A distributed wiring harness test system based on SCTP communication protocol

Abstract

The present invention discloses a distributed harness test system based on the SCTP communication protocol, comprising an interface layer, a network layer, a transport layer, an intermediate layer and an application layer; the interface layer is connected to the network layer, the network layer is connected to the transport layer, the transport layer is connected to the intermediate layer, and the intermediate layer is connected to the application layer. The present invention belongs to the technical field of harness testing, and specifically refers to a distributed harness test system based on the SCTP communication protocol, which can capture the timing relationship of data based on a recurrent neural network (RNN), can effectively monitor the timing relationship of SCTP multi-test communication data streams, and thus judge abnormal data timing behavior.

Description

Technical Field

[0001] The invention belongs to the technical field of wire harness testing, and in particular refers to a distributed wire harness testing system based on the SCTP communication protocol. Background Art

[0002] SCTP (Stream Control Transmission Protocol) is a message-oriented transport layer protocol, similar to TCP and UDP, but with some unique features, such as multi-stream support, partial reliability, and better message boundary control. SCTP allows multiple logical streams to exist in parallel on a connection, which enables multiple test data streams to be transmitted simultaneously in the test system. The establishment and maintenance of connections requires the system to support multipoint-to-multipoint communication. SCTP allows the creation of multiple independent logical streams on a connection, which is very useful for performing multiple test tasks or transmitting multiple data streams simultaneously. The test system can use this feature to improve test efficiency.

[0003] However, the following problems usually occur during the testing of SCTP communication harness test system:

[0004] SCTP communication usually involves continuous data streams, which usually produce certain abnormal behaviors in time series; currently there is no accurate method to detect abnormal behaviors in time series. Summary of the invention

[0005] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a distributed wiring harness testing system based on the SCTP communication protocol, which can capture the timing relationship of data based on a recurrent neural network (RNN), and can effectively monitor the timing relationship of SCTP multi-test communication data streams, thereby judging abnormal data timing behavior.

[0006] The technical solution adopted by the present invention is as follows: A distributed wiring harness test system based on the SCTP communication protocol provided by the present invention includes an interface layer, a network layer, a transport layer, an intermediate layer and an application layer;

[0007] The interface layer is connected to the network layer, the network layer is connected to the transport layer, the transport layer is connected to the intermediate layer, and the intermediate layer is connected to the application layer;

[0008] The interface layer is used to process the underlying hardware and network connection, and is used for LCR measurement, low resistance measurement, insulation resistance measurement, DC withstand voltage measurement, AC withstand voltage measurement, diode withstand voltage measurement and conduction measurement;

[0009] The network layer is used to use the SCTP communication protocol to communicate between the host and the slave;

[0010] The transport layer is used for data transmission;

[0011] The middle layer is used to embed the recurrent neural network into the SCTP communication protocol;

[0012] The application layer is used to provide a direct user interface.

[0013] Furthermore, a wiring harness testing method for improving the SCTP communication protocol based on a recurrent neural network comprises the following steps:

[0014] The host sends a command to the slave, and the slave responds to the command to start the conduction test;

[0015] The slave obtains the measurement data result;

[0016] The slave transmits measurement data to the master;

[0017] The host uses a recurrent neural network model to process the measurement results;

[0018] The host summarizes all test results, obtains the point conduction point and determines data anomalies.

[0019] Furthermore, in “the host uses a recurrent neural network model to process the measurement result”, the following steps are included:

[0020] Expand the SCTP protocol to transmit the information required by the recurrent neural network;

[0021] Preprocessing of measurement data to adapt to the model input requirements;

[0022] Pass the processed measurement data to the recurrent neural network model for analysis;

[0023] Get the output data of the recurrent neural network;

[0024] Define data structures;

[0025] Integrate the output data of the recurrent neural network into the defined data structure;

[0026] The integrated data is transmitted to the host.

[0027] Furthermore, in “slave device transmits measurement data to host device”, the following steps are included:

[0028] The host sends control signals to the slave;

[0029] The slave switches the information transmission channel;

[0030] Start the continuity test;

[0031] The slave sends the measurement data results to the master.

[0032] Further, in “obtaining measurement data results from the machine”, the method further includes:

[0033] Connect the power supply to the operational amplifier to output an operational voltage;

[0034] After connecting to the device under test, a current is generated;

[0035] After the current passes through the resistor, a measurement voltage is generated on the operational amplifier;

[0036] Connect the operation voltage and the measurement voltage to the comparator;

[0037] The comparator outputs a conducting signal or a non-conducting signal of the device under test.

[0038] Furthermore, in “the host uses a recurrent neural network model to process measurement results”, the architecture adopted by the recurrent neural network is a long short-term memory network.

[0039] The beneficial effects achieved by the present invention using the above structure are as follows:

[0040] Recurrent Neural Networks (RNNs) are a class of neural network structures specifically designed to process sequential data. Unlike traditional feedforward neural networks, RNNs have recurrent connections that allow information to be passed around the network and maintain state. This structure makes RNNs very effective at processing sequential data. SCTP communications typically involve continuous streams of data. RNNs can capture the temporal relationship of data and therefore may be useful in detecting abnormal behavior based on time series. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A wiring diagram of a wiring harness test system provided by the present invention;

[0042] Figure 2 A connection diagram between a host and a slave provided by the present invention;

[0043] Figure 3 SCTP protocol diagram between the host and the slave provided by the present invention;

[0044] Figure 4 A schematic diagram of the system architecture provided by the present invention;

[0045] Figure 5 A diagram of a wiring harness test system based on a recurrent neural network to improve the SCTP communication protocol provided by the present invention;

[0046] Figure 6 A flow chart of the SCTP communication protocol improved by the recurrent neural network provided by the present invention;

[0047] Figure 7A flow chart of a wiring harness testing method for improving the SCTP communication protocol based on a recurrent neural network provided by the present invention;

[0048] Figure 8 A flow chart of obtaining a return signal provided by the present invention.

[0049] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0051] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0052] Embodiment 1: Extending SCTP to support RNN;

[0053] Step 1: Add a new field in the SCTP header to pass the information required by the RNN;

[0054] Part of the code is as follows:

[0055]

[0056] Step 2: Simulate the timing relationship between multiple streams and maintain a sequence number or timestamp for each stream;

[0057]

[0058] Step 3: Modify the code for sending and receiving data to ensure that it is processed according to the new fields and timing relationships.

[0059] Send data code:

[0060]

[0061]

[0062] Code for receiving data:

[0063]

[0064] Example 2: Embed the trained RNN model into the middle layer;

[0065] Step 1: Import the deep learning framework;

[0066] Code process:

[0067] import tensorflow as tf#Use the TensorFlow deep learning framework in Python;

[0068] Step 2: In the middle layer code;

[0069] Define the same RNN model structure as during training; the code process is as follows:

[0070] model=tf.keras.models.Sequential([

[0071] tf.keras.layers.LSTM(units=64,input_shape=(sequence_length,input_dim)),

[0072] tf.keras.layers.Dense(units=output_dim,activation='softmax')

[0073] ])#Add other levels or custom structures;

[0074] Step 3: Load pre-trained weights;

[0075] Load the trained weights into the model, keeping the model structure the same as during training. The code is as follows:

[0076] model.load_weights('path / to / pretrained_weights.h5')#Change to the path of your pre-trained weight file;

[0077] Step 4: Data preprocessing;

[0078] In the middle layer, the data transmitted by the transport layer is preprocessed to meet the input requirements of the model; the code is as follows:

[0079] input_data = preprocess_data(input_data) # In TensorFlow, format the data;

[0080] Step 5: Model reasoning;

[0081] The code is as follows:

[0082] predictions = model.predict(input_data) #Using the model loaded with pre-trained weights for inference;

[0083] Step 6: Result processing;

[0084] Integrate the output of the model with the data of the transport layer; add the output of the RNN to the data of the transport layer so that the processed data is returned to the transport layer.

[0085] Step 7: Test;

[0086] Conduct comprehensive testing to ensure that the integrated RNN model in the middle layer functions properly, including unit testing, integration testing, and performance testing.

[0087] Embodiment 3: Conductivity data measurement and data timing abnormal behavior judgment;

[0088] During the continuity measurement, the host sends point control information. After receiving the control information, the slave returns a confirmation signal to the host. After receiving the confirmation signals from all slaves, the host starts to send the clock signal. After receiving the clock signal, the slave starts the continuity test. Vs is a 5V power supply, Rx is the device under test for the continuity test, R1 and R2 are voltage divider resistors, and vs is connected to the op amp A1, and then a voltage is output, which is used as the reference voltage ref of the DAC. When the device under test is connected, a current is generated. After the current passes through R2, a voltage is generated on the op amp A2. The output of the op amp A2 and the voltage output of the DAC are connected to a comparator. If the comparator value is 1, the device under test Rx is turned on. If the comparator value is 0, the device under test Rx is not turned on. The calculation formula for the resistance Rs value of the continuity test is as follows:

[0089] V1=Vs*DAC / (2*4096);-------------(1)

[0090] V2=k*R2*Is;-----------------------------(2)

[0091] V1=V2;

[0092] From (1) and (2) => Vs / Is = Rs = 2k*4096*R2 / DAC;

[0093] After the test is completed, the slave sends the measurement data results to the master. The test results are which points to be tested are connected with the points in the slave. The master summarizes the test results of all channel cards to determine which points of the device under test are connected.

[0094] When the above work is completed;

[0095] The host uses a recurrent neural network to judge abnormal data timing behavior, initializes the test results in Example 2, uses a recurrent neural network model to infer the initialized measurement data, and finally outputs the result data, outputs the result data to the host, and the host makes a data timing judgment.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants 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.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0098] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A wiring harness testing method based on a recurrent neural network to improve the SCTP communication protocol, using a distributed wiring harness testing system based on the SCTP communication protocol, the distributed wiring harness testing system based on the SCTP communication protocol comprising: Includes interface layer, network layer, transport layer, middle layer and application layer; The interface layer is connected to the network layer, the network layer is connected to the transport layer, the transport layer is connected to the intermediate layer, and the intermediate layer is connected to the application layer; The interface layer is used to process the underlying hardware and network connection, and is used for LCR measurement, low resistance measurement, insulation resistance measurement, DC withstand voltage measurement, AC withstand voltage measurement, diode withstand voltage measurement and conduction measurement; The network layer is used to use the SCTP communication protocol to communicate between the host and the slave; The transport layer is used for data transmission; The middle layer is used to embed the recurrent neural network into the SCTP communication protocol; The application layer is used to provide a direct user interface; The steps include: The host sends a command to the slave, and the slave responds to the command to start the conduction test; The slave obtains the measurement data result; The slave transmits measurement data to the master; The host uses a recurrent neural network model to process the measurement results; The host summarizes all test results, obtains the point conduction point and determines data abnormalities; In the host using the recurrent neural network model to process the measurement result, the method further includes: Expand the SCTP protocol to transmit the information required by the recurrent neural network; Preprocessing of measurement data to adapt to the model input requirements; Pass the processed measurement data to the recurrent neural network model for analysis; The host uses a recurrent neural network to judge abnormal behavior of data timing; Get the output data of the recurrent neural network; Define data structures; Integrate the output data of the recurrent neural network into the defined data structure; Transmit the integrated data to the host; In transmitting the measurement data from the slave to the master, the method further includes: The host sends control signals to the slave; The slave switches the information transmission channel; Start the continuity test; The slave sends the measurement data results to the master; In obtaining the measurement data result from the slave, the method further includes: Connect the power supply to the operational amplifier to output an operational voltage; After connecting to the device under test, a current is generated; After the current passes through the resistor, a measurement voltage is generated on the operational amplifier; Connect the operation voltage and the measurement voltage to the comparator; The comparator outputs a signal that the device under test is conducting or a signal that the device under test is not conducting; When the host uses a recurrent neural network model to process measurement results, the architecture adopted by the recurrent neural network is a long short-term memory network.

Images