates to a gas monitoring system, comprising: A gas-insulated switchgear (1) having a chamber (4) filled with an insulating gas (5) surrounding a high-voltage or medium-voltage component (6); a sensor (2) operatively connected to said chambers (4) and adapted to measure a physical property of said insulating gas (5) in each of said chambers (4) over time; a computer unit (3) adapted to perform statistical step detection on the sensor measurements to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred, The statistical steps and / or change detection may include identifying an expected range based on past sensor measurements and determining if a current sensor measurement is outside of the expected range.

Description

[Technical field]

[0001] Technical Field The present invention relates to a gas monitoring system comprising a gas-insulated switchgear having a chamber filled with an insulating gas surrounding a high or medium voltage component and a sensor operatively connected to the chamber and adapted to measure a physical property of the insulating gas in the respective chamber over time. The present invention further relates to a respective method for a gas-insulated switchgear comprising a chamber filled with an insulating gas surrounding a high or medium voltage component, the method comprising the step of measuring a physical property of the insulating gas in the respective chamber over time. [Background technology]

[0002] Background technology Gas-insulated switchgear (GIS) typically operates at medium voltages of 10 kV to 100 kV and / or high voltages of 100 kV to 1200 kV. To insulate high or medium voltage components such as disconnectors, fuses and / or circuit breakers, the GIS is filled with an insulating gas such as air or SF6. Other gases are also used and are currently being developed by several manufacturers. However, SF6 has traditionally been the gas of choice for high voltage applications due to a variety of advantages. The symmetrical arrangement of the molecules results in extreme stability and very high insulating capacity. SF6 therefore has approximately three times the dielectric strength of air at atmospheric pressure. As a result, GIS using SF6 as the insulating gas can be significantly smaller than equipment using air as the insulating medium. Furthermore, SF6 is a "self-healing" insulator in that it is not significantly damaged by breakdown. This makes it very suitable as an interrupting medium.

[0003] On the other hand, SF6 is also recognized as a very strong greenhouse gas. The USEPA reports that SF6 has an atmospheric lifetime of about 3200 years and a global warming potential (100-year value) 23,900 times that of CO2. Therefore, the use of SF6 is strictly controlled and regulated. Manufacturing defects within GIS can result in SF6 leakage, which impacts the environment and at the same time reduces the insulating performance of the GIS. Large amounts of leaked insulating gas can have potentially devastating effects through uncontrolled electrical discharge.

[0004] EP3790032A1 describes a method for monitoring a circuit breaker containing a gas having at least one gas property providing a numerical value comprising the steps of: a) collecting a data set referring to the at least one gas property in the circuit breaker, the data set comprising numerical values ​​of the at least one gas property for particular conditions or during a particular time of day, the method further comprising: b) calculating the standard deviation of the at least one gas property of the data set for at least three of the last ten days; c) comparing the standard deviation of the gas pressure with a predefined threshold; and d) taking initial action if the standard deviation exceeds the threshold.

[0005] To avoid such problems, technical gas monitoring systems for gas-insulated switchgear have typically been developed for monitoring the system concentration and / or pressure. The leakage rate is calculated based on these measurements. To increase the accuracy of this leakage rate calculation, the system can take into account measurements and / or data from, for example, the last four weeks. Especially for gas compartments in voltage transformers, if a gas handling operation occurs or if the load changes, a rapid change in the initially measured gas pressure and / or concentration results. Such a rapid change in pressure can result in an erroneously calculated leakage rate that causes undesired alarms and warnings, which is not favorable for the continuous operation of the GIS. Summary of the Invention [Problem to be solved by the invention]

[0006] It is therefore an object of the present invention to provide an improved gas monitoring system and respective method for preventing such "false alarms". [Means for solving the problem]

[0007] The object of the invention is solved by the features of the independent claims. Preferred embodiments are detailed in the dependent claims.

[0008] Therefore, the object is solved by a gas monitoring system, said gas monitoring system comprising: a gas-insulated switchgear having a chamber filled with an insulating gas surrounding a high-voltage or medium-voltage component; a sensor operatively connected to the chambers and adapted to measure a physical property of the insulating gas in each of the chambers over time; and a computer unit adapted to perform statistical step detection on the sensor measurements to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred.

[0009] Therefore, the key to the proposed solution is to automatically detect any other sudden changes in gas handling or physical properties, such as, for example, the pressure and / or concentration values ​​of the insulating gas. In such a case, it can be determined in turn whether the leakage rate calculation is restarted or whether the jump and / or change constitutes a "false alarm" that can be removed from the measurement data, for example in the case of voltage transformers as high or medium voltage components, as described below. In such a way, undesired alarms and warnings are significantly reduced or even eliminated.

[0010] Therefore, in contrast to prior art systems, it is not necessarily required to specifically manually restart the leakage rate and / or trend calculation algorithms, e.g. HAPG and competition, after a gas handling operation, e.g. refilling the insulating gas. Similar problems can occur in gas compartments of voltage transformers, where the gas pressure and / or concentration strongly depends on the load of the voltage transformer. So, if a line with a voltage transformer is switched on or off, the pressure and / or concentration change with a large increase and / or decrease, i.e. jumps and changes, respectively. Such jumps have the same effect as refills and can cause false alarms, which are avoided in a very convenient way by the proposed solution.

[0011] In other words, the proposed solution essentially adds detection of jumps and / or changes, particularly occurring prior to the actual leak rate and / or trend calculation that follows. The proposed jump and / or change detection may look at past measurements to identify if a current measurement constitutes a jump and / or change. The jump and / or change detection may identify an expected range based on past measurements and check if a current, particularly the most recent, measurement is outside of the expected range. In particular, the statistical step and / or change detection may include identifying an expected range based on past sensor measurements and checking if a current sensor measurement is outside of the expected range. If so, the jump and / or change can be detected and eliminated by adding an offset to the measurement, as described below.

[0012] In prior art embodiments, problems with leak rate calculations have been observed in the real world after jumps due to gas handling and in gas compartments of voltage transformers, where pressure and / or concentration strongly depend on the load of the line, especially on energization. In the proposed solution, instead of new data points, several recent data points can be used to detect jumps and / or changes. For this, the detection thresholds of jumps and / or steps can be calculated using past measurements or given by predefined configurations. Usually, in all compartments except voltage transformers, a jump up indicates a refill. A jump down indicates a large gas leak or discharge. Therefore, a jump down should be able to trigger a leak warning. In the case of voltage transformers, a jump up can be a refill or energization event. A jump down can be caused by a large leak or de-energization. A jump down due to de-energization should not trigger a leak warning. The proposed solution provides for the detection of these pressure drop events to provide false alarms.

[0013] The gas insulated switchgear (GIS) is preferably adapted to operate under medium voltages of 10 kV to 100 kV and / or high voltages of 100 kV to 1200 kV. The chamber is preferably provided gas tight and / or with a metal housing enclosing the medium or high voltage components. The medium or high voltage components are exemplarily provided as circuit breakers, fuses, disconnectors, etc. The physical properties measured may include pressure, concentration, sound speed, diffraction, etc. While there may be a single sensor, there may also be multiple sensors, measuring the same or different physical properties. The word over time preferably means that the sensor measures continuously, e.g. over one day, two days or more. The computer unit may be provided as any computerized means, e.g. a microprocessor.

[0014] The statistical step detection, also known as step smoothing, step filtering, shift detection, jump or edge detection, is to be understood as a process of finding abrupt changes in the average level of the measurement values. The statistical step detection is preferably performed as the measurement data arrives, preferably by using so-called online algorithms. The statistical change detection preferably involves identifying when the probability distribution of a stochastic process or a time series changes.

[0015] According to a preferred embodiment, the computer unit detects if a jump and / or change occurs. a) resetting the leakage rate calculation of the insulating gas in the chamber; or b) removing said jumps and / or changes by applying an offset to said sensor measurements; It is adapted to:

[0016] Thus, if jumps and / or changes are detected, the leak rate and / or trend calculations can be automatically reset and reinitialized, respectively. For example, refilling does not normally decrease the physical property, e.g., gas pressure, so in such cases, the calculations can be limited to jumps and / or changes that increase the physical property. Alternatively, in cases where the physical property, e.g., pressure and / or concentration, depends on e.g., line load or other parameters, e.g., gas compartment in a voltage transformer, the jumps and / or any changes can be removed from the measurements. In such cases, an offset can be applied, which specifically removes the jumps and / or changes before the leak rate calculation is applied. The offset is preferably equal to a value that essentially equalizes the jumps and / or changes in the measurements. Thus, the leak rate calculations can continue without the need to reinitialize and stop the leak rate calculation algorithm, respectively.

[0017] In a further preferred embodiment, the computer unit comprises: c) if a jump is determined, determining whether a) or b) is true depending on the height of said jump and / or the direction of said jump and / or on further measurement signals indicating the state of said gas-insulated switchgear; It is adapted to:

[0018] Especially for embodiments with a gas compartment where b) is possible, e.g. a voltage transformer, it is advantageous to decide whether to select option a) or b). The decision can be made based on at least one of the parameters, e.g. the height of the jump, the direction of the jump, e.g. the pressure change based on the energization in the voltage transformer is repeated, after a jump up the next jump is a jump down, and an additional measurement signal representing the state, e.g. the voltage of the voltage transformer, in analog or binary values.

[0019] In another preferred embodiment, the computer unit comprises: d) calculating the leakage rate of said insulating gas within said chamber; It is adapted to:

[0020] Such calculation is preferably performed after the determination of a) and b). The calculation of the leakage rate of the insulating gas may include calculation of a future predicted leakage rate. Such prediction allows maintaining a tight inventory of gases that are regulated by environmental standards or are very expensive. Furthermore, the operator of the GIS may identify future times when the operation of the GIS may become unsafe and the GIS will require repairs. Leakage of the insulating gas, e.g. SF6, may impact the environment and at the same time degrade the insulating performance of the GIS. Thus, a significant loss of the insulating gas may result in potentially catastrophic effects due to uncontrolled discharge.

[0021] According to a further preferred embodiment, jumps and / or changes are characterized by a measurement value being greater than or equal to 1.5, 2.0, 2.5, 5.0 or 10.0 of the standard deviation of the measurements of the last hour, last two hours, last day or last two days. Alternatively, the standard deviation of the measurements of the last 12 hours, last 18 hours, last 5 days, last 10 days can be taken into account. Thus, precise measurement rules are defined that determine if a jump and / or change has occurred.

[0022] In another preferred embodiment, the statistical change detection comprises sequentially discounting autoregression time series modelling (SDAR modelling). Such autoregressive modelling means preferably constructing a time series model for predicting a current measurement from past measurements in the same univariate time series. Alternatively, other statistical change detection algorithms known to those skilled in the art can be used.

[0023] According to a further preferred embodiment, the physical property of the insulating gas is indicative of a concentration of the insulating gas in the chamber.Thus, the sensor is preferably provided as a concentration measuring sensor.

[0024] In a further preferred embodiment, the gas-insulated switchgear comprises a voltage transformer. A voltage transformer (VT), also called a potential transformer (PT), is commonly understood as a parallel-connected type of instrument transformer. A voltage transformer is typically designed to provide a weak load to the source being measured and to have a precise voltage ratio and phase relationship to enable accurate secondary connection metering.

[0025] Generally, there are various options for the insulating gas. However, according to a preferred embodiment, the insulating gas is SF6. Sulfur hexafluoride (SF6) is traditionally the insulating gas of choice for high voltage applications. The symmetrical arrangement of the molecules results in extreme stability and very high insulating capacity. SF6 has approximately three times the dielectric strength of air at atmospheric pressure. As a result, GIS using SF6 as insulating gas can be significantly smaller than equipment using air as the insulating medium. Furthermore, SF6 is a "self-healing" insulator in that it is not significantly damaged by breakdown. This makes SF6 very suitable as an interrupting medium. In addition to SF6, other gases are being developed as more climate-friendly alternatives to SF6, which can equally be used in the proposed solution.

[0026] The object is completely solved by a method for detecting jumps and / or changes in a gas-insulated switchgear comprising a chamber filled with insulating gas surrounding a high-voltage or medium-voltage component, said method comprising the steps of: measuring a physical property of the insulating gas in each of the chambers over time; performing statistical step detection to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred in the sensor measurements; Equipped with.

[0027] According to a preferred embodiment, if a jump and / or change occurs, the method further comprises: a) resetting a leakage rate calculation of the insulating gas in the chamber; or b) removing said jumps and / or changes by applying an offset to said sensor measurements. Equipped with.

[0028] In another preferred embodiment, the method further comprises: c) if a jump is determined, deciding between a) or b) depending on further measurement signals indicating the height of the jump and / or the direction of the jump and / or the state of the gas-insulated switchgear. Equipped with.

[0029] According to a further preferred embodiment, the method comprises d) calculating the leakage rate of the insulating gas within the chamber. Equipped with.

[0030] In other preferred embodiments, jumps and / or changes are characterized by measurements that are greater than or equal to 1.5, 2.0 or 2.5 standard deviations of measurements over the last hour, last two hours, last day or last two days.

[0031] According to a further preferred embodiment, the statistical change detection comprises sequentially discounting autoregression time series modelling (SDAR modelling).

[0032] According to another preferred embodiment, the physical property of the insulating gas is indicative of a concentration of the insulating gas in the chamber.

[0033] Further implementations and advantages of said method are directly and clearly derived by the skilled person from the system as described above.

[0034] BRIEF DESCRIPTION OF THE DRAWINGS These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. [Brief description of the drawings]

[0035] [Figure 1] 1 illustrates a schematic diagram of a gas monitoring system according to an exemplary embodiment; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0036] Implementation Description FIG. 1 illustrates a physical diagram of a gas monitoring system according to an exemplary embodiment.

[0037] The gas monitoring system comprises a gas-insulated switchgear 1, a sensor 2 and a computer unit 3. The gas-insulated switchgear 1 comprises a tubular gas-tight chamber 4, which is filled with an insulating gas 5 surrounding high or medium voltage components 6. The insulating gas 5 is provided as sulfur hexafluoride (SF6), while other insulating gases with similar insulating capabilities may be used as well.

[0038] The high or medium voltage components 6 are provided as voltage transformers. The sensors 2 are provided as pressure sensors, operatively connected to the chambers 4 and measuring pressure over time as a physical property of the insulating gas 5 in each chamber 4. Although Figure 1 shows only one chamber 4, the gas-insulated switchgear 1 may comprise three or more chambers 4, one for each phase, each filled with an insulating gas 5 surrounding a respective high or medium voltage component 6, and each fitted with a respective sensor 2.

[0039] The computer unit 3 comprises a microprocessor or similar computing means and is configured to perform statistical step detection on the sensor measurements to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred, such that the computer unit 3 determines that a jump and / or change has occurred if the most recent measurement is greater than or equal to 1.5, 2.0 or 2.5 of the standard deviation of the measurements over the last hour, last two hours, last day or last two days.

[0040] In such a case, the computer unit 3 decides whether a) an already performed leak rate calculation of the insulating gas 5 in the chamber 4 is reset and reinitialized, respectively, or b) the detected jumps and / or changes are removed by applying an offset, respectively, to the sensor measurements and subsequently continuing the leak rate calculation, so that in the latter case an offset is applied before the leak rate calculation is performed in order to nullify the jumps and / or changes.

[0041] Step b) is possible for gas compartments that comprise a voltage transformer, whereby a decision can be made based on the height of the jumps and / or changes, the direction of the jumps and / or changes, e.g. the pressure change based on energization in the voltage transformer is repeated, the next jump after a rise is a fall, and also based on an additional measurement signal indicating the state, e.g. the voltage of the voltage transformer, as an analog or binary value.

[0042] Once the offset is applied, the leak rate of the insulating gas 5 in the chamber 4 can be continuously calculated. The size of the offset is selected to even out jumps and / or changes in the measurement data. Statistical step detection to determine if a jump occurs and / or statistical change detection to determine if a change occurs may include sequentially discounting autoregression time series modelling (SDAR modelling).

[0043] While the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustrations and descriptions are to be considered as illustrative or exemplary and not restrictive. The present invention is not limited to the embodiments not disclosed. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the term "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be interpreted as limiting the scope. [Explanation of symbols]

[0044] List of References 1 Gas insulated switchgear 2 Sensors 3 Computer Unit 4. Chamber 5. Insulating gas 6 Medium voltage components

Claims

1. 1. A gas monitoring system comprising: A gas-insulated switchgear (1) having a chamber (4) filled with an insulating gas (5) surrounding a high-voltage or medium-voltage component (6); a sensor (2) operatively connected to the chamber (4) and adapted to measure a physical property of the insulating gas (5) within the chamber (4) over time as a sensor measurement; a computer unit (3) adapted to perform statistical step detection on the sensor measurements to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred, The gas monitoring system, wherein the statistical steps and / or change detection include identifying an expected range based on past sensor measurements and determining if a current sensor measurement is outside of the expected range.

2. The computer unit (3) resets the leakage rate calculation of the insulating gas (5) in the chamber (4) if a jump and / or change occurs, or 2. The gas monitoring system of claim 1, wherein b) the gas monitoring system is adapted to remove said jumps and / or changes by applying an offset to the sensor measurements.

3. The computer unit (3) c) if a jump is determined, determining either a) or b) depending on the height of the jump and / or on the direction of the jump and / or on further measurement signals indicative of the state of the gas-insulated switchgear, and / or said computer unit (3) d) Calculate the leakage rate of the insulating gas (5) in the chamber (4).

3. The gas monitoring system of claim 2 adapted to:

4. 4. The gas monitoring system of claim 1, wherein the jumps and / or changes are characterized by measurements that are greater than or equal to 1.5, 2.0 or 2.5 standard deviations of measurements over the last hour, last two hours, last day or last two days.

5. The gas monitoring system of any preceding claim, wherein the statistical change detection comprises SDAR modeling.

6. The gas monitoring system of any one of claims 1 to 3, wherein the physical property of the insulating gas (5) is indicative of a concentration of the insulating gas (5) in the chamber (4).

7. The gas monitoring system according to any one of claims 1 to 3, wherein the gas insulated switchgear comprises a voltage transformer.

8. The gas monitoring system according to any one of claims 1 to 3, wherein the insulating gas (5) is SF6.

9. A method for detecting jumps and / or changes in a gas-insulated switchgear (1) comprising a chamber (4) filled with insulating gas (5) surrounding a high or medium voltage component (6), said method comprising the steps of: measuring a physical property of the insulating gas (5) in each of the chambers (4) over time as a sensor measurement; performing statistical step detection to determine if a jump has occurred and / or statistical change detection to determine if a change has occurred in the sensor measurements; The method, wherein the statistical steps and / or change detection include identifying an expected range based on past sensor measurements and determining if a current sensor measurement is outside of the expected range.

10. a) resetting the leakage rate calculation of the insulating gas (5) in the chamber (4), or The method of claim 9, comprising the step of: b) removing said jumps and / or changes by applying an offset to said sensor measurements.

11. 11. The method according to claim 10, comprising the step of: c) if a jump is determined, determining either a) or b) depending on further measurement signals indicative of the height of the jump and / or the direction of the jump and / or the state of the gas-insulated switchgear.

12. The method according to any one of claims 9 to 11, comprising the step of calculating the leakage rate of the insulating gas (5) in the chamber (4).

13. 12. The method of any one of claims 9 to 11, wherein the jumps and / or changes are characterized by measurements that are greater than or equal to 1.5, 2.0 or 2.5 standard deviations of measurements over the last hour, last two hours, last day or last two days.

14. The method of any one of claims 9 to 11, wherein the statistical change detection comprises SDAR modelling.

15. The method according to any one of claims 9 to 11, wherein the physical property of the insulating gas (5) is indicative of a concentration of the insulating gas (5) in the chamber (4).

Images