Method for detecting an electric arc in a resistive heating system

Abstract

The present description relates to a method for detecting an electric arc in an electric heating system comprising at least one heating electric resistance, said method comprising a step of detecting that the number of sign changes of the filtered time derivative of the electric power supplied to the heating electric resistance is greater than a first threshold.

Description

DESCRIPTION TITLE: METHOD FOR DETECTING AN ELECTRICAL ARC IN A RESISTIVE HEATING SYSTEM This patent application claims priority from French patent application FR23 / 06182, which will be considered as forming an integral part of this description. technical field

[0001] This description relates in general to methods of detecting an electric arc in a resistive heating system, particularly for a furnace. Prior art

[0002] A controlled crystallization furnace for semiconductor materials, particularly silicon used in photovoltaic applications, comprises a crucible containing the semiconductor material and surrounded by heating elements. The flow of current through each heating element results in the emission of heat via the Joule effect.

[0003] Depending on the intended application, the current flowing through each heating element can be high. There is a risk of arcing between a heating element and another component of the oven. It is advisable to detect the occurrence of an arc in order to cut off the power supply to the heating element and limit damage to the heating element and / or other oven components.

[0004] When electric heating elements are powered by a three-phase power supply, arc detection can be achieved by comparing the three phases of the current. However, for some applications, it is desirable that the Each electric heating element is powered by a single-phase current. Conventional methods for detecting an electric arc cannot then be used. Summary of the invention

[0005] One embodiment overcomes all or part of the drawbacks of known methods for detecting an electric arc in a resistive heating system.

[0006] According to one embodiment, the detection method is compatible with a single-phase current supply to the heating system.

[0007] According to one embodiment, the process allows for rapid detection of the electric arc from the beginning of its formation.

[0008] One embodiment provides a method for detecting an electric arc in an electric heating system comprising at least one electric heating resistance, including determining that the number of sign changes of the filtered time derivative of the electrical power supplied to the electric heating resistance is greater than a first threshold.

[0009] According to one embodiment, the method includes interrupting the power supply to the electric heating resistance when the number of sign changes of the filtered time derivative of the electrical power supplied to the electric heating resistance is greater than the first threshold or greater than a second threshold, strictly greater than the first threshold.

[0010] According to one embodiment, the method includes determining the time derivative of the electrical power supplied to the heating electrical resistance and determining the time derivative filtered by filtering the time derivative.

[0011] According to one embodiment, the method includes determining values ​​of the electrical power supplied to the heating electrical resistance and determining a new value of the time derivative of the electrical power supplied to the heating electrical resistance from the last two determined values ​​of the electrical power supplied to the heating electrical resistance.

[0012] According to one embodiment, the method includes determining a new value of the filtered time derivative equal to the last determined value of the time derivative when the last determined value of the time derivative is greater than a derivative value threshold, and equal to 0 when the last determined value of the time derivative is less than the derivative value threshold.

[0013] According to one embodiment, the method includes determining a new value of a filtered and corrected time derivative equal to the last determined value of the filtered time derivative when the last determined value of the filtered time derivative is different from 0, and equal to the last determined value of the filtered and corrected time derivative when the last determined value of the filtered time derivative is equal to 0.

[0014] According to one embodiment, the method comprises determining a logic value at a first logic state when the product of the last two determined values ​​of the filtered and corrected time derivative is strictly negative and at a second logic state, different from the first logic state, when the product of the last two determined values ​​of the filtered and corrected time derivative is positive or zero.

[0015] According to one embodiment, the method includes determining a new value of a meter equal to the last determined value of the meter at which it is added a first strictly positive increment when the last determined logic value is in the first logic state, optionally a second strictly positive increment strictly less than the first increment when the last determined logic value is in the second logic state and the last determined value of the filtered time derivative is not zero, or a third strictly negative increment when the last determined logic value is in the second logic state and the last determined value of the filtered time derivative is zero, and including the detection of an electric arc when the determined value of the counter is greater than the first threshold.

[0016] One embodiment also provides a system for detecting an electric arc in an electric heating system comprising at least one electric heating resistance configured to determine whether the number of sign changes of the filtered time derivative of the electrical power supplied to the electric heating resistance is greater than a first threshold.

[0017] One embodiment also provides for an oven comprising an electric heating system including at least one electric heating element and a system for detecting an electric arc in the electric heating system as defined above.

[0018] According to one embodiment, the oven includes a system for supplying the electric heating element with a single-phase current. Brief description of the drawings

[0019] These features and advantages, along with others, will be detailed in the following description of modes of specific implementations carried out, by way of non-exhaustive list, in relation to the attached figures, including:

[0020] Figure 1 represents, in a partial and schematic way, one method of implementing a crystallization furnace;

[0021] Figure 2 represents, in the form of a block diagram, one embodiment of a method for detecting an electric arc;

[0022] Figure 3 represents a curve showing the evolution of the electrical power supplying a heating electrical resistance of the oven in Figure 1 as a function of time;

[0023] Figure 4 represents a curve showing the evolution of the time derivative of the electrical power of Figure 3 as a function of time;

[0024] Figure 5 represents a curve showing the evolution of the time derivative of the electrical power in Figure 3 as a function of time after filtering; and

[0025] Figure 6 represents a curve showing the evolution of a counter used during the implementation of the embodiment of the process illustrated in Figure 2. Description of the embodiments

[0026] The same elements have been designated by the same references in the different figures. In particular, the structural and / or functional elements common to the different embodiments may have the same references and may have identical structural, dimensional and material properties.

[0027] For the sake of clarity, only the steps and elements necessary for understanding the described implementation methods have been shown and detailed. In particular, the The structure of a furnace including an electric heating system is known but not described in detail.

[0028] In the description that follows, when referring to absolute positional qualifiers, such as the terms "front", "back", "top", "bottom", "left", "right", etc., or relative positional qualifiers, such as the terms "above", "below", "superior", "inferior", etc., or to orientational qualifiers, such as the terms "horizontal", "vertical", etc., unless otherwise specified, reference is made to the orientation of the figures or to an oven in a normal operating position.

[0029] Unless otherwise specified, the expressions "approximately", "roughly", and "on the order of" mean within 10%, preferably within 5%.

[0030] Figure 1 shows, in part and schematically, an embodiment of a crystallization furnace 10. The furnace 10 comprises an enclosure 12 in which a crucible 14 is placed. The enclosure 12 may be a water-cooled enclosure in which a controlled atmosphere can be maintained. The crucible 14 may be placed in a support 16 that is a good conductor of heat. The furnace 10 further comprises electric heating elements 18 surrounding the crucible 14, in particular above the crucible 14, below the crucible 14, and around the side wall of the crucible 14. Thermally insulating walls 20 delimit, within the enclosure 12, a heating zone 22 containing the crucible 14 and the heating elements 18.

[0031] The oven 10 includes a power supply system 24 for the heating elements 18. The power supply system 24 includes, in particular, electrical transformers connected to the mains and switches for turning the power supply to each heating element 18 on or off. independently of each other. The power supply system 24 further includes sensors, in particular current intensity sensors and voltage sensors, enabling the determination of the instantaneous electrical power supplied to each heating element 18. The oven 10 further includes a processing module 26 configured to control the operation of the oven 10. The processing module 26 includes, for example, a processor. The processing module 26 is connected to the power supply system 24 and is configured, in particular, to control the activation or interruption of the power supply to each heating element 18 by the power supply system 24.

[0032] During operation, an electric arc can form between the heating elements 18 and components of the furnace 10 that are normally connected to a low reference potential, such as ground. This can occur, for example, when there is insufficient space between a heating element 18 and the thermally insulated walls 20, when there is contact between two components of the heating element 18, or when the charge in the crucible 15 is incorrectly positioned in the furnace and comes close to one of the heating elements 18. When an electric arc forms, it can cause damage to parts of the furnace 10. Furthermore, if the electric arc becomes self-sustaining, the damage to the furnace 10 can be significant, potentially compromising its safe operation.

[0033] There is therefore a need to quickly detect the formation of an electric arc, in particular to interrupt the power supply to the heating elements.

[0034] According to one embodiment, the 24 power supply system is configured to supply each heating element 18 with a single-phase current. According to one embodiment, the power supply system 24 includes a single-phase output electrical transformer connected to each heating element 18. A single-phase output transformer advantageously offers reduced size and cost compared to a two-phase or three-phase output transformer.

[0035] According to one embodiment, the processing module 26 is configured to determine the evolution of the electrical power supplied by the power supply system 24 to each heating resistance 18 as a function of time, in particular from the signals supplied by the sensors of the power supply system 24. The processing module 26 is further configured to determine the evolution of the time derivative of the electrical power supplied by the power supply system 24 to each heating resistance 18 as a function of time.

[0036] Figure 2 represents, in the form of a block diagram, an embodiment of a method for detecting an electric arc implemented by the processing module 26. According to one embodiment, the detection method is implemented by the processing module 26 for each heating resistance 18.

[0037] In step 30, the processing module 26 determines a new value Pn of the electrical power P supplied by the power supply system 24 to the heating element 18, where n is an integer greater than 1. According to one embodiment, the processing module 26 is connected to a voltage sensor from the power supply system 24 and to a current sensor from the power supply system 24, and determines successive power values. electrical P n at a sampling period T samp The sampling period T samp can be between 0.1 s and 10 s, and is, for example, equal to 1 s. The longer the sampling period T sampThe lower the value, the better the detection, but the greater the computing power of the processing module 26 must be. The process continues to step 32. On the first execution of step 30, the process proceeds to step 32 when two successive values ​​of the electrical power P are determined. On each subsequent execution of step 30, the process proceeds to step 32 as soon as a new value Pn of the electrical power P is determined.

[0038] At step 32, the processing module 26 determines a new value P' n of the time derivative P' of the electrical power supplied by the power supply system 24 to the heating element 18. According to one embodiment, the new value P' n the time derivative P' is determined from the last value P n of electrical power and the penultimate P n-i value of the electrical power determined in step 30. According to one embodiment, the new value P' n the time derivative P' is equal to the difference between the last value P n determined electrical power P and the penultimate value P n _! determined from the electrical power P divided by the sampling period T samp The process continues at step 34.

[0039] In step 34, the processing module 26 determines a filtered time derivative P'f obtained by filtering the time derivative P' of the electrical power supplied by the power supply system 24 to the heating element 18. According to one embodiment, the absolute value of the new value P' n the time derivative P' determined in step 32 is compared to a threshold TH_P' of derivative value. In the case where the absolute value of the new value P' nIf the time derivative P' is less than the threshold TH_P' of derivative value, the new value P' f n of the filtered time derivative P'f is equal to 0. In the case where the absolute value of the new value P' n If the time derivative P' is greater than the threshold TH_P' of derivative value, the new value P' f n the filtered time derivative P'f is equal to the new value P' n of the time derivative P' determined in step 32. In one embodiment, the threshold value of the derivative TH_P' is in the range of 0.01 kW / s to 100 kW / s, and is, for example, equal to 0.4 kW / s. In another embodiment, the threshold value of the derivative TH_P' can be parameterized, in particular, according to the furnace power. Step 34 advantageously allows for the exclusion of filtered time derivative values ​​P'f that correspond to normal variations in the electrical power P. The process continues in step 36.

[0040] In step 36, the processing module 26 determines a filtered and corrected time derivative P'f2 obtained from the filtered time derivative P'f of the electrical power supplied by the power supply system 24 to the heating element 18. According to one embodiment, if the new value P'f n of the filtered time derivative P'f is not equal to zero, then the new value P'f2 n the filtered and corrected time derivative P'f2 is equal to the new value P'f n of the filtered time derivative P'f determined in step 34. If the new value P'f n of the filtered time derivative P'f is equal to zero, then the new value P'f2 n the filtered and corrected time derivative P'f2 is equal to the previous value P'f n -i of the intermediate time derivative P'f2 determined in the previous execution of step 36. Step 36 prevents the new value P'f2 fromn of the filtered and corrected time derivative P'f2 is not equal to 0 as soon as a value determined previously of the filtered time derivative P'f is different from 0. The process continues at step 38.

[0041] In step 38, the processing module 26 determines whether the filtered time derivative P'f has changed sign. In one embodiment, the processing module 26 determines a new logical value Detectn from a logical data Detect. The processing module 26 sets the new logical value Detect n to the logical value "True" when the filtered time derivative P'f has changed sign and sets the logical data Detect n to the logical value "False" when the filtered time derivative P'f has not changed sign. According to one embodiment, if the product between the last value P'f2 n of the filtered and corrected time derivative P'f2 and the penultimate value P'f2 nIf the filtered and corrected time derivative P'f2 is positive, then processing module 26 sets the new logical value Detect n to the logical value "False", and, if the product enters the last value P'f2 n of the filtered and corrected time derivative P'f2 and the penultimate value P'f2 n The _! of the filtered and corrected time derivative P'f2 is negative, so the processing module 26 sets the new logical value Detect n to the logical value "True". The process continues at step 40.

[0042] At step 40, the processing module 26 determines a new CPT value n of a CPT counter from the last CPT value n _! of the CPT counter determined in the previous execution of step 40. The CPT counter is initially set to zero. If the new logical value Detect n determined in step 38 is "True", the last CPT value nThe value of the CPT counter is increased by a positive increment ACPT1. In one embodiment, the increment ACPT1 is in the range of 1 to 100, and is, for example, equal to 10. Furthermore, if the new value P'f n of the filtered time derivative P'f determined in step 34 is not zero, it is added to the last CPT value n _! of the CPT counter a positive increment ACPT2, strictly less than the increment ACPT1, and if the new value P' f n If the filtered time derivative P'f determined in step 34 is equal to zero, it is added to the last CPT value n -i of the CPT counter, a negative increment ACPT3. The new CPT value n The CPT counter value is therefore equal to the last CPT value n_! of the CPT counter determined in the previous execution of step 40, to which the ACPT1 increment is optionally added, and to which the ACPT2 or ACPT3 increment is added. However, if the result is negative, then the new CPT value nThe CPT counter is equal to zero. In one embodiment, the ACPT2 increment is strictly less than the ACPT1 increment. The ACPT1 increment is in the range of 0 to 10, and is, for example, equal to 0. In another embodiment, the ACPT3 increment is in the range of -10 to -1, and is, for example, equal to -1. The use of the negative ACPT3 increment allows the CPT counter to be brought back to 0 at a slow rate, particularly in the case where the CPT counter has been increased by one-off events not corresponding to an electric arc. The use of the positive ACPT2 increment, strictly less than the ACPT1 increment, allows the CPT counter to continue increasing at a slow rate when there is no change in sign of the time derivative P' but the value P'n of the time derivative remains above the threshold TH_P' of the derivative value.

[0043] The process continues at step 42.

[0044] In step 42, the processing module 26 determines if the new CPT value n The CPT counter value determined in step 40 has exceeded a first threshold of the TH1_CPT counter. In one embodiment, the first threshold of the TH1_CPT counter is in the range of 10 to 1000, and is, for example, equal to 40. If the new CPT value n the CPT counter has exceeded the first The TH1_CPT (Y) counter threshold, the process continues to step 44. If the new CPT value n If the CPT counter has not exceeded the first threshold of the TH1_CPT (N) counter, the process returns to step 30.

[0045] In step 44, the processing module 26 determines if the new CPT value nThe CPT counter value determined in step 40 has exceeded a second threshold, TH2_CPT, which is strictly greater than the first threshold, TH1_CPT. In one embodiment, the second threshold, TH2_CPT, is in the range of 10 to 2000, and is, for example, equal to 200. If the new CPT value n If the CPT counter value has not exceeded the second threshold of the TH2_CPT (N) counter, the process continues to step 46. If the new CPT value n The CPT counter has exceeded the second threshold of the TH2_CPT (Y) counter, the process continues at step 48.

[0046] In step 46, which corresponds to the case where the new CPT value nThe CPT counter reading exceeded the first threshold of the TH1_CPT counter but did not exceed the second threshold of the TH2_CPT counter. The processing module 26 then performs an action corresponding to a risk of arc flash. For example, the processing module 26 issues an alert on a human-machine interface. The process returns to step 30.

[0047] At step 48, which corresponds to the case where the new CPT value n When the CPT counter exceeds the second threshold of the TH2_CPT counter, the processing module 26 performs an action corresponding to the detection of an electrical arc. For example, the processing module 26 commands the power supply system 24 to interrupt the power supply to the heating element 18. Alternatively, the processing module 26 issues an alert on a human-machine interface. The process returns to step 30.

[0048] As an alternative, the process may use only one counter threshold. In this case, step 44 is replaced by step 46 and step 48 is omitted.

[0049] Figures 3, 4, 5, and 6 were obtained for an actual operating phase of the furnace 10 during which an electric arc formed in a heating element 18 of the furnace 10 between times t1, equal to 17707 s, and t2, equal to 19420 s. The method for detecting an electric arc according to the embodiment illustrated previously in relation to Figure 2 was implemented. The threshold TH_P' with derivative value is equal to 0.4. The increment ACPT1 is equal to 10. The increment ACPT2 is equal to 0. The increment ACPT3 is equal to -1. The first counter threshold TH1_CPT is equal to 40 and the second counter threshold TH2_CPT is equal to infinity to avoid interrupting the power supply to the heating element 18.

[0050] Figure 3 represents a time-dependent evolution curve of the electrical power P supplied by the electrical supply system 24 to the heating resistance 18 during an operating phase of the oven 10 during which an electric arc occurred.

[0051] Figure 4 represents a time-dependent evolution curve of the time derivative P' corresponding to the evolution curve of the power P in Figure 3.

[0052] Figure 5 represents a time evolution curve of the filtered time derivative P'f obtained from the time evolution curve of the time derivative P' of Figure 4.

[0053] Figure 6 shows a curve of the evolution of the CPT counter used during the implementation of the process embodiment illustrated in Figure 2. As can be seen in Figure 6, the CPT counter exceeds the first The TH1_CPT counter threshold was reached shortly after tl, precisely at 17715 s. Therefore, the detection of an electric arc formation was rapid. Furthermore, if the second TH2_CPT counter threshold had been set to 200, the power supply to the heating element 18 would have been interrupted shortly after time tl, precisely at 17737 s.

[0054] Various embodiments and variations have been described. Those skilled in the art will understand that certain features of these various embodiments and variations could be combined, and other variations will become apparent to them. Finally, the practical implementation of the described embodiments and variations is within the capabilities of those skilled in the art, based on the functional specifications given above.

Claims

CLAIMS 1. Method for detecting an electric arc in an electric heating system comprising at least one electric heating resistor (18) comprising determining that the number of changes in sign of the filtered time derivative of the electric power supplied to the electric heating resistor (18) is greater than a first threshold.

2. Method according to claim 1, comprising interrupting the electrical power supply to the electrical heating resistor (18) when the number of changes in sign of the filtered time derivative of the electrical power supplied to the electrical heating resistor (18) is greater than the first threshold or greater than a second threshold, strictly greater than the first threshold.

3. Method according to claim 1 or 2, comprising determining the time derivative of the electrical power supplied to the electrical heating resistor (18) and determining the filtered time derivative by filtering the time derivative.

4. Method according to any one of claims 1 to 3, comprising determining values ​​of the electrical power supplied to the electrical heating resistor (18) and determining a new value of the time derivative of the electrical power supplied to the electrical heating resistor (18) from the last two determined values ​​of the electrical power supplied to the electrical heating resistor (18).

5. Method according to claim 4, comprising determining a new value of the filtered time derivative equal to the last determined value. of the time derivative when the last determined value of the time derivative is greater than a derivative value threshold, and equal to 0 when the last determined value of the time derivative is less than the derivative value threshold.

6. Method according to claim 5, comprising determining a new value of a filtered and corrected time derivative equal to the last determined value of the filtered time derivative when the last determined value of the filtered time derivative is different from 0, and equal to the last determined value of the filtered and corrected time derivative when the last determined value of the filtered time derivative is equal to 0.

7. Method according to claim 6, comprising determining a logic value at a first logic state when the product of the last two determined values ​​of the filtered and corrected time derivative is strictly negative and at a second logic state, different from the first logic state, when the product of the last two determined values ​​of the filtered and corrected time derivative is positive or zero.

8. Method according to claim 7, comprising the determination of a new value of a counter equal to the last determined value of the counter to which is added a first strictly positive increment when the last determined logic value is in the first logic state, possibly a second strictly positive increment and strictly less than the first increment when the last determined logic value is in the second logic state and the last determined value of the filtered time derivative is different of zero, or a third strictly negative increment when the last determined logic value is in the second logic state and the last determined value of the filtered time derivative is equal to zero, and including the detection of an electric arc when the determined value of the counter is greater than the first threshold.

9. System for detecting (26) an electric arc in an electric heating system comprising at least one electric heating resistor (18) configured to determine whether the number of changes in sign of the filtered time derivative of the electric power supplied to the electric heating resistor (18) is greater than a first threshold.

10. Oven comprising an electric heating system comprising at least one electric heating resistor (18) and a system (26) for detecting an electric arc in the electric heating system according to claim 9.

11. Oven according to claim 10, comprising an electrical power supply system (24) of the electrical heating resistor (18) by a single-phase current.

Images