Application of Variable Frequency Speed Control Technology in Automated Gas Relay Booster Stations and Related Control Issues

1. Introduction In the steel industry, the gas produced by blast furnaces, coke ovens, and converters is an important high-calorific-value fuel. If emitted into the atmosphere, it will cause serious air pollution. Collecting and fully utilizing it as fuel for various industrial gas-fired heating furnaces offers considerable economic benefits. The general process involves collecting the various types of gas into separate storage tanks (or storage towers) and then pipelines to relay stations for pressurization before supplying them to users. When converting from primary energy (such as coal) to secondary energy (gas), the unstable quality of the primary energy inevitably leads to instability in the quality of the secondary energy (gas), meaning the gas quality has significant uncertainty. Users require that the quality of the gas be controlled within permissible limits and that the gas pressure be strictly stabilized within specified ranges for effective use. Taking the gas-fired rolling mill heating furnace in steel enterprises as an example, it requires stable gas quality and pressure; otherwise, the combustion system is difficult to control, directly affecting the product quality of subsequent automated production lines such as rolling mill production. Currently, a significant portion of gas booster stations utilize electrical instrumentation control systems, while some have been modified to collect data from the field and use PLCs as the main controller for equipment control. However, due to the relatively crude control methods, the results are unsatisfactory. Due to equipment limitations, automatic monitoring, automatic gas distribution, and automatic scheduling are not yet possible; they are essentially in a state of instrument monitoring and manual operation. Optimizing gas distribution and scheduling is out of the question. The core problem is that the control section suffers from numerous issues after adopting frequency converters. This article discusses these issues. [b]2 Object Characteristic Analysis and Control Strategy Selection[/b] 2.1 Object Characteristic Analysis Since mixed gas originates from high-temperature coke oven gas, coke oven gas, and converter gas, as mentioned above, the uncertainty and unpredictability of the gas quality are its main characteristics. Therefore, it is impossible to directly establish a mathematical model of the object using traditional mechanistic methods, nor is it possible to indirectly derive a mathematical model of control using system identification methods. Conventional control strategies are clearly ineffective for this semi-structured (or unstructured) control problem. The problems with gas booster stations appear superficially to be changes in pressure and flow, but in reality, they involve changes in pipeline characteristics. These changes require that speed adjustments to the booster units must adapt to these changes. Uncertainty and uncertainty manifest in several ways, such as the unknown, time-varying, random, and dispersed nature of system parameters; the unknown and time-varying nature of system time delays; severe nonlinearity of the system; the correlation between system variables; and the unknown nature of environmental disturbances. Traditional control methods are ineffective for objects with these characteristics, necessitating the search for effective control strategies. 2.2 Control Strategy Selection Due to the unique characteristics of steel enterprises, the geographical dispersion of various branches or workshops is significant, with considerable differences in resource allocation and a highly complex situation. A general discussion is impractical. This paper only discusses a specific situation: the collected various types of gas are sent to relay booster stations for pressurization, and then distributed and dispatched as needed. Due to the existence of uncertainty and unknown, the characteristics of the object are difficult to be mathematically described by rigorous quantitative methods. Therefore, from the perspective of control, the main target of interest should not be the controlled object, but the controller itself. In engineering, the method of identifying the dynamic characteristics of the system based on the rich experience and operating rules of the operator, making intuitive reasoning, and then determining the control strategy online is often adopted, which is the intelligent control strategy. Its characteristic is that the establishment of control rules and the control decision-making process are not based on the simple mathematical analytical model of the controlled object, but on knowledge (including quantitative and qualitative knowledge), which reflects the intelligence of humans (experts, skilled operators). Its control forms are diverse, such as expert control, humanoid intelligent control, fuzzy control, neural network control, self-learning control, etc. Among them, expert control and humanoid intelligent control are closer to reality. (1) Expert control strategy Real-time expert control system is essentially a computer intelligent software system. It simulates domain experts to process knowledge and solve problems. It is a system that can obtain feedback information and control online in real time. The processing of real-time data and feature identification are online, which can reflect the dynamic characteristics of the system in a timely manner. Its reasoning method generally adopts data-driven forward reasoning, and judges the conditions of each rule in turn. If the conditions are met, the rule is executed. The execution rule must give a control decision. The decision can be a combination of qualitative and quantitative methods. The real-time expert control system is mainly composed of a knowledge base, a database and an inference engine. Its knowledge base is divided into a static knowledge base and a dynamic knowledge base. The static knowledge base stores parameters such as system and operating condition setpoints. The dynamic knowledge base stores various expert rule sets. The rules can be divided into multiple subsets according to different types. The relationship between the subsets is hierarchical (or coordinated). In order to meet the time response requirements, the number of rules in each subset should not be too many. The difficulty lies in the fact that it is not easy to establish a knowledge base for uncertain and unknown systems. Even if a complete and complex knowledge base is established, it may not meet the real-time requirements. In fact, it is not advisable for the specific system being studied. (2) The basic idea of humanoid intelligent control is to use computer simulation of human control behavior functions in the control process, to identify and utilize the characteristic information provided by the dynamic process of the control system to the maximum extent, to carry out heuristic and intuitive reasoning, so as to achieve effective control of objects that lack precise mathematical models. Its physical implementation method is to identify the state, dynamic characteristics, and behavior of the controlled system based on the input and output information of the computer-controlled dynamic system. That is, by using the system error e and its first and second derivatives, a control algorithm can be constructed. Relatively speaking, it is simple and practical, with clear physical meaning. Compared with other intelligent control strategies (such as expert control strategies, fuzzy control strategies, neural control strategies, self-learning control strategies, multi-level hierarchical intelligent control strategies, multi-mode variable structure intelligent control strategies, etc.), its advantages are obvious. Therefore, in the gas relay automated pressurization station, a human-like intelligent control strategy was adopted. [b]3 Control Algorithm[/b] The basic idea is to imitate the common behavior of experienced operators in process control systems, such as issuing a strong action (closed-loop control) when the system error tends to increase; and canceling the control action and waiting for observation when the system error tends to decrease, etc. The more humans understand the state, dynamic characteristics, and behavior of the controlled system, the better the control effect will be. If en represents the error value at the current sampling time in the discretization, and en-1 and en-2 represent the error values at the previous and second sampling times respectively, then we can obtain more characteristic information from the dynamic process by considering the two basic characteristic changes of error e and error change Δe. (1) e·Δe The product of error e and error change Δe constitutes a new characteristic variable describing the dynamic process of the system. By using whether the value of this characteristic variable is greater than zero, we can describe the trend of the dynamic process of the system. When e·Δe<0, it indicates that the dynamic process of the system is changing in the direction of decreasing error, that is, the absolute value of the error is gradually decreasing. When e·Δe>0, it indicates that the dynamic process of the system is changing in the direction of increasing error, that is, the absolute value of the error is gradually increasing. In the control process, by identifying the sign of e·Δe, we can grasp the behavioral characteristics of the dynamic process of the system, so as to better formulate the next control strategy. (2) Δen·Δen-1 The product of two consecutive error changes, Δen·Δen-1, constitutes a characteristic quantity representing the extreme state of the error. If Δen·Δen-1 < 0, it indicates the occurrence of an extreme value. If Δen·Δen-1 > 0, it indicates the absence of an extreme value. (3) The absolute value of the ratio of error change Δe to error e describes the posture of error change during the dynamic process of the system. When used in conjunction with e·Δe, the dynamic process can be further divided. Through this division, different postures of the dynamic process can be captured. (4) Δ(Δe) The rate of change of error, i.e., the second difference, describes the dynamic process when it is in the overshoot or pullback segment. When Δ(Δe) > 0, it is in the overshoot segment; when Δ(Δe) < 0, it is in the pullback segment. Summarizing the above characteristics, its basic control algorithm can be summarized as follows: k is the suppression (attenuation) coefficient of the controller gain, generally taken as o.

# Variable Frequency | Speed Control | Technology | Gas

← Application of frequency converters in the control system of PVC film laminating machine

Delta Electronics – Application of Electromechanical Products in Corrugated Cardboard Cross-Cutting Control Systems →

Application of Variable Frequency Speed Control Technology in Automated Gas Relay Booster Stations and Related Control Issues

Read next

CATDOLL Dolly Hybrid Silicone Head

CATDOLL Maruko 109CM TPE (Soft Silicone Head)

CATDOLL Mimi Soft Silicone Head

CATDOLL 115CM Cici TPE (Natural Tone) Customer Photos