Algorithm for selecting residual error threshold value of levitation system of maglev train

Abstract

The invention discloses a selection algorithm for a residual threshold of a levitation system of a maglev train, and the algorithm comprises the steps: collecting and processing a normal working signal of the levitation system, and obtaining a residual value; selecting and setting an adaptive threshold based on the characteristics of the residual value; and the residual value is compared with a set threshold value, and fault information of the electromagnet of the maglev train suspension system is obtained. According to the algorithm, the dynamic self-adaptive threshold value model is constructed, and the dynamic adjustment of the threshold value is realized according to the real-time state and the operation condition of the system, so that whether a fault occurs or not is more accurately judged according to the real-time condition of the system. The Bayesian decision theory can consider the change of historical information and system states, so that the self-adaptive threshold method based on the theory can better adapt to fault detection requirements under different working conditions, and on the basis, the downtime and maintenance cost of the maglev train caused by faults can be further reduced, and the fault detection efficiency is improved. And the safety and the reliability of the maglev train are further improved.

Description

Technical Field

[0001] The invention relates to the technical field of vehicle engineering, and in particular to an algorithm for selecting a residual threshold value of a suspension system of a maglev train. Background Art

[0002] Maglev train is a means of transportation that uses the principle of electromagnetic force to suspend on the track and is driven by electromagnetic force to travel at high speed. Its advantages such as ultra-high speed, low energy consumption, environmental protection and reduction of urban congestion make it an important development direction of future transportation. The high-speed operation of maglev trains requires extremely high safety and reliability. Any minor fault may lead to serious consequences, so real-time monitoring and timely detection of faults are the key to ensure the safe operation of trains. The construction and operation costs of maglev trains are high, and any shutdown or accident caused by faults will bring huge economic losses. Therefore, the introduction of fault detection technology is also of great significance to reducing operating costs and improving economic benefits.

[0003] At present, the method of detecting sensor and actuator faults of magnetic levitation systems based on state observers has been widely used. This method generates residuals by designing the system observer, and analyzes and processes the residuals to achieve fault diagnosis. In the process of comparing the residual value and the threshold, the selection of the threshold is very critical. The selection of the threshold directly affects the accuracy and reliability of fault detection. If the threshold is set too low, it may cause false alarms, that is, the system also issues a fault alarm when there is no fault; if the threshold is set too high, it may miss alarms, that is, the system fails to issue an alarm in time when there is a fault. In the prior art, the selection of thresholds is often selected and adjusted according to the actual situation and noise level of the system, combined with engineering experience and professional knowledge. In actual engineering, the setting of static thresholds is relatively simple, and it can meet the basic fault detection requirements in some specific application scenarios and is widely used; but the suspension system is often in dynamic changes, and the noise level is often in fluctuations. At present, with the increasing requirements for detection accuracy, the static threshold can no longer meet the current fault detection needs. Summary of the invention

[0004] According to an embodiment of the present invention, in order to solve the above-mentioned deficiencies in the prior art, an algorithm for selecting a residual threshold of a maglev train suspension system is provided, comprising the following steps: Step 1: Collect and process the signal of the normal operation of the suspension system to obtain the residual value; Step 2: Based on the characteristics of the residual value, select and set the adaptive threshold; Step 3: Compare the residual value with the set threshold value to obtain the fault information of the electromagnet of the maglev train suspension system.

[0005] Preferably, in step three, when the residual value is less than a threshold value, there is no fault in the suspension system; when the residual value is greater than the threshold value, there is a fault in the suspension system.

[0006] Preferably, the selection of the adaptive threshold is based on Bayesian decision theory.

[0007] Preferably, step 2 further comprises the following steps: Based on the residual value, characteristic calculation is performed; The prior probability of the suspension system is obtained, and the decision quantity of the suspension system is established based on the Bayesian decision theory.

[0008] Preferably, the characteristic calculation includes determining a mean value and a variance of a difference between an integrated signal in a normal state and an actual signal based on the residual value.

[0009] Preferably, the mean and variance are: ; in, is the mean value of the difference between the integral signal and the actual signal of the electromagnet in the normal state, The square of is the variance of the difference between the integral signal and the actual signal of the electromagnet in the normal state, i is the electromagnet current, It represents the residual sequence when the system is normal, and N is the data length.

[0010] Preferably, the decision quantity is: ; in, is the decision amount, The square of is the variance of the intercepted residual of the signal to be diagnosed and the normal system, and k is the number.

[0011] Preferably, the specific method of comparison in step 3 is that the comparison between the decision amount and the adaptive threshold is: ; in, is the decision amount, is the data length in the running state, is the prior probability that the system is fault-free.

[0012] According to the selection algorithm of the residual threshold of the maglev train suspension system of the embodiment of the present invention, by constructing a dynamic adaptive threshold model, the threshold is dynamically adjusted according to the real-time state and operating conditions of the system, so as to more accurately judge whether a fault occurs according to the real-time situation of the system. This helps to improve the accuracy and reliability of fault detection and reduce the occurrence of false alarms and missed alarms. In practical applications, the noise level of the system may change due to the influence of various factors. The dynamic adaptive threshold can improve the robustness and adaptability of the system to better adapt to the changes in the noise level, as well as other dynamic operating conditions and fault conditions. After the adaptive real-time adjustment, the fault detection level will be more reliable, further improving the safety performance of the maglev train. Further, the Bayesian decision theory is used to select the adaptive threshold, which can take into account the changes in historical information and system status. Therefore, the adaptive threshold method based on this theory can better adapt to the fault detection requirements under different working conditions. Based on this, it helps to further reduce the downtime and maintenance costs caused by faults of the maglev train, and further improve the safety and reliability of the maglev train.

[0013] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flow chart of an algorithm for selecting a residual threshold value of a suspension system of a maglev train according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings to further illustrate the present invention.

[0016] In order to ensure the high stability and safety of the maglev train during operation. The first embodiment of the present application provides an algorithm for selecting the residual threshold of the maglev train suspension system, such as Figure 1 As shown, it includes the following steps: Step 1: Collect and process the signal of the normal operation of the suspension system to obtain the residual value; Step 2: Based on the characteristics of the residual value, select and set the adaptive threshold; Step 3: Compare the residual value with the set threshold value to obtain the fault information of the electromagnet of the maglev train suspension system.

[0017] The algorithm designs an adaptive mechanism to enable the threshold selection to meet the fault detection requirements under different working conditions, thereby improving the flexibility and adaptability of the algorithm. This method can more effectively distinguish between normal and fault states, reducing the possibility of false positives and false negatives. In reality, this method does not require additional hardware and is easy to implement. It also reduces the need for manual intervention and parameter adjustment, and reduces the difficulty and cost of system maintenance.

[0018] Preferably, in step 3, under normal operating conditions, due to the high precision of the observer and the stability of the system, the difference (residual value) between the actual state of the system and the estimated state of the observer will remain within a preset small range, which is defined by a dynamically set threshold. If the residual value is lower than the threshold, it indicates that the electromagnet is in normal working condition and no fault has occurred.

[0019] Preferably, the selection of the adaptive threshold is based on Bayesian decision theory. Some traditional adaptive threshold methods adjust the threshold only based on current or recent data, ignoring the cumulative effect of historical information. Bayesian decision theory takes into account previous parameter information and system status when updating the threshold, which makes the threshold adjustment more accurate and robust. The Bayesian optimization method can find a better threshold in a smaller number of iterations by continuously updating prior knowledge. This means that in practical applications, the adaptive threshold method based on Bayesian decision theory can adapt to changes in system status more quickly and improve the real-time performance of fault detection. By accurately adjusting the threshold, the adaptive threshold method based on Bayesian decision theory can more accurately identify the fault state of the electromagnet. It helps to reduce false positives and false negatives and improve the reliability of fault detection.

[0020] Furthermore, step three includes data processing for fault detection of the electromagnet of the maglev train suspension system, which specifically includes the following sub-steps: Based on the residual value, characteristic calculation is performed; The prior probability of the suspension system is obtained, and the decision quantity of the suspension system is established based on the Bayesian decision theory.

[0021] Preferably, the characteristic calculation includes determining the mean and variance of the difference between the integral signal and the actual signal under normal conditions based on the residual value. Bayesian decision theory can take into account changes in historical information and system states, so the adaptive threshold method based on this theory can better adapt to the fault detection needs under different working conditions, and based on this, it helps to further reduce the downtime and maintenance costs caused by faults in maglev trains. Determining the mean and variance of the difference between the integral signal and the actual signal under normal conditions based on the residual value as a characteristic can reflect the stability of the system, be sensitive to fault changes, be easy to implement and calculate, and provide a strong decision-making basis for fault detection. This characteristic calculation method has broad application prospects in the fault detection of electromagnets in the suspension system of maglev trains.

[0022] Preferably, the mean and variance are: ; in, is the mean value of the difference between the integral signal and the actual signal of the electromagnet in the normal state, The square of is the variance of the difference between the integral signal and the actual signal of the electromagnet in the normal state, i is the electromagnet current, It represents the residual sequence when the system is normal, and N is the data length.

[0023] Preferably, the decision quantity is: ; in, is the decision amount, The square of is the variance of the intercepted residual of the signal to be diagnosed and the normal system, and k is the number.

[0024] Preferably, the specific method of comparison in step 3 is that the comparison between the decision amount and the adaptive threshold is: ; in, is the decision amount, is the data length in the running state, is the prior probability that the system is fault-free.

[0025] Above, refer to Figure 1 The invention describes an algorithm for selecting the residual threshold of the suspension system of a maglev train according to an embodiment of the present invention. By constructing a dynamic adaptive threshold model, the threshold is dynamically adjusted according to the real-time state and operating conditions of the system, so as to more accurately determine whether a fault occurs according to the real-time situation of the system. This helps to improve the accuracy and reliability of fault detection and reduce the occurrence of false alarms and missed alarms. In practical applications, the noise level of the system may change due to a variety of factors. The dynamic adaptive threshold can improve the robustness and adaptability of the system to better adapt to changes in noise levels, as well as other dynamic operating conditions and fault conditions. Further, the threshold of fault diagnosis is selected by adopting Bayesian decision theory, which can take into account changes in historical information and system states. Therefore, the adaptive threshold method based on the theory can better adapt to the fault detection requirements under different working conditions. Based on this, it helps to further reduce the downtime and maintenance costs caused by faults of maglev trains, and further improve the safety and reliability of maglev trains. After adaptive real-time adjustment, the fault detection level will be more reliable, further improving the safety performance of maglev trains.

[0026] In the description of the present invention, it should be noted that, unless otherwise specified, “plurality” means two or more than two; the orientations or positional relationships indicated by terms such as “upper”, “lower”, “left”, “right”, “inner”, “outer”, “top”, and “bottom” are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0027] It should be noted that, in this specification, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprises..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0028] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. An algorithm for selecting residual threshold of maglev train suspension system, characterized in that: Include Step 1: Collect and process the signal when the suspension system works normally to obtain a residual value; Step 2: Based on the characteristics of the residual value, select and set an adaptive threshold; Step 3: Compare the residual value with the set threshold value to obtain fault information of the electromagnet of the maglev train suspension system.

2. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 1 is characterized in that: In step three, when the residual value is smaller than the threshold value, the suspension system has no fault; when the residual value is larger than the threshold value, the suspension system has a fault.

3. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 1 is characterized in that: The selection of the adaptive threshold is based on Bayesian decision theory.

4. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 3 is characterized in that: The step 2 further comprises the following steps: Based on the residual value, performing characteristic calculation; The prior probability of the suspension system is obtained, and a suspension system decision quantity is established based on Bayesian decision theory.

5. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 4 is characterized in that: The characteristic calculation includes determining a mean value and a variance of a difference between an integrated signal and an actual signal in a normal state based on the residual value.

6. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 5 is characterized in that: The mean and the variance are: ； in, is the mean value of the difference between the integral signal and the actual signal of the electromagnet in the normal state, The square of is the variance of the difference between the integral signal and the actual signal of the electromagnet in the normal state, i is the electromagnet current, represents the residual sequence when the system is normal, is the data length.

7. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 6 is characterized in that: The decision quantity is: ； in, is the decision amount, The square of is the variance of the intercepted residual of the signal to be diagnosed and the normal system, and k is the number.

8. The algorithm for selecting the residual threshold of the maglev train suspension system according to claim 7 is characterized in that: The specific method of comparison in step 3 is the comparison between the decision amount and the adaptive threshold: ； in, is the decision amount, is the data length in the running state, is the prior probability that the system is fault-free.

Images