Neqsim Utility Design

NeqSim Utility Design Skill

Size and cost the core plant utilities with deterministic NeqSim models in neqsim.process.util.utilitydesign. This skill covers the calculation side (duties → flows → power → CO₂ → cost); use neqsim-utilities-specification to decide which utility levels to provide and neqsim-heat-integration to derive the duties.

When to Use

• Sizing a fired steam boiler, deaerator, refrigeration package, or N₂ generator
• Building a multi-pressure steam header balance (HP→MP→LP letdown cascade)
• Aggregating heating/cooling/shaft-power/instrument-air demands from a flowsheet
• Estimating utility CO₂ emissions and annual operating cost for ESG/screening
• Exposing utility sizing through the MCP designUtilities tool

These are screening-level models for early design, not detailed mechanical design.

Classes

All classes live in neqsim.process.util.utilitydesign, expose public static final String SCHEMA_VERSION = "1.0", and provide calculate(), toResultsMap(), and toJson(). Each has a no-arg and a String name constructor.

Boiler

import neqsim.process.util.utilitydesign.Boiler;

Boiler boiler = new Boiler("HP Boiler");
boiler.setBoilerEfficiency(0.85);
boiler.setFuelLowHeatingValueMJperKg(47.0);
boiler.addSteamDuty("Reboiler", 5000.0); // kW; negative duty is rejected
boiler.calculate();
double fuelKgh = boiler.getFuelMassDemandKgh();
double steamKgh = boiler.getSteamGenerationKgh();
double co2 = boiler.getCo2TonnePerYear();
String json = boiler.toJson();

RefrigerationCycle

COP is (T_evap/ΔT) × cycleEfficiency (Carnot fraction). Condenser duty is the evaporator duty plus compressor power.

import neqsim.process.util.utilitydesign.RefrigerationCycle;

RefrigerationCycle cycle = new RefrigerationCycle("Propane Chiller");
cycle.setEvaporatorTempC(-35.0);
cycle.setCondenserTempC(35.0);
cycle.setCycleEfficiency(0.55);
cycle.addRefrigerationDuty("Gas Chilling", 3000.0); // kW
cycle.calculate();
double power = cycle.getCompressorPowerKW();
double cop = cycle.getCop();

NitrogenSystem

Generation method sets the specific energy and air-to-product ratio; purity above 99% adds a small energy penalty.

import neqsim.process.util.utilitydesign.NitrogenSystem;

NitrogenSystem n2 = new NitrogenSystem("N2 Generation");
n2.setNitrogenDemandNm3h(500.0);
n2.setPurityPercent(99.5);
n2.setGenerationMethod(NitrogenSystem.GenerationMethod.MEMBRANE); // or PSA, CRYOGENIC
n2.calculate();
double powerKW = n2.getPowerKW();
double airFeed = n2.getAirFeedNm3h();

SteamNetwork

Headers are sorted descending by pressure; the cascade balances each header top-down and the top header's make-up sets the boiler generation.

import neqsim.process.util.utilitydesign.SteamNetwork;

SteamNetwork net = new SteamNetwork("Steam System");
net.addLevel("HP", 42.0, 253.0);   // name, pressure bara, saturation T °C
net.addLevel("LP", 4.5, 148.0);
net.addDemand("HP", 8000.0);        // kg/h; unknown header throws
net.addDemand("LP", 5000.0);
net.setLocalGeneration("LP", 2000.0);
net.calculate();
double boilerKgh = net.getBoilerGenerationKgh();

UtilitySystemDesigner (flowsheet aggregator)

import neqsim.process.util.utilitydesign.UtilitySystemDesigner;

processSystem.run();
UtilitySystemDesigner u = UtilitySystemDesigner.fromProcessSystem(processSystem).design();
double heating = u.getTotalHeatingDutyKW();
double cooling = u.getTotalCoolingDutyKW();
double cwFlow = u.getCoolingWaterFlowM3h();
// For a ProcessModel: UtilitySystemDesigner.fromProcessModel(plant)

MCP designUtilities tool

The MCP server (neqsim-mcp-server) exposes designUtilities (Tier 2, Engineering Advanced). Dispatch on utilityType:

{ "utilityType": "boiler", "name": "HP Boiler", "boilerEfficiency": 0.85,
  "duties": [{ "name": "Reboiler", "dutyKW": 5000.0 }] }

The runner returns the class toJson() wrapped by the standard MCP response envelope. Backend: neqsim.mcp.runners.UtilityDesignRunner.

Agentic utility optimization

UtilityCompressionOptimizer (same package) drives a real two-stage compression flowsheet (feed → Compressor → intercooler Cooler → Compressor) through the agentic AgenticProcessOptimizer to find the interstage pressure that minimizes total shaft power. It rediscovers the analytic geometric-mean optimum sqrt(pIn·pOut) from the live model.

import neqsim.process.util.utilitydesign.UtilityCompressionOptimizer;

UtilityCompressionOptimizer opt = new UtilityCompressionOptimizer("Instrument Air");
opt.setInletPressureBara(5.0);
opt.setDeliveryPressureBara(60.0);
opt.setSeed(42L).setMaxEvaluations(70);
opt.optimize();
double pMid = opt.getOptimumInterstagePressureBara(); // ≈ sqrt(5*60) = 17.3 bara
double power = opt.getMinTotalPowerKW();
String json = opt.toJson();

Single decision variable (interstage pressure), custom objective = summed stage power, deterministic for a fixed seed, never throws. Use as the template for wiring any utility flowsheet into a closed optimization loop.

Gotchas

• All duties are kW; negative duties are rejected (boiler, refrigeration).
• SteamNetwork.addDemand/setLocalGeneration validate the header name exists.
• Refrigeration COP uses absolute temperatures — keep evaporator below condenser.
• These are screening models: no detailed mechanical, control, or transient design.
• Set the mixing rule and run the flowsheet before UtilitySystemDesigner.fromProcessSystem.

Related

• neqsim-utilities-specification — choosing utility levels
• neqsim-heat-integration — deriving duties from composite curves
• neqsim-power-generation — HRSG / steam-turbine network