You need to enable JavaScript to run this app.
优惠活动
大模型
产品
解决方案
定价
更多

如何用Python的FMPY库初始化OpenModelica热力学模型参数

avatar
API 探路者
2026-8-19
使用FMPY修改OpenModelica热力学模型的初始压力参数

我想了解如何用Python的FMPY库修改OpenModelica模型的内部参数，当前使用的是一个热力学模型，希望直接在Python中初始化压力参数。

待初始化参数

现有代码 

# define the model name and simulation parameters
# fmu_filename = 'Stockage_1.fmu'
fmu_filename = 'Stockage350.fmu'
start_time = 0
#threshold = 150.0
stop_time = 8000
step_size = 1
#step_size = 0.5

# download the FMU
#download_test_file('2.0', 'CoSimulation', 'MapleSim', '2016.2', 'CoupledClutches', fmu_filename)

# read the model description
model_description = read_model_description(fmu_filename)

# collect the value references
vrs = {}
for variable in model_description.modelVariables:
    vrs[variable.name] = variable.valueReference

# get the value references for the variables we want to get/set
in_consommation_H2   = vrs['Conso_Kg_s']      # normalized force on the 3rd clutch
in_cmd_1_R1   = vrs['cmd_1_R1']
in_cmd_1_R2   = vrs['cmd_1_R2']
in_cmd_1_R3   = vrs['cmd_1_R3']
in_cmd_1_R4   = vrs['cmd_1_R4']
in_cmd_1_R5   = vrs['cmd_1_R5']
in_cmd_1_R6   = vrs['cmd_1_R6']
in_cmd_1_R7   = vrs['cmd_1_R7']
in_cmd_1_R8   = vrs['cmd_1_R8']
in_cmd_2_R1   = vrs['cmd_2_R1']
in_cmd_2_R2   = vrs['cmd_2_R2']
in_cmd_2_R3   = vrs['cmd_2_R3']
in_cmd_2_R4   = vrs['cmd_2_R4']
in_cmd_2_R5   = vrs['cmd_2_R5']
in_cmd_2_R6   = vrs['cmd_2_R6']
in_cmd_2_R7   = vrs['cmd_2_R7']
in_cmd_2_R8   = vrs['cmd_2_R8']
out_Masse_H2 = vrs['Masse_H2']
out_P1 = vrs['P1']
out_P2 = vrs['P2']
out_P3 = vrs['P3']
out_P4 = vrs['P4']
out_P5 = vrs['P5']
out_P6 = vrs['P6']
out_P7 = vrs['P6']
out_P8 = vrs['P8']
out_P_1 = vrs['P_1']
out_P_2 = vrs['P_2']
out_P_3 = vrs['P_3']
out_P_4 = vrs['P_4']
out_P_5 = vrs['P_5']
out_P_6 = vrs['P_6']
out_P_7 = vrs['P_7']
out_P_8 = vrs['P_8']
out_P_HP = vrs['P_HP']
out_P_MP = vrs['P_MP']
out_T1 = vrs['T1']
out_T2 = vrs['T2']
out_T3 = vrs['T3']
out_T4 = vrs['T4']
out_T5 = vrs['T5']
out_T6 = vrs['T6']
out_T7 = vrs['T6']
out_T8 = vrs['T8']
out_T_1 = vrs['T_1']
out_T_2 = vrs['T_2']
out_T_3 = vrs['T_3']
out_T_4 = vrs['T_4']
out_T_5 = vrs['T_5']
out_T_6 = vrs['T_6']
out_T_7 = vrs['T_7']
out_T_8 = vrs['T_8']
out_T_HP = vrs['T_HP']
out_m_flow = vrs['m_Flow']

# extract the FMU
unzipdir = extract(fmu_filename)

fmu = FMU2Slave(guid=model_description.guid,
                unzipDirectory=unzipdir,
                modelIdentifier=model_description.coSimulation.modelIdentifier,
                instanceName='instance1')

# initialize
fmu.instantiate()
fmu.setupExperiment(startTime=start_time)
fmu.enterInitializationMode()
fmu.exitInitializationMode()

time = start_time

rows = []  # list to record the results

# simulation loop
while time <= stop_time:

    # NOTE: the FMU.get*() and FMU.set*() functions take lists of
    # value references as arguments and return lists of values
    
    print("\n\n\n\n\nwhile() time: " + str(time))

    # set the input
    fmu.setReal([in_consommation_H2], [0.00678 if time < 0.0 else 0.00678])

    # set the input
    fmu.setBoolean([in_cmd_1_R1], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R2], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R3], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R4], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R5], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R6], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R7], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_1_R8], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R1], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R2], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R3], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R4], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R5], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R6], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R7], [1 if time < 1.0 else 1])
    fmu.setBoolean([in_cmd_2_R8], [1 if time < 1.0 else 1])

    print("\nwhile() 2 time: " + str(time))
    # perform one step
    
    fmu.doStep(currentCommunicationPoint=time, communicationStepSize=step_size)
    
    print("\nwhile() 3 time: " + str(time))

    # advance the time
    time += step_size

    # get the values for 'inputs' and 'outputs[4]'
    resultat_FMU = fmu.getReal([out_Masse_H2,
                                out_P1, out_P2, out_P3, out_P4, out_P5, out_P6, out_P7, out_P8,
                                out_P_1, out_P_2, out_P_3, out_P_4, out_P_5, out_P_6, out_P_7, out_P_8, 
                                out_T1, out_T2, out_T3, out_T4, out_T5, out_T6, out_T7, out_T8,
                                out_T_1, out_T_2, out_T_3, out_T_4, out_T_5, out_T_6, out_T_7, out_T_8,
                                out_P_HP, out_P_MP,
                                out_T_HP,
                                out_m_flow])
                                                                        
    print("Sortie du FMU: " + str(resultat_FMU))

    # append the results
    rows.append((time,
                    resultat_FMU[0],
                    resultat_FMU[1], resultat_FMU[2], resultat_FMU[3], resultat_FMU[4], resultat_FMU[5], resultat_FMU[6], resultat_FMU[7], resultat_FMU[8],
                    resultat_FMU[9], resultat_FMU[10], resultat_FMU[11], resultat_FMU[12], resultat_FMU[13], resultat_FMU[14], resultat_FMU[15], resultat_FMU[16],
                    resultat_FMU[17], resultat_FMU[18], resultat_FMU[19], resultat_FMU[20], resultat_FMU[21], resultat_FMU[22], resultat_FMU[23], resultat_FMU[24],
                    resultat_FMU[25], resultat_FMU[26], resultat_FMU[27], resultat_FMU[28], resultat_FMU[29], resultat_FMU[30], resultat_FMU[31], resultat_FMU[32],
                    resultat_FMU[33], resultat_FMU[34],
                    resultat_FMU[35],
                    resultat_FMU[36],
                    ))

    # use the threshold to terminate the simulation
    # if outputs4 > threshold:
        # print("Threshold reached at t = %g s" % time)
        # break

fmu.terminate()
fmu.freeInstance()

# clean up
shutil.rmtree(unzipdir, ignore_errors=True)

# convert the results to a structured NumPy array
result = np.array(rows, dtype=np.dtype([('time', np.float64),
                                        ('Masse H2', np.float64),
                                        ('Pression 1', np.float64), ('Pression 2', np.float64), ('Pression 3', np.float64), ('Pression 4', np.float64), ('Pression 5', np.float64), ('Pression 6', np.float64), ('Pression 7', np.float64), ('Pression 8', np.float64), 
                                        ('Pression 2-1', np.float64), ('Pression 2-2', np.float64), ('Pression 2-3', np.float64), ('Pression 2-4', np.float64), ('Pression 2-5', np.float64), ('Pression 2-6', np.float64), ('Pression 2-7', np.float64), ('Pression 2-8', np.float64), 
                                        ('Température 1', np.float64), ('Température 2', np.float64), ('Température 3', np.float64), ('Température 4', np.float64), ('Température 5', np.float64), ('Température 6', np.float64), ('Température 7', np.float64), ('Température 8', np.float64),
                                        ('Température 2-1', np.float64), ('Température 2-2', np.float64), ('Température 2-3', np.float64), ('Température 2-4', np.float64), ('Température 2-5', np.float64), ('Température 2-6', np.float64), ('Température 2-7', np.float64), ('Température 2-8', np.float64), 
                                        ('Pression HP', np.float64), ('Pression MP', np.float64),
                                        ('Température HP', np.float64),
                                        ('Flux', np.float64)
                                        ]))

# plot the results
if show_plot:
    plot_result(result)

return time

修改方法

要初始化压力参数，核心是在FMU初始化阶段设置参数，而不是仿真循环中（后者是运行时输入），具体步骤如下：

  1. 定位初始参数变量
    OpenModelica导出FMU时，状态变量的初始值通常会生成带_start后缀的变量（比如截图中的P1_start）。你可以遍历model_description.modelVariables确认所有可修改的初始变量：
# 打印所有变量名，找到对应压力的初始变量
for variable in model_description.modelVariables:
    if 'start' in variable.name and 'P' in variable.name:
        print(variable.name)
  1. 获取变量的Value Reference
    在收集vrs字典时，加入目标初始参数的引用：
# 比如获取P1、P2的初始值变量引用
p1_start_vr = vrs['P1_start']
p2_start_vr = vrs['P2_start']
# 按需添加其他压力参数
  1. 在初始化阶段设置参数
    修改初始化代码块，在enterInitializationMode()之后、exitInitializationMode()之前调用setReal()设置初始值：
# initialize
fmu.instantiate()
fmu.setupExperiment(startTime=start_time)
fmu.enterInitializationMode()

# 设置初始压力值，注意单位匹配（示例为350bar转Pa）
fmu.setReal([p1_start_vr, p2_start_vr], [350e5, 350e5])

fmu.exitInitializationMode()

注意事项

  • 单位一致性：确保设置的数值单位与模型内部单位一致（如Pa、bar），否则会导致仿真结果异常。
  • 变量可写性：确认目标变量的variability属性为fixed或tunable，这类变量才能通过FMPY修改初始值。
  • 批量设置：可以一次性传入多个变量的valueReference和对应值，提高效率。

内容的提问来源于stack exchange，提问作者Yaya356

相关产品推荐
方舟 Agent Plan

超全模态模型 × Harness 升级，最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列、Doubao-Seedream-5.0-pro、Kimi-K3 (部分)， 限时 9.9 元起

产品介绍相关文档
相关产品推荐
方舟 Agent Plan

超全模态模型 × Harness 升级，最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列、Doubao-Seedream-5.0-pro、Kimi-K3 (部分)， 限时 9.9 元起

产品介绍相关文档
火山引擎 最新活动
全面开放
Doubao-Seedance-2.5全新模型升级

模型再升级,30秒超长叙事, 模态参考扩容

HOT
方舟Codingplan

模型自由，工具不限，最新支持 Deepseek-V4 系列、GLM-5.3 系列

HOT
方舟Agent plan

超全模态模型 × Harness 升级，最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列

HOT
新用户特惠专场

大模型19元起，Al应用9.9元畅享，新人首购爆款尽享优惠

最近更新时间：2026.08.19 04:40:17