Comparison Of Forecasts For Average Methane Concentration At Longwalls Using Autoregressive And Cause-Effect Models Using Daily Mining As Descriptive, Based On Data From The “Krupiński” Mine
Journal Title: International Journal of engineering Research and Applications - Year 2018, Vol 8, Issue 10
Abstract
This paper describes forecasts for the average concentration of methane for a specific day regarded as a mining cycle. Such forecasts are helpful in taking short-term measures to prevent methane hazards. It presents a comparison of forecast results using the autoregressive and cause-effect models based on a daily coal output at the longwall as a descriptive variable. In the autoregressive model, a descriptive variable was the average concentration of methane on the preceding day with the reference to the day of the forecast. Individual prognostic equations were performed for each day of the week. In the cause-effect model with daily coal output at the longwall as a descriptive variable, linear equations were used as prognostic models, in which the number of descriptive variables varied between one and three. The variables included the predicted coal output at the longwall on the day of forecast, on the preceding day and two days earlier. If any of the mentioned variables was insignificant for calculating parameters of the prognostic equation, then parameters were estimated excluding such a variable. Ex post forecasts based on both models were prepared for two longwalls. These forecasts differed in the processing method and the number of working days per week. Distributions of absolute and relative errors in forecasts for the average concentration of methane were compared for each longwall. Both forecasts were regarded as accurate and equivalent. Introducing a one-day forecast of methane concentration into the mining sector can enhance work safety by applying short-term preventive measures.
Authors and Affiliations
Dr Hab. Inż. Henryk Badura, Dr Zygmunt Łukaszczyk
Keywords
Related Articles
- EP ID EP406738
- DOI 10.9790/9622-0810057080.
- Views 139
- Downloads 0
Synthesis, Study and Characterization of Amidoximated Acrylic Copolymer
Copolymer membranes were fabricated into DMP and copolymer was treated with HA in aqueous solution at 60-80°C. In the amidoximation process, a part of nitrile groups of copolymer is converted into amidoxime groups. The c...
An Authentication Protocol based on Kerberos
In this paper a single-sign-on authentication protocol has been proposed. The protocol is a derivative of the Kerberos protocol that uses one server for authentication purposes, except that it is simpler in its’ implemen...
Performance Evaluation Of Support Vector Machines (Svms)And Convolutional Neural Network (Cnn) On Binary Classification Problem
Support vector machines (SVMs) have been around for decades, they have been used for a number of classification tasks. They actually have a very strong theory behind them, which make it relatively easy to choose the best...
Framework For Performance Evaluation Of Dealer’s In Car Service Industry Using CRM- Case Study
Indian automobile sector perform various task to attract customer for sales their vehicles, and try to fulfill of customers need and requirements with provide various services and facilities. Impact of customer relations...
Voltage Stability Control of Wind Farm Using STATCOM
The conversion of wind energy into electrical power generation is very useful, important and economical. Due to injection of wind power in electrical grid system, many power quality issues raisedsuch as reliability, harm...