Research and development of electrical reliability test monitoring system

Abstract: This paper discusses a circuit multi-parameter test system for real-time monitoring of vibration reliability test of automobile electrical system. The system takes LXI bus modular virtual instrument as the core, LabWindows/CVI as the software development platform, and effectively combines computer fault diagnosis technology to form a test monitoring system that can independently complete remote control, testing, data processing and real-time fault diagnosis, laying a foundation for the intelligence and traceability of the test process of automobile electrical products.

1 Introduction

Due to the existence of two major excitation sources, road excitation and engine vibration, the proportion of automotive electrical and electronic system failures in vehicle failures is extremely high, and the trend is increasing year by year. Conducting road simulation vibration tests on vehicles and their components in the laboratory is considered an effective means to accelerate product development and improve product quality. The traditional test process mostly uses manual supervision to record relevant data. This method has the following problems:
1. The test environment is harsh, often accompanied by noise, humidity and heat and other factors;
2. The time is long and the workload of the on-duty personnel is heavy;
3. Manually recorded data lacks integrity and consistency;
4. The fault phenomenon is not traceable and cannot provide sufficient basis for fault analysis;
these shortcomings have greatly affected the effectiveness of the test and cannot conduct in-depth analysis of the test results. Due to the complex control principle of the automotive electrical system itself, the large number and variety of components and various structural forms, there is no effective method for real-time monitoring of the test process of the vehicle electrical system with the help of automated test equipment. This paper focuses on a construction method of a test system for vehicle electrical system test monitoring and fault diagnosis applied to the "Automobile Electrical System Reliability Test Bench".

2 Working Principle of Monitoring System

The automotive electrical reliability test bench applies electrical stress and vibration stress with the help of auxiliary test equipment, so that the automotive electrical system on the test bench can simulate the actual working conditions of the automotive road test. The "Automobile Electrical System Reliability Test Real-time Monitoring System" uses accessible nodes in the circuit as monitoring points, and uses data acquisition equipment to track, measure and record signals such as voltage, current, and frequency in each electrical circuit; through application software, the unit data is processed and analyzed in real time, faults are discovered in time, sound and light alarms are realized, and the type and location of typical faults are diagnosed.

2.1 Basic system design concept

Due to the complexity of the system under test, the monitoring system is designed based on the system planning method of "UUT (Unit Under TEST) classification", as shown in Figure 1. The hardware and software design of the system is carried out according to the UUT type. The reasons for adopting this method are:
1. Conventional electrical equipment in the automotive electrical system adopts a parallel connection mode, and each electrical circuit is used as a test unit UUT. Then the entire system under test can be used as a circuit network system with multiple UUTs tested in parallel;
2. Some electrical equipment in the car has commonalities in its working mode and failure form. Combined with its own characteristics and relevant standard requirements, all UUT units can be divided into several typical categories such as lamps, motors, and instruments;
3. Designing for UUT classification rather than for a single UUT can reduce system complexity and improve versatility.

Figure 1: Basic system design concept

2.2 Implementation of parameter measurement

When conducting reliability tests, it is necessary to conduct comprehensive monitoring and accurate fault diagnosis of all electrical appliances in the tested system. The prerequisites are: first, sufficient accessible test nodes can be obtained from the system without destroying the integrity of the tested system; second, the signal I/O interface must be reliably connected and able to withstand high-intensity test stress without failing before the tested system. Determine the signal to be tested and the sampling node according to the UUT type, and use the connector actually used in automotive electrical appliances as the signal output interface. Design the sampling connector access circuit to obtain the signal required for monitoring. Figure 2 illustrates the sampling method of a resistive electrical unit. In the original state, the automotive electrical unit Rx is directly connected to the automotive wiring harness to form a working loop, and its positive and negative loop resistances are r1 and r2 respectively, and the resistance value is unknown; when testing, the "sampling connector" containing the circuit shown in the figure is connected, and the circuit state parameters can be obtained without affecting the operation of the original circuit.

Figure 2: Resistive electrical unit monitoring principle

By building a multi-channel data acquisition system and performing real-time measurement and data processing of electrical parameters at each node, accurate fault identification and location can be achieved.

3 Synthesis of system test instruments

The system uses a measurement platform based on the LXI (LAN eXtensions for Instrumentation) bus. LXI is a new generation of LAN-based modular platform standards suitable for automatic test systems. The LXI modular test standard combines the high performance of GPIB instruments, the small size of VXI/PXI card instruments, and the high-speed throughput of LAN, and takes into account instrument requirements such as timing, triggering, cooling, and electromagnetic compatibility. At the same time, it also has many advantages and characteristics, such as: it is an open industrial standard system with downward compatibility, low instrument development cost, good collaborative working ability, and scalability. Its data transmission gets rid of the limitations of traditional instruments on data transmission distance and bandwidth, and can easily realize remote control of instruments and long-distance high-bandwidth data transmission. In harsh test environments, such as high-noise environments in vibration tests, LXI has better applicability than other bus platforms due to its advantages in remote control and data transmission networks.

3.1 Main Controller

As shown in Figure 3, the system uses an industrial computer as the main controller, controls the modules through the LAN network, and completes the system's data processing, display, storage, etc. As the main controller of the entire test system, the industrial computer is also responsible for the task of stress loading control of the reliability test.

3.2 LXI Virtual Instrument Module

1) The test host system uses AGILENT 34980A switch/measurement unit as the hardware platform, and measures, converts and outputs the side signal through the built-in digital multimeter. The test host communicates and exchanges data with the main control computer through the LAN bus.
2) Multiplexer Since the signals in the system under test are mainly relatively stable DC analog signals, a public DMM time-sharing measurement method is adopted. Two optoelectronic isolation ETF switch modules are used to achieve 80 channels of two-wire measurement.
3) The jitter measurement module is used for transient monitoring and detecting transient voltage jumps in the circuit.
4) D/A conversion module The D/A converter provides drive signals for the working components of the test system, such as the voltage pulse signals required for the tachometer and speedometer, and the current signals required for the fuel meter and water temperature meter. The signal is controlled by the main control computer, output by the D/A converter, and input into the test system through the analog I/O interface after conditioning.
5) Digital Oscilloscope The digital oscilloscope communicates and exchanges data with the host computer through the LAN bus, and connects to the test host through the analog bus, sharing the switch module with the built-in instrument. It can realize high-frequency sampling and virtual oscilloscope of any signal of the 80 measurement channels.

Figure 3: System hardware composition

3.3 Key issues

1) Frequency signal measurement The measured signal has both frequency signals above 100Hz and below 3Hz. Since the system uses a common DMM isochronous scanning measurement method, the two signals need to be measured using different sampling methods. For high-frequency signals, the system scanning channel is set to direct output of frequency measurement. For low-frequency signals below 3Hz, since the frequency channel cannot measure directly due to its low frequency, a fitting method is required. This method has high requirements for the system scanning frequency.

Nyquist theorem: or
The single-channel sampling rate should be determined by the upper frequency limit of the signal to be measured;

Therefore,
if 80 analog channels are scanned and sampled, the total switching frequency of the switch should be greater than 480CH/s. The system sets the clock value of a single scan to 160ms, and the actual scanning frequency is 500CH/s, which enables the measurement of low-frequency signals.

2) Realization of transient interruption monitoring As a circuit transient phenomenon, the sampling rate of the DMM time-sharing sampling method is too low to monitor such signals, and the multi-channel parallel analog data sampling leads to a large amount of data redundancy and excessive system cost. The system uses a jitter measurement module to monitor the voltage jump of each channel in real time with a 32-channel parallel digital sampling method. The maximum sampling rate of a single channel is 0.1μs. The monitoring voltage threshold is set to 10.5V/21V optional according to the test voltage of the automotive electrical appliances.
4 System application software design
4.1 Software development environment

The system uses LabWindows/CVI as the software development platform. It has interactive programming methods and rich library functions, providing an ideal software development environment for developers to build data acquisition and process monitoring systems, and is a fast way to realize virtual instruments and networked instruments.

4.2 Multithreading Technology in Experiment Monitoring

Windows is a weak real-time operating system. It achieves preemption through thread priority and meets the real-time requirements of most test tasks by setting appropriate priorities for test threads. Test monitoring requires that system control, data acquisition, data display and data analysis functions be completed synchronously. The thread pool technology in LabWindows/CVI multithreading can well achieve the real-time performance of the system. With interface control as the main thread, control instructions are issued to other threads through interface operations, so that the system can respond to user operations in a timely manner; data acquisition, real-time display, and fault diagnosis are auxiliary threads, which are executed synchronously with the main thread. In the auxiliary thread, the real-time display thread and the data analysis thread communicate with the data acquisition thread in real time through the pipeline message drive mechanism to realize data sharing between threads.

4.3 Fault Diagnosis Method

According to the principle of logical identification: the fault cause function A, the fault characteristic function X and the decision rule E satisfy the Boolean function relationship. The essence of the fault diagnosis process is to solve A from the known X and E. It is expressed in logical language as: The implementation method is to divide the test system into 6 typical monitoring unit types according to the working characteristics of the UUT, and design corresponding fault identification subroutines for each type. The contents include:

1. Establish a typical fault mode database based on the monitoring unit type, that is, construct the fault cause function A;
2. Use the measurable physical quantities I, U, f and other parameters of the circuit to describe the fault mode and construct the fault characteristic function X;
3. Establish the fault decision rule E based on logical judgment and convert it into the corresponding fault identification subroutine.

Figure 4: Simplified troubleshooting flow chart

During operation, the system issues a collection instruction and retrieves data. The data processing thread first compares the real-time data of each physical quantity to be measured with the threshold value corresponding to the UUT state parameter in the threshold value library. When it is found that there is data exceeding the threshold value, it is considered that a fault has occurred, and the fault identification subroutine corresponding to the type of UUT is started to perform fault diagnosis. After the diagnosis subroutine is executed, the diagnosis result is output and recorded. Figure 4 illustrates the fault diagnosis process of the system.

4.4 System data management and database

The system uses Microsoft SQL Server as the underlying database, establishes an ODBC data source through SQL Toolkit, and performs operations on database connection and data information access. System configuration data can be generated by an open user interface, allowing users to configure the system accordingly for different test objects, thus ensuring the flexibility and versatility of the system.

5 Conclusion

The system was applied to the vibration reliability test of a certain automobile enterprise, solving the problem of intelligent monitoring of the automobile electrical system test process. The results show that the system can correctly measure, display, record and replay various test physical quantities; it can diagnose and alarm faults in real time and accurately, effectively improving many disadvantages of traditional test monitoring methods, and can meet the engineering requirements of real-time monitoring of the test system; based on LXI bus instruments combined with virtual instrument software development technology, it is an effective means to build a comprehensive test and measurement system. The integration of modern testing technology and computer technology has made the automatic monitoring of automobile electrical system reliability tests a reality, making the test process intelligent and scientific; it provides a scientific basis for the failure mechanism analysis of the test system, the evaluation of reliability test results, and the design and quality improvement of automobile products.

The author's innovations are as follows: 1. A scheme for building a circuit multi-parameter test monitoring system based on LXI bus is proposed; 2. A real-time diagnosis method for automotive electrical faults based on UUT classification is established.