Comprehensive Evaluation of Paddy Quality by Different Drying Methods, Based on Gray Relational Analysis

Abstract

If paddy is not dried in time after harvesting, it very easily becomes moldy, which causes substantial losses. Inappropriate drying methods also affect the quality and safety of paddy when it enters the drying process. In order to select the optimal drying method, paddy with different moisture content was treated with sun drying, mechanical drying, and late harvesting, and the quality indexes of paddy were tested for safe moisture content. The effects of different drying methods on the quality of paddy were analyzed in terms of burst rate, taste value, appearance quality, and pasting characteristics. A paddy quality index system was constructed, and gray relational analysis (GRA) was used to evaluate comprehensively the paddy quality. The results showed that when the moisture content of paddy was 24.4%, the best quality of paddy was obtained after mechanical drying, when the gray relational degree was 0.996. Timely harvesting and mechanical drying can not only reduce the loss of paddy, but can also ensure the quality and safety of paddy.

1. Introduction

Paddy is one of the most important food crops in the world [1], and China’s paddy production ranks first in the world, although the post-harvest loss of paddy is much higher than that in developed countries. Without timely drying, when harvested paddy is piled together a sudden rise in temperature results, and with high moisture content and no ventilation inside the piles, the paddy germinates and develops mold, causing huge losses [2].

After the paddy is harvested, it enters the drying process and can only be processed or stored after reaching a safe moisture content of about 14.5% to ensure its quality and safety [3]. Different drying methods affect the quality of paddy. As China’s agricultural production and management has progressed through the development stage from old to new, gradually turning to large-scale management [4], small farmers have mainly adopted methods of late harvesting or sun drying to bring their paddy to safe levels of moisture content, while large-scale operations have mainly adopted the mechanical drying method. Late harvest was a drying method commonly used by small-scale farmers, in which the paddy was harvested after reaching safe moisture content. The sun drying method was also popularly used by farmers, where newly harvested paddy was dried in the sun on open flat ground, and then cleaned and bagged when it reached safe moisture content. Mechanical drying has been a more popular way of drying used by large-scale enterprises, in which the newly harvested paddy is preliminarily screened and then dried by grain dryers to obtain safe moisture content.

Choosing a suitable drying method can not only reduce the post-harvest loss of paddy, but can also provide guarantees of its processing and storage qualities. Li et al. compared and analyzed three drying methods for paddy: sun drying, indoor drying in the shade, and mechanical drying, and concluded that mechanical drying was the preferred method [5]. Janaun et al. studied the effects of moisture and drying methods on amylose content in paddy [6]. Sahabuddin et al. analyzed the effect of paddy-drying methods on the performance of rice mills in Bangladesh. Weather dependency and long sun-drying time (48–96 h) of paddy were found to be responsible for the low capacity utilization of major rice mills. That study also proposed a single dryer for parboiled and aromatic paddy rather than following existing practices [7]. However, those studies analyzed the impact of different drying methods on paddy quality according to a single index, without comprehensive index evaluation, and the analyses were somewhat one-sided. Zhou et al. studied the paddy drying process based on comprehensive analysis of membership, but focused only different mechanical drying processes, ignoring other drying methods [8].

Therefore, it is necessary to comprehensively analyze paddy quality obtained from different drying methods and select the optimal drying method. In the current study, different drying methods were used to dry paddy with different moisture content, and the effects of different drying methods on paddy quality were compared using a single factor. GRA was employed to evaluate comprehensively the paddy quality obtained from different drying methods. Paddy samples obtained from different drying methods were tested for quality and analyzed in comparison with optimal samples, and the gray relational degree of each sample was derived by GRA. Finally, the gray relational value was used as a comprehensive evaluation result to select the optimal drying method to guarantee the quality and reduce the loss of paddy.

2. Materials and Methods

2.1. Materials and Main Instrumentation

The test samples comprised newly harvested paddy of the variety Wu Youdao4 (WYD4), originating from Gudianzi Town, Changyi District, Jilin City, Jilin Province, China.

The moisture content of paddy was detected using a DHS-20A moisture meter produced by Shanghai Jingqi Company, Shanghai, China. The readability of the moisture content measurement was 0.02~0.1%, and the moisture content measurement range was 0–100%.

The taste value of paddy was detected using a SATAKE STA1B taste meter produced by Sasaki Company, Ōsaka, Japan, and the measurement time was once per minute.

The appearance quality of paddy was evaluated with a JMWT12 appearance quality tester produced by Dongfu Jiuheng, Beijing, China. The testing time was less than 40 s, and the recommended sample quantity for a single test was 12 g.

The pasting characteristics of paddy were detected using an RVA Starchmaster2 fast viscosity analyzer produced by Broadcom Company, Stockholm, Sweden. The rise and fall rate was 14 °C/min, the rotational speed range was 20~1000 rpm, and the viscosity range was 40~12,000 cp at 80 rpm.

2.2. Dring Methods

2.2.1. Sun Drying

Paddy with standard planting and good growth conditions was selected, and the harvested paddy was threshed for sun drying. The paddy was dried on a concrete floor in an open space according to the drying method used by local farmers. The weight of the threshed paddy was 500 kg, and occupied 30.33 m² area with a thickness of about 0.04 m. The moisture content of the paddy was recorded once a day at 8:00, 12:00, and 16:00. The paddy was turned once a day at 9:00 and 14:00 according to technique used by farmers, and it was covered by thatched cloth in case of rain. Three batches of harvested paddy dried by the sun-drying method had different initial levels of moisture content: 32.6%, 24.4%, and 18.7%, respectively. The tested variety was japonica paddy, the safe moisture level of which is specified in the standard as 14.5% [3], so the quality of paddy was tested when the moisture content reached approximately 14.5% after drying.

2.2.2. Mechanical Drying

The paddy drying machinery selected was the Satake 5HNSDR-30 cross-flow circulating drying tower used by the enterprise operating at the test site, and the heat source was coal. Samples comprised WYD4 in the normal state of use during the actual drying process used by the enterprise. The hot air temperature of the dryer was measured at 45 °C, putting it in the category of low-temperature cycle drying. Sampling was conducted at appropriate intervals to measure the moisture content. The initial levels of moisture content in the three batches of paddy undergoing mechanical drying were 32.6%, 24.4%, and 18.7% respectively, which were the same as those receiving sun-drying treatment. The quality of the paddy was tested when the moisture content reached 14.5%.

2.2.3. Late Harvest

To ensure the accuracy of the test, paddy after sun drying or mechanical drying on a piece of land was used as the test samples. Late harvesting was carried out in the paddy field, using paddy selected from different locations. The sample points for the test were required to be representative, and were selected from within the field according to the plum sampling method from 5 sample points [9], as shown in Figure 1. Sampling was carried out at 8:30 every day during the test period; the samples were threshed and processed in the morning, and then tested for moisture content. In this study, three different initial moisture levels of harvested paddy were used for testing. The late harvest method required that the paddy was not harvested until it had been reduced to safe moisture content, so the moisture was tested daily from high moisture until it fell to about 14.5%, at which point it was harvested and threshed, and testing of paddy quality was conducted. The moisture of the paddy at the beginning of the test was 32.6%. The final quality of the paddy was tested for three different levels of initial moisture from the same batch of paddy.

The experiment was divided into three batches, the first batch was from the period of high moisture in the paddy, at an early harvesting point according to the actual local situation; the second batch was from the period of medium moisture in the paddy, representing timely harvesting of the paddy; the third batch was from the period of low moisture in the paddy, at a later harvesting time in the local area. The initial moisture content in the paddy was 32.6%, 24.4%, and 18.7%, respectively. Paddy samples with different initial moisture levels and having undergone different drying methods were labelled A–G respectively. The different drying methods and different moisture contents are indicated in

Table 1. Sample processing with different moisture content using different drying methods.

No.	Initial Moisture	Drying Method	Drying Time	Final Moisture
A	32.6%	sun drying	164 h 30 min	14.9%
B	32.6%	mechanical drying	28 h 30 min	14.8%
C	32.6%	late harvest	168 h	14.7%
D	24.4%	sun drying	27 h 30 min	14.8%
E	24.4%	mechanical drying	10 h	14.5%
F	18.7%	sun drying	25 h 30 min	14.7%
G	18.7%	mechanical drying	8 h	14.9%

.

2.3. Quality Testing

Moisture content was measured by halogen moisture meter; burst rate was measured by national standard GB/T 5496-1985 [10]; taste value was measured using a paddy huller to hull the paddy, then a paddy mill was utilized to make head rice, which was measured with a taste meter; pasting characteristics were measured using a fast viscosity analyzer after processing to rice and grinding the rice to powder. Appearance quality was determined using an appearance quality tester after the paddy was hulled and made into rice.

2.4. Comprehensive Evaluation

Assessment of paddy quality involves a complex comprehensive system, and each single index has different effects on paddy quality, which cannot be directly and accurately evaluated using single index. The gray relational analysis method was applied to calculate the relation degree between each sample and the optimal sample to obtain the evaluation results of paddy quality after different drying methods.

2.4.1. Index Normalization

In order to eliminate the influence of varying gauges of paddy quality indexes after different drying methods, and to ensure uniformity and comparability of evaluation, the indexes were normalized. According to the nature of the quality evaluation indexes, they were divided into positive and negative indexes [11].

A positive index means that the higher the value of x_i, the better is the quality of the paddy, and the normalized treatment formula is:

$$y_i=\frac{x_i-x_{min}}{x_{max}-x_{min}}$$

(1)

where x_i is the actual value of the index, x_max and x_min are the theoretical maximum and minimum values of the index x_i, respectively, which can be determined before the calculation.

A negative index means that the higher the value of x_i, the worse is the quality of the paddy, and the normalized treatment formula is:

$$y_i=\frac{x_{max}-x_i}{x_{max}-x_{min}}$$

(2)

The meanings of the items in the formula are the same as in the previous equation.

2.4.2. Gray Relational Analysis Method

For the analysis of factors between systems, the measurement of the magnitude of the relation that changes with time or with different variables is called the relation degree. If the change trend of the two factors is consistent, the degree of synchronous change of the two factors is high, and vice versa. According to the degree of similarity or difference of the development trends between factors, the gray relational analysis (GRA) method [12,13] can be used for measuring the relation between those factors.

The steps for using gray relational analysis were as follows [14,15]:

Paddy quality data for different drying methods were taken as the comparison data column X_ik, where i represents the ith sample and k represents the kth index value. The sample with the best index value in the original sample data was selected to form the reference data column X_0k, which provides an ideal comparison standard for the gray relational analysis.

(1) The evaluation samples and evaluation criteria were determined, setting the number of evaluation samples as m and the number of evaluation indicators as n. The comparison sequence is:

$$X_i=\left\{x_{ik}|i=1,2,\dots ,m;k=1,2,\dots ,n\right\}$$

(3)

The best value of each index was taken as X₀, and x_0k was the entity of the reference number sequence X_0, Then, the reference number sequence is:

$$X_0=\left\{x_{0k}|k=1,2,\dots ,n\right\}$$

(4)

where x_0k = optimum(x_ik), i = 1, 2, …, n.

(2) The relational coefficient was calculated:

$${\zeta }_{ik}=\frac{mi{n}_{i}mi{n}_k+ama{x}_{i}ma{x}_k\left|X_{0k}-X_{ik}\right|}{\left|X_{0k}-X_{ik}\right|+ama{x}_{i}ma{x}_k\left|X_{0k}-X_{ik}\right|}$$

(5)

where ${\zeta }_{ik}$ is the relative difference between the comparison sequence X_i and the reference sequence X₀ on the k_th evaluation index, which is called the relational coefficient of X_i to X₀ on the k index; $a$ is the resolution coefficient, 0 ≤ $a$ ≤ 1, usually 0.5.

(3) The weighted relational degree was calculated:

$$r_i=\frac{1}{n}{\displaystyle \sum }_{k=1}^n\omega {\zeta }_{i\left(k\right)}$$

(6)

where r_i is weighted relational degree, n is the number of indicators, ω is the weighting coefficient.

3. Results

3.1. Moisture Content Change

After treatment according to the test method, the moisture content of the samples was tested. The moisture content change during different drying methods of paddy with different initial moisture content is shown in Figure 2A–G.

Figure 2A–C shows the moisture content changes for three drying methods, for paddy with initial moisture content of 32.6%. Among the three drying methods, the drying rate during mechanical drying decreased more slowly (Figure 2B) and did not fluctuate with the environment, while sun drying (Figure 2A) and late harvest (Figure 2C) were affected by changes of weather and the environment, with sun drying less affected by changes of weather. Late harvest was more affected by changes in the weather. Low temperature phenomena were particularly serious in the test area during the test period, and it rained frequently in early October. The moisture content of the paddy increased after the rain, and the moisture dropped rapidly after wind and sun.

The moisture change processes in mechanically dried and sun-dried paddy with initial moisture content of 24.4% are shown in Figure 2D,E. The moisture content of sun-dried paddy was greatly affected by weather and the environment, while the moisture content of mechanical dried paddy was less affected by environmental factors.

Figure 2F,G shows the moisture decline of mechanical dried and sun-dried paddy with initial moisture content of 18.7%. The moisture decline in paddy in the late stage of sun drying showed large fluctuations due to the environmental influence of weather. The drying rate during mechanical drying was relatively smooth, and the process of moisture decline was more stable and less influenced by the environment.

3.2. Paddy Quality

3.2.1. Burst Rate

After the paddy was dried under the different drying methods, the quality of the paddy was tested. The burst rate of the paddy was tested first, and the results for the detected burst rates are shown in Figure 3.

From Figure 3, it can be seen that the burst rate of 11% for the sun-dried paddy with 32.6% initial moisture content (A) was the highest. The testing was carried out in accordance with the habits of farmers exposing the paddy to the sun, and drying it on the road, where road dust, vehicle waste, and other harmful substances pollute the paddy. During the sun-drying process, paddy absorbed moisture when rainy weather was encountered, followed by exposure to the sun during sunny weather, so the burst rate of the paddy increased more quickly. The burst rate of high-moisture paddy undergoing the mechanical drying process continued to increase, and the burst rate after drying reached 5% (B). This may be because when the moisture of the paddy was too high, and the mechanical drying process involving hot air stopped, the paddy came into contact with colder air causing shock which led to increased burst rate. The burst rate of high-moisture paddy after late harvest was the lowest at 3% (C), indicating that although the moisture of late-harvested paddy was affected by the weather, its burst rate was less affected by the environment because it had not been harvested earlier so the paddy remained in a natural state of growth. According to comparison of the three drying methods for high-moisture paddy, the highest burst rate was achieved by sun drying, followed by mechanical drying, and finally by late harvest, so considering the burst rate, drying of high-moisture paddy by late harvest is recommended.

In Figure 3, sun drying of paddy with a moisture content of 24.41% (D) produced a high burst rate. The reason for this phenomenon was the same as for the high burst rate of high-moisture paddy; both samples were affected by the weather, so that the burst rate of the paddy increased sharply. The burst rate of this batch of paddy after mechanical drying was 1%, which was comparatively low. Comparing the two drying methods of paddy with initial moisture content of 24.4% (D and E) and the late harvest method (C), the highest burst rate of paddy under sun drying was 3%, while the burst rate of paddy after mechanical drying and late harvest was 1%. Therefore, in terms of burst rate, it is recommended that paddy harvested with moderate moisture content should be dried mechanically or by late harvest.

Figure 3 shows that the burst rate for sun drying of paddy with initial moisture content of 18.7% was 3% (F), while the burst rate of paddy after both mechanical drying and late harvest was 1% (G), and the burst rate was higher after sun drying. The reasons for this phenomenon were the same as those applying to the previous two batches of paddy. It is therefore recommended, according to burst rate, that low-moisture paddy is dried mechanically or by late harvest.

3.2.2. Taste Value

Taste value is a comprehensive evaluation reflecting the palatability of paddy. The principle of taste-value testing is that the less light that is reflected from the paddy, and the more light that is transmitted, the lower the gelatinization of the paddy and the higher its taste value [16]. For taste-value testing of paddy after different drying methods, paddy was weighed after processing, then panned, steamed well with an appropriate amount of water, and tested using a taste meter [17]. Comparison of the taste values of paddy after different drying methods is shown in Figure 4a,b.

As shown in Figure 4a,b, the lowest scores for appearance and flavor and the lowest comprehensive scores were obtained after sun drying for paddy samples A, D, and F with 32.6%, 24.4%, and 18.7% initial moisture content, respectively. This indicates that sun drying caused changes in the qualitative characteristics of the paddy due to environmental pollution and the weather, leading to a decrease in the taste value of the paddy. The highest taste value was obtained with the late harvest method (sample C), where the appearance score was 8.9 and the flavor score was 9.1, giving a comprehensive score of 89. The difference in taste value between mechanical drying and the late harvest method was not significant, and the appearance and flavor of the paddy after late harvest or mechanical drying were better than those from sun drying. Therefore, in terms of taste value, sun drying is not recommended.

3.2.3. Appearance Quality

The appearance quality of paddy not only reflects its taste and palatability, but also affects the yield of processed paddy [18,19,20]. The appearance quality of paddy, like its taste value, influences the overall quality of paddy. The JMWT12 paddy appearance quality tester was employed to test the appearance quality characteristics of paddy after different drying methods. The results of appearance quality testing of paddy after different drying methods are shown in

Table 2. Appearance quality testing of paddy after different drying methods.

No.	Head Rice Rate (%)		Chalky Grain Rate (%)		Yellow Rice Kernel (%)		Unsound Kernel (%)
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
A	72.2	6.51	32.55	6.37	0.45	0.24	4.35	0.36
B	85.6	3.46	24.55	2.69	0.8	0.44	2.9	1.70
C	93.5	7.23	23.8	2.72	0.25	0.13	4.05	1.12
D	80.4	6.82	29.7	4.95	0.32	0.16	3.8	1.62
E	92.7	6.35	21.8	3.97	0.05	0.14	1.3	2.26
F	82.5	3.01	27.7	2.95	0.3	0.15	3.3	1.08
G	90.3	5.17	22.4	1.70	0.1	0.06	1.7	0.65

.

During the experiment, images were acquired showing rates of head rice from paddy after different drying methods, as shown in Figure 5.

In terms of appearance quality, the rate of head rice for late harvest (C) was 93.5%, followed by that of mechanically dried paddy with 24.4% moisture content (E) at 92.7%, while the rates of head rice for high-moisture paddy after mechanical drying (B) and sun drying (A) were 85.6% and 72.2%, respectively. Mechanical drying or sun drying of high-moisture paddy directly leads to an increase in broken paddy due to high moisture content and more serious loss of paddy during processing. The chalky grain rate of late harvest paddy (C) was 23.8%, while the lowest chalky grain rate was 21.8% for mechanically dried moderate-moisture paddy (E), followed by 21.8% for mechanically dried low-moisture paddy (G), while the highest rate of chalky grains was for sun-dried high-moisture paddy (A) at a rate of 32.55%.

Mechanically dried paddy with a moisture content of 24.4% (E) had the lowest rate of yellow rice kernels at 0.05%, followed by mechanical dried paddy with a moisture content of 18.7% (F) with 1% yellow rice kernels. Frequent weather changes, low temperatures and cold phenomena, rainfall, or other bad weather, resulting in a decline in the quality of late-harvested paddy and an increase in yellow rice kernels. The yellow rice kernel rates of mechanical dried (B) and sun-dried (A) high-moisture paddy were 0.8% and 0.45%, respectively, representing the most yellow rice kernels. The high moisture content of paddy dried by mechanical warm air and exposed to natural weather deteriorated the paddy quality and increased the rate of yellow rice kernels. Mechanical drying of paddy with moisture content of 24.4% (E) and 18.7% (G) resulted in the fewest unsound kernels, with 1.3% and 1.7%, respectively, while sun-dried paddy with moisture content of 32.6% (A) and late-harvested paddy (C) had the most unsound kernels, with 4.35% and 4.05%, respectively.

From the above analysis, it can be seen that neither mechanical drying nor sun drying are suitable for high-moisture paddy. The sun drying method is also unsuitable for moderate- and low-moisture paddy, and the late-harvest method had a higher rate of head rice, but other qualities were poor. Therefore, it is recommended that paddy should be harvested at moderate moisture, and mechanical drying is required to ensure it appearance quality.

3.2.4. Pasting Characteristics

Pasting characteristics directly affect the taste, texture, and stability of the product and are therefore considered important parameters for determining the quality of grains [21,22]. Usually, a lower the degree of pasting is associated with better reprocessing ability, cooking, and edibility characteristics [23]. The RVA method is currently the most commonly used approach for detecting pasting characteristics. The values of pasting characteristics for paddy after different drying methods are shown in

Table 3. Pasting characteristics of paddy after different drying methods.

No.	Paste Temperature (°C)		Peak Viscosity (cP)		Breakdown Value (cP)		Setback Value (cP)
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
A	68.4	0.41	1000	7.21	458	19.80	600	14.14
B	68.3	0.38	894	11.23	444	8.48	641	4.24
C	67.7	0.30	1022	19.90	525	48.80	542	12.72
D	67.9	0.23	996	5.13	483	12.02	569	6.36
E	67.4	0.29	1074	36.78	536	25.45	468	34.67
F	68.2	0.17	913	28.72	510	7.07	548	18.48
G	67.7	0.11	1036	35.25	524	16.97	475	17.68

.

The lower the paste temperature of grains, the better are their cooking characteristics, and the more easily they are digested by the human body, so lower paste temperatures are preferred. Higher peak viscosity indicates better palatability, higher breakdown value means a chewier grain, and a smaller setback value corresponds with a lower degree of aging [24,25,26].

According to the test results, the lowest paste temperature was 67.4 °C for mechanically dried paddy with 24.4% initial moisture content (E), suggesting that this had the best cooking characteristics. This was followed by low paste temperatures for late-harvested paddy (C) and mechanically dried paddy with 18.7% initial moisture content (G). The relatively high paste temperatures of mechanically dried and sun-dried paddy with high initial moisture indicates their poor cooking characteristics. In terms of peak viscosity, moderate- and low-moisture paddy had the highest peak viscosities after mechanical drying, at 1074 cp and 1032 cp, respectively. This suggests that moderate- and low-moisture paddy are more palatable after mechanical drying, followed by late harvested paddy. The worst result was for sun-dried high-moisture paddy (A), with a peak viscosity of 894 cp indicating the worst taste.

The breakdown values of mechanically dried paddy with moderate initial moisture (E) and late-harvested paddy (C) were both similarly high, 536 cp and 525 cp, respectively, indicating that the chewiness of rice obtained from both sets of paddy was good, while the breakdown value of sun-dried paddy with high initial moisture (A) was the lowest at 444 cp, indicating that its rice was less chewy. Mechanically dried paddy with moderate or low initial moisture content had the smallest setback values and the lowest aging rates, while the setback value for late-harvested paddy (C) was as high as 542 cp, indicating that leaving it standing in the field for a longer period of time and failing to harvest it in time would increase the aging of paddy to a higher degree.

Comprehensive analysis of the above results revealed that when paddy was harvested at high moisture, it had been harvested early and not at the right time. Then, mechanical drying caused extensive damage to the physical and chemical properties of the paddy, due to its high moisture content. Thus, the paddy’s taste value and quality became poor, the rate of broken rice and the number of yellow rice kernels increased, and the paddy’s pasting characteristics were poor. The sun-drying method is also undesirable for high-moisture paddy, resulting in poorer taste value, appearance quality, and pasting characteristics due to weather and environmental effects.

Although the late-harvested paddy had better appearance quality, taste value, and pasting characteristics, related studies have shown that the late-harvest method results in a delayed harvest that reduces the dry matter weight of the paddy, and this situation can reduce the yield of the paddy [27]. Therefore, the paddy should be harvested at the right time.

In the experiment, the mechanical drying method was most suitable for paddy with a moderate initial moisture of 24.4%, and that dried paddy had a lower burst rate, better taste value, better appearance, and better pasting characteristics, while its dry matter weight was higher. In summary, paddy should be harvested at moderate moisture and dried by mechanical drying, which can reduce the loss of dry matter while at the same time guaranteeing the quality characteristics of paddy after drying.

3.3. Gray Relational Degree

The quality index values were normalized for paddy after different drying methods. The processed results are shown in

Table 4. Normalized values of paddy quality indexes.

	Burst Rate	Taste Value	Head Rice Rate	Chalky Grain Rate	Yellow Rice Kernel	Unsound Kernel	Paste Temperature	Peak Viscosity	Breakdown Value	Setback Value
A	0.89	0.83	0.722	0.674	0.9955	0.9565	0.316	0.589	0.152	0.237
B	0.95	0.87	0.856	0.754	0.992	0.971	0.317	0.000	0.000	0.000
C	0.97	0.89	0.935	0.762	0.9975	0.9595	0.323	0.711	0.880	0.572
D	0.95	0.83	0.804	0.703	0.9968	0.962	0.321	0.567	0.424	0.416
E	0.99	0.88	0.927	0.782	0.9995	0.987	0.326	1.000	1.000	1.000
F	0.97	0.82	0.825	0.723	0.9997	0.967	0.318	0.106	0.717	0.538
G	0.99	0.87	0.903	0.776	0.9999	0.983	0.323	0.789	0.870	0.960

.

The index weight obtained by the entropy weight method [28] and the sample data were utilized to calculate the gray relational degree values. The gray relational analysis results for each sample are shown in

Table 5. Gray relational analysis results for each sample.

Sample	Gray Relational Degree	Sort
A	0.748	7
B	0.763	6
C	0.884	3
D	0.787	5
E	0.996	1
F	0.802	4
G	0.930	2

.

After calculation, the gray relational degree values of paddy quality after different drying methods were sorted as E > G > C > F > D > B > A. The quality of paddy harvested with moderate moisture content was best after mechanical drying, when its gray relational degree was 0.996; the quality of paddy harvested with low moisture content was high after mechanical drying, with a gray relational degree of 0.930; the worst quality was found in the paddy with high initial moisture after natural drying, with a gray relational degree of 0.748.

4. Conclusions

In this study, paddy with different initial levels of moisture content was dried by different drying methods, and its quality was tested after safe moisture content was reached. The effects of different drying methods on the quality of paddy were analyzed according to various testing indicators and were summarized. To analyze the effect of different drying methods on paddy quality according to a single index risks bias, so GRA was employed to determine the quality of paddy after different drying methods, and the optimal drying method was selected according to the gray relational degree. According to the comprehensive evaluation results, the gray relational degree of paddy quality after different drying methods was sorted as E > G > C > F > D > B > A. Sample E was the best quality paddy, having been harvested with moderate moisture content then dried mechanically; its gray relational degree was 0.996. Following that, the gray relational degree of low-moisture paddy after mechanical drying was 0.930. The lowest gray relational degree was for the sun-dried paddy with high initial moisture content, where the gray relational degree was 0.748; this drying method causes serious loss of paddy, so sun-drying of paddy should not be recommended to farmers.

It was particularly noted that the quality score of late-harvested paddy was also high, but this drying method does not finish until the moisture content falls to safe levels. By the time that safe moisture content was reached, the optimal harvesting time had been exceeded. Relevant studies have shown that delayed harvesting produces 3.5% dry matter loss [27] and the quality of the obtained paddy was not optimal. From the point of view of value regression, this lost value can be recovered by investment in drying machinery and equipment, to reduce the loss of dry matter incurred due to the absence of a dryer, and the loss of paddy quality caused by other drying methods.