A method and system for online detection of steam turbine valve closing time

Abstract

The present invention relates to the field of online detection of turbine valve closing time, and provides a method and system for online detection of turbine valve closing time, establishes an anti-interference network model between a turbine DCS system and a valve, and the turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time, wherein the valve closing instruction includes a valve closing limited time range and a valve closing limited power; monitors the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time; obtains a scatter model diagram based on multiple first closing times and multiple second closing times, and obtains the valve closing time based on the scatter model diagram and the instruction sending time. When used, the present invention can effectively avoid the phenomenon of instruction data retransmission and the phenomenon of inaccurate LVDT sensor measurement in a harsh environment inside a turbine.

Description

Technical Field

[0001] The invention relates to the field of online detection of closing time of a steam turbine valve, and more specifically to an online detection method and system for closing time of a steam turbine valve. Background Art

[0002] The valve closing time of the steam turbine is determined by the moment when the turbine trip command is issued, the moment when the valve leaves the fully open position, and the moment when the valve reaches the fully closed position. The factors that affect the accuracy of the valve closing time generally include the command transmission delay and the valve execution trip action detection time delay.

[0003] At present, the traditional command transmission method is generally that the turbine DCS system and the valve establish a point-to-point transmission channel to transmit the valve trip command, and the valve then executes the trip action, where the trip action refers to the process of the valve changing from the fully open position to the fully closed position, and the valve closing time is the trip action execution time plus the command data transmission time. The traditional trip action execution time detection method generally uses an LVDT sensor to detect the displacement change of the valve to obtain the trip action execution time.

[0004] However, the interior of a steam turbine is usually a high-temperature, high-pressure, and high-noise environment. The use of traditional point-to-point transmission of command data and LVDT sensors will often cause interference in the transmission process or data collection due to the harsh environment inside the steam turbine. The command data may not be received normally, resulting in multiple retransmissions and inaccurate measurements by the LVDT sensor, resulting in problems such as high delay and inaccurate valve closing time. Summary of the invention

[0005] The object of the present invention is to provide a method and system for online detection of turbine valve closing time, so as to solve the shortcomings of the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a method for online detection of turbine valve closing time, comprising the following steps:

[0007] An anti-interference network model between the steam turbine DCS system and the valve is established. The steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time, wherein the valve closing instruction includes the valve closing limited time range and the valve closing limited power;

[0008] The valve is monitored based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time;

[0009] A scatter model diagram is obtained based on a plurality of first gate closing moments and a plurality of second gate closing moments, and a valve closing time is obtained based on the scatter model diagram and the instruction sending moment.

[0010] In a preferred embodiment, the step of establishing an anti-interference network model between the steam turbine DCS system and the valve, and the steam turbine DCS system transmitting a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time includes:

[0011] Determine the valve closing type, which includes normal closing and emergency closing, and set the valve closing limited time range and valve closing limited power according to the valve closing type to obtain a valve closing instruction;

[0012] A plurality of continuous interference monitoring points are set on the data transmission path between the steam turbine DCS system and the valve, and an interference monitoring point selection interval is set. A corresponding number of instruction channels are set based on the number of the interference monitoring point selection intervals. A number of interference monitoring points are selected from all instruction channels according to the preset interference monitoring point intervals and distributed on the corresponding instruction channels. The corresponding instruction channels are numbered according to the order of the instruction transmission direction to obtain a plurality of monitoring channels. The interference monitoring point on the monitoring channel is connected with the next interference monitoring point with the same corresponding number on other monitoring channels. According to the distance between the interference monitoring points, the interference monitoring points with different numbers that are connected are prioritized from near to far and from high to low. The priority is marked on the corresponding monitoring channel to obtain an anti-interference network model. The anti-interference network model is replaced with the data transmission path, and the valve closing instruction is transmitted to the valve based on the anti-interference network model.

[0013] In a preferred embodiment, the step of transmitting the valve closing instruction to the valve based on the anti-interference network model includes:

[0014] According to the number of monitoring channels in the anti-interference network model, a corresponding number of valve closing instructions and a corresponding number of detection data are copied, wherein the detection data includes a preset data segment, and multiple monitoring channels simultaneously transmit the corresponding detection data and valve closing instructions and record the instruction sending time;

[0015] The interference monitoring point through which the detection data and the valve closing instruction pass is taken as the target monitoring point, and the detection data is transmitted to the interference monitoring point connected to the target monitoring point and numbered next to the target monitoring point as the test monitoring point. A preset section of data received by the test monitoring point is obtained as the inspection data and transmitted back to the target monitoring point. The target monitoring point compares the inspection data with the detection data. If they are consistent, the test monitoring point is marked as a normal monitoring point. If they are inconsistent, the test monitoring point is marked as an abnormal monitoring point. The target monitoring point transmits the detection data and the valve closing instruction to the normal monitoring point corresponding to the monitoring channel with the highest priority. This operation is repeated until the valve receives the first detection data and valve closing instruction and records the arrival time of the instruction.

[0016] In a preferred embodiment, the step of monitoring the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain the first closing time and the second closing time includes:

[0017] Binding multiple LVDT sensors to corresponding video acquisition devices and numbering them in sequence to obtain multiple online detection devices;

[0018] The valve image is obtained based on the video acquisition device, and the valve image is divided into multiple detection intervals according to the valve closing direction and the preset distance. The positions of all the detection intervals of the valve image are obtained and marked on the corresponding LVDT sensor as recording points;

[0019] The time when the instruction arrives is used as the starting time of the valve closing action, and multiple LVDT sensors synchronously record the time when the valve passes through the recording point as the passing point recording time. The multiple passing point recording times of the corresponding LVDT sensors are used as the first closing time. Multiple video acquisition devices synchronously record the time when the valve passes through the detection interval to obtain the second closing time of the corresponding video acquisition device.

[0020] In a preferred embodiment, the step of synchronously recording the moment when the valve passes through the detection interval by the plurality of video acquisition devices to obtain the second closing moment of the corresponding video acquisition device comprises:

[0021] An image of a valve body at a preset distance closest to the valve is obtained as a valve head image, a plurality of valve head inspection points at preset positions are set on the valve head image, a valve closing video is obtained in real time based on a video acquisition device as an inspection video, and a plurality of valve head inspection points at preset positions are set in all detection intervals of the inspection video as valve head feature points;

[0022] When the valve head reaches a detection interval and all valve head inspection points coincide with the valve head characteristic points, a moment is recorded and bound to the corresponding detection interval as the zone crossing recording moment, until the valve head coincides with the last inspection interval to obtain multiple zone crossing recording moments as the second gate closing moment.

[0023] In a preferred embodiment, the step of obtaining a scatter model diagram based on a plurality of first gate closing moments and a plurality of second gate closing moments, and obtaining the valve closing time based on the scatter model diagram and the instruction sending moment comprises:

[0024] Compare all the point-passing record times of the first gate-closing time with the corresponding zone-passing record times of the second gate-closing time, and delete the point-passing record times and zone-passing record times whose values ​​differ beyond a preset range to obtain a plurality of first qualified gate-closing times and a plurality of second qualified gate-closing times;

[0025] A scatter model diagram is established based on multiple first qualified gate closing moments and multiple second qualified gate closing moments, a fitting curve is calculated based on the scatter model diagram, and the valve closing time is obtained based on the fitting curve and the instruction sending time and the instruction arrival time.

[0026] In a preferred embodiment, the steps of establishing a scatter model diagram based on a plurality of first qualified gate closing moments and a plurality of second qualified gate closing moments, calculating a fitting curve based on the scatter model diagram, and obtaining the valve closing time based on the fitting curve and the instruction sending time and the instruction arrival time include:

[0027] Merging all the passing point record moments of the first qualified gate-closing moment and the corresponding passing zone record moments of the second qualified gate-closing moment on the two-dimensional image according to the preset positions to obtain a scatter point model diagram;

[0028] Based on the scatter plot and the command sending time, the valve closing time is obtained by the residual square sum formula and the polynomial regression formula.

[0029] In a preferred embodiment, the step of obtaining the valve closing time based on the scatter plot and the instruction sending time by using the residual square sum formula and the polynomial regression formula includes:

[0030] Based on the scatter model graph, the fitting coefficient corresponding to the preset distance is obtained by the residual square sum formula, and the residual square sum formula is: ,in is the fitting coefficient corresponding to the preset position in the scatter model diagram, The number of preset positions for the scatter plot. For the The passing time of the first qualified gate closing moment, For the The time of passing the second qualified gate;

[0031] A fitting curve is obtained based on the scatter model graph according to a polynomial regression formula, and the polynomial regression formula is: ,in For time, For the preset distance, is the order of the polynomial, is the fitting coefficient corresponding to the preset distance in the scatter model diagram;

[0032] The time when the first derivative of the fitting curve is zero is obtained as the valve closing time, and the difference between the valve closing time and the command sending time is obtained to obtain the valve closing time.

[0033] The present invention also provides a steam turbine valve closing time online detection system, comprising:

[0034] Establishing a module for establishing an anti-interference network model between a steam turbine DCS system and a valve, wherein the steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain an instruction sending time and an instruction arrival time, wherein the valve closing instruction includes a valve closing limited time range and a valve closing limited power;

[0035] An online monitoring module, connected to the establishment module, is used to monitor the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time;

[0036] The calculation module is connected to the online detection module and is used to obtain a scatter model diagram based on multiple first gate closing moments and multiple second gate closing moments, and to obtain the valve closing time based on the scatter model diagram and the instruction sending moment.

[0037] In a preferred embodiment, the establishment module includes:

[0038] A determination module, used to determine the valve closing type, wherein the valve closing type includes normal closing and emergency closing, and to set the valve closing limited time range and the valve closing limited power according to the valve closing type to obtain a valve closing instruction;

[0039] A setting module is connected to the determination module, and is used to set multiple continuous interference monitoring points on the data transmission path between the turbine DCS system and the valve, set the interference monitoring point selection interval, set the corresponding number of instruction channels based on the number of interference monitoring point selection intervals, select a number of interference monitoring points from all instruction channels according to the preset interference monitoring point intervals and distribute them on the corresponding instruction channels, mark the numbers on the corresponding instruction channels according to the order of the instruction transmission direction to obtain multiple monitoring channels, connect the interference monitoring point on the monitoring channel with the next interference monitoring point with the same corresponding number on other monitoring channels, set the priority from high to low for the interference monitoring points that are connected and have different numbers according to the distance between the interference monitoring points, and mark the priority on the corresponding monitoring channel to obtain an anti-interference network model, replace the anti-interference network model with the data transmission path, and transmit the valve closing instruction to the valve based on the anti-interference network model.

[0040] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0041] The present invention can better optimize the instruction transmission delay and the valve execution tripping action detection time delay. By establishing an anti-interference network model to transmit instruction data to the valve, the valve closing action is monitored by multiple online detection devices and a corresponding scatter model diagram is established. It can effectively avoid the instruction data retransmission phenomenon and the LVDT sensor measurement inaccuracy phenomenon in the harsh environment inside the turbine, reduce the instruction data transmission delay, improve the instruction data transmission efficiency, and ensure the accuracy of data measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0043] Figure 1 The present invention is a flow chart of the method.

[0044] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Example 1, please refer to Figure 1 As shown, the method for online detection of turbine valve closing time described in this embodiment includes the following steps:

[0047] S1. Establish an anti-interference network model between the steam turbine DCS system and the valve. The steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time, wherein the valve closing instruction includes the valve closing limited time range and the valve closing limited power;

[0048] S2, monitoring the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time;

[0049] S3, obtaining a scatter model diagram based on a plurality of first gate closing moments and a plurality of second gate closing moments, and obtaining a valve closing time based on the scatter model diagram and the instruction sending time;

[0050] As described in the above steps S1-S3, at present, the traditional command transmission method is generally that the steam turbine DCS system and the valve establish a point-to-point transmission channel to transmit the valve trip command, and the valve then executes the trip action, wherein the trip action refers to the process of the valve from the fully open position to the fully closed position, and the valve closing time is the trip action execution time plus the command data transmission time, and the traditional trip action execution time detection method generally uses the LVDT sensor to detect the displacement change of the valve to obtain the trip action execution time, but the interior of the steam turbine is usually a high temperature, high pressure, and high noise environment. The use of traditional point-to-point transmission of command data and LVDT sensors often causes the command data to be transmitted due to the harsh environment inside the steam turbine. If the process or data collection is disturbed, the command data cannot be received normally, resulting in multiple retransmissions and inaccurate measurement of the LVDT sensor, resulting in high delay and inaccurate valve closing time. The present application can better optimize the command transmission delay and the valve execution tripping action detection time delay, and transmit the command data to the valve by establishing an anti-interference network model, monitor the valve closing action through multiple online detection devices and establish a corresponding scatter model diagram, which can effectively avoid the command data retransmission phenomenon and the LVDT sensor inaccurate measurement in the harsh environment inside the turbine, reduce the command data transmission delay, improve the command data transmission efficiency, and ensure the accuracy of data measurement;

[0051] In one embodiment, the step S1 of establishing an anti-interference network model between the steam turbine DCS system and the valve, wherein the steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time, comprises:

[0052] S11, determining the valve closing type, wherein the valve closing type includes normal closing and emergency closing, and setting the valve closing limited time range and the valve closing limited power according to the valve closing type to obtain a valve closing instruction;

[0053] S12. Set a plurality of continuous interference monitoring points on the data transmission path between the steam turbine DCS system and the valve, set an interference monitoring point selection interval, set a corresponding number of instruction channels based on the number of interference monitoring point selection intervals, select a number of interference monitoring points from all instruction channels according to the preset interference monitoring point intervals and distribute them on the corresponding instruction channels, mark the numbers on the corresponding instruction channels according to the order of the instruction transmission direction to obtain a plurality of monitoring channels, connect the interference monitoring point on the monitoring channel with the next interference monitoring point with the same corresponding number on other monitoring channels, set the priority from high to low for the interference monitoring points with connection relationship and different numbers according to the distance between the interference monitoring points, and mark the priority on the corresponding monitoring channel to obtain the anti-interference network model, replace the anti-interference network model with the data transmission path, and transmit the valve closing instruction to the valve based on the anti-interference network model;

[0054] As described in the above steps S11-S12, first determine the valve closing type, where the valve closing type includes normal closing and emergency closing, set different valve closing time limit ranges and valve closing limit powers as valve closing instructions according to different closing types, and then replace the data transmission path between the turbine DCS system and the valve with an anti-interference network model. The anti-interference network model is established by setting multiple continuous interference monitoring points on the data transmission path, setting the interference monitoring point selection interval, and the selection range here is used to specify the spacing between each interference monitoring point on the subsequent instruction channel. According to the number of interference monitoring point selection intervals, a corresponding number of instruction channels are set, and then all instruction channels select a number of interference monitoring points according to the preset interference monitoring point intervals and distribute them on the corresponding instruction channels. The total number of interference monitoring points on all instruction channels here corresponds to the total number of interference monitoring points on the data transmission path, and the corresponding instruction channels are marked in order according to the instruction transmission direction. The above numbering obtains multiple monitoring channels, where the numbers are the same in different instruction channels. For example, the interference monitoring points of the first monitoring channel are numbered 0001, 0002, 0003..., and the second interference monitoring points are also numbered 0001, 0002, 0003..., and the interference monitoring points on the monitoring channel are connected with the next interference monitoring point of the corresponding interference monitoring point with the same number on other monitoring channels. For example, the 0001 interference monitoring point of the first monitoring channel is connected with the 0002 interference monitoring point of the second monitoring channel. According to the distance between the interference monitoring points, the interference monitoring points with different numbers that are connected are prioritized from near to far and from high to low, and the priority is marked on the corresponding monitoring channel to obtain an anti-interference network model, which is used to replace the original data transmission path with the anti-interference network model, which can better optimize and effectively avoid the retransmission of instruction data in the harsh environment inside the steam turbine, and assist in the subsequent transmission of valve closing instructions;

[0055] In one embodiment, the step S12 of transmitting the valve closing instruction to the valve based on the anti-interference network model includes:

[0056] S121, copying a corresponding number of valve closing instructions and a corresponding number of detection data according to the number of monitoring channels in the anti-interference network model, wherein the detection data includes a preset data segment, and multiple monitoring channels simultaneously transmit the corresponding detection data and valve closing instructions and record the time when the instructions are sent;

[0057] S122. The interference monitoring point through which the detection data and the valve closing instruction pass is taken as the target monitoring point, and the detection data is transmitted to the interference monitoring point connected to the target monitoring point and numbered next to the target monitoring point as the test monitoring point. A preset segment of data received by the test monitoring point is obtained as the inspection data and transmitted back to the target monitoring point. The target monitoring point compares the inspection data with the detection data. If they are consistent, the test monitoring point is marked as a normal monitoring point. If they are inconsistent, the test monitoring point is marked as an abnormal monitoring point. The target monitoring point transmits the detection data and the valve closing instruction to the normal monitoring point corresponding to the monitoring channel with the highest priority. This operation is repeated until the valve receives the first detection data and valve closing instruction and records the arrival time of the instruction.

[0058] As described in the above steps S121-S122, first, according to the number of monitoring channels in the anti-interference network model, a corresponding number of valve closing instructions and a corresponding number of detection data are copied. For example, three monitoring channels copy three copies of valve closing instructions and detection data. The detection data here is a preset piece of data. Multiple monitoring channels simultaneously transmit the corresponding detection data and valve closing instructions and record the time when the instructions are sent. Then, the interference monitoring point through which the detection data and the valve closing instruction pass is taken as the target monitoring point, and the detection data is transmitted to the interference monitoring point connected to the target monitoring point and numbered the next number of the target monitoring point as the test monitoring point. For example, the target monitoring point is numbered 0001, and the interference monitoring point numbered 0002 in all monitoring channels is taken as the test monitoring point. The detection data is transmitted to all the test monitoring points, and then the test monitoring point is obtained. A preset data is received as the inspection data and transmitted back to the target monitoring point. The target monitoring point compares the inspection data with the detection data. If they are consistent, the test monitoring point is marked as a normal monitoring point, otherwise the test monitoring point is marked as an abnormal monitoring point. For example, the detection data is 010001 in binary, and the inspection data of one of the test monitoring points is 010001 in binary, then the test detection point is marked as a normal monitoring point. Finally, the target monitoring point transmits the detection data and the valve closing instruction to the normal monitoring point corresponding to the monitoring channel with the highest priority. This operation is repeated until the valve receives the first detection data and the valve closing instruction and records the arrival time of the instruction, which is used to transmit the valve closing instruction from the turbine DCS system to the valve through the anti-interference network model, which can effectively reduce the instruction data transmission delay and improve the instruction data transmission efficiency.

[0059] In one embodiment, the step S2 of monitoring the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain the first closing time and the second closing time includes:

[0060] S21, binding multiple LVDT sensors with corresponding video acquisition devices and numbering them in sequence to obtain multiple online detection devices;

[0061] S22, acquiring a valve image based on a video acquisition device, dividing the valve image into multiple detection intervals according to the valve closing direction and a preset distance, acquiring positions of all detection intervals of the valve image and marking them on corresponding LVDT sensors as recording points;

[0062] S23, taking the command arrival time as the valve closing action start time, multiple LVDT sensors synchronously recording the time when the valve passes through the recording point as the passing point recording time, taking the multiple passing point recording times of the corresponding LVDT sensors as the first closing time, and multiple video acquisition devices synchronously recording the time when the valve passes through the detection interval to obtain the second closing time of the corresponding video acquisition device;

[0063] In one embodiment, the step S23 of synchronously recording the moment when the valve passes through the detection interval by the plurality of video acquisition devices to obtain the second closing moment of the corresponding video acquisition device includes:

[0064] S231, obtaining an image of a valve body at a preset distance closest to the valve as a valve head image, setting a plurality of valve head inspection points at preset positions on the valve head image, obtaining a valve closing video in real time based on a video acquisition device as an inspection video, and setting a plurality of valve head inspection points at preset positions in all detection intervals of the inspection video as valve head feature points;

[0065] S232, when the valve head reaches a detection interval and all the valve head inspection points coincide with the valve head characteristic points, a time is recorded and bound to the corresponding detection interval as the zone-crossing recording time, until the valve head coincides with the last inspection interval to obtain multiple zone-crossing recording times as the second gate closing time;

[0066] As described in the above steps S21-S23, firstly, a plurality of online detection devices are set up. Here, the online detection devices are obtained by binding a plurality of LVDT sensors with corresponding video acquisition devices and numbering them in sequence. Then, the valve image is obtained by using the video acquisition device. The valve image is divided into a plurality of detection intervals according to the valve closing direction and the preset distance. The positions of all detection intervals of the valve image are obtained and marked on the corresponding LVDT sensor as recording points. Then, the arrival time of the instruction is used as the start time of the valve closing action. The plurality of LVDT sensors synchronously record the time when the valve passes the recording point as the passing point recording time. The plurality of passing point recording times of the corresponding LVDT sensor are used as the first closing time. For example, the length of the valve closing direction is 10 cm, and the preset distance is 5 mm. The valve image is divided into 20 parts, and each of them is marked as a recording point at intervals of 5 mm. The LVDT sensor records a passing point recording time each time it detects that the valve passes the recording point. The plurality of passing point recording times are used as the first closing time. Then, an image of the valve body at a preset distance closest to the valve is obtained as the valve head image. A plurality of valve head inspection points at preset positions are set on the valve head image, and then a real-time valve closing video is obtained as an inspection video. A plurality of valve head inspection points at preset positions are set in all detection intervals of the inspection video as valve head feature points. When the valve head reaches a detection interval and all valve head inspection points coincide with the valve head feature points, a time is recorded and bound to the corresponding detection interval as a cross-zone recording time, until the valve head coincides with the last inspection interval to obtain a plurality of cross-zone recording times as the second gate closing time. For example, the inspection interval here has a range of 5mm. A valve head inspection point of 5mm at the valve head is copied to the inspection interval as a valve head feature point. When the valve starts to close, when the valve head inspection point coincides with the valve head feature point, a cross-zone recording time is recorded. The plurality of cross-zone recording times are combined into the second gate closing time, which is used to simultaneously monitor the valve closing action through multiple online detection devices to obtain the first gate closing time and the second gate closing time. This can better avoid the phenomenon of inaccurate measurement of LVDT sensors, enhance data accuracy and reliability, and assist in the subsequent establishment of a scatter model diagram to calculate the valve closing time.

[0067] In one embodiment, the step S3 of obtaining a scatter model diagram based on a plurality of first gate closing moments and a plurality of second gate closing moments, and obtaining the valve closing time based on the scatter model diagram and the instruction sending moment, comprises:

[0068] S31, comparing all the point-passing record moments of the first gate-closing moments with the corresponding zone-passing record moments of the second gate-closing moments, deleting the point-passing record moments and zone-passing record moments whose values ​​differ beyond a preset range to obtain a plurality of first qualified gate-closing moments and a plurality of second qualified gate-closing moments;

[0069] S32, establishing a scatter model diagram based on a plurality of first qualified gate closing moments and a plurality of second qualified gate closing moments, calculating a fitting curve based on the scatter model diagram, and obtaining the valve closing time based on the fitting curve and the instruction sending time and the instruction arrival time;

[0070] In one embodiment, the step S32 of establishing a scatter model diagram based on a plurality of first qualified gate closing moments and a plurality of second qualified gate closing moments, calculating a fitting curve based on the scatter model diagram, and obtaining the valve closing time based on the fitting curve and the instruction sending time and the instruction arrival time includes:

[0071] S321, merging the passing point record time of all first qualified gate-closing time and the passing zone record time of the corresponding second qualified gate-closing time on the two-dimensional image according to the preset position to obtain a scatter point model diagram;

[0072] S322, based on the scatter plot and the instruction sending time, the valve closing time is obtained by using the residual square sum formula and the polynomial regression formula;

[0073] In one embodiment, the step S322 of obtaining the valve closing time based on the scatter plot and the instruction sending time by using the residual square sum formula and the polynomial regression formula includes:

[0074] S3221. Obtain the fitting coefficient corresponding to the preset distance through the residual square sum formula based on the scatter model graph, and the residual square sum formula is: ,in is the fitting coefficient corresponding to the preset position in the scatter model diagram, The number of preset positions for the scatter plot. For the The passing time of the first qualified gate closing moment, For the The time of passing the second qualified gate;

[0075] S3222. Obtain a fitting curve based on the scatter model graph according to a polynomial regression formula, where the polynomial regression formula is: ,in For time, For the preset distance, is the order of the polynomial, is the fitting coefficient corresponding to the preset distance in the scatter model diagram;

[0076] S3223, obtaining the time when the first derivative of the fitting curve is zero as the valve closing time, and obtaining the difference between the valve closing time and the command sending time to obtain the valve closing time;

[0077] As described in the above steps S31-S32, firstly, all the point-passing record moments of the first gate closing moment are compared with the corresponding zone-passing record moments of the second gate closing moment, and the point-passing record moments and zone-passing record moments whose numerical differences exceed the preset range are deleted to obtain a plurality of first qualified gate closing moments and a plurality of second qualified gate closing moments. For example, a point-passing record moment is 20ms, and the corresponding zone-passing record moment is 50ms, and the preset range of difference within the interval is 10ms, then the point-passing record moment and the zone-passing record moment are deleted to eliminate the data with large errors, which can reduce the error of the overall data, and then all the point-passing record moments and the zone-passing record moments are marked at the corresponding preset positions in the two-dimensional image. A scatter model graph is obtained to analyze data features and assist in calculating a fitting curve. Then, the fitting coefficient corresponding to the preset distance is obtained based on the scatter model graph through the residual square sum formula. The fitting curve is obtained based on the scatter model graph according to the polynomial regression formula. The moment when the first derivative of the fitting curve is zero is obtained as the valve closing moment. Here, the moment when the first derivative of the fitting curve is zero represents that the valve has been completely closed. Finally, the difference between the valve closing moment and the instruction sending moment is collected to obtain the valve closing time, which is used to obtain the valve closing time according to the residual square sum formula and the polynomial regression formula. This can intuitively display the time characteristics of the point recording moment and the cross-zone recording moment at the preset distance, and better ensure the accuracy of data measurement.

[0078] Example 2, please refer to Figure 2 As shown, the steam turbine valve closing time online detection system described in this embodiment includes:

[0079] Establishing a module for establishing an anti-interference network model between a steam turbine DCS system and a valve, wherein the steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain an instruction sending time and an instruction arrival time, wherein the valve closing instruction includes a valve closing limited time range and a valve closing limited power;

[0080] An online monitoring module, connected to the establishment module, is used to monitor the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time;

[0081] A calculation module, connected to the online detection module, for obtaining a scatter model diagram based on a plurality of first gate closing moments and a plurality of second gate closing moments, and obtaining a valve closing time based on the scatter model diagram and the instruction sending moment;

[0082] In one embodiment, the establishment module includes:

[0083] A determination module, used to determine the valve closing type, wherein the valve closing type includes normal closing and emergency closing, and to set the valve closing limited time range and the valve closing limited power according to the valve closing type to obtain a valve closing instruction;

[0084] A setting module is connected to the determination module, and is used to set multiple continuous interference monitoring points on the data transmission path between the turbine DCS system and the valve, set the interference monitoring point selection interval, set the corresponding number of instruction channels based on the number of interference monitoring point selection intervals, select a number of interference monitoring points from all instruction channels according to the preset interference monitoring point intervals and distribute them on the corresponding instruction channels, mark the numbers on the corresponding instruction channels according to the order of the instruction transmission direction to obtain multiple monitoring channels, connect the interference monitoring point on the monitoring channel with the next interference monitoring point with the same corresponding number on other monitoring channels, set the priority from high to low for the interference monitoring points that are connected and have different numbers according to the distance between the interference monitoring points, and mark the priority on the corresponding monitoring channel to obtain an anti-interference network model, replace the anti-interference network model with the data transmission path, and transmit the valve closing instruction to the valve based on the anti-interference network model.

[0085] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for online detection of turbine valve closing time, characterized in that: The following steps are involved: An anti-interference network model between the steam turbine DCS system and the valve is established. The steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time, wherein the valve closing instruction includes the valve closing limited time range and the valve closing limited power; The valve is monitored based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time; A scatter model diagram is obtained based on a plurality of first gate closing moments and a plurality of second gate closing moments, and a valve closing time is obtained based on the scatter model diagram and the instruction sending time; The step of establishing an anti-interference network model between the steam turbine DCS system and the valve, and the steam turbine DCS system transmitting a valve closing instruction to the valve based on the anti-interference network model to obtain the instruction sending time and the instruction arrival time includes: Determine the valve closing type, which includes normal closing and emergency closing, and set the valve closing limited time range and valve closing limited power according to the valve closing type to obtain a valve closing instruction; A plurality of continuous interference monitoring points are set on the data transmission path between the steam turbine DCS system and the valve, and an interference monitoring point selection interval is set. A corresponding number of instruction channels are set based on the number of the interference monitoring point selection intervals. A number of interference monitoring points are selected from all instruction channels according to the preset interference monitoring point intervals and distributed on the corresponding instruction channels. A number is marked on the corresponding instruction channels according to the order of the instruction transmission direction to obtain a plurality of monitoring channels. The interference monitoring point on the monitoring channel is connected with the next interference monitoring point of the same corresponding number on other monitoring channels. According to the distance between the interference monitoring points, the interference monitoring points with different numbers that are connected are prioritized from near to far and from high to low. The priority is marked on the corresponding monitoring channel to obtain an anti-interference network model. The anti-interference network model is replaced with the data transmission path. Based on the anti-interference network model, the valve closing instruction is transmitted to the valve. The step of monitoring the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain the first closing time and the second closing time includes: Binding multiple LVDT sensors to corresponding video acquisition devices and numbering them in sequence to obtain multiple online detection devices; The valve image is obtained based on the video acquisition device, and the valve image is divided into multiple detection intervals according to the valve closing direction and the preset distance. The positions of all the detection intervals of the valve image are obtained and marked on the corresponding LVDT sensor as recording points; The command arrival time is used as the valve closing action start time, multiple LVDT sensors synchronously record the time when the valve passes the recording point as the passing point recording time, the multiple passing point recording times of the corresponding LVDT sensors are used as the first closing time, and multiple video acquisition devices synchronously record the time when the valve passes the detection interval to obtain the second closing time of the corresponding video acquisition device; The step of obtaining a scatter model diagram based on a plurality of first gate closing moments and a plurality of second gate closing moments, and obtaining the valve closing time based on the scatter model diagram and the instruction sending moment, comprises: Compare all the point-passing record times of the first gate-closing time with the corresponding zone-passing record times of the second gate-closing time, and delete the point-passing record times and zone-passing record times whose values ​​differ beyond a preset range to obtain a plurality of first qualified gate-closing times and a plurality of second qualified gate-closing times; Establish a scatter model diagram based on multiple first qualified gate closing moments and multiple second qualified gate closing moments, calculate a fitting curve based on the scatter model diagram, and obtain the valve closing time based on the fitting curve and the instruction sending time and the instruction arrival time; The steps of establishing a scatter model diagram based on a plurality of first qualified gate closing moments and a plurality of second qualified gate closing moments, calculating a fitting curve based on the scatter model diagram, and obtaining the valve closing time based on the fitting curve and the instruction sending moment and the instruction arrival moment include: Merging all the passing point record moments of the first qualified gate-closing moment and the corresponding passing zone record moments of the second qualified gate-closing moment on the two-dimensional image according to the preset positions to obtain a scatter point model diagram; Based on the scatter plot and the command sending time, the valve closing time is obtained by using the residual square sum formula and the polynomial regression formula; The step of obtaining the valve closing time based on the scatter plot and the instruction sending time by using the residual square sum formula and the polynomial regression formula includes: Based on the scatter model graph, the fitting coefficient corresponding to the preset distance is obtained by the residual square sum formula, and the residual square sum formula is: ,in is the fitting coefficient corresponding to the preset position in the scatter model diagram, The number of preset positions for the scatter plot. For the The passing time of the first qualified gate closing moment, For the The time of passing the second qualified gate; A fitting curve is obtained based on the scatter model graph according to a polynomial regression formula, and the polynomial regression formula is: ,in For time, For the preset distance, is the order of the polynomial, is the fitting coefficient corresponding to the preset distance in the scatter model diagram; The time when the first derivative of the fitting curve is zero is obtained as the valve closing time, and the difference between the valve closing time and the command sending time is obtained to obtain the valve closing time.

2. A method for online detection of turbine valve closing time according to claim 1, characterized in that: The step of transmitting the valve closing instruction to the valve based on the anti-interference network model includes: According to the number of monitoring channels in the anti-interference network model, a corresponding number of valve closing instructions and a corresponding number of detection data are copied, wherein the detection data includes a preset data segment, and multiple monitoring channels simultaneously transmit the corresponding detection data and valve closing instructions and record the instruction sending time; The interference monitoring point through which the detection data and the valve closing instruction pass is taken as the target monitoring point, and the detection data is transmitted to the interference monitoring point connected to the target monitoring point and numbered next to the target monitoring point as the test monitoring point. A preset section of data received by the test monitoring point is obtained as the inspection data and transmitted back to the target monitoring point. The target monitoring point compares the inspection data with the detection data. If they are consistent, the test monitoring point is marked as a normal monitoring point. If they are inconsistent, the test monitoring point is marked as an abnormal monitoring point. The target monitoring point transmits the detection data and the valve closing instruction to the normal monitoring point corresponding to the monitoring channel with the highest priority. This operation is repeated until the valve receives the first detection data and valve closing instruction and records the arrival time of the instruction.

3. A method for online detection of turbine valve closing time according to claim 1, characterized in that: The step of synchronously recording the moment when the valve passes through the detection interval by the plurality of video acquisition devices to obtain the second closing moment of the corresponding video acquisition device comprises: An image of a valve body at a preset distance closest to the valve is obtained as a valve head image, a plurality of valve head inspection points at preset positions are set on the valve head image, a valve closing video is obtained in real time based on a video acquisition device as an inspection video, and a plurality of valve head inspection points at preset positions are set in all detection intervals of the inspection video as valve head feature points; When the valve head reaches a detection interval and all valve head inspection points coincide with the valve head characteristic points, a moment is recorded and bound to the corresponding detection interval as the zone crossing recording moment, until the valve head coincides with the last inspection interval to obtain multiple zone crossing recording moments as the second gate closing moment.

4. A steam turbine valve closing time online detection system, used to implement a steam turbine valve closing time online detection method according to any one of claims 1 to 3, characterized in that: include: Establishing a module for establishing an anti-interference network model between a steam turbine DCS system and a valve, wherein the steam turbine DCS system transmits a valve closing instruction to the valve based on the anti-interference network model to obtain an instruction sending time and an instruction arrival time, wherein the valve closing instruction includes a valve closing limited time range and a valve closing limited power; An online monitoring module, connected to the establishment module, is used to monitor the valve based on multiple LVDT sensors and multiple video acquisition devices to obtain a first closing time and a second closing time; The calculation module is connected to the online detection module and is used to obtain a scatter model diagram based on multiple first gate closing moments and multiple second gate closing moments, and to obtain the valve closing time based on the scatter model diagram and the instruction sending moment.

5. The on-line detection system for closing time of a steam turbine valve according to claim 4, characterized in that: The establishment module includes: A determination module, used to determine the valve closing type, wherein the valve closing type includes normal closing and emergency closing, and to set the valve closing limited time range and the valve closing limited power according to the valve closing type to obtain a valve closing instruction; A setting module is connected to the determination module, and is used to set multiple continuous interference monitoring points on the data transmission path between the turbine DCS system and the valve, set the interference monitoring point selection interval, set the corresponding number of instruction channels based on the number of interference monitoring point selection intervals, select a number of interference monitoring points from all instruction channels according to the preset interference monitoring point intervals and distribute them on the corresponding instruction channels, mark the numbers on the corresponding instruction channels according to the order of the instruction transmission direction to obtain multiple monitoring channels, connect the interference monitoring point on the monitoring channel with the next interference monitoring point with the same corresponding number on other monitoring channels, set the priority from high to low for the interference monitoring points that are connected and have different numbers according to the distance between the interference monitoring points, and mark the priority on the corresponding monitoring channel to obtain an anti-interference network model, replace the anti-interference network model with the data transmission path, and transmit the valve closing instruction to the valve based on the anti-interference network model.

Images