Fuel Blending Properties in ProductMixer: Stream and Final Product Characteristics
ProductMixer Series
1 ProductMixer Complete Guide for Refinery Product Blending
2 ProductMixer: LP-Based Blend Optimization in Refinery Simulation
3 Product Mixer Automatic Blending Algorithm
4 Mathematical Formulation of Automatic Product Blend Optimization
5 Fuel Blending Properties in ProductMixer: Stream and Final Product Characteristics
6 Manual Product Blending Algorithm
7 Product Mixer Performance Metrics in Refinery Simulation
8 ProductMixer Control in Refinery Simulation
Solution: ProductMixer Simulation for Petroleum Refining Digital Twins
1 ProductMixer Complete Guide for Refinery Product Blending
2 ProductMixer: LP-Based Blend Optimization in Refinery Simulation
3 Product Mixer Automatic Blending Algorithm
4 Mathematical Formulation of Automatic Product Blend Optimization
5 Fuel Blending Properties in ProductMixer: Stream and Final Product Characteristics
6 Manual Product Blending Algorithm
7 Product Mixer Performance Metrics in Refinery Simulation
8 ProductMixer Control in Refinery Simulation
Solution: ProductMixer Simulation for Petroleum Refining Digital Twins
Fuel Blending Properties in ProductMixer
Modern refinery fuel blending requires simultaneous control of multiple fuel blending properties and fuel quality specifications while maximizing production or minimizing cost. ProductMixer provides a flexible property framework that separates stream properties from finished fuel product specifications, enabling the same optimization engine to support different products, grades, and operating conditions.
This architecture allows the same input streams to be blended into different products or product grades without modifying their quality data. Product specifications are defined independently and can include multiple fuel quality specification profiles, such as seasonal or customer-specific requirements. During optimization, every enabled property is translated into a linear programming constraint, allowing ProductMixer to determine blend ratios (blend proportions) that satisfy all active quality specifications while optimizing the selected objective. The mathematical formulation and optimization process are described in Automatic Product Blending Mathematical Algorithm. It should be noted that, in addition to automatic optimization, the ProductMixer can also operate in manual mode using predefined blend ratios. Configuration of product specifications, runtime properties, recalculation behavior, and ProductMixer API is described in ProductMixer Control in Refinery Simulation. Learn how ProductMixer uses stream properties and product specifications in the ProductMixer Complete Guide.
This architecture allows the same input streams to be blended into different products or product grades without modifying their quality data. Product specifications are defined independently and can include multiple fuel quality specification profiles, such as seasonal or customer-specific requirements. During optimization, every enabled property is translated into a linear programming constraint, allowing ProductMixer to determine blend ratios (blend proportions) that satisfy all active quality specifications while optimizing the selected objective. The mathematical formulation and optimization process are described in Automatic Product Blending Mathematical Algorithm. It should be noted that, in addition to automatic optimization, the ProductMixer can also operate in manual mode using predefined blend ratios. Configuration of product specifications, runtime properties, recalculation behavior, and ProductMixer API is described in ProductMixer Control in Refinery Simulation. Learn how ProductMixer uses stream properties and product specifications in the ProductMixer Complete Guide.
Stream Properties
Each stream entering ProductMixer is assigned a set of blending properties that describe its blending characteristics. Typical fuel blending properties include octane number, sulfur content, benzene concentration, Reid Vapor Pressure (RVP), aromatics, oxygen content, cetane number, and other refinery-specific characteristics.
Each blending property must be defined independently for every stream. If a property value is missing, ProductMixer throws an exception because the contribution of the stream to that blending property cannot be determined. The list of supported properties is configurable and can be extended to meet the requirements of a specific refinery or blending application. Learn how these stream properties are used by the optimization engine in Automatic Product Blending Mathematical Algorithm.
It should be noted that the ProductMixer also supports a manual operating mode. In this mode, the user manually considers the properties of the inlet streams when specifying the blend ratios. Learn how ProductMixer Statistics report stream properties throughout the blending process.
These stream and product properties become optimization inputs during refinery blending. Their application can be explored in the ProductMixer refinery blending simulation.
Each blending property must be defined independently for every stream. If a property value is missing, ProductMixer throws an exception because the contribution of the stream to that blending property cannot be determined. The list of supported properties is configurable and can be extended to meet the requirements of a specific refinery or blending application. Learn how these stream properties are used by the optimization engine in Automatic Product Blending Mathematical Algorithm.
It should be noted that the ProductMixer also supports a manual operating mode. In this mode, the user manually considers the properties of the inlet streams when specifying the blend ratios. Learn how ProductMixer Statistics report stream properties throughout the blending process.
These stream and product properties become optimization inputs during refinery blending. Their application can be explored in the ProductMixer refinery blending simulation.
Finished Product Specifications
While stream properties describe the quality of incoming blend components, finished fuel quality specifications define the required quality of the finished product. For each product, ProductMixer allows users to specify minimum and maximum limits for individual properties. Each constraint can be evaluated on either a mass or volume basis, depending on the property being controlled. Multiple specification groups can also be defined for the same product, making it easy to model seasonal fuel grades, regulatory standards, or customer-specific quality requirements without modifying the stream properties. During optimization, only the active specification group is included in the linear programming model, ensuring that the resulting blend satisfies all enabled quality constraints. ProductMixer also supports different optimization strategies. See Product Mixer Automatic Blending Algorithm for details.
It should be noted that the ProductMixer also supports a manual operating mode. In this mode, the user manually takes the properties of the inlet streams into account when specifying the blend ratios.
It should be noted that the ProductMixer also supports a manual operating mode. In this mode, the user manually takes the properties of the inlet streams into account when specifying the blend ratios.
Property Constraints in Optimization
During optimization, every enabled product specification is converted into a linear constraint. ProductMixer simultaneously considers all active blending properties, ensuring that the final blend satisfies every specification while optimizing the selected objective.
Although some fuel properties exhibit nonlinear blending behavior in real refinery operations, ProductMixer solves a linear programming (LP) model. In practice, nonlinear effects are represented using linearized correlations, empirical correction factors, statistical models, or engineering safety margins, which are incorporated directly into the optimization constraints. This approach combines computational efficiency with the flexibility required for industrial blending applications. The optimization model is generated automatically from the current simulation state. The complete LP formulation is described in Automatic Product Blending Mathematical Algorithm.
Although some fuel properties exhibit nonlinear blending behavior in real refinery operations, ProductMixer solves a linear programming (LP) model. In practice, nonlinear effects are represented using linearized correlations, empirical correction factors, statistical models, or engineering safety margins, which are incorporated directly into the optimization constraints. This approach combines computational efficiency with the flexibility required for industrial blending applications. The optimization model is generated automatically from the current simulation state. The complete LP formulation is described in Automatic Product Blending Mathematical Algorithm.
Property Management in Petroleum Refining Library
In Petroleum Refining Library, fuel blending properties are stored in a set of relational database tables that separate property definitions, stream quality data, and finished product specifications.
The main tables are:
The main tables are:
- a_product_mixer_constraints_names — defines the available blending properties (e.g., octane number, sulfur, benzene, Reid Vapor Pressure, aromatics, oxygen content, and cetane number). Each property includes its name, abbreviation, unit, and description.
- product_mixer_flows_constraints_values — stores the blending properties of incoming blend components. Each record defines the value of a specific property for a particular stream or blend component.
- a_product_mixer_constraints_values — stores the quality specifications for finished products. For each product, users can define minimum or maximum limits, select mass- or volume-based evaluation, enable or disable individual constraints, and organize specifications into multiple groups, such as seasonal or customer-specific quality standards.
Conclusion
ProductMixer separates stream properties from finished product specifications, providing a flexible and scalable framework for refinery fuel blending. Engineers can independently configure blend component properties, define multiple specification groups for each product, and adapt quality requirements to different operating conditions without changing the blending model.
This architecture enables ProductMixer to support a wide range of refinery blending applications while ensuring that every optimized blend satisfies the required quality specifications.
This architecture enables ProductMixer to support a wide range of refinery blending applications while ensuring that every optimized blend satisfies the required quality specifications.
FAQ
1. What fuel blending properties can ProductMixer optimize?
ProductMixer supports common fuel blending properties, including octane number, Reid Vapor Pressure (RVP), sulfur, benzene, aromatics, oxygen content, ether content, and cetane number.
2. Can different products use different quality specifications?
Yes. Each product has its own quality specifications. ProductMixer also supports multiple specification groups, allowing different requirements for seasonal fuel grades, regulatory standards, or customer-specific products.
3. What happens if a stream does not define a property?
Each blending property must be specified for every stream. If a property is not defined, ProductMixer throws an exception indicating that the required blending property value is missing.
4. Does ProductMixer support both minimum and maximum quality limits?
Yes. Each property can be constrained using minimum (No Less Than) or maximum (No More Than) limits and evaluated on either a mass or volume basis.
5. Does ProductMixer support nonlinear blending properties?
ProductMixer solves a linear programming (LP) problem. Nonlinear blending behavior is typically represented using linearized correlations, empirical correction factors, statistical models, or engineering safety margins incorporated directly into the optimization constraints.
ProductMixer supports common fuel blending properties, including octane number, Reid Vapor Pressure (RVP), sulfur, benzene, aromatics, oxygen content, ether content, and cetane number.
2. Can different products use different quality specifications?
Yes. Each product has its own quality specifications. ProductMixer also supports multiple specification groups, allowing different requirements for seasonal fuel grades, regulatory standards, or customer-specific products.
3. What happens if a stream does not define a property?
Each blending property must be specified for every stream. If a property is not defined, ProductMixer throws an exception indicating that the required blending property value is missing.
4. Does ProductMixer support both minimum and maximum quality limits?
Yes. Each property can be constrained using minimum (No Less Than) or maximum (No More Than) limits and evaluated on either a mass or volume basis.
5. Does ProductMixer support nonlinear blending properties?
ProductMixer solves a linear programming (LP) problem. Nonlinear blending behavior is typically represented using linearized correlations, empirical correction factors, statistical models, or engineering safety margins incorporated directly into the optimization constraints.