Q&A

These products are highly technological and always customized to the customer's business processes, so to determine the cost and provide a commercial offer, we need to gather information on an individual basis.

JSC "NBI" is the largest Russian provider of IT solutions aimed at improving the efficiency of production and distribution of electric and thermal energy in Russia. The company is also the developer of a unique solution that has already become the industry standard for automating business processes related to operations on the Wholesale Electricity Market (OREM). Our products have been implemented at more than 200 power plants with a total installed capacity of over 126 GW, which accounts for more than 50% of the entire Unified Energy System of the Russian Federation.

The system is not a closed black box. It is simple and intuitive to use. Further development of the system can be carried out either by the client’s team or by our specialists. During the commissioning phase, we provide training for various categories of users. We are constantly evolving, and enhancing the system to meet the core needs of the client is one of our key priorities.

We develop operational digital twins, whose primary function is to optimize performance. Based on an objective function, the model plans operating modes over various time periods (long-term, medium-term, short-term) and identifies the most advantageous mode among all possibilities. The digital twins also have several other functions, which can be explored in more detail here.

After the implementation of our products, a 12-month warranty support is provided. If necessary, a technical support agreement can be concluded.

The implementation timeline depends on the enterprise and the scope of production processes, but the approximate period for deploying a digital twin at a combined heat and power plant (CHP) is between 8 to 12 months.

The maximum number of users is unlimited. The main limitation is related to the automation objects, which corresponds to our internal licensing policy.

The modules form a unified EMAS platform that complement each other, but depending on the customer's business requirements, these modules are also easily separable. The modules within the EMAS platform can be described as Lego blocks. To answer the question: "Is it possible to purchase the EMAS.TEP or EMAS.OPT module separately?" — yes, it is.

At present, subscription-based use of services is provided only in the EMAS.FORECAST module. This is because the implementation of other EMAS platform products is quite labor-intensive and is carried out individually for each facility.

Our software connects to your systems via standard information and industrial protocols without the need to replace existing infrastructure. We have extensive experience in integration and have worked with a wide variety of system “zoos” at different facilities.

Why this matters: You preserve your current investment in equipment, and our "soft" integration technology reduces the risks of downtime during implementation. Unlike competitors, we support even outdated systems without costly upgrades.

You will only need a basic setup: a server with Linux (Astra or RedOS) and PostgreSQL; all special licenses are already included in our package.
Competitive advantage: We have deliberately opted out of expensive commercial imported DBMS, which reduces your costs by 25-30% compared to Oracle or MS SQL solutions. At the same time, performance remains at the same level thanks to our optimized data storage scheme, and you gain import independence.

You have several options for working with data, including automatic daily exports or on-demand data retrieval.
Important to note: Our CSV format automatically adapts to your accounting systems, saving up to 15 hours of IT specialists' work each month.

The recommended setup with two servers ensures uninterrupted operation. Data is synchronized in real time, and failover occurs automatically.
Why this is reliable: We have implemented a "hot" standby mechanism — in case of failure, the system virtually does not lose any critical data. For mission-critical facilities, a third server can be used for geographic distribution.

Our solution is developed in accordance with strict standards, and the development process is certified by FSTEC. Access rights are segmented for different employees, and all user actions are logged.
An important detail: Our system does not require data transmission outside your network — all processing is done locally. This eliminates risks associated with competitors’ cloud solutions.

The support service operates according to a clear schedule:

* Critical failures – resolved within 4 hours
* Medium-priority issues – resolved within the day
* General inquiries – up to 3 business days

24/7 support can also be included as part of the SLA.

Why we are better: Our engineers specialize specifically in the energy sector — the average problem resolution time is 2.3 times faster than that of general vendors. Each specialist understands the specifics of CHP plant operations.

The cost already includes:

* A 16-hour basic course
* Practical exercises with real-life examples
* Access to training materials
* Consultations during the first months of operation

What this means: Our methods and test models simulate real situations — after training, your operators make 60% fewer mistakes during critical operating modes.

We release regular updates and improvements including:

* New features based on client requests
* Enhancements to existing algorithms
* Adaptation to changes in legislation

Update policy: All new features are first tested on test environments. Only after confirming stability and effectiveness is the update offered to clients in production environments.

Our solution uses mixed-integer linear programming (MILP).

Key advantage: It provides an exact solution for problems with discrete parameters (equipment on/off) and continuous variables (fuel consumption).

Implementation features:

* Utilizes a modified branch-and-bound algorithm
* Speeds up calculations by pre-filtering non-optimal options
* Accuracy of the solution is up to 0.1% from the theoretical optimum

Comparison with other methods:

* Unlike genetic and dynamic algorithms, it guarantees finding the global optimum
* Allows explicit consideration of technological constraints

Our system employs a comprehensive approach to processing sensor data, ensuring high measurement accuracy even in the presence of inaccuracies:

PDR (Process Data Reconciliation) Technology

How it works:

• A statistical data reconciliation method that
• Analyzes relationships between all system sensors
• Detects and corrects conflicting readings
• Minimizes the sum of squared adjustments to measurements

Automatic Data Replacement Based on Rules

• A multi-level anomaly handling system:
• Primary filtering (excluding physically impossible values)
• Correction using nearby sensors measuring similar parameters
• Substitution of calculated values based on stored characteristics

Examples of replacement rules for a steam pressure sensor failure:

• Use data from a redundant sensor
• Calculate based on temperature and flow
• Substitute the last valid value (for a limited time)

Advantages of our approach:

• Reduces human error in data processing
• Minimizes false alarms in emergency signaling
• Allows fine-tuning of rules for specific equipment
• Maintains a log of all adjustments for further analysis

Our implementations demonstrate consistent results:

* Fuel savings of at least 2%
* Reduction in operating costs
* Payback period of less than one year

Financial aspect: We can formalize these metrics in the contract and are open to discussing penalties for non-achievement.