A 851Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE
Vegetation management is an important issue for the power distribution operators and for the whole electricity-dependent society. This topic surfaces after every storm that causes power outages. Since the overhead power lines are the most economic method for transmitting electricity it has been widely selected as a main transmission and distribution method.
With our society becoming more dependent on uninterrupted electricity distribution the need for more effective vegetation management has risen. While the transmission technology has developed quite rapidly over the past 60 years, the vegetation management techniques have stayed mostly the same. The cost of both vegetation management planning and work is substantial and this increases the pressure for vegetation management to be targeted efficiently and accurately.
In the past LIDAR detection algorithms were mostly designed for point clouds that were created from relatively high altitude. Those algorithms were mostly area and surface model based. In vegetation management there is a need for a different approach to object detection. The overhead lines that are LIDAR scanned from relatively low altitude create vastly more accurate point cloud data. This enables a completely new kind of LIDAR analysis algorithms.
To achieve the full benefits from LIDAR scanning the electricity distribution network needs to be modeled as an object based system for the electricity distribution operators to use the information efficiently in their processes. The traditional classifying point cloud data to several classes does not alone meet the requirements.
Traditionally LIDAR scanning has been roughly divided into two categories: the large area scanning done mostly from high altitude using airplanes; and precision scanning of objects done mostly using stationary scanners. The point clouds produced from high altitude aren’t precise enough for object recognition-based LIDAR analysis and although the point clouds created with stationary scanners can surely be used for automatic object recognition-based LIDAR analysis the data density from stationary scanners usually aren’t high enough to justify use of automatic recognition. Also the interest in most precision scanning is in defining exact details of the scanned object where automatic recognition has very few uses.
Power transmission line scanning is unique from a LIDAR scanning point of view in that it has very large amounts of relatively precise point clouds. These relatively unique characteristics have made possible completely new, objectbased LIDAR data analysis.
As even a smaller local distribution company can have tens of thousands of kilometers of overhead distribution lines it becomes essential to understand the connection between the asset data and observations from LIDAR. All observations from the vegetation analysis must be automatically and accurately linked to the existing asset data in order it to be useful for the distribution company. The LIDAR data alone–whether classified or nor–does not add value to the distribution company’s workflow. What matters are what actions can be planned based on this information and how well optimized the actions are given the real need in the field.
Object based LIDAR data analysis enables distribution companies to know each tree that can either fall on the overhead line or possibly touch the line in severe weather. The amount of vegetation observation data can be overwhelming, which emphasizes the actual problem of the distribution companies: which overhead lines should they prioritize for maintenance? Just knowing there is a problem is not enough. An optimized action plan is needed to eliminate those hazards where the reliability of distribution is at most risk.
Object-based LIDAR data analysis differs vastly from the traditional LIDAR data classification and analysis. LIDAR is a cost efficient method for observing the situation in the field and when combined with the logistic planning, business rules and budgeting can help to solve the equation of efficient vegetation management.
When the analysis of LIDAR for vegetation management planning is automated, there are specific requirements for LIDAR surveys. On one hand, the density and accuracy of the LIDAR needs to be at very high level–as individual overhead conductors need to be accurately detected. On the other hand, the process can be more relaxed and certain steps can be skipped in comparison with traditional LIDAR survey process.
This task of optimization can be daunting if done manually, and often the distribution operators revert to traditional area rotation method. This decades old manual planning method does not help to target the vegetation management effort where the effect on reliability would be the highest.
An automatic vegetation management planning solution is required, which uses the precise locations of each individual risk tree and creates logistically optimized plans for vegetation management using the precise locations of risk trees weighting the observations based on their actual risk level.
Changing to the optimized vegetation management is a process change. In order to execute the change successfully the results must be measured. The distribution operators need to understand how much the benefit of the change is. This is where LiDAR-based vegetation management solution needs to support the decision-makers. The optimized solution must provide understandable metrics based on measured facts and data and estimate the financial benefits of the optimized plan in comparison with the legacy method. Sharper Shape enables methods to validate the actual benefits by measuring the legacy methods against the new more economically optimized methods.
The benefits of smart vegetation management can vary according to the particular circumstances of a power distribution company. The statistics show that the vegetation management plans created by Sharper Shape solutions are 3-5 times more efficient in comparison to the vegetation management that would be done on area-based only method. The benefits of targeted and optimized vegetation management arise from weighting the different observations with the severity and estimated costs of the power outage, and the cost of vegetation management work. This is a unique capability which was not been available using the traditional area-based methods.
The object-based tree models also enable extremely accurate growth models for each individual tree. This information can be used to predict the growth of the vegetation. When the vegetation growth can be forecasted, the frequency of aerial LIDAR surveys can be optimized and targeted to areas where fresh information is needed. This enables the allocation of LIDAR survey resources where they benefit most. At the same time, the vegetation growth models are beneficial to predict the future need for vegetation management.
The object model-based approach enables extremely precise vegetation management maps. Because the solution reveals each individual tree that needs to be removed it enables new kind of vegetation management. Previously the distribution companies typically instructed their field personnel or their contractors to cut trees in certain overhead lines area. The field personnel make their own decisions in the field as to what trees needed to be trimmed or cut. The results would vary and are suboptimal.
The object model-based LIDAR analysis enables the operator to individualize each tree that needs to be cut down or trimmed. For other purposes a statistical summary of the trees can be used including the number of trees per kilometer/mile, average size of the trees to be cut, maximum height of branches that needing cutting, and further recommend the vegetation management method such as manual, forestry machine or branch trimming from helicopter.
The object model-based approach and knowledge of each individual tree that needs cutting can enable the usage of GPS-based mobile systems and applications that can guide field personnel to each observation site. The mobile tools can also assist in documenting the vegetation management results–right at the site of work. This way the distribution company can also track and ensure the vegetation management work was properly done.
Sharper Shape has created fully automatic LIDAR data analysis and vegetation management planning from raw LIDAR data and network information system data. It enables automatic vegetation management plans that take into consideration logistic issues and severity of each individual tree that is too close to an overhead line. It also enables a new kind of vegetation management where all field actions can be accurately planned beforehand and executed with the help of solution that targets and documents the field personnel’s work with utmost accuracy.
Ville Koivuranta has worked several years with Fortum, the largest Northern European power company. He now works for Sharper Shape as a senior technical specialist and consultant for leading northern European power companies.
A 851Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE