保持核心列分开，让a和b只返回新列：
举个简单的例子：

def a(df): 
    return df.select(
        when(col('speed') % 2 == 0, 'svm').otherwise('nn').alias('model_type')
    )

a（df）输出：

+----------+
|model_type|
+----------+
|        nn|
|       svm|
|        nn|
|       svm|
|        nn|
+----------+

简单b示例：

def b(df): 
    return df.select(
        when(col('speed') % 2 == 0, 'v1.0.1').otherwise('v1.0.0').alias('model_version')
    )

输出b（df）：

+-------------+
|model_version|
+-------------+
|       v1.0.0|
|       v1.0.1|
|       v1.0.0|
|       v1.0.1|
|       v1.0.0|
+-------------+

简单测试数据框：

df = spark.createDataFrame([{"speed": 1, "volume": 1}, 
                            {"speed": 2, "volume": 1}, 
                            {"speed": 3, "volume": 1}, 
                            {"speed": 4, "volume": 1}, 
                            {"speed": 5, "volume": 1}])

输出：

+-----+------+
|speed|volume|
+-----+------+
|    1|     1|
|    2|     1|
|    3|     1|
|    4|     1|
|    5|     1|
+-----+------+

zip和Map功能：

def merge(df1, df2): 
    schema = StructType([*df1.schema, *df2.schema]) 
    return spark.createDataFrame(df1.rdd.zip(df2.rdd).map(lambda x: x[0] + x[1]), schema)

结果：

result_df = merge(
    # Original DF with core columns
    df, 
    merge(
        # df with only model_type
        a(df),
        # df with only model_version
        b(df)
    )
)

输出：

+-----+------+----------+-------------+
|speed|volume|model_type|model_version|
+-----+------+----------+-------------+
|    1|     1|        nn|       v1.0.0|
|    2|     1|       svm|       v1.0.1|
|    3|     1|        nn|       v1.0.0|
|    4|     1|       svm|       v1.0.1|
|    5|     1|        nn|       v1.0.0|
+-----+------+----------+-------------+