An Approach to Weight-Based Battle AI

There are a number of ways to implement combatant AI in RPG battle games such as Dragon Quest or Pokemon. However, one simple approach is to develop a list of probabilities for all actions that a combatant can take, then draw a random action based on these probabilities. In some games, such probability lists may have been explicitly specified by designers and remained static regardless of the actual state of the battle. While in other games, more advanced logic, perhaps even incorporating combo attacks over the course of turns, may have been employed. Whatever the approach, the number of unique takes on battle AI throughout games is wide and varied. Ultimately, the implementation perhaps comes down to what features the game developers felt was most important for the gameplay, as well as what was technically and economically feasible at the time.

MNIST Fashion Classification

The MNIST fashion dataset is a popular dataset containing grayscale 28x28 pixel images of fashion items, such as shirts, shoes, and pants. This post explores the use of this dataset to train two neural network models in the identification of these garments.

MNIST Handwritten Digit Classification

The MNIST handwritten digit dataset is a popular dataset containing grayscale 28x28 pixel images of handwritten digits. This post explores the use of this dataset to train two neural network models in the identification of handwritten digits.

Transmission Line Structure LiDAR Comparison

Automated methods for comparing transmission line structure LiDAR open the way for a number of opportunities in the identification of trouble areas for designers to focus on. This post will explore an approach for separating structure point cloud clusters, pairing structure point clusters between data sets taken at different times, and comparing cluster pairs using the Hausdorrf distance to determine whether they significantly differ.

Steel Member Tensile Strength: Equivalent Number of Bolts for Net Area Calculations

Structural software, such as PLS-Tower, sometimes provides a field for the number of bolts to remove from the gross area of a steel member cross section for tensile strength calculations. For scenarios when bolts at a connection are not staggered, specifying this value is rather trivial. However, for scenarios when bolts are staggered, the $ s^2 / 4g $ factor specified by §D3.2 of [1] must be incorporated into the input value.

This post presents the equations for calculating the equivalent number of bolts incorporating the $ s^2 / 4g $ value. In addition, to reduce the number of failure paths that must be considered for complex bolt patterns, an equation for the critical pitch spacing, beyond which a staggered tensile failure path will not occur, is presented.

DBSCAN Clustering of Transmission Line Phase LiDAR

Sometimes utility standards or PLS-CADD graphical sagging requirements necessitate that the different transmission line phases be divided into different feature codes. However, depending on who processes the LiDAR data, the LiDAR received from transmission line surveys may or may not include the phases on different feature codes. Therefore, in these scenarios, it falls on the designer to separate the data themselves. In the event of large data sets or complex phase rolls, this process may be tedious and time consuming.

In order to determine some possible form of automation for this problem, I looked into the available clustering algorithms in the scikit-learn Python library. Based on my findings, the DBSCAN model seems to work for these purposes and is quite performant in comparison to some of the other clustering models that the library contains.

This post provides a sample of how the DBSCAN model can be leveraged to separate different wire phases within transmission line LiDAR.

Transmission Line 2D EMF Calculations

In the United States, the magnetic and electric fields of transmission lines are not nationally regulated. However, some states have their own set of regulations as to their maximum values within the right-of-way and/or at the edge of the right-of-way. For example, in the state of Florida, for new 500kV or greater transmission lines, the maximum electric and magnetic fields at the edge of the right-of-way, 1 meter above ground, are 5.5 kV/m and 250 mG, respectively. In addition, the maximum electric field within the right-of-way is 15 kV/m at 1 meter above the ground. To provide a clearer picture of what goes into these calculations, this post breaks down the steps necessary to calculate the magnetic and electric fields of a transmission line using 2D calculations.