ARTICLE | doi:10.20944/preprints202008.0725.v1
Subject: Biology, Agricultural Sciences & Agronomy Keywords: litchi; fruit bagging; bag colours; dates; quality
Online: 31 August 2020 (16:25:13 CEST)
Litchi orchards of 15 year age in Pantnagar were subjected to different fruit bagging treatments in study entitled “Impact of pre-harvest fruit bagging technology on growth and quality traits in litchi cv. Rose Scented under Indian prospective”. The combination includes white and pink polypropylene bags practiced on three dates i.e. 15, 25 and 30 days after fruit set and a control. Hence, study comprised of 7 treatment combination in total. The data of year 2017 and 2018 as well as pooled data revealed that T1 i.e. white polypropylene bags + bagging 15 days after fruit set was found to be promising in attributes such as fruit cracking (%) and Sun burn (%). T3 White Polypropylene bags + bagging 30 days after fruit set was found best for fruit Weight (g) and Acidity (%), T4 Pink Polypropylene bags + bagging 15 days after fruit set was found promising for TSS (0Brix), T6-Polypropylene Pink + 10th May (30 days after fruit set) was observed to be best for most of the desired attributes viz. Fruit breadth (mm), Yield (Kg/tree), Acidity (%), Anthocyanin (mg/100g), Fruit colour (visual), borer infestation (%) and B:C Ratio. However, fruits without bagging i.e. control were found to have inferior appearance and have maximum fruit cracking (%) and sun burn (%). Thus the bagging of litchi fruits with white polypropylene bags 15 days after fruit set resulted in lesser cracking and sunburn incidence. For other attributes, polypropylene pink bagged 30 days after fruit set was found promising. In Litchi under Indian condition, the novel technique of fruit bagging significantly enhance the fruit appearance and quality.
ARTICLE | doi:10.20944/preprints201811.0293.v1
Subject: Engineering, Energy & Fuel Technology Keywords: machine Learning (ML); artificial neutral network (ANN); bagging decision tree (BDT); SUpport Vector Machines (SVM); no free lunch theorem (NFLT); hyperparameter optimisation; model comparison; heat meter
Online: 13 November 2018 (04:41:07 CET)
Heat metres are used to calculate the consumed energy in central heating systems. The subject of this article is to prepare a method of predicting a failure of a heat meter in the next settlement period. Predicting failures is essential to coordinate the process of exchanging the heat metres and to avoid inaccurate readings, incorrect billing and additional costs. The reliability analysis of heat metres was based on historical data collected over many years. Three independent models of machine learning were proposed, and they were applied to predict failures of metres. The efficiency of the models was confirmed and compared using the selected metrics. The optimisation of hyperparameters characteristics for each of models was successfully applied. The article shows that the diagnostics of devices does not have to rely only on newly collected information, but it is also possible to use the existing big data sets.
ARTICLE | doi:10.20944/preprints201907.0319.v1
Subject: Engineering, Energy & Fuel Technology Keywords: heat meter; district heating; fault detection; predictive maintenance; Machine Learning (ML); Artificial Neural Network (ANN); Bagging Decision Tree (BDT); Support Vector Machines (SVM); hyperparameter optimisation; ensemble model
Online: 28 July 2019 (16:26:47 CEST)
The need to increase the energy efficiency of buildings as well as the use of local renewable heat sources has caused that heat meters are used not only to calculate the consumed energy but also for the active management of central heating systems. Increasing the reading frequency and the use of measurement data to control the heating system expands the requirements for the reliability of heat meters. The aim of the research is to analyse a large set of meters in the real network and predict their faults to avoid inaccurate readings, incorrect billing, heating system disruption and unnecessary maintenance. The reliability analysis of heat metres, based on historical data collected over several years, shows some regularities which cannot be easily described by physics-based models. The failure rate is almost constant and does depend on the past but is a non-linear combination of state variables. To predict meters' failures in the next settlement period, three independent machine learning models are implemented and compared with selected metrics because even the high performance of a single model (87\% True Positive for Neural Network) may be insufficient to make a maintenance decision. Additionally, performing hyperparameters optimisation boosts models' performance by a few percent. Finally, three improved models are used to build an ensemble classifier which outperforms the individual models. The proposed procedure ensures the high efficiency of fault detection (>95\%), while maintaining overfitting at the minimum level. The methodology is universal and can be utilised to study the reliability and predict faults of other types of meters and different objects with the constant failure rate.