What Is Marginal-Cost Pricing?

Marginal cost pricing refers to the increase in total supply cost caused by increasing unit output. It is generally divided into short-term marginal cost and long-term marginal cost.

In the electricity market, the market mechanism (price mechanism) coordinates and controls all aspects of electricity commodities from production to consumption. The formation mechanism of electricity prices and the structure of electricity prices determine the fair, just and efficient operation of the electricity market. Electricity prices play a central role in the realization of the electricity market.

The theoretical problem of the electricity market is focused on electricity prices, which is specifically expressed by the price (cost) of all actions from electricity production to consumption. For example, calculation of peak and valley electricity prices, determination of hydroelectricity prices, calculation of reactive electricity costs, calculation of maintenance costs, calculation of standby costs, calculation of transmission costs, calculation of connection costs, calculation of accident losses, calculation of reliability costs, and so on. Utilize the economic signals of electricity prices in the market environment to guide, regulate, and control electricity production and consumption, so as to achieve the goal of optimizing resource allocation, rationally organizing production, and improving social and economic benefits ^[1]

Electricity price is an important factor affecting the development of the power industry. The power sector sells its only product (electricity) to recover the invested capital and raise a considerable part of the expansion funds. The price of electricity is directly related to the profit of the power sector, which affects the ability of the power industry to expand reproduction and attract external investment.

For a long period of time, electricity prices were mainly determined based on financial accounting. Generally, a certain additional amount is added to the historical cost (or average cost). The size of the additional amount usually depends on the government. This pricing method is called comprehensive cost pricing.

The concept of marginal cost is focused on the future. It is defined as the cost of increasing the unit's power supply based on the system's optimized planning and optimized operation, which increases the system's power supply. It can be seen from the definition that the essence of electricity price based on marginal cost is to solve development problems and raise funds for system expansion. At the same time, the electricity price gives the user a choice, and the user decides whether to increase the electricity consumption according to the income obtained by increasing the electricity consumption and the increased electricity expenses, thereby realizing the function of load management.

The ability of the power industry to recover funds depends mainly on the price of electricity over a longer period of time. In the long run, the real cost that the power industry pays to meet growing load needs is long-term marginal cost. Long-term marginal cost calculation uses various investment decision-making models to find the optimal investment plan and corresponding total costs at different load levels, and then uses the ratio of total cost increments to load increments as an approximation of long-term marginal costs. Long-term marginal cost can be divided into three components: marginal capacity cost, marginal energy cost, and marginal user cost.

There is also a non-strict method for calculating the long-term marginal cost of power generation capacity. Optimal planning is not required. Instead, a certain unit or a certain type of unit is designated as the marginal unit, and its cost is approximately used to replace the marginal cost of the system.

Long-term marginal cost pricing is applicable to the establishment of electricity prices for power systems that are growing in demand. Combined with the development plan of the power grid, it is possible to obtain forecasts of changes in electricity prices over a longer period of time. The advantage is that it is easy to implement and does not require high hardware and software requirements. Electricity prices have shown a stable trend over a period of time, which is expected by some users. The shortcomings are also obvious. As the long-term change trend of the system is examined, the electricity price is insufficient for the short-term load management function ^[2]

Short-term marginal cost pricing is the electricity price based on the marginal cost of optimized operation of existing resources without considering the new fixed investment in the system. Real-time electricity prices are representative of short-term marginal cost electricity prices. Based on the real-time electricity price theory, the real-time electricity price system has been continuously improved and expanded. The possibility of real-time electricity price for frequency control of the system, charging of reactive power, the relationship between reliability index and electricity price, and payment of system backup costs are discussed theoretically.

Due to the formation of the power market, many of the power generation and services provided by traditional power companies are decomposed to facilitate transactions and competition, such as frequency / connected line control, reactive power / voltage control, power transmission, power quality, and reliability services. Wait. Further concepts such as transshipment electricity prices and ancillary service electricity prices. There is no uniform definition of the above concepts in the world. We can temporarily assume that in addition to the production of active power, all other production and services in order to allow users to obtain qualified power are collectively referred to as auxiliary services, and the fees charged are referred to as auxiliary services. price. Ancillary service electricity prices are also included in real-time electricity prices.

By forming a stochastic optimal control model, many factors such as frequency / tie line control, reactive power / voltage constraints, safety and reliability, maintenance and backup, harmonic control, and environmental protection can be included in the real-time electricity price, and the calculation steps can be summarized. for:

(1) Consider the calculation of electricity price as a solution to the optimization problem of social benefits, and establish the objective function of social benefits and various technical and economic constraint functions;

(2) Write the Lagrangian function of the optimization problem. When the objective function or the constraint condition contains a differential term, the Hamiltonian function is used;

(3) The Lagrangian function finds the first derivative of the control variables and makes them all equal to zero;

(4) Let the real-time active (reactive) electricity price be equal to the partial derivative of the user's benefit function to its active (reactive) electricity. By transforming the equation in step 3, the real-time active (reactive) electricity price expression is finally obtained. .

The real-time electricity price obtained has a good structural form, which is the sum of the marginal cost and constraint cost of each production.

There is another type of dual solution using linear programming, that is, the real-time electricity price calculation method with the endogenous shadow price as the marginal cost. The concept of shadow price is closer to the opportunity cost in economics. The real-time electricity price component calculated using the Lagrangian function is generally related to the actual cost incurred, while the opportunity cost may not actually occur. Comparing the two algorithms, the real-time electricity price financial balance function obtained by using the Lagrangian function is better than the real-time electricity price obtained by linear programming, and the economic signal function of the latter is better than the former.

Another notable issue in electricity price calculation is the response of the load to electricity prices. The premise of electricity price calculation is a certain load forecast. In fact, because the electricity price has a controlling effect on the load, the load will change as the electricity price changes. There are various loads in the market, and different loads have different responses to the electricity price. Not only does the current electricity price affect the size of the load, but the previous and future electricity price levels also affect the current load. This makes it difficult to establish an accurate load model. Therefore, not only the electricity price calculation adopts an iterative method, but the electricity price implementation process is also a cyclic process. That is, the initial electricity price is released, the load response is examined, and the electricity price is revised again and again.

Theoretically, the real-time electricity price is continuously updated as the system operating conditions change. The shorter the cycle of the electricity price update, the better it is for the realization of various control functions of the electricity price, and the more it is for the economic benefits of the system. However, the realization of real-time electricity prices has high requirements on the system's software and hardware, and the cost of information collection, transmission, and processing increases rapidly with the shortening of the electricity price update cycle. As a result of the balance between benefits and expenditure, the real-time electricity price update cycle is generally 30 minutes ^[2]

The peak-valley electricity price (time-of-use electricity price) can be considered as a simplified form after taking into account the difficulty of real-time electricity price implementation. It generally divides the system's operation into 3 periods of peak, waist, and valley according to the system operating conditions.

Charge electricity at an average marginal cost in each period. The price update period can be longer, such as a quarter or a year. Peak-to-valley electricity prices are the preferred solution for many electricity, software, hardware and equipment facilities that lack backward management experience, and will gradually transition to real-time electricity prices when conditions mature in the future.

The method for calculating the peak-to-valley electricity price including the cost of energy operation and capacity can be obtained through the random production simulation of the power system.

The marginal cost electricity price represented by real-time electricity price has a good economic signal function, reflects the true cost of electricity production, and is conducive to the economic benefits of the power system. However, China's power system does not have the conditions to implement real-time electricity prices. The main reasons are:

1. Real-time electricity prices have high requirements on system hardware, including communication, metering, and control operations.

2. It is necessary to accumulate experience for the operating characteristics of the power system in a market environment.

3. Regarding changes in electricity prices, there is still no effective control method for electricity production and use, so it will affect the economic benefits obtained after the implementation of real-time electricity prices ^[3] .