analysis and synthesis of data about load shedding

A Load-Shedding Algorithm for Power System Based on Transfer Learning

• Conference paper
• First Online: 24 January 2021
• Cite this conference paper

• Bian Daokuan   ORCID: orcid.org/0000-0001-6725-6578 37 ,
• Wang yong   ORCID: orcid.org/0000-0001-7469-5018 37 &
• Yang Ning   ORCID: orcid.org/0000-0002-0606-4625 38  

Part of the book series: Lecture Notes in Electrical Engineering ((LNEE,volume 718))

Included in the following conference series:

• The Purple Mountain Forum on Smart Grid Protection and Control

474 Accesses

In power system fault consequence analysis based on intelligent algorithms, usually set the load removal, as well as load-shedding as the main indicator. In general, it is necessary to use many fault samples generated offline to train the model, which can achieve good results under the support of a certain amount of data and training algebra. However, when the topological structure or operation mode of system changes, the accuracy of the trained model will decrease significantly. To solve this problem, transfer learning is introduced and a fast transfer adaptation method for the trained model is proposed based on the convolutional neural network. Firstly, using the data set obtained by the Monte Carlo simulation method to train the pre-trained model. When the structure of the system changes, the network structure of the pre-trained model is kept unchanged, and some network parameters are transferred to the new model to adapt to the system after changes. Finally, IEEE-RTS79 system was used to verify that the transfer learning method has better adaptability to the system with structural changes, and can effectively update the model and reduce the training time of the new model.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Xue, Y.: Integrated defense evolved from accidental failure to power disaster – warning of “8 · 14” blackout in North America. Autom. Electr. Power Syst. 27 (18), 1–5 (2003). (in Chinese)

Google Scholar  

Billinton, R.: Composite system reliability evaluation. IEEE Trans. Power Apparatus Syst. 88 (4), 276–281 (1969)

Article   Google Scholar  

Billinton, R., Allan, R.N.: Reliability Evaluation of Power Systems. Plenum Press, New York and London (1996)

Book   Google Scholar  

Song, H., Junyong, W., Linfeng, W.: Dynamic partitioning and autonomous disassembly strategy in power system emergency. Trans. China Electrotech. Soc. 27 , 224–230 (2012). (in Chinese)

Wang, B., Fang, W., Luo, X.: A fast algorithm for optimal cutting machine load cutting scheme under emergency control. Power Syst. Technol. 35 , 82–87 (2011). (in Chinese)

Sun, D., Zhou, H., Xiong, H., Ping Ju, Y., Zhao, J., et al.: Emergency load cutting control optimization method for DC receiver system based on sensitivity analysis. Proc. CSEE (38), 67–75 (2008). (in Chinese)

Yi, H., Wang, X., Teng, Y.: Optimal load shedding scheme for AC/DC hybrid receiving-end power grid after UHVDC blocking. Autom. Electr. Power Syst. 42 (22), 98–108 (2018). (in Chinese)

Lei, S., Sun, C., Zhou, Q., Zhang, X., Cheng, Q.: Based on RBF neural network and adaptive neural fuzzy system of electric power short-term load forecasting method. Proc. CSEE 22 , 81–85 (2005). (in Chinese)

Wang, H., Yang, D., Zhou, B.: Fault diagnosis of multi-terminal HVDC transmission line based on parallel convolutional neural network. Autom. Electr. Power Syst. 44 (12), 84–95 (2020). (in Chinese)

Jun, L., Chen, Z., Gong, G., Zhiqiang, X., Qi, B.: Classification and analysis method of residential electricity use behavior based on extreme learning machine. Autom. Electr. Power Syst. 143 , 97–104 (2019). (in Chinese)

Li, C., Li, H., Liu, Y.: Intelligent assessment system for dynamic security risk of large-scale power grid. Autom. Electr. Power Syst. 43 (22), 67–75 (2019). (in Chinese)

Zhu, Y., Liu, Y.: Fast search for high-risk cascading failures based on deep learning DC blocking judgment. Autom. Electr. Power Syst. 43 (22), 59–67 (2019). (in Chinese)

Wang, Y., Wang, Y., Ding, Y., Zhou, Y., Zhang, Z.: A fast load-shedding algorithm for power system based on artificial neural network. In: 2019 International Conference on IC Design and Technology (ICICDT), SUZHOU, China, pp. 1–3 (2019)

Liu, W., Zhang, D., Hou, X., Liu, L.: A decision making strategy for generating unit tripping under emergency circumstances based on deep reinforcement learning. Proc. CSEE 38 (1), 109–119+347 (2018). (in Chinese)

Zhang, C., Wang, Q., Liu, Z.: Pulmonary nodule auxiliary diagnosis method based on deep transfer learning. Comput. Eng. 46 , 271–278 (2020). (in Chinese)

Wang, J., Huang, H., Xu, H.: Low-resource Uyghur speech recognition based on transfer learning. Comput. Eng. (44), 281–285 + 291 (2018). (in Chinese)

Chen, J., Du, W., Wang, H.: A Method of locating the power system subsynchronous oscillation source unit with grid-connected PMSG using deep TRANSFER learning. Trans. China Electrotech. Soc. 1–12 (2020). (in Chinese)

Wang, Y.: Power Data Mining Technology Based on Deep Learning and Transfer Learning. Zhejiang University (2019). (in Chinese)

Lecun, Y., Bottou, L., Bengio, Y., et al.: Gradient-based learning applied to document recognition. Proc. IEEE 86 , 2278–2324 (1998)

Krizhevsky, A, Sutskever, I., Hinton, G.E.: Image net classification with deep convolutional neural networks. In: The 25th International Conference on Neural Information Processing systems, pp. 1097–1105. Curran Associates Inc., Lake Tahoe (2012)

Parkhi, O.M., Vedaldi, A., Zisserman, A.: Deep face recognition. In: British Machine Vision Conference, pp. 41–42. British Machine Vision Association, Swansea (2015)

Szegedy, C., Liu, W., Jia, Y.Q., et al.: Going deeper with convolutions. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9. IEEE, Boston (2015)

Li, Y., Hao, Z., Lei, H.: Survey of convolutional neural network. J. Comput. Appl. 36 (9), 2508–2515 (2016). (in Chinese)

Pan, S.J., Yang, Q.: A survey on transfer learning. IEEE Trans. Knowl. Data Eng. 22 (10), 1345–1359 (2010)

Download references

Author information

Authors and affiliations.

School of Electrical Engineering, Shandong University, Jinan, 250061, Shandong, China

Bian Daokuan & Wang yong

Huaneng Shandong Power Generation Co. Ltd., Jinan, 250014, China

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Bian Daokuan .

Editor information

Editors and affiliations.

NARI Group Co., Ltd., Nanjing, Jiangsu, China

Yusheng Xue

Yuping Zheng

School of EECS, Washington State University, Pullman, WA, USA

Rights and permissions

Reprints and permissions

Copyright information

© 2021 Springer Nature Singapore Pte Ltd.

About this paper

Cite this paper.

Daokuan, B., yong, W., Ning, Y. (2021). A Load-Shedding Algorithm for Power System Based on Transfer Learning. In: Xue, Y., Zheng, Y., Bose, A. (eds) Proceedings of 2020 International Top-Level Forum on Engineering Science and Technology Development Strategy and The 5th PURPLE MOUNTAIN FORUM (PMF2020). PMF 2020. Lecture Notes in Electrical Engineering, vol 718. Springer, Singapore. https://doi.org/10.1007/978-981-15-9746-6_46

Download citation

DOI : https://doi.org/10.1007/978-981-15-9746-6_46

Published : 24 January 2021

Publisher Name : Springer, Singapore

Print ISBN : 978-981-15-9745-9

Online ISBN : 978-981-15-9746-6

eBook Packages : Energy Energy (R0)

Share this paper

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Features/Analysis
• Transmission and Distribution
• Climate Change
• Battery Energy Storage
• Carbon Capture
• Research and Development
• Energy Efficiency
• Artificial Intelligence
• Internet of Things
• Smart Grids
• Future Energy
• Smart Mobility
• Southern Africa
• East Africa
• West Africa
• North Africa
• Central Africa
• International
• Women in Energy
• Submit content
• Press Releases
• Industry Insights
• Conference Papers
• Tenders library
• Subscribe to newsletter
• Subscribe to magazine
• Company Showcase

South Africa has been plagued by loadshedding since 2007, and 2021 was the worst year yet, with 1,165 hours of the year’s 8,760 lost. Though, to be fair, on 8 July 2022 the country was sitting at 1,008 hours of loadshedding so 2021 might be losing that title fairly soon.

UCT’s Energy System Research Group worked on a two-part feasibility strategy which they called Resolving the Power Crises. Part A: Insights from 2021 – SA’s Worst Load Shedding Year so far provides the analysis, while Part B: An Achievable Game Plan to end load shedding sets out proposals for how Eskom could get around loadshedding.

Instead of creating a projection proposing what the situation could be like if more energy is added to the grid going forward, the research group analysed what the impact would have been on loadshedding last year if there had been more generation capacity on the system, using Eskom’s own data, thus creating an empirical basis for their strategy.

Have you read? Ways already exist for DMRE to expand renewable energy onto the grid

Counterintuitively they discovered that rather than increasing system risk, the data shows that adding more wind and solar power to the existing distressed South African power system would have reduced loadshedding and increased system reliability.

“Empirical evidence demonstrates that an additional 5GW of renewable capacity would have essentially solved loadshedding in 2021 whilst enabling better and more efficient operation of our power system – all at a cost-saving to Eskom,” reads the report.

Adding 5GW of RE changes the loadshedding game

The report posits that adding 5GW of wind and solar – the approximate capacity of two REIPPP programme bidding windows – in the same proportion as the current installed capacity, would have allowed Eskom to eliminate 96.5% of loadshedding in 2021.

The additional wind and solar capacity would also then have reduced the amount of diesel used in the Open Cycle Gas Turbine peaker plants by more than 70%. More optimal use of Eskom’s pumped storage assets would also then have been possible, creating a further diesel saving.

“We find that the remaining fraction (3.5%) of loadshedding could have been eliminated by a modest expansion of Eskom’s ILS3 demand response programme or other aggregated demand response intervention. The very last few hours of loadshedding could be mitigated by 2GW of one-hour batteries,” declares the report.

Analysis of the cost impact of adding 5GW of renewable energy to the system also yielded a surprising result. Based on Eskom’s own 2020/1 financial year, dispatch cost of OCGT and coal power, and assuming renewable energy prices to be around R0.86kWH6, adding the renewable energy to the grid would have saved Eskom R2.5 billion.

That cost-saving doesn’t reflect the economic benefit to the country of avoiding loadshedding, it is the net cash saving Eskom could accrue because it would burn less diesel. “The saving is after provision for a hypothetical R6.08/kWh incentive for participating customers to reduce load under a demand response programmes (this equates to a 100% premium over the cost of running existing OCGTs), and the cost of 2GW of batteries,” the report qualified.

Have you read? Eskom agrees to wage deal but loadshedding energy continues

Resolving the crises

The second half of the report tries to demonstrate that containing, reducing and resolving loadshedding is possible. It demonstrates thought, that the possibility of achieving this with the current set of policy and procurement measures is low.

It then lays out the nature and extent of a suit of interventions that could help, mapping out a potential resource plan and game plant to implement said plan.

Resolving the power crises Part B: an Achievable Game Plan to end Load Shedding Part B of the report generally aligns with the National Planning Commission’s proposals, in particular removing the 100MW ceiling and local content requirements in the short term.

It also proposes that a ministerial determination be issued for the remaining capacity catered for in the current IRP2019 (13,600MW) by expanding Bid Window 6 and supports the updating of what it refers to as the now “outdated” Integrated Resource Plan.

Of reference Opportunities to expand the renewable energy market do exist

In addition to support Eskom’s continued maintenance schedule, Plan B also suggests investigating how to:

• Increase the likelihood that projects in the existing IPP Office procurement rounds can close;
• Increase incentives to ramp up renewable energy projects in the <1MW and 100MW market category;
• Use the existing and new projects to add even more additional energy to the grid;and
• Install additional thermal peaking capacity and expand diesel storage at existing peakers. ESI

You can read online

Resolving The Power Crisis Part A: Insights from 2021 – SA’s Worst Load Shedding Year Thus Far

Resolving The Power Crisis Part B: An Achievable Game Plan To End Load Shedding

*The reports were authored by Dr Grové Steyn, Dr Peter Klein, Adam Roff, Celeste Renaud, Lonwabo Mgoduso and Rian Brand

• demand response programmes
• Loadshedding
• power generation
• research and development
• South Africa

Share this article

RELATED ARTICLES

Africa energy bank: one ‘historic’ step closer to opening its doors, nigeria: government debt a threat to energy supply, say gencos, hybrid solar pv mini-grid with distribution line installed in chad, latest features, low-cost cooling solutions to support sustainability despite climate change, how to go about salvaging industry in south africa, the nuclear doppler effect – the reason for a ‘walk-away safe’..., using groundwater boreholes network to address drought in africa, what africa can learn from kenya about geothermal energy.

South Africa: coalition government won’t fix past failures – expect the private sector to play a bigger role in delivering power, transport and security

Lawrence Hamilton , University of the Witwatersrand

South African elections: research explores how disillusioned ANC supporters might use their vote

Michael Braun , University of the Witwatersrand

South Africa’s electricity crisis: a series of failures over 30 years have left a dim legacy

Mark Swilling , Stellenbosch University

South Africa’s electricity crisis: what political parties say in their election manifestos about solving it

Hartmut Winkler , University of Johannesburg

South African president Cyril Ramaphosa aims for upbeat tone in annual address, but fails to impress a jaundiced electorate

Keith Gottschalk , University of the Western Cape

South Africans are opting to go off-grid : how they’re being helped, and hindered, in their efforts

Germarié Viljoen , North-West University and Felix Dube , University of Pretoria

Tourists are returning to South Africa – but the sector will need to go green to deal with the country’s electricity crisis

Gijsbert Hoogendoorn , University of Johannesburg ; Irma Booyens , University of Strathclyde , and Kristy Langerman , University of Johannesburg

Johannesburg has been hit by severe water shortages: new plan to manage the crisis isn’t the answer

Anja du Plessis , University of South Africa

South Africa’s power crisis will continue until 2025 - and blackouts will take 5 years to phase out

South Africa’s power blackouts: solutions lie in solar farms and battery storage at scale, and an end to state monopoly

David Richard Walwyn , University of Pretoria

Corruption in South Africa: former CEO’s explosive book exposes how state power utility was destroyed

South Africa’s cold weather has arrived – some tips on how to stay warm and safe

Adriaan Van Der Walt , University of the Free State ; Jennifer Fitchett , University of the Witwatersrand , and Sarah Roffe , Agricultural Research Council

Farmers in South Africa face power cuts and a weak rand - but a number of factors are working in their favour too

Wandile Sihlobo , Stellenbosch University

Power cuts and food safety: how to avoid illness during loadshedding 

Lise Korsten , University of Pretoria

South Africans are fed up with their prospects, and their democracy, according to latest social attitudes survey

Joleen Steyn Kotze , University of the Free State and Benjamin Roberts , Human Sciences Research Council

Peter Hain: Neil Aggett died fighting apartheid – South Africa’s rulers have betrayed the struggle

Peter Hain , University of the Witwatersrand

Power cuts: South Africa’s state of disaster is being contested in court - COVID rulings give clues as to the outcome

Tracy-Lynn Field , University of the Witwatersrand

South Africa’s power crisis: going off the grid works for the wealthy – but could deepen injustice for the poor

Christina Culwick Fatti , Gauteng City-Region Observatory and Samkelisiwe Khanyile , Gauteng City-Region Observatory

Home power backup systems – electrical engineers answer your questions

MJ (Thinus) Booysen , Stellenbosch University and Arnold Rix , Stellenbosch University

Robberies surge as criminals take advantage of South Africa’s power outages

Guy Lamb , Stellenbosch University

Related Topics

• Cyril Ramaphosa
• Electricity
• Peacebuilding
• Renewable energy
• South Africa
• South Africa power cuts

Head of School, School of Arts & Social Sciences, Monash University Malaysia

Chief Operating Officer (COO)

Clinical Teaching Fellow

Data Manager

Director, Social Policy

Top contributors.

Professor of Physics, University of Johannesburg

Professor at the Graduate School of Business, University of Cape Town

Professor of Technology Management, University of Pretoria

Professor in Engineering, Chair in the Internet of Things, Stellenbosch University

Political Scientist, University of the Western Cape

Researcher in Renewable Energy, University of Cape Town

Professor in the Department of Microbiology and Plant Pathology and Co-Director at the Centre of Excellence in Food Security, University of Pretoria

Visiting Adjunct Professor at Wits Business School, University of the Witwatersrand

Senior Research Fellow, Johannesburg Institute for Advanced Study, University of Johannesburg

Senior Director: Social Impact & Transformation, Stellenbosch University

Chief Research Specialist in Democracy and Citizenship at the Human Science Research Council and a Research Fellow Centre for African Studies, University of the Free State

Professor of Physical Geography, University of the Witwatersrand

Consulting Economist; and Senior Research Fellow University of Johannesburg, University of Johannesburg

Research Director: Developmental, Capable and Ethical State (DCES) research division, and Coordinator of the South African Social Attitudes Survey (SASAS), Human Sciences Research Council

Senior researcher, urban sustainability transitions, environmental governance and resilience, Gauteng City-Region Observatory

• X (Twitter)
• Unfollow topic Follow topic

• Send Us News

Resolving the power crisis in South Africa: Insights from 2021, SA's worst load shedding year so far

This Executive Summary of a study report by Meridian Economics, published on 13 June 2022, was abridged by Chris Yelland, EE Business Intelligence

Empirical evidence demonstrates that an additional 5 GW of renewable energy would have essentially solved load shedding in 2021 whilst enabling better and more efficient operation of our power system, all at a cost saving to Eskom. Load shedding in 2021 was the worst on record, spanning 1165 hours with a total of 1,8 TWh of energy unserved – uncomfortably close to 1% of total electricity demand.

The report is in two parts.

Click here for Part A: insights from 2021 - SA’s worst load shedding year so far

Click here for Part B: An achievable game plan to end load shedding

The broader economic costs due to daily disruptions are difficult to quantify, but include lost production, lost investments, de-industrialisation, greater unemployment and declining livelihoods.

As the reliability of the existing fleet of generators continues to decline, and delays with procuring and connecting new capacity to the grid continue to mount, South Africa now faces the very real prospect of a return to Level 6 or even Level 8 load shedding in the foreseeable future. This situation is arguably the central manifestation of South Africa’s economic crisis, and a pathway to resolving it our greatest economic opportunity.

This Study Report is Part A of a two-part series exploring a feasible strategy to resolve load shedding and aims to provide a proper empirical basis for the development of such a strategy.

For this study, Eskom’s actual data is used to investigate the impact that additional generation capacity would have had on load shedding if it had been operational last year, focusing on the shortest lead-time and cheapest sources of power generation – wind and solar.

Confirmed by two separate modelling platforms, the results are startling – an additional 5 GW of wind and solar (the approximate capacity of two REIPPPP bidding rounds) in the same proportion as the currently installed capacity, would have allowed Eskom to eliminate 96,5% of load shedding in 2021.

Further to this, the additional wind and solar capacity would have reduced the amount of diesel burnt in the open cycle gas turbine (OCGT) peakers by more than 70%, simply by generating power at the time that these were running. More optimal use of Eskom’s pumped storage assets, enabled by the additional energy on the system, could have created further diesel savings – exceeding 80% in all.

The study finds that the remaining (3,5%) fraction of load shedding could have been eliminated by a modest expansion of Eskom’s ILS demand response programme or other aggregated Demand Response intervention, and the very last few hours by 2 GW of one-hour batteries.

This outcome is counterintuitive. Rather than increasing system risk as many observers expect, the analysis based on the empirical data shows unequivocally that adding variable renewable generators to the existing distressed South African power system will result in a disproportionate reduction in load shedding and increase in system reliability.

This insight is critical for mapping the way forward and avoiding expensive pitfalls and delays in doing so.

Analysis of the cost impact of adding 5 GW of renewable energy capacity to the system is equally surprising.

Based on Eskom’s 2020/21 Financial Year dispatch cost of OCGTs (R3,04/kWh) and coal power (R0,42/kWh) and assuming renewables prices around R0,68/kWh, the additional 5 GW of wind and solar would have created a net annual saving of R2,5-billion for Eskom.

This takes no account of the economic benefit of avoiding load shedding, but is merely the net cash saving to Eskom, driven primarily by reduction in the quantity of diesel burned.

The saving is after provision for a hypothetical R6,08/kWh incentive for participating customers to reduce load under a demand response programme (equating to a 100% premium over the cost of running existing OCGTs), and after provision of the cost of 2 GW of batteries.

The analysis in Part A of the Study Report demonstrates how avoidable the current load shedding crisis has been, and how cost-effectively it can be resolved based on hard evidence from the actual 2021 data.

Insights from this analysis also demonstrate that by taking adequate steps (starting immediately), solutions to resolving load shedding are within reach.

“In the absence of further urgent and drastic interventions load shedding is likely to increase substantially in the coming years”

An achievable game plan to end loadshedding in South Africa

South Africa’s ballooning load shedding problem is significantly worse than generally recognised, but insights from the empirical evidence demonstrate that practical pathways exist to contain and then resolve load shedding, and kickstart the country’s green industrialisation and decarbonisation ambitions. Unprecedented interventions are required.

The purpose of this report is two-fold, namely, to demonstrate to policy makers, regulators, and key stakeholders:

(a) how insistence on poorly conceived measures and regulatory rules has the direct effect of worsening the load shedding crisis by obstructing and delaying interventions that could reduce it; and 

(b) that by applying a laser focus to implementing a coherent set of strategically identified policy levers, government can establish a high level of confidence that the problem will be resolved in a reasonable period of time.

In contrast to conducting an  ex-post  analysis of historical data, developing a forward-looking plan to resolve load shedding is a more complex task – even over the short to medium term – due to the uncertainty associated with, and the continued evolution of the key drivers behind load shedding. 

The analysis covers the period up to 2026.

The first step was to analyse the nature of the problem, based on current trends and the interventions already being implemented to connect new generation capacity onto the grid (i.e., the outstanding Kusile units and IPP Office procurements up to Bid Window 6). This is referred to as the Base Case. 

Thereafter, a near optimised suite of additional resources was developed – as summarised in Table 6 on Page 8 of the full Study Report – that will have to be deployed to close the gap that remains. This is referred to as the Risk Adjusted Resource Plan, which is explained in more detail in the full Executive Summary and Study Report.

While there are a limited number of plausible scenarios where load shedding is resolved under the Base Case, this requires a decreasing demand trajectory and no further decline in the coal fleet performance.

In the more likely scenarios, load shedding in 2023 will see up to a 4-fold increase compared to 2021; up to 5-fold in 2024, 4-fold in 2025 and up to 10-fold in 2026, all when compared to 2021 – the worst year on record. In other words, in the absence of further urgent and drastic interventions load shedding is likely to increase substantially in the coming years.

Any credible plan to resolve load shedding cannot be based on best-case scenarios, it needs to respond effectively to most of the plausible downside scenarios outlined in the study report. Furthermore, the plan cannot be based on the same centralised “all eggs in one basket”-type approach that created the problem in the first place. The challenge is so large and complex that no single player will be able to solve it alone. 

The focus of Government’s intervention should be on mobilising thousands of economic actors throughout the economy to take the necessary steps to bring new capacity online urgently. This must be achieved by opening doors, removing policy obstacles and red tape, and creating powerful incentives for delivering the right outcomes. The solution must be diversified, contain contingency and avoid “single points of failure”. 

Furthermore, there is no time to start from scratch. To deliver expedited capacity we must work with what we have. This means, for instance, exploiting opportunities with the existing IPP Office procurement rounds, existing IPP projects, the 1 MW to 100 MW market segment, the <1 MW market segments, and Eskom and municipal procurements, etc.

Numerous resource expansion scenarios designed to resolve loadshedding were analysed. From this an ambitious Risk Adjusted Resource Plan was developed that also contains a modest amount of contingency to hedge against the high probability that not all aspects of a plan will be delivered in time.

Together this suite of additions in the Risk Adjusted Resource Plan can practically eliminate load shedding by 2024, with full security of supply reached by 2025. Ensuring this is delivered on time will be a substantial challenge. In practice the outcomes can be achieved by a “game plan” comprising a number of measures, detailed further in the full Executive Summary and Study Report.

This game plan to resolve load shedding consists of a combination of interdependent measures which, if all implemented, will result in a high probability that load shedding will practically be eliminated by 2024.

Implementing these measures will require the cooperation of different players – including some who do not always appreciate the negative impact of their current positions or behaviour on the ability of the power system to resolve load shedding. 

Implementing these reforms will also require political will at a scale that has not yet been demonstrated in dealing with South Africa’s power crisis. As with the 100 MW reform, substantial “arm twisting” will be required.

This is an abridged version of the Executive Summary of a study report by Meridian Economics, published on 13 June 2022.

Dire loss of export income looms unless SA decarbonises – De Ruyter

Groundbreaking agri-tech set to sweeten the deal for SA’s sustainable energy future

NERSA approves national free basic electricity rate for 2024/25

ABB launches new energy efficiency calculator

Survey explores reliability of lower-voltage generator step-up transformers in renewable energy systems

PPA secures wind energy for SA mining company

British port goes fully electric on new berth

South Africa registers 788 MW of new generation facilities in first quarter

The race is on to lead carbon-neutral logistics

Namibia and Belgium collaborate to build first green hydrogen plant in Walvis Bay

Heimdall “magic ball” optimises power grid with real-time monitoring

BP and Shell Downstream finalise sale of SAPREF precinct

Magazine click to view, energize issue 10 2024.

Energize Issue 9 2024

General Enquiries : Tel : + 27 12 841 2911 | Email :  [email protected]

• The CSIR in brief
• Our vision, mission and values
• Organisational structure
• History and highlights
• Facts and figures
• Collaborations and partnerships
• CSIR Code of Conduct
• CSIR Ethics Statement
• Partner with us
• Industry innovation support
• Specialised services
• Collaborate on technology commercialisation
• Contract R&D
• Alumni Programme
• CSIR International Convention Centre
• Students and graduates
• Professionals
• CSIR career guide
• Recruitment Agency Vacancies
• Multimedia Gallery
• Media releases
• Advanced Agriculture and Food
• Future production: Chemicals
• NextGen Health
• Future production: Manufacturing
• Future production: Mining
• Defence and security
• NextGen Enterprises and Institutions
• Smart mobility
• Smart places
• Scientific infrastructure
• CSIR Energy Research Centre
• CSIR Emerging Digital Technologies for 4IR Research Centre
• CSIR Information and Cybersecurity Research Centre
• CSIR Photonics Centre
• CSIR Robotics and Production Centre
• CSIR Water Research Centre
• Research groups
• Waste RDI Roadmap
• C4IR South Africa
• Annual report
• Corporate brochures
• eNewsletter
• Research publications
• ScienceScope
• Sector-specific publications
• Shareholder's compact
• Events for the Science Community
• CSIR upcoming conferences and events
• CSIR past conferences and events

Page Comment

Leave a comment.

We would love what you have to say

Page Unpublish

• Media Releases

CSIR releases statistics on power generation in South Africa for first half of 2022 and loadshedding data for first three quarters

The Council for Scientific and Industrial Research (CSIR) has released its annual statistics on power generation in South Africa for the first half of 2022 (1 January 2022 to 30 June 2022). Loadshedding and energy availability factor (EAF) data are also provided up until 30 September 2022.

The statistics include all utility-scale generation technologies in the analysis. Technologies include coal, nuclear, hydro, solar photovoltaics (PV), onshore wind, concentrated solar power (CSP), pumped storage and diesel-fueled open cycle gas turbines.

Contact Person

David Mandaha

072 126 8910

[email protected]

In the first half of 2022, the total system demand was similar to the year before, but still 3.0 TWh (2.5%) below the pre-lockdown levels of 2019. Coal still dominates the South African energy mix, providing more than 80% of the total system load. The contribution of renewable energy technologies (wind, solar PV and CSP) increased in 2022 to a total of 6.2 GW installed capacity and provided 6.5% of the total energy mix

The Eskom fleet EAF continued its declining trend in 2022, with an average EAF of 59.4%, compared to the EAF of 61.7% for 2021 and 65% for 2020. This is largely due to the increase of unplanned outages (detailed by the unplanned capacity loss factor) experienced by Eskom. This year overtook 2021 as the most loadshedding-intensive year yet, concentrated in July and September.

This year is the most intensive loadshedding year to date, concentrated in July and September.  The collective in the three months of July to September 2022 had more loadshedding in any year before. September 2022, the highest loadshedding month ever, on its own, had more loadshedding than the entire 2020. This year’s Stage 6 loadshedding has far surpassed 2019’s, the only other year that had Stage 6.

Click here for a full report [pdf]

Issued by CSIR Strategic Communications

For enquiries, contact:

David Mandaha:CSIR Media Relations Manager Tel: 012 841 3654 Mobile: 072 126 8910 Email: [email protected]  

About the CSIR:

The CSIR, an entity of the Ministry of Higher Education, Science and Innovation, is one of the leading scientific and technology research, development and implementation organisations in Africa. Constituted by an Act of Parliament in 1945 as a science council, the CSIR undertakes directed and multidisciplinary research and technological innovation, as well as industrial and scientific development to improve the quality of life of all South Africans. For more information, visit www.csir.co.za  

Follow us on social media:

Twitter: @CSIR. Facebook: CSIRSouthAfrica. Instagram: CSIRSouthAfrica. LinkedIn: Council for Scientific and Industrial Research (CSIR). YouTube: CSIRNewMedia

Related Information

Copyright © CSIR 2024. All Rights Reserved.