Crisis Probability Modelling in Electricity Systems

Detailed Explanation With Case Laws

1. Introduction

Crisis probability modelling in electricity systems means using mathematical, statistical and computational methods to estimate the likelihood of serious electricity-system events. These events may include blackouts, generation shortages, transmission failures, extreme price events, fuel shortages, cyber incidents and system instability.

The purpose is not to predict a crisis with certainty. Instead, probability models help system operators, regulators and governments understand which risks may occur, how serious they could be and what preparations may be required.

Modern electricity systems are becoming more complex because of renewable generation, battery storage, electric vehicles, distributed generation and digital control systems. Probability modelling has therefore become an important part of electricity-risk governance.

2. Meaning of Crisis Probability

A probability model estimates the likelihood of an event.

For example:

Probability of generation shortage = possible shortage events ÷ relevant system conditions

A model may estimate:

probability of insufficient generation;

probability of transmission failure;

expected frequency of outages;

probability of voltage instability;

probability of extreme electricity prices; and

probability of cascading failures.

The results can then support emergency planning and investment decisions.

3. Importance for Electricity Security

Electricity systems operate continuously and must maintain a balance between supply and demand.

A simple crisis chain can be:

Generator failure → supply shortage → frequency decline → emergency intervention → possible blackout.

Probability modelling helps identify the likelihood of such chains.

For example, a system operator may model thousands of possible combinations of:

generator outages;

wind and solar availability;

electricity demand;

transmission failures;

fuel availability; and

weather conditions.

This can provide an estimate of the system's overall risk.

4. Reliability Metrics

Electricity regulators and system operators use different reliability measures.

Loss of Load Probability (LOLP)

LOLP estimates the probability that available generation will be insufficient to meet demand.

Loss of Load Expectation (LOLE)

LOLE estimates the expected amount of time when available capacity may be insufficient.

Expected Energy Unserved (EEU)

EEU estimates the expected amount of electricity that consumers may not receive during shortage events.

These measures help regulators decide whether additional generation, storage, interconnection or demand response is necessary.

5. Monte Carlo Modelling

One common approach is Monte Carlo simulation.

The model generates many possible system conditions.

For example:

Generator A fails or operates;

wind generation is high or low;

demand is high or low;

transmission line operates or fails.

The computer runs thousands or millions of combinations.

The resulting distribution can provide estimates of crisis probabilities.

This is especially useful when electricity systems contain many uncertain variables.

6. Weather and Renewable Energy

Renewable electricity introduces additional uncertainty.

Wind and solar output depends heavily on weather.

Therefore, probability models may combine:

weather data + renewable generation + demand + network conditions.

For example, a prolonged period of low wind combined with high winter demand could increase the probability of a supply shortage.

This is sometimes referred to as renewable-energy adequacy modelling.

7. Extreme Events and Cascading Failures

Electricity crises may involve cascading failures.

For example:

Transmission line failure → power flows shift → second line overloads → additional failure → system separation → blackout.

Probability models can estimate the likelihood of these chains.

However, modelling cascading events is difficult because the system is highly interconnected.

Small changes in initial conditions can sometimes produce very different outcomes.

Therefore, models should be treated as risk-management tools rather than perfect predictions.

8. Cybersecurity Risk

Modern electricity systems also face cyber risks.

Probability modelling can assess the likelihood of:

cyber intrusion;

control-system compromise;

communication failure;

malicious switching;

data manipulation; and

combined cyber-physical failures.

The challenge is that historical data about successful attacks may be limited.

Consequently, models often need to combine historical evidence with scenario analysis and expert assessment.

9. Legal and Regulatory Importance

Probability modelling has become important in energy regulation because regulators must decide whether an electricity system has sufficient resilience.

Risk assessments can influence:

capacity-market requirements;

network investment;

emergency planning;

reserve requirements;

reliability standards;

transmission investment; and

government energy-security policy.

The legal question is therefore not simply whether the model is technically sophisticated.

Authorities must also consider whether the modelling process is lawful, transparent, evidence-based and appropriately explained.

10. Relevant Case Laws

R (National Grid Electricity Transmission plc) v Gas and Electricity Markets Authority [2018] EWCA Civ 1344

This case concerned regulatory arrangements affecting electricity transmission.

It demonstrates the importance of Ofgem exercising its regulatory powers within the statutory framework.

For probability modelling, the case is relevant because technical and economic assessments used by regulators ultimately operate within a legal decision-making framework. A sophisticated model cannot replace lawful regulatory authority.

R (on the application of Greenpeace Ltd) v Secretary of State for Business, Energy and Industrial Strategy [2022] EWHC 165 (Admin)

This case concerned the government's energy strategy and the evidence supporting its policy decisions.

The case is relevant to probability modelling because it illustrates the importance of evidence and reasoning in major energy-policy decisions.

Where government relies on forecasts, modelling or assumptions, those materials can become relevant to the legality of the decision-making process.

R (ClientEarth) v Secretary of State for Business, Energy and Industrial Strategy [2022] EWHC 2687 (Admin)

This case concerned the government's strategy for meeting statutory climate targets.

The court considered whether the government's strategy contained sufficient information to demonstrate how statutory targets would be achieved.

The case illustrates an important principle for modelling: quantitative assessments used in energy governance must be sufficiently explained to support accountable decision-making.

11. Models and Regulatory Decision-Making

Probability models should not automatically determine regulatory decisions.

A regulator may consider:

modelling results;

historical evidence;

engineering assessments;

economic analysis;

stakeholder evidence; and

expert judgment.

Models contain assumptions.

For example, a model may assume a particular:

weather distribution;

generator failure rate;

demand pattern;

fuel availability; or

interconnector availability.

If those assumptions change, the probability estimate may also change.

Therefore, model uncertainty is itself an important regulatory issue.

12. Transparency and Accountability

A regulator relying on probability modelling should ideally explain:

the purpose of the model;

data sources;

key assumptions;

methodology;

limitations;

sensitivity analysis; and

how the results affected the final decision.

This allows affected stakeholders and courts to understand the reasoning.

For PhD-level energy law, this creates an important connection between technical modelling and administrative law.

13. Problems With Crisis Probability Models

Probability modelling has several limitations.

Data Limitations

Rare events may have very little historical data.

Model Risk

The model itself may contain incorrect assumptions.

Black-Swan Events

Some events may be outside the model's historical experience.

Correlated Failures

Different failures may occur together rather than independently.

Cyber Risks

Cyberattacks may be difficult to quantify because attackers deliberately change their behaviour.

Therefore, regulators should use scenario testing and stress testing alongside probability models.

14. Scenario and Stress Testing

Probability modelling should be complemented by extreme scenarios.

For example:

Scenario 1: unusually cold winter + low wind generation

Scenario 2: major gas-supply interruption

Scenario 3: multiple transmission failures

Scenario 4: cyberattack + generation outage

Stress testing asks:

"What happens if several serious events occur at the same time?"

This can identify vulnerabilities that ordinary probability estimates may underestimate.

15. Conclusion

Crisis probability modelling in electricity systems provides a structured method for assessing the likelihood and potential consequences of electricity emergencies.

It can support decisions concerning generation adequacy, transmission resilience, reserve requirements, capacity markets, emergency planning and investment.

However, probability models are not perfect predictions. Their results depend on data, assumptions and modelling methodology. Therefore, regulators should combine quantitative models with scenario analysis, engineering expertise and stress testing.

The cases National Grid v GEMA, Greenpeace and ClientEarth demonstrate the wider legal importance of evidence, statutory authority and accountable decision-making in energy governance.

For PhD-level analysis, the central issue is the relationship between technical uncertainty and legal accountability. Electricity regulators increasingly depend on complex models, but the use of sophisticated modelling does not remove the need for transparent assumptions, reasoned decisions and meaningful regulatory scrutiny.