# Legacy documentation

## Operation modes

The legacy, MatLab-based routine offers two operation modes:

**Markovian chain-based sampling**Having a distinct household with several appliances, the time-discrete simulation proceeds in time and switches on different appliances based on the switch-on probabilities found in Markovian chains. If a device is switched on, a pre-defined time series for the appliance is replayed. The totality of time-series is then aggregated afterwards.

**Probability weighted aggregation**Once again, this mode is as well time-discrete. Based on the given switch-on probability, the rated apprent power of an appliance is weighted with this probability and the totality of consumed power is accumulated.

Refactoring note

For the refactoring, we will foremost focus on operation mode 1 and maybe add 2 later.

## Input Data

**Read probability**

Eingabeoberflaeche Wahrscheinlichkeit Leistung.xls

Loadprofile for appliance classes from “Synergy Potential of Smart Appliances” (2008)

Hourly switch-on-probability: $$p_{on,1h}$$ for appliance and season

Usage probability (days/year) $$p_{use}$$ for appliance and season

$$\underbrace {\begin{bmatrix} p_{1, spring} & \dots &p_{1,winter}\ \vdots &\vdots&\vdots \ p_{96, spring} & \dots &p_{96,winter}\end{bmatrix}}*{\mathbf{M*{Wk}}}= \frac {p_{ein,1h}} {4} \cdot p_{use}$$

**Read input interface**

Eingabeoberflaeche Haushalte.xls

Criteria: average household, number of residents, income, housetype

number of households to be simulated

number of appliaces according to criteria

**Define variables**

configuration parameters

timesteps per day

number of days to be simulated