Zhomart Mominbayev, MBA, InTech-Forecast Project Manager, Director of Modern Innovative Technologies LLP

The main reason is the unreliability and daily unevenness of electricity supply from renewable energy sources.

Current issues of forecasting electricity generation by renewable energy facilities

The basis of the balance of capacities of the Republic of Kazakhstan is eight national electric power stations and seven industrial power plants. The calculation of the forecast balance of electric power of the Unified Electric Power System of the Republic of Kazakhstan for an hour of combined maximum loads is made assumed that the available generation capacity of the Republic of Kazakhstan does not include the capacity of stations using renewable energy sources. The main reason is the unreliability and daily unevenness of electricity supply from renewable energy sources. Due to the instability of power output, RES stations are a very destabilizing factor for the System operator of the National Electric Grid of the Republic of Kazakhstan (KEGOC JSC).

Balance sheets and reserves

In 2020, the regulatory requirement for maneuverable capacity for balancing existing plants using renewable energy sources was 260 MW. If the need for maneuverable capacity to balance the green energy sector increases by 3 times in the coming years, the entire reserve of maneuverable capacity will be exhausted and the functioning of the National Electric Grid will noticeably deteriorate.

 Along with the development of maneuverable generation as separate independent energy projects, KEGOC JSC makes proposals to include

in the Technical Conditions for connecting RES facilities to the grid the provision of maneuverable sources to the System operator in the ratio of 2 MW of RES to 1 MW of maneuverable power. Exactly according to this mechanism, the technical specifications for connection to the grid of the first stage of the 50 MW SPP "Balkhash" were obtained, which is implemented outside the auction mechanism. At the same time, the System operator purchases services for the provision of maneuverable power, but this payment compensates only part of the costs associated with the installation and maintenance of the ALEC system (automatic load-frequency control).

No less urgent issue for reducing the destabilizing impact of RES on the UES of the Republic of Kazakhstan is the compliance of RES stations with the stated plans for generating electricity for a day or more. In accordance with the rules of power grid, the actual generation should not deviate by +/- 10% from the reported plan. For customary stations, this norm is strictly regulated, but there have been eases in relation to RES which may soon end. The Law of the Republic of Kazakhstan "On Support for the Development of Renewable Energy Sources" specifies two points — the primary dispatching of electricity from the RES and the commercial settlement of imbalances by the AEC. At the same time, there is no information that in case of non-compliance of RES facilities with the planned output values, the System operator does not have the right to impose grid restrictions on the output of power.

Two projection models

There are two models for organizing the forecasting of electricity generation from renewable energy sources — decentralized, used in Europe, and centralized, used in China. In a decentralized model, RES facilities themselves predict the weather and output, and are themselves responsible for the results of forecasting. In a centralized model, the System operator (State Grid of China) provides all stations with a weather prognosis and is responsible for the accuracy of the forecasts. But at the same time, it also easy restricts the output from RES stations, if this affects the stability of the grid operation modes.

In Kazakhstan, the decentralized model of regulation of forecasts for electricity generation is taken as a basis, while at the stage of attracting investors and developing the RES market, the issue of compliance with the declared regimes did not cause much concern to the "green energy" representatives. However, even now the System operator is faced with the problem of inconsistency between the forecast data and the actual output. The responsibility of RES facilities will only increase with the introduction of a balancing electricity market and the growth of RES projects, which may lead to the need to adjust their financial and technical models. Currently, RES stations predict electricity generation for a day or more in advance, using third-party software, or independently, taking into account historical data and received weather forecasts. Often the prediction results look like as follows:

The meeting held on March 18 this year under the auspices of the Ministry of Energy of the Republic of Kazakhstan and KEGOC JSC on the topic "Forecasting of renewable energy production" showed the great work carried out by the System operator in this direction. According to the presentation of KEGOC JSC, the volume of purchased/sold imbalances committed by the AFC for RES Support LLP in 2020 in the simulation mode amounted to 660 million kWh in the amount of 21.3 billion tenge. That approximately corresponds to 20% of all electricity purchased by AFC for RES Support LLP from RES sources during this period.

InTech Forecast IT-program

In order to create a Kazakhstan software product aimed at deep customization for local conditions, reporting on the technical readiness of the station to issue power and the requirements put forward by KEGOC JSC, Modern Innovative Technologies LLP has developed an InTech Forecast IT-program with the receipt of the corresponding copyrights.

The program is being tested at the 10 MW KengirSPP. The main elements of the program are historical weather and production data, as well as machine learning, which is necessary to constantly improve the validity of forecast data.

In the near future, the program will be ready to be launched both in Kazakhstan and in foreign markets. As a payment for the use of the IT product, it is possible to provide CO2 units, which can also be verified by us with the involvement of a partner specialized organization. Thus, units of CO2 that currently do not have an active demand can be involved in circulation as a means of exchange. And, of course, all operations will be performed in accordance with the Tax Code of the Republic of Kazakhstan. The program will be able to work, among other things, with such requirements of KEGOC JSC:

DUE TO EXISTING SITUATION IN THE REPUBLIC OF KAZAKHSTAN, 24-HOUR BALANCE AND SCHEDULES WILL BE DRAWN UP AS FOLLOWS:

March 15 — for March 16 and 17 March 16 — for March 18 and 19 ,

March 17 — for March 20 and 21 March 18 — for March 22 and 23 March 19 — for March 24 and 25

 First of all, the program is aimed at working with solar stations, and it can be scaled to work with wind stations in future as well. In order to attract grant funds to co-finance the completion of the IT program, the development team, which in addition to programmers includes a specialist in commercialization, a marketer and a financier, in February this year applied to Center for Engineering and Technology Transfer JSC (from March 09, 2021, it was renamed to National Agency for the Development of Innovations QazInnovations JSC. Our application has passed the first stage and the documents for the second stage have already been submitted. We will definitely inform you about the results. In total, 661 applications for innovative grants were sent to QazInnovations, including 92 for GreenTech.

We are ready to actively cooperate with the participants of the electricity and capacity market to make our own contribution to the sustainable development of renewable energy projects in the Republic of Kazakhstan. We also hope that the InTech-Forecast program will occupy its niche in the market of software products for forecasting electricity generation by solar and wind power plants.