JPH0486461A - Controller for absorption type freezer - Google Patents

Abstract

PURPOSE:To keep a cold water outlet temperature stably near its set value by a method wherein fuzzy rules are set in detail in a control device for controlling operation with a fuzzy inference in response to external conditions and at the same time membership functions at a conclusion part near zero value of a heating amount are set in detail. CONSTITUTION:A control device 26 of a fuel control valve 17 keeps information Q of a degree of opening of a valve 17 and then a degree of opening of a fuel control valve 17 is varied in reference to an operating amount from a fuzzy inference processor 27. A memory device 28 for control rules stores fuzzy rules run at the fuzzy inference processor 27, a condition part and a conclusion part membership function. A calculation device 30 takes temperature data of each of the parts, calculates a temperature variation rate or the like and sends them to a micro-computer 25. A degree of adaptability of the condition part of all control rules is checked and then the membership function at the corresponding conclusion part is corrected. A valve control device 26 calculates new information of a degree of opening so as to adjust the fuel control valve 17.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Application The present invention relates to a control device for an absorption refrigerating machine that controls the absorption refrigerating machine by fuzzy control.

(B) Prior Art Conventionally, various devices have been considered for performing fuzzy control using physical quantities representing various external conditions and internal conditions. for example,
In the case of absorption type water chillers and absorption chillers, in addition to multiple external conditions such as the deviation and rate of change of the chilled water outlet temperature from the set value and the deviation and rate of change of the chilled water inlet temperature, the temperature of the high temperature regenerator is Fuzzy control is performed by simultaneously taking in multiple internal conditions as input variables, such as the rate of change and the rate of change of the circulation pump drive frequency.

In such devices, fuzzy rules (hereinafter referred to as rules)
When creating a so-called production rule (I
Some input variables (membership functions) are taken into fuzzy inference alone, but in many cases, they are logically ANDed with other input variables, and the fuzzy inference is Make inferences. For example, if the input variable is A
, B, C1 When the output variable is D, Rule 1: IF A 1sPBAND B 1sZR
It is written as ANDCis NB THEN Dis is ZR (where PB indicates a positive large value, ZR indicates a fuzzy label of zero, and NB indicates a negative large value). In the case of an absorption chiller, the deviation of the actual chilled water outlet temperature from the set value is often used as an input variable to control the fuel supply to the evaporator.

(c) Problems to be Solved by the Invention In such devices, there is a problem in that control accuracy deteriorates when the cold water outlet temperature approaches a set value.

(d) Means for Solving the Problems The present invention has been developed in view of the above points, and consists of connecting an evaporator, an absorber, a generator, a condenser, etc. to form a refrigeration cycle. In a control device for an absorption chiller that uses fuzzy reasoning to control the amount of heating according to external conditions including the chilled water outlet temperature, fuzzy rules are set in detail in areas where the chilled water outlet temperature is close to the set value. At the same time, the membership function of the conclusion part in the vicinity of zero heating amount of the generator is set in detail.

(E) Function The device of the present invention provides fine control when the cold water outlet temperature is close to the set value.

(f) Example Figure 1 shows water as the refrigerant and lithium bromide as the absorbent (solution).
(LiBr) A double-effect absorption refrigerator using an aqueous solution, 1 is a high-temperature generator equipped with burner IB, 2 is a low-temperature generator, 3 is a condenser, 4 is an evaporator, 5 is an absorber, and 6 is a Absorption liquid pump, 7 and 8 are a low temperature heat exchanger and a high temperature heat exchanger, respectively, 10 is a box plate liquid collection pipe, 11 is an intermediate absorption liquid pipe, 12
is a concentrated liquid pipe, 13 is a refrigerant pipe, 14 is a refrigerant liquid down pipe,
15 is a refrigerant liquid circulation pipe, each of which is connected as shown in FIG. A refrigerant pump 15P is provided in the middle of the refrigerant liquid circulation pipe 15. Further, 16 is a fuel supply pipe connected to the burner IB, and a fuel control valve (heat amount control valve) 17 is provided in the middle of this fuel supply pipe 16.

Further, 20 is a cold water pipe, and an evaporator heat exchanger 21 is provided in the middle of this cold water pipe 20.

Furthermore, 22 is a cooling water pipe.

Reference numeral 23 denotes a control unit, and 24 denotes a cold water outlet temperature detector provided in the cold water pipe 20. This cold water temperature detector 24 and the fuel control valve 17 are connected to the control panel 23. The control panel 23 is provided with a microprocessor 25 and a control device 26 for the fuel control valve 17. The microprocessor 25 is composed of a fuzzy inference processor 27 and a control rule storage device 28. The fuzzy inference processor 27 performs a logical operation on the manipulated variable KQ to the fuel control valve 17 and outputs the obtained manipulated variable KQ to the control device 26 . The control device 26 corrects the opening degree of the fuel control valve 17 based on the manipulated variable KQ. Specifically, this control device 26
holds the opening degree information Q of the valve 23, and this opening degree information Q
The opening degree of the fuel control valve 17 is adjusted accordingly. Then, each time the manipulated variable KQ is received, new opening information Q is created using the previously set opening information Q, -, and the manipulated variable KQゎ.
. =Q-++KQ- is set. That is, in this embodiment, the opening degree of the fuel control valve 17 is changed by the manipulated variable KQ from the fuzzy inference processor 27. Further, the control rule storage device 28 stores control rules (fuzzy rules) necessary for fuzzy logic operations executed by the fuzzy inference processor 27.
, the condition part and the conclusion part membership functions are stored. Further, 30 is a calculation device, and 31 is a cold water inlet temperature detector provided in the cold water pipe 20 on the inlet side of the evaporator 4. 32 is a high-temperature regenerator temperature detector that detects the high-temperature regenerator temperature; 33
is a cooling water inlet temperature detector that detects the cooling water inlet temperature. The arithmetic unit 30 takes in the temperature data of the cold water outlet temperature detector 24, the cold water inlet temperature detector 31, the high temperature regenerator temperature detector 32, and the cooling water inlet temperature detector 33, and calculates the following data.

■ Deviation of chilled water outlet temperature (e t o) eto = current value - target value ■ Rate of change of deviation of chilled water outlet temperature (dtO) dto = current value - previous value ■ Rate of change of chilled water inlet temperature (dtci) dtci = Current value - Previous value ■ Rate of change in cold water inlet temperature (dti) dti = Current value - Previous value ■ Average of the past 40 samples of eto 1i (e) e7 = (
Σ eto, )/40 ■ High temperature regenerator temperature change rate (d t g) dtg = current value - previous value Next, the functional block diagram of the microprocessor 25 is shown in the second diagram.
As shown in the figure. In the figure, 34 indicates data eto, dto, d, t ci, dt i from the calculation unit 730.
, e, dtg, and calculates the fitness of each control rule from the condition part membership function and control rule stored in the control rule storage device 28. When the definition is made by a function, the minimum degree of fitness is taken as the degree of fitness. here,
As conditional membership functions, eto, dto, c
NB for lti, dtci, dtg, and e, respectively.
It is defined as shown in FIGS. 3 to 8 using NS, ZR, PS, and PB. As you can see, dti, dtc
In order to reduce the degree of influence of i and e, weighting of 0.4.0.5.0°5 is applied to each.

The control rules for eto, dto, and dtg are as shown in FIG. That is, here, rules are not defined for adjacent control rules, thereby reducing the time required to calculate the degree of suitability.

Furthermore, eta and dto are defined as shown in FIG. 1O. In this case, NZ and PZ, which will be described later, are added to the conclusion part membership function to better simulate the degree of convergence of the control amount KQ when eto approaches the set value. dtci is defined in FIG. 11, dtl is defined in FIG. 12, and e is defined in FIG. 13. Regarding this e, control performance and convergence are improved by taking the average of the past 40 samples of eto, as shown in (2) above, only when eto is near ZR.

Reference numeral 35 denotes a modification unit that modifies the conclusion part membership function in the control rule storage device 28 so as to cut the upper part of the conclusion part membership function in accordance with the degree of fitness obtained by the degree of fitness calculation unit 34. Note that the conclusion section membership function shown in FIG. 14 is defined. As can be seen from this figure, NZ and PZ are defined in the vicinity of ZR to improve control.

Reference numeral 36 denotes a disjunctive unit that superimposes the membership functions modified by the correcting unit 35 and calculates a logical sum; and 37 a centroid calculation unit that computes the centroid of the function generated by the disjunctive unit 36. , this calculated value is given to the control device 26 as the valve operation amount KQ.

In such a device, during operation of the absorption chiller, the arithmetic unit 30 detects the temperature from the chilled water outlet temperature detector 24, the chilled water inlet temperature detector 31, the high temperature regenerator temperature detector 32, and the chilled water inlet temperature detector 33. The signal is taken in, for example, every 5 seconds. From the temperature signal thus obtained, the deviation of the chilled water outlet temperature (eto), the rate of change of the deviation of the chilled water outlet temperature (dto), and the rate of change of the chilled water inlet temperature (dtci) are calculated.
, rate of change of chilled water inlet temperature (dti), past 40 of eto
Sample average value (e), high temperature regenerator temperature change rate (d
tg) and sends it to the microcomputer 25.

The suitability calculation section 34 in the microcomputer 25 examines the suitability of the condition parts of all control rules. Then, using this goodness of fit, the modification unit 35 modifies the membership function of the corresponding conclusion part shown in FIG. That is, the portion above the fitness level of each membership function is cut. A logical sum of the membership functions modified in this way is calculated by a logical sum section 36, and a center of gravity of the membership function is determined by a center of gravity calculation section 37. This center of gravity calculation section 3
The output of 7 is the manipulated variable KQ of the fuel control valve 17.

is output as

The valve control device 26 generates new opening information Q, =Q-- based on this manipulated variable KQ and the previous opening information Q*-1.
Calculate ++KQ-. And this opening information Q. Adjust the fuel control valve 17 accordingly.

These operations are repeated at the above-mentioned 5 second period.

(g) Effects of the Invention As described above, the absorption refrigerator control device of the present invention connects an evaporator, absorber, generator, condenser, etc. to form a refrigeration cycle, and the heating amount of the generator is In a control device for an absorption chiller that uses fuzzy reasoning to control external conditions including the chilled water outlet temperature, fuzzy rules are finely set and the error occurs when the chilled water outlet temperature is close to the set value. Since the membership function of the conclusion part near the zero heating amount of the container is set in detail, even if the cold water outlet temperature is near the set value, it will not oscillate and will remain stable.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an absorption chiller to which the control device of the present invention is applied, Fig. 2 is a functional block diagram of a microcomputer used in the device of the present invention, and Figs. Characteristic diagram of the conditional membership function,
9 to 13 are explanatory diagrams of the control rules, and FIG. 14 is a characteristic diagram of the membership function of the conclusion part. DESCRIPTION OF SYMBOLS 1... High temperature generator, 2... Low temperature generator, 3... Condenser, 4... Evaporator, 5... Absorber, 17... Fuel control valve, 23... Control Part, 24...Cold water outlet temperature detector, 25...Microprocessor, 26...Control device, 27...Fuzzy inference processor, 28...
- Control rule storage device, 30... arithmetic device 31...
- Chilled water inlet temperature detector, 32... High temperature regenerator temperature detector, 33... Chilled water inlet temperature detector 34... Compatibility calculation section, 35... Correction section, 36... Logical sum section, 37...
- Center of gravity calculation section.

Claims (1)

[Claims] (1) Absorption refrigeration, in which an evaporator, absorber, generator, condenser, etc. are connected to form a refrigeration cycle, and the amount of heat generated by the generator is controlled using fuzzy reasoning according to external conditions including the chilled water outlet temperature. In the control system of the machine, the fuzzy rules are set in detail in the part where the cold water outlet temperature is close to the set value, and the membership function in the conclusion part in the vicinity of zero heating amount of the generator is set in detail. A refrigerator control device characterized by: