Method and system for monitoring heat dissipation status of rail transit train converter

Abstract

The present invention discloses a method and system for monitoring the heat dissipation state of a rail transit train converter, comprising: taking the traction power and ambient temperature of the train under several operating conditions as sample data, and inputting the sample data into a neural network model for network training to obtain a trained network model; inputting the traction power and ambient temperature of the train under normal operating conditions into the trained network model, and outputting a dynamic temperature curve l of the converter under normal heat dissipation state. h ; Measure the dynamic temperature curve l of the converter under actual heat dissipation state; Calculate the dynamic temperature curve l h The difference D1 between the dynamic temperature curve l and the dynamic temperature curve l is calculated h Dynamic temperature curve f The difference D2 between them is obtained; the difference D1 is compared with the difference D2 to obtain a comparison result, and the heat dissipation state of the converter is identified through the comparison result. The present invention realizes non-intrusive online monitoring and ensures the long-term reliable operation of the heat dissipation system.

Description

Technical Field

[0001] The present invention relates to the field of converter heat dissipation monitoring, and in particular to a method and system for monitoring the heat dissipation status of a rail transit train converter. Background Art

[0002] At present, the heat dissipation status monitoring of the converter generally adopts regular manual purging and cleaning of the cooling system. However, this method requires a lot of manpower and material resources, and is extremely inefficient. In addition, due to the blindness of regular maintenance and the inaccuracy of manual operation, it is easy to cause "over-maintenance" or "under-maintenance", and it still cannot effectively avoid the occurrence of over-temperature failures. Other related heat dissipation status monitoring methods can be divided into direct monitoring and indirect monitoring.

[0003] In terms of direct monitoring, optical and physical contact methods can directly monitor the temperature of the system, such as infrared, thermal imagers, thermistors, etc. The monitoring results are intuitive and clear. However, directly obtaining internal temperature information requires changing the original structure of the system, which is invasive and may affect the operational reliability of the system.

[0004] In terms of indirect monitoring, based on the zero-input response curve of the system, the heat network parameters and natural response frequency are used as characteristic quantities to monitor the performance of the cooling system. However, during the braking of the train, the diode inside the power module will continue to work and cannot provide a heat source-free cooling curve. In addition, there are studies that propose to monitor parameters such as the degradation trend of ball bearings, the dynamic characteristics of fan speed, and the top oil temperature, and indirectly evaluate the health status of the cooling system through numerical analysis or intelligent algorithms. This type of method is mainly aimed at systems with controllable heat dissipation conditions and is applied to thermal dynamic equilibrium states. However, due to the design of running air cooling, the heat dissipation of urban rail trains itself changes with the speed of the train, which will affect the accuracy of the judgment results. At the same time, since the distance between urban transportation stations is usually close, trains are frequently started and stopped, and it is difficult for the converter temperature to reach a thermal equilibrium state. Therefore, applying these methods or ideas directly to rail trains ignores the dynamic response process and may result in misjudgment. Summary of the invention

[0005] In view of this, the purpose of the present invention is to overcome the defects in the prior art and provide a method and system for monitoring the heat dissipation status of a rail transit train inverter, which solves the problem that the prior art is not suitable for the operating conditions and heat dissipation design of rail trains, and realizes non-intrusive online monitoring, thereby ensuring the long-term reliable operation of the heat dissipation system.

[0006] The method for monitoring the heat dissipation status of a rail transit train converter of the present invention comprises the following steps:

[0007] S1. The traction power and ambient temperature of the train under several operating conditions are used as sample data, and the sample data is input into the neural network model for network training to obtain a trained network model;

[0008] S2. Input the traction power and ambient temperature of the train under normal working conditions into the trained network model, and output the dynamic temperature curve of the converter under normal heat dissipation state. h ;

[0009] S3. Measure the dynamic temperature curve l of the converter under the actual heat dissipation state;

[0010] S4. Calculate the dynamic temperature curve l h The difference D1 between the dynamic temperature curve l and the dynamic temperature curve l is calculated h Dynamic temperature curve f The difference between D2; where l f It is the dynamic temperature curve of the converter under the state of heat dissipation aging failure;

[0011] S5. Compare the difference D1 with the difference D2 to obtain a comparison result, and identify the heat dissipation state of the converter through the comparison result.

[0012] Furthermore, the dynamic temperature curve l h Including heat sink temperature curve h-s ;

[0013] The dynamic temperature curve 1 includes a heat sink temperature curve 1 s ;

[0014] The dynamic temperature curve l f Including heat sink temperature curve f-s .

[0015] Furthermore, the dynamic temperature curve l h Including the temperature difference curve of the radiator inlet and outlet h-w ;

[0016] The dynamic temperature curve l includes the radiator inlet and outlet temperature difference curve l w ;

[0017] The dynamic temperature curve l f Including the temperature difference curve of the radiator inlet and outlet f-w .

[0018] Further, the dynamic temperature curve l is determined according to the following formula h The difference between the dynamic temperature curve l and D1:

[0019]

[0020] Among them, the D point (p,l)=min q∈l (D Euclid (p,q)); D Euclid (p,q) is the Euclidean distance between point p and point q.

[0021] Further, the dynamic temperature curve l is determined according to the following formula h Dynamic temperature curve f The difference between D2:

[0022]

[0023] Among them, the D point (r,l f )=min s∈lf (D Euclid (r,s));D Euclid (r,s) is the Euclidean distance between point r and point s.

[0024] Further, the step S5 specifically includes:

[0025] Calculate the ratio of difference D1 to difference D2

[0026] Determine the ratio Whether it approaches 1, if so, the heat dissipation state of the converter is poor; if not, the heat dissipation state of the converter is normal.

[0027] A rail transit train converter heat dissipation status monitoring system includes a collection unit and a data analysis unit connected to the collection unit in communication;

[0028] The collection unit collects the traction power and ambient temperature of the train under several operating conditions, the radiator temperature of the train during operation, and the temperature of the radiator inlet and outlet;

[0029] The data analysis unit: uses the traction power and ambient temperature of the train under several operating conditions as training sample data, performs network training on the neural network model, and obtains a trained network model; and outputs a dynamic temperature curve l of the converter under normal heat dissipation state. h ; By comparing the difference D1 and the difference D2, the heat dissipation state of the converter is identified;

[0030] Wherein, the difference D1 is the dynamic temperature curve l h The difference D2 is the difference between the dynamic temperature curve l of the converter under the actual heat dissipation state, and the dynamic temperature curve l h Dynamic temperature curve of the converter under heat dissipation aging failure state f The difference between.

[0031] Furthermore, the acquisition unit includes a temperature sensor and / or a transmitter.

[0032] Furthermore, the data analysis unit includes a data acquisition card and a data analysis platform; the input end of the data acquisition card is connected to the output end of the acquisition unit, and the output end of the data acquisition card is connected to the input end of the data analysis platform.

[0033] The beneficial effects of the present invention are as follows: the present invention discloses a method and system for monitoring the heat dissipation status of a rail transit train inverter, which uses a dynamic temperature curve as a characteristic quantity, and evaluates the health status of the heat dissipation system by comparing the difference between the real-time monitoring curve and the reference curve. In order to solve the problem of diversity of operating conditions during train operation, an artificial neural network is used to match reference values ​​at different measurement times; in order to solve the problem that existing methods are difficult to apply to non-thermal equilibrium states, a one-way distance algorithm is used to take into account the dynamic temperature response process. The present invention fully considers the characteristics of train operating conditions and the running air-cooled heat dissipation design, and solves the problem of dust coverage monitoring of the heat dissipation system under the conditions of diversity of operating conditions and non-thermal equilibrium of system temperature. At the same time, the information required by the present invention is mostly existing information of the existing train control system, with simple software and hardware, simple operation, and non-invasive online monitoring, which greatly reduces the operation and maintenance costs and ensures the long-term reliable operation of the heat dissipation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0035] Figure 1 It is a schematic diagram of the monitoring method flow of the present invention;

[0036] Figure 2 This is a block diagram of the structure principle of the monitoring system of the present invention. DETAILED DESCRIPTION

[0037] The present invention is further described below in conjunction with the accompanying drawings, as shown in the drawings:

[0038] The method for monitoring the heat dissipation status of a rail transit train converter of the present invention comprises the following steps:

[0039] S1. The traction power and ambient temperature of the train under several operating conditions are taken as sample data, and the sample data are input into the neural network model for network training to obtain a trained network model; wherein, in the training process of the neural network model, the configuration parameters of the neural network are continuously adjusted so that the output result of the neural network model meets the predetermined requirements, thereby obtaining a trained network model; specifically, a large amount of sample data is input to train the neural network model so that the sample data obtains dynamic temperature data through mapping, thereby forming a dynamic temperature curve, and finally the neural network model outputs a dynamic temperature curve; the neural network model adopts an existing neural network structure, which will not be repeated here.

[0040] S2. Input the traction power and ambient temperature of the train under normal working conditions into the trained network model, and output the dynamic temperature curve of the converter under normal heat dissipation state. h By inputting the working parameters under normal conditions into the trained network model, the dynamic temperature curve of the converter cooling system under healthy conditions can be obtained. h , prepare data for subsequent analysis and comparison.

[0041] S3. Measuring the dynamic temperature curve l of the converter under the actual heat dissipation state; Under the actual working condition, the heat sink of the converter to be measured is measured to obtain the dynamic temperature curve l of the converter under the actual heat dissipation state, and the temperature curve l is the monitored temperature curve.

[0042] S4. Calculate the dynamic temperature curve l h The difference D1 between the dynamic temperature curve l and the dynamic temperature curve l is calculated h Dynamic temperature curve f The difference between D2; where l f is a dynamic temperature curve of the converter under the heat dissipation aging failure state; wherein, the heat dissipation aging failure state is set by the user according to the needs, including but not limited to the state where the junction temperature of the power module rises by 20%, the state where the total thermal resistance rises by 20%, and the state where the heat sink thermal resistance rises by 5%, etc.; these abnormal states are used as the threshold of heat dissipation aging failure, and when the threshold is met, it is considered that the heat dissipation aging failure; the junction temperature measurement, total thermal resistance measurement and heat sink thermal resistance measurement are all existing technologies and will not be repeated here; under the heat dissipation aging failure state, the measured dynamic temperature curve l f That is, the temperature curve when the cooling system is unhealthy or the cooling is abnormal;

[0043] S5. Compare the difference D1 with the difference D2 to obtain a comparison result, and identify the heat dissipation state of the converter through the comparison result.

[0044] In this embodiment, the dynamic temperature curve lh Including heat sink temperature curve h-s ;

[0045] The dynamic temperature curve 1 includes a heat sink temperature curve 1 s ;

[0046] The dynamic temperature curve l f Including heat sink temperature curve f-s .

[0047] The radiator temperature curve is a curve whose horizontal axis is time and whose vertical axis is radiator temperature.

[0048] In this embodiment, in addition to considering the radiator temperature as the dynamic reference temperature, the temperature difference between the radiator inlet and outlet can also be used as the dynamic reference temperature, that is:

[0049] The dynamic temperature curve l h Including the temperature difference curve of the radiator inlet and outlet h-w ;

[0050] The dynamic temperature curve l includes the radiator inlet and outlet temperature difference curve l w ;

[0051] The dynamic temperature curve l f Including the temperature difference curve of the radiator inlet and outlet f-w ; Wherein, the radiator inlet and outlet temperature difference is the difference between the radiator outlet temperature and the radiator inlet temperature. The radiator inlet and outlet temperature difference curve is a curve with the horizontal axis being time and the vertical axis being the radiator inlet and outlet temperature difference.

[0052] Through the above settings, according to the monitoring method of the present invention, a difference comparison result A with the radiator temperature as the dynamic temperature can be obtained, and a difference comparison result B with the temperature difference between the radiator inlet and outlet as the dynamic temperature can be obtained. In step S5, the heat dissipation state of the inverter is identified as a through the difference comparison result A, and the heat dissipation state of the inverter is identified as b through the difference comparison result B; the heat dissipation state a and the heat dissipation state b can be used as references to each other, thereby improving the accuracy and reliability of heat dissipation state monitoring.

[0053] In this embodiment, the dynamic temperature curve l is determined according to the following formula: h The difference between the dynamic temperature curve l and D1:

[0054]

[0055] Among them, the D point (p,l)=min q∈l (D Euclid (p,q)); D Euclid(p,q) is the Euclidean distance between point p and point q.

[0056] Determine the dynamic temperature curve l according to the following formula h Dynamic temperature curve f The difference between D2:

[0057]

[0058] Among them, the D point (r,l f )=min s∈lf (D Euclid (r,s));D Euclid (r,s) is the Euclidean distance between point r and point s.

[0059] In this embodiment, the step S5 specifically includes:

[0060] Calculate the ratio of difference D1 to difference D2

[0061] Determine the ratio Is it close to 1? If so, it means that the monitored temperature curve is closer to the dynamic temperature curve under the state of heat dissipation aging failure, that is, the heat dissipation system of the converter is unhealthy and the heat dissipation state of the converter is poor; if not, the heat dissipation system of the converter is healthy and the heat dissipation state of the converter is normal.

[0062] The present invention also relates to a heat dissipation state monitoring system for a rail transit train converter. The state monitoring system corresponds to the above-mentioned heat dissipation state monitoring method for a rail transit train converter, and can be understood as a system for implementing the above-mentioned state monitoring method. The state monitoring system includes a collection unit and a data analysis unit in communication with the collection unit;

[0063] The collection unit collects the traction power and ambient temperature of the train under several operating conditions, the radiator temperature of the train during operation, and the temperature of the radiator inlet and outlet;

[0064] The data analysis unit: uses the traction power and ambient temperature of the train under several operating conditions as training sample data, performs network training on the neural network model, and obtains a trained network model; and outputs a dynamic temperature curve l of the converter under normal heat dissipation state. h ; By comparing the difference D1 and the difference D2, the heat dissipation state of the converter is identified;

[0065] Wherein, the difference D1 is the dynamic temperature curve l h The difference D2 is the difference between the dynamic temperature curve l of the converter under the actual heat dissipation state, and the dynamic temperature curve l hDynamic temperature curve of the converter under heat dissipation aging failure state f The difference between.

[0066] In this embodiment, the acquisition unit includes a temperature sensor and / or a transmitter. The temperature sensor may be an optical fiber, a thermocouple, a PT100, etc., and according to actual working conditions and actual needs, a corresponding type of temperature sensor and / or transmitter is selected to detect the temperature of the measured object; during the data transmission process, the data type or format of the collected data needs to be converted so that the collected data can be applied to other devices, and the acquisition unit also includes an A / D converter.

[0067] In this embodiment, the data analysis unit includes a data acquisition card and a data analysis platform; the input end of the data acquisition card is connected to the output end of the acquisition unit, and the output end of the data acquisition card is connected to the input end of the data analysis platform.

[0068] The data acquisition card is used for receiving and transmitting data; the data analysis platform includes a host computer and a visualization platform that communicates with the host computer; of course, when performing data analysis and processing, the host computer is configured with data analysis and processing software, such as LabVIEW, MATLAB, etc.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for monitoring heat dissipation status of a rail transit train converter, characterized in that: The steps include: S1. The traction power and ambient temperature of the train under several operating conditions are used as sample data, and the sample data is input into the neural network model for network training to obtain a trained network model; S2. Input the traction power and ambient temperature of the train under normal working conditions into the trained network model, and output the dynamic temperature curve of the converter under normal heat dissipation state. h ; S3. Measure the dynamic temperature curve l of the converter under the actual heat dissipation state; S4. Calculate the dynamic temperature curve l h The difference D1 between the dynamic temperature curve l and the dynamic temperature curve l is calculated h Dynamic temperature curve f The difference between D2; where l f It is the dynamic temperature curve of the converter under the state of heat dissipation aging failure; S5. Compare the difference D1 with the difference D2 to obtain a comparison result, and identify the heat dissipation state of the converter through the comparison result, specifically including: Calculate the ratio of difference D1 to difference D2 Determine the ratio Whether it approaches 1, if so, the heat dissipation state of the converter is poor; if not, the heat dissipation state of the converter is normal.

2. The method for monitoring heat dissipation status of a rail transit train converter according to claim 1, characterized in that: The dynamic temperature curve l h Including heat sink temperature curve h-s ; The dynamic temperature curve 1 includes a heat sink temperature curve 1 s ; The dynamic temperature curve l f Including heat sink temperature curve f-s .

3. The method for monitoring heat dissipation status of a rail transit train converter according to claim 1, characterized in that: The dynamic temperature curve l h Including the temperature difference curve of the radiator inlet and outlet h-w ; The dynamic temperature curve l includes the radiator inlet and outlet temperature difference curve l w ; The dynamic temperature curve l f Including the temperature difference curve of the radiator inlet and outlet f-w .

4. The method for monitoring heat dissipation status of a rail transit train converter according to claim 1, characterized in that: Determine the dynamic temperature curve l according to the following formula h The difference between the dynamic temperature curve l and D1: Among them, the D point (p, l) = min q∈l (D Euclid (p, q)); D Euclid (p, q) is the Euclidean distance between point p and point q.

5. The method for monitoring heat dissipation status of a rail transit train converter according to claim 1, characterized in that: Determine the dynamic temperature curve l according to the following formula h Dynamic temperature curve f The difference between D2: Among them, the D Euclid (r, s) is the Euclidean distance between point r and point s.

6. A rail transit train converter heat dissipation status monitoring system, characterized in that: It includes a data collection unit and a data analysis unit that is in communication with the data collection unit; The collection unit collects the traction power and ambient temperature of the train under several operating conditions, the radiator temperature of the train during operation, and the temperature of the radiator inlet and outlet; The data analysis unit: uses the traction power and ambient temperature of the train under several operating conditions as training sample data, performs network training on the neural network model to obtain a trained network model; and outputs a dynamic temperature curve l of the converter under normal heat dissipation state. h ; By comparing the difference D1 and the difference D2, the heat dissipation status of the converter is identified, including: Calculate the ratio of difference D1 to difference D2 Determine the ratio Whether it approaches 1, if so, the heat dissipation state of the converter is poor; if not, the heat dissipation state of the converter is normal; Wherein, the difference D1 is the dynamic temperature curve l h The difference D2 is the difference between the dynamic temperature curve l of the converter under the actual heat dissipation state, and the dynamic temperature curve l h Dynamic temperature curve of the converter under heat dissipation aging failure state f The difference between.

7. The rail transit train converter heat dissipation status monitoring system according to claim 6, characterized in that: The acquisition unit includes a temperature sensor and / or a transmitter.

8. The rail transit train converter heat dissipation status monitoring system according to claim 6, characterized in that: The data analysis unit comprises a data acquisition card and a data analysis platform; the input end of the data acquisition card is connected to the output end of the acquisition unit, and the output end of the data acquisition card is connected to the input end of the data analysis platform.

Images