A Tour of ESP-r


ESP-r is a general purpose, multi-domain (building thermal, inter-zone air flow, intra-zone air movement, HVAC systems and electrical power flow) simulation environment which has been under development for three decades. It works on the principle of simulation follows description where additional technical domain solvers are invoked as the building and system descriptions evolve. ESP-r allows the user to define both the model resolution and extent of the model from the room level to a neighborhood, to address the specific needs of a project.

Explore the facilities and capabilities of ESP-r in the updated May 2018 document (an Italian translation is also available). At 315 pages this is more compact than the thousands of pages in manuals from other simulation tools. These pages provide highlights for the impatient. Select from the topics below to quickly scan for relevant information or drill-down for details.

A montage of images from various ESP-r models.
various model types


The big ideas about ESP-r can be found in this overview and a potted history highlights its evolution. ESP-r is distributed as a suite of applications. It can perform a range of simulation tasks, some of which are documented in these typical workflows. ESP-r is distributed with hundreds of exemplar models, some drawn from consulting projects. ESP-r has been deployed on computers ranging from a Raspberry Pi to servers, running multiple operating systems (see supported platforms).


ESP-r supports the near-simultaneous solution of zones, network mass flow, CFD and electrial domains. ESP-r uses numerous solution techniques across zonal, air flow, system and electrical power domains. ESP-r maintains (and reports on) energy balances throughout the model (at each zone, surface, and component), details of which may be found in the solver descriptions: the zone solver, the mass flow solver(air or water or mixed), the CFD solver, the system solver for dynamic system component models and the electrical power solver which deals with DC and AC electrical power distributions.


The art of creating models fit for purpose draws on our ability to select appropriate entities from within the simulation tool. In ESP-r the user is in control of the level of compositional resolution. Thus browsing the ESP-r data model is a good way to determine if the available entities will support your next project as well as improving the likelyhood of delivering information on-time and on-budget.

In ESP-r there are often multiple approaches to describing building and system entities. Depending on the information available, the resources available and the goals of the project there are a number of composition rules to guide users. For example, early in a design and when details for a component-based approach to environmental controls are not avaiable, users can select from a number of ideal environmental controls.

Ideally, the form and composition of buildings in simulation models are based on a finite set of well-understood entities which are thermophysically equivalent across a range of tools. To reduce ambiguities between entities in other simulation tool, users can browse the building entities and site entities available to represent the building site, the form and fabric of the built-environment as well as the furniture and fittings within buildings.

ESP-r is distributed with a range of databases for frequently-used materials, constructions, optical properties, weather, system components and even micotoxin growth patterns. Model-specific databases are also supported.

What happens inside the built-environment also is part of the ESP-r data model and schedule entities for user defined infiltration, inter-zone ventilation and casual gains can range from simplistic patterns to tracking minutely patterns recorded on site.

The creation and use of mass flow networks and electrical networks is predictated on the attributes and fluid/power flow characteristics of mass flow entities.

Other tools

ESP-r has a number of import and export facilities, for example, its facilities to drive Radiance to undertake a number of visual assessments. However, sharing data models with other tools sometimes requires careful planning of models to respect differences in the underlying data models.

Thermal Simulation

One of the core concepts of numerical simulation is how the solver derives and reports on the energy balances within the model. Another is the treatment conduction within buildng elements. The default regime uses a 1-D model but it is possible to describe 2-D and 3-D conduction (but almost no one does this). And convection regimes (there are dozens of options both for inside and outside faces of surfaces) is a topic of interest to many users. In ESP-r there are a variety of approaches to designing models to track both general and position-specific thermal comfort (we are still working on this section) :-(


Reporting of performance at many decimal places is false precision because many attributes of building models (thickness of construction layers, material properties, surface areas, heating capacities etc.) are uncertain. ESP-r provides a formal description of what is uncertain in the model, where these uncertanties are located in the model as well as a regime to use to explore uncertainties (Monti-Carlo, differential, factorial) as well as facilities in the res module to extract the impact of uncertainties. You can find out more HERE.


An overview of solar treatment will get you started. How close does simulation come to the evolving patterns we observe in and around buildings? It has much to do with how we plan our models and the level of resolution we choose to include. The performance of buildings is often dependent on how sunlight is transformed as it passes through facades and then is distributed within the building. The method of calculation and user directives controlling this might also be of intrest.


An overview of how environmental control systems are designed covers the range from ideal controls to user defined system component networks.

Infiltration, Ventilation, and Multizone Airflow

This overview of how air flow is treated in ESP-r describes the range of approaches taken by the program. Air which gets past the facade either via natural flows or mechanical flows is termed infiltration while flow between thermal zones is termed ventilation (whether natural or forced). Check out what air-flow networds actually look like.


At the end of the day, simulation is only useful if users can get access to indicators of performance. ESP-r differs from many simulation tools in its creation of performace data files and a bespoke application res which provides interactive access to performance data and facilities to analyse, extract and display such data.


The ESRU web site www.strath.ac.uk/esru has an extensive publications list (conference papers, thesis, and occassional papers).

How long does it take the kettle to boil

ESP-r's solvers are designed to be numerically efficient. Users have a lot of latitude to influence how long it takes the kettle to boil. If you want to have a look at performance data across a matrix of hardware, operating systems, model types and assessment goals and put this in context with the EnergyPlus suite, have a look here

Past and present staff of ESRU and ESP-r contributors

ESP-r is the result of the contributions from scores of people. This list includes many who have been associated with ESRU and ESP-r.

Common Issues for Noobs

ESP-r is a cutting edge software, which means sometimes the edge can bleed. When faced with any issue, our first advice is: DON'T PANIC. Then, look at this page, where we have tried to document the commonest newbie issues.

Back to top