With a predictive model that can predict failures of a manufacturing machine, many benefits can be obtained. Unnecessary downtime and accidents can be avoided. In this study a wood chipper which has 12 replaceable knives was examined. The specific task was to create a predictive model that can predict if a knife change is needed or not. To create a predictive model, supervised machine learning was used. Decision forest was the algorithm used in this study. Data samples were collected from vibration measurements. Each sample was labeled with help of ocular inspections of the knives. Microsoft Azure learning studio was the workspace used to train all models. The data set acquired consist of 106 samples, were only 9 samples belongs to the minority class. Two strategies of training a model were used, with and without oversampling. The result for the best model without oversampling obtained 87.5% precision and 77.8% recall. The best model with oversampling achieved 79% precision and 86.7% recall. This result indicates that the trained models can be useful. However, the validity of the result has been hurt by a small data set and many uncertainness of acquiring the data set.

The aim of this master's thesis is to design of wood chipper connected to the tractor ZETOR 3011. In the introduction of this thesis it is described the general use of wood chippers. First part of thesis contains recherche from wood chippers' area. Then it is followed with the description of the own chipper design and with the basic design calculations. Part of the thesis is 3D model of the machine and drawing documentation of selected parts. Next part is given economic evaluation. In the last part of thesis is made analysis of risk.

In both the chemical and mechanical pulping process, the logs are cut into wood chips by a disc chipper before fibre separation. To make the wood chipping process more efficient, one have to investigate in detail the coupling between process parameters and the quality of the chips. One objective of this thesis was to obtain an understanding of the fundamental mechanisms behind the creation of wood chips. Another objective with the thesis was to investigate whether it was possible to, in a way tailor the chipping process so as to reduce the energy consumption in a following mechanical refining process. Both experimental and analytical/numerical approaches have been taken in this work. The first part of the experimental investigations, were performed with an in-house developed chipping device and a digital speckle photography equipment. The results from the experimental investigation showed that the friction between the log and chipping tool is probably one crucial factor for the chip formation. Further more it was found that the indentation process is approximately self-similar, and that the stress field over the entire crack-plane is critical for chip creation. The developed analytical model predicts the normal and shear strain distribution and to be more specific, the model can predict the compressive stresses parallel to the fibre direction for an assumed linear elastic and orthotropic material. The analytical distributions were found to be in reasonable agreement with the corresponding distributions obtained from a finite element analysis. To be able to study the chipping process under realistic conditions, which for example means to use chipping rates representative for a real wood chipper, a laboratory chipper was developed. Details regarding the chipper and how to evaluate the force measurements are given together with an example of how the force on the cutting tool (the knife) varies with time during cutting. To investigate the influence of a certain chipping process parameter, the chips were after production in the laboratory chipper, refined in a pilot refiner during conditions optimized for TMP (thermomechanical pulp) and CTMP (chemithermomechanical pulp) processes. It was concluded that the details concerning the chip process had a large impact on e.g. the energy consumption in both first stage and second stage refining. Results showing this are given in this thesis. / För både kemisk och mekanisk pappersmassa så tillverkas flis av trädstockar med hjälp av en skivhugg innan fibrerna separeras. För att göra flisningsprocessen mer effektiv, måste kopplingen mellan processparametrar och fliskvalitet studeras. Ett mål med denna avhandling är att ge fundamental kunskap om mekanismerna bakom bildandet av träflis. Både experimentella och analytiska/numeriska metoder har använts i detta arbete. De experimentella undersökningarna har gjorts med hjälp av egen utvecklad utrustning. Resultaten från den experimentella undersökningen visar att friktionen mellan stammen och flisningsverktyget har betydelse vid flisning. Vidare observerades det att inträngnings processen är approximativt självlik (self similar) och att det är spänningsfältet över hela sprickplanet som är kritiskt för bildandet av en flis. Den utvecklade analytiska modellen förutsäger normal- och skjuvspänningsfördelningen över sprickplanet och kan mer specifikt förutsäga den kompressiva belastning som verkar parallellt fiberriktningen i ett linjärt elastiskt och ortotropt material (trä). De analytiskt bestämda fördelningarna stämmer relativt väl överens med motsvarande fördelningar beräknad med finit element analys. För att kunna studera flisningsprocessen under realistiska förhållanden, vilket bl.a. betyder att skärhastigheter som är representativa för en verklig process skall användas, så utvecklades inom ramen för avhandlingsarbetet, en laboratoriesflishugg. Detaljer rörande flishuggen samt hur uppmätta lastsignaler skall utvärderas ges tillsammans med ett exmpel på hur kraften på skärverktyget (kniven) varierar under ett skärförlopp. Inverkan av en viss flisningsprocessparameter undersöktes genom att flis tillverkades i laboratorieflishuggen varefter de raffinerades i en pilotraffinör under förhållanden som var optimerade för TMP (termomekanisk massa) och CTMP (kemitermomekanisk massa) processerna. Det konstaterades att detaljer i flisningsprocessen hade stor inverkan på t.ex. energiåtgången i både första stegs – och andrastegsraffinering. Resultat som verifierar detta ges i avhandlingen. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Submitted. Paper 4: Submitted. Paper 5: Submitted.</p>