A high-voltage interlocking system and device applied to a rail vehicle

Abstract

The present application relates to the technical field of track high-voltage interlocking control, in particular to a kind of applied to track vehicle high-voltage interlocking system and device.By monitoring the operating state and air path state of air path interlocking box in track train system in advance, and calculating operation data, time series model is established according to operation data, to predict the operation data of air path interlocking box under the time needing continuous unlocking in future, judge whether high-voltage interlocking system can safely respond control, improve the safety control precision of high-voltage interlocking control system;And when predicting whether air path interlocking box can be continuously unlocked in future, the running state of high-voltage related equipment and the severity of fault are integrated, so as to accurately evaluate the repair disposal time of high-voltage related equipment, by combining repair disposal time with time series model, accurately judge whether air path interlocking box can be safely and accurately unlocked and repair high-voltage related equipment, guarantee the personal safety of repair personnel.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of track high-voltage interlocking control, and particularly relates to a track vehicle high-voltage interlocking system and device. BACKGROUND

[0002] The key interlocking mechanism is a physical safety measure in the track high-voltage interlocking system, which works together with the circuit interlocking (such as the breaker interlocking) to ensure the safety and standardization of high-voltage equipment operation. Through the transmission and interlocking of physical keys, the operation process of high-voltage equipment is strictly controlled to prevent misoperation and dangerous conditions.

[0003] At present, in the track high-voltage interlocking system, the air path interlocking box is unlocked by a key, and the subsequent high-voltage related equipment is unlocked again based on the unlocked air path interlocking box. By unlocking layer by layer, the high-voltage equipment components in the track train system can be maintained and disposed. Therefore, when unlocking the air path interlocking box and the subsequent high-voltage related equipment, there is a lack of state monitoring of the air path interlocking box and the subsequent high-voltage related equipment. When the running state cannot be learned, if the air path interlocking box is abnormal and direct unlocking is performed, the abnormality of the air path interlocking box will be aggravated. Under this condition, during the maintenance process of the subsequent high-voltage related equipment after unlocking, there is inevitably a certain safety hazard. Therefore, how to realize the state monitoring of the air path interlocking box and the subsequent high-voltage related equipment, and build a relatively safe track high-voltage interlocking system, which is beneficial to the maintenance of high-voltage related equipment and guarantees the personal safety of maintenance personnel, is a problem to be solved. SUMMARY

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a track vehicle high-voltage interlocking system and device, which can effectively solve the problem that the state of the air path interlocking box and the high-voltage related equipment is not monitored in real time in the prior art, and the safety of the high-voltage interlocking control system cannot be improved.

[0005] To achieve the above purpose, the present application is realized by the following technical scheme:

[0006] The present application provides a track vehicle high-voltage interlocking system, which at least comprises:

[0007] A first interlocking monitoring module is used for monitoring the running state of the air path interlocking box in the track high-voltage interlocking system, wherein:

[0008] The operation state and the air path state of the air path interlocking box are monitored to determine the running state, and the running data is determined based on the running state;

[0009] The operation data of the air path interlocking box at different time points is obtained, time sequence data is constructed, a time sequence model is established based on the time sequence data, the time sequence model is combined with the time value of unlocking the air path interlocking box at a future time, future operation data of the air path interlocking box is determined, and it is judged whether the first level key is allowed to unlock the air path interlocking box;

[0010] The second interlocking monitoring module is used for determining the expected travel time according to the second level key and the high-voltage related equipment.

[0011] The time-frequency diagram is generated by obtaining the time sequence data of the high-voltage related equipment, the local time-frequency features are extracted to form a high-dimensional feature vector, the high-dimensional feature vector is input into the SVM classifier to determine the fault severity of the high-voltage related equipment, the corresponding repair disposal time is judged, the time value is constructed based on the repair disposal time and the expected travel time, and it is judged whether the air path interlocking box is unlocked.

[0012] Further, the operation data of the air path interlocking box is determined as follows:

[0013] The operation state of the air path interlocking box is monitored,

[0014]

[0015] Q FS represents the operation state data of the air path interlocking box, M run represents the current operation torque of the three-way valve, M sat represents the initial operation torque of the three-way valve, G cut represents the current mechanical gap of the three-way valve, G sat represents the initial mechanical gap of the three-way valve, F cut represents the current operation torque of the lock core, F sat represents the initial operation torque of the lock core, R cut represents the current rotation resistance of the lock core, R sat represents the initial rotation resistance of the lock core, GF represents the vibration time of the three-way valve, FR represents the vibration time of the lock core, and α1, α2, α3, α4, α5 and α6 are corresponding weight coefficients.

[0016] The air path state of the air path interlocking box is monitored,

[0017]

[0018] P FS represents the air path state data of the air path interlocking box, M eas represents the current air path pressure, M sat represents the initial air path pressure, F eas represents the current air path gas flow, and F satQp represents the initial gas path gas flow, MF represents a calculation term, β1 and β2 represent corresponding weight coefficients;

[0019] Gas path state data Q of the gas path interlocking box FS and operation state data P FS to obtain running data QP of the gas path interlocking box at the current time point through weighted summation t .

[0020] Further, the method for determining future running data of the gas path interlocking box is:

[0021] Obtain running data QP of the gas path interlocking box at multiple time points t to construct time series data;

[0022] Establish a time series model, that is,

[0023] (1-α1B-α2B 2 -···-α p B p )(1-B) d QP t =μ+(1+θ1B+θ2B 2 +···+θ q B q )εt

[0024] Wherein, QP t represents time series data, B represents a lag operator, α1, α2,..., α p represent coefficients of the autoregressive part respectively, θ1, θ2,..., θ q represent coefficients of the moving average part respectively, p represents the order of the autoregressive part, q represents the order of the moving average part, d represents the difference order, μ represents a constant term, and ε t represents an error term;

[0025] Predict running data QP of the gas path interlocking box at a future time t+h , that is,

[0026]

[0027] Wherein, α i represents a coefficient of the autoregressive part, θ j represents a coefficient of the moving average part, ε t+h-j represents an error term at the past j time steps, QP t+h-i represents running data at time t+h-i, and h represents the step length of prediction.

[0028] Further, the determination method of the predicted travel time is:

[0029] Determine the coordinates of the second level key, denoted as second coordinates, determine the coordinates of the high-voltage related equipment, denoted as high-voltage equipment coordinates, determine the distance based on the second coordinates and the high-voltage equipment coordinates, and calculate the expected travel time SS t .

[0030] Further, the expression of the time series data of the high-voltage related equipment generating a time-frequency diagram is:

[0031] Perform short-time Fourier transform on the time series data to generate a time-frequency diagram S(f,τ):

[0032]

[0033] Where ω(t-τ) represents a short-time window function, τ represents the center position of the short-time window function, represents the position of the time axis, f represents the frequency variable, e -j2πft represents a complex exponential basis function, and t represents a time variable.

[0034] Further, the method for forming a high-dimensional feature vector is:

[0035] Input the time-frequency diagram S(f,τ), and extract the local time-frequency features, so that

[0036]

[0037] Where represents the output of the lth layer and the kth convolution kernel, W l,k,K，m,n represents the weight matrix of the lth layer and the kth convolution kernel under the size K, S m,n represents the pixel value of the m,nth position in the input feature map, b k represents the bias of the kth convolution kernel in the convolution layer, and σ represents an activation function.

[0038] Reduce the feature dimension through the pooling layer, so that

[0039]

[0040] Where P i,j is the output result after the pooling operation, (m,n)∈window represents the position within the pooling window, window represents the window defining the size of the pooling region, F l,k,m,n represents the value of the kth output feature map in the convolution layer l at position m,n.

[0041] Output a high-dimensional feature vector through a fully connected layer, so that

[0042] h=σ(W f ·X+b f )

[0043] wherein h represents the high-dimensional feature vector extracted by the CNN, W f represents the weight matrix of the fully connected layer, X represents the feature input of the previous layer, b f represents the bias of the fully connected layer.

[0044] Further, the method for determining the fault severity of the high-voltage related equipment by the SVM classifier is:

[0045] Define the SVM objective function as

[0046]

[0047] wherein ω represents the weight vector of the classifier, b represents the bias, ξ i is the slack variable, C represents the penalty coefficient, and N represents the total number of samples in the sample data set.

[0048] Define the SVM constraint condition, and the form of the constraint condition is as follows,

[0049] y i (ω·h i +b)≥1-ξ i ,ξ i ≥0

[0050] wherein y i represents the true class label of the i-th sample, and y i is set as the class label representing the fault severity, h i represents the high-dimensional feature vector of the i-th sample, 1-ξ i represents that the right side of the constraint condition is related to the slack variable, and ξ i≥0 represents the non-negativity of the slack variable.

[0051] Further, the method for determining the time value is:

[0052] Determine the high-voltage related equipment controlled by the air circuit interlocking box in the current rail train system;

[0053] Determine the repair disposal time P GS corresponding to the fault severity of the controlled high-voltage related equipment.

[0054] Combine the repair disposal time P GS with the expected route time SS t to construct a time value U j .

[0055] The rail vehicle high-voltage interlocking device comprises a memory and a processor, the memory stores a computer program, and the processor implements the system according to any one of the above when executing the computer program.

[0056] A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the system of any one of the preceding claims.

[0057] Compared with the known prior art, the technical solution provided by the application has the following beneficial effects:

[0058] By monitoring the operation state and the air circuit state of the air circuit interlocking box in the rail train system in advance, calculating the operation data, establishing a time series model according to the operation data, predicting the operation data of the air circuit interlocking box at a time when the air circuit interlocking box needs to be continuously unlocked in the future, and judging whether the high-voltage interlocking system can safely respond to control, the safety control precision of the high-voltage interlocking control system is improved.

[0059] When predicting whether the air circuit interlocking box can be continuously unlocked in the future, the operation state and the fault severity of the high-voltage related equipment are integrated, so as to accurately evaluate the repair disposal time of the high-voltage related equipment, accurately judge whether the air circuit interlocking box can be safely and correctly unlocked and the high-voltage related equipment can be repaired by combining the repair disposal time with the time series model, and the personal safety of the repair personnel is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0061] Figure 1 The overall module block diagram of the present application. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0063] The current rail high-voltage interlocking system, for example:

[0064] Step one: driver's room operation and use of first level key (A key, blue)

[0065] Operation content: Turn the electric ground key switch of the driver's room fault switch panel to the locked position, and remove the first level key (A key)

[0066] The circuit interlock of the driver's cab ensures that the trolley pole raising and main breaker closing operations cannot be performed at this time, and the high-voltage power supply is cut off;

[0067] After the grounding switch is turned to the grounding position, the high-voltage system of the rail vehicle is in a grounded state, ensuring safety.

[0068] The removal of the first-level key indicates that the operation enters the physical key interlock process. Only after the grounding switch is locked can the first-level key be obtained to proceed to the subsequent steps.

[0069] Step 2: Unlock the first-level key to the gas path interlock box and cut off the gas path.

[0070] Operation content: Insert the first-level key into the trolley pole raising gas path interlock box, unlock the three-way valve handle, cut off the trolley pole raising gas path, lock the three-way valve handle, and take out the second-level key (B key, yellow).

[0071] High-voltage interlock function:

[0072] This step ensures that the trolley pole raising gas path is cut off through physical operation, physically eliminating the possibility of trolley pole raising, and further enhancing safety.

[0073] Only after the gas path is cut off and the three-way valve is locked can the second-level key be removed, ensuring the mandatory interlock between gas path cutting and subsequent high-voltage operations.

[0074] The acquisition of the second-level key is strictly limited by the previous operation, ensuring that only after the gas path cutting operation is completed can the operation continue.

[0075] The extraction of the second-level key indicates the start of the next stage of operation, i.e., unlocking the high-voltage related equipment.

[0076] Unlock the traction inverter, auxiliary inverter, medium-voltage bus box, and external power supply with the second-level key to operate and maintain the high-voltage related equipment.

[0077] It should be noted that the high-voltage related equipment is provided with physical interlocking (first-level key and second-level key), and only the corresponding key can be obtained to unlock.

[0078] Ensure that only after the previous operations are completed (such as grounding state confirmation and trolley pole raising cutting) can the high-voltage related equipment be accessed, preventing accidental contact with high-voltage during operation.

[0079] Therefore, in order to further improve the safety of the above-mentioned high-voltage interlocking system, the present application will be further described in conjunction with the embodiments.

[0080] Example 1 (see Figure 1 ): A high-voltage interlocking system applied to a rail vehicle, comprising:

[0081] The first interlock monitoring module is used to collect and monitor the relevant data of the air circuit interlock box after the driver's cabin fails to switch the electric ground key switch to the locked position and the first level key is removed, and the specific operation is as follows:

[0082] The running state of each air circuit interlock box in the track high-voltage interlocking system is monitored, including:

[0083] The operating state of the air circuit interlock box is monitored,

[0084]

[0085] Wherein, Q FS represents the operating state data of the air circuit interlock box, M run represents the current operating torque of the three-way valve, M sat represents the initial operating torque of the three-way valve, G cut represents the current mechanical gap (physical distance) of the three-way valve, G sat represents the initial mechanical gap of the three-way valve, F cut represents the current operating torque of the lock core, F sat represents the initial operating torque of the lock core, R cut represents the current rotational resistance of the lock core, R sa t represents the initial rotational resistance of the lock core, GF represents the vibration time of the three-way valve (monitored by the built-in vibration sensor), FR represents the vibration time of the lock core, and α1, α2, α3, α4, α5 and α6 are the corresponding weight coefficients.

[0086] The air circuit state of the air circuit interlock box is monitored,

[0087]

[0088] Wherein, P FS represents the air circuit state data of the air circuit interlock box, M eas represents the current air circuit pressure, M sat represents the initial air circuit pressure, F eas represents the current air circuit gas flow, F sat represents the initial air circuit gas flow, MF represents the calculation item, which is preset to 2, and β1 and β2 represent the corresponding weight coefficients.

[0089] Therefore, based on the air circuit state data Q FS and the operating state data P FS , the running data QP of the air circuit interlock box at the current time point is obtained by weighted summation (the corresponding weight coefficients of the air circuit state data and the operating state data are assigned and added). t, based on the above expression, the operating data QP of the gas interlocking box at each time point (which can be constructed according to the time point of the hour) is calculated by the same reasoning t , by obtaining the operating data QP of the gas interlocking box at multiple time points t , time series data is constructed, a time series model is established based on the time series data, and parameters (α1, α2, … α p , θ1, θ2, … θ q ) of the model are determined based on the time series model, so as to predict the operating data QP at future time t+h , wherein the expression of the time series model is:

[0090] (1-α1B-α2B 2 -···-α p B p )(1-B) d QP t =μ+(1+θ1B+θ2B 2 +···+θ q B q )εt

[0091] Wherein, QP t represents time series data, B represents a lag operator, α1, α2, … α p represent the coefficients of the autoregressive part respectively, θ1, θ2, … θ q represent the coefficients of the moving average part respectively, p represents the order of the autoregressive part, which describes the use of the value at the previous p time to predict the current time, q represents the order of the moving average part, which describes the use of the previous q error items to correct the prediction value at the current time, d represents the difference order, which describes the number of times the time series data is differentiated to ensure the stationarity of the time series data, μ represents the constant term, ε t represents the error term, which describes the part of the time series data that is not explained; wherein, α1, α2, … α p and θ1, θ2, … θ q are solved by least squares method or maximum likelihood estimation (MLE), which is the conventional method for solving at present, and the present application will not be described.

[0092] Thus, based on the prediction formula, the operating data QP of the gas interlocking box at future time is predicted t+h , then,

[0093]

[0094] Wherein, αi represents the coefficient of the autoregressive part, θ j represents the coefficient of the moving average part, ε t+h-j represents the error term at the previous j time steps, QP t+h-iLet represent the operational data at time t+hi, and h represent the prediction step size, describing the future time point to be predicted. Therefore, based on the above model expression, the operational data QP of the gas interlock box at future times can be predicted. t+h ;

[0095] Therefore, the time value U required to operate and unlock the pneumatic interlock box in this rail vehicle operation should be obtained. j This refers to the estimated time required to operate and unlock the pneumatic interlock box. Therefore, the time value U... j Substituting the above prediction formula, we can obtain the operating data QP of the gas interlock box at a future time. t+h (This refers to the time when the operation is completed, such as the end time of operations such as maintenance and repair of rail vehicles, describing the time from the current time to t+h (i.e., the time between the current time and U). j (Whether the combined time of the gas circuit interlock box operation data can support continuous unlocking) and further, based on future operation data QP t+h Determine whether it exceeds the preset operational risk threshold of the pneumatic interlock box:

[0096] If the limit is not exceeded, the first-level key is allowed to be inserted into the gas circuit interlock box and unlocked, thereby unlocking the three-way valve operating handle, cutting off the lifting air circuit, locking the three-way valve handle, and removing the second-level key;

[0097] If the limit is exceeded, a prohibition on unlocking command will be generated, which prohibits the first-level key from being inserted into the gas interlock box and unlocking (that is, providing a safety restriction for the subsequent unlocking of the high-voltage interlock control system).

[0098] This avoids unlocking and using the pneumatic interlock box when there is an operational risk, reducing the probability of the pneumatic interlock box malfunctioning further. Therefore, when a prohibition on unlocking command is generated, it is simultaneously output to the rail train system (management and control center) for early warning, indicating to the operator to handle the pneumatic interlock box malfunction. Unlocking and maintenance of high-voltage related equipment in the rail train system are only permitted after ensuring that the entire high-voltage interlock system of the rail vehicle is in a safe state. That is, in the high-voltage interlock system, the initial operation on the pneumatic interlock box is confirmed to be safe before subsequent handling steps of high-voltage related equipment are allowed (in the rail train system, the unlocking and closing of equipment must be strictly controlled), thereby greatly protecting the personal safety of the operators.

[0099] Generally, the need to unlock high-voltage equipment with a second-level key indicates an anomaly in the high-voltage equipment of the rail vehicle, requiring maintenance and repair. Therefore, to accurately assess the condition of the high-voltage equipment, determine the severity of the fault, and estimate the appropriate maintenance time, the following steps are also required:

[0100] A second interlock monitoring module is configured to determine a coordinate of the second key, denoted as a second coordinate, determine a coordinate of the high-voltage related device, denoted as a high-voltage device coordinate, determine a distance based on the second coordinate and the high-voltage device coordinate, and calculate a predicted travel time SS t , which is obtained according to a ratio of the travel distance and the speed, and will not be described herein again (the second key can be taken out after the air circuit interlock box is unlocked, and then it is determined that the first key and the second key are at the same coordinate position, and even if the air circuit interlock box is not unlocked, the distance based on the second coordinate and the high-voltage device coordinate can still be determined and the predicted travel time SS t and the following steps) can be calculated.

[0101] Subsequently, the state of the high-voltage related device is monitored and identified, wherein the high-voltage related device includes a traction converter, an auxiliary converter, a medium-voltage bus, and an external power supply.

[0102] For the state determination of the traction converter:

[0103] The time series data x(t) in the high-voltage related device (the traction converter, the auxiliary converter, the medium-voltage bus, or the external power supply) is collected, including temperature data (used to monitor whether the high-voltage related device is overheated, especially when the load is high, the temperature signal is an important indicator for evaluating faults), vibration data (reflecting mechanical faults, looseness, collisions, or other abnormal states), and current data (reflecting the fluctuation of current, especially high-frequency current fluctuation can reveal electrical faults or abnormal electronic components), the information of the running state of the key components of the high-voltage related device can be captured through the temperature data, the vibration data, and the current data, and then normalized processing is performed.

[0104] Then, short-time Fourier transform is performed on each time series data to generate a time-frequency graph S(f, τ):

[0105]

[0106] wherein ω(t-τ) represents a short-time window function (including a Hamming window or a Hanning window), τ represents the center position of the short-time window function, represents the position of the time axis, f represents a frequency variable, e -j2πft represents a complex exponential basis function, and t represents a time variable. By converting the time series data into a two-dimensional image form, both time information and frequency components are retained, which facilitates subsequent CNN extraction of time-frequency features and capture of complex dynamic characteristics in the running process of the high-voltage related device.

[0107] Further, the CNN model is trained, including:

[0108] The time-frequency graph S(f, τ) is inputted;

[0109] A convolutional layer is used to extract local time-frequency features, and then

[0110]

[0111] wherein, represents the output of the lth layer kth convolution kernel, W l,k,K，m,n represents the weight matrix of the lth layer kth convolution kernel in size K, describes the weight value of the m,n position in the convolution kernel, S m,n represents the pixel value of the m,n position in the input feature map, b k represents the bias top of the kth convolution kernel in the convolution layer, and σ represents the activation function, such as ReLU, and thus, features are extracted from the input data, i.e., from the time-frequency graph, through multiple convolution kernels and multi-scale convolution operations.

[0112] Further, the feature dimension is reduced by the pooling layer to extract the main features, that is,

[0113]

[0114] wherein, P i,j is the output result after the pooling operation, represents the pooling result corresponding to the i,j position in the pooling window, (m,n)∈window represents a position in the pooling window, window represents a window defining the size of the pooling region, and is usually a fixed-size rectangular region, F l,k,m,n represents the value of the kth output feature map in the convolution layer l at the m,n position, and thus, the maximum value F l,k,m,n in each window is selected in the pooling window, and is taken as the pooling result P i,j to reduce the size of the feature map and extract the most important features.

[0115] and a fully connected layer for outputting a high-dimensional feature vector, that is,

[0116] h = σ (W f · X + b f )

[0117] wherein, h represents a high-dimensional feature vector extracted by the CNN, W f represents the weight matrix of the fully connected layer, X represents the feature input of the previous layer (i.e., the flattened result of the feature map P i,j after the pooling), and b f represents the bias top of the fully connected layer.

[0118] Therefore, the above scheme extracts local features from the input time-frequency graph based on convolution operations, further simplifies the feature dimension with the pooling layer to retain important information and reduce noise interference, and then integrates the feature output of the convolution layer through the fully connected layer to form a high-dimensional feature vector, which is convenient for subsequent SVM classifier processing.

[0119] Further, the SVM classifier is trained with the high-dimensional feature vector h extracted by the CNN as input to classify the severity of the fault of the high-voltage related equipment. Assuming that the classes of the fault severity are continuous or discrete levels, they are defined as different class labels, including normal, slight fault, moderate fault, and severe fault, and thus,

[0120] The SVM objective function is defined as

[0121]

[0122] where ω represents the weight vector of the classifier, which determines the direction and position of the classification hyperplane, b represents the bias term, which is used to determine the position of the classification hyperplane, and ξ i is the slack variable, which represents the classification error of the i-th sample. If a certain sample is misclassified or is close to the hyperplane, the slack variable will be positive. C represents the penalty coefficient, which balances the classification margin and the error. N represents the total number of samples in the sample data set. Therefore, by the regularization term, overfitting is avoided, and the constructed classifier has good generalization ability. And through the penalty term of the slack variable, samples that cannot be completely correctly classified are processed.

[0123] The SVM constraint condition is defined, which ensures that the model can correctly classify the training data and at the same time allows a certain degree of misclassification. The form of the constraint condition is as follows,

[0124] y i (ω·h i +b)≥1-ξ i ,ξ i ≥0

[0125] where y i represents the true class label of the i-th sample, and y i is set to the class label representing the severity of the fault (such as 0, indicating normal; 1 indicating slight fault; 2 indicating moderate fault; 3 indicating severe fault). h i represents the high-dimensional feature vector of the i-th sample, 1-ξ i represents that the right side of the constraint condition is related to the slack variable, and ξ i ≥0 represents the non-negativity of the slack variable. Therefore, by training the SVM and optimizing the weight vector ω and the bias term b, the classifier can classify the severity of the fault of the high-voltage related equipment according to the high-dimensional feature vector h i The severity of the fault of the high-voltage related equipment is classified, and the classification result obtained is used to judge the severity of the fault, such as:

[0126] 0 represents normal; 1 represents slight fault; 2 represents moderate fault; and 3 represents severe fault.

[0127] In the above, when the label is 0 (normal), the temperature, vibration, and current indicators of the high-voltage related equipment are within the normal range. When the labels are 1, 2, and 3, they represent the degree to which the temperature, vibration, and current indicators of the high-voltage related equipment exceed the normal range, respectively, with the range gradually increasing. By setting a reasonable threshold range, the above classification can be performed, and this classification method is also a conventional method, so it will not be elaborated further. Therefore, by labeling the fault data of high-voltage related equipment and using SVM for multi-category classification, the fault status can be identified based on sensor data and classified into the above four levels, clarifying whether the current high-voltage related equipment is faulty and the severity of the current fault.

[0128] Therefore, once the severity of the fault in the high-voltage related equipment is determined, the preset maintenance and handling time P corresponding to the fault severity is obtained. GS Therefore, the corresponding maintenance and handling time P for the faulty high-voltage related equipment is obtained. GS It should be noted that in rail train systems, the following situations typically exist, each with a time value U. j The corresponding determination methods include:

[0129] Firstly, in a railcar system, a single pneumatic interlocking box synchronously controls multiple high-voltage related devices (traction converter, auxiliary converter, medium-voltage busbar, and external power supply).

[0130] In this situation, if multiple faulty high-voltage related devices are identified simultaneously, only the longest maintenance time P among the multiple high-voltage related devices (different types of high-voltage related devices have different preset maintenance and handling times) needs to be selected. GS The corresponding high-voltage related equipment is used as the target maintenance and handling time P. GS′ (Because multiple high-voltage related equipment with faults need to be inspected and repaired simultaneously), this is because even after the inspection and repair of other high-voltage related equipment is completed, it will still take the longest inspection and repair time P. GS The railcar system can only be restarted after the corresponding high-voltage equipment has been inspected and repaired; that is, the pneumatic interlock box can only be closed after that.

[0131] U j =P GS′ +SS t

[0132] Therefore, the time value U required for operating and unlocking the gas circuit interlock box in the above scheme is realized. j This is determined in order to obtain the future unlocking completion time U of the gas interlock box. jw And determine the time U after unlocking is completed. jw The following future running data OP W, thereby judging whether the air circuit interlocking box can be operated at the time value U j The first level key is inserted into the air circuit interlocking box and is unlocked, so that the scheme not only realizes the fault supervision of the high-voltage related equipment, but also simultaneously implements the safety state supervision of the air circuit interlocking box, thereby guaranteeing the safety of the entire high-voltage interlocking system.

[0133] Secondly, in the rail train system, one air circuit interlocking box controls one high-voltage related equipment (one air circuit interlocking box is interlocked with one of the traction converter, the auxiliary converter, the medium-voltage bus or the external power supply, and is independently controlled)

[0134] Therefore, the time value U

[0135] U j = P GS + SS t

[0136] The time value U j In the above formula, the air circuit interlocking box is interlocked with the corresponding high-voltage related equipment, so that only the maintenance disposal time P GS of the high-voltage related equipment corresponding to the air circuit interlocking box needs to be obtained.

[0137] In addition, the present application also provides (the following specific reference to the above system can be, hereinafter will not be repeated):

[0138] The rail vehicle high-voltage interlocking device comprises a memory and a processor, the memory stores a computer program, and the processor realizes the system of any one of the above when executing the computer program.

[0139] A computer readable storage medium, which stores a computer program, the computer program is executed by the processor to realize the system of any one of the above.

[0140] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit it; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements will not make the essence of the corresponding technical solutions deviate from the protection scope of the technical solutions of the embodiments of the present application.

Claims

1. An application to a high-voltage interlocking system of a railway vehicle, characterized by, The system comprises: a first interlock monitoring module for monitoring the running state of the air circuit interlock box in the track high-voltage interlocking system, wherein: the running state is determined by monitoring the operation state and the air circuit state of the air circuit interlock box; the running data of the air circuit interlock box at different time points is obtained and time series data is constructed, a time series model is established based on the time series data, the time series model is combined with the time value of the unlocking of the air circuit interlock box at the future time, the running data of the air circuit interlock box at the future time is determined, and it is determined whether the first level key is allowed to unlock the air circuit interlock box; and a second interlock monitoring module for determining the expected travel time according to the second level key and the high-voltage related equipment; time series data of the high-voltage related equipment is obtained to generate a time-frequency graph, local time-frequency features are extracted and a high-dimensional feature vector is formed, the high-dimensional feature vector is input into an SVM classifier to determine the fault severity of the high-voltage related equipment, the corresponding maintenance disposal time is determined, a time value is constructed based on the maintenance disposal time and the expected travel time, and it is determined whether the air circuit interlock box is unlocked.

2. The application of claim 1, wherein, The determination method of the running data of the air circuit interlock box is as follows: monitoring the operation state of the air circuit interlock box, wherein, Q FS represents the operating state data of the gas circuit interlocking box, M run represents the current operating torque of the three-way valve, M sat represents the initial operating torque of the three-way valve, G cut represents the current mechanical clearance of the three-way valve, G sat represents the initial mechanical clearance of the three-way valve, F cut represents the current operating torque of the lock cylinder, F sat represents the initial operating torque of the lock cylinder, R cut represents the current rotating resistance of the lock cylinder, R sat represents the initial rotating resistance of the lock cylinder, GF represents the vibration time of the three-way valve, FR represents the vibration time of the lock cylinder, and α1, α2, α3, α4, α5, and α6 are weight coefficients corresponding to the respective parameters. monitoring the air circuit state of the air circuit interlock box, Wherein, P FS represents the gas path state data of the gas path interlocking box, M eas represents the current gas path pressure, M sat represents the initial gas path pressure, F eas represents the current gas path gas flow, F sat represents the initial gas path gas flow, MF represents a calculation term, and β1 and β2 represent corresponding weight coefficients. Gas path state data Q based on the gas path interlock box FS and operating state data P FS to obtain operating data QP of the gas path interlock box at the current point in time t .

3. The application of claim 1, wherein, The method for determining the future running data of the air circuit interlock box is: Obtain the running data QP of the gas circuit interlocking box at multiple time points t to construct time series data; establishing a time series model, then, (1 - α1B - α2B 2 -···-α p B p )(1 - B) d QP t = μ + (1 + θ1B + θ2B 2 +···+θ q B q )εt where QP t denotes timing data, B denotes a lag operator, a1, a2, · · · a p denote coefficients of the autoregressive part, 1, 1, · · · 1 q denote coefficients of the moving average part, p denotes the order of the autoregressive part, q denotes the order of the moving average part, d denotes the order of differencing, m denotes a constant term, e t denotes an error term; Predicting operating data QP of an airpath interlock box at a future time t+h Then, where, α i denotes the coefficient of the autoregressive part, θ j denotes the coefficient of the moving average part, ε t+h-j denotes the error term of the past j time steps, QP t+h-i denotes the operational data at time t + h - i, h denotes the step length of the prediction.

4. The application of claim 1, wherein, The determination method of the expected travel time is: determining a coordinate in which the second level key is located and denoted as a second coordinate, determining a coordinate of the high-voltage related device and denoted as a high-voltage device coordinate, determining a distance based on the second coordinate and the high-voltage device coordinate and calculating an estimated travel time SS t .

5. The application of claim 1, wherein, The expression for generating a time-frequency graph from time series data of the high-voltage related equipment is: performing short-time Fourier transform on the time series data to generate a time-frequency graph S(f,τ): where ω(t - τ) represents a short-time window function, τ represents a center position of the short-time window function, t represents a position of a time axis, f represents a frequency variable, e -j2πft represents a complex exponential basis function, and t represents a time variable.

6. An application for a high-voltage interlocking system of a railway vehicle according to claim 5, characterized in that, The method for forming a high-dimensional feature vector is: inputting the time-frequency graph S(f,τ) and extracting local time-frequency features, then, wherein, represents the multi-channel feature map of the output of the lth layer kth convolution kernel, W l,k,K，m,n represents the weight matrix of the lth layer kth convolution kernel under the size K, S m,n represents the pixel value of the m,nth position in the input feature map, b k represents the bias top of the kth convolution kernel in the convolution layer, and σ represents the activation function. reducing the feature dimension through a pooling layer, then, wherein P i,j is the result output after the pooling operation, (m, n) e window represents the position within the pooling window, window represents the window defining the size of the pooling region, F l,k,m,n represents the value of the kth output feature map in the convolutional layer l at position m, n; outputting a high-dimensional feature vector through a fully connected layer, then, h = σ(W f • X + b f ) wherein h represents a high-dimensional feature vector extracted by the CNN, W f represents a weight matrix of the fully connected layer, X represents a feature input of the previous layer, b f represents a bias of the fully connected layer.

7. The application for a high-voltage interlock system for a rail vehicle according to claim 6, characterized in that The method for determining the fault severity of the high-voltage related equipment through an SVM classifier is: defining an SVM objective function, where ω represents the weight vector of the classifier, b represents the bias top, ξ represents the set of all sample relaxation variables, ξ represents the set of all sample relaxation variables, and ξ i is the relaxation variable of the i-th sample, C represents the penalty coefficient, and N represents the total number of samples in the sample data set. defining an SVM constraint condition, the form of the constraint condition is as follows, y i (ω·h i +b)≥1-ξ i ,ξ i ≥0 where y i represents the true class label of the i-th sample, and y i is set as the class label representing the severity of the fault, h i represents the high-dimensional feature vector of the i-th sample, 1-ξ i represents the right side of the constraint condition related to the slack variable, ξ i ≥ 0 represents the non-negativity of the slack variable.

8. The application of claim 7, wherein, The determination method of the time value is: determining the high-voltage related equipment controlled by the air circuit interlock box in the current track train system; to determine the severity of the fault of the high-voltage related device controlled by the control device GS ; P = P + P GS S = S + S t U = U + U j .

9. A high-voltage interlock device for a rail vehicle, comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to realize the system of any one of claims 1 to 8.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the system of any one of claims 1 to 8.

Images