Methods for predicting warp at a given condition
Methods are provided for predicting warp of a lumber product at a given condition, such as relative humidity, temperature, moisture content, or the like. Note that this methodology may apply to any of crook, bow, twist, cup, or combinations of these quantities. The method involves measuring an original warp of a first wood product at a known condition; changing the condition of the first wood product to a new condition; measuring a second warp of the first wood product at the new condition; developing a prediction model based on data from the original warp and condition and the second warp; measuring an original warp and condition of the given wood product; and inputting data from the original warp and condition of the given wood product to the prediction model to determine warp at a given condition.
1. A method for sorting two or more given wood products, the method comprising the steps of:
measuring original warp of at least one first wood product at a known or measured moisture content;
changing the moisture content of the at least one first wood product to a new moisture content;
measuring a second warp of the at least one first wood product at the new moisture content;
developing a prediction algorithm based on data from the original warp and the second warp and at least one of the original and new moisture contents;
measuring an original warp and moisture content for each of the two or more given wood products;
inputting data from the original warp and at least one of the original and given moisture contents from each of the two or more given wood products to the prediction algorithm to predict a predicted warp for each of the two or more given wood products at a given moisture content; and
sorting the two or more given wood products based on the predicted warp at the given moisture content.
2. The method of claim 1 wherein the original warp is measured based on one or more inputs and/or the original moisture content is measured based on one or more inputs.
3. The method of claim 2 wherein the inputs are selected from a group consisting of: moisture content measurement, camera imaging, laser range finding, acoustic range finding, mechanical displacement, electrical property measurement, infra-red or near infra-red measurement, acousto-ultrasonic property measurement, shape measurement, weight, nuclear magnetic resonance measurement, temperature, and microwave measurement.
4. The method of claim 1 wherein the prediction model is derived using at least one of the following techniques: regression trees, classification trees, linear discriminant analysis, quadratic discriminant analysis, logistic regression, Partial Least Squares and neural networks, linear regression, non-linear regression, generalized linear regression, generalized additive regression, projection pursuit regression, or look-up tables.
5. The method of claim 1 wherein the warp is characterized quantitatively.
6. The method of claim 1 wherein the warp is characterized qualitatively.
7. The method of claim 1 wherein the warp is at least one of: crook, bow, cup, and twist.
8. A method for sorting two or more given wood products, the method comprising the steps of:
measuring original warp of at least one first wood product, wherein the at least one first wood product has at least two known or measured conditions, the at least two known or measured conditions being selected from: moisture content, relative humidity and temperature;
changing at least one of the at least two known conditions of the at least one first wood product to a new condition;
measuring a second warp of the at least one first wood product at the new condition;
developing a prediction algorithm based on data from the original warp and the second warp and at least one of the original and new conditions;
measuring an original warp and condition for each of the two or more given wood products;
inputting data from the original warp and at least one of the original and given conditions from each of the two or more given wood products to the prediction algorithm to predict a predicted warp for each of the two or more given wood products at a given condition; and
sorting the given wood product based on the predicted warp at the given condition.
9. The method of claim 8 wherein the original warp is measured based on one or more inputs and/or the original moisture content is measured based on one or more inputs.
10. The method of claim 9 wherein the inputs are selected from a group consisting of: moisture content measurement, camera imaging, laser range finding, acoustic range finding, mechanical displacement, electrical property measurement, infra-red or near infra-red measurement, acousto-ultrasonic property measurement, shape measurement, weight, nuclear magnetic resonance measurement, temperature, and microwave measurement.
11. The method of claim 8 wherein the prediction algorithm is derived using at least one of the following techniques: regression trees, classification trees, linear discriminant analysis, quadratic discriminant analysis, logistic regression, Partial Least Squares and neural networks, linear regression, non-linear regression, generalized linear regression, generalized additive regression, projection pursuit regression, or look-up tables.
12. The method of claim 8 wherein the warp is characterized quantitatively.
13. The method of claim 8 wherein the warp is characterized qualitatively.
14. The method of claim 8 wherein the warp is at least one of: crook, bow, cup, and twist.