Method and device for locating fault of converter valve cooling system and storage medium

Abstract

A fault positioning method, device and medium of a converter valve cooling system are disclosed. In the method, a plurality of water pressure sensors and a plurality of voltage sensors are arranged in the converter valve cooling system, historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information includes fault types, fault positioning information and state information, water pressure signals and voltage sensor signals of the converter valve cooling system are collected in real time to form a sensor signal matrix, fault discrimination and fault positioning are performed based on the sensor signal matrix; a water leakage fault of a second positioning accuracy is identified and positioned, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database, the historical water leakage fault information includes fault positioning information and state information, a target function is established based on the second fault database, linear regression is used for solving, when a water leakage fault alarm occurs, state information is collected and a fault point is confirmed according to the target function.

Description

Technical Field

[0001] The present invention belongs to the technical field of direct current transmission system detection, and in particular to a fault locating method, device and storage medium for a converter valve cooling system. Background Art

[0002] During converter valve operation, power electronic components generate significant power losses. This significant power loss causes the valve's temperature to rise. If this temperature rise cannot be effectively and promptly controlled, it can lead to thermal failure of the power electronic components, threatening the safety of the entire HVDC Flexible system. Therefore, converter valves require a cooling system to circulate heat and discharge it outside the valve hall. The converter valve cooling system includes a wide variety of equipment, including water pipes, pumps, filters, ion exchangers, and three-way valves. Consequently, a wide variety of faults can occur during operation, making precise fault location difficult.

[0003] For example, at present, water leakage faults in the cooling system of the converter valve are mainly detected through valve tower water leakage detection devices and valve section water leakage detection devices. The valve tower water leakage detection device is set at the bottom of the valve tower of the converter valve, and the valve section water leakage detection device is set at the bottom of the valve section. In this way, the water leakage fault can only be located in a certain valve tower or a certain valve section, and further positioning is impossible.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] In response to the problems existing in the prior art, the present invention proposes a fault locating method, device and storage medium for a converter valve cooling system. Water pressure and voltage sensors are used to collect data in real time to identify and locate water leakage faults, sensor faults and power supply faults. A linear regression method is used to accurately locate water leakage faults, and a fault database is established to facilitate future inquiries and use.

[0006] The objective of the present invention is achieved through the following technical solutions: a fault location method for a converter valve cooling system includes:

[0007] Identifying a first-level positioning accuracy of a converter valve cooling system failure mode, including a blockage fault, a sensor fault, and a power supply fault, wherein historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information including fault type, fault location information, and status information; collecting water pressure signals and voltage sensor signals of the converter valve cooling system in real time to form a sensor signal matrix; and performing fault identification and fault location based on the sensor signal matrix;

[0008] Identify and locate water leakage faults with a second positioning accuracy, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database. The historical water leakage fault information includes water leakage fault location information and status information. An objective function is established based on the second fault database and solved using linear regression. When a water leakage fault alarm occurs, status information is collected and the fault point is confirmed according to the objective function.

[0009] In the fault location method for the converter valve cooling system, in identifying and locating a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy, the sensor signal matrix:

[0010] in,

[0011] The water pressure signal and voltage sensor signal include the sensor position signal x, the self-test signal y and the alarm signal z, respectively, with subscripts 1, 2,…, m-1, m, m+1,…, n, where m is the number of water pressure sensors, nm is the number of voltage sensors, y is 0 for a sensor failure, y is 1 for a sensor normal, the alarm signal z is 0 for an abnormality in the information detected by the sensor and the sensor issues an alarm signal, and z is 1 for a normality in the information detected by the sensor and the sensor does not issue an alarm signal.

[0012] In the fault location method of the converter valve cooling system, the second column of the sensor signal matrix is ​​queried to see whether there is y i is 0, 1≤i≤n. If yes, a sensor failure occurs. According to the position information x corresponding to the sensor i Position the sensor. If the second column y in the matrix i is 1, i=1, 2, ..., n, then query the third column, if the third column contains z i If it is 0, 1≤i≤n, then query the position information x corresponding to the sensor i , if 1≤x i ≤m, a blocking fault occurs. If m+1≤× i ≤n, a power failure occurs, according to the position information x i Locate the sensor to locate the fault.

[0013] In the fault location method for the converter valve cooling system, in identifying and locating a water leakage fault with a second positioning accuracy, n points are used to equally divide the water pipe of the converter valve cooling system into n positioning points 1, 2, ..., n. The positioning point i closest to the leakage point when the water leakage fault occurs, 1≤i≤n, is used as the water leakage fault location point, and the fault location information x is i.

[0014] In the fault location method of the converter valve cooling system, the state information when the water leakage fault occurs is collected, which includes the inlet water temperature t in, outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form a state matrix:

[0015]

[0016] A second fault database is constructed based on the state matrix, and an objective function is established:

[0017]

[0018] in: is the position information calculated according to the objective function, w1, w2, …, w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation;

[0019] Written in vector form as follows:

[0020]

[0021] in

[0022] Define the parameter matrix

[0023] In the fault location method for the converter valve cooling system, a training set is established based on the second fault database, and a loss function is defined;

[0024] Select a fixed number of small batch samples B, which contains |B| number of fault instances;

[0025] The loss function for a single fault instance i(i∈B) for:

[0026]

[0027] in is the loss function of the i-th fault instance, is the fault location information of fault instance i calculated according to the objective function, y (i) is the real position information x of fault instance i, w1, W2, ..., w6 are the weights of inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level respectively, b is the deviation, is the parameter matrix;

[0028] Loss function for mini-batch samples B for:

[0029]

[0030] in The loss function of the mini-batch sample B, |B| is the number of fault instances contained in the mini-batch sample B;

[0031] Select the initial value of the parameter:

[0032]

[0033] Iterative calculation is performed according to the gradient descent method, and iterative updates are performed as follows:

[0034]

[0035] Where η is the learning rate;

[0036] After each iteration, the loss function of sample B is calculated as:

[0037]

[0038] when Continue iterating, when When , stop iteration;

[0039] Where m is a parameter that determines whether to stop the iteration.

[0040] When the iteration stops, the final parameter matrix is ​​obtained

[0041] When a water leakage alarm occurs, collect the status information of the fault and the water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0042]

[0043] Substitute into the objective function Get location information according to The corresponding location point is queried to complete the water leakage fault location and the fault instance is recorded in the second fault database to update the second fault database.

[0044] A computer-readable storage medium is configured to execute the fault location method for a converter valve cooling system.

[0045] A fault locating device for a converter valve cooling system comprises:

[0046] a first identification unit, configured to identify a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy of a fault mode of a converter valve cooling system, wherein historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information including fault type, fault location information, and status information; water pressure signals and voltage sensor signals of the converter valve cooling system are collected in real time to form a sensor signal matrix; and fault identification and fault location are performed based on the sensor signal matrix;

[0047] The second identification unit is used to identify and locate water leakage faults with a second positioning accuracy, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database, the historical water leakage fault information includes water leakage fault location information and status information, an objective function is established based on the second fault database, and a linear regression solution is used. When a water leakage fault alarm occurs, the status information is collected and the fault point is confirmed according to the objective function.

[0048] In the fault location device for the converter valve cooling system, the first identification unit identifies and locates a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy, and the sensor signal matrix:

[0049]

[0050] The water pressure signal and voltage sensor signal include the sensor position signal x, the self-test signal y and the alarm signal z, respectively, with subscripts 1, 2,…, m-1, m, m+1,…, n, where m is the number of water pressure sensors, nm is the number of voltage sensors, y is 0 for a sensor failure, y is 1 for a sensor normal, the alarm signal z is 0 for an abnormality in the information detected by the sensor and the sensor issues an alarm signal, and z is 1 for a normality in the information detected by the sensor and the sensor does not issue an alarm signal.

[0051] In the fault location device for the converter valve cooling system, the first identification unit queries the sensor signal matrix for the presence of y in the second column. i is 0, 1≤i≤n. If yes, a sensor failure occurs. According to the position information x corresponding to the sensor i Position the sensor. If the second column y in the matrix i is 1, i=1, 2, ..., n, then query the third column, if the third column contains z i If it is 0, 1≤i≤n, then query the position information x corresponding to the sensor i , if 1≤x i ≤m, a blocking fault occurs, if m+1≤x i ≤n, a power failure occurs, according to the position information x i Locate the sensor to locate the fault.

[0052] In the fault locating device for the converter valve cooling system, the second identification unit identifies and locates a water leakage fault with a second positioning accuracy, divides the water pipe of the converter valve cooling system into n equal parts to determine n positioning points 1, 2, ..., n, and takes the positioning point i closest to the leakage point when the water leakage fault occurs, 1≤i≤n, as the water leakage fault positioning point, and the fault positioning information x is i.

[0053] In the fault location device of the converter valve cooling system, the second identification unit collects state information when the water leakage fault occurs, which includes the inlet water temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form a state matrix:

[0054]

[0055] A second fault database is constructed based on the state matrix, and an objective function is established:

[0056]

[0057] in: is the position information calculated according to the objective function, W1, W2, …, W6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation;

[0058] Written in vector form as follows:

[0059]

[0060] in

[0061] Define the parameter matrix

[0062] In the fault location device for the converter valve cooling system, the second identification unit establishes a training set based on the second fault database and defines a loss function;

[0063] Select a fixed number of small batch samples B, which contains |B| number of fault instances;

[0064] The loss function for a single fault instance i(i∈B) for:

[0065]

[0066] in is the loss function of the i-th fault instance, is the fault location information of fault instance i calculated according to the objective function, y (i) is the real position information of fault instance i×, w1, w2, ..., w6 are the weights of inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level respectively, b is the deviation, is the parameter matrix;

[0067] Loss function for mini-batch samples B for:

[0068]

[0069] in The loss function of the mini-batch sample B, |B| is the number of fault instances contained in the mini-batch sample B;

[0070] Select the initial value of the parameter:

[0071]

[0072] Iterative calculation is performed according to the gradient descent method, and iterative updates are performed as follows:

[0073]

[0074]

[0075] Where η is the learning rate;

[0076] After each iteration, the loss function of sample B is calculated as:

[0077]

[0078] when Continue iterating, when When , stop iteration;

[0079] Where m is a parameter that determines whether to stop the iteration.

[0080] When the iteration stops, the final parameter matrix is ​​obtained

[0081] When a water leakage alarm occurs, collect the status information of the fault and the water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0082]

[0083] Substitute into the objective function Get location information according to The corresponding location point is queried to complete the water leakage fault location and the fault instance is recorded in the second fault database to update the second fault database.

[0084] A fault locating device for a converter valve cooling system includes:

[0085] A plurality of water pressure sensors are provided in the converter valve cooling system;

[0086] A plurality of voltage sensors are provided in the converter valve cooling system;

[0087] A processor is connected to a plurality of water pressure sensors and a plurality of voltage sensors, and the processor executes the fault locating method of the converter valve cooling system.

[0088] In the fault locating method for the converter valve cooling system, the second locating accuracy is higher than the first locating accuracy.

[0089] Compared with the prior art, the present invention has the following advantages: the fault location method of the converter valve cooling system described in the present invention can identify and accurately locate the converter valve cooling system blockage fault, sensor fault, power supply fault, and water leakage fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Various other advantages and benefits of the present invention will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are intended only to illustrate preferred embodiments and are not to be construed as limiting the present invention. It should be understood that the drawings described below are merely examples of the present invention, and that those skilled in the art will be able to derive other drawings from these drawings without inventive effort. Throughout the drawings, identical reference numerals are used to denote identical components.

[0091] In the attached figure:

[0092] Figure 1 is a flow chart of a fault location method for a converter valve cooling system according to an embodiment of the present invention;

[0093] Figure 2 Schematic diagram of locating a water leakage point using a linear regression method of a fault locating method for a converter valve cooling system according to an embodiment of the present invention.

[0094] The present invention will be further explained below with reference to the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0095] The following will refer to the attached Figures 1 to 2Specific embodiments of the present invention will now be described in greater detail. Although specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention may be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to facilitate a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0096] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the invention. The scope of protection of the present invention shall be as defined in the attached claims.

[0097] To facilitate understanding of the embodiments of the present invention, further explanation will be given below using specific embodiments as examples in conjunction with the accompanying drawings, and the accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0098] For better understanding, Figures 1 to 2 As shown, in an embodiment of the present invention, a fault location method for a converter valve cooling system includes:

[0099] 1. Identify and classify the fault modes of the converter valve cooling system. According to the characteristics of the fault itself, the fault modes of the converter valve cooling system can be divided into water leakage fault, blockage fault, sensor fault, and power supply fault.

[0100] The fault modes are divided into faults of the first positioning accuracy that require simple positioning and faults of the second positioning accuracy that require detailed positioning. Faults that require simple positioning include blockage faults, sensor faults, and power supply faults, while faults that require detailed positioning include water leakage faults.

[0101] 2. For faults that require simple location, use the following methods:

[0102] 2.1 Collect historical fault information of the converter valve cooling system, including fault type, fault location information and status information, to form a first fault database.

[0103] The fault types include the blockage fault, sensor fault, and power supply fault mentioned above.

[0104] The fault location and status information are as follows:

[0105] Collect signals from the water pressure sensor and voltage sensor. The sensor signals include the sensor position signal x, the self-test signal y, and the alarm signal z. The position signal x = 1, 2, ..., m-1, m, m+1, ..., n (where m is the number of water pressure sensors and nm is the number of voltage sensors) and represents the sensor number. y = 0 indicates a sensor failure, and y = 1 indicates a normal sensor. Alarm signal z = 0 indicates an abnormal sensor detection and the sensor issues an alarm signal. z = 1 indicates normal sensor detection and the sensor does not issue an alarm signal.

[0106] Establish the first fault database for future reference.

[0107] For blockage faults, sensor faults, and power supply faults, when these three faults occur, water pressure and voltage sensor data are collected to form the following sensor signal matrix:

[0108]

[0109] 2.2 Real-time collection of water pressure and voltage sensor signals to form a sensor signal matrix for fault identification and location:

[0110]

[0111] Check if there is y in the second column of the matrix i =0(1≤i≤n), if yes, then sensor failure occurs, according to the position information x corresponding to the sensor i , the sensor can be located. If the second column y in the matrix i =1(i=1,2,…,n), then query the third column. If z exists in the third column, i =0(1≤i≤n), then query the position information x corresponding to the sensor i , if 1≤x i ≤m, a blocking fault occurs. If m+1≤× i ≤n, a power failure occurs, according to the position information x i Locate the sensor location and then locate the fault location.

[0112] 3. For faults that require detailed positioning, i.e. water leakage faults, use the following methods:

[0113] 3.1 Collect historical water leakage fault information of the converter valve cooling system, including fault type, fault location information and status information, to form a second fault database.

[0114] The fault type is water leakage.

[0115] The fault location information is as follows:

[0116] Use n points to divide the water pipe into equal parts, so that n positioning points 1, 2, ..., n are determined. The positioning point i (1≤i≤n) closest to the leakage point when the leakage fault occurs is used as the leakage fault positioning point, and the fault positioning information ×=i.

[0117] Establish a second fault database to collect the status information when the fault occurs. in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0118]

[0119] 3.2 Based on the second fault database, establish the objective function and solve it using linear regression.

[0120] 3.2.1 Establish the objective function.

[0121] as follows:

[0122]

[0123] in: is the position information calculated according to the objective function, w1, w2, …, w are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation.

[0124] Written in vector form as follows:

[0125]

[0126] in

[0127] Define the parameter matrix

[0128] 3.2.2 Establish a training set based on the fault database and define the loss function

[0129] In the fault database, a fixed number of small batch samples B are selected, which contains |B| number of fault instances.

[0130] The loss function of a single fault instance i (i∈B) is defined as:

[0131]

[0132] in is the fault location information of fault instance i calculated according to the objective function, y (i) is the real location information x of fault instance i.

[0133] The loss function of the small batch sample B is defined as:

[0134]

[0135] 3.2.3 Select the initial value of the parameter and perform iterative calculation according to the gradient descent method

[0136] Select the initial value of the parameter:

[0137]

[0138] Iterative updates are performed as follows:

[0139]

[0140] Right now:

[0141]

[0142] Where η is the learning rate,

[0143] After each iteration, the loss function of sample B is calculated as:

[0144]

[0145] when Continue iterating, when When , stop the iteration.

[0146] Where m is a manually set parameter used as a condition for stopping the iteration.

[0147] When the iteration stops, the final parameter matrix is ​​obtained

[0148] 3.3 When a water leakage fault alarm occurs, collect status information and confirm the fault point according to the objective function.

[0149] Collect status information when the fault occurs Water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0150]

[0151] Substitute into the objective function Get location information according to Query the corresponding location point to locate the water leakage fault. At the same time, record the fault instance in the fault database and update the fault database.

[0152] The above method realizes the fault identification and precise positioning of converter valve cooling system blockage faults, sensor faults, power supply faults, and precise positioning of water leakage faults.

[0153] In one embodiment, the converter valve is a flexible direct current converter valve.

[0154] In one embodiment, converter valve cooling system fault patterns are identified and classified. Based on the characteristics of the faults themselves, the fault patterns can be categorized into water leakage, blockage, sensor failure, and power supply failure. Fault patterns are categorized into those requiring simple location and those requiring detailed location. Simple location includes blockage, sensor, and power supply failures, while detailed location includes water leakage.

[0155] In one embodiment, for faults that require simple location, historical fault information of the converter valve cooling system is collected, including fault type, fault location information, and status information, to form a fault database.

[0156] The fault types include the blockage fault, sensor fault, and power supply fault mentioned above.

[0157] The fault location and status information are as follows:

[0158] Collect signals from the water pressure sensor and voltage sensor. The sensor signals include the sensor position signal x, the self-test signal y, and the alarm signal z. The position signal x = 1, 2, ..., m-1, m, m+1, ..., n (where m is the number of water pressure sensors and nm is the number of voltage sensors) and represents the sensor number. y = 0 indicates a sensor failure, and y = 1 indicates a normal sensor. Alarm signal z = 0 indicates an abnormal sensor detection and the sensor issues an alarm signal. z = 1 indicates normal sensor detection and the sensor does not issue an alarm signal.

[0159] Establish the first fault database for future reference.

[0160] For blockage faults, sensor faults, and power supply faults, when these three faults occur, water pressure and voltage sensor data are collected to form the following sensor signal matrix:

[0161]

[0162] The water pressure sensor and voltage sensor signals are collected in real time to form a sensor signal matrix as follows:

[0163]

[0164] Where m = 30, y2 = 0, y i =1(i≠2,1≤i≤35),zi =1(1≤i≤35), it can be determined that a sensor failure has occurred. The fault is in sensor number 2, and the fault can be located based on the number information.

[0165] In another embodiment, for a fault that requires detailed location, i.e., a water leakage fault, the water pipe is divided into equal parts using 100 points, thereby determining 100 location points 1, 2, ..., 100. The location point i (1≤i≤100) closest to the leakage point when the water leakage fault occurs is used as the water leakage fault location point, and the fault location information x=i.

[0166] Establish a fault database to collect status information when the fault occurs. Water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0167]

[0168] According to the fault database, the objective function is established and solved by linear regression.

[0169] Establish the objective function.

[0170] as follows:

[0171]

[0172] in: is the position information calculated according to the objective function, w1, w2, …, w are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation.

[0173] Written in vector form as follows:

[0174]

[0175] in

[0176] Let the parameter matrix

[0177] Establish a training set based on the fault database and define the loss function

[0178] In the fault database, a fixed number of small batch samples B are selected, which contains |B| number of fault instances.

[0179] The loss function of a single fault instance i (i∈B) is defined as:

[0180]

[0181] in is the fault location information of fault instance i calculated according to the objective function, y (i) is the real location information x of fault instance i.

[0182] The loss function of the small batch sample B is defined as:

[0183]

[0184] Select the initial value of the parameter and perform iterative calculation according to the gradient descent method

[0185] Select the initial value of the parameter:

[0186]

[0187] Iterative updates are performed as follows:

[0188]

[0189] Right now:

[0190]

[0191] Where η is the learning rate,

[0192] After each iteration, the loss function of sample B is calculated as:

[0193]

[0194] when Continue iterating, when When , stop the iteration.

[0195] Where m is a manually set parameter used as a condition for stopping the iteration.

[0196] When the iteration stops, the final parameter matrix is ​​obtained

[0197] According to a certain fault database, the parameter matrix is ​​finally obtained as:

[0198]

[0199] When a water leakage fault alarm occurs, status information is collected and the fault point is confirmed according to the objective function.

[0200] Collect status information when the fault occurs Water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0201]

[0202] Substitute into the objective function Get location information according to The corresponding locations 32 and 33 are queried to locate the water leakage fault. At the same time, the fault instance is recorded in the fault database and the fault database is updated.

[0203] A fault locating device for a converter valve cooling system includes:

[0204] Multiple water pressure sensors are installed in the converter valve cooling system.

[0205] Multiple voltage sensors are provided in the converter valve cooling system.

[0206] A processor is connected to a plurality of water pressure sensors and a plurality of voltage sensors, and the processor executes the fault locating method of the converter valve cooling system.

[0207] A computer-readable storage medium is configured to execute the fault location method for a converter valve cooling system.

[0208] A fault locating device for a converter valve cooling system comprises:

[0209] a first identification unit, configured to identify a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy of a fault mode of a converter valve cooling system, wherein historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information including fault type, fault location information, and status information; water pressure signals and voltage sensor signals of the converter valve cooling system are collected in real time to form a sensor signal matrix; and fault identification and fault location are performed based on the sensor signal matrix;

[0210] The second identification unit is used to identify and locate water leakage faults with a second positioning accuracy, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database, the historical water leakage fault information includes water leakage fault location information and status information, an objective function is established based on the second fault database, and a linear regression solution is used. When a water leakage fault alarm occurs, the status information is collected and the fault point is confirmed according to the objective function.

[0211] In a preferred embodiment of the fault location device for the converter valve cooling system, the first identification unit identifies and locates a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy, and the sensor signal matrix is:

[0212]

[0213] The water pressure signal and voltage sensor signal include the sensor position signal x, the self-test signal y and the alarm signal z, respectively, with subscripts 1, 2,…, m-1, m, m+1,…, n, where m is the number of water pressure sensors, nm is the number of voltage sensors, y is 0 for a sensor failure, y is 1 for a sensor normal, the alarm signal z is 0 for an abnormality in the information detected by the sensor and the sensor issues an alarm signal, and z is 1 for a normality in the information detected by the sensor and the sensor does not issue an alarm signal.

[0214] In a preferred embodiment of the fault location device for the converter valve cooling system, the first identification unit queries the sensor signal matrix to see whether there is a y in the second column. i is 0, 1≤i≤n. If yes, a sensor failure occurs. According to the position information x corresponding to the sensor i Position the sensor. If the second column y in the matrix i is 1, i=1, 2, ..., n, then query the third column, if the third column contains z i If it is 0, 1≤i≤n, then query the position information x corresponding to the sensor i , if 1≤× i ≤m, a blocking fault occurs, if m+1≤x i ≤n, a power failure occurs, according to the position information x i Locate the sensor to locate the fault.

[0215] In a preferred embodiment of the fault locating device for the converter valve cooling system, the second identification unit identifies and locates a water leakage fault with a second positioning accuracy, divides the water pipe of the converter valve cooling system into n equal parts to determine n positioning points 1, 2, ..., n, and takes the positioning point i closest to the leakage point when the water leakage fault occurs, 1≤i≤n, as the water leakage fault positioning point, and the fault positioning information x is i.

[0216] In a preferred embodiment of the fault location device for the converter valve cooling system, the second identification unit collects state information when the water leakage fault occurs, which includes the inlet water temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form a state matrix:

[0217]

[0218] A second fault database is constructed based on the state matrix, and an objective function is established:

[0219]

[0220] in: is the position information calculated according to the objective function, w1, w2, ..., w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, b is the deviation,

[0221] Written in vector form as follows:

[0222]

[0223] in

[0224] Define the parameter matrix

[0225] In a preferred embodiment of the fault location device for the converter valve cooling system, the second identification unit establishes a training set based on the second fault database, defines a loss function,

[0226] Select a fixed number of small batch samples B, which contains |B| number of fault instances.

[0227] The loss function for a single fault instance i(i∈B) for:

[0228]

[0229] in is the loss function of the i-th fault instance, is the fault location information of fault instance i calculated according to the objective function, y (i) is the real position information of fault instance i×, w1, w2, ..., w6 are the weights of inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level respectively, b is the deviation, is the parameter matrix;

[0230] Loss function for mini-batch samples B for:

[0231]

[0232] in The loss function of the mini-batch sample B, |B| is the number of fault instances contained in the mini-batch sample B;

[0233] Select the initial value of the parameter:

[0234]

[0235] Iterative calculation is performed according to the gradient descent method, and iterative updates are performed as follows:

[0236]

[0237] Where η is the learning rate,

[0238] After each iteration, the loss function of sample B is calculated as:

[0239]

[0240] when Continue iterating, when When , stop the iteration.

[0241] Where m is a parameter that determines whether to stop the iteration.

[0242] When the iteration stops, the final parameter matrix is ​​obtained

[0243] When a water leakage alarm occurs, collect the status information of the fault and the water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix:

[0244]

[0245] Substitute into the objective function Get location information according to The corresponding location point is queried to complete the water leakage fault location and the fault instance is recorded in the second fault database to update the second fault database.

[0246] A fault locating device for a converter valve cooling system includes:

[0247] Multiple water pressure sensors are installed in the converter valve cooling system.

[0248] Multiple voltage sensors are provided in the converter valve cooling system.

[0249] A processor is connected to a plurality of water pressure sensors and a plurality of voltage sensors, and the processor executes the fault locating method of the converter valve cooling system.

[0250] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal translation scripting language JavaScript, etc.

[0251] It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0252] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0253] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0254] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0255] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A fault location method for a converter valve cooling system, characterized in that: It includes: Identifying a first-level positioning accuracy of a converter valve cooling system failure mode, including a blockage fault, a sensor fault, and a power supply fault, wherein historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information including fault type, fault location information, and status information; collecting water pressure signals and voltage sensor signals of the converter valve cooling system in real time to form a sensor signal matrix; and performing fault identification and fault location based on the sensor signal matrix; Identifying and locating water leakage faults with a second positioning accuracy, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database, the historical water leakage fault information including water leakage fault location information and status information. An objective function is established based on the second fault database and solved using linear regression. When a water leakage fault alarm occurs, status information is collected and the fault point is confirmed based on the objective function. Establish a training set based on the second fault database and define a loss function; Select a fixed number of small batch samples B, which contains |B| number of fault instances; Single failure instance The loss function for: ， in For the The loss function for each fault instance is For fault instances According to the fault location information calculated by the objective function, For fault instances The real position information x, w1, w2, ..., w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively. b is the deviation. is the parameter matrix; Loss function for mini-batch samples B for: ， in The loss function of the mini-batch sample B, |B| is the number of fault instances contained in the mini-batch sample B; Select the initial value of the parameter: ， Iterative calculation is performed according to the gradient descent method, and iterative updates are performed as follows: ， ， Where η is the learning rate; After each iteration, the loss function of sample B is calculated as: ， when , continue to iterate, when When , stop iteration; Where m is a parameter that determines whether to stop the iteration. When the iteration stops, the final parameter matrix is ​​obtained ; When a water leakage alarm occurs, collect the status information of the fault and the water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix: ， Substitute into the objective function , get location information ,according to The corresponding location point is queried to complete the water leakage fault location and the fault instance is recorded in the second fault database to update the second fault database.

2. The fault location method of the converter valve cooling system according to claim 1, wherein: Identify and locate blockage faults, sensor faults, and power supply faults for first-level positioning accuracy. Sensor signal matrix: ,in, The water pressure signal and voltage sensor signal include the sensor position signal x, the self-test signal y and the alarm signal z, respectively. The subscripts are 1, 2,…, m-1, m, m+1,…, n, where m is the number of water pressure sensors, nm is the number of voltage sensors, y is 0 for a sensor failure, y is 1 for a sensor normal, the alarm signal z is 0 for an abnormal sensor detected information and the sensor issues an alarm signal, and z is 1 for a normal sensor detected information and the sensor does not issue an alarm signal.

3. The fault location method of the converter valve cooling system according to claim 2, wherein: Query the sensor signal matrix to see if the second column has y i is 0, 1≤i≤n. If yes, a sensor failure occurs. According to the position information x corresponding to the sensor i Position the sensor. If the second column y in the matrix i is 1, i=1, 2, …, n, then query the third column, if the third column exists z i If it is 0, 1≤i≤n, then query the position information x corresponding to the sensor i , if 1≤x i ≤m, a blocking fault occurs, if m+1≤x i ≤n, a power failure occurs, according to the position information x i Locate the sensor to locate the fault.

4. The fault location method of the converter valve cooling system according to claim 1, wherein: In identifying and locating water leakage faults with the second positioning accuracy, n points are used to equally divide the water pipes of the converter valve cooling system into n positioning points 1, 2, ..., n. The positioning point i closest to the leakage point when the water leakage fault occurs, 1≤i≤n, is used as the water leakage fault positioning point, and the fault positioning information x is i.

5. The fault location method of the converter valve cooling system according to claim 4, wherein: Collect status information when water leakage occurs, including water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form a state matrix: ， A second fault database is constructed based on the state matrix, and an objective function is established: ， in: is the position information calculated according to the objective function, w1, w2, …, w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation; Written in vector form as follows: ， in , Define the parameter matrix .

6. A computer-readable storage medium, characterized in that: The method is configured to execute the fault locating method for the converter valve cooling system according to any one of claims 1 to 5.

7. A fault location device for a converter valve cooling system, characterized in that: It includes: a first identification unit, configured to identify a blockage fault, a sensor fault, and a power supply fault with a first positioning accuracy of a fault mode of a converter valve cooling system, wherein historical fault information of the converter valve cooling system is collected to form a first fault database, the historical fault information including fault type, fault location information, and status information; water pressure signals and voltage sensor signals of the converter valve cooling system are collected in real time to form a sensor signal matrix; and fault identification and fault location are performed based on the sensor signal matrix; a second identification unit, configured to identify and locate a water leakage fault with a second positioning accuracy, wherein historical water leakage fault information of the converter valve cooling system is collected to form a second fault database, the historical water leakage fault information including water leakage fault location information and status information; an objective function is established based on the second fault database and solved using linear regression; when a water leakage fault alarm occurs, the status information is collected and the fault point is confirmed according to the objective function; The second recognition unit establishes a training set based on the second fault database and defines a loss function; Select a fixed number of small batch samples B, which contains |B| number of fault instances; Single failure instance The loss function for: ， in For the The loss function for each fault instance is For fault instances According to the fault location information calculated by the objective function, For fault instances The real position information x, w1, w2, ..., w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively. b is the deviation. is the parameter matrix; Loss function for mini-batch samples B for: ， in The loss function of the mini-batch sample B, |B| is the number of fault instances contained in the mini-batch sample B; Select the initial value of the parameter: ， Iterative calculation is performed according to the gradient descent method, and iterative updates are performed as follows: ， ， Where η is the learning rate; After each iteration, the loss function of sample B is calculated as: ， when , continue to iterate, when When , stop iteration; Where m is a parameter that determines whether to stop the iteration. When the iteration stops, the final parameter matrix is ​​obtained ; When a water leakage alarm occurs, collect the status information of the fault and the water inlet temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form the following state matrix: ， Substitute into the objective function , get location information ,according to The corresponding location point is queried to complete the water leakage fault location and the fault instance is recorded in the second fault database to update the second fault database.

8. The fault location device for a converter valve cooling system according to claim 7, wherein: The first identification unit identifies and locates a blocking fault, a sensor fault, and a power supply fault with a first positioning accuracy. The sensor signal matrix: ,in, The water pressure signal and voltage sensor signal include the sensor position signal x, the self-test signal y and the alarm signal z, respectively. The subscripts are 1, 2,…, m-1, m, m+1,…, n, where m is the number of water pressure sensors, nm is the number of voltage sensors, y is 0 for a sensor failure, y is 1 for a sensor normal, the alarm signal z is 0 for an abnormal sensor detected information and the sensor issues an alarm signal, and z is 1 for a normal sensor detected information and the sensor does not issue an alarm signal.

9. The fault locating device for a converter valve cooling system according to claim 8, wherein: The first recognition unit queries the sensor signal matrix for the presence of y in the second column. i is 0, 1≤i≤n. If yes, a sensor failure occurs. According to the position information x corresponding to the sensor i Position the sensor. If the second column y in the matrix i is 1, i=1, 2,…, n, then query the third column. If z exists in the third column, i If it is 0, 1≤i≤n, then query the position information x corresponding to the sensor i , if 1≤x i ≤m, a blocking fault occurs, if m+1≤x i ≤n, a power failure occurs, according to the position information x i Locate the sensor to locate the fault.

10. The fault location device for a converter valve cooling system according to claim 7, wherein: The second identification unit identifies and locates a water leakage fault with a second positioning accuracy, uses n points to equally divide the water pipes of the converter valve cooling system into n positioning points 1, 2, ..., n, and takes the positioning point i closest to the leakage point when the water leakage fault occurs, 1≤i≤n, as the water leakage fault positioning point, and the fault positioning information x is i.

11. The fault location device for a converter valve cooling system according to claim 10, wherein: The second identification unit collects status information when a water leakage fault occurs, including the inlet water temperature t in , outlet water temperature t out , inlet valve pressure p in , the main pump operating power P, the internal cooling water flow Q, the expansion tank water level f, form a state matrix: ， A second fault database is constructed based on the state matrix, and an objective function is established: ， in: is the position information calculated according to the objective function, w1, w2, …, w6 are the weights of the inlet water temperature, outlet water temperature, inlet valve pressure, main pump operating power, internal cooling water flow, and expansion tank water level, respectively, and b is the deviation; Written in vector form as follows: ， in , Define the parameter matrix .

12. A fault location device for a converter valve cooling system, characterized in that: It includes, A plurality of water pressure sensors are provided in the converter valve cooling system; A plurality of voltage sensors are provided in the converter valve cooling system; A processor is connected to a plurality of water pressure sensors and a plurality of voltage sensors, and the processor executes the fault locating method for the converter valve cooling system according to any one of claims 1 to 5.

Images