Applying multi-objective genetic algorithm (MOGA) to optimize the energy inputs and greenhouse gas emissions (GHG) in wetland rice production

In this study, energy use efficiency (EUE) and greenhouse gas emissions (GHG) in wheat production were defined, and the energy equivalents (EE) of the inputs in production per unit production area, EUE and GHG values of the product were computed. The data used in the study were obtained from 175 different wheat producing enterprises in 2021 by conducting face-to-face surveys according to the proportional sampling method. In the study, the amount of direct (DE) and indirect energy (IE) use in wheat production and their shares in total energy consumption were defined. According to the results of the study, total energy input (EI) in wheat production was com-puted as 19,024.21 MJ/ha and energy output (EO) as 80,585.40 MJ/ha. It was defined that the input with the highest energy consumption was fertili-zation with a value of 8748.38 MJ/ha. This was followed by seed energy input 4626.79 MJ/ha (24.32%), fuel energy 2697.25 MJ/ha (14.18%), irriga-tion energy 2362.50 MJ/ha (12.42%), chemicals energy 269.19 MJ/ha (1.41%), machinery energy 309.52 MJ/ha (1.63%), human labor energy 10.58 MJ/ha (0.06%). EUE, energy productivity (EP), specific energy (SE) and net energy (NE) yield values were 4.24, 0.29 kg/MJ, 3.47 MJ/kg and 61561.19 MJ/ha, respectively. Total GHG emission for wheat production was computed as 3784.60 kgCO2-eq/ha. The highest share of total GHG emissions belonged to seed (59.41%). Seed was followed by irrigation (16.84%), nitrogen fertilizer use (14.60%), phosphate fertilizer use (3.99%), fuel use (3.49%), chemicals use (0.98%), machinery use (0.58%) and hu-man labor (0.10%). In addition, the GHG ratio in wheat production was computed as 0.69 kgCO2-eq/ha.

Management of energy use and reduction of greenhouse gas emissions (GHG) in agricultural system is the important topic. For this purpose, many methods have been proposed in different researches for solution of these items in recent years. Obviously, the selection of appropriate method was a new concern for researchers. Accordingly, the energy inputs and GHG emissions of orange production in north of Iran were modeled and optimized by artificial neural networks (ANN) and multi-objective genetic algorithm (MOGA) in this study and the results obtained were compared with the results of data envelopment analysis (DEA) approach. Results showed that, on average, an amount of 25,582.50 MJ ha−1 was consumed in orange orchards in the region and the nitrogen fertilizer was accounted for 36.84 % of the total input energy. The outcomes of this study demonstrated that on average 803 kg carbon dioxide (kgCO2eq.) is emitted per ha and diesel fuel is responsible for 35.7 % of all emissions. The results of ANN signified that they were capable of modeling crop output and total GHG emissions where the model with a 13-4-2 topology had the highest accuracy in both training and testing steps. The optimization of energy consumption using MOGA revealed that the total energy consumption and GHG emissions of orange production can be reduced to the values of 13,519 MJ ha−1 and 261 kgCO2eq. ha−1, respectively. A comparison between MOGA and DEA clearly showed the better performance of MOGA due to simultaneous application of different objectives and the global optimum solutions produced by the last generation.

This study was conducted in order to model energy consumption and greenhouse gas emissions for peanut production in Guilan province of Iran using artificial neural network (ANN). Also, the multi-objective genetic algorithm was used for optimization of energy inputs and GHG emissions in the region. Data were randomly collected from 120 farms in Astaneh Ashrafiyeh city with face to face questionnaire method. The results illustrated that the total energy consumption and the average yield were calculated as 19248.04 MJ ha-1 and 3488.39 kg ha -1 , respectively. Moreover, the results showed that the share of chemical fertilizers (mainly nitrogen) and diesel fuel energy to the total energy input were the highest. Also, the energy used efficiency ratio calculated as 4.53. The results of GHG emissions analysis showed the total GHG emissions were 571.18 kgCO2eq. ha -1 and the diesel fuel has the main reasonable of GHG emissions in peanut production. In this study, several direct and indirect factors have been identified to create a model based on ANN to predict energy use and GHG emissions in peanut production. The ANN model with 9-22-2 structure was capable of predicting the peanut yield and GHG emissions. Moreover, the results of the best topology showed that R 2 was 0.994 and 0.999, RMSE was 0.076 and 0.003 and MAPE was 0.174 and 0.009 for peanut yield and GHG emissions, respectively. The results of optimization indicated the total energy consumption and GHG emissions generation was calculated about 6888 MJ ha -1 and 159.08 kgCO2eq. ha -1 , respectively. The total GHG emissions reduction was found to be 412.09 kgCO2eq. ha -1 in optimal generation toward present farms.

توسعه پایدار تولید یک محصول در هر منطقه مستلزم توجه به سیر انرژی سامانه تولیدی آن است، در عین حال توجه به نهاده‌های ورودی سامانه تولیدی با نگرش مدیریت محیط زیست نیز از اهمیت ویژه ای برخوردار است. در این تحقیق انرژی مصرفی و انتشار گازهای گلخانه‌ای تولید چای در استان گیلان مورد بررسی قرار گرفت. اطلاعات از طریق مصاحبه حضوری با 75 چای‌کار گیلانی و تطبیق اطلاعات با دفترچه چای هر کشاورز جمع‌آوری شد. مجموع انرژی‌های ورودی 60/39060 مگاژول بر هکتار بود. کارایی انرژی 22/0 محاسبه شد. کودهای شیمیایی بیش‌ترین سهم را در انرژی‌های مصرفی و انتشار گازهای گلخانه‌ای به ترتیب با 55/58 و 22/74 درصد در تولید چای به خود اختصاص دادند. مجموع انتشار گازهای گلخانه‌ای تولید چای در منطقه kgCO2eq. ha-1 82/1281 بود. نتایج استفاده از تابع کاب داگلاس و تحلیل حساسیت انرژی تولید چای در استان گیلان نشان داد که تأثیر تمامی نهاده‌های انرژی ورودی به غیر از سموم شیمیایی بر عملکرد مثبت بود و تأثیر نهاده انرژی نیروی کارگری بر عملکرد در سطح یک درصد معنی‌دار شد. نهاده انرژی نیروی کارگری، حساس‌ترین و همچنین بیش‌ترین تأثیر را بر عملکرد داشت و پس از آن نهاده‌های انرژی ماشین‌ها و سموم شیمیایی بیش‌ترین تأثیر را بر عملکرد چای در استان گیلان داشتند. توسعه پایدار تولید یک محصول در هر منطقه مستلزم توجه به سیر انرژی سامانه تولیدی آن است، در عین حال توجه به نهاده‌های ورودی سامانه تولیدی با نگرش مدیریت محیط زیست نیز از اهمیت ویژه ای برخوردار است. در این تحقیق انرژی مصرفی و انتشار گازهای گلخانه‌ای تولید چای در استان گیلان مورد بررسی قرار گرفت. اطلاعات از طریق مصاحبه حضوری با 75 چای‌کار گیلانی و تطبیق اطلاعات با دفترچه چای هر کشاورز جمع‌آوری شد. مجموع انرژی‌های ورودی 60/39060 مگاژول بر هکتار بود. کارایی انرژی 22/0 محاسبه شد. کودهای شیمیایی بیش‌ترین سهم را در انرژی‌های مصرفی و انتشار گازهای گلخانه‌ای به ترتیب با 55/58 و 22/74 درصد در تولید چای به خود اختصاص دادند. مجموع انتشار گازهای گلخانه‌ای تولید چای در منطقه kgCO2eq. ha-1 82/1281 بود. نتایج استفاده از تابع کاب داگلاس و تحلیل حساسیت انرژی تولید چای در استان گیلان نشان داد که تأثیر تمامی نهاده‌های انرژی ورودی به غیر از سموم شیمیایی بر عملکرد مثبت بود و تأثیر نهاده انرژی نیروی کارگری بر عملکرد در سطح یک درصد معنی‌دار شد. نهاده انرژی نیروی کارگری، حساس‌ترین و همچنین بیش‌ترین تأثیر را بر عملکرد داشت و پس از آن نهاده‌های انرژی ماشین‌ها و سموم شیمیایی بیش‌ترین تأثیر را بر عملکرد چای در استان گیلان داشتند.

In this research, the energy use efficiency (EUE) and greenhouse gas emissions (GHG) of cotton cultivation in Beşiri district of Batman province in Turkey were determined. This research was conducted through face-to-face surveys with 64 farms selected by simple random sampling method in the 2018-2019 cultivation season. The energy input (EI) and energy output (EO) in cotton cultivation were calculated as 52,302.62 MJ/ha and 60,341.03 MJ/ha. Energy inputs consist of electricity energy with 19,948.86 MJ/ha(38.14%), chemical fertilizers energy with 14,163.83 MJ/ha (27.08%), diesel fuel energy with 13,218.49 (25.27%), irrigation water energy with 2563.79 MJ/ha(4.90%), machinery energy with 1071.14 MJ/ha(2.05%), chemicals energy with 797.96 MJ/ha (1.53%), seed energy with 291.46 MJ/ha (0.56%) and human labour energy with 247.09 MJ/ha(0.47%), respectively. Total energy inputs in cotton cultivation can be categorized as 68.79% direct, 31.21% indirect, 5.93% renewable and 94.07% non-renewable. EUE, specific energy (SE), energy productivity (EP) and net energy (NE) in cotton cultivation were calculated as 1.15, 10.23 MJ/kg, 0.10 kg /MJ and 8038.41 MJ/ ha, respectively. Total GHG was calculated as 3742.59 kgCO2-eq ha-1 for cotton cultivation with the greatest share taken by nitrogen (26.19%). Nitrogen was followed by electricity (24.73%), irrigation water (18.48%), diesel fuel (17.31%), seed (5.04%), chemicals (2.93%), phosphorous (2.74%), human labour (2.36%), potassium (0.19%) and machinery (0.03%), respectively. GHG ratio value was calculated as 0.73 kgCO2-eq kg-1 in cotton cultivation.

In the past, we employed a multi-objective genetic algorithm (MOGA) for optimization of model parameters and feature selection, and then devised a stock scoring mechanism to rank and select stocks for forming a portfolio. With each chromosome representing a feasible portfolio, that adopted multi-objective genetic algorithm (MOGA) model thus decided good portfolios by considering their return and risk. In this paper, we further improve upon the MOGA model using financial knowledge to help selection of beneficial portfolios. Especially, we refine the evaluation criteria with the assistance of relevant domain knowledge from investment. Based on the promising results, we expect this improved MOGA methodology to advance the current state of research in soft computing for real-world stock selection applications.

To achieve the goal of "carbon peak and neutrality," the strict requirements for greenhouse gas (GHG) emissions control in the agricultural sector were recommended in relevant plans for Beijing during the 14th Five-Year Plan period. Through collecting agricultural activity data and calculating and screening the emission factors, the amount and emission characteristics of agricultural GHG emissions in Beijing in 2020 were estimated and set as the baseline condition. On this basis, the GHG emissions in 2025 with optimized measurements implemented, which were selected in combination with the natural conditions and planting-breeding mode of Beijing, were set as the reduction condition. The emission reduction potential and its distribution during the 14th Five-Year Plan Period were predicted simultaneously. Meanwhile, the reduction effects on the GHG emissions of optimized measurements were evaluated. In addition, relevant policy recommendations on GHG reduction were proposed accordingly. The results revealed that the total agricultural GHG emissions in Beijing were estimated to be 456000 t (CO2-eq) in 2020, primarily from sources of animal intestinal fermentation and manure management, with contribution rates of 50.7% and 26.7%, respectively. Spatially, it was mainly distributed in districts with large livestock and poultry breeding scales, such as Shunyi District, Miyun District, and Yanqing District, etc. It was predicted that in 2025, the total agricultural GHG emissions would be 349000 t (CO2-eq), and the emission reduction potential in the 14th Five-Year Plan period would be 107000 t (CO2-eq). Animal intestinal fermentation would be the emission source with the largest reduction potential (60000 tons, CO2-eq), followed by the emission source of animal manure management (37000 tons, CO2-eq). Adjusting fodder composition and optimizing manure management were analyzed to be the most effective optimized measurements for agricultural GHG emission reduction. Moreover, the emission reduction potential of CH4 would be greater than that of N2O. The emission reduction potential would be mainly distributed in Miyun District, Shunyi District, Yanqing District, Fangshan District, Tongzhou District, and other suburbs with large livestock and poultry breeding scales, accounting for more than 10% of the total emission reduction potential for each. These regions with large emission reduction potential should be prioritized and then the assessments should be extended to the whole city. The measurements were recommended as follows:① the research and promotion of technologies such as fodder optimization and the efficient treatment of manure should be strengthened, ② the scope of the combination of planting and breeding model should be expanded to promote the development of circular agriculture, and ③ relevant standards, guidelines, and specifications for green and low-carbon agriculture should be formulated, and the regulatory and policy system for synergy reduction of agricultural pollution and GHG should be developed.

In agricultural production, it is important to determine where input usage saving can be implemented by taking energy use into consideration and to analyze the greenhouse gas emissions of agricultural activities. This study has been conducted to review orange (Citrus sinensis L.) production in terms of energy balance and greenhouse gas (GHG) emissions. This study was carried out during the 2015/2016 production season in Adana, a province in Turkey. Energy balance and GHG emissions have been defined by calculating the inputs and outputs of agricultural nature used in orange production. The findings of the study indicate that the distribution of energy inputs in orange production are 11,880 MJ ha−1 (34.10%) of electricity, 10,079.75 MJ ha−1 (28.93%) of chemical fertilizer energy, 7630 MJ ha−1 (21.90%) of chemical energy, 3052 MJ ha−1 (8.76%) of diesel fuel energy, 1348.91 MJ ha−1 (3.87%) of human labor energy, 378 MJ ha−1 (1.09%) of irrigation water energy, 351.22 MJ ha−1 (1.01%) of machinery energy and 118.80 MJ ha−1 (0.34%) of lime energy. In total, input energy (IE) in orange production has been calculated as 34,838.68 MJ ha−1 and the output energy (OE) has been calculated as 95,000 MJ ha−1. Energy use efficiency (EUE), specific energy (SE), energy productivity (EP) and net energy (NE) have been calculated as 2.73, 0.70 MJ kg−1, 1.44 kg MJ−1 and 60,161.32 MJ ha−1, respectively. The total energy input in the production of oranges was divided into: 47.82% direct, 52.18% indirect, 4.96% from renewable sources and 95.04% from non-renewable sources. The GHG emissions figure for orange production was 3794.26 kg CO2–eq ha−1, with electricity having the greatest share, 1983.96 (52.29%); the GHG ratio was 0.08 kg CO2–eq kg−1. According to the results, the production of orange was considered to be profitable in terms of EUE.

Energy consumption, greenhouse gas emissions and assessment of sustainability index in corn agroecosystems of Iran

This study aimed to model energy use, energy efficiency, and greenhouse gas emissions in rain-fed wheat production by using a nonparametric data envelopment analysis (DEA) method. Data were collected through face-to-face interviews with 140 wheat farmers in 4 districts of Antalya Province. The energy inputs (independent variables) were human labor, seeds, chemical fertilizers, herbicides, and diesel fuel, and the energy output was the dependent variable. The results showed that the average energy consumption and the output energy for the studied wheat production system were 21. 07GJ ha−1 and 50. 99 GJ ha−1, respectively, and the total GHG emissions were calculated to be 592.12 kg CO2eq ha−1. Chemical fertilizer has the highest share of energy consumption and total GHG emissions. Based on the results from DEA, the technical efficiency of the farmers was found to be 0.81, while pure technical and scale efficiencies were 0.65 and 0.76, respectively. The results also highlighted that there is a potential opportunity to save approximately 14% (2.93 GJ ha−1) of the total energy consumption and consequently a 17% reduction in GHG emissions by following the optimal amounts of energy consumption while keeping the wheat yield constant. Efficient use of energy and reduction in GHG emissions will lead to resource efficiency and sustainable production, which is the main aim of the green economy.

Energy use is an important consideration for developing more sustainable agricultural practices. Identifying animal production methods that maximize energy efficiency and minimize greenhouse gas emissions is vital. This study determined the energy use and the energy use efficiency (EUE) of two specialized dairy farms with different barn planning systems in Konya, Turkey. Total energy use included both the direct energy and indirect energy consumed during the production of farm inputs. This study investigated changes in energy use and EUE between Dairy Farm A, with a freestall dairy cattle housing system (ADF), and Dairy Farm B, with loose housing systems (BDF).The results showed that animal feed accounted for a high percentage of total energy use and that electricity comprised the major portion of the direct energy used on both farms. The total energy use per hectare was lower for ADF compared with BDF. The energy productivity of the farms was set at 5.4 L milk per 100 MJ−1and 3.9 L milk per 100 MJ−1 for energy use on ADF and BDF, respectively. EUE was highly dependent on the energy inputs. For that reason, it is recommended that freestall dairy cattle barns be used for dairy cattle breeding and milk production. © 2012 American Institute of Chemical Engineers Environ Prog, 32: 1202–1208, 2013

In this study, the energy balance and Greenhouse Gas Emissions (GHG) of cotton cultivation in Bismil district of Diyarbakır province in Turkey was defined. The energy balance and GHG of cotton cultivation was computed by conducting face to face surveys with 73 farms in the 2018-2019 cultivation season, which were selected by simple random sampling method. The energy input and output in cotton cultivation were computed as 54 617.62 MJ ha-1 and 65 984.42 MJ ha-1, respectively. Energy inputs occurs of electricity energy with 18 608.40 MJ ha-1 (34.06%), chemical fertilizers energy with 15 254.67 MJ ha-1 (27.93%), diesel fuel energy with 14 364.68 (26.30%), irrigation water energy with 3 559.50 MJ ha-1 (6.53%), machinery energy with 1 152.79 MJ ha-1 (2.11%), chemicals energy with 1 075.76 MJ ha-1 (1.96%), seed energy with 307.98 MJ ha-1 (0.57%), human labour energy with 293.84 MJ ha-1 (0.54%), respectively. Total energy inputs in cotton cultivation can be classified as 67.43% direct, 32.57% indirect, 7.62% renewable and 92.38% non-renewable. Energy use efficiency, specific energy, energy productivity and net energy in cotton cultivation were computed as 1.21, 9.77 MJ kg-1, 0.10 kg MJ-1 and 11 366.80 MJ ha-1, respectively. Total GHG emissions were computed as 6 482.36 kgCO2-eqha-1 for cotton cultivation with the greatest input part for electricity with 47.94% (3 107.60 kgCO2-eqha-1). The electricity followed up nitrogen with 16.29% (1 055.67 kgCO2-eqha-1), irrigation water with 14.82% (960.50 kgCO2-eqha-1), diesel fuel with 10.86% (704.08 kgCO2-eqha-1), seed with 3.07% (199.14 kgCO2-eqha-1 ), chemicals with 2.28% (147.76 kgCO2-eqha-1), phosphorous with 1.78% (115.64 kgCO2-eqha-1), human labour with 1.62% (104.94 kgCO2-eqha-1), machinery with 1.26% (81.85 kgCO2-eqha-1) and potassium with 0.08% (5.18 kgCO2-eqha-1), respectively. Additionally, GHG ratio value was computed as 1.16 kgCO2-eqkg-1 in cotton cultivation.

Analysis of energy use and greenhouse gas emissions (GHG) of transplanting and broadcast seeding wetland rice cultivation

In order to achieve sustainable development in agriculture, it is necessary to quantify and compare the energy, economic, and environmental aspects of products. This paper studied the energy, economic, and greenhouse gas (GHG) emission patterns in broiler chicken farms in the Alborz province of Iran. We studied the effect of the broiler farm size as different production systems on the energy, economic, and environmental indices. Energy use efficiency (EUE) and benefit-cost ratio (BCR) were 0.16 and 1.11, respectively. Diesel fuel and feed contributed the most in total energy inputs, while feed and chicks were the most important inputs in economic analysis. GHG emission calculations showed that production of 1000 birds produces 19.13t CO2-eq and feed had the highest share in total GHG emission. Total GHG emissions based on different functional units were 8.5t CO2-eq per t of carcass and 6.83kg CO2-eq per kg live weight. Results of farm size effect on EUE revealed that large farms had better energy management. For BCR, there was no significant difference between farms. Lower total GHG emissions were reported for large farms, caused by better management of inputs and fewer bird losses. Large farms with more investment had more efficient equipment, resulting in a decrease of the input consumption. In view of our study, it is recommended to support the small-scale broiler industry by providing subsidies to promote the use of high-efficiency equipment. To decrease the amount of energy usage and GHG emissions, replacing heaters (which use diesel fuel) with natural gas heaters can be considered. In addition to the above recommendations, the use of energy saving light bulbs may reduce broiler farm electricity consumption.

This study examined the input energy, economic indices, and Greenhouse Gas (GHG) emissions in sunflower farm enterprises of Kermanshah province of Iran. Different mechanization production systems involving traditional, semi‐mechanized, and mechanized ones were statistically compared. Results revealed that mechanized farms consumed more total inputs energy, while possessed significantly higher yield and better economic indices. In which, the human labor, diesel fuel, and fertilizer were the most predominant inputs in GHG emissions. In particular, traditional, semi‐mechanized and mechanized farms emitted 358, 386, and 438 kg CO2/ha, respectively. Also, technical efficiencies were reported as 0.88, 0.86, and 0.96, for traditional, semi‐mechanized, and mechanized farms, respectively. The relationship among different variables including energy inputs, GHG emissions, output energy, and benefit to cost ratio was studied using econometric modeling. Data envelopment analysis (DEA) and multi‐objective genetic algorithm (MOGA) were also applied to detect a set of Pareto frontiers in the combination of energy, environmental, and economic indices (energy consumption, GHG emissions, and benefit to cost ratio as three selected output parameters) for sunflower production. It has been observed that the capability of MOGA for energy saving was higher than DEA. Application results of DEA and MOGA combined algorithms showed that diesel fuel and water had the highest and lowest potential for total energy savings, respectively.