将基于知识的规则模型和解析的数学模型有机结合，建立两级结构的实时专家控制系统（ECS）。由具有专家自学习功能的实时专家系统（ES）以专家优化方式寻找电解液最优酸锌含量比和最优温度，由实时控制系统（RTCS）控制新液流量、总地槽液位和冷却风机状态，保证生产过程在最优生产条件下进行。本文介绍了系统结构、专家优化技术、实时控制技术和系统实现。

Stable pressure of gas collectors is beneficial to prolonging coke oven's life-time, consuming less energy and decreasing pollution. However, it is difficult to stabilize the pressure with conventional control methods because the collectors' pressure system is a timevarying, nonlinear multi-variable system that is strongly inter-coupled and disturbed. In this paper, a hybrid control approach that incorporates PID control withfeedforward control and expert control is presented. It integrates simplicity, reliability, flexibility and promptness in restraining disturbance by making full use of each method's advantages. The hybrid intelligent control system has been built to control the pressure of gas collectors. The results of actual runs show the effectiveness.

With the increasing complexity of industrial application, an embedded control system (ECS) requires processing a number of hard real-time tasks and needs fault-tolerance to assure high reliability. Considering the characteristics of real-time tasks in ECS, an integrated algorithm is proposed to schedule real-time tasks and to guarantee that all real-time tasks are completed before their deadlines even in the presence of faults. Based on the nonpreemptive critical-section protocol (NCSP), this paper analyzes the blocking time introduced by resource con icts of relevancy tasks in fault-tolerant multiprocessor systems. An extended schedulability condition is presented to check the assignment feasibility of a given task to a processor. A primary/backup approach and on-line replacement of failed processors are used to tolerate processor failures. The analysis reveals that the integrated algorithm bounds the blocking time, requires limited overhead on the number of processors, and still assures good processor utilization. This is also demonstrated by simulation results. Both analysis and simulation show the e ectiveness of the proposed algorithm in ECS.

This paper describes an optimization prolem to minimize the cost of power consumption for the electrochemical porocess of zinc (EPZ) depending onvarying prices of electrical power. Ae series of conditional experiments was conducted to obtain enough data, whic reflect the complex relationships among the factors influecing power consumption. Two backpropagation neural networks are used to build a process model that describes these relationships. An equivalent Hopfield neural network is cnstructed to solve this nonlinear optimization poroblem with tcchnological constraints, a penalty function is introduced into the network energy function to meet the equality constraints, and inequality constraints are removed by altering the sigmoid function. An optimal power-dispatching control system (OPDCS) has been developed to provide an optimal power-dispatching scheme and keep the ePZ running economically. Since the OPDCS was put into service in a smeltery, the cost of power consumption has decreased significantly, and cost of power consumption has decreased significantly, and it also contrbutes to halancing the power grid load.

Depending on varying prices of electricity, an optimal power-dispatching system (OPDS) is developed to minimize the cost of power consumption in the electrochemical process of zinc (EPZ). Due to the complexity of the EPZ, the main factors influencing the power consumption are determined by qualitative analysis, and a series of conditional experiments is conducted to acquire sufficient data, then two backpropagation neural networks are used to describe these relationships quantitatively. An equivalent Hopfield neural network is constructed to solve the optimization problem where a penalty function is introduced into the network energy function so as to meet the equality constraints, and inequality constraints are removed by alteration of the Sigmoid function. This OPDS was put into service in a smeltery in 1998. The cost of power consumption has decreased significantly, the total electrical energy consumption is reduced, and it is also beneficial to balancing the load of the power grid. The actual results show the effectiveness of the OPDS. This paper introduces a successful industrial application and mainly presents how to utilize neural networks to solve particular problems for the real world.