跳到主要内容

read_history_processed (聚合)

最强大的 HistoryRead 模式, 配合标准聚合函数. 给区间 + 聚合函数 + 子区间宽度, 服务端返回每个子区间的聚合结果.

前置 / 配套

签名​

def read_history_processed(self,
                           node_id: str,
                           start_utc: datetime,
                           end_utc: datetime,
                           processing_interval_ms: float,
                           aggregate_node_id: str) -> List[OpcUaDataValue]: ...

HistoryAggregate 常量 (推荐)​

from darra_opcua import HistoryAggregate

HistoryAggregate.AVERAGE                    # i=2342  算术平均
HistoryAggregate.MINIMUM                    # i=2346  最小值
HistoryAggregate.MAXIMUM                    # i=2347  最大值
HistoryAggregate.COUNT                      # i=2352  Good 状态点数
HistoryAggregate.TOTAL                      # i=2348  累加 (代数和)
HistoryAggregate.STANDARD_DEVIATION_SAMPLE  # i=11427 样本标准差
HistoryAggregate.TIME_AVERAGE               # i=11285 时间加权平均
HistoryAggregate.RANGE                      # i=2353  Max - Min
HistoryAggregate.DELTA                      # i=11286 区间末-区间始

完整列表见 OPC UA Part 13 §5.4 Aggregate Functions.

用法​

24 小时温度的每小时平均​

import datetime as dt
from darra_opcua import HistoryAggregate

today = dt.datetime.now(dt.timezone.utc).replace(
    hour=0, minute=0, second=0, microsecond=0)
yesterday = today - dt.timedelta(days=1)

values = ua.read_history_processed(
    node_id="ns=2;s=Temperature",
    start_utc=yesterday,
    end_utc=today,
    processing_interval_ms=3_600_000,        # 1 hour
    aggregate_node_id=HistoryAggregate.AVERAGE)
# 返回 24 个 OpcUaDataValue, 每个是该小时的平均值

一周每天的最大温度​

maxes = ua.read_history_processed(
    "ns=2;s=Temperature",
    start_utc=today - dt.timedelta(days=7),
    end_utc=today,
    processing_interval_ms=86_400_000,        # 1 day
    aggregate_node_id=HistoryAggregate.MAXIMUM)

一小时累计流量​

import datetime as dt
now = dt.datetime.now(dt.timezone.utc)

totals = ua.read_history_processed(
    "ns=2;s=FlowRate",
    start_utc=now - dt.timedelta(hours=1),
    end_utc=now,
    processing_interval_ms=3_600_000,         # 1 hour bucket
    aggregate_node_id=HistoryAggregate.TOTAL)

性能优势​

  • 避免拉百万点原始数据再客户端算
  • 服务端常对历史数据建索引, 聚合极快
  • 跨日 / 跨周报表的标准方案

服务端要求​

  • Historian 必须实现该聚合函数 (服务端 ServerCapabilities/AggregateFunctions 列出支持的)
  • 不支持的聚合返回 BadAggregateNotSupported

下一步​