underway-spring.git

			@@ -1,9 +1,15 @@
			package com.flightfeather.uav.biz.sourcetrace.model

			import com.flightfeather.uav.biz.FactorFilter
			import com.flightfeather.uav.common.utils.MapUtil
			import com.flightfeather.uav.domain.entity.SceneInfo
			import com.flightfeather.uav.domain.repository.SceneInfoRep
			import com.flightfeather.uav.lightshare.bean.AreaVo
			import com.flightfeather.uav.lightshare.bean.SceneInfoVo
			import com.flightfeather.uav.lightshare.eunm.SceneType
			import com.flightfeather.uav.socket.eunm.FactorType
			import org.springframework.beans.BeanUtils
			import org.springframework.web.context.ContextLoader
			import kotlin.math.round

			/**
			* 污染来源
			@@ -19,22 +25,29 @@
			*/

			// 溯源企业
			var sceneList:List<SceneInfo?>? = null
			var sceneList: List<SceneInfoVo?>? = null

			// 溯源推理结论
			var conclusion: String? = null

			fun searchScenes(pollutedArea: PollutedArea, pollutedData: PollutedData) {
			ContextLoader.getCurrentWebApplicationContext()?.getBean(SceneInfoRep::class.java)?.run {
			searchScenes(pollutedArea, this, pollutedData)
			}
			}

			/**
			* 查找系统内部溯源范围内的污染企业
			*/
			fun searchScenes(pollutedArea: PollutedArea, sceneInfoRep: SceneInfoRep, factor: FactorFilter.SelectedFactor) {
			fun searchScenes(pollutedArea: PollutedArea, sceneInfoRep: SceneInfoRep, pollutedData: PollutedData) {
			// Fixme 2025.5.14: 污染源的坐标是高德地图坐标系（火星坐标系），而走航数据是WGS84坐标系
			// 按照区域检索内部污染源信息
			// 1. 首先按照四至范围从数据库初步筛选污染源，需要先将坐标转换为gcj02（火星坐标系），因为污染源场景信息都为此坐标系
			val polygonTmp = pollutedArea.polygon!!.map {
			MapUtil.gcj02ToWgs84(it)
			}
			var result = mutableListOf<SceneInfo>()
			// 1. 首先按照四至范围从数据库初步筛选污染源，此处的区域坐标已转换为火星坐标系
			val polygonTmp = pollutedArea.polygon!!
			val fb = MapUtil.calFourBoundaries(polygonTmp)
			val sceneList = sceneInfoRep.findByCoordinateRange(fb)
			// 2. 再精确判断是否在反向溯源区域多边形内部
			val result = mutableListOf<SceneInfo>()
			sceneList.forEach {
			val point = it!!.longitude.toDouble() to it.latitude.toDouble()
			if (MapUtil.isPointInPolygon(point, polygonTmp)) {
			@@ -42,9 +55,125 @@
			}
			}

			this.sceneList = result
			val closePolygonTmp = pollutedArea.closePolygon!!
			val closeFb = MapUtil.calFourBoundaries(closePolygonTmp)
			val closeSceneList = sceneInfoRep.findByCoordinateRange(closeFb)
			// 2. 再精确判断是否在反向溯源区域多边形内部
			closeSceneList.forEach {
			val point = it!!.longitude.toDouble() to it.latitude.toDouble()
			if (MapUtil.isPointInPolygon(point, closePolygonTmp)) {
			result.add(it)
			}
			}

			TODO("按照所选监测因子类型，区分污染源类型")
			// 根据污染因子的量级，计算主要的污染场景类型，筛选结果
			val mainSceneType = calSceneType(pollutedData)
			if (mainSceneType != null) {
			this.conclusion = mainSceneType.first
			result = result.filter {
			val r = mainSceneType.second.find { s->
			s.value == it.typeId.toInt()
			}
			r != null
			}.toMutableList()
			}

			this.sceneList = findClosestStation(sceneInfoRep, result)

			}

			/**
			* 计算可能的相关污染场景类型以及推理结论
			*/
			@Throws(Exception::class)
			private fun calSceneType(pollutedData: PollutedData): Pair<String, List<SceneType>>? {
			when (pollutedData.selectedFactor?.main) {
			// 氮氧化合物，一般由于机动车尾气，同步计算CO
			FactorType.NO2 -> {
			val coAvg = round(pollutedData.dataList.map { it.co!! }.average()) / 1000
			return "氮氧化合物偏高，CO的量级为${coAvg}mg/m³，一般由于机动车尾气造成，污染源以汽修、加油站为主" to
			listOf(SceneType.TYPE6, SceneType.TYPE10, SceneType.TYPE17)
			}

			FactorType.CO -> return null

			FactorType.H2S -> return null

			FactorType.SO2 -> return null

			FactorType.O3 -> return null
			// a) pm2.5、pm10特别高，两者在各情况下同步展示，pm2.5占pm10的比重变化，比重越高，越有可能是餐饮
			// b) pm10特别高、pm2.5较高，大颗粒扬尘污染，只展示pm10，pm2.5占pm10的比重变化，工地为主
			FactorType.PM25,
			FactorType.PM10,
			-> {
			val pm25Avg = round(pollutedData.dataList.map { it.pm25!! }.average() * 10) / 10
			val pm10Avg = round(pollutedData.dataList.map { it.pm10!! }.average() * 10) / 10
			// 计算异常数据的pm2.5占pm10比重的均值
			val percentageAvg = pollutedData.dataList.map {
			it.pm25!! / it.pm10!!
			}.average()
			val str =
			"PM2.5量级为${pm25Avg}μg/m³，PM10量级为${pm25Avg}μg/m³，PM2.5占PM10的比重为${round(percentageAvg * 100)}%"
			return if (percentageAvg > 0.666) {
			"${str}，比重较大，污染源以餐饮为主，工地次之" to
			listOf(SceneType.TYPE1, SceneType.TYPE2, SceneType.TYPE3, SceneType.TYPE14, SceneType.TYPE5)
			} else if (percentageAvg < 0.333) {
			"${str}，比重较小，属于大颗粒扬尘污染，污染源以工地为主" to
			listOf(SceneType.TYPE1, SceneType.TYPE2, SceneType.TYPE3, SceneType.TYPE14, SceneType.TYPE5)
			} else {
			"${str}，污染源以餐饮、工地为主" to
			listOf(SceneType.TYPE1, SceneType.TYPE2, SceneType.TYPE3, SceneType.TYPE14, SceneType.TYPE5)
			}
			}
			// c) VOC较高，同比计算pm2.5的量级，可能存在同步偏高（汽修、加油站）, 同步计算O3是否有高值
			// d) VOC较高，处于加油站（车辆拥堵情况），CO一般较高, 同步计算O3是否有高值
			FactorType.VOC -> {
			val pm25Avg = round(pollutedData.dataList.map { it.pm25!! }.average() * 10) / 10
			val coAvg = round(pollutedData.dataList.map { it.co!! }.average()) / 1000
			val o3Avg = round(pollutedData.dataList.map { it.o3!! }.average() * 10) / 10
			return "VOC偏高，同时PM2.5量级为${pm25Avg}μg/m³，CO量级为${coAvg}mg/m³，O3量级为${o3Avg}μg/m³，污染源以汽修、加油站为主" to
			listOf(SceneType.TYPE6, SceneType.TYPE17, SceneType.TYPE12)
			}

			else -> return null
			}
			}

			/**
			* 计算最近的监测站点
			*/
			private fun findClosestStation(sceneInfoRep: SceneInfoRep, sceneList: List<SceneInfo>): List<SceneInfoVo> {
			val res1 = sceneInfoRep.findByArea(AreaVo().apply {
			sceneTypeId = SceneType.TYPE19.value.toString()
			})

			val res2 = sceneInfoRep.findByArea(AreaVo().apply {
			sceneTypeId = SceneType.TYPE20.value.toString()
			})
			val res = res1.toMutableList().apply { addAll(res2) }

			return sceneList.map {
			var minLen = -1.0
			var selectedRes: SceneInfo? = null
			res.forEach { r ->
			val dis = MapUtil.getDistance(
			it.longitude.toDouble(),
			it.latitude.toDouble(),
			r!!.longitude.toDouble(),
			r.latitude.toDouble()
			)
			if (minLen < 0 \|\| dis < minLen) {
			minLen = dis
			selectedRes = r
			}
			}
			val vo = SceneInfoVo()
			BeanUtils.copyProperties(it, vo)
			vo.closestStation = selectedRes
			vo.length = minLen

			return@map vo
			}
			}
			}