【正文】 (12) Use ()qru and ()qsu to denote the values of ru and su at time point qt respectively (1 2 )qQ＃ . Let ( 1 ) ( 1 ) ( 1 ) ( 1 ) ( ) ( ) ( ) ( )1 1 1 1( , .. ., , , .. ., , .. ., , .. ., , , .. . )Q Q Q Q TR S R Sx u u v v u u v v= (13) ( 1 ) ( 2 ) ( 1 ) 211( , ..., , ..., , ..., )Q Q Q Q TSSy v v v v++= (14) Where ( ) ,n R S Q m QS= + =. The task of the prediction is to determine y of (14)according to the vector x of (13) .For the current time 0t ,all of the measured results can be used to construct the inputs of the prediction work. Suppose that all of the operating variables and state variables can be got by measuring ,and their values in the future are unknown. To construct a prediction sample ,the related time t should satisfy 0t t Q t?D .Otherwise, unknown values would be contained in the sample which would be unreasonable. Suppose that enough samples ( ) ( )( , ), 1, 2 , .. .,kkx y k K= have been got .First, calculate the parameters of the hidden units, then calculate the prediction value of the storage temperature. Case2 :Automatic control system is on At this time, the set of the input variables only contains the environmental temperature, humidity and quantum of the stored fruits and vegetables ,etc. Any of the input variables doesn’t appear in the control algorithm ,while the prediction variables are the stable values of the state variables. The nonlinear mapping function of the RBF neural work is used to design the stable models. When the stable values of the state variables have been obtained, the control algorithm is used to calculate the temperature of the storehouse, thus the set of the predicted variables wouldn’t contain any variable to be controlled. That’s why the set of the predicted variables and the set of the controlled variables under Case2 are different from those under Case1. ONLINE OPTIMUM CONTROL OF THE COLD STORAGE TEMPERATURE The mon PID control algorithm of a variable unit takes the following form 0( ) { ( ) ( ) [ ( ) ( 1 ) ] }ksdcsiisTTu k K e k e i e k e k u== + + +229。 (9) Let *T denote the optimum storage temperature in general .It should satisfy 2*( , ) 0L T ttT182。 = + 182。 (3) Here, the numerator is a traditional RBF interpolating algorithm expression, and the denominator is the interpolating expression of constant this denominator, the attenuation of exponent functions in the numerator is canceled out greatly by that of the denominator. In this way ,the improved RBF neural work has a better performance. 3. THE ONLINE CALCULATION OF THE COLD STORAGE TEMPERATURE To choose the target value of the cold storage temperature, it is needed to take overall considerations about all factors. In order to use energy reasonably, the refrigeration process should have a high performance coefficient 0e which is the ratio of the refrigeration quantum 0Q to the needed energy P satisfying 00 Qe P= (4) Research results show that 0e increases as the evaporation temperature increases or the condensation temperature decreases [4,6],and a higher evaporation temperature and a lower condensation temperature are beneficial to keep fruits and vegetable fresh . Thus the refrigeration system should run under a higher evaporation temperature and a lower condensation temperature. However the evaporation temperature is apparently limited by the temperature of the object under refrigeration. For a special kind of fruit or vegetable just entering the cold storage, its optimum storage temperature can be got with the orthogonal experimental method. The optimum storage temperature decreases with the increasing of the storage time. The loss of per unit of i th fruit or vegetable is (1) ( 2 )i i iL L L=+ (5) where (1)iL is produced by frostbiting, while (2)iL by deteriorating .When temperature increases , (1)iL decreases and (2)iL increases .Both of them are related to the storage time t , thus ( 1 ) ( 2 )( 1 ) 0 ( 2 ) 0[ , ] , [ , ]iii i i iLLf T t t f T t ttt抖 = = 抖 (6) where (1)if decreases and (2)if increases respectively when the temperature T increases , 0it denotes the time of entering the storage, while 0itt denotes the storage time, then we have 0( 1 ) 0 ( 2 ) 0{ [ , ] [ , ] }iti i i i itL f T t t f T t t d t= + 242。 Xue Guoxin Jiangsu Institution of Petrochemical Technology, Changzhou 213016, (Received November 26, 1999) Abstract :In recent years ,advanced control technologies have been for the optimum control of a cold storage. But there are still a lot of shortings. One of the main problems is that the traditional methods can’t realize the online predictive optimum control of a refrigerating system with simple and valid algorithms. An RBF neural work has a strong ability in nonlinear mapping, a good interpolating value performance, and a higher training speed. Thus a twostage RBF neural work is proposed in this paper .Combining the measured values with the predicted values , the twostage RBF neural work is used for the online predictive optimum control of the cold storage temperature. The application results of the new methods show a great success. Keywords: RBF neural work, Cold storage, Online prediction, optimum control. 1. INTRODUCTION The predictive optimum control of cold storage temperature has found a wide application in a agricultural engineering, especially for keeping fruits and vegetables fresh by cold storage. All of the currentlyused temperature control units face the problems on how to choose the optimum temperature as the controlled object, how to predict the temperature variation of the refrigerating storehouse and how to realize the optimum control. A lot of study efforts have been made. The earlier methods wer