wzy-warehouse
183 lines
7.2 KiB
Python
183 lines
7.2 KiB
Python
"""
|
||
暴雨灾害链预测接口
|
||
"""
|
||
import asyncio
|
||
from datetime import datetime
|
||
from typing import List, Dict, Any, Optional
|
||
|
||
from fastapi import APIRouter, HTTPException
|
||
|
||
from app.schemas.api_schemas import RainfallPredictRequest, PredictResponse, PredictData, UpdateMonitoringTimeRequest
|
||
from app.utils.api_deps import get_rainfall_model, get_prediction_semaphore
|
||
from app.repositories.dbn_repository import dbn_repository
|
||
from app.core.rainfall_manager import rainfall_manager
|
||
from app.config.paths import get_logger
|
||
from app.utils.time_converter import TimeConverter
|
||
|
||
router = APIRouter(prefix="/rainfall", tags=["暴雨灾害链"])
|
||
logger = get_logger("api.rainfall")
|
||
|
||
SOURCE_TYPE_MAP = {1: "隐患点", 2: "风险点"}
|
||
LEVEL_MAP = {"低": "低", "中": "中", "较高": "较高", "高": "高"}
|
||
|
||
|
||
def _build_prediction_map(results: List[Dict[str, Any]]) -> Dict[str, float]:
|
||
"""将模型原始结果转换为存储格式: {id_type: 概率百分比}"""
|
||
result_map = {}
|
||
for r in results:
|
||
probs = r.get("disaster_probabilities", {})
|
||
if not probs:
|
||
continue
|
||
|
||
source_id = r["source_id"]
|
||
source_type = r.get("source_type")
|
||
max_hazard = max(probs, key=probs.get)
|
||
key = f"{source_id}_{source_type}"
|
||
result_map[key] = round(probs[max_hazard] * 100, 2)
|
||
return result_map
|
||
|
||
|
||
def _build_prediction_map_with_location(results: List[Dict[str, Any]], threshold: float = 50.0) -> Dict[str, Dict[str, Any]]:
|
||
"""将模型原始结果转换为返回格式: {id_type: {probability, lon, lat}}"""
|
||
from config import settings
|
||
threshold = getattr(settings, 'PREDICT_PROBABILITY_THRESHOLD', threshold)
|
||
result_map = {}
|
||
for r in results:
|
||
probs = r.get("disaster_probabilities", {})
|
||
if not probs:
|
||
continue
|
||
|
||
source_id = r["source_id"]
|
||
source_type = r.get("source_type")
|
||
max_hazard = max(probs, key=probs.get)
|
||
prob_value = round(probs[max_hazard] * 100, 2)
|
||
# 低于阈值不返回
|
||
if prob_value < threshold:
|
||
continue
|
||
key = f"{source_id}_{source_type}"
|
||
result_map[key] = {
|
||
"probability": prob_value,
|
||
"lon": r.get("lon"),
|
||
"lat": r.get("lat")
|
||
}
|
||
return result_map
|
||
|
||
|
||
def _fetch_points(point_ids: Optional[List[int]], region_code: Optional[str]) -> List[Dict[str, Any]]:
|
||
"""获取点位列表"""
|
||
if point_ids:
|
||
return dbn_repository.get_points_by_ids(point_ids)
|
||
return dbn_repository.get_all_points(region_code)
|
||
|
||
|
||
def _predict_sync(point_ids: Optional[List[int]], region_code: Optional[str],
|
||
rainfall: Optional[float], duration: Optional[float],
|
||
operation_type: str, occurred_time: Optional[datetime] = None) -> tuple:
|
||
"""
|
||
同步执行暴雨预测（在线程池中运行）
|
||
|
||
Args:
|
||
occurred_time: 事件发生时间，用于查询降雨数据和DBN推理
|
||
|
||
Returns:
|
||
(存储用result_map, 返回用result_map_with_location, 传入的条件, 实际使用的事件时间)
|
||
"""
|
||
points = _fetch_points(point_ids, region_code)
|
||
if not points:
|
||
return {}, {}, {}, occurred_time or datetime.now()
|
||
|
||
# 使用传入的时间，如果没有传则使用 rainfall_manager 中的全局查询时间，最后才用当前时间
|
||
if occurred_time:
|
||
query_time = occurred_time
|
||
else:
|
||
from app.core.rainfall_manager import rainfall_manager
|
||
query_time = rainfall_manager.get_current_query_time() or datetime.now()
|
||
|
||
model = get_rainfall_model()
|
||
raw_results = model.predict_multiple_points(points, rainfall=rainfall, duration=duration, query_time=query_time)
|
||
result_map = _build_prediction_map(raw_results) # 用于存储
|
||
result_map_with_location = _build_prediction_map_with_location(raw_results) # 用于返回
|
||
|
||
# 存储传入的原始条件（降雨量和持续时间可能每个点不同，所以存储传入值）
|
||
condition = {
|
||
"point_ids": point_ids,
|
||
"region_code": region_code,
|
||
"rainfall": rainfall,
|
||
"duration": duration,
|
||
"occurred_time": query_time.isoformat() if hasattr(query_time, 'isoformat') else str(query_time)
|
||
}
|
||
|
||
return result_map, result_map_with_location, condition, query_time
|
||
|
||
|
||
@router.post("/update-monitoring-time", summary="更新降雨监测查询时间")
|
||
async def update_monitoring_time(req: UpdateMonitoringTimeRequest):
|
||
"""
|
||
更新降雨站点监测的查询时间，触发重新计算
|
||
|
||
- **query_time**: 新的查询时间，格式: YYYY-MM-DD HH:mm:ss
|
||
"""
|
||
try:
|
||
# 将字符串时间解析为 datetime 对象
|
||
new_time = TimeConverter.parse_input_time(req.query_time)
|
||
|
||
# 更新监测时间，触发重新计算
|
||
result = rainfall_manager.update_query_time(new_time)
|
||
|
||
logger.info(f"更新监测时间成功: {result}")
|
||
return {
|
||
"code": 200,
|
||
"message": "success",
|
||
"data": result
|
||
}
|
||
except ValueError as e:
|
||
logger.error(f"时间格式错误: {e}")
|
||
raise HTTPException(status_code=400, detail=f"时间格式错误: {e}")
|
||
except Exception as e:
|
||
logger.error(f"更新监测时间失败: {e}", exc_info=True)
|
||
raise HTTPException(status_code=500, detail=f"更新监测时间失败: {e}")
|
||
|
||
|
||
@router.post("/predict", response_model=PredictResponse, summary="暴雨灾害链预测")
|
||
async def predict_rainfall(req: RainfallPredictRequest):
|
||
"""
|
||
根据降雨量和持续时间，批量预测隐患点/风险点的灾害概率。
|
||
|
||
- **disaster_name**: 灾害名称
|
||
- **point_ids**: 点位ID列表（可选，不传则查询所有点）
|
||
- **region_code**: 行政区划代码（可选，不传则不限区域）
|
||
- **rainfall**: 累计降雨量(mm)，不传则从气象表自动获取
|
||
- **duration**: 降雨持续时间(h)，不传则从气象表自动获取
|
||
- **operation_type**: 操作类型（如 '实时监测', '情景模拟', '应急评估'）
|
||
"""
|
||
semaphore = get_prediction_semaphore()
|
||
|
||
async with semaphore:
|
||
loop = asyncio.get_event_loop()
|
||
try:
|
||
result_map, result_map_with_location, condition, occurred_time = await loop.run_in_executor(
|
||
None, _predict_sync, req.point_ids, req.region_code,
|
||
req.rainfall, req.duration, req.operation_type, req.occurred_time
|
||
)
|
||
except Exception as e:
|
||
logger.error(f"暴雨预测失败: {e}", exc_info=True)
|
||
raise HTTPException(status_code=500, detail=f"预测失败: {e}")
|
||
|
||
# 保存推理结果
|
||
record_id = None
|
||
if result_map:
|
||
try:
|
||
record_id = dbn_repository.save_inference_result(
|
||
disaster_name=req.disaster_name,
|
||
event_type="rainfall",
|
||
occurred_time=occurred_time,
|
||
operation_type=req.operation_type,
|
||
condition=condition,
|
||
result=result_map
|
||
)
|
||
logger.info(f"推理结果已保存，record_id={record_id}")
|
||
except Exception as e:
|
||
logger.error(f"保存推理结果失败: {e}", exc_info=True)
|
||
|
||
return PredictResponse(code=200, message="success", data=PredictData(record_id=record_id, list=result_map_with_location))
|