Regarding market and competition analysis, thorough desktop analysis was performed for mapping the European and North American urban gas pipeline infrastructure along with their current conditions, risk for failure, the target customers, and their preferences within using system monitoring solutions. Also, focus group interviews were performed with gas grid operators (target customers) in Spain, Finland, and Estonia. The key conclusion is that the gas pipeline leakages are infer serious cost, and that new generation continuous monitoring solutions are much needed. The concept of the TM Field Analyzer as a 3D Transition Measurement Solution intended for urban pipelines that transport natural gas was treated with great attention, and all of the interviewed end-users are considering to offer their facilities for becoming pilot customers during the full scale development phase.
Regarding laboratory tests and analysis, an experimental campaign was carried out in a simple test environment. The main aim of the campaign was to verify the performance of advanced information-theoretic measures in leak detection with the collected data. The experimental design consisted of seven experiments five experiments during which a simulated leak occurred, and two reference experiments without a leak, representing “normal flow” –conditions. The reference experiments were run in the middle and at the end of the test set.
Two information-theoretic measures, Shannon’s entropy and Jaccard’s index, were tested for the leak detection. Additionally, the required data pre-processing methods were implemented. With both measures, the leaks could be detected accurately. However, the Jaccard’s index combined with wavelet denoising and applied in monitoring the changes in the dynamics of the system demonstrated robustness against false alarms, variation in test environment and the orientation of the sensor. Thus, it was selected for further studies. During those it was then observed that the index could be further combined to the statistical process control (SPC) methods to obtain alarm and control limits automatically. Also, in some cases the monitoring algorithm seemed to demonstrate predictive capability: the algorithm could alarm about the development of the leak before its actual occurrence.
To conclude, the information-theoretic measures, especially the Jaccard’s index with proper data pre-processing steps demonstrated robust performance in fault detection. The use of control charts for monitoring the index values seems to be a viable option to be used in the monitoring application. The chosen approach is non-parametric, real-time capable and requires no models. Therefore, it is expected to be easy to install and maintain and scalable also to large systems. Further testing and a full-scale development will be needed in a controlled laboratory environment and industrial-scale systems to fully develop the algorithm.
Regarding TM Field Analyzer concept development, marketing and commercialization strategy, given that the market and competition analysis confirmed the need for the solution, and also that the laboratory tests validated the technical feasibility of utilizing adanced signal processing, the commercialisation strategy was further defined. First, this entailed a full-scale Freedom To Operate analysis along with developing an IPR protection strategy. The results confirmed the uniqueness of the TM Field Analyzer concept. Also, a financial feasibility analysis was constructed along with constructing the financial forecasts subsequent to the full-scale development. According to the current assumptions, the TM Field Analyzer business line has a potential to generate 150 mEUR worth of revenue by year 5, and increase the number of new employees to roughly 100. Nevertheless, the critical barrier is herein the need to demonstrate the solution within as many end-user facilities as possible. Also, substantial development is needed to be performed in order to come out with a fully commercial solution. Thus, applying for Phase 2 financing is critical in order to continue the TM Field Analyzer development, and eventually reach commercialization.
