2. Maximization/Minimization using Lagrange Multipliers

Having obtained the expression for the total shear component of traction, this needs to be maximized/minimized. As n₁, n₂, and n₃ are not independent of each other, we again use the method of Lagrange multipliers. So, we define a function V as given below: (4)

In this step, we are looking to optimize the shear component of traction we just derived. However, we cannot do this straightforwardly since n₁, n₂, and n₃ (components of the normal vector) rely on each other. Therefore, we use a mathematical technique known as Lagrange multipliers to help us maximize or minimize the function we have defined, denoted as V. This technique allows us to handle situations where some variables depend on others, making it suitable for our problem.

Think of it like planning a trip where you want to visit certain places (traffic rules) and you have to account for the distance between them. You want to maximize your enjoyment (function V), but your route options overlap (n₁, n₂, n₃) and you can't take all paths independently. Lagrange multipliers help you find the best route under the complex conditions.

We’re using α to denote the Lagrange multiplier here because we already have λ’s for principal stress components. This is the function that has to be maximized/minimized with respect to the 4 unknowns: n₁, n₂, n₃, and α. Let us take the derivative of V with respect to these four unknowns starting with the kth component of normal vector (nₖ): (5)

Here, we compute the derivative of our function V concerning the unknown variables: n₁, n₂, n₃, and α. By taking these derivatives, we can find critical points—values where the function is either at a maximum or minimum. This process helps us identify the conditions under which our shear component of traction is optimized. The use of the Kronecker delta function in our results allows us to simplify these derivatives significantly.

Imagine you're adjusting the settings on a machine to achieve the best performance. Each knob (n₁, n₂, n₃, α) needs to be fine-tuned. By observing how small adjustments impact the overall performance (taking derivatives), you can find the optimal settings that yield the best results.

In each of these equations, one among the two terms multiplied has to be zero for the product to be zero. There are multiple solutions to this problem and all solutions can be found by considering different cases. If suppose n₁ = 0 in the first equation, n₂ = 0 in the second, and n₃ = 0 in the third, we get a trivial solution for n₁, n₂, and n₃ but that will not give us a valid direction as the magnitude of the direction vector will not be 1. So, we take the first term in the first equation and the second terms in the other two equations to be zero. So, we have the following equations at hand now: (10, 11, 12)

This step involves analyzing the system of equations generated from our derivatives. We find that at least one term in each equation must equal zero for the product to resolve to zero, helping us eliminate possibilities to find our solution. We note that if we assume variables can equal zero, it leads us to trivial and non-useful solutions. Instead, we strategically select terms to zero in a way that allows us to derive meaningful non-trivial solutions.

Consider a group of friends deciding what movie to watch. If everyone says 'I don’t like that genre' (setting terms to zero), they might end up with no options at all. Instead, they should explore combinations (selecting useful terms) that lead to exciting choices, ensuring a fun movie night!

Equations (11) and (12) have been obtained by substitution of n₁=0. We need to find n₂ and n₃ now. To eliminate α, we subtract (12) from (11) to get (13)

Once we have established possible non-trivial settings for n₁, we look for values for n₂ and n₃ using algebraic manipulation. By subtracting one equation from the other, we are essentially eliminating the Lagrange multiplier α from our equations, allowing us to isolate the variables we want to solve. This moves us closer to the exact conditions fulfilling our maximization criteria.

Think of it like a puzzle; once some pieces are placed correctly (n₁), you can determine what pieces fit best in the other spaces (n₂ and n₃). By focusing on the remaining pieces after removing unnecessary ones, you can achieve a complete picture of your solution.

Thus, we get the values of n₁, n₂, and n₃ as (15). The above equation gives us four solutions for the direction as n₂ and n₃ both can take two values independently. But these are not the only solutions. In equations (7), (8), and (9), if we would have assumed either n₁=0 or n₃=0 instead of n₂=0, our analysis would essentially have remained exactly similar leading to four solutions each given by: (16, 17)

After processing our equations, we summarize our findings for the normal vectors n₁, n₂, and n₃ that give us the different directions where the shear component of traction is maximized. In total, we have 12 unique combinations of solutions from our analysis, each significant in understanding how the shear component behaves relative to principal stress components.

Imagine trying to figure out how to optimize your budget for a shopping spree. You have main categories (n₁, n₂, n₃) but within each category, there are alternatives that can change the total balance. You find multiple ways to achieve a fantastic shopping experience (12 unique solutions) that fit within your constraints (the stress values).

--