Fast non-technical losses identification through Optimum-Path Forest

Fraud detection in energy systems by illegal consumers is the most actively pursued study in non-technical losses by electric power companies. Commonly used supervised pattern recognition techniques, such as Artificial Neural Networks and Support Vector Machines have been applied for automatic commercial frauds identification, however they suffer from slow convergence and high computational burden. We introduced here the Optimum-Path Forest classifier for a fast non-technical losses recognition, which has been demonstrated to be superior than neural networks and similar to Support Vector Machines, but much faster. Comparisons among these classifiers are also presented. © 2009 IEEE.

Air-assistance and low volume application to control of asian rust on soybean crop

Currently, one of factors that cause the production cost increase of soybean crop is the pesticide application. The most important disease in soybean crop is Asian rust, caused by Phakopsora pachyrhizi Sydon & P. Sydon fungus, which can cause significant loss of the production. Therefore, this work aimed at evaluation of different spraying techniques on the spray deposits and some parameters of soybean crop: grain size, weight of 1 000 seeds and the crop productivity. Two experiments were carried out in the experimental area of FCA/UNESP (Faculdade de Ciencias Agronomicas/Universidade Estadual Paulista Julio de Mesquita Filho) - Botucatu, S P, Brazil, in soybean crop, Conquista variety, in the 2007/2008 season. In the first experiment, three air levels (0, 9 and 29 km/h of the air speed generated by fan) with flat fan nozzle XR 8002 with a spray volume of 130 l/ha were compared with a rotating nozzle - using low volume oily - LVO at 40 l/ha of spray volume. The second experiment was carried out under the same conditions as the previous experiment, including a control treatment (untreated plants). The disease severity was evaluated using a diagrammatic scale with a visual evaluation of the disease on 15 leaves of each plot. The grades varied between 0.6 and 78.5% of the disease severity. The use of air assistance when compared with the rotating system nozzle did not show significant differences for spray deposits on adaxial and abaxial surface of the leaves in bottom part of the plant. The air assistance with maximum air speed (29 km/h) increased the productivity with respect of the other treatments.

Carbono orgânico e carbono residual do solo em sistema de plantio direto, submetido a diferentes manejos

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

What is the importance of selecting features for non-technical losses identification?

Although non-technical losses automatic identification has been massively studied, the problem of selecting the most representative features in order to boost the identification accuracy has not attracted much attention in this context. In this paper, we focus on this problem applying a novel feature selection algorithm based on Particle Swarm Optimization and Optimum-Path Forest. The results demonstrated that this method can improve the classification accuracy of possible frauds up to 49% in some datasets composed by industrial and commercial profiles. © 2011 IEEE.

Identification of foliar diseases in cotton crop

The pathogens manifestation in plantations are the largest cause of damage in several cultivars, which may cause increase of prices and loss of crop quality. This paper presents a method for automatic classification of cotton diseases through feature extraction of leaf symptoms from digital images. Wavelet transform energy has been used for feature extraction while Support Vector Machine has been used for classification. Five situations have been diagnosed, namely: Healthy crop, Ramularia disease, Bacterial Blight, Ascochyta Blight, and unspecified disease. © 2012 Taylor & Francis Group.

Características morfoagronômicas do milho submetido a diferentes níveis de desfolha manual

The present study was conducted to evaluate the damage caused by defoliation in maize. The experiment was carried out in 2005/2006 at Roma Farm, Ituverava-SP, using the triple hybrid CODETEC 304®. The experiment consisted of split plots (split-plot), which were distributed at random in three blocks, with the main treatment manual defoliation (30%, 40%, 60% and 80% defoliation) in the plots and as the secondary treatment the phenological phases (D1, D2 and D3) with three replicates for each treatment and a control plot (0% defoliation) for each block. It was evaluated the production of plots (g/m2), the average size of the spikes (cm) and the average weight of thousand grains (g). According to the results, it was concluded that the source-sink relations were negatively affected, and the treatment with 80% defoliation was the one that most affected all variables, thus, it is a level of defoliation that is responsible for the largest losses, with no means of compensation for the maize crop.

A genetic algorithm for crop rotation

In the last few years, crop rotation has gained attention due to its economic, environmental and social importance which explains why it can be highly beneficial for farmers. This paper presents a mathematical model for the Crop Rotation Problem (CRP) that was adapted from literature for this highly complex combinatorial problem. The CRP is devised to find a vegetable planting program that takes into account green fertilization restrictions, the set-aside period, planting restrictions for neighboring lots and for crop sequencing, demand constraints, while, at the same time, maximizing the profitability of the planted area. The main aim of this study is to develop a genetic algorithm and test it in a real context. The genetic algorithm involves a constructive heuristic to build the initial population and the operators of crossover, mutation, migration and elitism. The computational experiment was performed for a medium dimension real planting area with 16 lots, considering 29 crops of 10 different botanical families and a two-year planting rotation. Results showed that the algorithm determined feasible solutions in a reasonable computational time, thus proving its efficacy for dealing with this practical application.

Ammonia volatilization losses from surface-applied urea with urease and nitrification inhibitors

Urease inhibitor (UI) and nitrification inhibitor (NI) have the potential to improve N-use efficiency of applied urea and minimize N losses via gaseous emissions of ammonia (NH 3) to the atmosphere and nitrate (NO3-) leaching into surface and ground water bodies. There is a growing interest in the formulations of coating chemical fertilizers with both UI and NI. However, limited information is available on the combined use of UI and NI applied with urea fertilizer. Therefore the aim of this study was to investigate the effects of treating urea with both UI and NI to minimize NH 3 volatilization. Two experiments were set up in volatilization chambers under controlled conditions to examine this process. In the first experiment, UR was treated with the urease inhibitor NBPT [N-(n-butyl) thiophosphoric acid triamide] at a rate of 1060 mg kg -1 urea and/or with the nitrification inhibitor DCD (dicyandiamide) at rates equivalent to 5 or 10% of the urea N. A randomized experimental design with five treatments and five replicates was used: 1) UR, 2) UR + NBPT, 3) UR + DCD 10%, 4) UR + NBPT + DCD 5%, and 5) UR + NBPT + DCD 10%. The fertilizer treatments were applied to the surface of an acidic Red Latosol soil moistened to 60% of the maximum water retention and placed inside volatilization chambers. Controls chambers were added to allow for NH 3 volatilized from unfertilized soil or contained in the air that swept over the soil surface. The second experiment had an additional treatment with surface-applied DCD. The chambers were glass vessels (1.5 L) fit with air inlet and outlet tubings to allow air to pass over the soil. Ammonia volatilized was swept and carried to a flask containing a boric acid solution to trap the gas and then measured daily by titration with a standardized H 2SO 4 solution. Continuous measurements were recorded for 19 and 23 days for the first and second experiment, respectively. The soil samples were then analyzed for UR-, NH4+-, and NO3--N. Losses of NH 3 by volatilization with unamended UR ranged from 28 to 37% of the applied N, with peak of losses observed the third day after fertilization. NBPT delayed the peak of NH 3 losses due to urease inhibition and reduced NH 3 volatilization between 54 and 78% when compared with untreated UR. Up to 10 days after the fertilizer application, NH 3 losses had not been affected by DCD in the UR or the UR + NBPT treatments; thereafter, NH 3 volatilization tended to decrease, but not when DCD was present. As a consequence, the addition of DCD caused a 5-16% increase in NH 3 volatilization losses of the fertilizer N applied as UR from both the UR and the UR + NBPT treatments. Because the effectiveness of NBPT to inhibit soil urease activity was strong only in the first week, it could be concluded that DCD did not affect the action of NBPT but rather, enhanced volatilization losses by maintaining higher soil NH4+ concentration and pH for a longer time. Depending on the combination of factors influencing NH 3 volatilization, DCD could even offset the beneficial effect of NBPT in reducing NH 3 volatilization losses. © 2012 Elsevier Ltd.

Identification and feature selection of non-technical losses for industrial consumers using the software WEKA

This work has as objectives the implementation of a intelligent computational tool to identify the non-technical losses and to select its most relevant features, considering information from the database with industrial consumers profiles of a power company. The solution to this problem is not trivial and not of regional character, the minimization of non-technical loss represents the guarantee of investments in product quality and maintenance of power systems, introduced by a competitive environment after the period of privatization in the national scene. This work presents using the WEKA software to the proposed objective, comparing various classification techniques and optimization through intelligent algorithms, this way, can be possible to automate applications on Smart Grids. © 2012 IEEE.

Irrigation management in pruned coffee tree crop

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

CO2-C losses and carbon quality of selected Maritime Antarctic soils

Polar Regions are the most important soil carbon reservoirs on Earth. Monitoring soil carbon storage in a changing global climate context may indicate possible effects of climate change on terrestrial environments. In this regard, we need to understand the dynamics of soil organic matter in relation to its chemical characteristics. We evaluated the influence of chemical characteristics of humic substances on the process of soil organic matter mineralization in selected Maritime Antarctic soils. A laboratory assay was carried out with soils from five locations from King George Island. We determined the contents of total organic carbon, oxidizable carbon fractions of soil organic matter, and humic substances. Two in situ field experiments were carried out during two summers, in order to evaluate the CO2-C emissions in relation to soil temperature variations. The overall low amounts of soil organic matter in Maritime Antarctic soils have a low humification degree and reduced microbial activity. CO2-C emissions showed significant exponential relationship with temperature, suggesting a sharp increase in CO2-C emissions with a warming scenario, and Q10 values (the percentage increase in emission for a 10°C increase in soil temperature) were higher than values reported from elsewhere. The sensitivity of the CO2-C emission in relation to temperature was significantly correlated with the humification degree of soil organic matter and microbial activity for Antarctic soils. © 2012 Antarctic Science Ltd.

Effects of superficial liming and silicate application on soil fertility and crop yield under rotation

Soil acidity and low natural fertility are the main problems for grain production in Brazilian 'cerrado'. Although lime has been the most applied source for soil correction, silicate may be an alternative material due to its lower solubility and Si supply, which is beneficial to several crops. This work aimed to evaluate the efficiency of superficial liming and calcium/magnesium silicate application on soil chemical attributes, plant nutrition, yield components and final yield of a soybean/white oat/maize/bean rotation under no-tillage system in a dry-winter region. The experiment was conducted under no tillage system in a deep acid clayey Rhodic Hapludox, Botucatu-SP, Brazil. The design was the completely randomized block with sixteen replications. Treatments consisted of two sources for soil acidity correction (dolomitic lime: ECC=90%, CaO=36% and MgO=12%; calcium/magnesium silicate: ECC=80%, CaO=34%, MgO=10% and SiO2=22%) applied in October 2006 to raise base saturation up to 70% and a control, with no soil correction. Soybean and white oat were sown in 2006/2007 as the main crop and off-season, respectively. Maize and bean were cropped in the next year (2007/2008). Products from silicate dissociation reach deeper soil layers after 18months from the application, compared to liming. Additionally, silicate is more efficient than lime to increasing phosphorus availability and reducing toxic aluminum. Such benefits in soil chemical attributes were only evidenced during bean cropping, when grain yield was higher after silicate application comparatively to liming. Both correction sources were improved mineral nutrition of all the other crops, mainly Ca and Mg levels and agronomical characteristics, reflecting in higher yield. © 2012 Elsevier B.V.

Evaluation of different foliar fertilizers on the crop production of sugarcane

Influence of different foliar fertilizers (phosphite, micronutrients, biostimulant, phosphite + micronutrients, phosphite + biostimulant, micronutrients + biostimulant and phosphite + micronutrients + biostimulant) on yield of sugarcane was evaluated after fertilization at 30, 90 and 150 days after harvesting two-year-old sugarcane. The experiment was carried out in a commercial crop employing a randomized block design in four replicates. Higher stalk masses were observed for fertilization at 30 days after harvest, and the higher content of sucrose, total recoverable sugar and Brix degrees were observed for sugarcane fertilized after 150 days. Statistical analysis (Duncan's test) revealed no significant variation (P & 0.05) in Brix degree, sucrose content and total recoverable sugar. For total recoverable sugar x stalk weight (the main payment type for sugarcane producers), the following sequence (time treatment, fertilizer composition) 30-days, micronutrient + biostimulant; 150-days, biostimulant; and 90-days, biostimulant increased 11%, 17%, and 21% the yield of sugarcane. © 2013 Copyright Taylor and Francis Group, LLC.

Influéncia da adubação nitrogenada sobre a interferéncia de plantas daninhas em feijoeiro1

The competition between weeds and bean plants, especially because of nutrients in the environment, is extremely important to minimize potential losses of crop productivity. Thus, the aim of this work was to evaluate the effects of using nitrogen fertilization on the timing and extent of the period before interference (PAI), total period of interference (PTPI) and the critical period of interference control (PCPI) in bean cultivar 'Ruby'. The seeds were sown in the conventional system and the trial was conducted in two areas at the same time, in which the nitrogen fertilization was performed only in one of them. Each area received 14 treatments consisting of two groups: first, the crop was free of interference of weeds from emergence to 10, 20, 30, 40, 50, 60 and 70 days; and the second: the crop remained free of interference from sowing to the same periods described above. The experimental design was randomized blocks with four replications. Raphanus raphanistrum showed greater relative importance in the two areas, mainly due to the accumulation of dry mass, followed by Amaranthus deflexus. There was no difference in diversity and evenness of weed populations in response to the fertilization. In the area without N PAI occurred 18 days after emergence (DAE), and up to 24 DAE PTPI, resulting in PCPI 18 to 24 DAE, this area had 58% of reduction in crop yield. In the area where with nitrogen fertilization, PAI occurred at 38 DAE and PTPI occurred at 19 DAE. In this area there was 56% of reduction in crop yield. The nitrogen fertilization increased crop productivity and favored competitiveness in relation to weeds.

Phosphorus and potassium balance in a corn-soybean rotation under no-till and chiseling

Nutrient use efficiency has become an important issue in agriculture, and crop rotations with deep vigorous rooted cover crops under no till may be an important tool in increasing nutrient conservation in agricultural systems. Ruzigrass (Brachiaria ruziziensis) has a vigorous, deep root system and may be effective in cycling P and K. The balance of P and K in cropping systems with crop rotations using ruzigrass, pearl millet (Pennisetum glaucum) and ruzigrass + castor bean (Ricinus communis), chiseled or not, was calculated down to 0.60 m in the soil profile for 2 years. The cash crops were corn in the first year and soybean in the second year. Crop rotations under no-till increased available P amounts in the soil-plant system from 80 to 100 %, and reduced K losses between 4 and 23 %. The benefits in nutrient balance promoted by crop rotations were higher in the second year and under without chiseling. Plant residues deposited on the soil surface in no-till systems contain considerable nutrient reserve and increase fertilizer use efficiency. However, P release from ruzigrass grown as a sole crop is not synchronized with soybean uptake rate, which may result in decreased yields. © 2013 Springer Science+Business Media Dordrecht.