Continuously optimizing efficient energy infrastructures

← Cases
Westland Infra Netbeheer

As a regional network operator, Westland Infra Netbeheer ensures efficient energy infrastructures with a high degree of safety and availability. They also support growth in the field of sustainable use of energy and energy transition. In this area they are constantly looking for innovative techniques to keep the quality of their services and insight and control of energy distribution as safe as possible.

Background

As a regional network operator, Westland Infra Netbeheer ensures efficient energy infrastructures with a high degree of safety and availability. They also support growth in the field of sustainable use of energy and energy transition. In this area they are constantly looking for innovative techniques to keep the quality of their services and insight and control of energy distribution as safe as possible.

What did we do?

West IT Solutions and Westland Infra Netbeheer have been working together for a long time already in optimizing energy consumption. For example, West developed a web portal in 2006, which made it possible for market gardeners to quickly respond to changes in the energy sector and have optimum control over their own energy consumption and costs. By this web portal, gardeners can set their boiler buffers to produce heat when energy is cheap. Gardeners can store this heat and then use it when energy is expensive. The desired energy transport capacity can also be selected per hour; all this accompanied by statistical information relevant to growers.

Westland Infra Netbeheer remains keen on quality and innovations for their services and customers. Westland Infra Netbeheer and West have recently taken new steps in optimizing their electricity grid. A project was carried out in 2018 to provide insights into individual power consumption caps and to exonerate them in certain circumstances. High-volume business customers have contracted maximum power that they can use (per 15-minute time interval), even though their physical connection would allow higher consumption. These individual restrictions should protect the supplier's infrastructure against a demand overload when all users simultaneously consume a large amount of electricity.

In this project, different data models are combined to provide insights into three repetitive patterns (annual cycle, weekly cycle and 24-hour cycle) in power consumption and to make them analyzable. Now the demand peak can be predicted and it can be determined at which moment a certain group of customers can be offered extra energy consumption.

Technology

The 2006 project already contained state-of-the-art web technology: portal and application servers, SOAP and XML. In addition, an attempt was made to use of standard components as much as possible. In order to increase operational reliability, the entire system has been implemented twice. The business logic, implemented in a Windows NT environment at Westland Infra Netbeheer, runs both in the computercenter in Poeldijk and in the fall-back center in Westerlee. The application servers are placed in the computercenter of Westland Infra Netbeheer and in the hosting center of West.

The application server that was used, running under Sun Solaris, is based on the open-source software product Jakarta Tomcat. The various functionalities were built as web applications that use Java servlets and JavaServer Pages (JSPs). These collect data via SOAP (XML over HTTP) at Westland Energie.

A SOAP cache was built to increase the scalability of the solution and to exonerate the back office. Frequently occurring SOAP calls can therefore be answered without having to consult the back office. The most recent project uses various modern data models and analysis methods. For example, we started with a forecasting method with which we could remove the seasonal components "manually" and then use a standard autoregression model (Arima) to train the model for local effects, predict the coming hours and finally add the seasonal components again.

After that, to record the changing daily pattern, the average daily pattern was calculated per month. In this way the average daily pattern of the corresponding month could be subtracted. The autoregression model then expresses the expected value of the next value as a linear combination of the previous one. During the training phase the optimal weights are estimated by regression. In the prediction phase, the future values ​​are iteratively calculated based on these weights.

Result

The web portal has reinforced the relationship with Westland Energie and its business customers. Only a year after the start of the web portal, about half of all contracts were concluded by gardeners via the portal. In addition, the portal provides services with unique added value. In a time of increasing market forces in the energy sector, adding such value to a product is of strategic importance.

Met behulp van de analyse methoden die gebruikt zijn in het meest recente project, kan Westland Infra Netbeheer de individuele stroomverbruik caps in bepaalde omstandigheden ontlasten. Deze omstandigheden zijn meestal de uren vóór en na de stroompieken in de ochtend. Op deze manier verwacht Westland Infra Netbeheer dat ze hun industriële klanten kunnen overtuigen om hun vraag te spreiden, wat de noodzaak om hun netwerk te upgraden zou verminderen - een dure operatie. Tegelijkertijd stelt het de klant ook in staat om af en toe zijn gecontracteerde maximale stroombehoefte te overschrijden zonder extra kosten, dus het is een win-win situatie.

Using the analysis methods used in the most recent project, Westland Infra Netbeheer can exonerate individual power consumption caps in certain circumstances. These conditions are usually the hours before and after the morning surges. In this way, Westland Infra Netbeheer expects to be able to convince industrial customers to spread their demand, which would reduce the need to upgrade their network - an expensive operation. At the same time, it also allows the customer to occasionally exceed his contracted maximum power requirement without additional costs, so it is a win-win situation.