【正文】 here some electrical resistances are located. To exactly obtain the same temperaturefixed for the air, a regulator is used to control the electricalresistances supply. In some experimental tests, the coolingload in the cold store is simulated by means of some electricheaters linked to a regulator and the electric power ismeasured by means of a Wattmeter. Table 1 lists thespecifications of the transducers used. The test apparatus isequipped with 32 bit A/D acquisition cards linked to apersonal puter allowing a high sampling rate and amonitoring of all the measures carriedout by means of thetransducers. The data acquisition software has been realizedin a Labview environment and the R407C and R507thermodynamic properties have been evaluated using adedicated software that has also been used to determine theenergy and exergy balances.3. Experimental procedure descriptionTo evaluate the plant performances when an inverteris used, it is necessary to pare the plant energyconsumption when the refrigeration capacity is regulatedby on/off cycles of the pressor that works at a supplycurrent frequency of 50 Hz, and when the refrigerationcapacity is controlled by the fuzzy algorithm. In theexperimental tests, different types of cooling loads havebeen considered. First of all some experimental tests whenthe cooling load is due both to the periodic opening of thecold store door and to the inevitable heat exchanges withoutdoor air, even when the cold store door is closed, havebeen realized. These tests have been performed at varioustemperature levels for the air in the coldstore and, precisely,at 5, 0 and 25 8C, opening the cold store door every 20 minfor about 5 min with an outdoor air temperature of about18 8C. Moreover, in some tests the cooling load has beenobtained by means of controllable electrical heaters locatedin the cold store, while in other tests a real cooling load hasbeen considered represented by 200 kg of fruit andvegetables for whose preservation the temperature hasbeen fixed at 5 8C in the cold store. In these last twoFig. 1. Sketch of the experimental plant.C. Aprea et al. / International Journal of Refrigeration 27 (2020) 639–648 641situations, the cold store door has been opened every 10 minto simulate a real working condition。d240。f222。 1–30 bar ^。 this happens when the cold store dooris open. To increase the sensitivity of the controller withrespect to the rate of the change of the derivative with timeof the temperature difference, the fuzzy subset might have asmaller definition, perhaps ranging from to .However, satisfactory results can be obtained with theprevious definition of the fuzzy subsets. The values ofthe pressor motor supply frequency considered in theoutput fuzzy subset membership function (Fig. 5), arelocated in the range 30–50 Hz. It was not possible toconsider values under 30 Hz because the pressorvibrations and the noise increase considerably togetherwith the lubrication troubles due to the splash systemincrease. The inference mechanism employed has been theproduct inferencing method whereby the minimum operatorfor the ‘a(chǎn)nd’ is replaced by the product operation [29–31].This mechanism allows a better interpolative reasoningamong the input and output variables because the effect ofthese variables on each other is obtained more effectively.The adopted defuzzification method is based on thedetermination of the mass center of a pound set。 this last value is acceptable as itcorresponds to the differential band of a thermostat.To explain the reason for the energy saving obtainablewith a fuzzy control algorithm that allows to regulatecontinuously the pressor velocity in parison with athermostatic control, an exergetic analysis of the ponents of the refrigeration plant has been realized onvarying the pressor speed. For this purpose, it resultsmore correct to realize the exergetic analysis in the steadystate conditions instead of the transient conditions, whichderive from the pressor speed fuzzy control, selectingagain the same conditions in terms of pressor refrigeration capacity. In particular, the refrigeration powercorresponding to each frequency of 30, 35, 40, 45, 50 Hzunder the transient conditions is considered again under thesteadystate conditions to allow a correct measurementprocess. This experimental analysis has been performedrelated to the summer season, with the outdoor temperatureat the condenser kept at about 32 8C, but similar results havebeen obtained also in other working conditions. Inparticular, the tests realized in the winter season verify thecorrect refrigerant lamination when, varying the pressorspeed, the pression ratio across the valve is low.The exergetic analysis allows to obtain importantinformation about the plant total irreversibility distributionamong the ponents. The overall plant exergeticefficiency has been evaluated as the ratio between theexergy output and the exergy input and can be expressed as:hex188。An accurate analysis can be realized evaluating the exergydestroyed for each single ponent of the plant. Inparticular, the exergy flow destroyed in the condenser isevaluated as:_Exdes。co2 exout。and the exergy flow destroyed in the evaporator is evaluatedas:_Exdes。ev2 exout。where the dimensionless exergetic temperature can bedefined as:t 188。where T0is the environmental temperature while Tmt。exin。_Lcp240。 _mref240。240。i_Lcp240。8222。 as the COPrevisfixed, referring to the temperatures shown in Fig. 9, theexergetic efficiency follows the trend of the COP. It is to benoted that when the pressor speed decreases the COPincreases. In particular, it is necessary to observe that, whenthe pressor speed decreases, the global heat transfercoefficient of the heat exchangers is practically constantbecause the