Intelligent temperature transmitter automatic line selection method with fault self-checking function

Abstract

The invention discloses an automatic line selection method for an intelligent temperature transmitter with fault self-detection, and the method is based on an automatic line selection module, and the automatic line selection module comprises first to sixth change-over switches and a platinum resistor temperature detector. When excitation is applied to the first to sixth change-over switches according to a preset excitation signal combination, a predetermined measurement value is obtained, and then it is judged that the A terminal and the B terminal are connected with an equipotential line of the three-wire platinum thermistor, and the C terminal is connected with a common line of the three-wire platinum thermistor. The input lines are switched through the cascade change-over switch, all possible line selection modes are automatically switched, the wiring sequence is intelligently recognized according to detection data, wiring is automatically switched to the correct line sequence, the field wiring work efficiency is improved, and the fault rate caused by wiring errors can be reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of power transformers and relates to an automatic line selection method for an intelligent temperature transmitter with fault self-detection. Background Art

[0002] At present, three-wire platinum thermal resistors are widely used in the field of power transformer temperature measurement. The "three wires" can be divided into two equipotential wires and one "common wire". When used on site, the three-wire platinum thermal resistor is usually connected to the temperature transmitter. The "three wires" of the three-wire platinum thermal resistor need to be connected one by one with the ports on the transmitter. In actual use, it is necessary to manually check the wiring status. This method is not only time-consuming, but also prone to wrong wiring, resulting in abnormal temperature display, causing unnecessary trouble, and greatly affecting on-site work efficiency. At the same time, the existing temperature transmitter does not have a remote self-test function, and cannot achieve remote self-test and fault location, delaying the time for fault elimination and forming unnecessary hidden dangers. Summary of the invention

[0003] The present invention discloses an automatic line selection method for an intelligent temperature transmitter with fault self-checking, wherein the automatic line selection method for an intelligent temperature transmitter with fault self-checking is based on an automatic line selection module, wherein the automatic line selection module comprises first to sixth switching switches and a platinum resistance temperature detector;

[0004] The normally closed contacts of the first to third switching switches are all connected to the A terminal, and the normally closed contacts are all connected to the B terminal. The output ends of the first to third switching switches are respectively connected to the normally closed contacts of the fourth to sixth switching switches, and the normally open contacts of the fourth to sixth switching switches are all connected to the C terminal.

[0005] The A terminal, the B terminal and the C terminal are respectively connected to any one wire of a three-wire platinum thermal resistor;

[0006] The output end of the fourth switch is connected to the FORCE+ terminal of the platinum resistance temperature detector, the output end of the fifth switch is connected to the RTDIN+ terminal of the platinum resistance temperature detector, and the output end of the sixth switch is connected to the RTDIN- and FORCE- terminals of the platinum resistance temperature detector;

[0007] When the first to sixth switching switches are excited according to the preset excitation signal combination, a predetermined measurement value is obtained, and it is determined that the A terminal and the B terminal are connected to the equipotential line of the three-wire platinum thermal resistor, and the C terminal is connected to the common line of the three-wire platinum thermal resistor.

[0008] Furthermore, the preset excitation signal combination includes line sequence 1 and line sequence 2. In the line sequence 1, the excitation signal of the first switch is low, the excitation signal of the second switch is high, the excitation signal of the fourth switch is low, the excitation signal of the fifth switch is low, and the excitation signal of the sixth switch is high. The measurement value corresponding to the line sequence 1 is +;

[0009] In the line sequence 2, the excitation signal of the first switching switch is low level, the excitation signal of the third switching switch is high level, the excitation signal of the fourth switching switch is low level, the excitation signal of the fifth switching switch is high level, and the excitation signal of the sixth switching switch is low level. The measurement value corresponding to the line sequence 2 is +.

[0010] Furthermore, in the line sequence 3, the excitation signal of the first switch is high, the excitation signal of the second switch is low, the excitation signal of the fourth switch is low, the excitation signal of the fifth switch is low, and the excitation signal of the sixth switch is high. The measurement value corresponding to the line sequence 3 is -;

[0011] In the line sequence 5, the excitation signal of the second switch is low level, the excitation signal of the third switch is high level, the excitation signal of the fourth switch is high level, the excitation signal of the fifth switch is low level, the excitation signal of the sixth switch is low level, and the measurement value corresponding to the line sequence 5 is 0;

[0012] Terminals A and C are connected to the equipotential lines, and terminal B is connected to the common line.

[0013] Furthermore, in the line sequence 4, the excitation signal of the first switch is high, the excitation signal of the third switch is high, the excitation signal of the fourth switch is low, the excitation signal of the fifth switch is high, and the excitation signal of the sixth switch is low, and the measurement value corresponding to the line sequence 4 is -;

[0014] In the line sequence 6, the excitation signal of the second switch is high, the excitation signal of the third switch is low, the excitation signal of the fourth switch is high, the excitation signal of the fifth switch is low, and the excitation signal of the sixth switch is low. The measurement value corresponding to the line sequence 6 is 0;

[0015] The B and C terminals are connected to the equipotential lines, and the A terminal is connected to the common line.

[0016] Furthermore, the automatic line selection module includes a seventh switch, the normally closed contact of the seventh switch is connected to the output end of the sixth switch, the normally open contact of the seventh switch is connected to the output end of the fifth switch through a resistor, and the output end of the seventh switch is connected to the RTDIN- and FORCE- terminals of the platinum resistance temperature detector;

[0017] When the excitation signal of the seventh switch is at a high level, if the temperature value output by the platinum resistance temperature detector is 0°C, the fault is that the three-wire platinum resistance function is abnormal, otherwise it is a fault of the intelligent temperature transmitter.

[0018] Furthermore, the platinum resistance temperature detector further comprises an excitation resistor, the excitation resistor is arranged between the BIAS terminal and the REFIN- terminal of the platinum resistance temperature detector, and the BIAS terminal is connected to the REFIN+ terminal;

[0019] The excitation resistor is used to calculate the resistance of the platinum resistor:

[0020] R p =R4ΔV / (V REFIN+ -V REFIN- )

[0021] V REFIN+ is the voltage of REFIN+ terminal of platinum resistance temperature detector, V REFIN- is the voltage at the REFIN-terminal of the platinum resistance temperature detector, R4 is the resistance of the excitation resistor, and ΔV is the voltage difference:

[0022] ΔV=IR p

[0023] I provides excitation current for the platinum resistor, R p is the resistance value of platinum resistor.

[0024] Furthermore, when the measured temperature T∈(0,850):

[0025] R p =R0(1+aT+bT 2 )

[0026] R0 is the resistance value at 0°C, T is the measured temperature, a is the first platinum material coefficient, b is the second platinum material coefficient, usually a=3.90830×10 -3 ℃ -1 , b = -5.77500 × 10 -7 ℃ -2 .

[0027] Furthermore, when the measured temperature T∈(-200,0):

[0028] R p =R0(1+aT+bT 2 +c(T-100)T 3 )

[0029] Where R0 is the resistance value at 0°C, T is the measured temperature, a is the first platinum material coefficient, b is the second platinum material coefficient, and c is the third platinum material coefficient. Usually a=3.90830×10 -3 ℃ -1 , b = -5.77500 × 10 -7 ℃ -2 , c = -4.18301 × 10 -12 ℃ -1 .

[0030] Furthermore, the platinum resistance temperature correction method comprises the following steps:

[0031] Step 1: Place the platinum resistor equipped with the temperature transmitter in a constant temperature box. The platinum resistor temperature detector of the temperature transmitter measures the voltage value ΔV corresponding to different temperatures. Perform multiple tests in sequence to form a ΔV-T mapping table between the ambient temperature and the voltage difference between the REFIN+ and REFIN- terminals.

[0032] Step 2: Import the ΔV-T mapping table as sample data. The system selects the sample data file and reads the sample data. Suppose n sets of data are collected. Where, ΔV i is the ith measured voltage difference, T i is the i-th measured temperature value;

[0033] Step 3: Normalize the sample data, set the data of T∈(0,850) as data set 1, and set the data of T∈(-200,0) in the sample data as data set 2;

[0034] Step 4: Preliminarily determine whether the ambient temperature T is a positive value. If so, select 70% of the data in data set 1 as the training set and 30% of the data as the test set; if not, select 70% of the data in data set 2 as the test set and 30% of the data as the test set;

[0035] Step 5: Use the training set to train the radial basis function network model and determine the network parameters;

[0036] Step 6: Use the test set to evaluate the model performance, using the mean square error (MSE) as the evaluation indicator;

[0037]

[0038] Among them, n t is the number of samples in the test set, y i.pred is the model's predicted value for the i-th test sample, y i is the true value of the i-th test sample;

[0039] Step 7: The voltage difference ΔV newly measured for the platinum resistance temperature detector new , normalize it and calculate the model output to get the corrected temperature value T new.corrected , and then denormalize it to get the corrected actual temperature value;

[0040] Step 8. Save the data.

[0041] Furthermore, the input layer of the radial basis function network model is 1 neuron, that is, the normalized ΔV value, the output layer is 1 neuron, that is, the corrected T value, and the hidden layer is 10 neurons. The hidden layer uses the K-Means clustering algorithm to determine the center vector c of the hidden layer neurons from the voltage difference data of the training set. j ;

[0042] The output of the hidden layer neurons uses a radial basis function in the form of a Gaussian function, namely:

[0043]

[0044] Among them, σ j is the width parameter of the function. The weights between the hidden layer and the output layer are calculated using the least squares method;

[0045] The model after compensation by the temperature nonlinear correction method is:

[0046] f(T)+ΔR=430ΔV / (V REFIN+ -V REFIN- )

[0047] Wherein f(T) is equation (1) or equation (2), and ΔR is the difference between the total resistance of RTDIN+ line and the total resistance of the common line.

[0048] The beneficial effects achieved by the present invention are:

[0049] The present invention switches input lines through cascade switching switches, automatically switches all possible line selection modes, intelligently identifies wiring sequence according to detection data, and automatically switches wiring to a correct sequence, thereby improving on-site wiring work efficiency and helping to reduce the failure rate caused by wiring errors.

[0050] The present invention automatically detects the circuit through standard resistance and effectively realizes fault location.

[0051] The present invention uses a multi-layer perceptron model for software compensation, which helps to reduce the variable fixed deviation caused by contact resistance wear, pollution, temperature and humidity changes of the cascade switching switch to the measurement of the platinum resistor, and reduces the errors caused by not considering the material characteristics of the platinum resistor, temperature unevenness, electromagnetic interference, fitting accuracy, high-order effects, etc. when solving the temperature using an ideal equation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of an automatic line selection method for an intelligent temperature transmitter with fault self-detection;

[0053] Figure 2 The invention discloses an automatic line selection calculation process of an automatic line selection method of an intelligent temperature transmitter with fault self-checking;

[0054] Figure 3 It is a wiring diagram of automatic line selection line sequence 1 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0055] Figure 4 It is a wiring diagram of automatic line selection line sequence 2 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0056] Figure 5 It is a wiring diagram of automatic line selection line sequence 3 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0057] Figure 6 It is a wiring diagram of automatic line selection line sequence 4 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0058] Figure 7 It is a wiring diagram of automatic line selection line sequence 5 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0059] Figure 8 It is a wiring diagram of automatic line selection line sequence 6 in an automatic line selection method of an intelligent temperature transmitter with fault self-detection;

[0060] Fig. 9 It is a fault self-checking flow chart of an automatic line selection method of an intelligent temperature transmitter with fault self-checking;

[0061] Fig.10 The present invention is a flow chart of temperature nonlinear correction for an automatic line selection method of an intelligent temperature transmitter with fault self-detection. DETAILED DESCRIPTION

[0062] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, these embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications and replacements all fall within the scope of protection of the present invention.

[0063] Embodiment 1

[0064] like Figure 1As shown, the present invention provides an automatic line selection method for an intelligent temperature transmitter with fault self-detection, and the method is implemented based on an automatic line selection module.

[0065] The automatic line selection module is used to detect the wiring status, automatically adjust the line sequence, and convert the voltage into digital quantity and transmit it to the microcontroller.

[0066] The microcontroller is used to output excitation to the automatic line selection module, judge the wiring status through calculation, calculate the measured temperature value, and issue an alarm signal.

[0067] User interface, used to display and confirm line selection results and display alarm status.

[0068] The automatic line selection module is used to detect the wiring status, automatically adjust the line sequence, and convert the voltage into digital quantity and transmit it to the microcontroller.

[0069] The automatic line selection module is connected to a microcontroller, and the user interface reads the data and control instructions of the microcontroller via Bluetooth or WIFI.

[0070] The automatic line selection module consists of 7 switches and a platinum resistance temperature detector. The schematic diagram of the automatic line selection module is as follows: Figure 1 In this embodiment, the switch model is the relay HF32F / 005-ZS3, and the platinum resistance temperature detector model is the digital output converter MAX31865.

[0071] Switches U1~U7 respectively receive excitation signals S1~S7 from the controller, the input ends of switches U1~U3 are respectively connected to end A and end B, the output ends are connected to one of the input ends of switches U4~U6, the other input ends of switches U4~U6 are connected to end C, the output end of switch U4 is connected to FORCE+ of the platinum resistance temperature detector, the output end of switch U5 is connected to RTDIN+ of the platinum resistance temperature detector and the standard resistor R5, the output end of switch U6 is connected to one of the input ends of switch U7, the standard resistor R7 is connected to the other input end of switch U7, and the output end of switch U7 is connected to RTDIN- and FORCE- of the platinum resistance temperature detector.

[0072] In this embodiment, the RTDIN+ and RTDIN- terminals of the platinum resistance temperature detector in the automatic line selection module form a pair of differential inputs, which can effectively offset the noise and interference introduced by the long wires, improve the signal-to-noise ratio of the signal, and thus obtain a more accurate temperature reading.

[0073] When the A, B, and C terminals of the automatic line selection module are connected to any of the three terminals of the three-wire platinum thermal resistor, the intelligent temperature transmitter performs the line selection process as follows:

[0074] The microcontroller applies excitation signals to switches U1 to U6 according to the six combinations shown in Table 1, where 0 is a low level and 1 is a high level.

[0075] Table 1 Excitation signal combination

[0076]

[0077]

[0078] The platinum resistance temperature detector of the automatic line selection module detects the voltage difference between RTDIN+ and RTDIN-, amplifies and processes it, and converts the detected signal into a digital signal and sends it to the microcontroller.

[0079] The microcontroller calculates the voltage difference ΔV under different combinations.

[0080] Furthermore, FORCE+ and FORCE- constitute the positive and negative phase ports of the differential excitation signal, providing an excitation current I to the platinum resistor, and the resistance of the platinum resistor is R p .

[0081] Furthermore, when the voltage difference between RTDIN+ and RTDIN- is positive, the corresponding Figure 3 The line sequence shown is 1 and Figure 4 As shown in line sequence 2, for line sequence 1 and line sequence 2, the current source is connected to the A pin and B pin of the three-wire platinum thermal resistor respectively, which has no effect on the measurement of ΔV:

[0082] ΔV=IR p

[0083] I provides excitation current for the platinum resistor, R p is the resistance value of platinum resistor.

[0084] When the voltage difference between RTDIN+ and RTDIN- is negative, the corresponding Figure 5 The line sequence shown is 3 and Figure 6 As shown in line sequence 4, for line sequence 3 and line sequence 4, the current source is connected to the A pin and B pin of the three-wire platinum thermal resistor respectively, which has no effect on the measurement of RTDIN+:

[0085] ΔV=-IR p

[0086] When the voltage difference between RTDIN+ and RTDIN- is zero, the corresponding Figure 7 The line sequence shown is 5 and Figure 8 As shown in line sequence 6, for line sequence 5 and line sequence 6, RTDIN+ is connected to the A pin and B pin of the three-wire platinum thermal resistor respectively, which has no effect on the measurement of RTDIN+:

[0087] ΔV=0

[0088] When the user clicks the "Start Line Sequence Detection" button on the menu bar, the microcontroller controls the above line sequence detection program to run; when the line sequence detection program detects that the current RTDIN+ pin voltage value is positive during operation, the user interface pops up a pop-up box, prompting "The current line sequence is the correct line sequence"; after the user clicks the pop-up box, the microcontroller sends a trigger control signal to maintain the switch state as the normal wiring line sequence. At this time, ΔV=IR p According to the measured value and the current I value, the resistance value R of the platinum resistor can be obtained. p .

[0089] The excitation resistor R4 is connected between the BIAS terminal and REFIN- terminal of the platinum resistance temperature detector. R4 is usually 430Ω. The current I=(V REFIN+ -V REFIN- ) / 430. V REFIN+ is the voltage of REFIN+ terminal of platinum resistance temperature detector, V REFIN- It is the voltage of REFIN- terminal of platinum resistance temperature detector.

[0090] Therefore, the resistance value of the platinum resistor R p =430ΔV / (V REFIN+ -V REFIN- )

[0091] The microcontroller calculates the corresponding temperature value based on the resistance value of the platinum resistor.

[0092] Furthermore, in this embodiment, the platinum resistor is PT100, and the measurement range is -200°C-850°C. Assuming that the measured temperature T∈(0,850), the formula method is applied to find the equation:

[0093] R p =R0(1+aT+bT 2 ) (1)

[0094] Where R0 is the resistance value at 0°C, R0 = 100Ω, a is the first platinum material coefficient, b is the second platinum material coefficient, a = 3.90830 × 10 -3 ℃ -1 ; b = -5.77500 × 10 -7 ℃ -2 .

[0095] If the temperature value obtained is T∈(0,850), the process ends and the user interface displays the measured temperature value. Otherwise, assuming T∈(-200,0), the Newton-Raphson method is used to solve the equation:

[0096] R p =R0(1+aT+bT 2 +c(T-100)T 3 ) (2)

[0097] Where c is the third platinum material coefficient, c = -4.18301 × 10 -12 ℃ -1 .

[0098] If the obtained temperature value T∈(-200, 0), the process ends and the user interface displays the measured temperature value. Otherwise, the user interface prompts an alarm message.

[0099] The above realizes that any wiring method can be connected to the intelligent temperature transmitter to achieve adaptive matching and calculate the correct temperature value.

[0100] Embodiment 2

[0101] In this embodiment, a method is provided as follows: Fig. 9 The fault self-check method shown in the figure. When the temperature data reading on the user interface is abnormal, perform the fault self-check according to the following process:

[0102] In the first step, the microcontroller applies an excitation signal S7=0 to the switch U7, executes the automatic line selection process according to Table 1, and adaptively adjusts the line selection mode to the correct line sequence.

[0103] When the connection is correct, the circuit connecting FORCE+ and RTDIN+ should be an equipotential line, and the circuit connecting FORCE- and RTDIN- should be a common line.

[0104] Apply excitation signals to switches U1~U6 according to the 6 combinations in Table 1, where 0 is a low level and 1 is a high level. Take line sequence 1 as an example: U1 and U4 excitation signals are 0, the switches are connected to normally closed contacts, and the A terminal signal is connected to FORCE+; U2 excitation signal is 1, the switch is actuated, and the B terminal is connected to U2; U5 excitation signal is 0, the switch is connected to normally closed contacts, and the B terminal is connected to U5 as the output of U2. At this time, the B terminal signal is connected to RTDIN+; U3 inputs any signal, U6 excitation signal is 1, the switch is actuated, and the C terminal signal is connected to FORCE- and RTDIN-.

[0105] The circuit can be simplified as Figure 3 The voltage value calculated for this topology is +Ir p .

[0106] By analogy, the voltage value calculated according to line sequence 2 is +Ir p ; The voltage value calculated according to line sequence 3 is -IR p ; The voltage value calculated according to line sequence 4 is -IR p ; The voltage value calculated according to line sequence 5 is 0; the voltage value calculated according to line sequence 6 is 0.

[0107] Only line sequence 1 and line sequence 2 are correct line sequences, that is, the voltage value is + for the correct control mode. When the input is in accordance with the excitation of line sequence 1 and 2, the measured value is +IR p This means that the A and B terminals are connected to the equipotential line, and the C terminal is connected to the common line. If the voltage value is the same as the measured value corresponding to the line sequence 3 and 5, it means that the A and C terminals are connected to the equipotential line, and the B terminal is connected to the common line; if the voltage value is the same as the measured value corresponding to the line sequence 4 and 6, it means that the B and C terminals are connected to the equipotential line, and the A terminal is connected to the common line.

[0108] The second step is to determine whether the user interface temperature returns to normal. If it is normal, the fault is caused by an incorrect line selection method, otherwise it is other faults.

[0109] In the third step, the microcontroller applies an excitation signal S7=1 to the switch U7. At this time, the standard resistor R5 (R5=100Ω) is connected to the RTDIN+ and RTDIN- terminals of the platinum resistance temperature detector.

[0110] Step 4: If the temperature value displayed on the user interface is 0°C, the fault is caused by abnormal function of the platinum resistance, otherwise it is caused by a fault of the intelligent temperature transmitter.

[0111] Optionally, in this embodiment, a 100nF capacitor is connected between the RTDIN+ and RTDIN- terminals of the platinum resistance temperature detector to improve signal quality and anti-interference capability, and ensure that the signal received from the platinum resistance is as pure as possible.

[0112] Optionally, in this embodiment, a 100nF decoupling capacitor is connected between the power supply and ground of the platinum resistance temperature detector to absorb high-frequency noise on the power line, thereby reducing the impact on the platinum resistance temperature detector, and providing instantaneous current when the internal current changes suddenly to avoid large fluctuations in the power supply voltage of the platinum resistance temperature detector, and by reducing the voltage ripple on the power line, the overall stability of the system is improved.

[0113] Embodiment 3

[0114] like Fig.10 As shown, when calculating the resistance of the platinum resistor in Example 1, it is assumed that the resistances on the three lines connected to the platinum resistor temperature detector are equal, without considering that the cascade switch actually has contact resistance. Switches from different batches, different contact wear, pollution, temperature and humidity changes, etc. will cause the contact resistances of the switches on the three lines to be different, which will cause deviations in the calculation of the platinum resistor. This embodiment provides a correction method that takes into account the errors caused by the material characteristics of the platinum resistor, temperature unevenness, electromagnetic interference, fitting accuracy, high-order effects, etc., and reduces the above errors through correction, which specifically includes the following steps:

[0115] Step 1, place the platinum resistor equipped with the temperature transmitter in a constant temperature box, and the platinum resistance temperature detector of the temperature transmitter measures the voltage value ΔV corresponding to different temperatures, and conduct multiple tests in sequence to form a ΔV-T mapping table (ΔV-T table) between the ambient temperature and the voltage difference between the REFIN+ and REFIN- terminals;

[0116] Step 2: Import the ΔV-T mapping table as sample data. The system selects the sample data file and reads the sample data. Suppose n sets of data are collected. Where, ΔV i is the ith measured voltage difference, T i is the i-th measured temperature value;

[0117] Step 3: Normalize the sample data, set the data of T∈(0,850) as data set 1, and set the data of T∈(-200,0) in the sample data as data set 2;

[0118] Step 4: Preliminarily determine whether the ambient temperature T is a positive value. If so, select 70% of the data in data set 1 as the training set and 30% of the data as the test set; if not, select 70% of the data in data set 2 as the test set and 30% of the data as the test set;

[0119] Step 5: Use the training set to train the radial basis function network model and determine the network parameters;

[0120] The input layer of the radial basis function network model is 1 neuron, that is, the normalized ΔV value, the output layer is 1 neuron, that is, the corrected T value, and the hidden layer is 10 neurons. The hidden layer uses the K-Means clustering algorithm to determine the center vector c of the hidden layer neurons from the voltage difference data of the training set. j The output of the hidden layer neurons uses a radial basis function in the form of a Gaussian function, namely:

[0121]

[0122] Among them, σ j is the width parameter of the function. The weights between the hidden layer and the output layer are calculated using the least squares method.

[0123] The model after compensation by the temperature nonlinear correction method is:

[0124] f(T)+ΔR=430ΔV / (V REFIN+ -V REFIN- )

[0125] Wherein f(T) is equation (1) or equation (2), and ΔR is the difference between the total resistance of RTDIN+ line and the total resistance of the common line.

[0126] Step 6: Use the test set to evaluate the model performance, using the mean square error (MSE) as the evaluation indicator;

[0127]

[0128] Among them, n t is the number of samples in the test set, y i.pred is the model's predicted value for the i-th test sample, y i is the true value of the i-th test sample.

[0129] Step 7: The voltage difference ΔV newly measured for the platinum resistance temperature detector new , normalize it and calculate the model output to get the corrected temperature value T new.corrected , and then denormalize it to get the corrected actual temperature value.

[0130] Step 8. Save the data.

[0131] The above are only specific steps of the present invention and do not constitute any limitation to the protection scope of the present invention; any technical solutions formed by equivalent transformation or equivalent replacement shall fall within the protection scope of the present invention; some parts not elaborated in detail in the present invention belong to the known technologies of those skilled in the art.

Claims

1. An automatic line selection method for an intelligent temperature transmitter with fault self-detection, characterized in that: The automatic line selection method of the intelligent temperature transmitter with fault self-checking is based on an automatic line selection module, and the automatic line selection module includes first to sixth switching switches and a platinum resistance temperature detector; The normally closed contacts of the first to third switching switches are all connected to the A terminal, and the normally closed contacts are all connected to the B terminal. The output ends of the first to third switching switches are respectively connected to the normally closed contacts of the fourth to sixth switching switches, and the normally open contacts of the fourth to sixth switching switches are all connected to the C terminal. The A terminal, the B terminal and the C terminal are respectively connected to any one wire of a three-wire platinum thermal resistor; The output end of the fourth switch is connected to the FORCE+ terminal of the platinum resistance temperature detector, the output end of the fifth switch is connected to the RTDIN+ terminal of the platinum resistance temperature detector, and the output end of the sixth switch is connected to the RTDIN- and FORCE- terminals of the platinum resistance temperature detector; When the first to sixth switching switches are excited according to the preset excitation signal combination, a predetermined measurement value is obtained, and it is determined that the A terminal and the B terminal are connected to the equipotential line of the three-wire platinum thermal resistor, and the C terminal is connected to the common line of the three-wire platinum thermal resistor.

2. According to the automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 1, it is characterized in that: The preset excitation signal combination includes line sequence 1 and line sequence 2. In line sequence 1, the excitation signal of the first switch is low level, the excitation signal of the second switch is high level, the excitation signal of the fourth switch is low level, the excitation signal of the fifth switch is low level, and the excitation signal of the sixth switch is high level. The measurement value corresponding to line sequence 1 is +; In the line sequence 2, the excitation signal of the first switching switch is low level, the excitation signal of the third switching switch is high level, the excitation signal of the fourth switching switch is low level, the excitation signal of the fifth switching switch is high level, and the excitation signal of the sixth switching switch is low level. The measurement value corresponding to the line sequence 2 is +.

3. According to the automatic line selection method of the intelligent temperature transmitter with fault self-detection as claimed in claim 1, it is characterized in that: In the line sequence 3, the excitation signal of the first switch is high, the excitation signal of the second switch is low, the excitation signal of the fourth switch is low, the excitation signal of the fifth switch is low, and the excitation signal of the sixth switch is high. The measurement value corresponding to the line sequence 3 is -; In the line sequence 5, the excitation signal of the second switch is low level, the excitation signal of the third switch is high level, the excitation signal of the fourth switch is high level, the excitation signal of the fifth switch is low level, the excitation signal of the sixth switch is low level, and the measurement value corresponding to the line sequence 5 is 0; Terminals A and C are connected to the equipotential lines, and terminal B is connected to the common line.

4. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 3 is characterized in that: In the line sequence 4, the excitation signal of the first switch is high, the excitation signal of the third switch is high, the excitation signal of the fourth switch is low, the excitation signal of the fifth switch is high, and the excitation signal of the sixth switch is low. The measurement value corresponding to the line sequence 4 is -; In the line sequence 6, the excitation signal of the second switch is high, the excitation signal of the third switch is low, the excitation signal of the fourth switch is high, the excitation signal of the fifth switch is low, and the excitation signal of the sixth switch is low. The measurement value corresponding to the line sequence 6 is 0; The B and C terminals are connected to the equipotential lines, and the A terminal is connected to the common line.

5. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 1 is characterized in that: The automatic line selection module includes a seventh switch, the normally closed contact of the seventh switch is connected to the output end of the sixth switch, the normally open contact of the seventh switch is connected to the output end of the fifth switch through a resistor, and the output end of the seventh switch is connected to the RTDIN- and FORCE- terminals of the platinum resistance temperature detector; When the excitation signal of the seventh switch is at a high level, if the temperature value output by the platinum resistance temperature detector is 0°C, the fault is that the three-wire platinum resistance function is abnormal, otherwise it is a fault of the intelligent temperature transmitter.

6. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 5 is characterized in that: The platinum resistance temperature detector further includes an excitation resistor, which is arranged between the BIAS terminal and the REFIN- terminal of the platinum resistance temperature detector, and the BIAS terminal is connected to the REFIN+ terminal; The excitation resistor is used to calculate the resistance of the platinum resistor: R p =|4ΔV / (V REFIN+ -V REFIN- ) V REFIN+ is the voltage of REFIN+ terminal of platinum resistance temperature detector, V REFIN- is the voltage at the REFIN- terminal of the platinum resistance temperature detector, R4 is the resistance of the excitation resistor, and ΔV is the voltage difference: ΔV=IR p I provides excitation current for the platinum resistor, R p is the resistance value of platinum resistor.

7. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 6 is characterized in that: When the measured temperature T∈(0,850): R p =R0(1+aT+bT 2 ) R0 is the resistance value at 0°C, T is the measured temperature, a is the first platinum material coefficient, b is the second platinum material coefficient, usually a=3.90830×10 -3 ℃ -1 , b = -5.77500 × 10 -7 ℃ -2 .

8. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 6 is characterized in that: When the measured temperature T∈(-200,0): R p =R0(1+aT+bT 2 +c(T-100)T 3 ) Where R0 is the resistance value at 0°C, T is the measured temperature, a is the first platinum material coefficient, b is the second platinum material coefficient, and c is the third platinum material coefficient. Usually a=3.90830×10 -3 ℃ -1 , b = -5.77500 × 10 -7 ℃ -2 , c = -4.18301 × 10 -12 ℃ -1 .

9. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to any one of claims 7 or 8, characterized in that: The platinum resistance temperature correction method comprises the following steps: Step 1: Place the platinum resistor equipped with the temperature transmitter in a constant temperature box. The platinum resistor temperature detector of the temperature transmitter measures the voltage value ΔV corresponding to different temperatures. Perform multiple tests in sequence to form a ΔV-T mapping table between the ambient temperature and the voltage difference between the REFIN+ and REFIN- terminals. Step 2: Import the ΔV-T mapping table as sample data. The system selects the sample data file and reads the sample data. Suppose n sets of data are collected. Where, ΔV i is the ith measured voltage difference, T i is the i-th measured temperature value; Step 3: Normalize the sample data, set the data of T∈(0,850) as data set 1, and set the data of T∈(-200,0) in the sample data as data set 2; Step 4: Preliminarily determine whether the ambient temperature T is a positive value. If so, select 70% of the data in data set 1 as the training set and 30% of the data as the test set; if not, select 70% of the data in data set 2 as the test set and 30% of the data as the test set; Step 5: Use the training set to train the radial basis function network model and determine the network parameters; Step 6: Use the test set to evaluate the model performance, using the mean square error (MSE) as the evaluation indicator; Among them, n t is the number of samples in the test set, y i.pred is the model's predicted value for the i-th test sample, y i is the true value of the i-th test sample; Step 7: The voltage difference ΔV newly measured for the platinum resistance temperature detector new , normalize it and calculate the model output to get the corrected temperature value T new.corrected , and then denormalize it to get the corrected actual temperature value; Step 8. Save the data.

10. The automatic line selection method of the intelligent temperature transmitter with fault self-detection according to claim 9 is characterized in that: The input layer of the radial basis function network model is 1 neuron, that is, the normalized ΔV value, the output layer is 1 neuron, that is, the corrected T value, and the hidden layer is 10 neurons. The hidden layer uses the K-Means clustering algorithm to determine the center vector c of the hidden layer neurons from the voltage difference data of the training set. j ; The output of the hidden layer neurons uses a radial basis function in the form of a Gaussian function, namely: Among them, σ j is the width parameter of the function. The weights between the hidden layer and the output layer are calculated using the least squares method; The model after compensation by the temperature nonlinear correction method is: f(T)+ΔR=430ΔV / (V REFIN+ -V REFIN- ) Where f(T) is the resistance expression of the platinum resistor, and ΔR is the difference between the total resistance of RTDIN+ line and the total resistance of the common line.

Images