新增tests；新增管道健康预测方法；更新复合查询函数输出格式

timescaledb/composite_queries.py

@@ -1,4 +1,4 @@
-from typing import List, Optional, Any
+from typing import List, Optional, Any, Dict
 from datetime import datetime
 from psycopg import AsyncConnection
 import pandas as pd
@@ -23,7 +23,7 @@ class CompositeQueries:
         device_ids: List[str],
         start_time: datetime,
         end_time: datetime,
-    ) -> List[Optional[Any]]:
+    ) -> Dict[str, List[Dict[str, Any]]]:
         """
         获取 SCADA 关联的 link/node 模拟值
 
@@ -39,12 +39,12 @@ class CompositeQueries:
             field: 要查询的字段名
 
         Returns:
-            模拟数据值列表，如果没有找到则对应位置返回 None
+            模拟数据字典，以 device_id 为键，值为数据列表，每个数据包含 time, value 和 scada_id
 
         Raises:
             ValueError: 当 SCADA 设备未找到或字段无效时
         """
-        results = []
+        result = {}
         # 1. 查询所有 SCADA 信息
         scada_infos = await PostgreScadaRepository.get_scadas(postgres_conn)
 
@@ -75,8 +75,11 @@ class CompositeQueries:
                 )
             else:
                 raise ValueError(f"Unknown SCADA type: {scada_type}")
-            results.append(res)
-        return results
+            # 添加 scada_id 到每个数据项
+            for item in res:
+                item["scada_id"] = device_id
+            result[device_id] = res
+        return result
 
     @staticmethod
     async def get_scada_associated_scheme_simulation_data(
@@ -87,7 +90,7 @@ class CompositeQueries:
         end_time: datetime,
         scheme_type: str,
         scheme_name: str,
-    ) -> List[Optional[Any]]:
+    ) -> Dict[str, List[Dict[str, Any]]]:
         """
         获取 SCADA 关联的 link/node 模拟值
 
@@ -103,12 +106,12 @@ class CompositeQueries:
             field: 要查询的字段名
 
         Returns:
-            模拟数据值列表，如果没有找到则对应位置返回 None
+            模拟数据字典，以 device_id 为键，值为数据列表，每个数据包含 time, value 和 scada_id
 
         Raises:
             ValueError: 当 SCADA 设备未找到或字段无效时
         """
-        results = []
+        result = {}
         # 1. 查询所有 SCADA 信息
         scada_infos = await PostgreScadaRepository.get_scadas(postgres_conn)
 
@@ -151,8 +154,11 @@ class CompositeQueries:
                 )
             else:
                 raise ValueError(f"Unknown SCADA type: {scada_type}")
-            results.append(res)
-        return results
+            # 添加 scada_id 到每个数据项
+            for item in res:
+                item["scada_id"] = device_id
+            result[device_id] = res
+        return result
 
     @staticmethod
     async def get_element_associated_scada_data(
@@ -282,31 +288,31 @@ class CompositeQueries:
             ]
 
             # 处理pressure数据
-            # if pressure_ids:
-            #     pressure_df = df[pressure_ids]
-            #     # 重置索引，将 time 变为普通列
-            #     pressure_df = pressure_df.reset_index()
-            #     # 移除 time 列，准备输入给清洗方法
-            #     value_df = pressure_df.drop(columns=["time"])
-            #     # 调用清洗方法
-            #     cleaned_value_df = clean_pressure_data_df_km(value_df)
-            #     # 添加 time 列到首列
-            #     cleaned_df = pd.concat([pressure_df["time"], cleaned_value_df], axis=1)
-            #     # 将清洗后的数据写回数据库
-            #     for device_id in pressure_ids:
-            #         if device_id in cleaned_df.columns:
-            #             cleaned_values = cleaned_df[device_id].tolist()
-            #             time_values = cleaned_df["time"].tolist()
-            #             for i, time_str in enumerate(time_values):
-            #                 time_dt = datetime.fromisoformat(time_str)
-            #                 value = cleaned_values[i]
-            #                 await ScadaRepository.update_scada_field(
-            #                     timescale_conn,
-            #                     time_dt,
-            #                     device_id,
-            #                     "cleaned_value",
-            #                     value,
-            #                 )
+            if pressure_ids:
+                pressure_df = df[pressure_ids]
+                # 重置索引，将 time 变为普通列
+                pressure_df = pressure_df.reset_index()
+                # 移除 time 列，准备输入给清洗方法
+                value_df = pressure_df.drop(columns=["time"])
+                # 调用清洗方法
+                cleaned_value_df = clean_pressure_data_df_km(value_df)
+                # 添加 time 列到首列
+                cleaned_df = pd.concat([pressure_df["time"], cleaned_value_df], axis=1)
+                # 将清洗后的数据写回数据库
+                for device_id in pressure_ids:
+                    if device_id in cleaned_df.columns:
+                        cleaned_values = cleaned_df[device_id].tolist()
+                        time_values = cleaned_df["time"].tolist()
+                        for i, time_str in enumerate(time_values):
+                            time_dt = datetime.fromisoformat(time_str)
+                            value = cleaned_values[i]
+                            await ScadaRepository.update_scada_field(
+                                timescale_conn,
+                                time_dt,
+                                device_id,
+                                "cleaned_value",
+                                value,
+                            )
 
             # 处理flow数据
             if flow_ids: