Investigation of small hydroelectric power projects in remote areas of Sub-Saharan Africa: The case of Kimbau

K. Goutis, Investigation of small hydroelectric power projects in remote areas of Sub-Saharan Africa: The case of Kimbau, Diploma thesis, 281 pages, Department of Water Resources and Environmental Engineering – National Technical University of Athens, Athens, March 2013.



The thesis that follows is an overview of the social, technical and power problems of the most undeveloped countries of the third world, the countries of sub-Saharan Africa, a synopsis of the great opportunities of small hydropower and an analysis of the failure of a small hydroelectric project in Kimbau, a remote village in the Democratic Republic of Congo. The last part of the study is a continuation of the first effort to the one the Italian University of Bari started in 2009, to study, analyze and investigate the causes of failure of the small hydroelectric project in Kimbau, by remote technical assistance. The study here goes through multiple paths via a multidisciplinary approach. The first goal here is to study and represent carefully and obtain knowledge about the working, technical, economic and social environment of sub-Saharan Africa that the foreigner engineer confronts, combined with the identification of the distinctive features of the region and the aggravating factors that are holding back, in terms of general development and progress, both sub-Saharan Africa and the Democratic Republic of Congo. The second goal is to analyze the power sector, mainly comparing the extention of the national grid electricity capacity and the efforts in the electrification of remote areas of these countries through small hydropower projects. The third goal is to find and attribute the failure factors of the small hydroelectric project of Kimbau, where people were given as an act of help by AIFO (an Italian non government organization) a Francis water turbine which made the project of building a a stand-alone hydroelectric power system achievable. Up to date the power plant has not been working for a while because failure causes persisted. Here we supplement the analysis made by our Italian colleagues first attempt and we present the renewed results in the light of a broader technical approach. We started our technical analysis by the Italian report and the pictures of the field study made by the engineers of Kimbau. We enriched the results of the hydrological analysis by comparing the climatological and pluviometric data provided by the Congolese rain gauges near Kimbau so that we can estimate the uniformity in the distribution of the water discharge and the expected high values of the runoff coefficient. We were also in the position to deduce valuable conclusions about the stromatographic and geotechnical analysis as the nature of the upper ground is concerned. First we could estimate given their known relative density, the specific gravity of the loose sandstones, which we could incorporate to our analysis as related to the anchorage block. From the comparison of other geological data about the sedimentation transport we could correlate both the medium size of grain in the river Nzasi (the river of the microhydro plant) and its sedimentation transport for the analysis of the desander. And in the end we could also assure that due to the high porosity, the high gradients in contrast to the lower angle of repose, there must be a considerable pore pressure from the water that indicates a more or less, steady discharge throughout the year. We examined the operational efficiency of all elements of the hydroelectric power system, under various hypothetical scenarios of the flow field, except the weir and the headrace, both of which are typical to their operations and are not designed to assume the burden of the changes in the flow field or the sediment transport. By the Italian report, we already knew that there were some the plant failed due to detachment and rupture of the penstock twice, in sections that implied both concentration of supperpressions and water-hammer in action. This is valid for both the first penstock made of steel and the second one made of PVC. Furthermore, there were signs of constant interruption of flow, so one can assume that there was an undergoing mechanism or that was causing the valve to shut down and thus, resulting to a water hammer, or at least an accident that resulted that could explain why the first penstock failed to rupture. Also, the poor geological condition of the upper layer of sandstones which is gravely eroded explains the sinking of the anchorage block, so we were led to fully examine this hypothetical cause. In the light of these of events, we examined a series of parameters. First we examined the whole system in normal circumstances, meaning in normal flow field. There were found no pressure problems at this point. Secondly we examined the operation of the desander addressing the problem of the settling rate of the medium sized grain compared to the dimensions of the desander and the problem of the needed rate of wash out of the desander compared to the level of the sediment transportation. There were found no designing restraints in the dimensional aspects of the desander that could lead eventually to endanger the safety of the desanding process. At a third stage, we followed our effort by investigating any problems considering the theoretical aspect of a water hammer hitting the penstock, and what consequences could this had in the hydraulic subsystem of the forebay and the feed pipe. For the water hammer hypothesis, we distinguished two scenarios that can occur, that have to do with the time of stopping the water flow in the penstock. If the time of stopping the water flow is long compared to the propagation time for a pressure wave to travel the length of the pipe, then one should use the formula of Michaud – this hypothesis applies to the “normal” (the manufacturer’s) closing time of the actual valve of the Kimbau power plant. Instead, if by an accident (mechanical or electrical fault, human force by hand etc), something causes the valve or the guide vanes to close more abruptly, then one should use the formula of Allievi. By this analysis, we reached to the conclusion that the penstock is surely vunerable to rupture by an abrupt stoppage of the flow, but it can endure theoretically a pressure wave caused by a longer stoppage of the flow. The second situation however, progressively can cause failure to fatigue of the material of the pipe, because as we found out, the PVC pipe’s thickness is actually unadequate. Also, because of the fact that the Francis turbine is actually overdimentional to the power needs of the village, and the fact that assuming a steady discharge of water the turbine is actually working often near 40% of its capacities, it means that the useful load is often not high enough and it gets rejected so that the valve stops. Also another case to consider is that the power demand in the village may be too small or even nullified for many hours during the day, so this could cause continuous disruptions of the ater flow in the penstock. As for the hydraulic subsystem of the forebay and the feed pipe, we recalculated and reassured the results of our Italian colleagues, and we can confirm that its behavior presents no problem whatsoever. Finally, we checked the stability of the anchorage block in the possibility of toppling, sliding and ground fracturing, for various lengths of reference of the penstock that the particular block could support, and estimating roughly from picture material the size of the block, the gradients of the ground and the moment of forces for various positions inside the block. As a result, we were able to prove that the block is causing the ground to fracture in any possible combination of factors, mainly because it is inside the area of failure – to prove that we studied it as a single foundation. Also toppling is not out of the question, it depends on the eccentricity of the applied forces. As a conclusion, one should say that because of the gravity of erosion in the upper layer and its poor resistance, so that it cannot support a more dense positioning of too many blocks, above - ground installation of the penstock should be avoided and the penstock should be buried. In that basis, there is a number of factors contributing simultaneously to the ultimate stage of failure, so we propose a series of technical solutions that take measures both for the water hammer scenario and avoid uncertainties with the foundations. As a final proposition, we suggest investigating further the ground condition by a verification visit and to study thoroughly or make a better planning about the power demand / supply distribution system throughout the year.

PDF Full text (23906 KB)