ProcessDesignAgents/main.py at main · may3rd/ProcessDesignAgents · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
from __future__ import annotations

from processdesignagents.graph.process_design_graph import ProcessDesignGraph
from processdesignagents.default_config import DEFAULT_CONFIG

config = DEFAULT_CONFIG.copy()

config["llm_provider"] = "openrouter"
# config["quick_think_llm"] = "google/gemini-2.5-flash-preview-09-2025"
# config["deep_think_llm"] = "google/gemini-2.5-flash-preview-09-2025"

# config["quick_think_llm"] = "google/gemini-2.5-flash-lite-preview-09-2025"
# config["deep_think_llm"] = "google/gemini-2.5-flash-lite-preview-09-2025"

# config["quick_think_llm"] = "qwen/qwen3-32b"
# config["deep_think_llm"] = "qwen/qwen3-32b"

# config["deep_think_llm"] = "anthropic/claude-sonnet-4.5"
# config["deep_think_llm"] = "z-ai/glm-4.6"

config["quick_think_llm"] = "x-ai/grok-4-fast"
config["deep_think_llm"] = "x-ai/grok-4-fast"

config["quick_think_llm"] = "google/gemini-2.5-flash-lite"
# config["deep_think_llm"] = "google/gemini-2.5-flash"

delay_time = 0.1

# config["llm_provider"] = "google"
# config["quick_think_llm"] = "gemini-2.5-flash"
# config["deep_think_llm"] = "gemeni-2.5-flash"

# config["llm_provider"] = "ollama"
# config["quick_think_llm"] = "llama3.2"
# config["deep_think_llm"] = "llama3.2"

config["quick_think_temperature"] = 0.5
config["deep_think_temperature"] = 0.5

def main():
    graph = ProcessDesignGraph(debug=False, config=config, delay_time=delay_time, save_graph_image=True, graph_image_filename="graph.png")
    problem_statement = ""
    problem_statement = "design generic compressed air unit for refinery with capacity 300 Nm3/h for plant air and instrument air."
    # problem_statement = "design carbon capture unit with capacity 100 ton per day of captured carbon product, the feed is flue gas with CO2 around 12.4 wt%."
    # problem_statement = "design MWCNT production, CVD of Methane in Fludized Bed Reactor, feed stock is LNG with 93% mole Methane (do need feed vaporization), the by-product H2 generated will be collected to be blue hydrogen (use PSA for H2 collection). Target capacity is 200 Ton per year of CNT production."
    # problem_statement = f"design the energy recovery from flue gas of LNG burner, 10,000 SCFD, 300°C, 0.1 barg and use it to produce electricity with 30% efficiency."
    # problem_statement = "design the carbon capture modular package, the feed to the package can be flue gas from various burner type. The target CO2 purity is 99.5% with the capacity of CO2 capture is 30 Ton per day"
    # problem_statement = "design heat exchanger to cool the ethanol product (99.5% purity molar basis, impurity is water) from 70°C to 45°C (at 2.0 barg). Design flowrate of ethanol feed is 2500 kg/hr. Supply CW temp is 30°C and return temp is 40°C."

    _ = graph.propagate(
        problem_statement=problem_statement,
        save_markdown="reports/latest_run.md",
        save_word_doc="reports/latest_run.docx",
        resume_from_last_run=False,
        manual_concept_selection=False
        )

if __name__ == "__main__":
    main()