1. Introduction Programmable logic controllers (PLCs) are increasingly widely used in industrial control due to their strong anti-interference capabilities, high reliability, simple programming, and high performance-price ratio. The industrial PLC, as the central control unit, is equipped with configuration software and a large-screen real-time monitoring interface to achieve dynamic display, data modification, fault diagnosis, and automatic alarm for each control point. It can also display and query historical event records, cumulative operating time of major system components, process flow diagrams of each device, and structural diagrams of each device. Data exchange between the central control unit and the lower-level PLCs is conducted via serial communication. Typically, 485 twisted-pair communication is used for distances within 1000m, fiber optic communication can be used for longer distances, and wireless communication can be used for even longer distances. The lower-level PLCs are controlled by PLCs. Depending on the number and range of controlled objects, one or more PLCs can be used for control. Data exchange between PLCs is achieved through internal link registers. Because PLCs offer high reliability for real-time monitoring of the field and are simple and flexible to program, they are receiving increasing attention. 2. Main Reasons for Reduced Control System Reliability Although industrial control computers and programmable logic controllers (PLCs) themselves have high reliability, errors in the digital input signals to the PLC, significant deviations in analog signals, and failure of the actuators controlled by the PLC output ports to operate as required can all cause control errors and irreparable economic losses. The main reasons for errors in field input signals to the PLC include: 1) Short circuits or open circuits in the transmission signal lines (due to mechanical pulling, aging of the lines themselves, and especially rodent damage). When the transmission signal lines malfunction, field signals cannot be transmitted to the PLC, causing control errors; 2) Mechanical contact jitter. Although the field contacts may only close once, the PLC may interpret it as multiple closures. Even with added filtering circuits in the hardware and differential instructions in the software, the short scan cycle of the PLC can still lead to errors in counting, accumulating, and shifting instructions, resulting in incorrect control results; 3) Malfunctions in the field transmitters or mechanical switches themselves, such as poor contact, large deviations in the transmitter's non-electrical quantity readings, or malfunctions. These faults can also prevent the control system from functioning properly. The main reasons for actuator errors are: 1) The control load contacts cannot operate reliably. Although the PLC issues an action command, the actuator does not operate as required; 2) The frequency converter starts, but due to a fault in the frequency converter itself, the motor driven by the frequency converter does not work as required; 3) Various electric valves and solenoid valves fail to open when they should, or fail to close completely when they should. Because the actuator fails to operate according to the PLC's control requirements, the system cannot work normally, reducing system reliability. To improve the reliability of the entire control system, it is necessary to improve the reliability of input signals and the accuracy of actuator actions. Otherwise, the PLC should be able to detect problems in a timely manner and alert operators with audible and visual alarms to troubleshoot the fault as soon as possible, allowing the system to work safely, reliably, and correctly. 3. Design a comprehensive fault alarm system In the design of the automatic control system, we designed a three-level fault display alarm system. Level 1 is set on the control panel of each control cabinet at the control site, using indicator lights to indicate normal operation and fault status of the equipment. When the equipment is operating normally, the corresponding indicator light is on; when the equipment malfunctions, the indicator light flashes at a frequency of 1Hz. To prevent indicator light bulbs from malfunctioning and failing to accurately reflect equipment operation, a fault reset/lamp test button is specifically installed. Pressing this button continuously for 3 seconds at any time during system operation should illuminate all indicator lights. If any indicator light remains off, it indicates a faulty indicator light that should be replaced immediately. After resetting the button, the indicator lights will resume their original operating status. Level 2 fault displays are located on the large-screen monitor in the central control room. When equipment malfunctions, the fault type is displayed in text, the corresponding equipment on the process flow diagram flashes, and the fault is recorded in the historical event table. Level 3 fault displays are located in the signal box in the central control room. When equipment malfunctions, the signal box will alert personnel with audible and visual alarms, allowing for timely fault handling. Faults are further categorized during troubleshooting. Some faults require system shutdown, while others have minimal impact on system operation and can be resolved during operation. This significantly reduces overall system downtime and improves system reliability. 4. Input Signal Reliability Study To improve the reliability of signals input to the PLC from the field, firstly, high-reliability transmitters and switches should be selected to prevent short circuits, open circuits, or poor contact in the signal transmission lines caused by various reasons. Secondly, a digital filtering program should be added during program design to increase the reliability of the input signals. A timer should be added after the field input contact. The timing period should be determined based on the contact bounce and the required response speed of the system, generally within tens of milliseconds. This ensures that other responses only occur after the contacts have stably closed. Analog signal filtering can be performed using the programming method shown in Figure 2b. The field analog signal is sampled three times consecutively, with the sampling interval determined by the A/D conversion speed and the rate of change of the analog signal. The three sampled data are stored in data registers DT10, DT11, and DT12, respectively. After the last sampling, the maximum and minimum values ​​are removed using data comparison, data exchange, and data segment comparison instructions, retaining the intermediate value as the sampling result and storing it in data register DT0. Improving the reliability of field signals read from the PLC can also be achieved by utilizing the characteristics of the control system itself and judging the reliability of the signals based on the relationships between them. For example, in liquid level control, since the tank size, inlet/outlet valve openings, and pressures are known, the approximate range of liquid level changes within the tank over a certain time is also known. If the data sent to the PLC by the level gauge differs significantly from the estimated liquid level, it's suspected that the level gauge is faulty. The fault alarm system then notifies the operator to check the level gauge. Similarly, if each tank has upper and lower liquid level limit protection, a signal is sent to the PLC when the switch activates. To verify the reliability of this signal, the program design compares this signal with the signal from the tank's level gauge. If the level gauge reading is also at the limit position, the signal is considered valid; if the reading is not at the limit position, it's suspected that the level limit switch or the signal transmission line is faulty. Again , the alarm system notifies the operator to handle the fault. By employing these methods in the program design, the reliability of the input signal is greatly improved. 5. Actuator Reliability Study After the field signal is accurately input to the PLC, the PLC executes the program and uses the actuators to adjust and control the field devices. How to ensure the actuator operates according to control requirements, and how to detect faults when the actuator does not operate as required? We take the following measures: When the load is controlled by a contactor, starting or stopping this type of load switches to controlling the contactor coil. We are concerned with whether the contactor reliably engages during startup and reliably releases during shutdown. X0 is the contactor operating condition, Y0 is the control coil output, X1 is the normally open auxiliary contact of the contactor returned to the PLC input, and the timer's timing duration is greater than the contactor operating time. R0 is the set fault position; R0 ON indicates a fault and triggers an alarm; R0 OFF indicates no fault. The fault has a memory function and is cleared by the fault reset button. When opening or closing the electric valve, a delay time is set according to the different valve opening and closing times. After the delay, the open or closed signal is detected. If these signals cannot be returned to the PLC accurately and on time, it indicates that the valve may be faulty, and a valve fault alarm is triggered. The program design is shown in Figure 3b. X2 represents the valve opening condition, Y1 represents the control valve action output, the timer duration is greater than the valve opening time, X3 is the valve opening return signal, and R1 represents the valve fault position. 6. Conclusion We adopted the above methods in the design of the automatic control system for the Shengtuo polymer injection station in the Shengli Oilfield's Shengli Oil Production Plant. Nearly two years of operation have proven that these methods are effective in improving the system's reliable operation.